--- bash-2.05b-2.orig/debian/README.bash_completion +++ bash-2.05b-2/debian/README.bash_completion @@ -0,0 +1,17 @@ + Programmable Completion + ======================= + +[BUG REPORTING: Please direct all enhancement requests directly to the + bash_completion maintainer, the completion script won't be changed for + Debian specific enhancements. If you find a real bug, report it the usual + way. The bash_completion author can be reached at + http://freshmeat.net/projects/bashcompletion/ + + If you find a newer version on freshmeat, you can replace it, if you did + not change the Debian version. +] + +To enable programmable completion for bash on Debian, uncomment the +bash_completion lines in /etc/bash.bashrc to source /etc/bash_completion. +/etc/bash_completion sources ~/.bash_completion, if it exists. + --- bash-2.05b-2.orig/debian/skel.bash_logout +++ bash-2.05b-2/debian/skel.bash_logout @@ -0,0 +1,7 @@ +# ~/.bash_logout: executed by bash(1) when login shell exits. + +# when leaving the console clear the screen to increase privacy + +case "`tty`" in + /dev/tty[0-9]*) clear +esac --- bash-2.05b-2.orig/debian/inputrc.arrows +++ bash-2.05b-2/debian/inputrc.arrows @@ -0,0 +1,30 @@ +# This file controls the behaviour of line input editing for +# programs that use the Gnu Readline library. +# +# Arrow keys in keypad mode +# +"\C-[OD" backward-char +"\C-[OC" forward-char +"\C-[OA" previous-history +"\C-[OB" next-history +# +# Arrow keys in ANSI mode +# +"\C-[[D" backward-char +"\C-[[C" forward-char +"\C-[[A" previous-history +"\C-[[B" next-history +# +# Arrow keys in 8 bit keypad mode +# +"\C-M-OD" backward-char +"\C-M-OC" forward-char +"\C-M-OA" previous-history +"\C-M-OB" next-history +# +# Arrow keys in 8 bit ANSI mode +# +"\C-M-[D" backward-char +"\C-M-[C" forward-char +"\C-M-[A" previous-history +"\C-M-[B" next-history --- bash-2.05b-2.orig/debian/README.abs-guide +++ bash-2.05b-2/debian/README.abs-guide @@ -0,0 +1,26 @@ +[ This is just a pointer to a document, which you might find helpful] + + Advanced Bash-Scripting Guide + + A complete guide to shell scripting, using Bash + + Mendel Cooper - Brindlesoft + + thegrendel@theriver.com + + +This tutorial assumes no previous knowledge of scripting or programming, but +progresses rapidly toward an intermediate/advanced level of instruction +(...all the while sneaking in little snippets of UNIX wisdom and lore). It +serves as a textbook, a manual for self-study, and a reference and source +of knowledge on shell scripting techniques. The exercises and heavily- +commented examples invite active reader participation, under the premise that +the only way to really learn scripting is to write scripts. + +The guide is availabe at http://tldp.org/LDP/abs/html/ + +The latest update of this document, as an archived "tarball" including both +the SGML source and rendered HTML, may be downloaded from the author's home +site (http://personal.riverusers.com/~thegrendel/abs-guide-2.1.tar.bz2). See +the change log for a revision history +(http://personal.riverusers.com/~thegrendel/Change.log). --- bash-2.05b-2.orig/debian/etc.inputrc +++ bash-2.05b-2/debian/etc.inputrc @@ -0,0 +1,13 @@ +# /etc/inputrc - global inputrc for libreadline +# See readline(3readline) and `info readline' for more information. + +# Be 8 bit clean. +set input-meta on +set output-meta on + +# To allow the use of german 8bit-characters like the german umlauts, comment +# out the two lines below. However this makes the meta key not work as a meta +# key, which is annoying to those which don't need to type in 8-bit characters. + +#set meta-flag on +#set convert-meta off --- bash-2.05b-2.orig/debian/patches/rl-header.dpatch +++ bash-2.05b-2/debian/patches/rl-header.dpatch @@ -0,0 +1,28 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: include stdio.h in readline.h +--- lib/readline/readline.h~ Thu Aug 5 14:10:59 1999 ++++ lib/readline/readline.h Mon Oct 9 20:32:56 2000 +@@ -32,6 +32,7 @@ + # include "keymaps.h" + # include "tilde.h" + #else ++# include + # include + # include + # include --- bash-2.05b-2.orig/debian/patches/deb-examples.dpatch +++ bash-2.05b-2/debian/patches/deb-examples.dpatch @@ -0,0 +1,30 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: document readline header location on Debian systems + +diff -u ./examples/loadables/README.orig ./examples/loadables/README +--- ./examples/loadables/README.orig Thu May 7 20:31:34 1998 ++++ ./examples/loadables/README Thu Nov 11 20:32:57 1999 +@@ -31,3 +31,6 @@ + the canonical example. There is no real `builtin writers' programming + guide'. The file template.c provides a template to use for creating + new loadable builtins. ++ ++On Debian GNU/Linux systems, the bash headers are in /usr/include/bash. ++The appropriate options are already set in the example Makefile. + --- bash-2.05b-2.orig/debian/patches/man-fignore.dpatch +++ bash-2.05b-2/debian/patches/man-fignore.dpatch @@ -0,0 +1,32 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash(1): mention quoting when assigning to FIGNORE + +--- doc/bash.1~ Fri May 31 00:45:57 2002 ++++ doc/bash.1 Fri May 31 00:48:52 2002 +@@ -1456,7 +1456,9 @@ + is excluded from the list of matched filenames. + A sample value is + .if t \f(CW".o:~"\fP. +-.if n ".o:~". ++.if n ".o:~" ++(Quoting is needed when assigning a value to this variable, ++which contains tildes). + .TP + .B GLOBIGNORE + A colon-separated list of patterns defining the set of filenames to --- bash-2.05b-2.orig/debian/patches/template.dpatch +++ bash-2.05b-2/debian/patches/template.dpatch @@ -0,0 +1,18 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: --- bash-2.05b-2.orig/debian/patches/rl-inputrc.dpatch +++ bash-2.05b-2/debian/patches/rl-inputrc.dpatch @@ -0,0 +1,91 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: Support for /etc/inputrc in addition to ~/inputrc + +diff -r -u readline4-4.2.orig/bind.c readline4-4.2/bind.c +--- readline4-4.2.orig/lib/readline/bind.c Wed Nov 8 18:39:01 2000 ++++ readline4-4.2/lib/readline/bind.c Tue Feb 6 18:24:42 2001 +@@ -593,6 +593,9 @@ + /* The last key bindings file read. */ + static char *last_readline_init_file = (char *)NULL; + ++/* Flag to read system init file */ ++static int read_system_init_file = 0; ++ + /* The file we're currently reading key bindings from. */ + static const char *current_readline_init_file; + static int current_readline_init_include_level; +@@ -670,7 +673,7 @@ + to the first non-null filename from this list: + 1. the filename used for the previous call + 2. the value of the shell variable `INPUTRC' +- 3. ~/.inputrc ++ 3. /etc/inputrc and ~/.inputrc + If the file existed and could be opened and read, 0 is returned, + otherwise errno is returned. */ + int +@@ -681,14 +684,23 @@ + if (filename == 0) + { + filename = last_readline_init_file; +- if (filename == 0) ++ if (filename == 0) { + filename = sh_get_env_value ("INPUTRC"); +- if (filename == 0) ++ read_system_init_file = 0; ++ } ++ if (filename == 0) { + filename = DEFAULT_INPUTRC; ++ read_system_init_file = 1; ++ } + } + +- if (*filename == 0) ++ if (*filename == 0) { + filename = DEFAULT_INPUTRC; ++ read_system_init_file = 1; ++ } ++ ++ if (read_system_init_file) ++ _rl_read_init_file (SYSTEM_INPUTRC, 0); + + #if defined (__MSDOS__) + if (_rl_read_init_file (filename, 0) == 0) +--- readline4-4.2.orig/lib/readline/doc/rluser.texinfo Mon Feb 5 15:38:11 2001 ++++ readline4-4.2/lib/readline/doc/rluser.texinfo Tue Feb 6 18:24:42 2001 +@@ -332,7 +332,8 @@ + @ifclear BashFeatures + file is taken from the value of the environment variable @env{INPUTRC}. If + @end ifclear +-that variable is unset, the default is @file{~/.inputrc}. ++that variable is unset, Readline will read both @file{/etc/inputrc} and ++@file{~/.inputrc}. + + When a program which uses the Readline library starts up, the + init file is read, and the key bindings are set. +diff -r -u readline4-4.2.orig/rlconf.h readline4-4.2/rlconf.h +--- readline4-4.2.orig/lib/readline/rlconf.h Thu Aug 5 14:11:14 1999 ++++ readline4-4.2/lib/readline/rlconf.h Tue Feb 6 18:24:42 2001 +@@ -39,6 +39,7 @@ + + /* The final, last-ditch effort file name for an init file. */ + #define DEFAULT_INPUTRC "~/.inputrc" ++#define SYSTEM_INPUTRC "/etc/inputrc" + + /* If defined, expand tabs to spaces. */ + #define DISPLAY_TABS --- bash-2.05b-2.orig/debian/patches/man-vx-opts.dpatch +++ bash-2.05b-2/debian/patches/man-vx-opts.dpatch @@ -0,0 +1,34 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: document -v / -x options + +--- ./doc/bash.1~ Wed Jan 20 22:48:04 1999 ++++ ./doc/bash.1 Sun Nov 14 13:26:59 1999 +@@ -104,6 +104,12 @@ + This option allows the positional parameters to be set + when invoking an interactive shell. + .TP ++.B \-v ++Print shell input lines as they are read. ++.TP ++.B \-x ++Print commands and their arguments as they are executed. ++.TP + .B \-D + A list of all double-quoted strings preceded by \fB$\fP + is printed on the standard ouput. --- bash-2.05b-2.orig/debian/patches/bashbug-editor.dpatch +++ bash-2.05b-2/debian/patches/bashbug-editor.dpatch @@ -0,0 +1,29 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: send bug reports to Debian bash maintainer too. + +--- ./support/bashbug.sh~ Mon Nov 27 17:15:21 2000 ++++ ./support/bashbug.sh Tue Feb 6 18:54:48 2001 +@@ -98,6 +98,7 @@ + esac ;; + esac + ++BUGBASH="${BUGBASH},bash@packages.debian.org" + BUGADDR="${1-$BUGBASH}" + + if [ -z "$DEFEDITOR" ] && [ -z "$EDITOR" ]; then --- bash-2.05b-2.orig/debian/patches/deb-bash-config.dpatch +++ bash-2.05b-2/debian/patches/deb-bash-config.dpatch @@ -0,0 +1,83 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: Changed compile time configuration options: +# DP: +# DP: - Set the default path to comply with Debian policy +# DP: +# DP: - Define DONT_REPORT_SIGPIPE: We don't want to see `Broken pipe' messages +# DP: when a job like `cat jobs.c | exit 1' is executed. Fixes part of #7047, +# DP: #10259, #10433 and #10494. Comment from the upstream author: "The +# DP: default bash behavior with respect to the exit status of a pipeline +# DP: will not change. Changing it as suggested in the discussion of #10494 +# DP: would render bash incompatible with every other shell out there.". +# DP: +# DP: - Enable System-wide .bashrc file for interactive shells. +# DP: +# DP: - Enable System-wide .bash_logout file for interactive shells. +# DP: +# DP: - make non-interactive shells begun with argv[0][0] == '-' +# DP: run the startup files when not in posix mode. +# DP: +# DP: - try to check whether bash is being run by sshd and source +# DP: the .bashrc if so (like the rshd behavior). + +--- bash-2.05a.orig/config-top.h Wed Oct 24 19:28:49 2001 ++++ bash-2.05a/config-top.h Fri Nov 2 11:16:00 2001 +@@ -29,19 +29,19 @@ + + /* Define DONT_REPORT_SIGPIPE if you don't want to see `Broken pipe' messages + when a job like `cat jobs.c | exit 1' is executed. */ +-/* #define DONT_REPORT_SIGPIPE */ ++#define DONT_REPORT_SIGPIPE + + /* The default value of the PATH variable. */ + #ifndef DEFAULT_PATH_VALUE + #define DEFAULT_PATH_VALUE \ +- "/usr/gnu/bin:/usr/local/bin:/usr/ucb:/bin:/usr/bin:." ++ "/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin" + #endif + + /* The value for PATH when invoking `command -p'. This is only used when + the Posix.2 confstr () function, or CS_PATH define are not present. */ + #ifndef STANDARD_UTILS_PATH + #define STANDARD_UTILS_PATH \ +- "/bin:/usr/bin:/usr/ucb:/sbin:/usr/sbin:/etc:/usr/etc" ++ "/bin:/usr/bin:/sbin:/usr/sbin" + #endif + + /* Default primary and secondary prompt strings. */ +@@ -49,15 +49,15 @@ + #define SPROMPT "> " + + /* System-wide .bashrc file for interactive shells. */ +-/* #define SYS_BASHRC "/etc/bash.bashrc" */ ++#define SYS_BASHRC "/etc/bash.bashrc" + + /* System-wide .bash_logout for login shells. */ +-/* #define SYS_BASH_LOGOUT "/etc/bash.bash_logout" */ ++#define SYS_BASH_LOGOUT "/etc/bash.bash_logout" + + /* Define this to make non-interactive shells begun with argv[0][0] == '-' + run the startup files when not in posix mode. */ +-/* #define NON_INTERACTIVE_LOGIN_SHELLS */ ++#define NON_INTERACTIVE_LOGIN_SHELLS + + /* Define this if you want bash to try to check whether it's being run by + sshd and source the .bashrc if so (like the rshd behavior). */ +-/* #define SSH_SOURCE_BASHRC */ ++#define SSH_SOURCE_BASHRC --- bash-2.05b-2.orig/debian/patches/rl-deb-build.dpatch +++ bash-2.05b-2/debian/patches/rl-deb-build.dpatch @@ -0,0 +1,85 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: Debian specific patches that allow libreadline to be built +# DP: as a static _and_ a shared library. + +--- ./lib/readline/Makefile.in~ Thu Feb 18 18:12:41 1999 ++++ ./lib/readline/Makefile.in Thu Nov 11 13:35:06 1999 +@@ -5,7 +5,8 @@ + ############################################################################# + + srcdir = @srcdir@ +-VPATH = .:@srcdir@ ++vpath %.c $(srcdir) ++vpath %.h $(srcdir) + topdir = @top_srcdir@ + BUILD_DIR = @BUILD_DIR@ + +@@ -32,12 +33,16 @@ + CPPFLAGS = @CPPFLAGS@ + LDFLAGS = @LDFLAGS@ + +-DEFS = @DEFS@ ++ifndef NOFPIC ++PICFLAGS = -fPIC ++endif ++ ++DEFS = @DEFS@ -D_REENTRANT + LOCAL_DEFS = @LOCAL_DEFS@ + + INCLUDES = -I. -I$(BUILD_DIR) -I$(topdir) -I$(topdir)/lib + +-CCFLAGS = $(DEFS) $(LOCAL_DEFS) $(APP_CFLAGS) $(CPPFLAGS) ${INCLUDES} $(LOCAL_CFLAGS) $(CFLAGS) ++CCFLAGS = $(DEFS) $(LOCAL_DEFS) $(APP_CFLAGS) $(CPPFLAGS) ${INCLUDES} $(LOCAL_CFLAGS) $(CFLAGS) $(PICFLAGS) + + .c.o: + ${RM} $@ +@@ -86,7 +91,22 @@ + + ########################################################################## + +-all: libreadline.a libhistory.a ++all: libreadline.so libhistory.so staticlibs ++ ++staticlibs: ++ test -d static || mkdir static ++ $(MAKE) -C static -f $(BUILD_DIR)/lib/readline/Makefile \ ++ srcdir=$(srcdir) NOFPIC=1 libreadline.a libhistory.a ++ ++libreadline.so: $(OBJECTS) ++ $(RM) -f $@ ++ $(CC) $(LDFLAGS) -shared -Wl,-soname,libreadline.so.$(SOVERSION) \ ++ -o libreadline.so $(OBJECTS) -lncurses ++ ++libhistory.so: history.o ++ $(RM) -f $@ ++ $(CC) $(LDFLAGS) -shared -Wl,-soname,libhistory.so.$(SOVERSION) \ ++ -o libhistory.so $(HISTOBJ) xmalloc.o + + libreadline.a: $(OBJECTS) + $(RM) $@ +@@ -117,7 +137,8 @@ + $(CTAGS) $(CSOURCES) $(HSOURCES) + + clean: force +- $(RM) $(OBJECTS) *.a ++ -$(RM) $(OBJECTS) *.a *.so ++ -$(RM) -r static + -( cd doc && $(MAKE) $(MFLAGS) $@ ) + + mostlyclean: clean --- bash-2.05b-2.orig/debian/patches/rl-examples.dpatch +++ bash-2.05b-2/debian/patches/rl-examples.dpatch @@ -0,0 +1,43 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: Modified readline examples to properly build with Debian setup. + +--- ./lib/readline/examples/Makefile.orig Fri Nov 26 23:13:35 1999 ++++ ./lib/readline/examples/Makefile Fri Nov 26 23:15:19 1999 +@@ -1,8 +1,8 @@ + # This is the Makefile for the examples subdirectory of readline. -*- text -*- + # + EXECUTABLES = fileman rltest rl +-CFLAGS = -g -I../.. -I.. -DREADLINE_LIBRARY +-LDFLAGS = -g -L.. ++CFLAGS = -g ++LDFLAGS = -g + + .c.o: + $(CC) $(CFLAGS) -c $< +--- ./lib/readline/examples/rl.c~ Mon Nov 8 22:22:03 1999 ++++ ./lib/readline/examples/rl.c Fri Dec 31 17:34:30 1999 +@@ -11,7 +11,7 @@ + + #include + #include +-#include "posixstat.h" ++#include + + #if defined (READLINE_LIBRARY) + # include "readline.h" --- bash-2.05b-2.orig/debian/patches/rl-del-backspace-policy.dpatch +++ bash-2.05b-2/debian/patches/rl-del-backspace-policy.dpatch @@ -0,0 +1,47 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: Handle Debian Backspace/Delete keyboard policy + +--- lib/readline/terminal.c.orig Mon Mar 4 18:23:09 2002 ++++ lib/readline/terminal.c Fri Jul 19 08:25:48 2002 +@@ -148,6 +148,9 @@ + /* Insert key */ + static char *_rl_term_kI; + ++/* The key sequence sent by the Delete key, if any. */ ++static char *_rl_term_kD; ++ + /* Cursor control */ + static char *_rl_term_vs; /* very visible */ + static char *_rl_term_ve; /* normal */ +@@ -314,6 +317,7 @@ + { "ic", &_rl_term_ic }, + { "im", &_rl_term_im }, + { "kH", &_rl_term_kH }, /* home down ?? */ ++ { "kD", &_rl_term_kD }, /* delete */ + { "kI", &_rl_term_kI }, /* insert */ + { "kd", &_rl_term_kd }, + { "ke", &_rl_term_ke }, /* end keypad mode */ +@@ -496,6 +500,7 @@ + + _rl_bind_if_unbound (_rl_term_kh, rl_beg_of_line); /* Home */ + _rl_bind_if_unbound (_rl_term_at7, rl_end_of_line); /* End */ ++ _rl_bind_if_unbound (_rl_term_kD, rl_delete); /* Delete */ + + _rl_keymap = xkeymap; + } --- bash-2.05b-2.orig/debian/patches/privmode.dpatch +++ bash-2.05b-2/debian/patches/privmode.dpatch @@ -0,0 +1,36 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: XXX missing description +# DP: +# DP: Comment from Chet Ramey : +# DP: +# DP: Nope. This will allow setuid scripts if not called as `sh' and not +# DP: called with the -p option. I won't install this. + + +--- bash-2.02.1.orig/shell.c ++++ bash-2.02.1/shell.c +@@ -378,7 +378,7 @@ + if (dump_translatable_strings) + read_but_dont_execute = 1; + +- if (running_setuid && privileged_mode == 0) ++ if (running_setuid && privileged_mode == 0 && act_like_sh == 0) + disable_priv_mode (); + + /* Need to get the argument to a -c option processed in the --- bash-2.05b-2.orig/debian/patches/various.dpatch +++ bash-2.05b-2/debian/patches/various.dpatch @@ -0,0 +1,29 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: a header include change + +--- bash-2.02.1.orig/general.h ++++ bash-2.02.1/general.h +@@ -21,6 +21,7 @@ + #if !defined (_GENERAL_H_) + #define _GENERAL_H_ + ++#include + #include "stdc.h" + + #include "bashtypes.h" --- bash-2.05b-2.orig/debian/patches/man-builtin.dpatch +++ bash-2.05b-2/debian/patches/man-builtin.dpatch @@ -0,0 +1,29 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: fix man page include + +--- doc/builtins.1~ Mon Nov 29 22:30:13 1999 ++++ doc/builtins.1 Tue Aug 1 21:54:06 2000 +@@ -10,6 +10,6 @@ + ulimit, umask, unalias, unset, until, wait, while \- bash built-in commands, see \fBbash\fR(1) + .SH BASH BUILTIN COMMANDS + .nr zZ 1 +-.so bash.1 ++.so /usr/share/man/man1/bash.1 + .SH SEE ALSO + bash(1), sh(1) --- bash-2.05b-2.orig/debian/patches/man-arithmetic.dpatch +++ bash-2.05b-2/debian/patches/man-arithmetic.dpatch @@ -0,0 +1,31 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: document deprecated syntax for arithmetic evaluation. + +--- doc/bash.1~ Sun Apr 29 23:35:03 2001 ++++ doc/bash.1 Sat Jul 21 21:48:33 2001 +@@ -2301,6 +2301,9 @@ + \fB$((\fP\fIexpression\fP\fB))\fP + .RE + .PP ++The old format \fB$[\fP\fIexpression\fP\fB]\fP is deprecated and will ++be removed in upcoming versions of bash. ++.PP + The + .I expression + is treated as if it were within double quotes, but a double quote --- bash-2.05b-2.orig/debian/patches/report-155436.dpatch +++ bash-2.05b-2/debian/patches/report-155436.dpatch @@ -0,0 +1,62 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +From: Satoshi Koike +Sender: Satoshi Koike +To: submit@bugs.debian.org +Subject: Bug#155436: bash: Completion by TAB in quoted sequence(`...`) with Multibyte Locale +Date: Mon, 05 Aug 2002 06:29:19 +0900 + +Package: bash +Version: 2.05b-2 +Severity: normal + +# DP: Completion by TAB in quoted sequence(`...`) with Multibyte Locale + +I use bash with Japanese Locale (ja_JP.eucJP). When I use the command +completion by [TAB] key in quoted sequence using `...`, bash freeze +and I must push Ctrl-C for getting bash prompt. + +The followed is example. + +> $ echo `pwd[TAB] +[bash is freezing and pushing Ctrl-C will get prompt] + +I checked the source code, and I knew that bash stopped in function +unclosed_pair(char *, int, char *) in subst.c. unclosed_pair() uses +the macro ADVANCE_CHAR defined in include/shmbutil.h. This macro +increases the pointer variable, but it doesn't increase in the +multibyte environment. Because in the multibyte environment, it +increases the pointer using the return value of mbrlen(3). mbrlen() +returns ZERO when the given word is the termination. + +So, I think that the macro ADVANCED_CHAR will be change like the +followed patch. + +------------------------------ patch ------------------------------ +--- include/shmbutil.h.orig 2002-08-05 06:26:13.000000000 +0900 ++++ include/shmbutil.h 2002-08-05 06:19:37.000000000 +0900 +@@ -120,6 +120,8 @@ + state = state_bak; \ + (_i)++; \ + } \ ++ else if (mblength == 0) \ ++ (_i)++; \ + else \ + (_i) += mblength; \ + } \ +------------------------------ patch ------------------------------ --- bash-2.05b-2.orig/debian/patches/bash205b-001.dpatch +++ bash-2.05b-2/debian/patches/bash205b-001.dpatch @@ -0,0 +1,52 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash-2.05b upstream patch 001 + + BASH PATCH REPORT + ================= + +Bash-Release: 2.05b +Patch-ID: bash205b-001 + +Bug-Reported-by: taviso@sdf.lonestar.org +Bug-Reference-ID: <200208021107.g72B723d021352@darkstar.example.net> +Bug-Reference-URL: http://mail.gnu.org/pipermail/bug-bash/2002-August/004980.html + +Bug-Description: + +Bash goes into an infinite loop and eventually crashes with a SIGSEGV +when some keys on the numeric keypad are pressed, possibly in combination +with the shift key. + +Patch: + +*** ../bash-2.05b/lib/readline/bind.c Thu Jan 24 11:15:52 2002 +--- lib/readline/bind.c Wed Jul 31 09:11:18 2002 +*************** +*** 312,316 **** + and the function bound to `a' to be executed when the user + types `abx', leaving `bx' in the input queue. */ +! if (k.function /* && k.type == ISFUNC */) + { + map[ANYOTHERKEY] = k; +--- 312,316 ---- + and the function bound to `a' to be executed when the user + types `abx', leaving `bx' in the input queue. */ +! if (k.function && ((k.type == ISFUNC && k.function != rl_do_lowercase_version) || k.type == ISMACR)) + { + map[ANYOTHERKEY] = k; --- bash-2.05b-2.orig/debian/patches/bash205b-002.dpatch +++ bash-2.05b-2/debian/patches/bash205b-002.dpatch @@ -0,0 +1,46 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash-2.05b upstream patch 002 + + BASH PATCH REPORT + ================= + +Bash-Release: 2.05b +Patch-ID: bash205b-002 + +Bug-Reported-by: rwhron@earthlink.net +Bug-Reference-ID: <20020724000454.GA15210@rushmore> (bug-bash) +Bug-Reference-URL: http://mail.gnu.org/pipermail/bug-bash/2002-July/004856.html + +Bug-Description: + +Repeating an edit in vi-mode with `.' does not work. + +Patch: + +*** ../bash-2.05b/lib/readline/readline.c Wed Mar 13 17:10:46 2002 +--- lib/readline/readline.c Tue Jul 30 17:46:44 2002 +*************** +*** 685,688 **** +--- 685,689 ---- + #if defined (VI_MODE) + if (rl_editing_mode == vi_mode && _rl_keymap == vi_movement_keymap && ++ key != ANYOTHERKEY && + _rl_vi_textmod_command (key)) + _rl_vi_set_last (key, rl_numeric_arg, rl_arg_sign); + --- bash-2.05b-2.orig/debian/patches/bash205b-003.dpatch +++ bash-2.05b-2/debian/patches/bash205b-003.dpatch @@ -0,0 +1,77 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash-2.05b upstream patch 003 + + BASH PATCH REPORT + ================= + +Bash-Release: 2.05b +Patch-ID: bash205b-003 + +Bug-Reported-by: jan.q.bruun-andersen@posten.se +Bug-Reference-ID: +Bug-Reference-URL: http://mail.gnu.org/pipermail/bug-bash/2002-July/004789.html + +Bug-Description: + +In certain cases, Bash appends a space instead of a slash to a directory +name relative to the current directory when performing command name +completion. This affects partial completion of intermediate directory +names. + +Patch: + +*** ../bash-2.05b/bashline.c Tue May 7 15:52:42 2002 +--- bashline.c Sat Aug 3 11:40:16 2002 +*************** +*** 1045,1049 **** +--- 1045,1052 ---- + else + { ++ #define CMD_IS_DIR(x) (absolute_pathname(x) == 0 && *(x) != '~' && test_for_directory (x)) ++ + matches = rl_completion_matches (text, command_word_completion_function); ++ + /* If we are attempting command completion and nothing matches, we + do not want readline to perform filename completion for us. We +*************** +*** 1053,1057 **** + if (matches == (char **)NULL) + rl_ignore_some_completions_function = bash_ignore_filenames; +! else if (matches[1] == 0 && *matches[0] != '/') + /* Turn off rl_filename_completion_desired so readline doesn't + append a slash if there is a directory with the same name +--- 1056,1060 ---- + if (matches == (char **)NULL) + rl_ignore_some_completions_function = bash_ignore_filenames; +! else if (matches[1] == 0 && CMD_IS_DIR(matches[0])) + /* Turn off rl_filename_completion_desired so readline doesn't + append a slash if there is a directory with the same name +*************** +*** 1062,1066 **** + conflict. */ + rl_filename_completion_desired = 0; +! else if (matches[0] && matches[1] && STREQ (matches[0], matches[1]) && *matches[0] != '/') + /* There are multiple instances of the same match (duplicate + completions haven't yet been removed). In this case, all of +--- 1065,1069 ---- + conflict. */ + rl_filename_completion_desired = 0; +! else if (matches[0] && matches[1] && STREQ (matches[0], matches[1]) && CMD_IS_DIR (matches[0])) + /* There are multiple instances of the same match (duplicate + completions haven't yet been removed). In this case, all of --- bash-2.05b-2.orig/debian/patches/bash205b-004.dpatch +++ bash-2.05b-2/debian/patches/bash205b-004.dpatch @@ -0,0 +1,60 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash-2.05b upstream patch 004 + + BASH PATCH REPORT + ================= + +Bash-Release: 2.05b +Patch-ID: bash205b-004 + +Bug-Reported-by: c.f.a.johnson@rogers.com +Bug-Reference-ID: +Bug-Reference-URL: http://mail.gnu.org/pipermail/bug-bash/2002-August/005074.html + +Bug-Description: + +A bug in the bash variable expansion code causes leading blanks in a +variable's value to be ignored when computing its length. + +Patch: + +*** ../bash-2.05b/subst.c Mon Jun 24 07:59:45 2002 +--- subst.c Sat Aug 17 17:28:46 2002 +*************** +*** 1639,1647 **** + /* This performs word splitting and quoted null character removal on + STRING. */ +! #if 0 +! #define issep(c) ((separators)[1] ? (member ((c), separators)) : (c) == (separators)[0]) +! #else +! #define issep(c) ((separators)[1] ? isifs(c) : (c) == (separators)[0]) +! #endif + + WORD_LIST * +--- 1639,1646 ---- + /* This performs word splitting and quoted null character removal on + STRING. */ +! #define issep(c) \ +! (((separators)[0]) ? ((separators)[1] ? isifs(c) \ +! : (c) == (separators)[0]) \ +! : 0) + + WORD_LIST * + + --- bash-2.05b-2.orig/debian/patches/s390-build.dpatch +++ bash-2.05b-2/debian/patches/s390-build.dpatch @@ -0,0 +1,82 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: eliminate build warning on s390 + +From: Chet Ramey +To: Gerhard Tonn +Cc: chet@po.cwru.edu, doko@debian.org +Subject: Debian Bash bug #160731 +Date: Thu, 26 Sep 2002 14:04:56 -0400 + +Try this patch, and see what happens. It should eliminate the warning, +at least. Your line numbers will vary. + +Chet + +*** ../bash-2.05b/parse.y Tue May 21 11:57:30 2002 +--- ./parse.y Thu Sep 26 12:08:19 2002 +*************** +*** 4525,4529 **** + set_line_mbstate () + { +! int i, previ, len; + mbstate_t mbs, prevs; + size_t mbclen; +--- 4534,4538 ---- + set_line_mbstate () + { +! int i, previ, len, c; + mbstate_t mbs, prevs; + size_t mbclen; +*************** +*** 4540,4544 **** + mbs = prevs; + +! if (shell_input_line[i] == EOF) + { + int j; +--- 4549,4554 ---- + mbs = prevs; + +! c = shell_input_line[i]; +! if (c == EOF) + { + int j; +*************** +*** 4564,4568 **** + else + { +! /* mbrlen doesn't return any other values */ + } + +--- 4574,4582 ---- + else + { +! /* XXX - what to do if mbrlen returns 0? (null wide character) */ +! int j; +! for (j = i; j < len; j++) +! shell_input_line_property[j] = 1; +! break; + } + + +-- +``The lyf so short, the craft so long to lerne.'' - Chaucer +( ``Discere est Dolere'' -- chet ) + +Chet Ramey, ITS, CWRU chet@po.CWRU.Edu http://cnswww.cns.cwru.edu/~chet/ --- bash-2.05b-2.orig/debian/patches/rl-8bit-init.dpatch +++ bash-2.05b-2/debian/patches/rl-8bit-init.dpatch @@ -0,0 +1,57 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +On Tue, 29 Oct 2002 19:11:32 +0900 +Fumitoshi UKAI wrote: + +--- bash-2.05b.orig/lib/readline/nls.c 2001-10-16 03:32:29.000000000 +0900 ++++ bash-2.05b/lib/readline/nls.c 2002-10-29 19:07:36.000000000 +0900 +@@ -87,7 +87,8 @@ + char *t; + + /* Set the LC_CTYPE locale category from environment variables. */ +- t = setlocale (LC_CTYPE, ""); ++ // t = setlocale (LC_CTYPE, ""); ++ t = setlocale (LC_CTYPE, NULL); + if (t && *t && (t[0] != 'C' || t[1]) && (STREQ (t, "POSIX") == 0)) + { + _rl_meta_flag = 1; + +I explain about this patch. + +0) bash invokes with system locale setting configured in /etc/environment + or C locale(from login, sshd or others). +1) eval ~/.bashrc and others. + if there are locale setting, such as LANG=ll_CC, it will call + setlocale(LC_ALL, value); in bash/locale.c:set_lang() +2) after startup script done, bash initializes terminal. + while terminal initialization, it checks current locale + in lib/readline/nls.c:_rl_init_eightbit(). + However, it will call setlocale(LC_CTYPE, ""), so it will + *modify* locale category LC_CTYPE according to process environment. + Here, bash runs in LANG=C or something configured at starting up - 0). + Locale configuration in 1) doesn't change bash's environment, it only + call setlocale(LC_ALL, value) and set environment variables for bash's + child process. + So, readline consider that bash runs in C locale (configued in 0)), + so that it disable 8bit or multibyte features. + +It can be solved with this patch. In lib/readline/nls.c:_rl_init_eightbit(), +it doesn't need to *modify* LC_CTYPE. What is required here is to *query* +locale category LC_CTYPE. So, it should use NULL for locale args for +setlocale() here. + --- bash-2.05b-2.orig/debian/patches/rbash-manpage.dpatch +++ bash-2.05b-2/debian/patches/rbash-manpage.dpatch @@ -0,0 +1,29 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: doc/rbash.1: fix bash(1) reference + +--- doc/rbash.1~ 1999-11-29 22:30:03.000000000 +0100 ++++ doc/rbash.1 2003-02-21 16:15:59.000000000 +0100 +@@ -3,6 +3,6 @@ + rbash \- restricted bash, see \fBbash\fR(1) + .SH RESTRICTED SHELL + .nr zY 1 +-.so bash.1 ++.so man1/bash.1 + .SH SEE ALSO + bash(1) --- bash-2.05b-2.orig/debian/patches/builtins-shift.dpatch +++ bash-2.05b-2/debian/patches/builtins-shift.dpatch @@ -0,0 +1,41 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +From: Ken Shan +Cc: 159996@bugs.debian.org +Subject: Bug#159996: bash: shift with no argument causes segfault +Date: Tue, 22 Oct 2002 01:09:31 -0400 + +# DP: Subject: Bug#159996: bash: shift with no argument causes segfault +# DP: The bug is only triggered by "shopt -s shift_verbose". It is a null +# DP: pointer dereference caused by an erroneous assumption that, when the +# DP: number of arguments to shift exceeds the number of arguments available +# DP: for shifting, the latter is always explicitly specified on the command +# DP: line rather than left implicit at the default 1. Here is a patch: + +--- builtins/shift.def.orig 2002-10-22 01:05:10.000000000 -0400 ++++ builtins/shift.def 2002-10-22 01:05:06.000000000 -0400 +@@ -68,7 +68,7 @@ + else if (times > number_of_args ()) + { + if (print_shift_error) +- sh_erange (list->word->word, "shift count"); ++ sh_erange (list ? list->word->word : NULL, "shift count"); + return (EXECUTION_FAILURE); + } + + --- bash-2.05b-2.orig/debian/patches/execute-cmd.dpatch +++ bash-2.05b-2/debian/patches/execute-cmd.dpatch @@ -0,0 +1,121 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p1 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p1 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: execute_cmd.c: conditionalize on enabled bash features + +--- src/execute_cmd.c~ 2002-03-18 19:24:22.000000000 +0100 ++++ src/execute_cmd.c 2003-05-25 22:08:02.000000000 +0200 +@@ -286,12 +286,18 @@ + { + if (currently_executing_command->type == cm_simple) + return currently_executing_command->value.Simple->line; ++#if defined (COND_COMMAND) + else if (currently_executing_command->type == cm_cond) + return currently_executing_command->value.Cond->line; ++#endif ++#if defined (DPAREN_ARITHMETIC) + else if (currently_executing_command->type == cm_arith) + return currently_executing_command->value.Arith->line; ++#endif ++#if defined (ARITH_FOR_COMMAND) + else if (currently_executing_command->type == cm_arith_for) + return currently_executing_command->value.ArithFor->line; ++#endif + else + return line_number; + } +@@ -801,7 +807,7 @@ + exec_result = execute_cond_command (command->value.Cond); + break; + #endif +- ++ + case cm_function_def: + exec_result = execute_intern_function (command->value.Function_def->name, + command->value.Function_def->command); +@@ -916,7 +922,7 @@ + characters after the decimal point, the optional `l' means to format + using minutes and seconds (MMmNN[.FF]s), like the `times' builtin', + and the last character is one of +- ++ + R number of seconds of `real' time + U number of seconds of `user' time + S number of seconds of `system' time +@@ -2351,7 +2357,7 @@ + #if 0 + debug_print_cond_command (cond_command); + #endif +- last_command_exit_value = result = execute_cond_node (cond_command); ++ last_command_exit_value = result = execute_cond_node (cond_command); + return (result); + } + #endif /* COND_COMMAND */ +@@ -2808,7 +2814,7 @@ + push_scope (VC_BLTNENV, temporary_env); + if (subshell == 0) + add_unwind_protect (pop_scope, "1"); +- temporary_env = (HASH_TABLE *)NULL; ++ temporary_env = (HASH_TABLE *)NULL; + } + } + +@@ -2881,7 +2887,7 @@ + + debug_trap = TRAP_STRING(DEBUG_TRAP); + error_trap = TRAP_STRING(ERROR_TRAP); +- ++ + /* The order of the unwind protects for debug_trap and error_trap is + important here! unwind-protect commands are run in reverse order + of registration. If this causes problems, take out the xfree +@@ -2960,7 +2966,7 @@ + bitmap = new_fd_bitmap (FD_BITMAP_DEFAULT_SIZE); + begin_unwind_frame ("execute-shell-function"); + add_unwind_protect (dispose_fd_bitmap, (char *)bitmap); +- ++ + ret = execute_function (var, words, 0, bitmap, 0, 0); + + dispose_fd_bitmap (bitmap); +@@ -3115,7 +3121,7 @@ + if (saved_undo_list) + dispose_redirects (saved_undo_list); + redirection_undo_list = exec_redirection_undo_list; +- saved_undo_list = exec_redirection_undo_list = (REDIRECT *)NULL; ++ saved_undo_list = exec_redirection_undo_list = (REDIRECT *)NULL; + discard_unwind_frame ("saved_redirects"); + } + +@@ -3363,7 +3369,7 @@ + /* If there is more text on the line, then it is an argument for the + interpreter. */ + +- if (STRINGCHAR(i)) ++ if (STRINGCHAR(i)) + { + for (start = i; STRINGCHAR(i); i++) + ; +@@ -3458,7 +3464,7 @@ + len = -1; \ + } \ + while (0) +- ++ + /* Call execve (), handling interpreting shell scripts, and handling + exec failures. */ + int --- bash-2.05b-2.orig/debian/patches/bash205b-005.dpatch +++ bash-2.05b-2/debian/patches/bash205b-005.dpatch @@ -0,0 +1,105 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash-2.05b upstream patch 005 + + BASH PATCH REPORT + ================= + +Bash-Release: 2.05b +Patch-ID: bash205b-005 + +Bug-Reported-by: Jim Meyering +Bug-Reference-ID: <87bs6v8iib.fsf@pixie.eng.ascend.com> +Bug-Reference-URL: http://mail.gnu.org/archive/html/bug-bash/2002-09/msg00047.html + +Bug-Description: + +When in a locale with multibyte characters, the readline display updater +will occasionally cause a segmentation fault when attempting to compute +the length of the first multibyte character on the line. + +Patch: + +*** ../bash-2.05b/lib/readline/mbutil.c Tue Jun 4 11:54:29 2002 +--- lib/readline/mbutil.c Mon Aug 5 11:20:39 2002 +*************** +*** 206,210 **** + { + /* shorted to compose multibyte char */ +! memset (ps, 0, sizeof(mbstate_t)); + return -2; + } +--- 206,211 ---- + { + /* shorted to compose multibyte char */ +! if (ps) +! memset (ps, 0, sizeof(mbstate_t)); + return -2; + } +*************** +*** 213,217 **** + /* invalid to compose multibyte char */ + /* initialize the conversion state */ +! memset (ps, 0, sizeof(mbstate_t)); + return -1; + } +--- 214,219 ---- + /* invalid to compose multibyte char */ + /* initialize the conversion state */ +! if (ps) +! memset (ps, 0, sizeof(mbstate_t)); + return -1; + } +*************** +*** 226,232 **** + int + _rl_compare_chars (buf1, pos1, ps1, buf2, pos2, ps2) +! char *buf1, *buf2; +! mbstate_t *ps1, *ps2; +! int pos1, pos2; + { + int i, w1, w2; +--- 228,237 ---- + int + _rl_compare_chars (buf1, pos1, ps1, buf2, pos2, ps2) +! char *buf1; +! int pos1; +! mbstate_t *ps1; +! char *buf2; +! int pos2; +! mbstate_t *ps2; + { + int i, w1, w2; +*************** +*** 277,282 **** + /* clear the state of the byte sequence, because + in this case effect of mbstate is undefined */ +! memset (ps, 0, sizeof (mbstate_t)); + } + else + pos += tmp; +--- 282,290 ---- + /* clear the state of the byte sequence, because + in this case effect of mbstate is undefined */ +! if (ps) +! memset (ps, 0, sizeof (mbstate_t)); + } ++ else if (tmp == 0) ++ pos++; + else + pos += tmp; --- bash-2.05b-2.orig/debian/patches/bash205b-006.dpatch +++ bash-2.05b-2/debian/patches/bash205b-006.dpatch @@ -0,0 +1,127 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash-2.05b upstream patch 005 + + BASH PATCH REPORT + ================= + +Bash-Release: 2.05b +Patch-ID: bash205b-006 + +Bug-Reported-by: clowenst@ucsd.edu +Bug-Reference-ID: <156388ec.0212021151.51a48df1@posting.google.com> +Bug-Reference-URL: + +Bug-Description: + +When running in a locale with multibyte characters, the readline display +updater will use carriage returns when drawing the line, overwriting any +partial output already on the screen and not terminated by a newline. + +Patch: + +*** ../bash-2.05b/lib/readline/display.c Tue Jun 4 10:54:47 2002 +--- lib/readline/display.c Fri Sep 13 16:22:57 2002 +*************** +*** 71,75 **** + + #if defined (HANDLE_MULTIBYTE) +! static int _rl_col_width PARAMS((char *, int, int)); + static int *_rl_wrapped_line; + #else +--- 71,75 ---- + + #if defined (HANDLE_MULTIBYTE) +! static int _rl_col_width PARAMS((const char *, int, int)); + static int *_rl_wrapped_line; + #else +*************** +*** 1349,1355 **** + _rl_output_some_chars (nfd + lendiff, temp - lendiff); + #if 0 +- _rl_last_c_pos += _rl_col_width (nfd+lendiff, 0, temp-lendiff) - col_lendiff; +- #else + _rl_last_c_pos += _rl_col_width (nfd+lendiff, 0, temp-col_lendiff); + #endif + } +--- 1349,1355 ---- + _rl_output_some_chars (nfd + lendiff, temp - lendiff); + #if 0 + _rl_last_c_pos += _rl_col_width (nfd+lendiff, 0, temp-col_lendiff); ++ #else ++ _rl_last_c_pos += _rl_col_width (nfd+lendiff, 0, temp-lendiff); + #endif + } +*************** +*** 1511,1516 **** + /* If we have multibyte characters, NEW is indexed by the buffer point in + a multibyte string, but _rl_last_c_pos is the display position. In +! this case, NEW's display position is not obvious. */ +! if ((MB_CUR_MAX == 1 || rl_byte_oriented ) && _rl_last_c_pos == new) return; + #else + if (_rl_last_c_pos == new) return; +--- 1511,1523 ---- + /* If we have multibyte characters, NEW is indexed by the buffer point in + a multibyte string, but _rl_last_c_pos is the display position. In +! this case, NEW's display position is not obvious and must be +! calculated. */ +! if (MB_CUR_MAX == 1 || rl_byte_oriented) +! { +! if (_rl_last_c_pos == new) +! return; +! } +! else if (_rl_last_c_pos == _rl_col_width (data, 0, new)) +! return; + #else + if (_rl_last_c_pos == new) return; +*************** +*** 1595,1603 **** + { + if (MB_CUR_MAX > 1 && rl_byte_oriented == 0) +! { +! tputs (_rl_term_cr, 1, _rl_output_character_function); +! for (i = 0; i < new; i++) +! putc (data[i], rl_outstream); +! } + else + _rl_backspace (_rl_last_c_pos - new); +--- 1602,1606 ---- + { + if (MB_CUR_MAX > 1 && rl_byte_oriented == 0) +! _rl_backspace (_rl_last_c_pos - _rl_col_width (data, 0, new)); + else + _rl_backspace (_rl_last_c_pos - new); +*************** +*** 2118,2122 **** + static int + _rl_col_width (str, start, end) +! char *str; + int start, end; + { +--- 2121,2125 ---- + static int + _rl_col_width (str, start, end) +! const char *str; + int start, end; + { +*************** +*** 2194,2196 **** + } + #endif /* HANDLE_MULTIBYTE */ +- +--- 2197,2198 ---- --- bash-2.05b-2.orig/debian/patches/bash205b-007.dpatch +++ bash-2.05b-2/debian/patches/bash205b-007.dpatch @@ -0,0 +1,68 @@ +#! /bin/sh -e + +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) patch $pdir -f --no-backup-if-mismatch -p0 < $0;; + -unpatch) patch $pdir -f --no-backup-if-mismatch -R -p0 < $0;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: bash-2.05b upstream patch 005 + + BASH PATCH REPORT + ================= + +Bash-Release: 2.05b +Patch-ID: bash205b-007 + +Bug-Reported-by: dman@dman.ddts.net +Bug-Reference-ID: <15893.26358.129589.503364@gargle.gargle.HOWL> +Bug-Reference-URL: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=175127 + +Bug-Description: + +Using the vi editing mode's case-changing commands in a locale with +multibyte characters will cause garbage characters to be inserted into +the editing buffer. + +Patch: + +*** ../bash-2.05b/lib/readline/vi_mode.c Thu May 23 13:27:58 2002 +--- lib/readline/vi_mode.c Tue Feb 4 15:11:07 2003 +*************** +*** 681,685 **** + { + wchar_t wc; +! char mb[MB_LEN_MAX]; + mbstate_t ps; + +--- 681,686 ---- + { + wchar_t wc; +! char mb[MB_LEN_MAX+1]; +! int mblen; + mbstate_t ps; + +*************** +*** 704,708 **** + if (wc) + { +! wctomb (mb, wc); + rl_begin_undo_group (); + rl_delete (1, 0); +--- 705,711 ---- + if (wc) + { +! mblen = wctomb (mb, wc); +! if (mblen >= 0) +! mb[mblen] = '\0'; + rl_begin_undo_group (); + rl_delete (1, 0); --- bash-2.05b-2.orig/debian/patches/random.dpatch +++ bash-2.05b-2/debian/patches/random.dpatch @@ -0,0 +1,111 @@ +#! /bin/sh -e + +dir=. +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" + dir=$3 +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) + patch $pdir -f --no-backup-if-mismatch -p1 < $0 + echo '2.05' > $dir/_distribution + echo '0' > $dir/_patchlevel + cd $dir && autoconf + rm -f $dir/_distribution $dir/_patchlevel + ;; + -unpatch) + patch $pdir -f --no-backup-if-mismatch -R -p1 < $0 + rm -f $dir/configure $dir/_distribution $dir/_patchlevel + ;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# DP: use the system random functions + +--- bash-2.05-old/configure.in Mon Nov 6 14:21:35 2000 ++++ bash-2.05/configure.in Sun Nov 26 20:42:06 2000 +@@ -479,6 +479,9 @@ + siginterrupt memmove ttyname gethostbyname getservbyname \ + inet_aton strpbrk setvbuf pathconf) + ++dnl checks for random functions ++AC_CHECK_FUNCS(random srandom) ++ + dnl checks for locale functions + AC_CHECK_HEADERS(libintl.h) + AC_CHECK_FUNCS(gettext textdomain bindtextdomain) +--- bash-2.05-old/config.h.in Mon Oct 2 15:44:22 2000 ++++ bash-2.05/config.h.in Sun Nov 26 21:51:18 2000 +@@ -462,6 +462,9 @@ + /* Define if you have the putenv function. */ + #undef HAVE_PUTENV + ++/* Define if you have the random function. */ ++#undef HAVE_RANDOM ++ + /* Define if you have the readlink function. */ + #undef HAVE_READLINK + +@@ -524,6 +527,9 @@ + + /* Define if you have the strsignal function or macro. */ + #undef HAVE_STRSIGNAL ++ ++/* Define if you have the srandom function. */ ++#undef HAVE_SRANDOM + + /* Define if you have the sysconf function. */ + #undef HAVE_SYSCONF +--- bash-2.05b/variables.c~ 2002-06-25 15:43:33.000000000 +0200 ++++ bash-2.05b/variables.c 2004-04-14 08:26:10.000000000 +0200 +@@ -1038,17 +1038,26 @@ + /* The random number seed. You can change this by setting RANDOM. */ + static unsigned long rseed = 1; + static int last_random_value; ++#if defined(HAVE_RANDOM) ++static char random_state[32]; ++#endif + +-/* A linear congruential random number generator based on the example +- one in the ANSI C standard. This one isn't very good, but a more +- complicated one is overkill. */ ++/* Use the random number genrator provided by the standard C library, ++ else use a linear congruential random number generator based on the ++ ANSI C standard. This one isn't very good (the values are alternately ++ odd and even, for example), but a more complicated one is overkill. */ + + /* Returns a pseudo-random number between 0 and 32767. */ + static int + brand () + { ++#if defined(HAVE_RANDOM) ++ rseed = (unsigned int) ((32767.0 * random()) / (RAND_MAX + 1.0)); ++ return rseed; ++#else + rseed = rseed * 1103515245 + 12345; + return ((unsigned int)((rseed >> 16) & 32767)); /* was % 32768 */ ++#endif + } + + /* Set the random number generator seed to SEED. */ +@@ -1056,8 +1065,14 @@ + sbrand (seed) + unsigned long seed; + { ++#if defined(HAVE_SRANDOM) ++ initstate((unsigned int) seed, random_state, sizeof(random_state)); ++ setstate(random_state); ++ srandom((unsigned int) seed); ++#else + rseed = seed; + last_random_value = 0; ++#endif + } + + static SHELL_VAR * --- bash-2.05b-2.orig/debian/patches/bashdb.dpatch +++ bash-2.05b-2/debian/patches/bashdb.dpatch @@ -0,0 +1,44033 @@ +#! /bin/sh -e + +dir=. +if [ $# -eq 3 -a "$2" = '-d' ]; then + pdir="-d $3" + dir=$3 +elif [ $# -ne 1 ]; then + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +fi +case "$1" in + -patch) + patch $pdir -f --no-backup-if-mismatch -p1 < $0 + rm -f $dir/aclocal.m4 + rm -f $dir/configure + rm -f $dir/doc/bashref.info + rm -f $dir/examples/bashdb/bashdb.el + rm -f $dir/lib/readline/doc/Makefile + rm -f $dir/parser-built + rm -f $dir/pathnames.h + rm -f $dir/y.tab.[ch] + (cd $dir && aclocal -I . -I debugger) + (cd $dir && autoconf) + for i in config.guess config.sub elisp-comp install-sh \ + missing mkinstalldirs + do + cp -p /usr/share/automake-1.7/$i $dir/debugger/ + done + for i in mdate-sh texinfo.tex; do + cp -p /usr/share/automake-1.7/$i $dir/debugger/doc/ + done + (cd $dir/debugger && aclocal -I ..) + cp -p $dir/debugger/elisp-comp $dir/debugger/emacs/. + (cd $dir/debugger && automake) + (cd $dir/debugger && autoconf) + ;; + -unpatch) + patch $pdir -f --no-backup-if-mismatch -R -p1 < $0 + ;; + *) + echo >&2 "`basename $0`: script expects -patch|-unpatch as argument" + exit 1 +esac +exit 0 + +# how to construct this file: +# - unpack the upstream bash source +# - unpack the bashdb source +# - apply bash the seven bash upstream patches +# - diff -ur --unidirectional-new-file \ + --exclude CVS --exclude=.cvsignore --exclude=/db/ \ +# bash-2.05b bashdb + +# DP: bashdb changes + +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/ChangeLog cvs/ChangeLog +--- bash-2.05b/ChangeLog 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/ChangeLog 2003-06-01 03:18:28.000000000 +0200 +@@ -0,0 +1,4559 @@ ++2003-05-27 07:32 rockyb ++ ++ * lib/readline/histfile.c: Bug: uninitalized initial bp1 ++ ++2003-05-27 07:22 rockyb ++ ++ * lib/readline/histfile.c: This time, for sure. ++ ++2003-05-27 07:19 rockyb ++ ++ * lib/readline/histfile.c: Wasn't updating bp1. There's still a bug ++ in truncating the history though. ++ ++2003-05-24 16:45 rockyb ++ ++ * tests/: history.right, run-histexpand, run-history: Update ++ history regression tests for new code. ++ ++2003-05-24 16:29 rockyb ++ ++ * configure.in: Get ready for yet another release... ++ ++2003-05-24 16:22 rockyb ++ ++ * builtins/history.def: Small typo. ++ ++2003-05-24 16:13 rockyb ++ ++ * ChangeLog, bashhist.c, builtins/history.def, doc/bashref.texi: ++ Add timestamp to history file. Environment variable HISTTIMEFORMAT ++ controls how to display. ++ ++2003-05-24 16:11 rockyb ++ ++ * lib/readline/examples/.cvsignore: Useful file. ++ ++2003-05-24 16:10 rockyb ++ ++ * lib/readline/: histfile.c, history.c, history.h, ++ examples/Makefile, examples/histexamp.c: Add timestamps to history ++ files. Makefile: add targets for histexamp ++ ++2003-05-21 03:45 rockyb ++ ++ * doc/bashdb.texi: Clarification of "stepi" for scripting ++ languages. ++ ++2003-05-21 03:43 rockyb ++ ++ * configure.in: [no log message] ++ ++2003-05-20 14:02 rockyb ++ ++ * debugger/dbg-init.inc: More OS-independent (output independent) ++ test if a terminal is attached. ++ ++2003-05-07 19:12 rockyb ++ ++ * execute_cmd.c: Cater to older C compilers. ++ ++2003-04-29 20:33 rockyb ++ ++ * debugger/NEWS: [no log message] ++ ++2003-04-29 20:22 rockyb ++ ++ * debugger/: bashdb, dbg-cmds.inc, dbg-fns.inc, dbg-init.inc: Bug: ++ Was using source filename rather than canonicalized (expanded) ++ filename. ++ ++2003-04-29 20:21 rockyb ++ ++ * debugger/tests/: debug.right, list.right, sig.right: Basename ++ made to work like basename. ++ ++2003-04-07 23:29 rockyb ++ ++ * MANIFEST, configure.in, debugger/ddd/ddd-3.3.1-bashdb.diff: We ++ now have this all in CVS as well as a file release. ++ ++2003-04-07 23:28 rockyb ++ ++ * doc/: bashdb.texi, version.texi.in: [no log message] ++ ++2003-04-07 23:27 rockyb ++ ++ * debugger/tests/sig.right: Line number always changing on me. ++ ++2003-04-07 23:26 rockyb ++ ++ * debugger/tests/: misc.cmd, misc.right: Give invalid command to ++ get "Invalid command" back. ++ ++2003-04-07 23:24 rockyb ++ ++ * debugger/dbg-cmds.inc: Use gdb undefined command message. ++ ++2003-04-02 02:45 rockyb ++ ++ * doc/: bashdb.texi, bashdb.texi: [no log message] ++ ++2003-04-02 02:34 rockyb ++ ++ * doc/bashdb.texi: Lots of little typos. Apparently @xxx{} doesn't ++ work as expected. ++ ++2003-04-01 10:12 rockyb ++ ++ * doc/bashdb.texi: Add section on using bashdb from DDD. ++ ++ Breakpoints now start at 1. ++ ++ Update to reflect changed breakpoint output. ++ ++ Document "set annotate" ++ ++ Minor corrections. ++ ++2003-03-31 10:09 rockyb ++ ++ * debugger/: NEWS, dbg-brk.inc, tests/action.right, ++ tests/brkpt1.right, tests/brkpt2.right, tests/display.right, ++ tests/tbreak.right, tests/watch1.right: Numbers in ++ breakpoint/watchpoint/action numbers are printed in a "%-3d" rather ++ than (I think the nicer) %3d" to match gdb's output. This helps ddd ++ parse breakpoints. ++ ++2003-03-30 22:04 rockyb ++ ++ * debugger/: NEWS, dbg-brk.inc, tests/action.right, ++ tests/brkpt1.cmd, tests/brkpt1.right, tests/brkpt2.cmd, ++ tests/brkpt2.right, tests/brkpt3.cmd, tests/brkpt3.right, ++ tests/bugIFS.right, tests/debug.right, tests/display.right, ++ tests/finish.right, tests/parm.right, tests/tbreak.right, ++ tests/watch2.right: Breakpoints are now are origin 1 to match gdb. ++ It also helps ddd. ++ ++2003-03-27 00:34 rockyb ++ ++ * debugger/dbg-pre.inc.in: Ooops - changed dbg-pre.inc but should ++ have changed this one instead. ++ ++2003-03-26 09:09 rockyb ++ ++ * debugger/tests/: list.cmd, list.right: Add test of filename ++ canonicalization ++ ++2003-03-26 09:05 rockyb ++ ++ * debugger/: dbg-file.inc, dbg-io.inc: Add routine to resolve based ++ on dir and then expand a filename. Hopefully will be more ++ permissive in specified and work more like gdb. ++ ++2003-03-26 09:01 rockyb ++ ++ * debugger/dbg-brk.inc: Expand filename only the breakpoint exists. ++ ++2003-03-25 23:30 rockyb ++ ++ * debugger/dbg-list.inc: When set annotate=1 expand full filenames ++ when listing source lines. ++ ++2003-03-25 23:29 rockyb ++ ++ * debugger/dbg-file.inc: Bug in testing absolute file name. ++ ++2003-03-25 18:46 rockyb ++ ++ * debugger/: dbg-brk.inc, dbg-file.inc, dbg-list.inc, ++ tests/brkpt3.cmd, tests/brkpt3.right, tests/list.right: ++ Canonicalize filenames in list and breakpoint commands and update ++ regression tests to make sure we don't slip back. ++ ++2003-03-21 19:16 rockyb ++ ++ * debugger/: dbg-file.inc, dbg-set.inc: Small stuff. ++ ++2003-03-21 18:54 rockyb ++ ++ * debugger/: bashdb, dbg-brk.inc, dbg-file.inc, dbg-fns.inc, ++ dbg-help.inc, dbg-init.inc, dbg-set.inc, tests/brkpt2.right, ++ tests/debug.right, tests/finish.right, tests/parm.right, ++ tests/sig.right, tests/tbreak.right: Add "show dir" and directory ++ search in filename lookup (such as for breakpoints or line ++ listing). ++ ++2003-03-21 12:59 rockyb ++ ++ * debugger/: dbg-sig.inc, tests/sig.right: Signal received message ++ is now gdb's. This simplifies ddd handling. ++ ++2003-03-21 12:56 rockyb ++ ++ * debugger/TODO: Current sitiation. ++ ++2003-03-02 13:52 rockyb ++ ++ * debugger/ddd/ddd-3.3.1-bashdb.diff: Spelled g++ option ++ incorrectly! ++ ++2003-02-27 05:14 rockyb ++ ++ * debugger/: dbg-brk.inc, dbg-file.inc, dbg-fns.inc, dbg-help.inc, ++ dbg-io.inc: _filenames -> _Dbg_filenames. Array now has expanded ++ name Add routine to print filename in form appropriate to ++ annotation and basename only ++ ++2003-02-27 02:33 rockyb ++ ++ * debugger/NEWS, doc/bashdb.texi, tests/dbg-fns, ++ tests/dbg-is-fn.right, tests/dbg-is-fn.tests: _bashdb_ -> _Dbg_ ++ ++2003-02-27 02:27 rockyb ++ ++ * debugger/: bashdb, dbg-brk.inc, dbg-cmds.inc, dbg-file.inc, ++ dbg-fns.inc, dbg-help.inc, dbg-hist.inc, dbg-init.inc, dbg-io.inc, ++ dbg-list.inc, dbg-main.inc, dbg-pre.inc.in, dbg-set-d-vars.inc, ++ dbg-set.inc, dbg-sig-ret.inc, dbg-sig.inc, dbg-stack.inc, ++ tests/bugIFS.cmd, tests/misc.cmd, tests/misc.right, ++ tests/sig.right: All _bashdb_ variables renamed _Dbg_. Might help ++ for debuggers in other POSIX shells. ++ ++2003-02-26 23:34 rockyb ++ ++ * debugger/: NEWS, dbg-help.inc, dbg-init.inc, dbg-set.inc, ++ tests/misc.cmd, tests/misc.right: Add "set annotate." More better ++ set/show regression tests. ++ ++2003-02-25 11:22 rockyb ++ ++ * debugger/: dbg-brk.inc, tests/brkpt2.right, tests/watch1.right: A ++ little closer to gdb. Mesage reads breakpoint already hit 1 time ++ or breakpoint already hit n times ++ ++2003-02-25 10:55 rockyb ++ ++ * debugger/: dbg-brk.inc, tests/action.right, tests/brkpt1.right, ++ tests/brkpt2.right, tests/brkpt3.right, tests/bugIFS.right, ++ tests/debug.right, tests/display.right, tests/finish.right, ++ tests/parm.right, tests/tbreak.right, tests/watch1.right, ++ tests/watch2.right: Make breakpoints look more like gdb. It helps ++ ddd parsing. ++ ++2003-02-24 22:21 rockyb ++ ++ * doc/bashdb.texi: Clarify use of -L option. ++ ++2003-02-23 15:40 rockyb ++ ++ * debugger/dbg-file.inc: Add tilde to list of translation ++ characters. ++ ++2003-02-19 18:45 rockyb ++ ++ * debugger/dbg-brk.inc, debugger/dbg-cmds.inc, debugger/dbg-io.inc, ++ debugger/dbg-sig.inc, doc/bashdb.texi: Some declare -a's turned ++ into typeset -a on the hope (from David Korn) that this will be ++ legal ksh. ++ ++2003-02-17 19:23 rockyb ++ ++ * debugger/.cvsignore: bashdb- -> dbg- ++ ++2003-02-17 19:23 rockyb ++ ++ * MANIFEST, configure.in, debugger/Makefile.am, ++ debugger/bashdb-sig-ret.inc, debugger/dbg-sig-ret.inc: ++ Miscellaneous small changes for new version and bashdb- -> dbg- ++ changes. ++ ++2003-02-17 19:09 rockyb ++ ++ * tests/dbg-is-fn.tests: bashdb- -> dbg- change ++ ++2003-02-17 19:02 rockyb ++ ++ * debugger/: Makefile.am, TODO, bashdb, bashdb-brk.inc, ++ bashdb-cmds.inc, bashdb-file.inc, bashdb-fns.inc, bashdb-help.inc, ++ bashdb-hist.inc, bashdb-init.inc, bashdb-io.inc, bashdb-list.inc, ++ bashdb-main.inc, bashdb-set-d-vars.inc, bashdb-set.inc, ++ bashdb-sig.inc, bashdb-stack.inc, dbg-brk.inc, dbg-cmds.inc, ++ dbg-file.inc, dbg-fns.inc, dbg-help.inc, dbg-hist.inc, ++ dbg-init.inc, dbg-io.inc, dbg-list.inc, dbg-main.inc, dbg-pre.inc, ++ dbg-set-d-vars.inc, dbg-set.inc, dbg-sig.inc, dbg-stack.inc, ++ tests/sig.right, bashdb-pre.inc.in, dbg-pre.inc, dbg-pre.inc.in: ++ bashdb -> dbg to facilitate porting to other POSIX shells. ++ ++2003-02-17 17:01 rockyb ++ ++ * MANIFEST, execute_cmd.c, subst.c, trap.c, trap.h, ++ debugger/bashdb-fns.inc, support/mksignames.c, ++ tests/dbg-support.right, tests/dbg-support.tests, ++ tests/errors.right: Remove SUBEXIT trap. ++ ++2003-02-17 17:00 rockyb ++ ++ * doc/bashdb.texi: Decscribe "print" command a little better. Also, ++ it's 2003. ++ ++2003-02-17 16:18 rockyb ++ ++ * builtins/source.def: Small doc fix to mention optional ARGUMENTS. ++ Thanks to Chet Ramey for pointing this out. ++ ++2003-02-17 13:15 rockyb ++ ++ * debugger/: bashdb-brk.inc, tests/brkpt3.cmd, tests/brkpt3.right, ++ tests/tbreak.cmd, tests/tbreak.right: I think I've gotten over the ++ hurdle of passing information back. Removing breakpoints in a ++ subshell and having that persist now works. ++ ++2003-02-17 07:34 rockyb ++ ++ * debugger/bashdb-brk.inc: Perhaps some cleanup. ++ ++2003-02-17 06:51 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-file.inc, ++ bashdb-sig.inc, tests/brkpt3.cmd, tests/brkpt3.right, ++ tests/sig.right, tests/subshell3.right: State saving bugs: save ++ state of more breakpoint/action variables, step/next variables. ++ ++2003-02-16 20:08 rockyb ++ ++ * debugger/: NEWS, bashdb-brk.inc, bashdb-file.inc, bashdb-sig.inc: ++ Probably just beating around problems. Perhaps closer. Perhaps not. ++ ++2003-02-16 20:05 rockyb ++ ++ * debugger/tests/: Makefile.am, brkpt3.cmd, brkpt3.right, ++ brkpt3.tests, run-brkpt, run-brkpt2, run-subshell, sig.right, ++ subshell1.cmd, subshell1.right, subshell2.cmd, subshell2.right, ++ subshell3.cmd, subshell3.right, subshell4.cmd, subshell4.right, ++ subshell4.tests: run-brkpt and run-subshell does all 1.. tests. ++ Add another subshell and brkpt test. ++ ++2003-02-16 12:18 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-file.inc, ++ bashdb-fns.inc, bashdb-sig.inc: More state recorded in subshell ++ journal: Enable/disable fntrace options ++ ++ having breakpoints persist inside a subshell now works. (Removing ++ has problems though, as does saving/restoring the step counter) ++ ++2003-02-16 12:14 rockyb ++ ++ * MANIFEST, debugger/tests/subshell2.right: multi->subshell ++ ++2003-02-16 12:13 rockyb ++ ++ * debugger/tests/: Makefile.am, multi.sh, multi1.cmd, multi1.right, ++ multi1.tests, multi2.cmd, multi2.right, multi2.tests, multi3.cmd, ++ multi3.right, multi3.tests, run-subshell, sig.right, subshell.sh, ++ subshell1.cmd, subshell1.right, subshell1.tests, subshell2.cmd, ++ subshell2.tests, subshell3.cmd, subshell3.right, subshell3.tests, ++ subshell4.cmd, subshell4.right: multi -> subshell ++ ++2003-02-16 08:34 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-file.inc, bashdb-fns.inc, ++ bashdb-sig.inc, tests/multi4.right: Back to the future. Back to a ++ single journal file. Can't get SUBEXIT working and this seems to be ++ more reliable (albeit more work/slower). ++ ++2003-02-14 19:04 rockyb ++ ++ * MANIFEST, debugger/tests/sig.right: Small changes. ++ ++2003-02-14 18:49 rockyb ++ ++ * debugger/tests/: multi4.cmd, multi4.right, multi4.tests: Add ++ another multi test for quit 0 n. ++ ++2003-02-14 18:47 rockyb ++ ++ * debugger/: NEWS, TODO, bashdb-sig.inc, tests/multi3.cmd, ++ tests/run-multi: Bug in quit 0 n. ++ ++2003-02-14 18:38 rockyb ++ ++ * debugger/CHANGES: Note BASH_SUBSHELL. Some typographical mistakes ++ corrected. ++ ++2003-02-14 18:33 rockyb ++ ++ * execute_cmd.c, subst.c, trap.c, trap.h, builtins/caller.def, ++ doc/bashref.texi, support/mksignames.c, tests/dbg-support.right, ++ tests/dbg-support.tests: Bug: caller() now works like it is ++ supposed to. It had mismatched fn/lineno/source levels. It also now ++ returns null if no caller. ++ ++ Add SUBEXIT trap to perform action when a subshell is exited. ++ ++2003-02-14 18:31 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2003-02-14 18:25 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-file.inc, bashdb-fns.inc, ++ bashdb-set-d-vars.inc, bashdb-sig.inc: Better journal mechanism: ++ Now have journal per subshell. (Journal is used to pass information ++ between subshell and parent) ++ ++ There were escaping bugs in putting history entries into the ++ journal and retrieving them. ++ ++ Debugger "Print" now works like quote does. We were having ++ substituion problems before, e.g. "This can't happen" (the single ++ quote in the string). ++ ++ bashdb-fns.inc: New fn _bashdb_esq_dq. Save last_xxx state even if ++ we don't enter the debugger. ++ ++2003-02-14 18:16 rockyb ++ ++ * debugger/tests/: action.cmd, brkpt1.cmd, brkpt1.right, ++ brkpt2.cmd, brkpt2.right, display.cmd, finish.cmd, finish.right, ++ list.cmd, misc.cmd, misc.right, multi3.cmd, multi3.right, parm.cmd, ++ run-multi, search.cmd, search.right, sig.cmd, sig.right, skip.cmd, ++ skip.right, tbreak.cmd, tbreak.right, watch1.cmd, watch1.right: ++ Print now expands * so add quotes around. Some small ++ regularizations. multi: quit 0 2 now works so note that.a ++ ++2003-02-13 15:22 rockyb ++ ++ * print_cmd.c: "\n" snuck in on last change in ++ print_case_command_head. ++ ++2003-02-12 19:55 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-file.inc, ++ bashdb-fns.inc, bashdb-help.inc, bashdb-hist.inc, bashdb-init.inc, ++ bashdb-io.inc, bashdb-list.inc, bashdb-main.inc, bashdb-set.inc, ++ bashdb-sig.inc, bashdb-stack.inc: Try to make more ksh-93 ++ compatible. ++ ++ declare -> typeset (except when using declare -p or declare -a). ++ source -> . ++ ++2003-02-12 01:47 rockyb ++ ++ * debugger/bashdb: Small changes to make a little more consistent ++ with ksh93 ++ ++2002-12-31 23:25 rockyb ++ ++ * debugger/bashdb-pre.inc.in: Comment more. ++ ++2002-12-31 23:17 rockyb ++ ++ * debugger/bashdb-file.inc: Add % to list of translated file ++ characters. ++ ++2002-12-31 23:12 rockyb ++ ++ * debugger/bashdb-fns.inc: [no log message] ++ ++2002-12-31 11:57 rockyb ++ ++ * debugger/: bashdb-fns.inc, bashdb-list.inc, bashdb-sig.inc: Make ++ all bashdb functions begin _bashdb - were were missing a couple of ++ routines. ++ ++2002-12-30 23:23 rockyb ++ ++ * debugger/NEWS: [no log message] ++ ++2002-12-30 23:20 rockyb ++ ++ * debugger/bashdb-file.inc: Add + to list of translated file2var ++ characters. ++ ++2002-12-29 05:08 masata-y ++ ++ * doc/bashdb.texi: Added my name again. ++ ++2002-12-25 10:30 rockyb ++ ++ * debugger/bashdb-cmds.inc: More state saved. ++ ++2002-12-25 10:20 rockyb ++ ++ * debugger/NEWS: Recent stuff. ++ ++2002-12-25 10:18 rockyb ++ ++ * debugger/: TODO, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-file.inc, bashdb-hist.inc, bashdb-set.inc, bashdb-sig.inc, ++ tests/misc.right, tests/multi3.cmd, tests/multi3.right, ++ tests/run-multi, tests/sig.right: Start the tedious task of having ++ a mechanism for setting variables that in a subshell so that its ++ value appears in its parent shell. ++ ++2002-12-25 00:01 rockyb ++ ++ * debugger/TODO: [no log message] ++ ++2002-12-24 23:14 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc: Fix setting ++ breakpoints over restart. Needed to save the _brkpt_filevar and ++ _action_filevar arrays as well as _brkpt_count and _brkpt_cond ++ arrays. ++ ++ If we are in a subshell, we need to save BASHB_RESTART_FILE so it ++ will get set in sup-shells. ++ ++2002-12-24 22:13 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-12-24 22:13 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-file.inc, bashdb-sig.inc: ++ Bug/Feature fix: Restart now quits all subshells before issuing its ++ restart (exec) command. ++ ++2002-12-20 21:31 rockyb ++ ++ * debugger/bashdb-fns.inc: Small typo. ++ ++2002-12-20 13:13 rockyb ++ ++ * debugger/tests/Makefile.am: Add run-bugIFS. ++ ++2002-12-20 12:22 rockyb ++ ++ * debugger/TODO: Update. ++ ++2002-12-20 12:17 rockyb ++ ++ * debugger/NEWS: - Bug: if IFS set character in filename: the ++ filename would be truncated. Handle IFS save/restore more ++ pervasively by saving on DEBUG handler entry. ++ ++2002-12-20 12:13 rockyb ++ ++ * MANIFEST: Add bug-IFS tests. ++ ++2002-12-20 12:12 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-fns.inc: Save and restore IFS ++ on debug signal handler. It means we can remove custom ++ save/restores elsewhere in the code. ++ ++2002-12-20 12:11 rockyb ++ ++ * debugger/tests/: bugIFS, bugIFS.cmd, bugIFS.right, bugIFS.sh, ++ bugIFS.tests, run-bugIFS: Regression test of recent bug when not ++ setting IFS on signal entry. ++ ++2002-12-20 12:09 rockyb ++ ++ * debugger/tests/sig.right: Update for IFS bug fix. ++ ++2002-12-10 04:45 rockyb ++ ++ * debugger/bashdb-brk.inc: More bash-oriented way to strip off a ++ suffix. ++ ++2002-12-10 01:45 rockyb ++ ++ * execute_cmd.c, subst.c: Let BASH_SUBSHELL catch more cases. In ++ particular, command substitution. ++ ++2002-12-10 01:44 rockyb ++ ++ * doc/bashdb.texi: Even more on quit/return. ++ ++2002-12-10 01:28 rockyb ++ ++ * doc/bashdb.texi: More on how quit and return are similar and ++ different. ++ ++2002-12-10 01:06 rockyb ++ ++ * doc/bashref.texi: Spelling mistake. ++ ++2002-12-10 01:05 rockyb ++ ++ * debugger/emacs/gud.el: Eval expression command is "x" not "pe" ++ ++2002-12-10 01:04 rockyb ++ ++ * debugger/TODO: [no log message] ++ ++2002-12-09 08:16 rockyb ++ ++ * MANIFEST, execute_cmd.c, externs.h, print_cmd.c, ++ debugger/CHANGES: "set -x" now show "for", "case" and "select" ++ headers ( "for x in", "case c in", "select s") ++ ++2002-12-09 08:12 rockyb ++ ++ * tests/: run-set-x, set-x.right, set-x.tests: Add "set -x" tests ++ ++2002-12-09 07:13 rockyb ++ ++ * debugger/tests/version: not needed? ++ ++2002-12-08 01:26 rockyb ++ ++ * doc/bashdb.texi: Note DDD and GNU Emacs frontents in sample ++ session. ++ ++2002-12-08 01:21 rockyb ++ ++ * doc/bashdb.texi: Another example mistake. "pe" doesn't exist any ++ more. We use "x". ++ ++2002-12-08 01:17 rockyb ++ ++ * doc/bashdb.texi: Output "echo 1" in wrong place. ++ ++2002-12-08 00:58 rockyb ++ ++ * htdocs/index.html: Add screenshot for ddd. ++ ++2002-12-08 00:54 rockyb ++ ++ * htdocs/bashdb-ddd.png: Show an example that is less obviously ++ buggy! ++ ++2002-12-08 00:38 rockyb ++ ++ * htdocs/bashdb-ddd.png: Screenshot in DDD. ++ ++2002-12-07 23:21 rockyb ++ ++ * doc/bashdb.texi: Correct sample session a bit. It really needs ++ more work ++ ++2002-12-07 22:44 rockyb ++ ++ * ChangeLog: [no log message] ++ ++2002-12-07 22:43 rockyb ++ ++ * MANIFEST: Add Changelog. ++ ++2002-12-07 22:42 rockyb ++ ++ * configure.in, debugger/ChangeLog, debugger/NEWS: Get ready for ++ yet another release. ++ ++2002-12-07 22:11 rockyb ++ ++ * doc/bashdb.texi: Small typos ++ ++2002-12-07 21:59 rockyb ++ ++ * doc/bashdb.texi: Revise "restart" command to reflect current ++ state. Document BASHDB_RESTART_FILE and other BASHDB_ and BASH_ ++ variables. ++ ++2002-12-07 16:54 rockyb ++ ++ * debugger/ChangeLog: [no log message] ++ ++2002-12-07 16:49 rockyb ++ ++ * debugger/bashdb-main.inc: Make sure we can read the restart file. ++ ++2002-12-07 16:48 rockyb ++ ++ * debugger/bashdb-cmds.inc: Forgot to unset flag which says to ++ reread the file. ++ ++2002-12-07 16:43 rockyb ++ ++ * debugger/: ChangeLog, bashdb, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-main.inc, tests/brkpt2.right, tests/debug.right, ++ tests/finish.right, tests/parm.right, tests/sig.right: Preserve ++ breakpoints, watchpoints, debugger set variables on a restart. ++ ++2002-12-07 16:41 rockyb ++ ++ * debugger/Makefile.am: Remove compile of bashdb.elc. It's going ++ away anyway. ++ ++2002-12-07 10:24 rockyb ++ ++ * debugger/NEWS: As appropriate. ++ ++2002-12-07 10:23 rockyb ++ ++ * execute_cmd.c, externs.h, print_cmd.c, debugger/ChangeLog: print ++ select head - "select s in" ++ ++2002-12-07 08:35 rockyb ++ ++ * debugger/bashdb-io.inc: [no log message] ++ ++2002-12-07 08:18 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-12-07 08:16 rockyb ++ ++ * debugger/bashdb-io.inc: bashdb-io.inc (_bashdb_readlin): - When ++ reading large files which takes a long time, print out status every ++ 1000 lines. ++ ++2002-12-07 08:02 rockyb ++ ++ * MANIFEST, parse.y, debugger/NEWS, tests/heredoc.right, ++ tests/heredoc.tests, tests/run-heredoc: Bug: (parse.y) line number ++ was not updated properly when in a here documented with an unquoted ++ delimter and containing backslash newline. ++ ++ Update and modernize tests and make executable. ++ ++2002-12-06 22:57 rockyb ++ ++ * tests/dbg-multicmd.right: A tad more correct than the preview ++ "right" check. ++ ++2002-12-06 22:51 rockyb ++ ++ * bashline.c, subst.c, lib/readline/bind.c, ++ lib/readline/readline.c: Add Official bash 2.05b patches ++ ++2002-12-06 22:49 rockyb ++ ++ * print_cmd.c: Add routines to print for and case headers "for i ++ in" or "case c in". ++ ++2002-12-06 22:48 rockyb ++ ++ * externs.h: Add definitions for print_for_head and ++ print_case_head. ++ ++2002-12-06 22:48 rockyb ++ ++ * execute_cmd.c: set BASH_COMMAND to for_head (for i in xxx) and ++ case_head (case c in). ++ ++2002-12-06 22:47 rockyb ++ ++ * debugger/bashdb-sig-ret.inc: Not used, but it if it were, this ++ would be right. Need to save/restore IFS around a "set -" ++ ++2002-12-06 22:44 rockyb ++ ++ * debugger/bashdb-list.inc: Typo. ++ ++2002-12-06 22:43 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-12-06 11:27 rockyb ++ ++ * debugger/emacs/: gud.el, gud.el.diff: A typo. Also modified DB ++ when I meant to modify bashdb. ++ ++2002-12-03 19:46 rockyb ++ ++ * debugger/bashdb-help.inc: Typo. ++ ++2002-12-03 19:34 rockyb ++ ++ * debugger/bashdb-fns.inc: Remove previous variable test. Not ++ needed? ++ ++2002-12-03 19:24 rockyb ++ ++ * debugger/bashdb-fns.inc: Bug in set was set of IFS messing up ++ "set" parsing. This time for sure!? ++ ++2002-12-03 18:36 rockyb ++ ++ * debugger/bashdb-fns.inc: More precise function comment. ++ ++2002-12-03 18:34 rockyb ++ ++ * debugger/bashdb-fns.inc: Bug: if _bashdb_old_set_opts was null. ++ Test now. ++ ++2002-12-02 09:42 rockyb ++ ++ * doc/bashdb.texi: Document set showcommand ++ ++2002-12-02 08:20 rockyb ++ ++ * debugger/TODO: Update as appropriate. ++ ++2002-12-02 06:58 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-12-02 06:48 rockyb ++ ++ * MANIFEST: MANIFEST: add debugger multi tests version.texi.in: new ++ version of manuals. ++ ++2002-12-02 06:42 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-file.inc, bashdb-init.inc: ++ bashdb-file.inc: source persistent only if it exists. ++ bashdb-init.inc/bashdb-cmds.inc: read persistent, write persistent. ++ ++2002-12-02 06:41 rockyb ++ ++ * debugger/bashdb-help.inc: Update doc for "x/examine" ++ ++2002-12-02 06:41 rockyb ++ ++ * debugger/bashdb-sig.inc: Hack (but works more often) cleanup of ++ persistent file. ++ ++2002-12-02 06:40 rockyb ++ ++ * debugger/: CHANGES, NEWS, TODO: As appropriate. ++ ++2002-12-02 06:39 rockyb ++ ++ * debugger/tests/: Makefile.am, run-misc, run-sig: Makefile.am: add ++ run-multi run-sig: line number changes run-misc: bad testing of ++ return codes. Wasn't doing second test. ++ ++2002-12-02 05:50 rockyb ++ ++ * doc/bashdb.texi: Document quit command. Has been put in command ++ reference and cross-referenced with return. Subshell parenthesis ++ noted. ++ ++ Some typos corrected. Help listing updated. ++ ++2002-12-02 05:15 rockyb ++ ++ * doc/.cvsignore: Add bashdb.tmp ++ ++2002-12-02 05:14 rockyb ++ ++ * doc/: bashref.texi, version.texi.in: Update for this version. ++ ++2002-12-02 04:55 rockyb ++ ++ * doc/version.texi.in: Typo ++ ++2002-12-02 04:54 rockyb ++ ++ * configure.in: Time for another release. ++ ++2002-12-02 04:53 rockyb ++ ++ * doc/bashref.texi: Update CHANGES - include BASH_SUBSHELL for ++ example. ++ ++2002-12-02 04:32 rockyb ++ ++ * debugger/CHANGES: Update as appropriate. ++ ++2002-12-01 16:26 rockyb ++ ++ * debugger/tests/: multi3.cmd, multi3.tests: Regression test for ++ multi-statement lines and quit multiple levels ++ ++2002-12-01 16:24 rockyb ++ ++ * debugger/tests/: misc.right, multi.sh, multi1.cmd, multi1.right, ++ multi1.tests, multi2.cmd, multi2.right, multi2.tests, multi3.right, ++ run-multi, sig.right: Regression tests for multi-statement lines ++ and new quit command. ++ ++2002-11-30 19:34 rockyb ++ ++ * debugger/tests/misc.right: Was checking buggy history command. ++ ++2002-11-30 19:13 rockyb ++ ++ * debugger/bashdb-hist.inc: Typo.. "H count" was broken. ++ ++2002-11-30 01:16 masata-y ++ ++ * debugger/bashdb: Fix a typo. ++ ++2002-11-26 10:47 rockyb ++ ++ * debugger/: bashdb-init.inc, bashdb-set-d-vars.inc: More elegant ++ (and I wonder if not more robust) way to set $1 via set. Use "set ++ -- xx yy" rather than "set xx yy" ++ ++2002-11-26 07:24 rockyb ++ ++ * MANIFEST: Add README, add examples/bashdb/bashdb.el ++ ++2002-11-26 07:23 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-11-26 07:20 rockyb ++ ++ * examples/bashdb/: README, bashdb.el: Remove bashdb.el as it is ++ likely to cause confusion. Set expectations for the code in this ++ directory. ++ ++2002-11-26 07:09 rockyb ++ ++ * debugger/emacs/gud.el.diff: Set default minibuf-mod-map in bashdb ++ read call from Masatake YAMATO and add link to ++ http://bashdb.sourceforge.net ++ ++2002-11-26 07:04 rockyb ++ ++ * debugger/emacs/gud.el: Add HTTP reference. ++ ++2002-11-26 06:34 rockyb ++ ++ * debugger/emacs/gud.el: Change from Masatake YAMATO in running ++ bash initially. Uses gud-minibuffer-local-map now. ++ ++2002-11-23 00:17 rockyb ++ ++ * debugger/bashdb-sig.inc: Don't wipe temp files unless subshell is ++ 0. Likewise don't enter command shell if subshell is not 0. Don't ++ know that this will make much difference, but I guess it's the ++ right thing to do. ++ ++2002-11-23 00:15 rockyb ++ ++ * debugger/tests/misc.right: Output changed because of "set ++ showcommand" ++ ++2002-11-22 22:02 rockyb ++ ++ * debugger/bashdb-help.inc: Document help "set showcommand" ++ ++2002-11-22 21:50 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-11-22 21:50 rockyb ++ ++ * debugger/bashdb-init.inc: Bug: if no script we were running "set" ++ to dump environment. ++ ++2002-11-22 20:57 rockyb ++ ++ * execute_cmd.c, variables.c, debugger/bashdb-cmds.inc, ++ debugger/bashdb-list.inc, debugger/bashdb-set.inc, ++ debugger/emacs/bashdb.el, debugger/emacs/gud.el, ++ debugger/tests/sig.right, doc/bashref.texi: We now keep track of ++ the subshell level: dynamic veriable BASH_SUBSHELL and the debugger ++ prints these as ()'s inside of the prompt. ++ ++2002-11-21 00:44 rockyb ++ ++ * debugger/ddd/ddd-3.3.1-bashdb.diff: Display bash variables of the ++ form ${foo} or even ${foo[bar]##/*} now. ++ ++2002-11-20 12:00 rockyb ++ ++ * debugger/TODO: Current state of things and known bugs. ++ ++2002-11-20 10:11 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-11-20 10:09 rockyb ++ ++ * support/Makefile.in: Remove man2html objects and actually all ++ objects. ++ ++2002-11-20 10:00 rockyb ++ ++ * support/.cvsignore: Add man2html binary ++ ++2002-11-20 09:59 rockyb ++ ++ * support/man2html: This is an OS-specific *binary* ++ ++2002-11-20 09:55 rockyb ++ ++ * debugger/bashdb-init.inc: "bash --debugger script args" was not ++ setting args properly. ++ ++2002-11-19 23:42 rockyb ++ ++ * doc/Makefile.in: Failiers in this directory aren't important. ++ ++2002-11-19 23:31 rockyb ++ ++ * debugger/tests/sig.right: Output chagned yet again. ++ ++2002-11-19 23:03 rockyb ++ ++ * debugger/tests/debug.right: Change to have it work with different ++ paths. Use basename setting. ++ ++2002-11-19 23:02 rockyb ++ ++ * debugger/bashdb-cmds.inc: Changes to have regression tests work ++ with different paths. ++ ++2002-11-19 22:27 rockyb ++ ++ * debugger/tests/run-sig: Typo. ++ ++2002-11-19 22:22 rockyb ++ ++ * debugger/tests/sig.right: [no log message] ++ ++2002-11-19 22:20 rockyb ++ ++ * debugger/tests/: brkpt2.right, debug.right, finish.right, ++ parm.right: Shortend bashdb. ++ ++2002-11-19 22:17 rockyb ++ ++ * debugger/bashdb: We have a better expand_fn now if we need it. ++ ++2002-11-19 21:44 rockyb ++ ++ * debugger/Makefile.am: This is a pain. Get date and autoregen ++ right on bashdb-pre.inc. ++ ++2002-11-19 21:41 rockyb ++ ++ * doc/Makefile.in: No really! ++ ++2002-11-19 21:39 rockyb ++ ++ * doc/Makefile.in: This time, for sure! ++ ++2002-11-19 21:37 rockyb ++ ++ * debugger/tests/misc.right: Made more like GDB output for ddd. ++ ++2002-11-19 21:26 rockyb ++ ++ * MANIFEST, configure.in, debugger/TODO, doc/Makefile.in: Mostly ++ hacks to make a distribution work smoothly. MANIFEST: add DVI's. ++ This reduce TeX dependency, okay? TODO: document bugs configure.in ++ touch files so that we have right the first time. ++ ++2002-11-19 20:27 rockyb ++ ++ * configure.in: Remove touch in output. Is now handled in ++ Makefiles. ++ ++2002-11-19 20:26 rockyb ++ ++ * debugger/bashdb-cmds.inc: Variable name typo. When rerunning, ++ show full command now. Is messing up a bit too much. ++ ++2002-11-19 20:25 rockyb ++ ++ * debugger/bashdb-set.inc: Update help. ++ ++2002-11-19 20:21 rockyb ++ ++ * debugger/bashdb: Remove readonly attributes because when we run ++ as a subshell these want to get set again. May have to rethink how ++ to do. ++ ++2002-11-19 20:20 rockyb ++ ++ * doc/Makefile.in: Put touch of version.texi to avoid redoing ++ unnecessary work and force users to have TeX installed. ++ ++2002-11-19 20:20 rockyb ++ ++ * debugger/Makefile.am: Put touch inside build of bashd-pre.inc to ++ avoid redoing work. ++ ++2002-11-19 20:19 rockyb ++ ++ * debugger/bashdb-help.inc: Add set documentation. Put messages in ++ a form that DDD will understand without any hassle. ++ ++2002-11-19 16:03 rockyb ++ ++ * tests/dbg-multicmd.right: Added case tests. ++ ++2002-11-19 16:02 rockyb ++ ++ * MANIFEST: Add newest member of debugger bashdb-file.inc. ++ ++2002-11-19 16:01 rockyb ++ ++ * debugger/NEWS: Update as appropriate. Gee there are lots of ++ changes, mostly bug fixes. ++ ++2002-11-19 15:55 rockyb ++ ++ * tests/: trap.right, trap.tests: Remove SIGCHILD in trap list ++ since FreeBSD lists in another order. ++ ++2002-11-19 15:52 rockyb ++ ++ * tests/: dbg-multicmd.right, dbg-multicmd.tests, run-dbg-multicmd: ++ New regression test to check BASH_COMMAND. ++ ++2002-11-19 15:47 rockyb ++ ++ * execute_cmd.c: Preserve line number selector, i.e. the "foo, bar, ++ baz" of in "case foo, bar baz in ..." ++ ++2002-11-19 15:32 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-help.inc, bashdb-io.inc, bashdb-main.inc: Canonicalize ++ filevars via a file expand routine. Allows multpile to refer to a ++ given file. ++ ++ Add option to have files show only short (basename) names, useful ++ in testing ++ ++ Add set/show of these. ++ ++ Bug in not setting fntrace correctly on debugger exit. ++ ++ Add -c option on bashdb. Analogous to "bash -c string" ++ ++2002-11-19 15:28 rockyb ++ ++ * debugger/tests/: brkpt2.cmd, brkpt2.right, misc.right, parm.cmd, ++ parm.right: *.cmd: added a couple of info files to check that we ++ read in files only once. misc: added "set/show debugger and ++ basename" *.right: corresponding new output ++ ++2002-11-19 15:16 rockyb ++ ++ * debugger/: bashdb-help.inc, bashdb-set.inc: Add set/show options ++ to debug debugger and have filenames get listed as basename only. ++ ++2002-11-19 15:15 rockyb ++ ++ * debugger/: Makefile.am, bashdb-file.inc: Break out more routines ++ into bashdb-file.inc. Add _bashdb_expand_filename to canonicalize ++ file names. ++ ++2002-11-19 10:40 rockyb ++ ++ * debugger/tests/sig.right: Numbers keep changing... ++ ++2002-11-19 10:36 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, tests/brkpt2.right: ++ Fix bug in continue comand. Moved code from bashdb-brk.inc to ++ bashdb-cmds.inc ++ ++2002-11-19 03:07 rockyb ++ ++ * configure.in: Remove AC_CONFIG_HEADER of non-h files since it ++ munges first line. Boooooo! Setting time back now with touch. ++ ++2002-11-19 03:03 rockyb ++ ++ * debugger/tests/: brkpt2.cmd, brkpt2.right: Small change to make ++ regression test work on FreeBSD. ++ ++2002-11-19 02:58 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-init.inc, bashdb-list.inc, ++ bashdb-pre.inc.in: Use common routines now. Spelled "Bourne" ++ incorrectly in bashdb-{pre,init}.inc ++ ++2002-11-19 02:57 rockyb ++ ++ * debugger/tests/brkpt1.right: some output changed: linespecs ++ aren't necessarily lines. Erroneously "break" instead of "clear" as ++ a result of cut-and-paste. ++ ++2002-11-19 02:55 rockyb ++ ++ * debugger/bashdb-fns.inc: Add common routines _bashdb_check_line ++ and _bashdb_get_maxline. Reduces code bulk while making things ++ more reliable. _get_maxline now reads in the file if it hasn't been ++ read in yet. ++ ++2002-11-19 02:53 rockyb ++ ++ * debugger/bashdb-io.inc: Test to see if file exists before ++ reading. We could probably do better, but will need more code ++ reorganization for this. ++ ++2002-11-18 22:59 rockyb ++ ++ * configure.in: Take care by using AC_CONFIG_HEADERS not to update ++ a release file if it doesn't need updating. This allows the ++ maintainer/distributor to create a DVI and have it not get ++ recreated if that's not needed. In this way users are not forced ++ say to have TeX installed. ++ ++2002-11-18 22:15 rockyb ++ ++ * debugger/tests/brkpt1.right: Ditto. ++ ++2002-11-18 22:14 rockyb ++ ++ * debugger/bashdb-brk.inc: A little closer for FreeBSD... ++ ++2002-11-18 22:08 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-stack.inc, tests/brkpt1.right, ++ tests/brkpt2.right, tests/debug.right, tests/finish.right, ++ tests/parm.right, tests/sig.right: Even closer to getting FreeBSD ++ regression tests work okay. ++ ++2002-11-18 21:50 rockyb ++ ++ * debugger/: bashdb-list.inc, tests/list.right: More changes to ++ make FreeBSD regression tests work out -- more to come... ++ ++2002-11-18 21:44 rockyb ++ ++ * debugger/tests/: brkpt2.right, dbg-test1.sub, watch2.right: More ++ changes to make FreeBSD regression tests work. ++ ++2002-11-18 21:31 rockyb ++ ++ * debugger/: bashdb, bashdb-brk.inc, bashdb-init.inc, ++ bashdb-list.inc, bashdb-stack.inc, tests/action.right, ++ tests/action.tests, tests/brkpt1.right, tests/brkpt1.tests, ++ tests/brkpt2.right, tests/brkpt2.tests, tests/debug.right, ++ tests/debug.tests, tests/display.right, tests/display.tests, ++ tests/finish.right, tests/finish.tests, tests/list.right, ++ tests/list.tests, tests/misc.right, tests/misc.tests, ++ tests/parm.right, tests/parm.tests, tests/search.right, ++ tests/search.tests, tests/sig.right, tests/sig.tests, ++ tests/skip.right, tests/skip.tests, tests/tbreak.right, ++ tests/tbreak.tests, tests/watch1.right, tests/watch1.tests, ++ tests/watch2.right, tests/watch2.tests: Add -B option to list only ++ basename in debugger output. Is useful for FreeBSD which list files ++ as foo rather than ./foo. ++ ++2002-11-18 19:42 rockyb ++ ++ * MANIFEST, configure.in: Get ready for version 0.33. Derived files ++ have been shuffled around. I think though the overall organization ++ is a tad better. ++ ++2002-11-18 19:40 rockyb ++ ++ * debugger/: .cvsignore, bashdb-pre.inc.in: Release information ++ included via bashdb-pre.inc (via bashdb-pre.inc.in) ++ ++2002-11-18 19:38 rockyb ++ ++ * doc/: .cvsignore, Makefile.in, bashdb.texi, bashdb.texi.in, ++ bashref.texi, version.texi.in: Add release information ++ automatically via version.texi include file. ++ ++2002-11-18 19:36 rockyb ++ ++ * debugger/: Makefile.am, NEWS, bashdb, bashdb-init.inc, ++ bashdb-init.inc.in, bashdb.in, tests/brkpt2.right, ++ tests/debug.right, tests/finish.right, tests/misc-output.right, ++ tests/parm.right, tests/sig.right: Add release information via new ++ file bashdb-pre.in rather than modify bashdb-init.inc and bashdb ++ ++2002-11-18 09:32 rockyb ++ ++ * debugger/NEWS: typo ++ ++2002-11-18 09:30 rockyb ++ ++ * debugger/NEWS: Note configure script bug fix. ++ ++2002-11-18 08:59 rockyb ++ ++ * configure.in: Bug: remove --with-debugger-start-file since this ++ isn't a "package." Instead we set via variable DEBUGGER_START_FILE. ++ Thanks to c_thomps@ecolinux.no-ip.com for reporting the problem! ++ ++ Clean up a little, version and release numbers are now set in one ++ place. ++ ++2002-11-18 08:57 rockyb ++ ++ * debugger/configure.in: Clean up a little. ++ ++2002-11-17 20:36 rockyb ++ ++ * Makefile.in: Add doc to list of things in all made. I *think* ++ this is correct... ++ ++2002-11-17 20:26 rockyb ++ ++ * MANIFEST, configure.in, debugger/Makefile.am, doc/Makefile.in, ++ doc/bashref.texi: Changes to make .in targets get recreated. ++ ++2002-11-17 20:23 rockyb ++ ++ * doc/bashdb.texi: Is now derived from bashdb.texi.in ++ ++2002-11-17 20:22 rockyb ++ ++ * doc/bashdb.texi.in: More stuff was trunctated. Wonder if it has ++ to do with sourceforge move? ++ ++ Make references to version get updated automatically. ++ ++2002-11-17 19:44 rockyb ++ ++ * debugger/: NEWS, TODO: Update to reflect current situation. ++ ++2002-11-17 19:42 rockyb ++ ++ * debugger/tests/: brkpt2.right, debug.right, finish.right, ++ parm.right, sig.right: Line numbers inside signal handlers changed. ++ ++2002-11-17 19:41 rockyb ++ ++ * debugger/bashdb-sig.inc: Bug: don't set exit handler inside debug ++ trap handler. Seems to kick to be inhereted on subshells and then ++ we call the exit routine too often. ++ ++ exit_handler more like the other handlers. Understands ++ "stop/nostop, print/noprint stack/nostack". Does more of the same ++ things before entering cmdloop that debug_handler does via a more ++ common debugger enter routine. ++ ++ Saving _curline now done in common debbber enter routine ++ ++2002-11-17 19:38 rockyb ++ ++ * debugger/bashdb-cmds.inc: File got trunctated. Like ++ bashdb-help.inc -- Hmm. Something's fishy. ++ ++ Anyway, "debug" with no arguments now uses the last command rather ++ the last statement. This should be correct more often. ++ ++2002-11-17 19:34 rockyb ++ ++ * debugger/: bashdb-fns.inc, bashdb-main.inc: _curline is now set ++ by BASH_LINENO rather than passing as a parameter to the trap. ++ ++ Add this common thing inside common debugger enter routine. ++ ++2002-11-17 19:30 rockyb ++ ++ * debugger/bashdb-help.inc: info signals now shows EXIT in list. ++ ++ Some help text got truncated. ++ ++ Move CVS Id line to end to facilitate debugging. ++ ++2002-11-17 19:26 rockyb ++ ++ * debugger/Makefile.am: Add new .in files to distribution ++ ++2002-11-17 19:23 rockyb ++ ++ * debugger/bashdb: Derived now from bashdb.in ++ ++2002-11-17 19:23 rockyb ++ ++ * debugger/bashdb-init.inc: Derived now from bashdb-init.in ++ ++2002-11-17 19:22 rockyb ++ ++ * MANIFEST, configure.in, debugger/bashdb-init.inc.in, ++ debugger/bashdb.in: bashdb and bashdb-init.inc are now derived so ++ that we can automatically put in the release number. ++ ++2002-11-16 11:44 rockyb ++ ++ * htdocs/index.html: Add link to Bash reference manual which has ++ better error reporting and debugging support. ++ ++2002-11-16 11:18 rockyb ++ ++ * doc/bashref.texi: Go over debugger changes. Add debug RETURN, and ++ BASH_COMMAND for example. ++ ++2002-11-16 11:17 rockyb ++ ++ * debugger/CHANGES: Update as appropriate ++ ++2002-11-16 09:55 rockyb ++ ++ * execute_cmd.c, externs.h, print_cmd.c, trap.c, variables.c: Add ++ feature to save in BASH_COMMAND the command that is about to be ++ executed (provided it's not a "trap" statement). This is useful ++ *inside* a trap handler to know what was being run at the time of ++ the trap: the source code line isn't enough since there can be many ++ commands on one line. And in fact for an "eval" the command may ++ change from say "eval $foo" to "echo this is my life." ++ ++2002-11-16 09:50 rockyb ++ ++ * debugger/tests/: brkpt2.right, debug.right, finish.right, ++ parm.right, sig.right, skip.right: Recent changes to bashdb-sig.inc ++ changed call-stack line numbers. These tests notice the line ++ number change. ++ ++2002-11-16 09:49 rockyb ++ ++ * debugger/bashdb-list.inc: We now print the statement to be ++ executed if it was on the same line/file and the statement part is ++ different from the last time we ran this. ++ ++2002-11-16 09:48 rockyb ++ ++ * debugger/emacs/gud.el: Patched version for emacs 21.2 of gud.el ++ with bash support. ++ ++2002-11-16 09:47 rockyb ++ ++ * debugger/emacs/gud.el.diff: Patches that I hope will be on the ++ next version of Emacs (assuming the paperwork isn't lost, people ++ don't forget about the patch as a result of the time delay to do ++ the paperwork and so on.) ++ ++2002-11-16 09:45 rockyb ++ ++ * debugger/emacs/bashdb.el: Move from parent directory here. May ++ have also added provision for "finish" command. ++ ++2002-11-16 09:40 rockyb ++ ++ * debugger/: bashdb-main.inc, bashdb-sig.inc: Now save the command ++ that is to be executed next. ++ ++2002-11-16 09:38 rockyb ++ ++ * debugger/bashdb-fns.inc: Add common function ++ _bashdb_set_to_return_from_debugger to regularize the things that ++ need to be done on exit. There are more now, since we are saving ++ the last command executed on this line of the source file. ++ ++2002-11-16 09:36 rockyb ++ ++ * debugger/bashdb-init.inc: Need to skip more initially for "bashdb ++ -c" setup. ++ ++2002-11-16 09:25 rockyb ++ ++ * debugger/bashdb: Add -c option analogous to "bash -c". ++ ++2002-11-15 08:31 rockyb ++ ++ * doc/bashdb.texi: Update edition number and correct note x takes ++ an expression but repeats old one if not given. ++ ++2002-11-15 07:52 rockyb ++ ++ * doc/bashdb.texi: Document better how x command works. Mention ++ that DDD is available. Numerous small improvements. ++ ++2002-11-14 22:25 rockyb ++ ++ * debugger/ddd/ddd-3.3.1-bashdb.diff: This time, for sure! ++ ++2002-11-14 19:16 rockyb ++ ++ * MANIFEST, execute_cmd.c, tests/dbg-support.right: Bug: Somehow a ++ statement is messing up the simple cmd's Simple.line. So don't use ++ that in retrieving the line number. Would like to understand why ++ this is so, and if we could really simplify executing_line_number. ++ ++2002-11-14 15:56 rockyb ++ ++ * debugger/ddd/ddd-3.3.1-bashdb.diff: Final diff patch for ddd? ++ ++2002-11-14 15:45 rockyb ++ ++ * debugger/: NEWS, bashdb-cmds.inc, ddd/ddd-3.3.1-bashdb.diff, ++ tests/brkpt2.cmd, tests/brkpt2.right, tests/sig.right: More ++ improvements to "x" command. Use print as a fallback... ++ ++2002-11-14 09:25 rockyb ++ ++ * configure.in, debugger/bashdb-help.inc, debugger/tests/sig.right, ++ doc/bashdb.texi: configuure.in: get ready for version 0.32 ++ bashdb.texi: minor typo changes debugger/help/sig: Remove "Num" ++ from title as it is no longer there. ++ ++2002-11-14 09:08 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-11-14 09:03 rockyb ++ ++ * tests/run-all: Bug: was not using right shell to run tests. ++ ++2002-11-14 08:15 rockyb ++ ++ * debugger/: bashdb-help.inc, bashdb-sig.inc, tests/sig.right: A ++ tad more like gdb: no signal numbers listed. No longer trap on ++ SIGWINCH. ++ ++2002-11-14 01:09 rockyb ++ ++ * MANIFEST: bashdb.el listed in old location. ++ ++2002-11-14 01:08 rockyb ++ ++ * tests/: run-dbg-support, run-dbg-support2, run-type: Remove ++ "export" -- messes up occasionally. ++ ++2002-11-13 23:39 rockyb ++ ++ * debugger/: NEWS, bashdb-help.inc, bashdb-sig.inc, ++ tests/sig.right: Add help for "handle". Tidy generic signal handler ++ more and make more general (e.g unconditinally turn off set ++ tracing). ++ ++2002-11-13 23:22 rockyb ++ ++ * debugger/: bashdb-fns.inc, bashdb-sig.inc, tests/sig.right: Clean ++ up signal handling code a small amount by creating a common routine ++ to set some global variables. ++ ++2002-11-13 19:42 rockyb ++ ++ * MANIFEST, doc/bashdb.texi: MANIFEST: add new ddd-3.3.1 patch. ++ bashdb.texi: docment handle command ++ ++2002-11-13 19:33 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-help.inc, bashdb-sig.inc, ++ tests/sig.cmd, tests/sig.right, tests/sig.sh: Signal handling is ++ more like gdb now. No doubt there're room for improvement, but this ++ is probably a big improvement. ++ ++2002-11-13 19:32 rockyb ++ ++ * debugger/ddd/ddd-3.3.1-bashdb.diff: Patches to make ddd work. ++ ++2002-11-12 15:20 rockyb ++ ++ * debugger/: NEWS, TODO, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-init.inc, bashdb-sig.inc, tests/sig.right: Some work on ++ signal handling. More is there, more to do. So what else is new? ++ ++2002-11-12 12:22 rockyb ++ ++ * doc/bashdb.texi: Add "info break count". More examples. Embolden ++ typed example output. Remove some duplicate index entries. ++ ++2002-11-12 11:50 rockyb ++ ++ * debugger/tests/: brkpt1.cmd, brkpt1.right: Add error tests on ++ "info break". And add test for "info break n." ++ ++2002-11-12 09:30 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-11-12 09:29 rockyb ++ ++ * debugger/tests/: misc.right, sig.right: Recent changes in code ++ are reflected changed output (line numbers or help text) ++ ++2002-11-12 09:26 rockyb ++ ++ * doc/bashdb.texi: Add documentation for "condition" command. ++ ++2002-11-12 07:45 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc: Add ++ gdb's "condition" command. ++ ++2002-11-12 07:42 rockyb ++ ++ * debugger/bashdb-sig.inc: Oops --- typo in variable name. ++ ++2002-11-12 07:41 rockyb ++ ++ * debugger/bashdb-sig.inc: Add provision for debugging debugger. ++ ++2002-11-12 07:40 rockyb ++ ++ * debugger/bashdb-io.inc: Add more file translation characters for ++ ", [, and ] and reorder letters to make mnemonic. ++ ++2002-11-12 07:39 rockyb ++ ++ * debugger/tests/: brkpt1.cmd, brkpt1.right: Add regression tests ++ for new "condition" command. ++ ++2002-11-12 07:37 rockyb ++ ++ * tests/: dbg-support.right, dbg-support2.right: Got these two ++ output "right" files reversed, - sorta. ++ ++2002-11-11 23:56 rockyb ++ ++ * debugger/bashdb-sig.inc: Caught another routine that didn't start ++ _bashdb_ ++ ++2002-11-11 23:45 rockyb ++ ++ * debugger/TODO: Cross many things off list. ++ ++2002-11-11 23:45 rockyb ++ ++ * debugger/NEWS: Update as appropriate - major changes. ++ ++2002-11-11 23:44 rockyb ++ ++ * debugger/Makefile.am: Move bashdb.el into emacs directory. ++ ++2002-11-11 23:42 rockyb ++ ++ * MANIFEST: Add debugger emacs and ddd directories. ++ ++2002-11-11 23:41 rockyb ++ ++ * execute_cmd.c: Line number reporting improved. ++ ++ Fix bug in using trap line numbers inside traps. BASH_LINENO had ++ been wrong for the parts inside of a trap. ++ ++ Fix bug in case line numbers reporting line number of "esac" rather ++ than case clause. ++ ++ The organization of the how line numbers work should really be ++ rethought. Here we've done a lot of saving and restoring when that ++ shoudl probably be done elsewhere. ++ ++2002-11-11 23:39 rockyb ++ ++ * tests/: dbg-support.right, dbg-support2.tests: Would you believe ++ we were checking for erroneous values. Put in the trap reporting a ++ line number that (when working properly) *should* change. ++ ++2002-11-11 23:37 rockyb ++ ++ * tests/run-dbg-support2: Make it so we can run standalone. ++ ++2002-11-11 23:36 rockyb ++ ++ * debugger/tests/sig.right: Had had wrong line numbers. This time, ++ for sure! ++ ++2002-11-11 16:45 rockyb ++ ++ * debugger/bashdb.el: Moved inside emacs directory now. ++ ++2002-11-11 07:28 rockyb ++ ++ * execute_cmd.c: Simplifer and I think a tad more line-number ++ accurate. ++ ++2002-11-10 21:26 rockyb ++ ++ * execute_cmd.c, trap.c, debugger/NEWS: Believe it or not, the C ++ files have gotten a little bit *simpler*. We are closer towards ++ having line numbers recorded accurately in BASH_LINENO (although ++ things are not perfect yet). ++ ++2002-11-10 21:06 rockyb ++ ++ * MANIFEST: Make tests 755 so they can more easily be run ++ standalone. ++ ++2002-11-10 21:05 rockyb ++ ++ * debugger/tests/brkpt2.right: Would you believe that we had been ++ checking an erroneous down all along? ++ ++2002-11-10 21:04 rockyb ++ ++ * tests/: dbg-support.right, dbg-support.tests, run-dbg-support: ++ Would you believe that we had been checking wrong stuff all along? ++ ++2002-11-10 14:39 rockyb ++ ++ * debugger/: bashdb-help.inc, bashdb-stack.inc, bashdb.el, ++ tests/brkpt2.right, tests/misc.right: bashdb-help: Make more like ++ gdb - facilitates use in ddd. bashdb-stack.inc: Fix bug in ++ "stack_down". (There's another one though lurking) bashdb.el: add ++ key mapping for finish and other keymappings tests/misc: change in ++ "show" output tests/brkpt2: perhaps better although not quite ++ perfect. ++ ++2002-11-10 11:35 rockyb ++ ++ * debugger/bashdb-help.inc: Add help for finish command. Invalid ++ help command now reports what gdb does. ++ ++2002-11-09 23:54 rockyb ++ ++ * debugger/bashdb-help.inc: Missed a place that should have added ++ _bashdb. ++ ++2002-11-09 20:28 rockyb ++ ++ * debugger/bashdb-set.inc: Make show command work like gdb. ++ ++2002-11-09 20:16 rockyb ++ ++ * debugger/bashdb-io.inc: Missed a couple more "builtin printf"'s ++ ++2002-11-09 20:15 rockyb ++ ++ * debugger/bashdb-io.inc: Specify *builtin* printf. ++ ++2002-11-09 20:12 rockyb ++ ++ * debugger/tests/finish.cmd: gdb "finish" (= perl5db "return") ++ regression test. ++ ++2002-11-09 20:07 rockyb ++ ++ * tests/: run-trap, run-type: If there's an error we save in a file ++ specific for this test. Is more standalone now. ++ ++2002-11-09 10:07 rockyb ++ ++ * tests/shopt.right: Update for add set -o errtrace option ++ ++2002-11-09 07:38 rockyb ++ ++ * debugger/: Makefile.am, NEWS, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-hist.inc, bashdb-io.inc, bashdb-main.inc, bashdb-set.inc, ++ tests/misc.cmd, tests/misc.right, tests/sig.cmd, tests/sig.right, ++ tests/sig.sh: echo -> builtin echo Add "info signals" and "info ++ handler" Break out history into new file bashdb-hist.inc. Help ++ command change v -> M ++ ++2002-11-09 07:36 rockyb ++ ++ * doc/bashdb.texi: Revise info on signal handling and expand ++ history command section. ++ ++2002-11-09 07:34 rockyb ++ ++ * configure.in: Get redy for a new release. ++ ++2002-11-09 07:34 rockyb ++ ++ * MANIFEST: Update with current list. ++ ++2002-11-09 07:34 rockyb ++ ++ * trap.c: A hack to stop the core dump until I figure out why we ++ are trying to free a mangled string. ++ ++2002-11-08 21:29 rockyb ++ ++ * debugger/tests/sig.sh: Regression test for signal handling. ++ ++2002-11-08 21:24 rockyb ++ ++ * tests/: dbg-support2.right, dbg-support2.tests, run-dbg-support2: ++ Regression tests for skipping an instruction. ++ ++2002-11-08 21:22 rockyb ++ ++ * MANIFEST: Add run-support2 and +x tests/run-* ++ ++2002-11-08 21:08 rockyb ++ ++ * debugger/tests/sig.right: [no log message] ++ ++2002-11-08 21:08 rockyb ++ ++ * debugger/bashdb-sig.inc: Previous flie was horribly truncated. ++ ++2002-11-07 05:34 rockyb ++ ++ * debugger/tests/run-sig: Fix bug in reporting a pass when we fail. ++ ++2002-11-05 22:43 rockyb ++ ++ * MANIFEST, execute_cmd.c, flags.c, flags.h, trap.c, trap.h, ++ builtins/set.def, debugger/bashdb-sig-ret.inc, ++ debugger/tests/Makefile.am, debugger/tests/finish.right, ++ debugger/tests/finish.tests, debugger/tests/run-finish, ++ debugger/tests/run-sig, doc/bashdb.texi: bashdb.texi: Document ++ finish command. ++ ++ debugger/tests*, MANIFEST, Makefile.am: Add finish command test. ++ ++ *sig: add signal save/restor test ++ ++2002-11-02 15:17 rockyb ++ ++ * debugger/NEWS: Today's hacks. ++ ++2002-11-02 14:47 rockyb ++ ++ * debugger/tests/run-sig: Signal handler testing. ++ ++2002-11-02 14:46 rockyb ++ ++ * debugger/: Makefile.am, bashdb-list.inc, bashdb-sig-ret.inc, ++ bashdb-sig.inc, tests/Makefile.am, tests/sig.cmd, tests/sig.right, ++ tests/sig.sh, tests/sig.tests: Save/restore signals on entering ++ debugger. ++ ++2002-11-02 10:17 rockyb ++ ++ * trap.c: Clear trap RETURN on function calls. ++ ++2002-11-02 10:16 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-sig.inc: Workaround? for bash ++ bug introduced implementing "trap RETURN" ++ ++2002-11-02 09:17 rockyb ++ ++ * trap.c: Save and restore return code $? across run_return_trap ++ calls. ++ ++2002-11-02 09:15 rockyb ++ ++ * debugger/bashdb-sig.inc: [no log message] ++ ++2002-11-02 09:15 rockyb ++ ++ * debugger/bashdb-sig.inc: Small comment changes. ++ ++2002-11-02 09:12 rockyb ++ ++ * debugger/CHANGES: Reflects better the changes. ++ ++2002-11-02 09:03 rockyb ++ ++ * debugger/CHANGES: Update as appropriate. ++ ++2002-11-02 08:12 rockyb ++ ++ * builtins/return.def, debugger/NEWS, debugger/bashdb-cmds.inc, ++ debugger/bashdb-sig.inc, debugger/bashdb-stack.inc, ++ debugger/tests/brkpt2.right, debugger/tests/parm.right: Fix bug ++ where stack trace current line entry was sometimes wrong. No ++ longer mangle line number in stack changing commands. ++ ++ Printing source line on "RETURN" debugger call. ++ ++2002-11-01 07:14 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-io.inc, bashdb-list.inc, bashdb-main.inc, bashdb-sig.inc: ++ Add gdb's finish or perl5db's return. Line number on return ++ statement is funny and there are some bugs to make fully work. ++ ++ Now all debugger routines start _bashdb. ++ ++2002-11-01 07:11 rockyb ++ ++ * execute_cmd.c, trap.c, trap.h, support/mksignames.c, ++ tests/dbg-is-fn.right, tests/dbg-is-fn.tests: Add: trap fn ... ++ RETURN ++ ++ Line numbers when trap is called are a little off (for use in a ++ debugger), but this works. ++ ++2002-10-31 05:59 rockyb ++ ++ * debugger/TODO: [no log message] ++ ++2002-10-30 22:11 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-10-30 22:03 rockyb ++ ++ * configure.in, debugger/bashdb-cmds.inc, debugger/bashdb-help.inc, ++ debugger/tests/misc.cmd, debugger/tests/misc.right, ++ doc/bashdb.texi: Allow negative numbers in history commands. Get ++ ready for release 0.30. ++ ++2002-10-30 07:55 rockyb ++ ++ * doc/bashdb.texi: /tmp/fact.sh -> fact.sh in example. ++ ++2002-10-30 07:46 rockyb ++ ++ * doc/bashdb.texi: Index history commands. ++ ++2002-10-30 07:13 rockyb ++ ++ * debugger/TODO: Cross some things off the list. ++ ++2002-10-30 07:02 rockyb ++ ++ * doc/bashdb.texi: Add documentation on history command. Minor ++ typographical improvements. ++ ++2002-10-29 08:54 rockyb ++ ++ * debugger/: NEWS, bashdb-cmds.inc, bashdb-help.inc, ++ tests/misc.cmd, tests/misc.right: Make history command more robust ++ and allow for more features. How allows for !n:p. We now give an ++ error message if the history numbers are out of range. Separate ++ history parsing into a subroutine to tidy and facilitate future ++ features. ++ ++2002-10-27 00:05 rockyb ++ ++ * debugger/NEWS: Update as appropriate. ++ ++2002-10-27 00:02 rockyb ++ ++ * debugger/bashdb-set-d-vars.inc: Better way to set dollar ++ parameters. It handles more than $1..$9 and sets $# correctly too. ++ ++2002-10-26 23:17 rockyb ++ ++ * debugger/bashdb-init.inc: bug in bashdb --debugger some local's ++ must be declares. ++ ++2002-10-26 10:29 rockyb ++ ++ * doc/bashdb.texi: Add doc for -t option and "set editing" Revise ++ example a little for current output. ++ ++2002-10-26 10:27 rockyb ++ ++ * debugger/: NEWS, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-help.inc, bashdb-init.inc, bashdb-set.inc: Fix/change ++ history. Numbers now match prompt numbers. Can use !n as well as hi ++ n ++ ++ Add GDB's "set editing" command. ++ ++ _bashdb_help_set added to show set commands. Made more like GDB's ++ output. ++ ++2002-10-26 10:25 rockyb ++ ++ * debugger/tests/: misc.cmd, misc.right: Output changes as a result ++ of: ++ ++ - Fix/change history. Numbers now match prompt numbers. ++ - Add test of GDB's "set args" command. - Add test of GDB's ++ "set editing" command. ++ ++2002-10-25 08:34 rockyb ++ ++ * doc/bashdb.texi: Add info about set args, show args, controlling ++ tty, setting and showing the prompt. ++ ++2002-10-25 08:15 rockyb ++ ++ * debugger/: bashdb-cmds.inc, tests/misc.cmd, tests/misc.right: ++ Make H be like perl's. Fix history (hi) command. Don't but some ++ commands like comment and history in the history. Test. ++ ++2002-10-25 08:05 rockyb ++ ++ * debugger/: bashdb-cmds.inc, tests/misc.cmd, tests/misc.right: ++ Make H be more like perl's. Make hi work. ++ ++2002-10-25 07:07 rockyb ++ ++ * debugger/: bashdb-set.inc, tests/brkpt2.right, tests/debug.right, ++ tests/list.right, tests/misc.right, tests/parm.right, ++ tests/run-misc: Tidy up show output. Tests changed largely as a ++ result of size increase in bashdb from last change. run-misc: ++ wasn't giving an error is diff failed on first diff check. ++ ++2002-10-25 06:44 rockyb ++ ++ * debugger/NEWS: Latest news... ++ ++2002-10-25 06:41 rockyb ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-set.inc, ++ tests/misc-output.right, tests/misc.cmd, tests/misc.right, ++ tests/run-misc: Add "set/show prompt." Make prompt customizable. ++ bashdb: all -t to be a filename. ++ ++2002-10-25 06:39 rockyb ++ ++ * MANIFEST: One more file... ++ ++2002-10-24 09:33 rockyb ++ ++ * debugger/: bashdb-set.inc, bashdb-sig.inc: Comment about why ++ slightly more complex loop to set args. ++ ++2002-10-24 09:08 rockyb ++ ++ * debugger/: NEWS, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-help.inc, bashdb-init.inc, bashdb-io.inc, bashdb-set.inc, ++ tests/brkpt2.right, tests/debug.right, tests/misc.cmd, ++ tests/misc.right, tests/parm.right: - Fix bugs in debugger output ++ when debugged program redirects I/O ++ ++ - Add GDB's "set args" command ++ ++2002-10-23 09:04 rockyb ++ ++ * debugger/bashdb: Remove debug statement accidentally committed. ++ ++2002-10-23 08:51 rockyb ++ ++ * debugger/: bashdb, bashdb-init.inc, bashdb-io.inc, ++ tests/brkpt2.right, tests/debug.right, tests/parm.right: bashdb, ++ bashdb-init.inc: Add -t option to set tty bashdb-io.inc: bug in not ++ testing _bashdb_tty in _bashdb_printf tests: output changes bashdb ++ script has gotten larger. ++ ++2002-10-23 08:49 rockyb ++ ++ * debugger/TODO: Update as appropriate. ++ ++2002-10-23 08:26 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-init.inc, bashdb-io.inc, bashdb-set.inc, tests/misc.cmd, ++ tests/misc.right: Change variable $_tty to $_bashdb_tty ++ ++ All "read -e -p" now have input and output redirected. ++ ++ Implement tty command to set tty. ++ ++2002-10-22 13:36 masata-y ++ ++ * htdocs/: bashdb-break.png, bashdb-where.png, index.html: Added ++ new screenshot. ++ ++2002-10-22 06:25 rockyb ++ ++ * debugger/bashdb-cmds.inc: Fix bug in printing debugger prompt ++ when debugged program has redirected stderr. Thanks to Masatake ++ YAMATO. ++ ++2002-10-21 10:59 rockyb ++ ++ * debugger/bashdb-cmds.inc: Remove trailing blanks. ++ ++2002-10-21 10:48 rockyb ++ ++ * debugger/: bashdb-io.inc, bashdb-list.inc, bashdb-sig.inc: Just ++ format changes: remove blanks at the ends of some lines. ++ ++2002-10-21 10:26 rockyb ++ ++ * debugger/.cvsignore: Allow for different versions of autotools. ++ ++2002-10-21 10:25 rockyb ++ ++ * debugger/bashdb-help.inc: Bug in "help return" -- missing some ++ lines there. ++ ++2002-10-21 10:00 rockyb ++ ++ * .cvsignore: Handle embedded version numbers for automate. ++ ++2002-10-19 09:26 rockyb ++ ++ * debugger/bashdb-help.inc, doc/bashdb.texi: Add documentation for ++ new command return. bashdb.texi: had duplicate help screens ++ ++2002-10-19 09:08 rockyb ++ ++ * debugger/NEWS: Minor typographical changes. ++ ++2002-10-18 23:24 rockyb ++ ++ * debugger/: bashdb-cmds.inc, tests/debug.right: Make so regression ++ test is portable. A hack for now ++ ++2002-10-18 23:16 rockyb ++ ++ * debugger/: bashdb-cmds.inc, tests/debug.cmd, tests/debug.right: ++ Correct debug test. ++ ++2002-10-18 23:14 rockyb ++ ++ * debugger/tests/debug2.cmd: Need to have a separate file for debug ++ after nested one level. ++ ++2002-10-18 23:13 rockyb ++ ++ * MANIFEST: Need yet another command file to test "debug" nested ++ one level deep. ++ ++2002-10-18 22:40 rockyb ++ ++ * debugger/bashdb-set-d-vars.inc: To set dollar variables ($1, $2, ++ ... $?) before evaluation of a user expression. ++ ++2002-10-18 22:37 rockyb ++ ++ * MANIFEST: This time, for sure! ++ ++2002-10-18 22:34 rockyb ++ ++ * MANIFEST: More of the same. ++ ++2002-10-18 22:31 rockyb ++ ++ * MANIFEST: Add missing file. ++ ++2002-10-18 22:14 rockyb ++ ++ * MANIFEST, debugger/tests/Makefile.am, debugger/tests/run-watch2, ++ debugger/tests/watch2.cmd, debugger/tests/watch2.right, ++ debugger/tests/watch2.tests: Add more tests. ++ ++2002-10-18 21:41 rockyb ++ ++ * MANIFEST: Fix filename typo ++ ++2002-10-18 21:31 rockyb ++ ++ * MANIFEST, trap.c, debugger/Makefile.am, debugger/bashdb-brk.inc, ++ debugger/bashdb-cmds.inc, debugger/bashdb-fns.inc, ++ debugger/bashdb-io.inc, debugger/bashdb-sig.inc, ++ debugger/tests/debug.cmd, debugger/tests/debug.right, ++ debugger/tests/debug.sh, debugger/tests/debug.tests, ++ debugger/tests/parm.right, debugger/tests/run-debug, ++ doc/bashdb.texi, doc/bashref.texi: A better way of handling setting ++ dollar variables. ++ ++ Add debug test. ++ ++2002-10-18 08:44 rockyb ++ ++ * debugger/: bashdb-brk.inc, NEWS: Set $1, $2 before ++ watch-expression evaluation ++ ++2002-10-18 08:10 rockyb ++ ++ * debugger/: NEWS, bashdb-cmds.inc, bashdb-sig.inc: $? is saved and ++ preserved for use inside "eval" and "print" ++ ++2002-10-18 08:06 rockyb ++ ++ * debugger/tests/: parm.cmd, parm.sh: Test that $? is saved and ++ preserved in eval and print. ++ ++2002-10-18 06:38 rockyb ++ ++ * debugger/bashdb-set.inc: Correct show version to add more ++ sources. ++ ++2002-10-18 06:33 rockyb ++ ++ * MANIFEST, execute_cmd.c, trap.c, doc/bashdb.texi: execute_cmd.c: ++ small bug fixes to "return" statement bashdb.texi: document return ++ ++ MANIFEST: add skip.sh ++ ++2002-10-18 06:32 rockyb ++ ++ * debugger/: NEWS, bashdb-cmds.inc, bashdb-sig.inc: Fix bug in skip ++ *count*. ++ ++2002-10-18 06:31 rockyb ++ ++ * debugger/tests/: action.cmd, tbreak.cmd: Add CVS Id lines. ++ ++2002-10-18 06:30 rockyb ++ ++ * debugger/tests/: skip.cmd, skip.right, skip.sh, skip.tests: More ++ thorough tests of skip command. ++ ++2002-10-18 06:30 rockyb ++ ++ * debugger/tests/: parm.cmd, parm.right, parm.sh: Add return tests. ++ More detailed dollar tests. ++ ++2002-10-17 08:56 rockyb ++ ++ * debugger/: NEWS, TODO, bashdb-cmds.inc, bashdb-main.inc, ++ tests/brkpt2.right, tests/parm.right: bashdb-main.inc: Bug in ++ changing to _bashdb_libdir bashdb-cmds.inc: number of times nested ++ in debug listed via exported variable BASHDB_LEVEL ++ ++2002-10-17 08:11 rockyb ++ ++ * debugger/bashdb-cmds.inc: Remove debugging statments. ++ ++2002-10-17 00:10 rockyb ++ ++ * debugger/: NEWS, TODO, bashdb, bashdb-cmds.inc: Get "debug" ++ working much better: variables (e.g. $0) are substituted now. ++ "bash --debugger" invocations work since we no longer fork ++ _libdir renamed to _bashdb_libdir (in bashdb). We add "-L ++ _bashdb_libdir" when executing via bashdb script. ++ ++2002-10-16 23:12 rockyb ++ ++ * doc/Makefile.in: Correct make uninstall -- this time for sure! ++ ++2002-10-16 23:12 rockyb ++ ++ * Makefile.in: Correct make uninstall target - add debugger ++ directory. ++ ++2002-10-16 22:55 rockyb ++ ++ * doc/Makefile.in: Bug in uninstall targets. Was missing trailing ++ ;'s ++ ++2002-10-16 08:41 rockyb ++ ++ * debugger/bashdb-help.inc, doc/bashdb.texi: Document skip and ++ debug. ++ ++2002-10-16 08:40 rockyb ++ ++ * debugger/bashdb-cmds.inc: Make skip work like next/step. ++ ++2002-10-16 08:39 rockyb ++ ++ * configure.in: Get ready for another release. ++ ++2002-10-16 08:22 rockyb ++ ++ * debugger/NEWS: As appropriate. ++ ++2002-10-15 23:40 rockyb ++ ++ * debugger/tests/skip.right: Correct output from skip test. ++ ++2002-10-15 23:29 rockyb ++ ++ * debugger/bashdb-help.inc, doc/bashdb.texi, doc/bashref.texi: ++ Document skip command. ++ ++2002-10-15 22:57 rockyb ++ ++ * MANIFEST, execute_cmd.c, trap.c, trap.h, ++ debugger/bashdb-cmds.inc, debugger/bashdb-sig.inc, ++ debugger/tests/Makefile.am, debugger/tests/run-skip, ++ debugger/tests/skip.cmd, debugger/tests/skip.tests, ++ doc/bashref.texi: Add ability to skip execution of a command. ++ Debugger command: skip ++ ++2002-10-15 22:55 rockyb ++ ++ * builtins/caller.def: Documentation improvement. ++ ++2002-10-12 09:26 rockyb ++ ++ * doc/bashdb.texi: Update stack example and help display to match ++ current settings. ++ ++2002-10-11 23:18 rockyb ++ ++ * debugger/NEWS: Update to reflect current sitiation ++ ++2002-10-11 23:17 rockyb ++ ++ * debugger/: bashdb-list.inc, tests/search.cmd, tests/search.right: ++ Fix bug in search forward not resetting line to start. ++ ++ Search now lists only one line same as gdb and perl5db do. Update ++ tests as appropriate. ++ ++2002-10-11 08:36 rockyb ++ ++ * configure.in: Get ready for release 0.28 ++ ++2002-10-11 08:27 rockyb ++ ++ * doc/bashref.texi: Document BASH_ARGC and BASH_ARGV. Change ++ workign on BASH_LINENO and BASH_SOURCE a little. ++ ++2002-10-10 21:40 rockyb ++ ++ * MANIFEST: Add parm tests. ++ ++2002-10-10 21:39 rockyb ++ ++ * debugger/tests/parm.sh: To keep run-parm and parm.right company. ++ ++2002-10-10 21:21 rockyb ++ ++ * debugger/tests/: parm.cmd, parm.right: Add an action command. ++ Forgot to check in parm.right. ++ ++2002-10-10 21:08 rockyb ++ ++ * debugger/TODO: Update to reflect current situtation: parameters ++ now work. ++ ++2002-10-10 20:57 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-main.inc, bashdb-sig.inc, ++ tests/parm.cmd: Fix bug so that $1 preserve embedded blanks (or ++ more precisely, embedded IFS) that they might have had. ++ ++2002-10-10 08:47 rockyb ++ ++ * debugger/bashdb-cmds.inc: Bug in _bashdb_cmd_eval. Was not ++ erasing evalfile before starting if $1.. was empty, so we'd have ++ accumulated output. ++ ++2002-10-10 08:34 rockyb ++ ++ * .cvsignore: Add more good derived stuff (from autotools) ++ ++2002-10-10 08:34 rockyb ++ ++ * array.c, execute_cmd.c, shell.c, variables.c, variables.h, ++ builtins/common.c, builtins/common.h, builtins/evalfile.c, ++ tests/array.right: Add ability to show function and shell ++ parameters. Added BASH arrays BASH_ARGC and BASH_ARGV. ++ ++ array.c: handle more error contitions. array.right: change for new ++ array variables ++ ++2002-10-10 08:30 rockyb ++ ++ * debugger/.cvsignore: Add more good derived stuff (from autotools) ++ ++2002-10-10 08:26 rockyb ++ ++ * debugger/tests/: run-brkpt1, run-brkpt2, run-display, run-list, ++ run-misc, run-parm, run-search, run-tbreak, run-watch1: Make all ++ run* have these two properties: ++ ++ File checked against is now custom for that test (and is called ++ *test*.check) to facilitate running batch tests and moving that to ++ *test*.right or for just comparison. ++ ++ Globbing turned off on diff in case there is an discrepency and the ++ difference has a pattern character like *. ++ ++2002-10-10 08:20 rockyb ++ ++ * debugger/tests/run-action: File checked against is now custom for ++ that test (and is called *test*.check) to facilitate running batch ++ tests and moving that to *test*.right or for just comparison. ++ ++2002-10-10 08:17 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-io.inc, bashdb-sig.inc, ++ bashdb-stack.inc, tests/Makefile.am, tests/misc.cmd, ++ tests/parm.cmd, tests/parm.tests, tests/run-display: Add parameters ++ in stack trace. And can now refer to $1 $2 $3 in print and eval ++ commands. ++ ++2002-10-10 08:06 rockyb ++ ++ * debugger/tests/: brkpt2.right, misc.right: Updated for new stack ++ tracing format and $1, $2 evaluation. ++ ++2002-10-10 08:05 rockyb ++ ++ * debugger/tests/: run-brkpt1, run-brkpt2, run-list, run-misc, ++ run-parm: File checked against is now custom for that test (and is ++ called *test*.check) to facilitate running batch tests and moving ++ that to *test*.right or for just comparison. ++ ++2002-10-03 08:40 rockyb ++ ++ * debugger/: bashdb-list.inc, tests/search.cmd, tests/search.right: ++ Fix up search command to do the right thing more often. However ++ need to make another pass to match gdb semantics more perfectly. ++ Save last search pattern and can be repeated if no pattern given. ++ ++2002-10-03 01:12 rockyb ++ ++ * MANIFEST: Forgot another. ++ ++2002-10-03 01:10 rockyb ++ ++ * MANIFEST: Forgot one: run-action. ++ ++2002-10-03 01:08 rockyb ++ ++ * doc/bashdb.texi: Document search. ++ ++2002-10-03 01:06 rockyb ++ ++ * MANIFEST: Break out list/search routines. Add action and search ++ regression tests. ++ ++2002-09-24 13:30 rockyb ++ ++ * debugger/: Makefile.am, NEWS, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-help.inc, bashdb-io.inc, bashdb-list.inc, bashdb-main.inc, ++ bashdb-sig.inc, tests/Makefile.am, tests/action.cmd, ++ tests/action.right, tests/action.tests, tests/brkpt1.right, ++ tests/brkpt2.right, tests/run-action, tests/run-search, ++ tests/search.cmd, tests/search.right, tests/search.tests, ++ tests/tbreak.right, tests/watch1.right: Add "search" "reverse" ++ (forward/backward searching) and Perl's "a" (action) commands. ++ bashdb-list.inc split off for the debugger growth. ++ ++2002-09-24 12:29 rockyb ++ ++ * configure.in: In preparation for yet another release - when I get ++ back from hiking. ++ ++2002-09-24 01:26 rockyb ++ ++ * debugger/bashdb-help.inc: Add tb to help list. ++ ++2002-09-23 18:45 rockyb ++ ++ * debugger/NEWS: [no log message] ++ ++2002-09-23 18:19 rockyb ++ ++ * debugger/NEWS: [no log message] ++ ++2002-09-23 18:09 rockyb ++ ++ * debugger/NEWS: Make mroe readible. ++ ++2002-09-23 17:57 rockyb ++ ++ * cvscompile.sh: Spelling typo. ++ ++2002-09-23 17:52 rockyb ++ ++ * debugger/bashdb.el: Small typo ++ ++2002-09-23 17:47 rockyb ++ ++ * debugger/bashdb.el: Messed up on author and didn't comment ++ properly. ++ ++2002-09-23 17:41 rockyb ++ ++ * debugger/MANIFEST: This really isn't used. Autotools rules. (But ++ really we do a make dist in the parent directory.) ++ ++2002-09-23 17:39 rockyb ++ ++ * MANIFEST: Add display tests. ++ ++2002-09-23 17:06 rockyb ++ ++ * configure.in: Time for yet another release. (Really should figure ++ out a way to do this by changing another file...) ++ ++2002-09-23 09:30 rockyb ++ ++ * doc/bashdb.texi: Document display routines. Some unfinished ++ things had been finished Add Masatake YAMATO. ++ ++2002-09-23 09:09 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-io.inc, tests/display.right: ++ Change format of display output to identify display number and ++ expression. ++ ++2002-09-22 07:40 rockyb ++ ++ * debugger/: bashdb-help.inc, tests/Makefile.am, tests/display.cmd, ++ tests/display.right, tests/display.tests, tests/run-display: Add ++ display regression test. bashdb-help.inc: change lines to ++ linespecs in help. Add missing _bashdb_msg. ++ ++2002-09-21 22:46 rockyb ++ ++ * debugger/bashdb.el: Don't know how to spell my own name! ++ ++2002-09-21 22:35 rockyb ++ ++ * debugger/bashdb-help.inc: Add help for display. correct en/dis ++ help to watchpoint and display. ++ ++2002-09-21 22:07 rockyb ++ ++ * htdocs/index.html: Add Masatake YAMATO's screenshot link. ++ ++2002-09-21 22:01 rockyb ++ ++ * htdocs/index.html: Adding with g+w. ++ ++2002-09-21 22:00 rockyb ++ ++ * htdocs/index.html: Remove so I can recreate with g+w. ++ ++2002-09-21 21:41 rockyb ++ ++ * debugger/: NEWS, TODO, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-help.inc, tests/misc.right: Add gdb's auto display commands ++ (display, undisplay, enable/disable display) ++ ++2002-09-21 06:41 rockyb ++ ++ * debugger/bashdb-cmds.inc: debug command a tad better -- now ++ allows for command to get run. ++ ++2002-09-21 06:17 rockyb ++ ++ * debugger/bashdb-cmds.inc: Okay, now get out some of the bugs in ++ the "debug" command. More work is needed to make this more ++ seamless. (In particular need to skip over execution after debugged ++ execution happens). ++ ++2002-09-21 05:50 rockyb ++ ++ * debugger/: NEWS, bashdb-cmds.inc, bashdb-sig.inc, ++ tests/tbreak.cmd, tests/tbreak.right: Add provision to debug into a ++ script ("debug" command), and leave script without going back into ++ debugger command loop (BASHDB_QUIT_ON_QUIT environment variable). ++ ++2002-09-19 14:50 rockyb ++ ++ * builtins/caller.def: Add Perl-like caller function ++ ++2002-09-19 14:49 rockyb ++ ++ * MANIFEST, builtins/Makefile.in, debugger/NEWS, ++ tests/dbg-support.right, tests/dbg-support.tests: Add caller ++ builtin ++ ++2002-09-19 13:01 masata-y ++ ++ * debugger/: bashdb-help.inc, bashdb.el: bashdb-help.inc: fix ++ typo(tbreak, down), added [count] in up command's help message. ++ bashdb.el: added gud-tbreak. ++ ++2002-09-19 12:51 masata-y ++ ++ * debugger/bashdb.el: added simple comments about how to install ++ bashdb.el. provided bashdb. ++ ++2002-09-19 12:39 masata-y ++ ++ * autogen.sh, cvscompile.sh: autogen.sh: invoked bootstrap.sh in ++ the current directory(.). cvscompile.sh: invoked autogen.sh in the ++ current directory(.). ++ ++2002-09-18 23:04 rockyb ++ ++ * doc/bashdb.texi: Document tbreak, enable and disable. Remove ++ references to catchpoints. ++ ++2002-09-18 22:04 rockyb ++ ++ * cvscompile.sh: Basically changed name to autogen.sh. This is for ++ compatibility oniy. ++ ++2002-09-18 22:04 rockyb ++ ++ * MANIFEST: Add tbreak tests ++ ++2002-09-18 22:03 rockyb ++ ++ * debugger/tests/: Makefile.am, run-tbreak, tbreak.cmd, ++ tbreak.right, tbreak.tests: Regression test for temporary break: ++ tbreak ++ ++2002-09-18 21:37 rockyb ++ ++ * debugger/: NEWS, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-help.inc, bashdb-sig.inc, tests/brkpt1.right, ++ tests/brkpt2.right: Add temporary break. ++ ++2002-09-18 21:06 rockyb ++ ++ * autogen.sh, bootstrap.sh: Creates derived files when starting a ++ CVS project. autogen.sh is bootstrap with an invocation of ++ configure. ++ ++ Names suggested from autotools book of Gary V. Vaughan et. al by ++ way of information provided by Masatake YAMATO. ++ ++2002-09-18 21:02 rockyb ++ ++ * INSTALL: Update instructions as per information from Masatake ++ YAMATO. ++ ++2002-09-18 02:03 rockyb ++ ++ * debugger/bashdb-help.inc: Document "break" command change args ++ help linespec in online help where appropriate. ++ ++2002-09-17 09:01 rockyb ++ ++ * debugger/NEWS: typo. ++ ++2002-09-17 08:03 rockyb ++ ++ * debugger/NEWS: Last minute additions for version 0.25 - recent ++ doc changes. ++ ++2002-09-17 08:00 rockyb ++ ++ * configure.in: Get ready for version 0.25. ++ ++2002-09-17 08:00 rockyb ++ ++ * INSTALL: Modify for debugger-enabled version of bash. ++ ++2002-09-17 07:57 rockyb ++ ++ * README: Customize for the debug-enabled version of bash. Is more ++ helpful, I think. ++ ++2002-09-17 07:56 rockyb ++ ++ * debugger/INSTALL: Remove generic instructions and customize for ++ something specific for the debugger. Should be more helpful, I ++ think. ++ ++2002-09-17 07:07 rockyb ++ ++ * README: Revise for BASH debugger. ++ ++2002-09-17 06:54 rockyb ++ ++ * MANIFEST: Add new miscellaneous tests. ++ ++2002-09-16 17:37 rockyb ++ ++ * debugger/: NEWS, bashdb-fns.inc, bashdb-main.inc, bashdb-sig.inc: ++ Move debug trap handler to bashdb-sig.inc and rename it, well, ++ debug_trap_handler. ++ ++ Save restart set parameters and inside set -xv (at least ++ initially). ++ ++2002-09-16 17:33 rockyb ++ ++ * debugger/tests/: Makefile.am, dbg-test2.sh, misc.cmd, misc.right, ++ misc.tests, prof1.cmd, prof2.cmd, run-misc: Add tests of ++ miscellaneous commands: source, info args, show args, show ++ warranty, and show copying. ++ ++2002-09-14 18:26 rockyb ++ ++ * debugger/NEWS: Update. ++ ++2002-09-14 18:25 rockyb ++ ++ * MANIFEST: Add new brkpt2 tests ++ ++2002-09-14 18:24 rockyb ++ ++ * doc/bashdb.texi: Change documentation of "info args" and ARG. ++ Reorganize sections a bit so there's more in the command reference. ++ Document _bashdb_ variables. ++ ++2002-09-14 18:14 rockyb ++ ++ * debugger/bashdb-help.inc: Change info args to match closer gdb. ++ Is now $1, $2 and so on rather than initial program arguments. ++ ++2002-09-14 17:54 rockyb ++ ++ * debugger/: bashdb-fns.inc, bashdb-main.inc: Now support $1 $2 as ++ ${ARG[1}}, ${ARG[2]}. ++ ++2002-09-14 17:53 rockyb ++ ++ * debugger/tests/: Makefile.am, brkpt1.right, brkpt2.cmd, ++ brkpt2.right, dbg-test1.sh, list.right, watch1.right: Modify ++ dbg-test1.sh so we can test parameters. In the future though, ++ probably time to start a dbg-test2.sh ++ ++ Makefile.am: add brkpt2 test. ++ ++2002-09-14 15:39 rockyb ++ ++ * debugger/tests/: brkpt2.cmd, brkpt2.right, brkpt2.tests, ++ run-brkpt2: Now that we have a real command file mechanism, use it ++ to test correct functioning of break, watch, watche, step, next, ++ continue and stack handling. ++ ++2002-09-14 14:51 rockyb ++ ++ * debugger/bashdb-cmds.inc: Missed place in renaming _hi to ++ _bashdb_hi ++ ++2002-09-14 12:19 rockyb ++ ++ * debugger/: NEWS, TODO, bashdb, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-init.inc, bashdb-sig.inc: Variable name cleanup: more now ++ start with _bashdb ++ ++2002-09-14 12:09 rockyb ++ ++ * debugger/bashdb-cmds.inc: Some variable name cleanup and removal. ++ ++2002-09-14 11:40 rockyb ++ ++ * debugger/bashdb-cmds.inc: Correct "restart" command when running ++ via "bash -c ..." ++ ++2002-09-14 07:36 rockyb ++ ++ * debugger/: NEWS, bashdb-init.inc, bashdb-io.inc: Small cleanups. ++ BOGUS file renamed to *BOGUS*. ++ ++2002-09-14 07:21 rockyb ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-init.inc, bashdb-main.inc: Clean up code for debugger source ++ command. Add "info source" ++ ++2002-09-14 01:09 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-init.inc: Slightly cleaner ++ code. ++ ++2002-09-14 00:51 rockyb ++ ++ * debugger/: NEWS, bashdb-cmds.inc, bashdb-init.inc: Command files ++ now nest and pick up where they left off in between script ++ execution. There are still some minor cleanups and a small bug here ++ or there. ++ ++2002-09-13 23:04 rockyb ++ ++ * shell.c, shell.h, variables.c, debugger/bashdb-cmds.inc, ++ debugger/bashdb-fns.inc, debugger/bashdb-init.inc, ++ debugger/bashdb-io.inc: bash --debugger -c "..." now works. (Bug in ++ restarting currently though. restart now picks up full bash name ++ from BASH environment variable. ++ ++ To accomplish the above, environment variable BASH_EXECUTION_STRING ++ was added which contains the contents what was passed after -c. ++ ++2002-09-13 21:38 rockyb ++ ++ * error.c: Fix bug when no source file and have an error. ++ ++2002-09-13 21:24 rockyb ++ ++ * debugger/: NEWS, bashdb-brk.inc: Fix bug where "watche x > 24" ++ was creating output to file 24 ++ ++2002-09-13 21:23 rockyb ++ ++ * debugger/tests/watch1.right: Now, more right. ++ ++2002-09-13 15:34 rockyb ++ ++ * configure.in, doc/bashdb.texi: Getting ready for 0.24 release. ++ ++2002-09-13 15:24 rockyb ++ ++ * tests/: func.right, trap.right: Updated for new function. ++ ++2002-09-13 15:24 rockyb ++ ++ * debugger/tests/: brkpt1.right, list.right: Update for merged ++ commands. ++ ++2002-09-13 15:09 rockyb ++ ++ * debugger/NEWS: Add: Function names now supported by changing the ++ semantics of declare -F. ++ ++2002-09-13 15:09 rockyb ++ ++ * builtins/declare.def, doc/bashref.texi: Change semantics of ++ declare -F: now adds source file and line number. ++ ++2002-09-13 15:06 rockyb ++ ++ * execute_cmd.c, debugger/NEWS, doc/bashdb.texi, ++ tests/dbg-support.right, tests/dbg-support.tests: Function names ++ now supported. As a result, can now set breakpoints on them. ++ However in non-interactive mode the line number is wrong. It is the ++ caller line. ++ ++2002-09-13 15:03 rockyb ++ ++ * debugger/: TODO, bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ tests/list.right: Function names now supported. As a result, can ++ now set breakpoints on them. However in non-interactive mode the ++ line number is wrong. It is the caller line. ++ ++ As a result of the above, now extended these commands to include ++ linespecs: continue, break, list. ++ ++2002-09-13 11:12 rockyb ++ ++ * debugger/tests/dbg-test1.sh: Remove CVS id line which messes up ++ regression testing. ++ ++2002-09-12 18:35 rockyb ++ ++ * debugger/tests/: list.cmd, list.right: More tests, better doc on ++ tests. ++ ++2002-09-12 18:21 rockyb ++ ++ * builtins/evalstring.c: Just doc/grammar changes. ++ ++2002-09-12 18:19 rockyb ++ ++ * MANIFEST: Add new file: dbg-test1.sub ++ ++2002-09-12 18:18 rockyb ++ ++ * debugger/: NEWS, bashdb-cmds.inc, bashdb-io.inc: Add list ++ file:line. Rename internal routines from print to list since that's ++ what the command is called. ++ ++2002-09-12 18:17 rockyb ++ ++ * debugger/tests/dbg-test1.sub: Now have sourced file in debugger ++ test. ++ ++2002-09-12 18:16 rockyb ++ ++ * debugger/tests/: brkpt1.right, dbg-test1.sh, list.cmd, ++ list.right: Add check of list file:line. ++ ++2002-09-12 18:10 rockyb ++ ++ * tests/dbg-support.sub: Minor. ++ ++2002-09-12 16:17 rockyb ++ ++ * debugger/: bashdb-cmds.inc, bashdb-fns.inc: Change names of ++ variables so as not to conflict with user variables. Thus more of ++ them are prefaced with "_bashdb_." May have introduced a bug in the ++ renaming though... ++ ++2002-09-12 16:16 rockyb ++ ++ * execute_cmd.c: List lines in "for var in" before each time it is ++ executed rather than just once. Line number setting also tries to ++ do the right thing so that error messages on the line are reported ++ correctly too. ++ ++2002-09-12 16:10 rockyb ++ ++ * tests/dbg-support.right: Make sure "for word in ..." gives line ++ number of "for" every time the for loop is executed, not just the ++ *first* time. ++ ++2002-09-12 16:09 rockyb ++ ++ * tests/: errors.right, errors.tests: Add checks on errors in "for ++ word in ..." loops. Should have line number of "for" ++ ++2002-09-12 06:50 rockyb ++ ++ * tests/: dbg-support.right, dbg-support.tests: Check that we list ++ a "for" line number. ++ ++2002-09-12 06:44 rockyb ++ ++ * command.h, execute_cmd.c, make_cmd.c, make_cmd.h, parse.y, ++ debugger/NEWS, debugger/TODO: Now list/stop line of "for" or ++ "select". ++ ++2002-09-12 06:41 rockyb ++ ++ * doc/bashdb.texi: Minor. ++ ++2002-09-12 06:40 rockyb ++ ++ * MANIFEST: Remove brkpt tests - usable ones under debugger/tests ++ ++2002-09-11 22:36 rockyb ++ ++ * debugger/TODO: Note more line number problems. ++ ++2002-09-11 22:36 rockyb ++ ++ * debugger/bashdb-sig.inc: Add cmd-loop loop in exit handler so we ++ in fact don't exit! ++ ++2002-09-11 22:13 rockyb ++ ++ * debugger/bashdb-fns.inc: Don't go into debugger command loop when ++ called from the debugger: (the caller is _bashdb_). ++ ++2002-09-11 19:32 rockyb ++ ++ * htdocs/index.html: Now list download by date descending. ++ ++2002-09-11 18:40 rockyb ++ ++ * doc/bashdb.texi: Small grammar change. ++ ++2002-09-11 18:33 rockyb ++ ++ * htdocs/index.html: Add link to documentation. ++ ++2002-09-11 17:24 rockyb ++ ++ * tests/: brkpt.right, brkpt.tests, brkpt1.test, brkpt2.test, ++ brkpt3.test, brkpt4.test, brkpt5.test, run-brkpt: Some of these are ++ now in debugger/tests ++ ++2002-09-11 17:12 rockyb ++ ++ * MANIFEST, configure.in: Getting ready for release 0.23 ++ ++2002-09-11 08:43 rockyb ++ ++ * doc/bashdb.texi: Document recent changes to debugger: W suffix ++ on breakpoint/watchpoint numbers, break file:line Condition ++ 9999. ++ ++2002-09-11 07:39 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-fns.inc, bashdb-init.inc, ++ tests/brkpt1.cmd, tests/brkpt1.right, tests/watch1.right: Add break ++ file:line syntax. Add counts on watchpoints and display of this on ++ listing. Display breakpoint number when adding. i ++ ++2002-09-11 07:38 rockyb ++ ++ * debugger/NEWS: Version 0.23 changes (so far) ++ ++2002-09-11 00:36 rockyb ++ ++ * debugger/bashdb-cmds.inc: Make safe for tampering with IFS. ++ ++2002-09-10 13:29 rockyb ++ ++ * debugger/tests/: list.cmd, list.right: Add tests of set/show ++ listsize. ++ ++2002-09-10 13:25 rockyb ++ ++ * debugger/bashdb-cmds.inc: Fix bug in comment without space after ++ #, e.g. #foo. ++ ++2002-09-10 13:13 rockyb ++ ++ * debugger/bashdb-cmds.inc: Improve comment pattern so "#foo" works ++ as does "# foo." ++ ++2002-09-10 12:53 rockyb ++ ++ * debugger/tests/Makefile.am: Add watch1 test. ++ ++2002-09-10 12:49 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc: Bugs fixed in getting ++ watch command regression tests hammered out; eval now checks for ++ tty, listing watchpoints quotes expressions. ++ ++2002-09-10 12:40 rockyb ++ ++ * debugger/tests/: run-watch1, watch1.cmd, watch1.right, ++ watch1.tests: Add first watchpoint tests (enough that we can handle ++ with incomplete command files). ++ ++2002-09-10 11:47 rockyb ++ ++ * debugger/TODO: Cross a couple of things off the list. ++ ++2002-09-10 11:39 rockyb ++ ++ * Makefile.in: Make check now runs check in debugger as well. ++ ++2002-09-10 11:02 rockyb ++ ++ * debugger/bashdb-brk.inc: Test to see that line breaks are inside ++ lines Remove bug in not setting break condition correctly. Message ++ if no breakpoints have been set. ++ ++2002-09-10 11:00 rockyb ++ ++ * debugger/bashdb-cmds.inc: More crud in comments. ++ ++2002-09-10 10:59 rockyb ++ ++ * debugger/bashdb-cmds.inc: Removed some crud. ++ ++2002-09-10 10:58 rockyb ++ ++ * debugger/bashdb-cmds.inc: Rename to _bashdb_cmd_break to be ++ consistent in naming. ++ ++2002-09-10 10:56 rockyb ++ ++ * debugger/tests/: Makefile.am, brkpt1.cmd, brkpt1.right, ++ brkpt1.tests, run-brkpt1: Add breakpoint tests - or those that we ++ can test out right now... ++ ++2002-09-10 10:55 rockyb ++ ++ * debugger/tests/list.tests: More generic format. ++ ++2002-09-10 09:47 rockyb ++ ++ * tests/: dbg-list.right, dbg-list.tests, list.cmd, run-dbg-list: ++ Now located under debugger/tests ++ ++2002-09-10 09:08 rockyb ++ ++ * debugger/: Makefile.am, configure.in: Add support to create/run ++ regression tests in tests directory. ++ ++2002-09-10 09:07 rockyb ++ ++ * debugger/tests/.cvsignore: A fact of life. ++ ++2002-09-10 09:06 rockyb ++ ++ * debugger/tests/: Makefile.am, dbg-test1.sh, list.cmd, list.right, ++ list.tests, run-all, run-list, version: Debugger tests are now ++ separated from bash tests, so we can test that separately. And we ++ now use automake's code too! ++ ++2002-09-10 00:06 rockyb ++ ++ * doc/bashdb.texi: Remove some more gdb things with respect to I/O ++ redirection that just aren't here. ++ ++2002-09-09 23:12 rockyb ++ ++ * doc/bashdb.texi: More on the differences between x, eval and ++ print. Clean up some ugly grammar. ++ ++2002-09-09 22:50 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc: Enable/disable of ++ watchpoints done. delete, enable and disable will allow suffix "w" ++ for watchpoint. ++ ++2002-09-09 20:55 rockyb ++ ++ * debugger/bashdb-help.inc: Add help for set command. Minor text ++ changes. ++ ++2002-09-09 19:45 rockyb ++ ++ * debugger/NEWS: Add case statement work. ++ ++2002-09-09 19:41 rockyb ++ ++ * .cvsignore: This too is (now) derived. ++ ++2002-09-09 19:41 rockyb ++ ++ * parse.y, tests/dbg-support.right, tests/dbg-support.tests: Wasn't ++ handling nested cases. ++ ++2002-09-09 19:09 rockyb ++ ++ * debugger/TODO: Update as appropriate. ++ ++2002-09-09 19:02 rockyb ++ ++ * debugger/TODO: Update as appropriate. ++ ++2002-09-09 19:01 rockyb ++ ++ * tests/: dbg-list.right, dbg-list.tests: Tests debugger list ++ command. ++ ++2002-09-09 19:00 rockyb ++ ++ * tests/: list.cmd, run-dbg-list: Test debugger list command. ++ ++2002-09-09 18:59 rockyb ++ ++ * command.h, execute_cmd.c, make_cmd.c, make_cmd.h, parse.y, ++ tests/dbg-support.right, tests/dbg-support.tests: Add line number ++ for beginning of case/select; debug now lists those lines. ++ ++2002-09-09 18:56 rockyb ++ ++ * tests/: brkpt.right, brkpt1.test, brkpt3.test, brkpt4.test, ++ brkpt5.test: Modify for current situation. These need to be redone ++ since they're wrong anyway. Split out status dbg tests? ++ ++2002-09-09 17:16 rockyb ++ ++ * Makefile.in: Add rule to make derived aclocal.m4. ++ ++ Should get rid of Makefile.in and go with Makefile.am. But this ++ diverges bash even more. ++ ++2002-09-09 14:26 rockyb ++ ++ * MANIFEST: Add new file bashlocal.m4 ++ ++2002-09-09 14:25 rockyb ++ ++ * bashlocal.m4, configure.in, cvscompile.sh: Add --lispdir to ++ configure and pass this along to debugger. Derived file aclocal.m4 ++ no longer in repository. BASH-specific macros moved to bashlocal.m4 ++ ++2002-09-09 12:58 rockyb ++ ++ * cvscompile.sh: Also need to run aclocal. ++ ++2002-09-09 12:50 rockyb ++ ++ * cvscompile.sh: Add missing debugger autoconf. ++ ++2002-09-09 12:39 rockyb ++ ++ * MANIFEST, cvscompile.sh, debugger/Makefile.am, debugger/TODO, ++ debugger/configure.in: Use Automake AM_LISPDIR rather than trying ++ to roll this code on my own (and failing). Thanks to Masatake ++ YAMATO! ++ ++2002-09-09 12:37 rockyb ++ ++ * debugger/NEWS: Add 0.22 features. ++ ++2002-09-09 12:27 rockyb ++ ++ * doc/bashdb.texi: Document watche. Put more command names in info ++ nodes. Move doc order around a little. ++ ++2002-09-09 12:22 rockyb ++ ++ * htdocs/index.html: Project page now points to Sourceforge ++ Summary, not Admin. Thanks to Masatake YAMATO. ++ ++2002-09-08 18:57 rockyb ++ ++ * debugger/bashdb-set.inc: Set/show commands ++ ++2002-09-07 14:43 rockyb ++ ++ * doc/bashdb.texi: Doc frame command. Small corrections to editing ++ in emacs. ++ ++2002-09-07 14:41 rockyb ++ ++ * debugger/bashdb-cmds.inc: Add frame command. Version command is ++ now a separate command. ++ ++2002-09-07 14:35 rockyb ++ ++ * debugger/bashdb-stack.inc: Get frame with args command hammered ++ out properly. ++ ++2002-09-07 09:36 rockyb ++ ++ * debugger/bashdb-io.inc: Some error correction and error message ++ on list command. ++ ++2002-09-07 09:35 rockyb ++ ++ * debugger/bashdb-cmds.inc: Ooops - forgot to allow set/show as ++ valid commands here. Move CVS Id to end. ++ ++2002-09-07 09:09 rockyb ++ ++ * MANIFEST, configure.in, debugger/Makefile.am, ++ debugger/bashdb-cmds.inc, debugger/bashdb-fns.inc, ++ debugger/bashdb-help.inc, debugger/bashdb-init.inc, ++ debugger/bashdb-io.inc, debugger/bashdb-main.inc, doc/bashdb.texi: ++ Start set/show commands. List now takes up where it left off. ++ Move some CVS id lines to the end to facilitate debugging Some ++ minor improvements. ++ ++2002-09-06 03:06 rockyb ++ ++ * debugger/: bashdb-brk.inc, bashdb-help.inc, bashdb-main.inc: ++ Minor formating changes or typos. ++ ++2002-09-06 01:59 rockyb ++ ++ * configure.in: Release change ++ ++2002-09-06 01:56 rockyb ++ ++ * debugger/Makefile.am: Include new guy bashdb-sig.inc. ++ ++2002-09-06 00:52 rockyb ++ ++ * doc/bashdb.texi: Today's update. All sections are at least ++ minimally done, so Manual minimally finished. ++ ++2002-09-05 22:39 rockyb ++ ++ * MANIFEST, debugger/bashdb-cmds.inc, debugger/bashdb-main.inc, ++ debugger/bashdb-sig.inc, tests/brkpt.right: Add signal handlers for ++ interrupt and exit. Now we catch when the program stopped and ++ report that. ++ ++2002-09-05 22:37 rockyb ++ ++ * debugger/TODO: Update as appropriate. ++ ++2002-09-05 10:54 rockyb ++ ++ * debugger/bashdb-main.inc: Slightly better? ++ ++2002-09-05 10:53 rockyb ++ ++ * debugger/bashdb-main.inc: Test to see if we are using a ++ debugger-enabled bash, and give up if not. ++ ++2002-09-05 09:54 rockyb ++ ++ * doc/bashdb.texi: Closer... ++ ++2002-09-05 08:56 rockyb ++ ++ * debugger/bashdb-brk.inc: Add null line + if in break command. ++ ++2002-09-05 08:55 rockyb ++ ++ * tests/: brkpt.right, brkpt2.test: Add test for null line + if ++ "break if" command ++ ++2002-09-05 00:47 rockyb ++ ++ * MANIFEST: Add in debugger breakpoint regression test. ++ ++2002-09-05 00:47 rockyb ++ ++ * tests/brkpt.tests: Put debugger bash in explicitly. ++ ++2002-09-05 00:27 rockyb ++ ++ * tests/run-all: To make work on Solaris's (and probably other's) ++ /bin/sh. ++ ++2002-09-04 23:31 rockyb ++ ++ * htdocs/index.html: Slight mods. Still plain as sin. ++ ++2002-09-04 23:25 rockyb ++ ++ * htdocs/index.html: Run through ispell. ++ ++2002-09-04 23:22 rockyb ++ ++ * htdocs/index.html: Correct group id. ++ ++2002-09-04 23:10 rockyb ++ ++ * htdocs/index.html: Lousy Web page. ++ ++2002-09-04 22:24 rockyb ++ ++ * debugger/TODO: Update. ++ ++2002-09-04 22:12 rockyb ++ ++ * configure.in: Rename release name. ++ ++2002-09-04 22:11 rockyb ++ ++ * tests/: bashdb-bp.tests, brkpt.right, brkpt.tests, brkpt1.test, ++ brkpt2.test, brkpt3.test, brkpt4.test, brkpt5.test, run-brkpt: ++ Regression tests for breakpoint. Actually the "right" file isn't ++ right due to a bug in debugger comand-file processing. However it ++ is largely right and this keeps us from breaking things even more. ++ ++2002-09-04 22:09 rockyb ++ ++ * tests/: run-dbg-is-fn, run-dbg-support: Just #!../bash lines and ++ perhaps clue Emacs in that these are scripts. ++ ++2002-09-04 22:08 rockyb ++ ++ * debugger/: bashdb, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-fns.inc, bashdb-help.inc, bashdb-init.inc, bashdb-io.inc: ++ Redo the way breakpoints are handled. Is more like gdb than ++ perltdb. ++ ++ Now have enable/disable and breakpoint conditions. ++ ++ Some other minor changes. Like testing to see if we have an ++ attached tty. ++ ++2002-09-03 13:49 rockyb ++ ++ * MANIFEST: We install as bash.info, not bashref. ++ ++2002-09-03 13:47 rockyb ++ ++ * doc/bashdb.texi: Minor corrections. Mostly to options-processing ++ area. ++ ++2002-09-03 13:22 rockyb ++ ++ * debugger/bashdb: Change option -s to -q as it is in gdb. ++ ++2002-09-03 13:09 rockyb ++ ++ * doc/.cvsignore: Forgot yet another... ++ ++2002-09-03 13:08 rockyb ++ ++ * doc/.cvsignore: Boy, doc's have lots of temporary files! ++ ++2002-09-03 13:00 rockyb ++ ++ * cvscompile.sh: Script for kicking everything off from CVS ++ initially. ++ ++2002-09-03 12:47 rockyb ++ ++ * debugger/: bashdb, bashdb-init.inc: Add more options - x: command ++ file run, n: no init file run, -V show version, and -s silient. All ++ compatible with GDB. ++ ++2002-09-03 12:36 rockyb ++ ++ * debugger/bashdb-help.inc: Minor change to refer to doc. ++ ++2002-09-03 12:35 rockyb ++ ++ * doc/fdl.texi: Perhaps should not be in CVS since I didn't write. ++ But it is part of distribution. Remove later if can figure out a ++ way to get it from somewhere else. ++ ++2002-09-03 12:34 rockyb ++ ++ * doc/gpl.texi: Perhaps should not be in CVS since I didn't write. ++ However it is part of distribution. Remove later if can figure out ++ a better way. ++ ++2002-09-03 12:30 rockyb ++ ++ * debugger/Makefile.am: Add newly-created file bashdb-io.inc ++ ++2002-09-03 12:28 rockyb ++ ++ * MANIFEST: Add missing files to make bashdb.texi documentation and ++ build debugger. ++ ++2002-09-03 12:25 rockyb ++ ++ * doc/Makefile.in: Incorrectly spelled directory in TEXINPUTS. Now ++ should build bashdb.texi correctly. ++ ++2002-09-03 07:25 rocky ++ ++ * .cvsignore: Ignore distribution file. ++ ++2002-09-03 07:24 rocky ++ ++ * Makefile.in: Remove distribution tar.gz file. Also tilde files ++ (from CVS checkouts). ++ ++2002-09-03 07:22 rocky ++ ++ * doc/.cvsignore: Many derived files here. ++ ++2002-09-03 07:18 rocky ++ ++ * MANIFEST: Add missing files for making a distribution. ++ ++2002-09-03 07:00 rocky ++ ++ * tests/errors.right: Previous version was completely bogus. ++ ++2002-09-02 23:22 rocky ++ ++ * doc/bashdb.texi: Changes for today: sections reorganized, gdb ++ stuff chopped out. Now has a command index. And text modified for ++ the BASH debugger. ++ ++ About 2/3 of the way correct. ++ ++2002-09-02 23:19 rocky ++ ++ * doc/Makefile.in: Change to allow automatic build of bashdb doc. ++ ++2002-09-02 23:18 rocky ++ ++ * support/texi2dvi: Perhaps shouldn't be in CVS, but do so for now. ++ ++2002-09-02 23:18 rocky ++ ++ * support/man2html: Perhaps shouldn't be in CVS - but do so for ++ now...x ++ ++2002-09-02 23:17 rocky ++ ++ * support/Makefile.in: Add missing CC_FOR_BUILD ++ ++2002-09-02 22:37 rocky ++ ++ * debugger/: TODO, bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-help.inc, bashdb-init.inc, bashdb-io.inc, bashdb-main.inc, ++ bashdb-stack.inc: Major cleanup/expansion and a ++ bug/misunderstanding removed. Now reads in files once. Unset ++ breakpoint code no longer inlined. "info files" now works Stack ++ trace bug when via "info stack" fixed. bashdb-io.inc created. ++ ++2002-09-02 15:29 rocky ++ ++ * debugger/: bashdb, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-fns.inc, bashdb-help.inc, bashdb-init.inc: Add info command. ++ In the process, we now allow getting tty and args. ++ ++2002-09-02 00:36 rocky ++ ++ * support/: mkdist, texi2html: Not sure if this should be here. For ++ now okay. ++ ++2002-09-02 00:36 rocky ++ ++ * MANIFEST: Add more good stuff like the bashdb reference. ++ ++2002-09-02 00:35 rocky ++ ++ * debugger/TODO: Revise as appropriate. ++ ++2002-09-02 00:32 rocky ++ ++ * doc/bashdb.texi: [no log message] ++ ++2002-09-02 00:29 rocky ++ ++ * doc/bashdb.texi: Grossly inaccurate but there's stuff in here ++ that is in fact correct. More later... ++ ++2002-09-02 00:28 rocky ++ ++ * debugger/bashdb-help.inc: Update to reflect current situation. ++ ++2002-09-02 00:16 rocky ++ ++ * debugger/: bashdb-cmds.inc, bashdb-help.inc, bashdb-stack.inc: ++ Add bt backtrace as an alias for where, or T. Add parameter to ++ specify how many frames to list on bt. ++ ++2002-09-01 19:57 rocky ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-stack.inc: ++ Mostly formatting. Change basic indent from 4 to 2. Stack commands ++ were duplicated in bashdb-cmds.inc, and a file identifier incorrect ++ in bashdb-brk.inc ++ ++2002-09-01 11:05 rocky ++ ++ * debugger/: bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-help.inc, bashdb-main.inc: Minor: correct function comments ++ at top of file. ++ ++2002-09-01 11:05 rocky ++ ++ * debugger/Makefile.am: Break into more smaller files. ++ ++2002-09-01 11:04 rocky ++ ++ * debugger/bashdb-stack.inc: Break away from bashdb-cmds.inc ++ ++2002-09-01 10:41 rocky ++ ++ * debugger/bashdb-brk.inc: Things with breakpoints are now in a ++ separate file. ++ ++2002-09-01 10:40 rocky ++ ++ * debugger/bashdb-help.inc: Things with breakpoints and watchpoints ++ (and later actions) are now their own separate file ++ ++2002-09-01 10:39 rocky ++ ++ * debugger/bashdb-help.inc: Help is now it's own separate file. ++ ++2002-09-01 05:46 rocky ++ ++ * MANIFEST: Add debugger files ++ ++2002-09-01 05:32 rocky ++ ++ * debugger/: bashdb-cmds.inc, bashdb-fns.inc: Change pe to x. On ++ watch, set initial value (and display it). Allow dump of variable. ++ Add internal function _is_var. ++ ++2002-09-01 04:08 rocky ++ ++ * debugger/bashdb-fns.inc: Separate watch variables from watch ++ expressions ++ ++2002-09-01 04:07 rocky ++ ++ * debugger/bashdb-init.inc: Add array to note if watchpoint is ++ arithmetic or not. ++ ++2002-09-01 04:06 rocky ++ ++ * debugger/bashdb-cmds.inc: Separate out watch variable and watch ++ expressions. ++ ++2002-09-01 02:47 rocky ++ ++ * debugger/bashdb-init.inc: Set default next/step default value. ++ ++2002-08-31 23:34 rocky ++ ++ * debugger/: bashdb-cmds.inc, bashdb-init.inc: Make more like gdb. ++ Implement aliases: run, shell. Add Init file startup ~.bashdbinit. ++ Save args on print and pe. ++ ++2002-08-30 22:33 rocky ++ ++ * tests/: dbg-support.right, dbg-support.tests: Add tests to make ++ sure command substitution traces properly and produces the correct ++ call stack. ++ ++2002-08-30 22:31 rocky ++ ++ * tests/: errors.right, errors.sub, errors.tests: Add tests that ++ make sure sourced file names appear. ++ ++2002-08-30 20:44 rocky ++ ++ * error.c: Now correctly reports the file location in the presence ++ of sourced files. ++ ++2002-08-30 20:44 rocky ++ ++ * debugger/CHANGES: Note watchpoints, and line number changes. ++ ++2002-08-30 16:17 rocky ++ ++ * debugger/: bashdb-cmds.inc, bashdb-fns.inc: Small cleanups. ++ Remove function break code which doesn't work anyway. ++ ++2002-08-30 12:46 rocky ++ ++ * debugger/: bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc: Add ++ watchpoints. Help and commands now allow unique prefix, e.g. del ++ for delete. ++ ++2002-08-30 12:20 rocky ++ ++ * debugger/AUTHORS: Again. ++ ++2002-08-30 12:18 rocky ++ ++ * debugger/AUTHORS: More precies. ++ ++2002-08-30 12:18 rocky ++ ++ * subst.c: Reset line number if interactive. ++ ++2002-08-30 12:17 rocky ++ ++ * trap.c: Respect function tracing if "set -o fntrace is in ++ effect." ++ ++2002-08-30 00:55 rocky ++ ++ * debugger/TODO: Add watchpoints as requested by Bert Caridad. ++ ++2002-08-28 23:27 rocky ++ ++ * builtins/fc.def: Added new extra parameter (renumber_line) on ++ parse_and_execute. ++ ++2002-08-27 20:56 rocky ++ ++ * debugger/bashdb-cmds.inc: Fixed bug Inadvertently added by change ++ \? to ?. (Is now '?') to help emacs script editing. ++ ++2002-08-27 20:52 rocky ++ ++ * debugger/bashdb-cmds.inc: Doc continue enhancement. ++ ++2002-08-27 20:51 rocky ++ ++ * debugger/: bashdb-cmds.inc, bashdb-fns.inc: clearing breakpoints ++ now works via a workaround. ++ ++2002-08-27 20:48 rocky ++ ++ * debugger/TODO: Update to reflect current situation. ++ ++2002-08-26 23:53 rocky ++ ++ * debugger/TODO: Did "up" and "down" - remove from list. ++ ++2002-08-26 23:52 rocky ++ ++ * debugger/bashdb.el: Add back in "up" and "down" which we have ++ (but perldb doesn't). ++ ++2002-08-26 23:51 rocky ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-init.inc: "Up [count} and down {count] working. Stack trace ++ is now more like gdb than perl. Probably will move more in that ++ direction, too. ++ ++ Note: There may be some funniness in FUNCNAME BASH_SOURCE and ++ BASH_LINENO. ++ ++2002-08-26 09:22 rocky ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-init.inc: S !pat now works. Some alphabetization of commands ++ and more subroutings in bashdb-cmds start _bashdb_cmd_. init Has ++ where we are in stack for getting ready to put in up and down. ++ ++2002-08-26 09:20 rocky ++ ++ * debugger/TODO: Update. ++ ++2002-08-26 09:20 rocky ++ ++ * debugger/NEWS: More features - note that we're useful even if you ++ don't use the debugger. ++ ++2002-08-26 08:06 rocky ++ ++ * debugger/bashdb-cmds.inc: This time, for sure! ++ ++2002-08-26 08:05 rocky ++ ++ * debugger/bashdb-cmds.inc: - now lists up to _curline Bug in ++ restart for bash --debugger ++ ++2002-08-26 06:55 rocky ++ ++ * debugger/bashdb: Minor changes. ++ ++2002-08-26 05:45 rocky ++ ++ * builtins/evalstring.c, debugger/CHANGES, debugger/TODO, ++ tests/errors.right: Fix off-by-one line number error in command ++ subtitutions $(), `` and {}. ++ ++2002-08-25 23:10 rocky ++ ++ * debugger/bashdb-cmds.inc: Help for individual commands. ++ ++2002-08-25 23:09 rocky ++ ++ * debugger/TODO: Added emacs lisp install. help on commands and ++ fixed substutution line numbers. ++ ++2002-08-25 23:08 rocky ++ ++ * debugger/: .cvsignore, Makefile.am, configure.in: batch-compile ++ and install emacs lisp files. (First cut: no doubt there's room for ++ improvement.) ++ ++2002-08-25 20:27 rocky ++ ++ * debugger/bashdb-cmds.inc: A couple more. ++ ++2002-08-25 20:17 rocky ++ ++ * debugger/bashdb-cmds.inc: Add long-form of commands. ++ ++2002-08-25 20:17 rocky ++ ++ * debugger/bashdb: Replace a [ with [[ ++ ++2002-08-25 17:50 rocky ++ ++ * debugger/CHANGES: Note line number change in command ++ substitution. ++ ++2002-08-25 17:42 rocky ++ ++ * bashline.c, jobs.c, parse.y, shell.c, subst.c, trap.c, ++ variables.c, builtins/common.h, builtins/eval.def, ++ builtins/evalfile.c, builtins/evalstring.c: Line number fix on ++ sourced files and command substitution. Added additional parameter ++ to parse_and_execute to indicate whether or not to reset the line ++ number. ++ ++2002-08-25 17:23 rocky ++ ++ * tests/dbg-support.right: Really this is the same as version 1.3. ++ Changes in output were really bugs! Sourced file needs line number ++ resetting. ++ ++2002-08-25 16:12 rocky ++ ++ * tests/dbg-support.tests: A couple more command substitution tests ++ and comments. ++ ++2002-08-25 13:56 rocky ++ ++ * tests/: dbg-support.right, dbg-support.tests, errors.right: ++ Update for line number change in running subshells. ++ ++2002-08-25 13:13 rocky ++ ++ * builtins/evalstring.c: Don't reset line number in running ++ subshells. This makes it hard for a debugger to track down where ++ you were. ++ ++ Would like to understand the benefit of this though. ++ ++2002-08-25 12:14 rocky ++ ++ * doc/.cvsignore: add another derived file. ++ ++2002-08-25 12:13 rocky ++ ++ * debugger/TODO: Another bug with subshells found. ++ ++2002-08-25 09:58 rocky ++ ++ * Makefile.in: Remake pathnames.h if it is missing ++ ++2002-08-25 09:58 rocky ++ ++ * debugger/: TODO, bashdb-cmds.inc: Fix restart in bash --debugger ++ ++2002-08-25 02:18 rocky ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-init.inc, ++ bashdb-main.inc: Minor variable name corrections and changes. ++ Mostly to do with version command. Typo in bashdb. ++ ++2002-08-25 02:04 rocky ++ ++ * configure.in, debugger/AUTHORS, debugger/Makefile.am, ++ debugger/PERMISSION, debugger/TODO, debugger/bashdb, ++ debugger/bashdb-cmds.inc, debugger/bashdb-fns.inc, ++ debugger/bashdb-init.inc, debugger/bashdb-main.inc: Last minute ++ cleanups before initial Savannah CVS import. ++ ++2002-08-24 18:13 rocky ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-fns.inc: Mostly small ++ extensions to make this match perl5db: - Add Subroutine list (S) ++ - Add list functions (l func) - Add variable dump (V) - Note ++ where we haven't finished implementing parameters - Fix help ++ ++ bashdb: rename functions to more system-like names (so won't appear ++ in function list) ++ ++2002-08-24 18:08 rocky ++ ++ * debugger/: TODO, NEWS: Update as appropriate. ++ ++2002-08-24 16:57 rocky ++ ++ * tests/: dbg-fns, dbg-fns.dat, dbg-is-fn.right, dbg-is-fn.tests, ++ dbg-support.right, run-dbg-is-fn: New tests: dbg-is-fn. Correct ++ output on dbg-support.right and set up for another new test ++ ++2002-08-24 16:51 rocky ++ ++ * tests/: dbg-support.right, dbg-support.tests, run-dbg-support: ++ CVS cleanup: add CVS Id line. Correct ooutput and allow tests to ++ be executable as is. ++ ++2002-08-24 00:58 rocky ++ ++ * debugger/TODO: Fixed "continue" ++ ++2002-08-24 00:55 rocky ++ ++ * debugger/: bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc: ++ Breakpoints now work! setting, deleting, clearing all, listing. ++ ++2002-08-23 14:15 rocky ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-fns.inc: CVS cleanup: ++ add CVS lines ++ ++2002-08-23 14:13 rocky ++ ++ * debugger/bashdb-fns.inc: Command "Continue" ('c") now runs ++ silently. ++ ++2002-08-23 14:11 rocky ++ ++ * .cvsignore: More of the same. ++ ++2002-08-23 12:16 rocky ++ ++ * tests/: dbg-support.right, dbg-support.sub, dbg-support.tests, ++ run-dbg-support: Debugging extensions to bash test. ++ ++2002-08-23 08:39 rocky ++ ++ * tests/.cvsignore: The usual. ++ ++2002-08-23 08:35 rocky ++ ++ * .cvsignore, builtins/.cvsignore: CVS cleanup. ++ ++2002-08-23 08:31 rocky ++ ++ * lib/: tilde/.cvsignore, sh/.cvsignore: The usual. ++ ++2002-08-23 08:30 rocky ++ ++ * lib/termcap/.cvsignore: This usual. ++ ++2002-08-23 08:27 rocky ++ ++ * .cvsignore, lib/readline/.cvsignore, lib/readline/doc/Makefile, ++ support/texi2dvi, support/texi2html: CVS cleanup. ++ ++2002-08-23 08:22 rocky ++ ++ * shell.c: Move execution of debugger init after setting up ++ argument list $0 and $*. ++ ++2002-08-23 07:57 rocky ++ ++ * debugger/: TODO, bashdb, bashdb-fns.inc, bashdb-init.inc: Get ++ dual execution of "bashdb" and "bash --debugger" with respect to ++ command-line arguments. In bashdb however, $0 is bashdb not the ++ program. ++ ++2002-08-23 07:56 rocky ++ ++ * debugger/bashdb.el: Remove looking for "--emacs" option in ++ calling bashdb. This is used in perldb but I don't know if we need ++ it here. It might get added later when we know what this is about ++ and decide it's necessary. ++ ++2002-08-23 00:19 rocky ++ ++ * examples/loadables/.cvsignore, ++ examples/loadables/perl/.cvsignore, lib/.cvsignore, ++ lib/glob/.cvsignore, lib/malloc/.cvsignore: The usual. ++ ++2002-08-23 00:18 rocky ++ ++ * doc/.cvsignore: The usual ++ ++2002-08-23 00:17 rocky ++ ++ * .cvsignore, array.h, command.h, configure.in, flags.h, ++ pathnames.h.in, shell.c, variables.h, builtins/evalfile.c, ++ builtins/source.def, doc/bashref.texi: "bash --debugger" now kicks ++ things off. ++ ++2002-08-23 00:13 rocky ++ ++ * .cvsignore, builtins/.cvsignore, support/.cvsignore, ++ examples/.cvsignore: The usual. ++ ++2002-08-23 00:08 rocky ++ ++ * debugger/: bashdb, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-init.inc, bashdb-main.inc, bashdb.el: Changes so that we can ++ support "bash --debugger" ++ ++2002-08-22 23:29 rocky ++ ++ * debugger/: CHANGES, TODO: Update to reflect current state of ++ things. ++ ++2002-08-22 08:13 rocky ++ ++ * debugger/bashdb: _cleanup -> _bashdb_cleanup ++ ++2002-08-22 08:03 rocky ++ ++ * debugger/bashdb: Bug: not handling -T correctly. Also add emacs ++ mode. ++ ++2002-08-22 07:58 rocky ++ ++ * debugger/: Makefile.am, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-init.inc, bashdb-main.inc: Split bashdb-fns into multiple ++ *.inc files ++ ++2002-08-22 07:56 rocky ++ ++ * debugger/.cvsignore: Yet more of the usual ignored files... ++ ++2002-08-22 07:54 rocky ++ ++ * debugger/.cvsignore: CVS: Useful file. ++ ++2002-08-21 19:53 rocky ++ ++ * debugger/: CHANGES, bashdb, bashdb-main.inc: Now source the file ++ to be debugged rather than use that hacky temp file. x ++ ++2002-08-20 13:48 rocky ++ ++ * Makefile.in: Include debugger in installing ++ ++2002-08-20 12:56 rocky ++ ++ * tests/array.right: add BASH_LINENO ++ ++2002-08-20 12:17 rocky ++ ++ * tests/: array.right, run-all, shopt.right: Revise for new ++ options. run-all now allows running a specified set of tests ++ ++2002-08-20 12:15 rocky ++ ++ * builtins/set.def: Add fntrace or -d option ++ ++2002-08-20 12:14 rocky ++ ++ * array.c, execute_cmd.c, flags.c, make_cmd.c, shell.c, ++ variables.c, builtins/evalfile.c: Add BASH_LINENO, BASH_SOURCE_FILE ++ arrays Turn FUNCNAME into an ARRAY. Add fntrace or -d set option ++ ++2002-08-20 12:10 rocky ++ ++ * debugger/: bashdb, bashdb-main.inc: Bogosity in assuming ++ single-file script removed. Now tracks source file correctly (in ++ conjunction with changes in bash) ++ ++2002-08-20 12:08 rocky ++ ++ * debugger/bashdb.el: [no log message] ++ ++2002-08-20 12:07 rocky ++ ++ * debugger/bashdb.el: Add GPL ++ ++2002-08-20 12:05 rocky ++ ++ * debugger/README: Add CVS id line ++ ++2002-08-20 12:05 rocky ++ ++ * debugger/README: No longer a modification but a rewrite. Note ++ that bashdb now requires my patches. ++ ++2002-08-20 12:04 rocky ++ ++ * debugger/: AUTHORS, CHANGES, COPYING, ChangeLog, INSTALL, ++ Makefile.am, NEWS, TODO, configure.in: Automake'd and autoconf'd ++ ++2002-08-20 11:34 rocky ++ ++ * debugger/bashdb-main.inc: Add stack trace. Some typos corrected. ++ ++2002-08-20 11:31 rocky ++ ++ * debugger/bashdb-main.inc: Add GPL ++ ++2002-08-20 11:16 rocky ++ ++ * AUTHORS, CHANGES, COMPAT, COPYING, INSTALL, MANIFEST, ++ Makefile.in, NEWS, NOTES, POSIX, RBASH, README, Y2K, alias.c, ++ alias.h, array.c, array.h, arrayfunc.c, arrayfunc.h, bashansi.h, ++ bashhist.c, bashhist.h, bashintl.h, bashjmp.h, bashline.c, ++ bashline.h, bashtypes.h, bracecomp.c, braces.c, builtins.h, ++ command.h, config-bot.h, config-top.h, config.h.in, configure.in, ++ conftypes.h, copy_cmd.c, dispose_cmd.c, dispose_cmd.h, error.c, ++ error.h, eval.c, execute_cmd.c, execute_cmd.h, expr.c, externs.h, ++ findcmd.c, findcmd.h, flags.c, flags.h, general.c, general.h, ++ hashcmd.c, hashcmd.h, hashlib.c, hashlib.h, input.c, input.h, ++ jobs.c, jobs.h, list.c, locale.c, mailcheck.c, mailcheck.h, ++ make_cmd.c, make_cmd.h, mksyntax.c, nojobs.c, parse.y, parser.h, ++ patchlevel.h, pathexp.c, pathexp.h, pcomplete.c, pcomplete.h, ++ pcomplib.c, print_cmd.c, quit.h, redir.c, redir.h, shell.c, ++ shell.h, sig.c, sig.h, siglist.c, siglist.h, stringlib.c, subst.c, ++ subst.h, syntax.h, test.c, test.h, trap.c, trap.h, unwind_prot.c, ++ unwind_prot.h, variables.c, variables.h, version.c, xmalloc.c, ++ xmalloc.h, CWRU/PLATFORMS, CWRU/README, CWRU/changelog, ++ CWRU/mh-folder-comp, CWRU/sh-redir-hack, CWRU/misc/bison, ++ CWRU/misc/errlist.c, CWRU/misc/hpux10-dlfcn.h, ++ CWRU/misc/open-files.c, CWRU/misc/sigs.c, CWRU/misc/sigstat.c, ++ builtins/Makefile.in, builtins/alias.def, builtins/bashgetopt.c, ++ builtins/bashgetopt.h, builtins/bind.def, builtins/break.def, ++ builtins/builtin.def, builtins/cd.def, builtins/colon.def, ++ builtins/command.def, builtins/common.c, builtins/common.h, ++ builtins/complete.def, builtins/declare.def, builtins/echo.def, ++ builtins/enable.def, builtins/eval.def, builtins/evalfile.c, ++ builtins/evalstring.c, builtins/exec.def, builtins/exit.def, ++ builtins/fc.def, builtins/fg_bg.def, builtins/getopt.c, ++ builtins/getopt.h, builtins/getopts.def, builtins/hash.def, ++ builtins/help.def, builtins/history.def, builtins/inlib.def, ++ builtins/jobs.def, builtins/kill.def, builtins/let.def, ++ builtins/mkbuiltins.c, builtins/printf.def, builtins/psize.c, ++ builtins/psize.sh, builtins/pushd.def, builtins/read.def, ++ builtins/reserved.def, builtins/return.def, builtins/set.def, ++ builtins/setattr.def, builtins/shift.def, builtins/shopt.def, ++ builtins/source.def, builtins/suspend.def, builtins/test.def, ++ builtins/times.def, builtins/trap.def, builtins/type.def, ++ builtins/ulimit.def, builtins/umask.def, builtins/wait.def, ++ cross-build/beos-sig.h, cross-build/cygwin32.cache, ++ cross-build/opennt.cache, cross-build/win32sig.h, ++ cross-build/x86-beos.cache, doc/FAQ, doc/INTRO, doc/Makefile.in, ++ doc/README, doc/article.ms, doc/bash.1, doc/bashbug.1, ++ doc/bashref.texi, doc/builtins.1, doc/htmlpost.sh, doc/rbash.1, ++ doc/texinfo.tex, examples/bashdb/PERMISSION, ++ examples/bashdb/bashdb, examples/bashdb/bashdb.el, ++ examples/complete/complete-examples, ++ examples/complete/complete.freebsd, ++ examples/complete/complete.gnu-longopt, ++ examples/complete/complete.ianmac, ++ examples/complete/complete2.ianmac, examples/functions/array-stuff, ++ examples/functions/array-to-string, examples/functions/autoload, ++ examples/functions/autoload.v2, examples/functions/autoload.v3, ++ examples/functions/basename, examples/functions/basename2, ++ examples/functions/coproc.bash, examples/functions/coshell.README, ++ examples/functions/coshell.bash, examples/functions/csh-compat, ++ examples/functions/dirfuncs, examples/functions/dirname, ++ examples/functions/emptydir, examples/functions/exitstat, ++ examples/functions/external, examples/functions/fact, ++ examples/functions/fstty, examples/functions/func, ++ examples/functions/gethtml, examples/functions/getoptx.bash, ++ examples/functions/inetaddr, examples/functions/inpath, ++ examples/functions/isnum.bash, examples/functions/isnum2, ++ examples/functions/isvalidip, examples/functions/jdate.bash, ++ examples/functions/jj.bash, examples/functions/keep, ++ examples/functions/ksh-cd, examples/functions/ksh-compat-test, ++ examples/functions/kshenv, examples/functions/login, ++ examples/functions/lowercase, examples/functions/manpage, ++ examples/functions/mhfold, examples/functions/notify.bash, ++ examples/functions/pathfuncs, examples/functions/recurse, ++ examples/functions/repeat2, examples/functions/repeat3, ++ examples/functions/seq, examples/functions/seq2, ++ examples/functions/shcat, examples/functions/shcat2, ++ examples/functions/sort-pos-params, examples/functions/substr, ++ examples/functions/substr2, examples/functions/term, ++ examples/functions/whatis, examples/functions/whence, ++ examples/functions/which, examples/functions/xalias.bash, ++ examples/functions/xfind.bash, examples/loadables/Makefile.in, ++ examples/loadables/README, examples/loadables/basename.c, ++ examples/loadables/cat.c, examples/loadables/cut.c, ++ examples/loadables/dirname.c, examples/loadables/finfo.c, ++ examples/loadables/getconf.c, examples/loadables/getconf.h, ++ examples/loadables/head.c, examples/loadables/hello.c, ++ examples/loadables/id.c, examples/loadables/ln.c, ++ examples/loadables/logname.c, examples/loadables/mkdir.c, ++ examples/loadables/necho.c, examples/loadables/pathchk.c, ++ examples/loadables/print.c, examples/loadables/printenv.c, ++ examples/loadables/push.c, examples/loadables/realpath.c, ++ examples/loadables/rmdir.c, examples/loadables/sleep.c, ++ examples/loadables/sync.c, examples/loadables/tee.c, ++ examples/loadables/template.c, examples/loadables/truefalse.c, ++ examples/loadables/tty.c, examples/loadables/uname.c, ++ examples/loadables/unlink.c, examples/loadables/whoami.c, ++ examples/loadables/perl/Makefile.in, ++ examples/loadables/perl/README, examples/loadables/perl/bperl.c, ++ examples/loadables/perl/iperl.c, examples/misc/aliasconv.bash, ++ examples/misc/aliasconv.sh, examples/misc/cshtobash, ++ examples/misc/suncmd.termcap, examples/obashdb/PERMISSION, ++ examples/obashdb/README, examples/obashdb/bashdb, ++ examples/obashdb/bashdb.fns, examples/obashdb/bashdb.pre, ++ examples/scripts/adventure.sh, examples/scripts/bcsh.sh, ++ examples/scripts/cat.sh, examples/scripts/center, ++ examples/scripts/dd-ex.sh, examples/scripts/fixfiles.bash, ++ examples/scripts/hanoi.bash, examples/scripts/inpath, ++ examples/scripts/krand.bash, examples/scripts/line-input.bash, ++ examples/scripts/nohup.bash, examples/scripts/precedence, ++ examples/scripts/randomcard.bash, examples/scripts/scrollbar, ++ examples/scripts/scrollbar2, examples/scripts/self-repro, ++ examples/scripts/showperm.bash, examples/scripts/shprompt, ++ examples/scripts/spin.bash, examples/scripts/timeout, ++ examples/scripts/vtree2, examples/scripts/vtree3, ++ examples/scripts/vtree3a, examples/scripts/websrv.sh, ++ examples/scripts/xterm_title, examples/scripts/zprintf, ++ examples/scripts.noah/PERMISSION, examples/scripts.noah/README, ++ examples/scripts.noah/aref.bash, ++ examples/scripts.noah/bash.sub.bash, ++ examples/scripts.noah/bash_version.bash, ++ examples/scripts.noah/meta.bash, examples/scripts.noah/mktmp.bash, ++ examples/scripts.noah/number.bash, ++ examples/scripts.noah/prompt.bash, ++ examples/scripts.noah/remap_keys.bash, ++ examples/scripts.noah/require.bash, ++ examples/scripts.noah/send_mail.bash, ++ examples/scripts.noah/shcat.bash, ++ examples/scripts.noah/source.bash, ++ examples/scripts.noah/string.bash, examples/scripts.noah/stty.bash, ++ examples/scripts.noah/y_or_n_p.bash, ++ examples/scripts.v2/PERMISSION, examples/scripts.v2/README, ++ examples/scripts.v2/arc2tarz, examples/scripts.v2/bashrand, ++ examples/scripts.v2/cal2day.bash, examples/scripts.v2/cdhist.bash, ++ examples/scripts.v2/corename, examples/scripts.v2/fman, ++ examples/scripts.v2/frcp, examples/scripts.v2/lowercase, ++ examples/scripts.v2/ncp, examples/scripts.v2/newext, ++ examples/scripts.v2/nmv, examples/scripts.v2/pages, ++ examples/scripts.v2/pf, examples/scripts.v2/pmtop, ++ examples/scripts.v2/ren, examples/scripts.v2/rename, ++ examples/scripts.v2/repeat, examples/scripts.v2/shprof, ++ examples/scripts.v2/untar, examples/scripts.v2/uudec, ++ examples/scripts.v2/uuenc, examples/scripts.v2/vtree, ++ examples/scripts.v2/where, examples/startup-files/Bash_aliases, ++ examples/startup-files/Bash_profile, ++ examples/startup-files/Bashrc.bfox, examples/startup-files/README, ++ examples/startup-files/bash-profile, examples/startup-files/bashrc, ++ examples/startup-files/apple/README, ++ examples/startup-files/apple/aliases, ++ examples/startup-files/apple/bash.defaults, ++ examples/startup-files/apple/environment, ++ examples/startup-files/apple/login, ++ examples/startup-files/apple/logout, ++ examples/startup-files/apple/rc, include/ansi_stdlib.h, ++ include/chartypes.h, include/filecntl.h, include/maxpath.h, ++ include/memalloc.h, include/ocache.h, include/posixdir.h, ++ include/posixjmp.h, include/posixstat.h, include/posixtime.h, ++ include/posixwait.h, include/shmbutil.h, include/shtty.h, ++ include/stdc.h, include/systimes.h, include/typemax.h, ++ include/unionwait.h, lib/glob/Makefile.in, lib/glob/collsyms.h, ++ lib/glob/glob.c, lib/glob/glob.h, lib/glob/glob_loop.c, ++ lib/glob/ndir.h, lib/glob/sm_loop.c, lib/glob/smatch.c, ++ lib/glob/strmatch.c, lib/glob/strmatch.h, lib/glob/xmbsrtowcs.c, ++ lib/glob/doc/Makefile, lib/glob/doc/glob.texi, ++ lib/malloc/Makefile.in, lib/malloc/alloca.c, ++ lib/malloc/getpagesize.h, lib/malloc/i386-alloca.s, ++ lib/malloc/imalloc.h, lib/malloc/malloc.c, lib/malloc/mstats.h, ++ lib/malloc/shmalloc.h, lib/malloc/stats.c, lib/malloc/stub.c, ++ lib/malloc/table.c, lib/malloc/table.h, lib/malloc/trace.c, ++ lib/malloc/watch.c, lib/malloc/watch.h, lib/malloc/x386-alloca.s, ++ lib/malloc/xleaktrace, lib/malloc/xmalloc.c, lib/readline/COPYING, ++ lib/readline/ChangeLog, lib/readline/Makefile.in, ++ lib/readline/README, lib/readline/STANDALONE, ++ lib/readline/ansi_stdlib.h, lib/readline/bind.c, ++ lib/readline/callback.c, lib/readline/chardefs.h, ++ lib/readline/compat.c, lib/readline/complete.c, ++ lib/readline/display.c, lib/readline/emacs_keymap.c, ++ lib/readline/funmap.c, lib/readline/histexpand.c, ++ lib/readline/histfile.c, lib/readline/histlib.h, ++ lib/readline/history.c, lib/readline/history.h, ++ lib/readline/histsearch.c, lib/readline/input.c, ++ lib/readline/isearch.c, lib/readline/keymaps.c, ++ lib/readline/keymaps.h, lib/readline/kill.c, lib/readline/macro.c, ++ lib/readline/mbutil.c, lib/readline/misc.c, lib/readline/nls.c, ++ lib/readline/parens.c, lib/readline/posixdir.h, ++ lib/readline/posixjmp.h, lib/readline/posixstat.h, ++ lib/readline/readline.c, lib/readline/readline.h, ++ lib/readline/rlconf.h, lib/readline/rldefs.h, ++ lib/readline/rlmbutil.h, lib/readline/rlprivate.h, ++ lib/readline/rlshell.h, lib/readline/rlstdc.h, ++ lib/readline/rltty.c, lib/readline/rltty.h, ++ lib/readline/rltypedefs.h, lib/readline/rlwinsize.h, ++ lib/readline/savestring.c, lib/readline/search.c, ++ lib/readline/shell.c, lib/readline/signals.c, lib/readline/tcap.h, ++ lib/readline/terminal.c, lib/readline/text.c, lib/readline/tilde.c, ++ lib/readline/tilde.h, lib/readline/undo.c, lib/readline/util.c, ++ lib/readline/vi_keymap.c, lib/readline/vi_mode.c, ++ lib/readline/xmalloc.c, lib/readline/xmalloc.h, ++ lib/readline/doc/Makefile, lib/readline/doc/hist.texinfo, ++ lib/readline/doc/hstech.texinfo, lib/readline/doc/hsuser.texinfo, ++ lib/readline/doc/manvers.texinfo, lib/readline/doc/rlman.texinfo, ++ lib/readline/doc/rltech.texinfo, lib/readline/doc/rluser.texinfo, ++ lib/readline/doc/rluserman.texinfo, lib/readline/examples/Inputrc, ++ lib/readline/examples/Makefile, lib/readline/examples/excallback.c, ++ lib/readline/examples/fileman.c, lib/readline/examples/histexamp.c, ++ lib/readline/examples/manexamp.c, lib/readline/examples/rl.c, ++ lib/readline/examples/rlcat.c, lib/readline/examples/rltest.c, ++ lib/sh/Makefile.in, lib/sh/clktck.c, lib/sh/clock.c, ++ lib/sh/fmtullong.c, lib/sh/fmtulong.c, lib/sh/fmtumax.c, ++ lib/sh/getcwd.c, lib/sh/getenv.c, lib/sh/inet_aton.c, ++ lib/sh/itos.c, lib/sh/mailstat.c, lib/sh/makepath.c, ++ lib/sh/memset.c, lib/sh/mktime.c, lib/sh/netconn.c, ++ lib/sh/netopen.c, lib/sh/oslib.c, lib/sh/pathcanon.c, ++ lib/sh/pathphys.c, lib/sh/rename.c, lib/sh/setlinebuf.c, ++ lib/sh/shquote.c, lib/sh/shtty.c, lib/sh/snprintf.c, ++ lib/sh/spell.c, lib/sh/strcasecmp.c, lib/sh/strerror.c, ++ lib/sh/strftime.c, lib/sh/strindex.c, lib/sh/stringlist.c, ++ lib/sh/stringvec.c, lib/sh/strpbrk.c, lib/sh/strtod.c, ++ lib/sh/strtoimax.c, lib/sh/strtol.c, lib/sh/strtoll.c, ++ lib/sh/strtoul.c, lib/sh/strtoull.c, lib/sh/strtoumax.c, ++ lib/sh/strtrans.c, lib/sh/times.c, lib/sh/timeval.c, ++ lib/sh/tmpfile.c, lib/sh/vprint.c, lib/sh/xstrchr.c, ++ lib/sh/zcatfd.c, lib/sh/zread.c, lib/sh/zwrite.c, ++ lib/termcap/Makefile.in, lib/termcap/ltcap.h, ++ lib/termcap/termcap.c, lib/termcap/termcap.h, lib/termcap/tparam.c, ++ lib/termcap/version.c, lib/termcap/grot/COPYING, ++ lib/termcap/grot/ChangeLog, lib/termcap/grot/INSTALL, ++ lib/termcap/grot/Makefile.in, lib/termcap/grot/NEWS, ++ lib/termcap/grot/README, lib/termcap/grot/configure, ++ lib/termcap/grot/configure.in, lib/termcap/grot/termcap.info, ++ lib/termcap/grot/termcap.info-1, lib/termcap/grot/termcap.info-2, ++ lib/termcap/grot/termcap.info-3, lib/termcap/grot/termcap.info-4, ++ lib/termcap/grot/termcap.texi, lib/termcap/grot/texinfo.tex, ++ lib/tilde/Makefile.in, lib/tilde/README, lib/tilde/shell.c, ++ lib/tilde/tilde.c, lib/tilde/tilde.h, lib/tilde/doc/Makefile, ++ lib/tilde/doc/tilde.texi, support/Makefile.in, support/SYMLINKS, ++ support/bash.xbm, support/bashbug.sh, support/bashversion.c, ++ support/config.guess, support/config.sub, support/fixlinks, ++ support/install.sh, support/man2html.c, support/missing, ++ support/mkclone, support/mkconffiles, support/mkdirs, ++ support/mksignames.c, support/mkversion.sh, support/printenv.c, ++ support/printenv.sh, support/recho.c, support/rlvers.sh, ++ support/shobj-conf, support/texi2dvi, support/texi2html, ++ support/xenix-link.sh, support/zecho.c, tests/README, ++ tests/arith-for.right, tests/arith-for.tests, tests/arith.right, ++ tests/arith.tests, tests/array-at-star, tests/array.right, ++ tests/array.tests, tests/array2.right, tests/braces-tests, ++ tests/braces.right, tests/builtins.right, tests/builtins.tests, ++ tests/builtins1.sub, tests/builtins2.sub, tests/cond.right, ++ tests/cond.tests, tests/cprint.right, tests/cprint.tests, ++ tests/dollar-at-star, tests/dollar.right, tests/dstack.right, ++ tests/dstack.tests, tests/dstack2.right, tests/dstack2.tests, ++ tests/errors.right, tests/errors.tests, tests/exec.right, ++ tests/exec1.sub, tests/exec2.sub, tests/exec3.sub, tests/exec4.sub, ++ tests/exec5.sub, tests/exec6.sub, tests/execscript, ++ tests/exp-tests, tests/exp.right, tests/extglob.right, ++ tests/extglob.tests, tests/extglob2.right, tests/extglob2.tests, ++ tests/func.right, tests/func.tests, tests/func1.sub, ++ tests/func2.sub, tests/func3.sub, tests/getopts.right, ++ tests/getopts.tests, tests/getopts1.sub, tests/getopts2.sub, ++ tests/getopts3.sub, tests/getopts4.sub, tests/getopts5.sub, ++ tests/getopts6.sub, tests/getopts7.sub, tests/glob-test, ++ tests/glob.right, tests/glob1.sub, tests/heredoc.right, ++ tests/heredoc.tests, tests/herestr.right, tests/herestr.tests, ++ tests/histexp.right, tests/histexp.tests, tests/history.list, ++ tests/history.right, tests/history.tests, tests/ifs.right, ++ tests/ifs.tests, tests/input-line.sh, tests/input-line.sub, ++ tests/input.right, tests/invert.right, tests/invert.tests, ++ tests/jobs.right, tests/jobs.tests, tests/jobs1.sub, ++ tests/jobs2.sub, tests/jobs3.sub, tests/more-exp.right, ++ tests/more-exp.tests, tests/new-exp.right, tests/new-exp.tests, ++ tests/new-exp1.sub, tests/new-exp2.sub, tests/new-exp3.sub, ++ tests/nquote.right, tests/nquote.tests, tests/nquote1.right, ++ tests/nquote1.tests, tests/nquote2.right, tests/nquote2.tests, ++ tests/nquote3.right, tests/nquote3.tests, tests/posix2.right, ++ tests/posix2.tests, tests/posixpat.right, tests/posixpat.tests, ++ tests/prec.right, tests/precedence, tests/printf.right, ++ tests/printf.tests, tests/quote.right, tests/quote.tests, ++ tests/read.right, tests/read.tests, tests/read1.sub, ++ tests/read2.sub, tests/read3.sub, tests/read4.sub, ++ tests/redir.right, tests/redir.tests, tests/redir1.sub, ++ tests/redir2.sub, tests/redir3.in1, tests/redir3.in2, ++ tests/redir3.sub, tests/redir4.in1, tests/redir4.sub, ++ tests/redir5.sub, tests/rhs-exp.right, tests/rhs-exp.tests, ++ tests/rsh.right, tests/rsh.tests, tests/run-all, tests/run-arith, ++ tests/run-arith-for, tests/run-array, tests/run-array2, ++ tests/run-braces, tests/run-builtins, tests/run-cond, ++ tests/run-cprint, tests/run-dirstack, tests/run-dollars, ++ tests/run-errors, tests/run-execscript, tests/run-exp-tests, ++ tests/run-extglob, tests/run-extglob2, tests/run-func, ++ tests/run-getopts, tests/run-glob-test, tests/run-heredoc, ++ tests/run-herestr, tests/run-histexpand, tests/run-history, ++ tests/run-ifs, tests/run-input-test, tests/run-invert, ++ tests/run-jobs, tests/run-minimal, tests/run-more-exp, ++ tests/run-new-exp, tests/run-nquote, tests/run-nquote1, ++ tests/run-nquote2, tests/run-nquote3, tests/run-posix2, ++ tests/run-posixpat, tests/run-precedence, tests/run-printf, ++ tests/run-quote, tests/run-read, tests/run-redir, ++ tests/run-rhs-exp, tests/run-rsh, tests/run-set-e, tests/run-shopt, ++ tests/run-strip, tests/run-test, tests/run-tilde, tests/run-trap, ++ tests/run-type, tests/run-varenv, tests/set-e-test, ++ tests/set-e.right, tests/shopt.right, tests/shopt.tests, ++ tests/source1.sub, tests/source2.sub, tests/source3.sub, ++ tests/source4.sub, tests/source5.sub, tests/strip.right, ++ tests/strip.tests, tests/test.right, tests/test.tests, ++ tests/tilde-tests, tests/tilde.right, tests/trap.right, ++ tests/trap.tests, tests/trap1.sub, tests/trap2.sub, ++ tests/trap2a.sub, tests/type.right, tests/type.tests, ++ tests/varenv.right, tests/varenv.sh, tests/varenv1.sub, ++ tests/varenv2.sub, tests/version, tests/version.mini, ++ tests/misc/dev-tcp.tests, tests/misc/perf-script, ++ tests/misc/perftest, tests/misc/read-nchars.tests, ++ tests/misc/redir-t2.sh, tests/misc/run-r2.sh, ++ tests/misc/sigint-1.sh, tests/misc/sigint-2.sh, ++ tests/misc/sigint-3.sh, tests/misc/sigint-4.sh, ++ tests/misc/test-minus-e.1, tests/misc/test-minus-e.2, ++ tests/misc/wait-bg.tests: Initial revision ++ ++2002-08-20 11:16 rocky ++ ++ * AUTHORS, CHANGES, COMPAT, COPYING, INSTALL, MANIFEST, ++ Makefile.in, NEWS, NOTES, POSIX, RBASH, README, Y2K, alias.c, ++ alias.h, array.c, array.h, arrayfunc.c, arrayfunc.h, bashansi.h, ++ bashhist.c, bashhist.h, bashintl.h, bashjmp.h, bashline.c, ++ bashline.h, bashtypes.h, bracecomp.c, braces.c, builtins.h, ++ command.h, config-bot.h, config-top.h, config.h.in, configure.in, ++ conftypes.h, copy_cmd.c, dispose_cmd.c, dispose_cmd.h, error.c, ++ error.h, eval.c, execute_cmd.c, execute_cmd.h, expr.c, externs.h, ++ findcmd.c, findcmd.h, flags.c, flags.h, general.c, general.h, ++ hashcmd.c, hashcmd.h, hashlib.c, hashlib.h, input.c, input.h, ++ jobs.c, jobs.h, list.c, locale.c, mailcheck.c, mailcheck.h, ++ make_cmd.c, make_cmd.h, mksyntax.c, nojobs.c, parse.y, parser.h, ++ patchlevel.h, pathexp.c, pathexp.h, pcomplete.c, pcomplete.h, ++ pcomplib.c, print_cmd.c, quit.h, redir.c, redir.h, shell.c, ++ shell.h, sig.c, sig.h, siglist.c, siglist.h, stringlib.c, subst.c, ++ subst.h, syntax.h, test.c, test.h, trap.c, trap.h, unwind_prot.c, ++ unwind_prot.h, variables.c, variables.h, version.c, xmalloc.c, ++ xmalloc.h, CWRU/PLATFORMS, CWRU/README, CWRU/changelog, ++ CWRU/mh-folder-comp, CWRU/sh-redir-hack, CWRU/misc/bison, ++ CWRU/misc/errlist.c, CWRU/misc/hpux10-dlfcn.h, ++ CWRU/misc/open-files.c, CWRU/misc/sigs.c, CWRU/misc/sigstat.c, ++ builtins/Makefile.in, builtins/alias.def, builtins/bashgetopt.c, ++ builtins/bashgetopt.h, builtins/bind.def, builtins/break.def, ++ builtins/builtin.def, builtins/cd.def, builtins/colon.def, ++ builtins/command.def, builtins/common.c, builtins/common.h, ++ builtins/complete.def, builtins/declare.def, builtins/echo.def, ++ builtins/enable.def, builtins/eval.def, builtins/evalfile.c, ++ builtins/evalstring.c, builtins/exec.def, builtins/exit.def, ++ builtins/fc.def, builtins/fg_bg.def, builtins/getopt.c, ++ builtins/getopt.h, builtins/getopts.def, builtins/hash.def, ++ builtins/help.def, builtins/history.def, builtins/inlib.def, ++ builtins/jobs.def, builtins/kill.def, builtins/let.def, ++ builtins/mkbuiltins.c, builtins/printf.def, builtins/psize.c, ++ builtins/psize.sh, builtins/pushd.def, builtins/read.def, ++ builtins/reserved.def, builtins/return.def, builtins/set.def, ++ builtins/setattr.def, builtins/shift.def, builtins/shopt.def, ++ builtins/source.def, builtins/suspend.def, builtins/test.def, ++ builtins/times.def, builtins/trap.def, builtins/type.def, ++ builtins/ulimit.def, builtins/umask.def, builtins/wait.def, ++ cross-build/beos-sig.h, cross-build/cygwin32.cache, ++ cross-build/opennt.cache, cross-build/win32sig.h, ++ cross-build/x86-beos.cache, doc/FAQ, doc/INTRO, doc/Makefile.in, ++ doc/README, doc/article.ms, doc/bash.1, doc/bashbug.1, ++ doc/bashref.texi, doc/builtins.1, doc/htmlpost.sh, doc/rbash.1, ++ doc/texinfo.tex, examples/bashdb/PERMISSION, ++ examples/bashdb/bashdb, examples/bashdb/bashdb.el, ++ examples/complete/complete-examples, ++ examples/complete/complete.freebsd, ++ examples/complete/complete.gnu-longopt, ++ examples/complete/complete.ianmac, ++ examples/complete/complete2.ianmac, examples/functions/array-stuff, ++ examples/functions/array-to-string, examples/functions/autoload, ++ examples/functions/autoload.v2, examples/functions/autoload.v3, ++ examples/functions/basename, examples/functions/basename2, ++ examples/functions/coproc.bash, examples/functions/coshell.README, ++ examples/functions/coshell.bash, examples/functions/csh-compat, ++ examples/functions/dirfuncs, examples/functions/dirname, ++ examples/functions/emptydir, examples/functions/exitstat, ++ examples/functions/external, examples/functions/fact, ++ examples/functions/fstty, examples/functions/func, ++ examples/functions/gethtml, examples/functions/getoptx.bash, ++ examples/functions/inetaddr, examples/functions/inpath, ++ examples/functions/isnum.bash, examples/functions/isnum2, ++ examples/functions/isvalidip, examples/functions/jdate.bash, ++ examples/functions/jj.bash, examples/functions/keep, ++ examples/functions/ksh-cd, examples/functions/ksh-compat-test, ++ examples/functions/kshenv, examples/functions/login, ++ examples/functions/lowercase, examples/functions/manpage, ++ examples/functions/mhfold, examples/functions/notify.bash, ++ examples/functions/pathfuncs, examples/functions/recurse, ++ examples/functions/repeat2, examples/functions/repeat3, ++ examples/functions/seq, examples/functions/seq2, ++ examples/functions/shcat, examples/functions/shcat2, ++ examples/functions/sort-pos-params, examples/functions/substr, ++ examples/functions/substr2, examples/functions/term, ++ examples/functions/whatis, examples/functions/whence, ++ examples/functions/which, examples/functions/xalias.bash, ++ examples/functions/xfind.bash, examples/loadables/Makefile.in, ++ examples/loadables/README, examples/loadables/basename.c, ++ examples/loadables/cat.c, examples/loadables/cut.c, ++ examples/loadables/dirname.c, examples/loadables/finfo.c, ++ examples/loadables/getconf.c, examples/loadables/getconf.h, ++ examples/loadables/head.c, examples/loadables/hello.c, ++ examples/loadables/id.c, examples/loadables/ln.c, ++ examples/loadables/logname.c, examples/loadables/mkdir.c, ++ examples/loadables/necho.c, examples/loadables/pathchk.c, ++ examples/loadables/print.c, examples/loadables/printenv.c, ++ examples/loadables/push.c, examples/loadables/realpath.c, ++ examples/loadables/rmdir.c, examples/loadables/sleep.c, ++ examples/loadables/sync.c, examples/loadables/tee.c, ++ examples/loadables/template.c, examples/loadables/truefalse.c, ++ examples/loadables/tty.c, examples/loadables/uname.c, ++ examples/loadables/unlink.c, examples/loadables/whoami.c, ++ examples/loadables/perl/Makefile.in, ++ examples/loadables/perl/README, examples/loadables/perl/bperl.c, ++ examples/loadables/perl/iperl.c, examples/misc/aliasconv.bash, ++ examples/misc/aliasconv.sh, examples/misc/cshtobash, ++ examples/misc/suncmd.termcap, examples/obashdb/PERMISSION, ++ examples/obashdb/README, examples/obashdb/bashdb, ++ examples/obashdb/bashdb.fns, examples/obashdb/bashdb.pre, ++ examples/scripts/adventure.sh, examples/scripts/bcsh.sh, ++ examples/scripts/cat.sh, examples/scripts/center, ++ examples/scripts/dd-ex.sh, examples/scripts/fixfiles.bash, ++ examples/scripts/hanoi.bash, examples/scripts/inpath, ++ examples/scripts/krand.bash, examples/scripts/line-input.bash, ++ examples/scripts/nohup.bash, examples/scripts/precedence, ++ examples/scripts/randomcard.bash, examples/scripts/scrollbar, ++ examples/scripts/scrollbar2, examples/scripts/self-repro, ++ examples/scripts/showperm.bash, examples/scripts/shprompt, ++ examples/scripts/spin.bash, examples/scripts/timeout, ++ examples/scripts/vtree2, examples/scripts/vtree3, ++ examples/scripts/vtree3a, examples/scripts/websrv.sh, ++ examples/scripts/xterm_title, examples/scripts/zprintf, ++ examples/scripts.noah/PERMISSION, examples/scripts.noah/README, ++ examples/scripts.noah/aref.bash, ++ examples/scripts.noah/bash.sub.bash, ++ examples/scripts.noah/bash_version.bash, ++ examples/scripts.noah/meta.bash, examples/scripts.noah/mktmp.bash, ++ examples/scripts.noah/number.bash, ++ examples/scripts.noah/prompt.bash, ++ examples/scripts.noah/remap_keys.bash, ++ examples/scripts.noah/require.bash, ++ examples/scripts.noah/send_mail.bash, ++ examples/scripts.noah/shcat.bash, ++ examples/scripts.noah/source.bash, ++ examples/scripts.noah/string.bash, examples/scripts.noah/stty.bash, ++ examples/scripts.noah/y_or_n_p.bash, ++ examples/scripts.v2/PERMISSION, examples/scripts.v2/README, ++ examples/scripts.v2/arc2tarz, examples/scripts.v2/bashrand, ++ examples/scripts.v2/cal2day.bash, examples/scripts.v2/cdhist.bash, ++ examples/scripts.v2/corename, examples/scripts.v2/fman, ++ examples/scripts.v2/frcp, examples/scripts.v2/lowercase, ++ examples/scripts.v2/ncp, examples/scripts.v2/newext, ++ examples/scripts.v2/nmv, examples/scripts.v2/pages, ++ examples/scripts.v2/pf, examples/scripts.v2/pmtop, ++ examples/scripts.v2/ren, examples/scripts.v2/rename, ++ examples/scripts.v2/repeat, examples/scripts.v2/shprof, ++ examples/scripts.v2/untar, examples/scripts.v2/uudec, ++ examples/scripts.v2/uuenc, examples/scripts.v2/vtree, ++ examples/scripts.v2/where, examples/startup-files/Bash_aliases, ++ examples/startup-files/Bash_profile, ++ examples/startup-files/Bashrc.bfox, examples/startup-files/README, ++ examples/startup-files/bash-profile, examples/startup-files/bashrc, ++ examples/startup-files/apple/README, ++ examples/startup-files/apple/aliases, ++ examples/startup-files/apple/bash.defaults, ++ examples/startup-files/apple/environment, ++ examples/startup-files/apple/login, ++ examples/startup-files/apple/logout, ++ examples/startup-files/apple/rc, include/ansi_stdlib.h, ++ include/chartypes.h, include/filecntl.h, include/maxpath.h, ++ include/memalloc.h, include/ocache.h, include/posixdir.h, ++ include/posixjmp.h, include/posixstat.h, include/posixtime.h, ++ include/posixwait.h, include/shmbutil.h, include/shtty.h, ++ include/stdc.h, include/systimes.h, include/typemax.h, ++ include/unionwait.h, lib/glob/Makefile.in, lib/glob/collsyms.h, ++ lib/glob/glob.c, lib/glob/glob.h, lib/glob/glob_loop.c, ++ lib/glob/ndir.h, lib/glob/sm_loop.c, lib/glob/smatch.c, ++ lib/glob/strmatch.c, lib/glob/strmatch.h, lib/glob/xmbsrtowcs.c, ++ lib/glob/doc/Makefile, lib/glob/doc/glob.texi, ++ lib/malloc/Makefile.in, lib/malloc/alloca.c, ++ lib/malloc/getpagesize.h, lib/malloc/i386-alloca.s, ++ lib/malloc/imalloc.h, lib/malloc/malloc.c, lib/malloc/mstats.h, ++ lib/malloc/shmalloc.h, lib/malloc/stats.c, lib/malloc/stub.c, ++ lib/malloc/table.c, lib/malloc/table.h, lib/malloc/trace.c, ++ lib/malloc/watch.c, lib/malloc/watch.h, lib/malloc/x386-alloca.s, ++ lib/malloc/xleaktrace, lib/malloc/xmalloc.c, lib/readline/COPYING, ++ lib/readline/ChangeLog, lib/readline/Makefile.in, ++ lib/readline/README, lib/readline/STANDALONE, ++ lib/readline/ansi_stdlib.h, lib/readline/bind.c, ++ lib/readline/callback.c, lib/readline/chardefs.h, ++ lib/readline/compat.c, lib/readline/complete.c, ++ lib/readline/display.c, lib/readline/emacs_keymap.c, ++ lib/readline/funmap.c, lib/readline/histexpand.c, ++ lib/readline/histfile.c, lib/readline/histlib.h, ++ lib/readline/history.c, lib/readline/history.h, ++ lib/readline/histsearch.c, lib/readline/input.c, ++ lib/readline/isearch.c, lib/readline/keymaps.c, ++ lib/readline/keymaps.h, lib/readline/kill.c, lib/readline/macro.c, ++ lib/readline/mbutil.c, lib/readline/misc.c, lib/readline/nls.c, ++ lib/readline/parens.c, lib/readline/posixdir.h, ++ lib/readline/posixjmp.h, lib/readline/posixstat.h, ++ lib/readline/readline.c, lib/readline/readline.h, ++ lib/readline/rlconf.h, lib/readline/rldefs.h, ++ lib/readline/rlmbutil.h, lib/readline/rlprivate.h, ++ lib/readline/rlshell.h, lib/readline/rlstdc.h, ++ lib/readline/rltty.c, lib/readline/rltty.h, ++ lib/readline/rltypedefs.h, lib/readline/rlwinsize.h, ++ lib/readline/savestring.c, lib/readline/search.c, ++ lib/readline/shell.c, lib/readline/signals.c, lib/readline/tcap.h, ++ lib/readline/terminal.c, lib/readline/text.c, lib/readline/tilde.c, ++ lib/readline/tilde.h, lib/readline/undo.c, lib/readline/util.c, ++ lib/readline/vi_keymap.c, lib/readline/vi_mode.c, ++ lib/readline/xmalloc.c, lib/readline/xmalloc.h, ++ lib/readline/doc/Makefile, lib/readline/doc/hist.texinfo, ++ lib/readline/doc/hstech.texinfo, lib/readline/doc/hsuser.texinfo, ++ lib/readline/doc/manvers.texinfo, lib/readline/doc/rlman.texinfo, ++ lib/readline/doc/rltech.texinfo, lib/readline/doc/rluser.texinfo, ++ lib/readline/doc/rluserman.texinfo, lib/readline/examples/Inputrc, ++ lib/readline/examples/Makefile, lib/readline/examples/excallback.c, ++ lib/readline/examples/fileman.c, lib/readline/examples/histexamp.c, ++ lib/readline/examples/manexamp.c, lib/readline/examples/rl.c, ++ lib/readline/examples/rlcat.c, lib/readline/examples/rltest.c, ++ lib/sh/Makefile.in, lib/sh/clktck.c, lib/sh/clock.c, ++ lib/sh/fmtullong.c, lib/sh/fmtulong.c, lib/sh/fmtumax.c, ++ lib/sh/getcwd.c, lib/sh/getenv.c, lib/sh/inet_aton.c, ++ lib/sh/itos.c, lib/sh/mailstat.c, lib/sh/makepath.c, ++ lib/sh/memset.c, lib/sh/mktime.c, lib/sh/netconn.c, ++ lib/sh/netopen.c, lib/sh/oslib.c, lib/sh/pathcanon.c, ++ lib/sh/pathphys.c, lib/sh/rename.c, lib/sh/setlinebuf.c, ++ lib/sh/shquote.c, lib/sh/shtty.c, lib/sh/snprintf.c, ++ lib/sh/spell.c, lib/sh/strcasecmp.c, lib/sh/strerror.c, ++ lib/sh/strftime.c, lib/sh/strindex.c, lib/sh/stringlist.c, ++ lib/sh/stringvec.c, lib/sh/strpbrk.c, lib/sh/strtod.c, ++ lib/sh/strtoimax.c, lib/sh/strtol.c, lib/sh/strtoll.c, ++ lib/sh/strtoul.c, lib/sh/strtoull.c, lib/sh/strtoumax.c, ++ lib/sh/strtrans.c, lib/sh/times.c, lib/sh/timeval.c, ++ lib/sh/tmpfile.c, lib/sh/vprint.c, lib/sh/xstrchr.c, ++ lib/sh/zcatfd.c, lib/sh/zread.c, lib/sh/zwrite.c, ++ lib/termcap/Makefile.in, lib/termcap/ltcap.h, ++ lib/termcap/termcap.c, lib/termcap/termcap.h, lib/termcap/tparam.c, ++ lib/termcap/version.c, lib/termcap/grot/COPYING, ++ lib/termcap/grot/ChangeLog, lib/termcap/grot/INSTALL, ++ lib/termcap/grot/Makefile.in, lib/termcap/grot/NEWS, ++ lib/termcap/grot/README, lib/termcap/grot/configure, ++ lib/termcap/grot/configure.in, lib/termcap/grot/termcap.info, ++ lib/termcap/grot/termcap.info-1, lib/termcap/grot/termcap.info-2, ++ lib/termcap/grot/termcap.info-3, lib/termcap/grot/termcap.info-4, ++ lib/termcap/grot/termcap.texi, lib/termcap/grot/texinfo.tex, ++ lib/tilde/Makefile.in, lib/tilde/README, lib/tilde/shell.c, ++ lib/tilde/tilde.c, lib/tilde/tilde.h, lib/tilde/doc/Makefile, ++ lib/tilde/doc/tilde.texi, support/Makefile.in, support/SYMLINKS, ++ support/bash.xbm, support/bashbug.sh, support/bashversion.c, ++ support/config.guess, support/config.sub, support/fixlinks, ++ support/install.sh, support/man2html.c, support/missing, ++ support/mkclone, support/mkconffiles, support/mkdirs, ++ support/mksignames.c, support/mkversion.sh, support/printenv.c, ++ support/printenv.sh, support/recho.c, support/rlvers.sh, ++ support/shobj-conf, support/texi2dvi, support/texi2html, ++ support/xenix-link.sh, support/zecho.c, tests/README, ++ tests/arith-for.right, tests/arith-for.tests, tests/arith.right, ++ tests/arith.tests, tests/array-at-star, tests/array.right, ++ tests/array.tests, tests/array2.right, tests/braces-tests, ++ tests/braces.right, tests/builtins.right, tests/builtins.tests, ++ tests/builtins1.sub, tests/builtins2.sub, tests/cond.right, ++ tests/cond.tests, tests/cprint.right, tests/cprint.tests, ++ tests/dollar-at-star, tests/dollar.right, tests/dstack.right, ++ tests/dstack.tests, tests/dstack2.right, tests/dstack2.tests, ++ tests/errors.right, tests/errors.tests, tests/exec.right, ++ tests/exec1.sub, tests/exec2.sub, tests/exec3.sub, tests/exec4.sub, ++ tests/exec5.sub, tests/exec6.sub, tests/execscript, ++ tests/exp-tests, tests/exp.right, tests/extglob.right, ++ tests/extglob.tests, tests/extglob2.right, tests/extglob2.tests, ++ tests/func.right, tests/func.tests, tests/func1.sub, ++ tests/func2.sub, tests/func3.sub, tests/getopts.right, ++ tests/getopts.tests, tests/getopts1.sub, tests/getopts2.sub, ++ tests/getopts3.sub, tests/getopts4.sub, tests/getopts5.sub, ++ tests/getopts6.sub, tests/getopts7.sub, tests/glob-test, ++ tests/glob.right, tests/glob1.sub, tests/heredoc.right, ++ tests/heredoc.tests, tests/herestr.right, tests/herestr.tests, ++ tests/histexp.right, tests/histexp.tests, tests/history.list, ++ tests/history.right, tests/history.tests, tests/ifs.right, ++ tests/ifs.tests, tests/input-line.sh, tests/input-line.sub, ++ tests/input.right, tests/invert.right, tests/invert.tests, ++ tests/jobs.right, tests/jobs.tests, tests/jobs1.sub, ++ tests/jobs2.sub, tests/jobs3.sub, tests/more-exp.right, ++ tests/more-exp.tests, tests/new-exp.right, tests/new-exp.tests, ++ tests/new-exp1.sub, tests/new-exp2.sub, tests/new-exp3.sub, ++ tests/nquote.right, tests/nquote.tests, tests/nquote1.right, ++ tests/nquote1.tests, tests/nquote2.right, tests/nquote2.tests, ++ tests/nquote3.right, tests/nquote3.tests, tests/posix2.right, ++ tests/posix2.tests, tests/posixpat.right, tests/posixpat.tests, ++ tests/prec.right, tests/precedence, tests/printf.right, ++ tests/printf.tests, tests/quote.right, tests/quote.tests, ++ tests/read.right, tests/read.tests, tests/read1.sub, ++ tests/read2.sub, tests/read3.sub, tests/read4.sub, ++ tests/redir.right, tests/redir.tests, tests/redir1.sub, ++ tests/redir2.sub, tests/redir3.in1, tests/redir3.in2, ++ tests/redir3.sub, tests/redir4.in1, tests/redir4.sub, ++ tests/redir5.sub, tests/rhs-exp.right, tests/rhs-exp.tests, ++ tests/rsh.right, tests/rsh.tests, tests/run-all, tests/run-arith, ++ tests/run-arith-for, tests/run-array, tests/run-array2, ++ tests/run-braces, tests/run-builtins, tests/run-cond, ++ tests/run-cprint, tests/run-dirstack, tests/run-dollars, ++ tests/run-errors, tests/run-execscript, tests/run-exp-tests, ++ tests/run-extglob, tests/run-extglob2, tests/run-func, ++ tests/run-getopts, tests/run-glob-test, tests/run-heredoc, ++ tests/run-herestr, tests/run-histexpand, tests/run-history, ++ tests/run-ifs, tests/run-input-test, tests/run-invert, ++ tests/run-jobs, tests/run-minimal, tests/run-more-exp, ++ tests/run-new-exp, tests/run-nquote, tests/run-nquote1, ++ tests/run-nquote2, tests/run-nquote3, tests/run-posix2, ++ tests/run-posixpat, tests/run-precedence, tests/run-printf, ++ tests/run-quote, tests/run-read, tests/run-redir, ++ tests/run-rhs-exp, tests/run-rsh, tests/run-set-e, tests/run-shopt, ++ tests/run-strip, tests/run-test, tests/run-tilde, tests/run-trap, ++ tests/run-type, tests/run-varenv, tests/set-e-test, ++ tests/set-e.right, tests/shopt.right, tests/shopt.tests, ++ tests/source1.sub, tests/source2.sub, tests/source3.sub, ++ tests/source4.sub, tests/source5.sub, tests/strip.right, ++ tests/strip.tests, tests/test.right, tests/test.tests, ++ tests/tilde-tests, tests/tilde.right, tests/trap.right, ++ tests/trap.tests, tests/trap1.sub, tests/trap2.sub, ++ tests/trap2a.sub, tests/type.right, tests/type.tests, ++ tests/varenv.right, tests/varenv.sh, tests/varenv1.sub, ++ tests/varenv2.sub, tests/version, tests/version.mini, ++ tests/misc/dev-tcp.tests, tests/misc/perf-script, ++ tests/misc/perftest, tests/misc/read-nchars.tests, ++ tests/misc/redir-t2.sh, tests/misc/run-r2.sh, ++ tests/misc/sigint-1.sh, tests/misc/sigint-2.sh, ++ tests/misc/sigint-3.sh, tests/misc/sigint-4.sh, ++ tests/misc/test-minus-e.1, tests/misc/test-minus-e.2, ++ tests/misc/wait-bg.tests: Imported sources ++ ++2000-04-03 08:54 rocky ++ ++ * debugger/bashdb-main.inc: Typo. Cut over to version control id. ++ ++2000-04-02 23:26 rocky ++ ++ * debugger/: MANIFEST, PERMISSION, README, bashdb, bashdb-main.inc, ++ bashdb.el: Initial revision ++ ++2000-04-02 23:26 rocky ++ ++ * debugger/: MANIFEST, PERMISSION, README, bashdb, bashdb-main.inc, ++ bashdb.el: Imported sources ++ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/INSTALL cvs/INSTALL +--- bash-2.05b/INSTALL 2002-06-26 23:09:18.000000000 +0200 ++++ cvs/INSTALL 2002-09-19 03:02:08.000000000 +0200 +@@ -1,9 +1,42 @@ + Basic Installation + ================== + +-These are installation instructions for Bash. ++These are installation instructions for Bash. + +-The simplest way to compile Bash is: ++To get the most recent version of this software from CVS anonymously... ++ ++ cvs -d:pserver:anonymous@cvs.bashdb.sourceforge.net:/cvsroot/bashdb login ++ ++when asked for a password just hit the or key. ++ ++ cd directory-of-place-you-want-this-to-be-put-under ++ ++The command below will create a directory underneath this. ++ ++ cvs -z3 -d:pserver:anonymous@cvs.bashdb.sourceforge.net:/cvsroot/bashdb checkout bashdb ++ ++After this you need to run either bootstrap.sh or autogen.sh to create ++some files which are used in the build process. This needs to be done ++only once. Use bootstrap.sh if you don't want to run configure; this ++might be the case in cross-compiling. autogen.sh runs bootstrap.sh and ++then runs configure. ++ ++If you have already have the CVS files checked out (i.e. you did the ++above once), but now want to just receive recent updates... ++ ++ cd directory-where-above-checkout-put-files ++ ++Note: this is not the directory you were in but one directory down ++from that; probably something-.../bashdb. ++ ++ cvs -d:pserver:anonymous@cvs.bashdb.sourceforge.net:/cvsroot/bashdb login ++ ++when asked for a password just hit the or key. ++ ++ cvs -z3 -d:pserver:anonymous@cvs.bashdb.sourceforge.net:/cvsroot/bashdb update . ++ ++If you have obtained this as a source tarball, or have followed the ++instructions above, the simplest way to compile Bash is: + + 1. `cd' to the directory containing the source code and type + `./configure' to configure Bash for your system. If you're using +@@ -11,16 +44,21 @@ + ./configure' instead to prevent `csh' from trying to execute + `configure' itself. + ++ To see ways to customization configuration, type ++ ./configure --help. ++ + Running `configure' takes some time. While running, it prints + messages telling which features it is checking for. + + 2. Type `make' to compile Bash and build the `bashbug' bug reporting + script. + +- 3. Optionally, type `make tests' to run the Bash test suite. ++ 3. Optionally, type `make check' to run the Bash test suite. + +- 4. Type `make install' to install `bash' and `bashbug'. This will +- also install the manual pages and Info file. ++ 4. Type `make install' to install `bash', 'bashdb' and `bashbug'. ++ This will also install the manual pages and Info file. Please ++ also see debugger/INSTALL for information on installation of the ++ debugger portion. + + The `configure' shell script attempts to guess correct values for + various system-dependent variables used during compilation. It uses +@@ -44,8 +82,8 @@ + + If you need to do unusual things to compile Bash, please try to figure + out how `configure' could check whether or not to do them, and mail +-diffs or instructions to so they can be +-considered for the next release. ++diffs or instructions to so they ++can be considered for the next release. + + The file `configure.in' is used to create `configure' by a program + called Autoconf. You only need `configure.in' if you want to change it +@@ -171,6 +209,10 @@ + Print the version of Autoconf used to generate the `configure' + script, and exit. + ++`--with-lispdir' Override the default lisp directory to use for installing ++ the Emacs bashdb scripts ++ ++ + `configure' also accepts some other, not widely used, boilerplate + options. `configure --help' prints the complete list. + +@@ -365,3 +407,5 @@ + not meant to be changed; beware of the consequences if you do. Read + the comments associated with each definition for more information about + its effect. ++ ++$Id: INSTALL,v 1.3 2002/09/19 01:02:08 rockyb Exp $ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/MANIFEST cvs/MANIFEST +--- bash-2.05b/MANIFEST 2002-05-13 19:09:00.000000000 +0200 ++++ cvs/MANIFEST 2003-09-01 21:57:21.000000000 +0200 +@@ -8,6 +8,10 @@ + CWRU/misc d + builtins d + cross-build d ++debugger d ++debugger/emacs d ++debugger/doc d ++debugger/test d + doc d + examples d + examples/bashdb d +@@ -38,6 +42,7 @@ + support d + tests d + tests/misc d ++ChangeLog f + CHANGES f + COMPAT f + COPYING f +@@ -56,6 +61,7 @@ + config-top.h f + config-bot.h f + config.h.in f ++bashlocal.m4 f + aclocal.m4 f + array.c f + arrayfunc.c f +@@ -140,7 +146,7 @@ + redir.h f + bashtypes.h f + mailcheck.h f +-pathnames.h f ++pathnames.h.in f + xmalloc.h f + y.tab.c f + y.tab.h f +@@ -150,6 +156,7 @@ + builtins/bind.def f + builtins/break.def f + builtins/builtin.def f ++builtins/caller.def f + builtins/cd.def f + builtins/colon.def f + builtins/command.def f +@@ -312,7 +319,6 @@ + lib/readline/rlstdc.h f + lib/readline/rlprivate.h f + lib/readline/xmalloc.h f +-lib/readline/doc/Makefile f + lib/readline/doc/manvers.texinfo f + lib/readline/doc/rlman.texinfo f + lib/readline/doc/rltech.texinfo f +@@ -417,6 +423,182 @@ + CWRU/changelog f + CWRU/sh-redir-hack f + CWRU/mh-folder-comp f ++debugger/AUTHORS f ++debugger/CHANGES f ++debugger/COPYING f ++debugger/ChangeLog f ++debugger/INSTALL f ++debugger/Makefile f ++debugger/Makefile.am f ++debugger/Makefile.in f ++debugger/NEWS f ++debugger/README f ++debugger/THANKS f ++debugger/TODO f ++debugger/aclocal.m4 f ++debugger/acinclude.m4 f ++debugger/bashdb f 755 ++debugger/bashdb.in f 755 ++debugger/dbg-brk.inc f ++debugger/dbg-cmds.inc f ++debugger/dbg-file.inc f ++debugger/dbg-fns.inc f ++debugger/dbg-help.inc f ++debugger/dbg-hist.inc f ++debugger/dbg-init.inc f ++debugger/dbg-io.inc f ++debugger/dbg-list.inc f ++debugger/dbg-main.inc.in f ++debugger/dbg-pre.inc.in f ++debugger/dbg-set.inc f ++debugger/dbg-set-d-vars.inc f ++debugger/dbg-sig.inc f ++debugger/dbg-sig-ret.inc f ++debugger/dbg-stack.inc f ++debugger/configure f 755 ++debugger/configure.ac f ++debugger/elisp-comp f ++debugger/install-sh f 755 ++debugger/missing f 755 ++debugger/mkinstalldirs f 755 ++debugger/doc/Makefile f ++debugger/doc/Makefile.am f ++debugger/doc/Makefile.in f ++debugger/doc/bashdb.1 f ++debugger/doc/bashdb.info f ++debugger/doc/bashdb.info-1 f ++debugger/doc/bashdb.info-2 f ++debugger/doc/bashdb.info-3 f ++debugger/doc/bashdb.info-4 f ++debugger/doc/bashdb.texi f ++debugger/doc/bashdb.html f ++debugger/doc/bashdb-man.html f ++debugger/doc/bashdb-man.pod f ++debugger/doc/fdl.texi f ++debugger/doc/gpl.texi f ++debugger/doc/mdate-sh f ++debugger/doc/version.texi f ++debugger/doc/version.texi.in f ++debugger/emacs/Makefile f ++debugger/emacs/Makefile.am f ++debugger/emacs/Makefile.in f ++debugger/emacs/gud.el f ++debugger/emacs/gud.el.diff f ++debugger/emacs/bashdb.el f ++debugger/test/Makefile f ++debugger/test/Makefile.am f ++debugger/test/Makefile.in f ++debugger/test/check_common f ++debugger/test/check_common.in f ++debugger/test/dbg-test1.sh f 755 ++debugger/test/dbg-test1.sub f ++debugger/test/dbg-test2.sh f 755 ++debugger/test/action.cmd f ++debugger/test/action.right f ++debugger/test/action.tests f 755 ++debugger/test/brkpt1.cmd f ++debugger/test/brkpt1.right f ++debugger/test/brkpt1.tests f 755 ++debugger/test/brkpt2.cmd f ++debugger/test/brkpt2.right f ++debugger/test/brkpt2.tests f 755 ++debugger/test/brkpt3.cmd f ++debugger/test/brkpt3.right f ++debugger/test/brkpt3.tests f 755 ++debugger/test/bugIFS f ++debugger/test/bugIFS.cmd f ++debugger/test/bugIFS.right f ++debugger/test/bugIFS.sh f 755 ++debugger/test/bugIFS.tests f 755 ++debugger/test/debug.cmd f ++debugger/test/debug2.cmd f ++debugger/test/debug.right f ++debugger/test/debug.sh f 755 ++debugger/test/debug.tests f 755 ++debugger/test/display.cmd f ++debugger/test/display.right f ++debugger/test/display.tests f 755 ++debugger/test/finish.cmd f ++debugger/test/finish.right f ++debugger/test/finish.tests f 755 ++debugger/test/list.cmd f ++debugger/test/list.right f ++debugger/test/list.tests f 755 ++debugger/test/misc.cmd f ++debugger/test/misc.right f ++debugger/test/misc-output.right f ++debugger/test/multi.sh f 755 ++debugger/test/multi1.tests f ++debugger/test/multi1.cmd f ++debugger/test/multi1.right f ++debugger/test/multi2.tests f ++debugger/test/multi2.cmd f ++debugger/test/multi2.right f ++debugger/test/multi3.tests f ++debugger/test/multi3.cmd f ++debugger/test/multi3.right f ++debugger/test/multi4.tests f ++debugger/test/multi4.cmd f ++debugger/test/multi4.right f ++debugger/test/misc-output.right f ++debugger/test/misc.tests f 755 ++debugger/test/subshell.sh f 755 ++debugger/test/subshell1.cmd f ++debugger/test/subshell1.right f ++debugger/test/subshell1.tests f 755 ++debugger/test/subshell2.cmd f ++debugger/test/subshell2.right f ++debugger/test/subshell2.tests f 755 ++debugger/test/subshell3.cmd f ++debugger/test/subshell3.right f ++debugger/test/subshell3.tests f 755 ++debugger/test/subshell4.cmd f ++debugger/test/subshell4.right f ++debugger/test/subshell4.tests f 755 ++debugger/test/parm.cmd f ++debugger/test/parm.right f ++debugger/test/parm.sh f 755 ++debugger/test/parm.tests f 755 ++debugger/test/prof1.cmd f ++debugger/test/prof2.cmd f ++debugger/test/search.cmd f ++debugger/test/search.right f ++debugger/test/search.tests f 755 ++debugger/test/sig.cmd f ++debugger/test/sig.right f ++debugger/test/sig.sh f 755 ++debugger/test/sig.tests f 755 ++debugger/test/skip.cmd f ++debugger/test/skip.right f ++debugger/test/skip.sh f 755 ++debugger/test/skip.tests f 755 ++debugger/test/tbreak.cmd f ++debugger/test/tbreak.right f ++debugger/test/tbreak.tests f 755 ++debugger/test/watch1.cmd f ++debugger/test/watch1.right f ++debugger/test/watch1.tests f 755 ++debugger/test/watch2.cmd f ++debugger/test/watch2.right f ++debugger/test/watch2.tests f 755 ++debugger/test/run-action f 755 ++debugger/test/run-all f 755 ++debugger/test/run-brkpt f 755 ++debugger/test/run-bugIFS f 755 ++debugger/test/run-debug f 755 ++debugger/test/run-display f 755 ++debugger/test/run-finish f 755 ++debugger/test/run-list f 755 ++debugger/test/run-misc f 755 ++debugger/test/run-multi f 755 ++debugger/test/run-parm f 755 ++debugger/test/run-search f 755 ++debugger/test/run-sig f 755 ++debugger/test/run-skip f 755 ++debugger/test/run-subshell f 755 ++debugger/test/run-tbreak f 755 ++debugger/test/run-watch1 f 755 ++debugger/test/run-watch2 f 755 + doc/FAQ f + doc/Makefile.in f + doc/bash.1 f +@@ -427,7 +609,13 @@ + doc/INTRO f + doc/texinfo.tex f + doc/bashref.texi f +-doc/bashref.info f ++doc/bashref.dvi f ++doc/bashref.html f ++doc/bash.info f ++doc/version.texi f ++doc/version.texi.in f ++doc/fdl.texi f ++doc/gpl.texi f + doc/article.ms f + doc/htmlpost.sh f 755 + support/Makefile.in f +@@ -441,6 +629,7 @@ + support/mkclone f 755 + support/mkconffiles f 755 + support/mkdirs f 755 ++support/mkdist f 755 + support/mkversion.sh f 755 + support/mksignames.c f + support/bashbug.sh f +@@ -456,8 +645,8 @@ + support/shobj-conf f 755 + support/rlvers.sh f 755 + examples/bashdb/PERMISSION f ++examples/bashdb/README f + examples/bashdb/bashdb f +-examples/bashdb/bashdb.el f + examples/obashdb/PERMISSION f + examples/obashdb/README f + examples/obashdb/bashdb f +@@ -611,10 +800,13 @@ + tests/array2.right f + tests/braces-tests f + tests/braces.right f ++tests/bashdb-bp.tests f 755 + tests/builtins.tests f + tests/builtins.right f + tests/builtins1.sub f + tests/builtins2.sub f ++tests/check_common f ++tests/check_common.in f + tests/source1.sub f + tests/source2.sub f + tests/source3.sub f +@@ -622,15 +814,25 @@ + tests/source5.sub f + tests/cond.tests f + tests/cond.right f +-tests/cprint.tests f ++tests/dbg-fns f ++tests/dbg-fns.dat f ++tests/dbg-is-fn.right f ++tests/dbg-is-fn.tests f 755 ++tests/dbg-support.sub f ++tests/dbg-support.right f ++tests/dbg-support.tests f 755 ++tests/dbg-support2.right f ++tests/dbg-support2.tests f 755 + tests/cprint.right f ++tests/cprint.tests f + tests/dollar-at-star f + tests/dollar.right f + tests/dstack.tests f + tests/dstack.right f + tests/dstack2.tests f + tests/dstack2.right f +-tests/errors.tests f ++tests/errors.sub f ++tests/errors.tests f 755 + tests/errors.right f + tests/execscript f + tests/exec.right f +@@ -663,7 +865,7 @@ + tests/glob-test f + tests/glob1.sub f + tests/glob.right f +-tests/heredoc.tests f ++tests/heredoc.tests f 755 + tests/heredoc.right f + tests/herestr.tests f + tests/herestr.right f +@@ -695,9 +897,9 @@ + tests/nquote.right f + tests/nquote1.tests f + tests/nquote1.right f +-tests/nquote2.tests f ++tests/nquote2.tests f 755 + tests/nquote2.right f +-tests/nquote3.tests f ++tests/nquote3.tests f 755 + tests/nquote3.right f + tests/posix2.tests f + tests/posix2.right f +@@ -709,13 +911,13 @@ + tests/printf.right f + tests/quote.tests f + tests/quote.right f +-tests/read.tests f ++tests/read.tests f 755 + tests/read.right f + tests/read1.sub f + tests/read2.sub f + tests/read3.sub f + tests/read4.sub f +-tests/redir.tests f ++tests/redir.tests f 755 + tests/redir.right f + tests/redir1.sub f + tests/redir2.sub f +@@ -725,77 +927,83 @@ + tests/redir4.sub f + tests/redir4.in1 f + tests/redir5.sub f +-tests/rhs-exp.tests f ++tests/rhs-exp.tests f 755 + tests/rhs-exp.right f + tests/rsh.tests f + tests/rsh.right f +-tests/run-all f +-tests/run-minimal f +-tests/run-arith-for f +-tests/run-arith f +-tests/run-array f +-tests/run-array2 f +-tests/run-braces f +-tests/run-builtins f +-tests/run-cond f +-tests/run-cprint f +-tests/run-dirstack f +-tests/run-dollars f +-tests/run-errors f +-tests/run-execscript f +-tests/run-exp-tests f +-tests/run-extglob f +-tests/run-extglob2 f +-tests/run-func f +-tests/run-getopts f +-tests/run-glob-test f +-tests/run-heredoc f +-tests/run-herestr f +-tests/run-histexpand f +-tests/run-history f +-tests/run-ifs f +-tests/run-input-test f +-tests/run-invert f +-tests/run-jobs f +-tests/run-more-exp f +-tests/run-new-exp f +-tests/run-nquote f +-tests/run-nquote1 f +-tests/run-nquote2 f +-tests/run-nquote3 f +-tests/run-posix2 f +-tests/run-posixpat f +-tests/run-precedence f +-tests/run-printf f +-tests/run-quote f +-tests/run-read f +-tests/run-redir f +-tests/run-rhs-exp f +-tests/run-rsh f +-tests/run-set-e f +-tests/run-shopt f +-tests/run-strip f +-tests/run-test f +-tests/run-tilde f +-tests/run-trap f +-tests/run-type f +-tests/run-varenv f ++tests/run-all f 755 ++tests/run-minimal f 755 ++tests/run-arith-for f 755 ++tests/run-arith f 755 ++tests/run-array f 755 ++tests/run-array2 f 755 ++tests/run-braces f 755 ++tests/run-builtins f 755 ++tests/run-cond f 755 ++tests/run-cprint f 755 ++tests/run-dirstack f 755 ++tests/run-dbg-is-fn f 755 ++tests/run-dbg-support f 755 ++tests/run-dbg-support2 f 755 ++tests/run-dollars f 755 ++tests/run-errors f 755 ++tests/run-execscript f 755 ++tests/run-exp-tests f 755 ++tests/run-extglob f 755 ++tests/run-extglob2 f 755 ++tests/run-func f 755 ++tests/run-getopts f 755 ++tests/run-glob-test f 755 ++tests/run-heredoc f 755 ++tests/run-herestr f 755 ++tests/run-histexpand f 755 ++tests/run-history f 755 ++tests/run-ifs f 755 ++tests/run-input-test f 755 ++tests/run-invert f 755 ++tests/run-jobs f 755 ++tests/run-more-exp f 755 ++tests/run-new-exp f 755 ++tests/run-nquote f 755 ++tests/run-nquote1 f 755 ++tests/run-nquote2 f 755 ++tests/run-nquote3 f 755 ++tests/run-posix2 f 755 ++tests/run-posixpat f 755 ++tests/run-precedence f 755 ++tests/run-printf f 755 ++tests/run-quote f 755 ++tests/run-read f 755 ++tests/run-redir f 755 ++tests/run-rhs-exp f 755 ++tests/run-rsh f 755 ++tests/run-set-e f 755 ++tests/run-set-x f 755 ++tests/run-shopt f 755 ++tests/run-strip f 755 ++tests/run-test f 755 ++tests/run-tilde f 755 ++tests/run-trap f 755 ++tests/run-type f 755 ++tests/run-varenv f 755 + tests/set-e-test f + tests/set-e.right f +-tests/shopt.tests f ++tests/set-x.tests f 755 ++tests/set-x.right f ++tests/shopt.tests f 755 + tests/shopt.right f +-tests/strip.tests f ++tests/strip.tests f 755 + tests/strip.right f +-tests/test.tests f ++tests/test.tests f 755 + tests/test.right f +-tests/tilde-tests f ++tests/tilde-tests f 755 + tests/tilde.right f +-tests/trap.tests f ++tests/trap.tests f 755 + tests/trap.right f + tests/trap1.sub f 755 + tests/trap2.sub f 755 + tests/trap2a.sub f 755 +-tests/type.tests f ++tests/type.tests f 755 + tests/type.right f + tests/varenv.right f + tests/varenv.sh f +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/Makefile.in cvs/Makefile.in +--- bash-2.05b/Makefile.in 2002-05-31 19:44:23.000000000 +0200 ++++ cvs/Makefile.in 2003-09-08 08:40:14.000000000 +0200 +@@ -38,6 +38,7 @@ + man3dir = $(mandir)/$(manpfx)3 + + htmldir = @htmldir@ ++transform = @program_transform_name@ + + # Support an alternate destination root directory for package building + DESTDIR = +@@ -395,6 +396,7 @@ + DEFSRC=$(BUILTIN_SRCDIR) + BUILTIN_ABSSRC=${topdir}/builtins + DEFDIR = $(dot)/builtins ++DEBUGGERDIR = $(dot)/debugger + + BUILTIN_DEFS = $(DEFSRC)/alias.def $(DEFSRC)/bind.def $(DEFSRC)/break.def \ + $(DEFSRC)/builtin.def $(DEFSRC)/cd.def $(DEFSRC)/colon.def \ +@@ -464,7 +466,7 @@ + # Keep GNU Make from exporting the entire environment for small machines. + .NOEXPORT: + +-.made: $(Program) bashbug ++.made: .prebuild $(Program) bashbug doc debugger + @echo "$(Program) last made for a $(Machine) running $(OS)" >.made + + $(Program): .build $(OBJECTS) $(BUILTINS_DEP) $(LIBDEP) +@@ -473,6 +475,14 @@ + ls -l $(Program) + size $(Program) + ++.prebuild: ++ @(oldpwd=`pwd`; cd $(srcdir) &&\ ++ if test "X`pwd`" != "X$(BUILD_DIR)" ; then \ ++ ${CP} $$oldpwd/pathnames.h $(srcdir)/ ; \ ++ ${CP} $$oldpwd/doc/version.texi $(srcdir)/doc/ ; \ ++ ${CP} $$oldpwd/doc/version.texi $(srcdir)/debugger/doc/ ; \ ++ fi) ++ + .build: $(SOURCES) config.h Makefile version.h $(VERSPROG) + @echo + @echo " ***********************************************************" +@@ -498,6 +508,9 @@ + lint: + ${MAKE} ${MFLAGS} CFLAGS='${GCC_LINT_FLAGS}' .made + ++aclocal.m4: bashlocal.m4 ++ aclocal -I @srcdir@ ++ + version.h: $(SOURCES) config.h Makefile + $(SHELL) $(SUPPORT_SRC)mkversion.sh -b -S ${topdir} -s $(RELSTATUS) -d $(Version) -o newversion.h \ + && mv newversion.h version.h +@@ -594,8 +607,8 @@ + # For the justification of the following Makefile rules, see node + # `Automatic Remaking' in GNU Autoconf documentation. + +-Makefile makefile: config.status $(srcdir)/Makefile.in +- CONFIG_FILES=Makefile CONFIG_HEADERS= $(SHELL) ./config.status ++Makefile makefile pathnames.h: config.status $(srcdir)/Makefile.in ++ CONFIG_HEADERS= $(SHELL) ./config.status + + Makefiles makefiles: config.status $(srcdir)/Makefile.in + @for mf in $(CREATED_MAKEFILES); do \ +@@ -627,6 +640,9 @@ + doc documentation: force + @(cd $(DOCDIR) ; $(MAKE) $(MFLAGS) ) + ++debugger: force ++ @( cd $(DEBUGGERDIR) ; $(MAKE) $(MFLAGS) info man html ) ++ + info dvi ps: force + @(cd $(DOCDIR) ; $(MAKE) $(MFLAGS) CFLAGS='$(CCFLAGS)' $@ ) + +@@ -646,13 +662,14 @@ + @${SHELL} $(SUPPORT_SRC)mkdirs $(DESTDIR)$(infodir) + + install: .made installdirs +- $(INSTALL_PROGRAM) $(INSTALLMODE) $(Program) $(DESTDIR)$(bindir)/$(Program) +- $(INSTALL_SCRIPT) $(INSTALLMODE2) bashbug $(DESTDIR)$(bindir)/bashbug ++ $(INSTALL_PROGRAM) $(INSTALLMODE) $(Program) $(DESTDIR)$(bindir)/`echo $(Program) | sed '$(transform)'` ++ $(INSTALL_SCRIPT) $(INSTALLMODE2) bashbug $(DESTDIR)$(bindir)/`echo bashbug | sed '$(transform)'` + -( cd $(DOCDIR) ; $(MAKE) $(MFLAGS) \ + man1dir=$(man1dir) man1ext=$(man1ext) \ + man3dir=$(man3dir) man3ext=$(man3ext) \ + infodir=$(infodir) htmldir=$(htmldir) DESTDIR=$(DESTDIR) $@ ) + -( cd $(DEFDIR) ; $(MAKE) $(MFLAGS) DESTDIR=$(DESTDIR) $@ ) ++ -( cd $(DEBUGGERDIR) ; $(MAKE) $(MFLAGS) DESTDIR=$(DESTDIR) $@ ) + + install-strip: + $(MAKE) $(MFLAGS) INSTALL_PROGRAM='$(INSTALL_PROGRAM) -s' \ +@@ -660,17 +677,21 @@ + DESTDIR=$(DESTDIR) install + + uninstall: .made +- $(RM) $(DESTDIR)$(bindir)/$(Program) $(DESTDIR)$(bindir)/bashbug ++ for p in $(DESTDIR)$(bindir)/$(Program) $(DESTDIR)$(bindir)/bashbug; do \ ++ $(RM) $(DESTDIR)$(bindir)/`echo $$p | sed '$(transform)'`; \ ++ done + -( cd $(DOCDIR) ; $(MAKE) $(MFLAGS) \ + man1dir=$(man1dir) man1ext=$(man1ext) \ + man3dir=$(man3dir) man3ext=$(man3ext) \ + infodir=$(infodir) htmldir=$(htmldir) DESTDIR=$(DESTDIR) $@ ) ++ -( cd $(DEBUGGERDIR) ; $(MAKE) $(MFLAGS) $@ ) + + .PHONY: basic-clean clean realclean maintainer-clean distclean mostlyclean maybe-clean + + basic-clean: + $(RM) $(OBJECTS) $(Program) bashbug +- $(RM) .build .made version.h ++ $(RM) .build .made version.h pathname.h ++ $(RM) *.tar.gz *~ */*~ + + clean: basic-clean + ( cd $(DOCDIR) && $(MAKE) $(MFLAGS) $@ ) +@@ -747,16 +768,16 @@ + @cp $(TESTS_SUPPORT) tests + @( cd $(srcdir)/tests && \ + PATH=$$PATH:$(BUILD_DIR)/tests THIS_SH=$(THIS_SH) $(SHELL) ${TESTSCRIPT} ) ++ @( cd $(srcdir)/debugger && make check ) + + symlinks: + $(SHELL) $(SUPPORT_SRC)fixlinks -s $(srcdir) + + dist: force +- @echo Bash distributions are created using $(srcdir)/support/mkdist. +- @echo Here is a sample of the necessary commands: +- @echo $(Program) $(srcdir)/support/mkdist -m $(srcdir)/MANIFEST -s $(srcdir) -r ${Program} $(Version)-${RELSTATUS} +- @echo tar cf $(Program)-$(Version)-${RELSTATUS}.tar ${Program}-$(Version)-${RELSTATUS} +- @echo gzip $(Program)-$(Version)-${RELSTATUS}.tar ++ $(Program) $(srcdir)/support/mkdist -m $(srcdir)/MANIFEST -s $(srcdir) -r ${Program} $(Version)-${RELSTATUS} ++ tar cf $(Program)-$(Version)-${RELSTATUS}.tar ${Program}-$(Version)-${RELSTATUS} ++ gzip $(Program)-$(Version)-${RELSTATUS}.tar ++ rm -fr $(Program)-$(Version)-${RELSTATUS} + + depend: depends + +@@ -767,6 +788,10 @@ + hashtest: hashlib.c + $(CC) -DTEST_HASHING $(CCFLAGS) -o $@ $(srcdir)/hashlib.c + ++arraytest: array.c xmalloc.o syntax.o lib/malloc/libmalloc.a lib/sh/libsh.a ++ $(CC) -c -DTEST_ARRAY $(CCFLAGS) -o arraytest.o $(srcdir)/array.c ++ $(CC) $(CCFLAGS) -o $@ $(srcdir)/arraytest.o xmalloc.o syntax.o lib/malloc/libmalloc.a lib/sh/libsh.a list.o ++ + ############################ DEPENDENCIES ############################### + + # Files that depend on the definitions in config-top.h, which are not meant +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/README cvs/README +--- bash-2.05b/README 2002-04-04 18:02:55.000000000 +0200 ++++ cvs/README 2002-09-17 13:57:46.000000000 +0200 +@@ -1,15 +1,20 @@ + Introduction + ============ + +-This is GNU Bash, version 2.05b. Bash is the GNU Project's Bourne +-Again SHell, a complete implementation of the POSIX.2 shell spec, +-but also with interactive command line editing, job control on +-architectures that support it, csh-like features such as history +-substitution and brace expansion, and a slew of other features. +-For more information on the features of Bash that are new to this +-type of shell, see the file `doc/bashref.texi'. There is also a +-large Unix-style man page. The man page is the definitive description +-of the shell's features. ++This is GNU Bash, version 2.05b-debugger. A patched version of Bash ++2.05b to provide debugging support and better error reporting. ++ ++Bash is the GNU Project's Bourne Again SHell, a complete ++implementation of the POSIX.2 shell spec, but also with interactive ++command line editing, job control on architectures that support it, ++csh-like features such as history substitution and brace expansion, ++and a slew of other features. For more information on the features of ++Bash that are new to this type of shell, see the file ++`doc/bashref.texi'. There is also a large Unix-style man page. The ++man page is the definitive description of the shell's features. ++ ++For more information on the features of the Bash debugger, see ++doc/bashdb.texi. + + See the file CWRU/POSIX.NOTES for a discussion of how Bash differs + from the POSIX.2 spec and a description of the Bash `posix mode'. +@@ -27,8 +32,13 @@ + `doc/FAQ'. + + To compile Bash, try typing `./configure', then `make'. Bash +-auto-configures the build process, so no further intervention +-should be necessary. Bash builds with `gcc' by default if it is ++auto-configures the build process, so no most intervention is not ++necessary. However one can customize where the binaries should go or ++where to install debugger lisp files. To see the list of ++customizations available, type ./configure --help. ++ ++ ++Bash builds with `gcc' by default if it is + available. If you want to use `cc' instead, type + + CC=cc ./configure +@@ -50,8 +60,21 @@ + Reporting Bugs + ============== + +-Bug reports for bash should be sent to: ++If the problem is in the debugger rather than bash (i.e. you don't ++have the same problem using an unpatched version of Bash, send bug ++reports to: ++ ++ bug-bashdb@sourceforge.net. ++ ++The discussion list `bug-devel@gnu.org' contains developer information ++and discussions of new features or behavior changes that people would ++like. This mailing list is also available as a usenet newsgroup: ++gnu.bash.bug. + ++If the problem is not debugger-related (i.e. you do have the same ++problem using an unmodified version bash), send bug reports to ++ ++Bug reports for bash should be sent to: + bug-bash@gnu.org + + using the `bashbug' program that is built and installed at the same +@@ -88,3 +111,8 @@ + + Chet Ramey + chet@po.cwru.edu ++ ++with modifications for the Bash Debugger by Rocky Bernstein ++rocky@panix.com ++ ++$Id: README,v 1.3 2002/09/17 11:57:46 rockyb Exp $ +Only in bash-2.05b: aclocal.m4 +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/array.c cvs/array.c +--- bash-2.05b/array.c 2002-03-18 19:16:50.000000000 +0100 ++++ cvs/array.c 2002-10-10 14:34:00.000000000 +0200 +@@ -365,8 +365,10 @@ + array_dispose_element(ae) + ARRAY_ELEMENT *ae; + { +- FREE(ae->value); +- free(ae); ++ if (NULL != ae) { ++ FREE(ae->value); ++ free(ae); ++ } + } + + /* +@@ -650,10 +652,67 @@ + } + #endif + ++/* ++ Delete the beginning of array A, subtracting one from all other ++ array element indices. ++ Return the element. The caller should dispose of it. ++*/ ++ARRAY_ELEMENT * ++array_unshift_element(a) ++ARRAY *a; ++{ ++ register ARRAY_ELEMENT *ae, *first; ++ ++ if (!a || array_empty(a)) ++ return((ARRAY_ELEMENT *) NULL); ++ first=element_forw(a->head); ++ first->next->prev = first->prev; ++ first->prev->next = first->next; ++ a->num_elements--; ++ a->max_index--; ++ ++ /* Renumber the element array indices. */ ++ for (ae = element_forw(a->head); ae != a->head; ae = element_forw(ae)) ++ element_index(ae)--; ++ ++ return(first); ++} ++ ++/* ++ Add a new element with index to beginning of array a, adding one to ++ all other array indices. Return the number of elements after adding. ++ */ ++int ++array_shift_element(a, v) ++ARRAY *a; ++char *v; ++{ ++ register ARRAY_ELEMENT *new, *ae, *temp; ++ arrayind_t i=0; ++ ++ if (!a) ++ return(-1); ++ new = array_create_element(0, v); ++ temp=element_forw(a->head); ++ ADD_BEFORE(temp, new); ++ ++ a->num_elements++; ++ a->max_index++; ++ ++ /* ++ * Renumber all the other elements in the array. ++ */ ++ for (ae = element_forw(new); ae != a->head; ae = element_forw(ae)) { ++ element_index(ae) = ++i; ++ } ++ return (a->num_elements); ++} ++ ++ + #if defined (TEST_ARRAY) + /* + * To make a running version, compile -DTEST_ARRAY and link with: +- * xmalloc.o syntax.o lib/malloc/libmalloc.a lib/sh/libsh.a ++ * xmalloc.o syntax.o list.o lib/malloc/libmalloc.a lib/sh/libsh.a + */ + int interrupt_immediately = 0; + +@@ -723,21 +782,6 @@ + return (REVERSE_LIST (wl, WORD_LIST *)); + } + +-GENERIC_LIST * +-list_reverse (list) +-GENERIC_LIST *list; +-{ +- register GENERIC_LIST *next, *prev; +- +- for (prev = 0; list; ) { +- next = list->next; +- list->next = prev; +- prev = list; +- list = next; +- } +- return prev; +-} +- + char * + pat_subst(s, t, u, i) + char *s, *t, *u; +@@ -772,7 +816,7 @@ + + main() + { +- ARRAY *a, *new_a, *copy_of_a; ++ ARRAY *a, *new_a, *copy_of_a, *stack; + ARRAY_ELEMENT *ae, *aew; + char *s; + +@@ -784,24 +828,29 @@ + array_insert(a, 12, "twelve"); + array_insert(a, 42, "forty-two"); + print_array(a); ++ printf("\n=> s = array_to_string(a, \" \", 0);\n"); + s = array_to_string (a, " ", 0); + printf("s = %s\n", s); ++ printf("\n=> copy_of_a = array_from_string(s, \" \");\n"); + copy_of_a = array_from_string(s, " "); +- printf("copy_of_a:"); + print_array(copy_of_a); + array_dispose(copy_of_a); + printf("\n"); + free(s); ++ printf("=> s = array_remove(a, 4);\n"); + ae = array_remove(a, 4); + array_dispose_element(ae); ++ printf("=> s = array_remove(a, 1029);\n"); + ae = array_remove(a, 1029); + array_dispose_element(ae); ++ printf("=> s = array_insert(a, 16, \"sixteen\"\n);\n"); + array_insert(a, 16, "sixteen"); + print_array(a); ++ printf("\n=> s = array_to_string(a, \" \", 0);\n"); + s = array_to_string (a, " ", 0); + printf("s = %s\n", s); ++ printf("\n=> copy_of_a = array_from_string(s, \" \");\n"); + copy_of_a = array_from_string(s, " "); +- printf("copy_of_a:"); + print_array(copy_of_a); + array_dispose(copy_of_a); + printf("\n"); +@@ -811,10 +860,11 @@ + array_insert(a, 0, "zero"); + array_insert(a, 134, ""); + print_array(a); ++ printf("\n=> s = array_to_string(a, \":\", 0);\n"); + s = array_to_string (a, ":", 0); + printf("s = %s\n", s); ++ printf("\n=> copy_of_a = array_from_string(s, \" \");\n"); + copy_of_a = array_from_string(s, ":"); +- printf("copy_of_a:"); + print_array(copy_of_a); + array_dispose(copy_of_a); + printf("\n"); +@@ -857,6 +907,23 @@ + printf("\n"); + array_dispose(a); + array_dispose(new_a); ++ ++ stack = array_create(); ++ array_shift_element(stack, "first"); ++ print_array(stack); ++ printf("\n=> array_shift_element(stack, \"second\");\n"); ++ array_shift_element(stack, "second"); ++ print_array(stack); ++ printf("\n=> array_unshift_element(stack);\n"); ++ ae = array_unshift_element(stack); ++ array_dispose_element(ae); ++ print_array(stack); ++ printf("\n=> array_unshift_element(stack);\n"); ++ ae = array_unshift_element(stack); ++ array_dispose_element(ae); ++ print_array(stack); ++ printf("** should see empty nothing above.***\n"); ++ array_dispose(stack); + } + + #endif /* TEST_ARRAY */ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/array.h cvs/array.h +--- bash-2.05b/array.h 2002-03-14 14:42:51.000000000 +0100 ++++ cvs/array.h 2002-08-23 06:17:00.000000000 +0200 +@@ -51,6 +51,7 @@ + extern void array_walk __P((ARRAY *, sh_ae_map_func_t *)); + + extern ARRAY_ELEMENT *array_shift __P((ARRAY *, int, int)); ++extern ARRAY_ELEMENT *array_unshift_element __P((ARRAY *)); + extern int array_rshift __P((ARRAY *, int, char *)); + extern ARRAY *array_quote __P((ARRAY *)); + +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/autogen.sh cvs/autogen.sh +--- bash-2.05b/autogen.sh 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/autogen.sh 2002-09-19 18:39:08.000000000 +0200 +@@ -0,0 +1,20 @@ ++#!/bin/sh ++# $Id: autogen.sh,v 1.2 2002/09/19 16:39:08 masata-y Exp $ ++# Run this to generate all the initial Makefiles, etc. ++ ++./bootstrap.sh $@ ++ ++echo " + Running 'configure $@':" ++if [ -z "$*" ]; then ++ echo " ** If you wish to pass arguments to ./configure, please" ++ echo " ** specify them on the command line." ++fi ++ ++./configure "$@" && \ ++ echo "Now type 'make' to compile $PKG_NAME" || exit 1 ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** ++ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/bashhist.c cvs/bashhist.c +--- bash-2.05b/bashhist.c 2002-03-12 16:29:56.000000000 +0100 ++++ cvs/bashhist.c 2003-05-24 22:13:47.000000000 +0200 +@@ -206,6 +206,7 @@ + history_quotes_inhibit_expansion = 1; + history_search_delimiter_chars = ";&()|<>"; + history_inhibit_expansion_function = bash_history_inhibit_expansion; ++ history_comment_char = '#'; + } + + void +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/bashline.c cvs/bashline.c +--- bash-2.05b/bashline.c 2002-05-07 21:52:42.000000000 +0200 ++++ cvs/bashline.c 2002-12-07 04:51:32.000000000 +0100 +@@ -747,7 +747,7 @@ + /* Now, POSIX.1-2001 and SUSv3 say that the commands executed from the + temporary file should be placed into the history. We don't do that + yet. */ +- r = parse_and_execute (command, (editing_mode == VI_EDITING_MODE) ? "v" : "C-xC-e", SEVAL_NOHIST); ++ r = parse_and_execute (command, (editing_mode == VI_EDITING_MODE) ? "v" : "C-xC-e", SEVAL_NOHIST, 0); + + current_command_line_count = cclc; + +@@ -1044,7 +1044,10 @@ + } + else + { ++#define CMD_IS_DIR(x) (absolute_pathname(x) == 0 && *(x) != '~' && test_for_directory (x)) ++ + matches = rl_completion_matches (text, command_word_completion_function); ++ + /* If we are attempting command completion and nothing matches, we + do not want readline to perform filename completion for us. We + still want to be able to complete partial pathnames, so set the +@@ -1052,7 +1055,7 @@ + filenames and leave directories in the match list. */ + if (matches == (char **)NULL) + rl_ignore_some_completions_function = bash_ignore_filenames; +- else if (matches[1] == 0 && *matches[0] != '/') ++ else if (matches[1] == 0 && CMD_IS_DIR(matches[0])) + /* Turn off rl_filename_completion_desired so readline doesn't + append a slash if there is a directory with the same name + in the current directory, or other filename-specific things. +@@ -1061,7 +1064,7 @@ + looking in the current directory anyway, so there's no + conflict. */ + rl_filename_completion_desired = 0; +- else if (matches[0] && matches[1] && STREQ (matches[0], matches[1]) && *matches[0] != '/') ++ else if (matches[0] && matches[1] && STREQ (matches[0], matches[1]) && CMD_IS_DIR (matches[0])) + /* There are multiple instances of the same match (duplicate + completions haven't yet been removed). In this case, all of + the matches will be the same, and the duplicate removal code +@@ -2717,7 +2720,7 @@ + ts = save_token_state (); + + cmd = savestring (cmd); +- parse_and_execute (cmd, "bash_execute_unix_command", SEVAL_NOHIST); ++ parse_and_execute (cmd, "bash_execute_unix_command", SEVAL_NOHIST, 0); + + current_command_line_count = old_line_count; + restore_token_state (ts); +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/bashlocal.m4 cvs/bashlocal.m4 +--- bash-2.05b/bashlocal.m4 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/bashlocal.m4 2002-09-09 20:25:01.000000000 +0200 +@@ -0,0 +1,1792 @@ ++dnl ++dnl Bash specific tests ++dnl ++dnl Some derived from PDKSH 5.1.3 autoconf tests ++dnl ++ ++AC_DEFUN(BASH_C_LONG_LONG, ++[AC_CACHE_CHECK(for long long, ac_cv_c_long_long, ++[if test "$GCC" = yes; then ++ ac_cv_c_long_long=yes ++else ++AC_TRY_RUN([ ++int ++main() ++{ ++long long foo = 0; ++exit(sizeof(long long) < sizeof(long)); ++} ++], ac_cv_c_long_long=yes, ac_cv_c_long_long=no) ++fi]) ++if test $ac_cv_c_long_long = yes; then ++ AC_DEFINE(HAVE_LONG_LONG, 1, [Define if the `long long' type works.]) ++fi ++]) ++ ++dnl ++dnl This is very similar to AC_C_LONG_DOUBLE, with the fix for IRIX ++dnl (< changed to <=) added. ++dnl ++AC_DEFUN(BASH_C_LONG_DOUBLE, ++[AC_CACHE_CHECK(for long double, ac_cv_c_long_double, ++[if test "$GCC" = yes; then ++ ac_cv_c_long_double=yes ++else ++AC_TRY_RUN([ ++int ++main() ++{ ++ /* The Stardent Vistra knows sizeof(long double), but does not ++ support it. */ ++ long double foo = 0.0; ++ /* On Ultrix 4.3 cc, long double is 4 and double is 8. */ ++ /* On IRIX 5.3, the compiler converts long double to double with a warning, ++ but compiles this successfully. */ ++ exit(sizeof(long double) <= sizeof(double)); ++} ++], ac_cv_c_long_double=yes, ac_cv_c_long_double=no) ++fi]) ++if test $ac_cv_c_long_double = yes; then ++ AC_DEFINE(HAVE_LONG_DOUBLE, 1, [Define if the `long double' type works.]) ++fi ++]) ++ ++dnl ++dnl Check for . This is separated out so that it can be ++dnl AC_REQUIREd. ++dnl ++dnl BASH_HEADER_INTTYPES ++AC_DEFUN(BASH_HEADER_INTTYPES, ++[ ++ AC_CHECK_HEADERS(inttypes.h) ++]) ++ ++dnl ++dnl check for typedef'd symbols in header files, but allow the caller to ++dnl specify the include files to be checked in addition to the default ++dnl ++dnl BASH_CHECK_TYPE(TYPE, HEADERS, DEFAULT[, VALUE-IF-FOUND]) ++AC_DEFUN(BASH_CHECK_TYPE, ++[ ++AC_REQUIRE([AC_HEADER_STDC])dnl ++AC_REQUIRE([BASH_HEADER_INTTYPES]) ++AC_MSG_CHECKING(for $1) ++AC_CACHE_VAL(bash_cv_type_$1, ++[AC_EGREP_CPP($1, [#include ++#if STDC_HEADERS ++#include ++#include ++#endif ++#if HAVE_INTTYPES_H ++#include ++#endif ++$2 ++], bash_cv_type_$1=yes, bash_cv_type_$1=no)]) ++AC_MSG_RESULT($bash_cv_type_$1) ++ifelse($#, 4, [if test $bash_cv_type_$1 = yes; then ++ AC_DEFINE($4) ++ fi]) ++if test $bash_cv_type_$1 = no; then ++ AC_DEFINE_UNQUOTED($1, $3) ++fi ++]) ++ ++dnl ++dnl BASH_CHECK_DECL(FUNC) ++dnl ++dnl Check for a declaration of FUNC in stdlib.h and inttypes.h like ++dnl AC_CHECK_DECL ++dnl ++AC_DEFUN(BASH_CHECK_DECL, ++[ ++AC_REQUIRE([AC_HEADER_STDC]) ++AC_REQUIRE([BASH_HEADER_INTTYPES]) ++AC_CACHE_CHECK([for declaration of $1], bash_cv_decl_$1, ++[AC_TRY_LINK( ++[ ++#if STDC_HEADERS ++# include ++#endif ++#if HAVE_INTTYPES_H ++# include ++#endif ++], ++[return !$1;], ++bash_cv_decl_$1=yes, bash_cv_decl_$1=no)]) ++bash_tr_func=HAVE_DECL_`echo $1 | tr 'abcdefghijklmnopqrstuvwxyz' 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'` ++if test $bash_cv_decl_$1 = yes; then ++ AC_DEFINE_UNQUOTED($bash_tr_func, 1) ++else ++ AC_DEFINE_UNQUOTED($bash_tr_func, 0) ++fi ++]) ++ ++AC_DEFUN(BASH_DECL_PRINTF, ++[AC_MSG_CHECKING(for declaration of printf in ) ++AC_CACHE_VAL(bash_cv_printf_declared, ++[AC_TRY_RUN([ ++#include ++#ifdef __STDC__ ++typedef int (*_bashfunc)(const char *, ...); ++#else ++typedef int (*_bashfunc)(); ++#endif ++main() ++{ ++_bashfunc pf; ++pf = (_bashfunc) printf; ++exit(pf == 0); ++} ++], bash_cv_printf_declared=yes, bash_cv_printf_declared=no, ++ [AC_MSG_WARN(cannot check printf declaration if cross compiling -- defaulting to yes) ++ bash_cv_printf_declared=yes] ++)]) ++AC_MSG_RESULT($bash_cv_printf_declared) ++if test $bash_cv_printf_declared = yes; then ++AC_DEFINE(PRINTF_DECLARED) ++fi ++]) ++ ++AC_DEFUN(BASH_DECL_SBRK, ++[AC_MSG_CHECKING(for declaration of sbrk in ) ++AC_CACHE_VAL(bash_cv_sbrk_declared, ++[AC_EGREP_HEADER(sbrk, unistd.h, ++ bash_cv_sbrk_declared=yes, bash_cv_sbrk_declared=no)]) ++AC_MSG_RESULT($bash_cv_sbrk_declared) ++if test $bash_cv_sbrk_declared = yes; then ++AC_DEFINE(SBRK_DECLARED) ++fi ++]) ++ ++dnl ++dnl Check for sys_siglist[] or _sys_siglist[] ++dnl ++AC_DEFUN(BASH_DECL_UNDER_SYS_SIGLIST, ++[AC_MSG_CHECKING([for _sys_siglist in signal.h or unistd.h]) ++AC_CACHE_VAL(bash_cv_decl_under_sys_siglist, ++[AC_TRY_COMPILE([ ++#include ++#include ++#ifdef HAVE_UNISTD_H ++#include ++#endif], [ char *msg = _sys_siglist[2]; ], ++ bash_cv_decl_under_sys_siglist=yes, bash_cv_decl_under_sys_siglist=no, ++ [AC_MSG_WARN(cannot check for _sys_siglist[] if cross compiling -- defaulting to no)])])dnl ++AC_MSG_RESULT($bash_cv_decl_under_sys_siglist) ++if test $bash_cv_decl_under_sys_siglist = yes; then ++AC_DEFINE(UNDER_SYS_SIGLIST_DECLARED) ++fi ++]) ++ ++AC_DEFUN(BASH_UNDER_SYS_SIGLIST, ++[AC_REQUIRE([BASH_DECL_UNDER_SYS_SIGLIST]) ++AC_MSG_CHECKING([for _sys_siglist in system C library]) ++AC_CACHE_VAL(bash_cv_under_sys_siglist, ++[AC_TRY_RUN([ ++#include ++#include ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++#ifndef UNDER_SYS_SIGLIST_DECLARED ++extern char *_sys_siglist[]; ++#endif ++main() ++{ ++char *msg = (char *)_sys_siglist[2]; ++exit(msg == 0); ++}], ++ bash_cv_under_sys_siglist=yes, bash_cv_under_sys_siglist=no, ++ [AC_MSG_WARN(cannot check for _sys_siglist[] if cross compiling -- defaulting to no) ++ bash_cv_under_sys_siglist=no])]) ++AC_MSG_RESULT($bash_cv_under_sys_siglist) ++if test $bash_cv_under_sys_siglist = yes; then ++AC_DEFINE(HAVE_UNDER_SYS_SIGLIST) ++fi ++]) ++ ++AC_DEFUN(BASH_SYS_SIGLIST, ++[AC_REQUIRE([AC_DECL_SYS_SIGLIST]) ++AC_MSG_CHECKING([for sys_siglist in system C library]) ++AC_CACHE_VAL(bash_cv_sys_siglist, ++[AC_TRY_RUN([ ++#include ++#include ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++#ifndef SYS_SIGLIST_DECLARED ++extern char *sys_siglist[]; ++#endif ++main() ++{ ++char *msg = sys_siglist[2]; ++exit(msg == 0); ++}], ++ bash_cv_sys_siglist=yes, bash_cv_sys_siglist=no, ++ [AC_MSG_WARN(cannot check for sys_siglist if cross compiling -- defaulting to no) ++ bash_cv_sys_siglist=no])]) ++AC_MSG_RESULT($bash_cv_sys_siglist) ++if test $bash_cv_sys_siglist = yes; then ++AC_DEFINE(HAVE_SYS_SIGLIST) ++fi ++]) ++ ++dnl Check for the various permutations of sys_siglist and make sure we ++dnl compile in siglist.o if they're not defined ++AC_DEFUN(BASH_CHECK_SYS_SIGLIST, [ ++AC_REQUIRE([BASH_SYS_SIGLIST]) ++AC_REQUIRE([BASH_DECL_UNDER_SYS_SIGLIST]) ++AC_REQUIRE([BASH_FUNC_STRSIGNAL]) ++if test "$bash_cv_sys_siglist" = no && test "$bash_cv_under_sys_siglist" = no && test "$bash_cv_have_strsignal" = no; then ++ SIGLIST_O=siglist.o ++else ++ SIGLIST_O= ++fi ++AC_SUBST([SIGLIST_O]) ++]) ++ ++dnl Check for sys_errlist[] and sys_nerr, check for declaration ++AC_DEFUN(BASH_SYS_ERRLIST, ++[AC_MSG_CHECKING([for sys_errlist and sys_nerr]) ++AC_CACHE_VAL(bash_cv_sys_errlist, ++[AC_TRY_LINK([#include ], ++[extern char *sys_errlist[]; ++ extern int sys_nerr; ++ char *msg = sys_errlist[sys_nerr - 1];], ++ bash_cv_sys_errlist=yes, bash_cv_sys_errlist=no)])dnl ++AC_MSG_RESULT($bash_cv_sys_errlist) ++if test $bash_cv_sys_errlist = yes; then ++AC_DEFINE(HAVE_SYS_ERRLIST) ++fi ++]) ++ ++dnl ++dnl Check if dup2() does not clear the close on exec flag ++dnl ++AC_DEFUN(BASH_FUNC_DUP2_CLOEXEC_CHECK, ++[AC_MSG_CHECKING(if dup2 fails to clear the close-on-exec flag) ++AC_CACHE_VAL(bash_cv_dup2_broken, ++[AC_TRY_RUN([ ++#include ++#include ++main() ++{ ++ int fd1, fd2, fl; ++ fd1 = open("/dev/null", 2); ++ if (fcntl(fd1, 2, 1) < 0) ++ exit(1); ++ fd2 = dup2(fd1, 1); ++ if (fd2 < 0) ++ exit(2); ++ fl = fcntl(fd2, 1, 0); ++ /* fl will be 1 if dup2 did not reset the close-on-exec flag. */ ++ exit(fl != 1); ++} ++], bash_cv_dup2_broken=yes, bash_cv_dup2_broken=no, ++ [AC_MSG_WARN(cannot check dup2 if cross compiling -- defaulting to no) ++ bash_cv_dup2_broken=no]) ++]) ++AC_MSG_RESULT($bash_cv_dup2_broken) ++if test $bash_cv_dup2_broken = yes; then ++AC_DEFINE(DUP2_BROKEN) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_STRSIGNAL, ++[AC_MSG_CHECKING([for the existence of strsignal]) ++AC_CACHE_VAL(bash_cv_have_strsignal, ++[AC_TRY_LINK([#include ++#include ], ++[char *s = (char *)strsignal(2);], ++ bash_cv_have_strsignal=yes, bash_cv_have_strsignal=no)]) ++AC_MSG_RESULT($bash_cv_have_strsignal) ++if test $bash_cv_have_strsignal = yes; then ++AC_DEFINE(HAVE_STRSIGNAL) ++fi ++]) ++ ++dnl Check to see if opendir will open non-directories (not a nice thing) ++AC_DEFUN(BASH_FUNC_OPENDIR_CHECK, ++[AC_REQUIRE([AC_HEADER_DIRENT])dnl ++AC_MSG_CHECKING(if opendir() opens non-directories) ++AC_CACHE_VAL(bash_cv_opendir_not_robust, ++[AC_TRY_RUN([ ++#include ++#include ++#include ++#ifdef HAVE_UNISTD_H ++# include ++#endif /* HAVE_UNISTD_H */ ++#if defined(HAVE_DIRENT_H) ++# include ++#else ++# define dirent direct ++# ifdef HAVE_SYS_NDIR_H ++# include ++# endif /* SYSNDIR */ ++# ifdef HAVE_SYS_DIR_H ++# include ++# endif /* SYSDIR */ ++# ifdef HAVE_NDIR_H ++# include ++# endif ++#endif /* HAVE_DIRENT_H */ ++main() ++{ ++DIR *dir; ++int fd, err; ++err = mkdir("/tmp/bash-aclocal", 0700); ++if (err < 0) { ++ perror("mkdir"); ++ exit(1); ++} ++unlink("/tmp/bash-aclocal/not_a_directory"); ++fd = open("/tmp/bash-aclocal/not_a_directory", O_WRONLY|O_CREAT|O_EXCL, 0666); ++write(fd, "\n", 1); ++close(fd); ++dir = opendir("/tmp/bash-aclocal/not_a_directory"); ++unlink("/tmp/bash-aclocal/not_a_directory"); ++rmdir("/tmp/bash-aclocal"); ++exit (dir == 0); ++}], bash_cv_opendir_not_robust=yes,bash_cv_opendir_not_robust=no, ++ [AC_MSG_WARN(cannot check opendir if cross compiling -- defaulting to no) ++ bash_cv_opendir_not_robust=no] ++)]) ++AC_MSG_RESULT($bash_cv_opendir_not_robust) ++if test $bash_cv_opendir_not_robust = yes; then ++AC_DEFINE(OPENDIR_NOT_ROBUST) ++fi ++]) ++ ++dnl ++AC_DEFUN(BASH_TYPE_SIGHANDLER, ++[AC_MSG_CHECKING([whether signal handlers are of type void]) ++AC_CACHE_VAL(bash_cv_void_sighandler, ++[AC_TRY_COMPILE([#include ++#include ++#ifdef signal ++#undef signal ++#endif ++#ifdef __cplusplus ++extern "C" ++#endif ++void (*signal ()) ();], ++[int i;], bash_cv_void_sighandler=yes, bash_cv_void_sighandler=no)])dnl ++AC_MSG_RESULT($bash_cv_void_sighandler) ++if test $bash_cv_void_sighandler = yes; then ++AC_DEFINE(VOID_SIGHANDLER) ++fi ++]) ++ ++dnl ++dnl A signed 16-bit integer quantity ++dnl ++AC_DEFUN(BASH_TYPE_BITS16_T, ++[ ++if test "$ac_cv_sizeof_short" = 2; then ++ AC_CHECK_TYPE(bits16_t, short) ++elif test "$ac_cv_sizeof_char" = 2; then ++ AC_CHECK_TYPE(bits16_t, char) ++else ++ AC_CHECK_TYPE(bits16_t, short) ++fi ++]) ++ ++dnl ++dnl An unsigned 16-bit integer quantity ++dnl ++AC_DEFUN(BASH_TYPE_U_BITS16_T, ++[ ++if test "$ac_cv_sizeof_short" = 2; then ++ AC_CHECK_TYPE(u_bits16_t, unsigned short) ++elif test "$ac_cv_sizeof_char" = 2; then ++ AC_CHECK_TYPE(u_bits16_t, unsigned char) ++else ++ AC_CHECK_TYPE(u_bits16_t, unsigned short) ++fi ++]) ++ ++dnl ++dnl A signed 32-bit integer quantity ++dnl ++AC_DEFUN(BASH_TYPE_BITS32_T, ++[ ++if test "$ac_cv_sizeof_int" = 4; then ++ AC_CHECK_TYPE(bits32_t, int) ++elif test "$ac_cv_sizeof_long" = 4; then ++ AC_CHECK_TYPE(bits32_t, long) ++else ++ AC_CHECK_TYPE(bits32_t, int) ++fi ++]) ++ ++dnl ++dnl An unsigned 32-bit integer quantity ++dnl ++AC_DEFUN(BASH_TYPE_U_BITS32_T, ++[ ++if test "$ac_cv_sizeof_int" = 4; then ++ AC_CHECK_TYPE(u_bits32_t, unsigned int) ++elif test "$ac_cv_sizeof_long" = 4; then ++ AC_CHECK_TYPE(u_bits32_t, unsigned long) ++else ++ AC_CHECK_TYPE(u_bits32_t, unsigned int) ++fi ++]) ++ ++AC_DEFUN(BASH_TYPE_PTRDIFF_T, ++[ ++if test "$ac_cv_sizeof_int" = "$ac_cv_sizeof_char_p"; then ++ AC_CHECK_TYPE(ptrdiff_t, int) ++elif test "$ac_cv_sizeof_long" = "$ac_cv_sizeof_char_p"; then ++ AC_CHECK_TYPE(ptrdiff_t, long) ++elif test "$ac_cv_type_long_long" = yes && test "$ac_cv_sizeof_long_long" = "$ac_cv_sizeof_char_p"; then ++ AC_CHECK_TYPE(ptrdiff_t, [long long]) ++else ++ AC_CHECK_TYPE(ptrdiff_t, int) ++fi ++]) ++ ++dnl ++dnl A signed 64-bit quantity ++dnl ++AC_DEFUN(BASH_TYPE_BITS64_T, ++[ ++if test "$ac_cv_sizeof_char_p" = 8; then ++ AC_CHECK_TYPE(bits64_t, char *) ++elif test "$ac_cv_sizeof_double" = 8; then ++ AC_CHECK_TYPE(bits64_t, double) ++elif test -n "$ac_cv_type_long_long" && test "$ac_cv_sizeof_long_long" = 8; then ++ AC_CHECK_TYPE(bits64_t, [long long]) ++elif test "$ac_cv_sizeof_long" = 8; then ++ AC_CHECK_TYPE(bits64_t, long) ++else ++ AC_CHECK_TYPE(bits64_t, double) ++fi ++]) ++ ++AC_DEFUN(BASH_TYPE_LONG_LONG, ++[ ++AC_CACHE_CHECK([for long long], bash_cv_type_long_long, ++[AC_TRY_LINK([ ++long long ll = 1; int i = 63;], ++[ ++long long llm = (long long) -1; ++return ll << i | ll >> i | llm / ll | llm % ll; ++], bash_cv_type_long_long='long long', bash_cv_type_long_long='long')]) ++if test "$bash_cv_type_long_long" = 'long long'; then ++ AC_DEFINE(HAVE_LONG_LONG, 1) ++fi ++]) ++ ++AC_DEFUN(BASH_TYPE_UNSIGNED_LONG_LONG, ++[ ++AC_CACHE_CHECK([for unsigned long long], bash_cv_type_unsigned_long_long, ++[AC_TRY_LINK([ ++unsigned long long ull = 1; int i = 63;], ++[ ++unsigned long long ullmax = (unsigned long long) -1; ++return ull << i | ull >> i | ullmax / ull | ullmax % ull; ++], bash_cv_type_unsigned_long_long='unsigned long long', ++ bash_cv_type_unsigned_long_long='unsigned long')]) ++if test "$bash_cv_type_unsigned_long_long" = 'unsigned long long'; then ++ AC_DEFINE(HAVE_UNSIGNED_LONG_LONG, 1) ++fi ++]) ++ ++dnl ++dnl Type of struct rlimit fields: some systems (OSF/1, NetBSD, RISC/os 5.0) ++dnl have a rlim_t, others (4.4BSD based systems) use quad_t, others use ++dnl long and still others use int (HP-UX 9.01, SunOS 4.1.3). To simplify ++dnl matters, this just checks for rlim_t, quad_t, or long. ++dnl ++AC_DEFUN(BASH_TYPE_RLIMIT, ++[AC_MSG_CHECKING(for size and type of struct rlimit fields) ++AC_CACHE_VAL(bash_cv_type_rlimit, ++[AC_TRY_COMPILE([#include ++#include ], ++[rlim_t xxx;], bash_cv_type_rlimit=rlim_t,[ ++AC_TRY_RUN([ ++#include ++#include ++#include ++main() ++{ ++#ifdef HAVE_QUAD_T ++ struct rlimit rl; ++ if (sizeof(rl.rlim_cur) == sizeof(quad_t)) ++ exit(0); ++#endif ++ exit(1); ++}], bash_cv_type_rlimit=quad_t, bash_cv_type_rlimit=long, ++ [AC_MSG_WARN(cannot check quad_t if cross compiling -- defaulting to long) ++ bash_cv_type_rlimit=long])]) ++]) ++AC_MSG_RESULT($bash_cv_type_rlimit) ++if test $bash_cv_type_rlimit = quad_t; then ++AC_DEFINE(RLIMTYPE, quad_t) ++elif test $bash_cv_type_rlimit = rlim_t; then ++AC_DEFINE(RLIMTYPE, rlim_t) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_LSTAT, ++[dnl Cannot use AC_CHECK_FUNCS(lstat) because Linux defines lstat() as an ++dnl inline function in . ++AC_CACHE_CHECK([for lstat], bash_cv_func_lstat, ++[AC_TRY_LINK([ ++#include ++#include ++],[ lstat(".",(struct stat *)0); ], ++bash_cv_func_lstat=yes, bash_cv_func_lstat=no)]) ++if test $bash_cv_func_lstat = yes; then ++ AC_DEFINE(HAVE_LSTAT) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_INET_ATON, ++[ ++AC_CACHE_CHECK([for inet_aton], bash_cv_func_inet_aton, ++[AC_TRY_LINK([ ++#include ++#include ++#include ++struct in_addr ap;], [ inet_aton("127.0.0.1", &ap); ], ++bash_cv_func_inet_aton=yes, bash_cv_func_inet_aton=no)]) ++if test $bash_cv_func_inet_aton = yes; then ++ AC_DEFINE(HAVE_INET_ATON) ++else ++ AC_LIBOBJ(inet_aton) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_GETENV, ++[AC_MSG_CHECKING(to see if getenv can be redefined) ++AC_CACHE_VAL(bash_cv_getenv_redef, ++[AC_TRY_RUN([ ++#ifdef HAVE_UNISTD_H ++# include ++#endif ++#ifndef __STDC__ ++# ifndef const ++# define const ++# endif ++#endif ++char * ++getenv (name) ++#if defined (__linux__) || defined (__bsdi__) || defined (convex) ++ const char *name; ++#else ++ char const *name; ++#endif /* !__linux__ && !__bsdi__ && !convex */ ++{ ++return "42"; ++} ++main() ++{ ++char *s; ++/* The next allows this program to run, but does not allow bash to link ++ when it redefines getenv. I'm not really interested in figuring out ++ why not. */ ++#if defined (NeXT) ++exit(1); ++#endif ++s = getenv("ABCDE"); ++exit(s == 0); /* force optimizer to leave getenv in */ ++} ++], bash_cv_getenv_redef=yes, bash_cv_getenv_redef=no, ++ [AC_MSG_WARN(cannot check getenv redefinition if cross compiling -- defaulting to yes) ++ bash_cv_getenv_redef=yes] ++)]) ++AC_MSG_RESULT($bash_cv_getenv_redef) ++if test $bash_cv_getenv_redef = yes; then ++AC_DEFINE(CAN_REDEFINE_GETENV) ++fi ++]) ++ ++# We should check for putenv before calling this ++AC_DEFUN(BASH_FUNC_STD_PUTENV, ++[ ++AC_REQUIRE([AC_HEADER_STDC]) ++AC_REQUIRE([AC_C_PROTOTYPES]) ++AC_CACHE_CHECK([for standard-conformant putenv declaration], bash_cv_std_putenv, ++[AC_TRY_LINK([ ++#if STDC_HEADERS ++#include ++#include ++#endif ++#ifndef __STDC__ ++# ifndef const ++# define const ++# endif ++#endif ++#ifdef PROTOTYPES ++extern int putenv (char *); ++#else ++extern int putenv (); ++#endif ++], ++[return (putenv == 0);], ++bash_cv_std_putenv=yes, bash_cv_std_putenv=no ++)]) ++if test $bash_cv_std_putenv = yes; then ++AC_DEFINE(HAVE_STD_PUTENV) ++fi ++]) ++ ++# We should check for unsetenv before calling this ++AC_DEFUN(BASH_FUNC_STD_UNSETENV, ++[ ++AC_REQUIRE([AC_HEADER_STDC]) ++AC_REQUIRE([AC_C_PROTOTYPES]) ++AC_CACHE_CHECK([for standard-conformant unsetenv declaration], bash_cv_std_unsetenv, ++[AC_TRY_LINK([ ++#if STDC_HEADERS ++#include ++#include ++#endif ++#ifndef __STDC__ ++# ifndef const ++# define const ++# endif ++#endif ++#ifdef PROTOTYPES ++extern int unsetenv (const char *); ++#else ++extern int unsetenv (); ++#endif ++], ++[return (unsetenv == 0);], ++bash_cv_std_unsetenv=yes, bash_cv_std_unsetenv=no ++)]) ++if test $bash_cv_std_unsetenv = yes; then ++AC_DEFINE(HAVE_STD_UNSETENV) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_ULIMIT_MAXFDS, ++[AC_MSG_CHECKING(whether ulimit can substitute for getdtablesize) ++AC_CACHE_VAL(bash_cv_ulimit_maxfds, ++[AC_TRY_RUN([ ++main() ++{ ++long maxfds = ulimit(4, 0L); ++exit (maxfds == -1L); ++} ++], bash_cv_ulimit_maxfds=yes, bash_cv_ulimit_maxfds=no, ++ [AC_MSG_WARN(cannot check ulimit if cross compiling -- defaulting to no) ++ bash_cv_ulimit_maxfds=no] ++)]) ++AC_MSG_RESULT($bash_cv_ulimit_maxfds) ++if test $bash_cv_ulimit_maxfds = yes; then ++AC_DEFINE(ULIMIT_MAXFDS) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_GETCWD, ++[AC_MSG_CHECKING([if getcwd() calls popen()]) ++AC_CACHE_VAL(bash_cv_getcwd_calls_popen, ++[AC_TRY_RUN([ ++#include ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++ ++#ifndef __STDC__ ++#ifndef const ++#define const ++#endif ++#endif ++ ++int popen_called; ++ ++FILE * ++popen(command, type) ++ const char *command; ++ const char *type; ++{ ++ popen_called = 1; ++ return (FILE *)NULL; ++} ++ ++FILE *_popen(command, type) ++ const char *command; ++ const char *type; ++{ ++ return (popen (command, type)); ++} ++ ++int ++pclose(stream) ++FILE *stream; ++{ ++ return 0; ++} ++ ++int ++_pclose(stream) ++FILE *stream; ++{ ++ return 0; ++} ++ ++main() ++{ ++ char lbuf[32]; ++ popen_called = 0; ++ getcwd(lbuf, 32); ++ exit (popen_called); ++} ++], bash_cv_getcwd_calls_popen=no, bash_cv_getcwd_calls_popen=yes, ++ [AC_MSG_WARN(cannot check whether getcwd calls popen if cross compiling -- defaulting to no) ++ bash_cv_getcwd_calls_popen=no] ++)]) ++AC_MSG_RESULT($bash_cv_getcwd_calls_popen) ++if test $bash_cv_getcwd_calls_popen = yes; then ++AC_DEFINE(GETCWD_BROKEN) ++AC_LIBOBJ(getcwd) ++fi ++]) ++ ++dnl ++dnl This needs BASH_CHECK_SOCKLIB, but since that's not called on every ++dnl system, we can't use AC_PREREQ ++dnl ++AC_DEFUN(BASH_FUNC_GETHOSTBYNAME, ++[if test "X$bash_cv_have_gethostbyname" = "X"; then ++_bash_needmsg=yes ++else ++AC_MSG_CHECKING(for gethostbyname in socket library) ++_bash_needmsg= ++fi ++AC_CACHE_VAL(bash_cv_have_gethostbyname, ++[AC_TRY_LINK([#include ], ++[ struct hostent *hp; ++ hp = gethostbyname("localhost"); ++], bash_cv_have_gethostbyname=yes, bash_cv_have_gethostbyname=no)] ++) ++if test "X$_bash_needmsg" = Xyes; then ++ AC_MSG_CHECKING(for gethostbyname in socket library) ++fi ++AC_MSG_RESULT($bash_cv_have_gethostbyname) ++if test "$bash_cv_have_gethostbyname" = yes; then ++AC_DEFINE(HAVE_GETHOSTBYNAME) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_FNMATCH_EXTMATCH, ++[AC_MSG_CHECKING(if fnmatch does extended pattern matching with FNM_EXTMATCH) ++AC_CACHE_VAL(bash_cv_fnm_extmatch, ++[AC_TRY_RUN([ ++#include ++ ++main() ++{ ++#ifdef FNM_EXTMATCH ++ exit (0); ++#else ++ exit (1); ++#endif ++} ++], bash_cv_fnm_extmatch=yes, bash_cv_fnm_extmatch=no, ++ [AC_MSG_WARN(cannot check FNM_EXTMATCH if cross compiling -- defaulting to no) ++ bash_cv_fnm_extmatch=no]) ++]) ++AC_MSG_RESULT($bash_cv_fnm_extmatch) ++if test $bash_cv_fnm_extmatch = yes; then ++AC_DEFINE(HAVE_LIBC_FNM_EXTMATCH) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_POSIX_SETJMP, ++[AC_REQUIRE([BASH_SYS_SIGNAL_VINTAGE]) ++AC_MSG_CHECKING(for presence of POSIX-style sigsetjmp/siglongjmp) ++AC_CACHE_VAL(bash_cv_func_sigsetjmp, ++[AC_TRY_RUN([ ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++#include ++#include ++#include ++ ++main() ++{ ++#if !defined (_POSIX_VERSION) || !defined (HAVE_POSIX_SIGNALS) ++exit (1); ++#else ++ ++int code; ++sigset_t set, oset; ++sigjmp_buf xx; ++ ++/* get the mask */ ++sigemptyset(&set); ++sigemptyset(&oset); ++sigprocmask(SIG_BLOCK, (sigset_t *)NULL, &set); ++sigprocmask(SIG_BLOCK, (sigset_t *)NULL, &oset); ++ ++/* save it */ ++code = sigsetjmp(xx, 1); ++if (code) ++ exit(0); /* could get sigmask and compare to oset here. */ ++ ++/* change it */ ++sigaddset(&set, SIGINT); ++sigprocmask(SIG_BLOCK, &set, (sigset_t *)NULL); ++ ++/* and siglongjmp */ ++siglongjmp(xx, 10); ++exit(1); ++#endif ++}], bash_cv_func_sigsetjmp=present, bash_cv_func_sigsetjmp=missing, ++ [AC_MSG_WARN(cannot check for sigsetjmp/siglongjmp if cross-compiling -- defaulting to missing) ++ bash_cv_func_sigsetjmp=missing] ++)]) ++AC_MSG_RESULT($bash_cv_func_sigsetjmp) ++if test $bash_cv_func_sigsetjmp = present; then ++AC_DEFINE(HAVE_POSIX_SIGSETJMP) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_STRCOLL, ++[ ++AC_MSG_CHECKING(whether or not strcoll and strcmp differ) ++AC_CACHE_VAL(bash_cv_func_strcoll_broken, ++[AC_TRY_RUN([ ++#include ++#if defined (HAVE_LOCALE_H) ++#include ++#endif ++ ++main(c, v) ++int c; ++char *v[]; ++{ ++ int r1, r2; ++ char *deflocale, *defcoll; ++ ++#ifdef HAVE_SETLOCALE ++ deflocale = setlocale(LC_ALL, ""); ++ defcoll = setlocale(LC_COLLATE, ""); ++#endif ++ ++#ifdef HAVE_STRCOLL ++ /* These two values are taken from tests/glob-test. */ ++ r1 = strcoll("abd", "aXd"); ++#else ++ r1 = 0; ++#endif ++ r2 = strcmp("abd", "aXd"); ++ ++ /* These two should both be greater than 0. It is permissible for ++ a system to return different values, as long as the sign is the ++ same. */ ++ ++ /* Exit with 1 (failure) if these two values are both > 0, since ++ this tests whether strcoll(3) is broken with respect to strcmp(3) ++ in the default locale. */ ++ exit (r1 > 0 && r2 > 0); ++} ++], bash_cv_func_strcoll_broken=yes, bash_cv_func_strcoll_broken=no, ++ [AC_MSG_WARN(cannot check strcoll if cross compiling -- defaulting to no) ++ bash_cv_func_strcoll_broken=no] ++)]) ++AC_MSG_RESULT($bash_cv_func_strcoll_broken) ++if test $bash_cv_func_strcoll_broken = yes; then ++AC_DEFINE(STRCOLL_BROKEN) ++fi ++]) ++ ++AC_DEFUN(BASH_FUNC_PRINTF_A_FORMAT, ++[AC_MSG_CHECKING([for printf floating point output in hex notation]) ++AC_CACHE_VAL(bash_cv_printf_a_format, ++[AC_TRY_RUN([ ++#include ++#include ++ ++int ++main() ++{ ++ double y = 0.0; ++ char abuf[1024]; ++ ++ sprintf(abuf, "%A", y); ++ exit(strchr(abuf, 'P') == (char *)0); ++} ++], bash_cv_printf_a_format=yes, bash_cv_printf_a_format=no, ++ [AC_MSG_WARN(cannot check printf if cross compiling -- defaulting to no) ++ bash_cv_printf_a_format=no] ++)]) ++AC_MSG_RESULT($bash_cv_printf_a_format) ++if test $bash_cv_printf_a_format = yes; then ++AC_DEFINE(HAVE_PRINTF_A_FORMAT) ++fi ++]) ++ ++AC_DEFUN(BASH_STRUCT_TERMIOS_LDISC, ++[ ++AC_CHECK_MEMBER(struct termios.c_line, AC_DEFINE(TERMIOS_LDISC), ,[ ++#include ++#include ++]) ++]) ++ ++AC_DEFUN(BASH_STRUCT_TERMIO_LDISC, ++[ ++AC_CHECK_MEMBER(struct termio.c_line, AC_DEFINE(TERMIO_LDISC), ,[ ++#include ++#include ++]) ++]) ++ ++dnl ++dnl Like AC_STRUCT_ST_BLOCKS, but doesn't muck with LIBOBJS ++dnl ++dnl sets bash_cv_struct_stat_st_blocks ++dnl ++dnl unused for now; we'll see how AC_CHECK_MEMBERS works ++dnl ++AC_DEFUN(BASH_STRUCT_ST_BLOCKS, ++[ ++AC_MSG_CHECKING([for struct stat.st_blocks]) ++AC_CACHE_VAL(bash_cv_struct_stat_st_blocks, ++[AC_TRY_COMPILE( ++[ ++#include ++#include ++], ++[ ++main() ++{ ++static struct stat a; ++if (a.st_blocks) return 0; ++return 0; ++} ++], bash_cv_struct_stat_st_blocks=yes, bash_cv_struct_stat_st_blocks=no) ++]) ++AC_MSG_RESULT($bash_cv_struct_stat_st_blocks) ++if test "$bash_cv_struct_stat_st_blocks" = "yes"; then ++AC_DEFINE(HAVE_STRUCT_STAT_ST_BLOCKS) ++fi ++]) ++ ++AC_DEFUN(BASH_CHECK_LIB_TERMCAP, ++[ ++if test "X$bash_cv_termcap_lib" = "X"; then ++_bash_needmsg=yes ++else ++AC_MSG_CHECKING(which library has the termcap functions) ++_bash_needmsg= ++fi ++AC_CACHE_VAL(bash_cv_termcap_lib, ++[AC_CHECK_LIB(termcap, tgetent, bash_cv_termcap_lib=libtermcap, ++ [AC_CHECK_LIB(tinfo, tgetent, bash_cv_termcap_lib=libtinfo, ++ [AC_CHECK_LIB(curses, tgetent, bash_cv_termcap_lib=libcurses, ++ [AC_CHECK_LIB(ncurses, tgetent, bash_cv_termcap_lib=libncurses, ++ bash_cv_termcap_lib=gnutermcap)])])])]) ++if test "X$_bash_needmsg" = "Xyes"; then ++AC_MSG_CHECKING(which library has the termcap functions) ++fi ++AC_MSG_RESULT(using $bash_cv_termcap_lib) ++if test $bash_cv_termcap_lib = gnutermcap && test -z "$prefer_curses"; then ++LDFLAGS="$LDFLAGS -L./lib/termcap" ++TERMCAP_LIB="./lib/termcap/libtermcap.a" ++TERMCAP_DEP="./lib/termcap/libtermcap.a" ++elif test $bash_cv_termcap_lib = libtermcap && test -z "$prefer_curses"; then ++TERMCAP_LIB=-ltermcap ++TERMCAP_DEP= ++elif test $bash_cv_termcap_lib = libtinfo; then ++TERMCAP_LIB=-ltinfo ++TERMCAP_DEP= ++elif test $bash_cv_termcap_lib = libncurses; then ++TERMCAP_LIB=-lncurses ++TERMCAP_DEP= ++else ++TERMCAP_LIB=-lcurses ++TERMCAP_DEP= ++fi ++]) ++ ++dnl ++dnl Check for the presence of getpeername in libsocket. ++dnl If libsocket is present, check for libnsl and add it to LIBS if ++dnl it's there, since most systems with libsocket require linking ++dnl with libnsl as well. This should only be called if getpeername ++dnl was not found in libc. ++dnl ++dnl NOTE: IF WE FIND GETPEERNAME, WE ASSUME THAT WE HAVE BIND/CONNECT ++dnl AS WELL ++dnl ++AC_DEFUN(BASH_CHECK_LIB_SOCKET, ++[ ++if test "X$bash_cv_have_socklib" = "X"; then ++_bash_needmsg= ++else ++AC_MSG_CHECKING(for socket library) ++_bash_needmsg=yes ++fi ++AC_CACHE_VAL(bash_cv_have_socklib, ++[AC_CHECK_LIB(socket, getpeername, ++ bash_cv_have_socklib=yes, bash_cv_have_socklib=no, -lnsl)]) ++if test "X$_bash_needmsg" = Xyes; then ++ AC_MSG_RESULT($bash_cv_have_socklib) ++ _bash_needmsg= ++fi ++if test $bash_cv_have_socklib = yes; then ++ # check for libnsl, add it to LIBS if present ++ if test "X$bash_cv_have_libnsl" = "X"; then ++ _bash_needmsg= ++ else ++ AC_MSG_CHECKING(for libnsl) ++ _bash_needmsg=yes ++ fi ++ AC_CACHE_VAL(bash_cv_have_libnsl, ++ [AC_CHECK_LIB(nsl, t_open, ++ bash_cv_have_libnsl=yes, bash_cv_have_libnsl=no)]) ++ if test "X$_bash_needmsg" = Xyes; then ++ AC_MSG_RESULT($bash_cv_have_libnsl) ++ _bash_needmsg= ++ fi ++ if test $bash_cv_have_libnsl = yes; then ++ LIBS="-lsocket -lnsl $LIBS" ++ else ++ LIBS="-lsocket $LIBS" ++ fi ++ AC_DEFINE(HAVE_LIBSOCKET) ++ AC_DEFINE(HAVE_GETPEERNAME) ++fi ++]) ++ ++AC_DEFUN(BASH_STRUCT_DIRENT_D_INO, ++[AC_REQUIRE([AC_HEADER_DIRENT]) ++AC_MSG_CHECKING(if struct dirent has a d_ino member) ++AC_CACHE_VAL(bash_cv_dirent_has_dino, ++[AC_TRY_COMPILE([ ++#include ++#include ++#ifdef HAVE_UNISTD_H ++# include ++#endif /* HAVE_UNISTD_H */ ++#if defined(HAVE_DIRENT_H) ++# include ++#else ++# define dirent direct ++# ifdef HAVE_SYS_NDIR_H ++# include ++# endif /* SYSNDIR */ ++# ifdef HAVE_SYS_DIR_H ++# include ++# endif /* SYSDIR */ ++# ifdef HAVE_NDIR_H ++# include ++# endif ++#endif /* HAVE_DIRENT_H */ ++],[ ++struct dirent d; int z; z = d.d_ino; ++], bash_cv_dirent_has_dino=yes, bash_cv_dirent_has_dino=no)]) ++AC_MSG_RESULT($bash_cv_dirent_has_dino) ++if test $bash_cv_dirent_has_dino = yes; then ++AC_DEFINE(STRUCT_DIRENT_HAS_D_INO) ++fi ++]) ++ ++AC_DEFUN(BASH_STRUCT_DIRENT_D_FILENO, ++[AC_REQUIRE([AC_HEADER_DIRENT]) ++AC_MSG_CHECKING(if struct dirent has a d_fileno member) ++AC_CACHE_VAL(bash_cv_dirent_has_d_fileno, ++[AC_TRY_COMPILE([ ++#include ++#include ++#ifdef HAVE_UNISTD_H ++# include ++#endif /* HAVE_UNISTD_H */ ++#if defined(HAVE_DIRENT_H) ++# include ++#else ++# define dirent direct ++# ifdef HAVE_SYS_NDIR_H ++# include ++# endif /* SYSNDIR */ ++# ifdef HAVE_SYS_DIR_H ++# include ++# endif /* SYSDIR */ ++# ifdef HAVE_NDIR_H ++# include ++# endif ++#endif /* HAVE_DIRENT_H */ ++],[ ++struct dirent d; int z; z = d.d_fileno; ++], bash_cv_dirent_has_d_fileno=yes, bash_cv_dirent_has_d_fileno=no)]) ++AC_MSG_RESULT($bash_cv_dirent_has_d_fileno) ++if test $bash_cv_dirent_has_d_fileno = yes; then ++AC_DEFINE(STRUCT_DIRENT_HAS_D_FILENO) ++fi ++]) ++ ++AC_DEFUN(BASH_STRUCT_TIMEVAL, ++[AC_MSG_CHECKING(for struct timeval in sys/time.h and time.h) ++AC_CACHE_VAL(bash_cv_struct_timeval, ++[ ++AC_EGREP_HEADER(struct timeval, sys/time.h, ++ bash_cv_struct_timeval=yes, ++ AC_EGREP_HEADER(struct timeval, time.h, ++ bash_cv_struct_timeval=yes, ++ bash_cv_struct_timeval=no)) ++]) ++AC_MSG_RESULT($bash_cv_struct_timeval) ++if test $bash_cv_struct_timeval = yes; then ++ AC_DEFINE(HAVE_TIMEVAL) ++fi ++]) ++ ++AC_DEFUN(BASH_STRUCT_WINSIZE, ++[AC_MSG_CHECKING(for struct winsize in sys/ioctl.h and termios.h) ++AC_CACHE_VAL(bash_cv_struct_winsize_header, ++[AC_TRY_COMPILE([#include ++#include ], [struct winsize x;], ++ bash_cv_struct_winsize_header=ioctl_h, ++ [AC_TRY_COMPILE([#include ++#include ], [struct winsize x;], ++ bash_cv_struct_winsize_header=termios_h, bash_cv_struct_winsize_header=other) ++])]) ++if test $bash_cv_struct_winsize_header = ioctl_h; then ++ AC_MSG_RESULT(sys/ioctl.h) ++ AC_DEFINE(STRUCT_WINSIZE_IN_SYS_IOCTL) ++elif test $bash_cv_struct_winsize_header = termios_h; then ++ AC_MSG_RESULT(termios.h) ++ AC_DEFINE(STRUCT_WINSIZE_IN_TERMIOS) ++else ++ AC_MSG_RESULT(not found) ++fi ++]) ++ ++dnl Check type of signal routines (posix, 4.2bsd, 4.1bsd or v7) ++AC_DEFUN(BASH_SYS_SIGNAL_VINTAGE, ++[AC_REQUIRE([AC_TYPE_SIGNAL]) ++AC_MSG_CHECKING(for type of signal functions) ++AC_CACHE_VAL(bash_cv_signal_vintage, ++[ ++ AC_TRY_LINK([#include ],[ ++ sigset_t ss; ++ struct sigaction sa; ++ sigemptyset(&ss); sigsuspend(&ss); ++ sigaction(SIGINT, &sa, (struct sigaction *) 0); ++ sigprocmask(SIG_BLOCK, &ss, (sigset_t *) 0); ++ ], bash_cv_signal_vintage=posix, ++ [ ++ AC_TRY_LINK([#include ], [ ++ int mask = sigmask(SIGINT); ++ sigsetmask(mask); sigblock(mask); sigpause(mask); ++ ], bash_cv_signal_vintage=4.2bsd, ++ [ ++ AC_TRY_LINK([ ++ #include ++ RETSIGTYPE foo() { }], [ ++ int mask = sigmask(SIGINT); ++ sigset(SIGINT, foo); sigrelse(SIGINT); ++ sighold(SIGINT); sigpause(SIGINT); ++ ], bash_cv_signal_vintage=svr3, bash_cv_signal_vintage=v7 ++ )] ++ )] ++) ++]) ++AC_MSG_RESULT($bash_cv_signal_vintage) ++if test "$bash_cv_signal_vintage" = posix; then ++AC_DEFINE(HAVE_POSIX_SIGNALS) ++elif test "$bash_cv_signal_vintage" = "4.2bsd"; then ++AC_DEFINE(HAVE_BSD_SIGNALS) ++elif test "$bash_cv_signal_vintage" = svr3; then ++AC_DEFINE(HAVE_USG_SIGHOLD) ++fi ++]) ++ ++dnl Check if the pgrp of setpgrp() can't be the pid of a zombie process. ++AC_DEFUN(BASH_SYS_PGRP_SYNC, ++[AC_REQUIRE([AC_FUNC_GETPGRP]) ++AC_MSG_CHECKING(whether pgrps need synchronization) ++AC_CACHE_VAL(bash_cv_pgrp_pipe, ++[AC_TRY_RUN([ ++#ifdef HAVE_UNISTD_H ++# include ++#endif ++main() ++{ ++# ifdef GETPGRP_VOID ++# define getpgID() getpgrp() ++# else ++# define getpgID() getpgrp(0) ++# define setpgid(x,y) setpgrp(x,y) ++# endif ++ int pid1, pid2, fds[2]; ++ int status; ++ char ok; ++ ++ switch (pid1 = fork()) { ++ case -1: ++ exit(1); ++ case 0: ++ setpgid(0, getpid()); ++ exit(0); ++ } ++ setpgid(pid1, pid1); ++ ++ sleep(2); /* let first child die */ ++ ++ if (pipe(fds) < 0) ++ exit(2); ++ ++ switch (pid2 = fork()) { ++ case -1: ++ exit(3); ++ case 0: ++ setpgid(0, pid1); ++ ok = getpgID() == pid1; ++ write(fds[1], &ok, 1); ++ exit(0); ++ } ++ setpgid(pid2, pid1); ++ ++ close(fds[1]); ++ if (read(fds[0], &ok, 1) != 1) ++ exit(4); ++ wait(&status); ++ wait(&status); ++ exit(ok ? 0 : 5); ++} ++], bash_cv_pgrp_pipe=no,bash_cv_pgrp_pipe=yes, ++ [AC_MSG_WARN(cannot check pgrp synchronization if cross compiling -- defaulting to no) ++ bash_cv_pgrp_pipe=no]) ++]) ++AC_MSG_RESULT($bash_cv_pgrp_pipe) ++if test $bash_cv_pgrp_pipe = yes; then ++AC_DEFINE(PGRP_PIPE) ++fi ++]) ++ ++AC_DEFUN(BASH_SYS_REINSTALL_SIGHANDLERS, ++[AC_REQUIRE([AC_TYPE_SIGNAL]) ++AC_REQUIRE([BASH_SYS_SIGNAL_VINTAGE]) ++AC_MSG_CHECKING([if signal handlers must be reinstalled when invoked]) ++AC_CACHE_VAL(bash_cv_must_reinstall_sighandlers, ++[AC_TRY_RUN([ ++#include ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++ ++typedef RETSIGTYPE sigfunc(); ++ ++int nsigint; ++ ++#ifdef HAVE_POSIX_SIGNALS ++sigfunc * ++set_signal_handler(sig, handler) ++ int sig; ++ sigfunc *handler; ++{ ++ struct sigaction act, oact; ++ act.sa_handler = handler; ++ act.sa_flags = 0; ++ sigemptyset (&act.sa_mask); ++ sigemptyset (&oact.sa_mask); ++ sigaction (sig, &act, &oact); ++ return (oact.sa_handler); ++} ++#else ++#define set_signal_handler(s, h) signal(s, h) ++#endif ++ ++RETSIGTYPE ++sigint(s) ++int s; ++{ ++ nsigint++; ++} ++ ++main() ++{ ++ nsigint = 0; ++ set_signal_handler(SIGINT, sigint); ++ kill((int)getpid(), SIGINT); ++ kill((int)getpid(), SIGINT); ++ exit(nsigint != 2); ++} ++], bash_cv_must_reinstall_sighandlers=no, bash_cv_must_reinstall_sighandlers=yes, ++ [AC_MSG_WARN(cannot check signal handling if cross compiling -- defaulting to no) ++ bash_cv_must_reinstall_sighandlers=no] ++)]) ++AC_MSG_RESULT($bash_cv_must_reinstall_sighandlers) ++if test $bash_cv_must_reinstall_sighandlers = yes; then ++AC_DEFINE(MUST_REINSTALL_SIGHANDLERS) ++fi ++]) ++ ++dnl check that some necessary job control definitions are present ++AC_DEFUN(BASH_SYS_JOB_CONTROL_MISSING, ++[AC_REQUIRE([BASH_SYS_SIGNAL_VINTAGE]) ++AC_MSG_CHECKING(for presence of necessary job control definitions) ++AC_CACHE_VAL(bash_cv_job_control_missing, ++[AC_TRY_RUN([ ++#include ++#ifdef HAVE_SYS_WAIT_H ++#include ++#endif ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++#include ++ ++/* Add more tests in here as appropriate. */ ++main() ++{ ++/* signal type */ ++#if !defined (HAVE_POSIX_SIGNALS) && !defined (HAVE_BSD_SIGNALS) ++exit(1); ++#endif ++ ++/* signals and tty control. */ ++#if !defined (SIGTSTP) || !defined (SIGSTOP) || !defined (SIGCONT) ++exit (1); ++#endif ++ ++/* process control */ ++#if !defined (WNOHANG) || !defined (WUNTRACED) ++exit(1); ++#endif ++ ++/* Posix systems have tcgetpgrp and waitpid. */ ++#if defined (_POSIX_VERSION) && !defined (HAVE_TCGETPGRP) ++exit(1); ++#endif ++ ++#if defined (_POSIX_VERSION) && !defined (HAVE_WAITPID) ++exit(1); ++#endif ++ ++/* Other systems have TIOCSPGRP/TIOCGPRGP and wait3. */ ++#if !defined (_POSIX_VERSION) && !defined (HAVE_WAIT3) ++exit(1); ++#endif ++ ++exit(0); ++}], bash_cv_job_control_missing=present, bash_cv_job_control_missing=missing, ++ [AC_MSG_WARN(cannot check job control if cross-compiling -- defaulting to missing) ++ bash_cv_job_control_missing=missing] ++)]) ++AC_MSG_RESULT($bash_cv_job_control_missing) ++if test $bash_cv_job_control_missing = missing; then ++AC_DEFINE(JOB_CONTROL_MISSING) ++fi ++]) ++ ++dnl check whether named pipes are present ++dnl this requires a previous check for mkfifo, but that is awkward to specify ++AC_DEFUN(BASH_SYS_NAMED_PIPES, ++[AC_MSG_CHECKING(for presence of named pipes) ++AC_CACHE_VAL(bash_cv_sys_named_pipes, ++[AC_TRY_RUN([ ++#include ++#include ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++ ++/* Add more tests in here as appropriate. */ ++main() ++{ ++int fd, err; ++ ++#if defined (HAVE_MKFIFO) ++exit (0); ++#endif ++ ++#if !defined (S_IFIFO) && (defined (_POSIX_VERSION) && !defined (S_ISFIFO)) ++exit (1); ++#endif ++ ++#if defined (NeXT) ++exit (1); ++#endif ++err = mkdir("/tmp/bash-aclocal", 0700); ++if (err < 0) { ++ perror ("mkdir"); ++ exit(1); ++} ++fd = mknod ("/tmp/bash-aclocal/sh-np-autoconf", 0666 | S_IFIFO, 0); ++if (fd == -1) { ++ rmdir ("/tmp/bash-aclocal"); ++ exit (1); ++} ++close(fd); ++unlink ("/tmp/bash-aclocal/sh-np-autoconf"); ++rmdir ("/tmp/bash-aclocal"); ++exit(0); ++}], bash_cv_sys_named_pipes=present, bash_cv_sys_named_pipes=missing, ++ [AC_MSG_WARN(cannot check for named pipes if cross-compiling -- defaulting to missing) ++ bash_cv_sys_named_pipes=missing] ++)]) ++AC_MSG_RESULT($bash_cv_sys_named_pipes) ++if test $bash_cv_sys_named_pipes = missing; then ++AC_DEFINE(NAMED_PIPES_MISSING) ++fi ++]) ++ ++AC_DEFUN(BASH_SYS_DEFAULT_MAIL_DIR, ++[AC_MSG_CHECKING(for default mail directory) ++AC_CACHE_VAL(bash_cv_mail_dir, ++[if test -d /var/mail; then ++ bash_cv_mail_dir=/var/mail ++ elif test -d /var/spool/mail; then ++ bash_cv_mail_dir=/var/spool/mail ++ elif test -d /usr/mail; then ++ bash_cv_mail_dir=/usr/mail ++ elif test -d /usr/spool/mail; then ++ bash_cv_mail_dir=/usr/spool/mail ++ else ++ bash_cv_mail_dir=unknown ++ fi ++]) ++AC_MSG_RESULT($bash_cv_mail_dir) ++AC_DEFINE_UNQUOTED(DEFAULT_MAIL_DIRECTORY, "$bash_cv_mail_dir") ++]) ++ ++AC_DEFUN(BASH_HAVE_TIOCGWINSZ, ++[AC_MSG_CHECKING(for TIOCGWINSZ in sys/ioctl.h) ++AC_CACHE_VAL(bash_cv_tiocgwinsz_in_ioctl, ++[AC_TRY_COMPILE([#include ++#include ], [int x = TIOCGWINSZ;], ++ bash_cv_tiocgwinsz_in_ioctl=yes,bash_cv_tiocgwinsz_in_ioctl=no)]) ++AC_MSG_RESULT($bash_cv_tiocgwinsz_in_ioctl) ++if test $bash_cv_tiocgwinsz_in_ioctl = yes; then ++AC_DEFINE(GWINSZ_IN_SYS_IOCTL) ++fi ++]) ++ ++AC_DEFUN(BASH_HAVE_TIOCSTAT, ++[AC_MSG_CHECKING(for TIOCSTAT in sys/ioctl.h) ++AC_CACHE_VAL(bash_cv_tiocstat_in_ioctl, ++[AC_TRY_COMPILE([#include ++#include ], [int x = TIOCSTAT;], ++ bash_cv_tiocstat_in_ioctl=yes,bash_cv_tiocstat_in_ioctl=no)]) ++AC_MSG_RESULT($bash_cv_tiocstat_in_ioctl) ++if test $bash_cv_tiocstat_in_ioctl = yes; then ++AC_DEFINE(TIOCSTAT_IN_SYS_IOCTL) ++fi ++]) ++ ++AC_DEFUN(BASH_HAVE_FIONREAD, ++[AC_MSG_CHECKING(for FIONREAD in sys/ioctl.h) ++AC_CACHE_VAL(bash_cv_fionread_in_ioctl, ++[AC_TRY_COMPILE([#include ++#include ], [int x = FIONREAD;], ++ bash_cv_fionread_in_ioctl=yes,bash_cv_fionread_in_ioctl=no)]) ++AC_MSG_RESULT($bash_cv_fionread_in_ioctl) ++if test $bash_cv_fionread_in_ioctl = yes; then ++AC_DEFINE(FIONREAD_IN_SYS_IOCTL) ++fi ++]) ++ ++dnl ++dnl See if speed_t is declared in . Some versions of linux ++dnl require a definition of speed_t each time is included, ++dnl but you can only get speed_t if you include (on some ++dnl versions) or (on others). ++dnl ++AC_DEFUN(BASH_CHECK_SPEED_T, ++[AC_MSG_CHECKING(for speed_t in sys/types.h) ++AC_CACHE_VAL(bash_cv_speed_t_in_sys_types, ++[AC_TRY_COMPILE([#include ], [speed_t x;], ++ bash_cv_speed_t_in_sys_types=yes,bash_cv_speed_t_in_sys_types=no)]) ++AC_MSG_RESULT($bash_cv_speed_t_in_sys_types) ++if test $bash_cv_speed_t_in_sys_types = yes; then ++AC_DEFINE(SPEED_T_IN_SYS_TYPES) ++fi ++]) ++ ++AC_DEFUN(BASH_CHECK_GETPW_FUNCS, ++[AC_MSG_CHECKING(whether getpw functions are declared in pwd.h) ++AC_CACHE_VAL(bash_cv_getpw_declared, ++[AC_EGREP_CPP(getpwuid, ++[ ++#include ++#ifdef HAVE_UNISTD_H ++# include ++#endif ++#include ++], ++bash_cv_getpw_declared=yes,bash_cv_getpw_declared=no)]) ++AC_MSG_RESULT($bash_cv_getpw_declared) ++if test $bash_cv_getpw_declared = yes; then ++AC_DEFINE(HAVE_GETPW_DECLS) ++fi ++]) ++ ++AC_DEFUN(BASH_CHECK_DEV_FD, ++[AC_MSG_CHECKING(whether /dev/fd is available) ++AC_CACHE_VAL(bash_cv_dev_fd, ++[if test -d /dev/fd && test -r /dev/fd/0; then ++ bash_cv_dev_fd=standard ++ elif test -d /proc/self/fd && test -r /proc/self/fd/0; then ++ bash_cv_dev_fd=whacky ++ else ++ bash_cv_dev_fd=absent ++ fi ++]) ++AC_MSG_RESULT($bash_cv_dev_fd) ++if test $bash_cv_dev_fd = "standard"; then ++ AC_DEFINE(HAVE_DEV_FD) ++ AC_DEFINE(DEV_FD_PREFIX, "/dev/fd/") ++elif test $bash_cv_dev_fd = "whacky"; then ++ AC_DEFINE(HAVE_DEV_FD) ++ AC_DEFINE(DEV_FD_PREFIX, "/proc/self/fd/") ++fi ++]) ++ ++AC_DEFUN(BASH_CHECK_DEV_STDIN, ++[AC_MSG_CHECKING(whether /dev/stdin stdout stderr are available) ++AC_CACHE_VAL(bash_cv_dev_stdin, ++[if test -d /dev/fd && test -r /dev/stdin; then ++ bash_cv_dev_stdin=present ++ elif test -d /proc/self/fd && test -r /dev/stdin; then ++ bash_cv_dev_stdin=present ++ else ++ bash_cv_dev_stdin=absent ++ fi ++]) ++AC_MSG_RESULT($bash_cv_dev_stdin) ++if test $bash_cv_dev_stdin = "present"; then ++ AC_DEFINE(HAVE_DEV_STDIN) ++fi ++]) ++ ++dnl ++dnl Check if HPUX needs _KERNEL defined for RLIMIT_* definitions ++dnl ++AC_DEFUN(BASH_CHECK_KERNEL_RLIMIT, ++[AC_MSG_CHECKING([whether $host_os needs _KERNEL for RLIMIT defines]) ++AC_CACHE_VAL(bash_cv_kernel_rlimit, ++[AC_TRY_COMPILE([ ++#include ++#include ++], ++[ ++ int f; ++ f = RLIMIT_DATA; ++], bash_cv_kernel_rlimit=no, ++[AC_TRY_COMPILE([ ++#include ++#define _KERNEL ++#include ++#undef _KERNEL ++], ++[ ++ int f; ++ f = RLIMIT_DATA; ++], bash_cv_kernel_rlimit=yes, bash_cv_kernel_rlimit=no)] ++)]) ++AC_MSG_RESULT($bash_cv_kernel_rlimit) ++if test $bash_cv_kernel_rlimit = yes; then ++AC_DEFINE(RLIMIT_NEEDS_KERNEL) ++fi ++]) ++ ++dnl ++dnl Check for 64-bit off_t -- used for malloc alignment ++dnl ++dnl C does not allow duplicate case labels, so the compile will fail if ++dnl sizeof(off_t) is > 4. ++dnl ++AC_DEFUN(BASH_CHECK_OFF_T_64, ++[AC_CACHE_CHECK(for 64-bit off_t, bash_cv_off_t_64, ++AC_TRY_COMPILE([ ++#ifdef HAVE_UNISTD_H ++#include ++#endif ++#include ++],[ ++switch (0) case 0: case (sizeof (off_t) <= 4):; ++], bash_cv_off_t_64=no, bash_cv_off_t_64=yes)) ++if test $bash_cv_off_t_64 = yes; then ++ AC_DEFINE(HAVE_OFF_T_64) ++fi]) ++ ++AC_DEFUN(BASH_CHECK_RTSIGS, ++[AC_MSG_CHECKING(for unusable real-time signals due to large values) ++AC_CACHE_VAL(bash_cv_unusable_rtsigs, ++[AC_TRY_RUN([ ++#include ++#include ++ ++#ifndef NSIG ++# define NSIG 64 ++#endif ++ ++main () ++{ ++ int n_sigs = 2 * NSIG; ++#ifdef SIGRTMIN ++ int rtmin = SIGRTMIN; ++#else ++ int rtmin = 0; ++#endif ++ ++ exit(rtmin < n_sigs); ++}], bash_cv_unusable_rtsigs=yes, bash_cv_unusable_rtsigs=no, ++ [AC_MSG_WARN(cannot check real-time signals if cross compiling -- defaulting to yes) ++ bash_cv_unusable_rtsigs=yes] ++)]) ++AC_MSG_RESULT($bash_cv_unusable_rtsigs) ++if test $bash_cv_unusable_rtsigs = yes; then ++AC_DEFINE(UNUSABLE_RT_SIGNALS) ++fi ++]) ++ ++dnl ++dnl check for availability of multibyte characters and functions ++dnl ++AC_DEFUN(BASH_CHECK_MULTIBYTE, ++[ ++AC_CHECK_HEADERS(wctype.h) ++AC_CHECK_HEADERS(wchar.h) ++AC_CHECK_HEADERS(langinfo.h) ++ ++AC_CHECK_FUNC(mbsrtowcs, AC_DEFINE(HAVE_MBSRTOWCS)) ++AC_CHECK_FUNC(wcwidth, AC_DEFINE(HAVE_WCWIDTH)) ++ ++AC_CACHE_CHECK([for mbstate_t], bash_cv_have_mbstate_t, ++[AC_TRY_RUN([ ++#include ++int ++main () ++{ ++ mbstate_t ps; ++ return 0; ++}], bash_cv_have_mbstate_t=yes, bash_cv_have_mbstate_t=no)]) ++if test $bash_cv_have_mbstate_t = yes; then ++ AC_DEFINE(HAVE_MBSTATE_T) ++fi ++ ++AC_CACHE_CHECK([for nl_langinfo and CODESET], bash_cv_langinfo_codeset, ++[AC_TRY_LINK( ++[#include ], ++[char* cs = nl_langinfo(CODESET);], ++bash_cv_langinfo_codeset=yes, bash_cv_langinfo_codeset=no)]) ++if test $bash_cv_langinfo_codeset = yes; then ++ AC_DEFINE(HAVE_LANGINFO_CODESET) ++fi ++ ++]) ++ ++dnl need: prefix exec_prefix libdir includedir CC TERMCAP_LIB ++dnl require: ++dnl AC_PROG_CC ++dnl BASH_CHECK_LIB_TERMCAP ++ ++AC_DEFUN(RL_LIB_READLINE_VERSION, ++[ ++AC_REQUIRE([BASH_CHECK_LIB_TERMCAP]) ++ ++AC_MSG_CHECKING([version of installed readline library]) ++ ++# What a pain in the ass this is. ++ ++# save cpp and ld options ++_save_CFLAGS="$CFLAGS" ++_save_LDFLAGS="$LDFLAGS" ++_save_LIBS="$LIBS" ++ ++# Don't set ac_cv_rl_prefix if the caller has already assigned a value. This ++# allows the caller to do something like $_rl_prefix=$withval if the user ++# specifies --with-installed-readline=PREFIX as an argument to configure ++ ++if test -z "$ac_cv_rl_prefix"; then ++test "x$prefix" = xNONE && ac_cv_rl_prefix=$ac_default_prefix || ac_cv_rl_prefix=${prefix} ++fi ++ ++eval ac_cv_rl_includedir=${ac_cv_rl_prefix}/include ++eval ac_cv_rl_libdir=${ac_cv_rl_prefix}/lib ++ ++LIBS="$LIBS -lreadline ${TERMCAP_LIB}" ++CFLAGS="$CFLAGS -I${ac_cv_rl_includedir}" ++LDFLAGS="$LDFLAGS -L${ac_cv_rl_libdir}" ++ ++AC_TRY_RUN([ ++#include ++#include ++ ++main() ++{ ++ FILE *fp; ++ fp = fopen("conftest.rlv", "w"); ++ if (fp == 0) exit(1); ++ fprintf(fp, "%s\n", rl_library_version ? rl_library_version : "0.0"); ++ fclose(fp); ++ exit(0); ++} ++], ++ac_cv_rl_version=`cat conftest.rlv`, ++ac_cv_rl_version='0.0', ++ac_cv_rl_version='4.2') ++ ++CFLAGS="$_save_CFLAGS" ++LDFLAGS="$_save_LDFLAGS" ++LIBS="$_save_LIBS" ++ ++RL_MAJOR=0 ++RL_MINOR=0 ++ ++# ( ++case "$ac_cv_rl_version" in ++2*|3*|4*|5*|6*|7*|8*|9*) ++ RL_MAJOR=`echo $ac_cv_rl_version | sed 's:\..*$::'` ++ RL_MINOR=`echo $ac_cv_rl_version | sed -e 's:^.*\.::' -e 's:[[a-zA-Z]]*$::'` ++ ;; ++esac ++ ++# ((( ++case $RL_MAJOR in ++[[0-9][0-9]]) _RL_MAJOR=$RL_MAJOR ;; ++[[0-9]]) _RL_MAJOR=0$RL_MAJOR ;; ++*) _RL_MAJOR=00 ;; ++esac ++ ++# ((( ++case $RL_MINOR in ++[[0-9][0-9]]) _RL_MINOR=$RL_MINOR ;; ++[[0-9]]) _RL_MINOR=0$RL_MINOR ;; ++*) _RL_MINOR=00 ;; ++esac ++ ++RL_VERSION="0x${_RL_MAJOR}${_RL_MINOR}" ++ ++# Readline versions greater than 4.2 have these defines in readline.h ++ ++if test $ac_cv_rl_version = '0.0' ; then ++ AC_MSG_WARN([Could not test version of installed readline library.]) ++elif test $RL_MAJOR -gt 4 || { test $RL_MAJOR = 4 && test $RL_MINOR -gt 2 ; } ; then ++ # set these for use by the caller ++ RL_PREFIX=$ac_cv_rl_prefix ++ RL_LIBDIR=$ac_cv_rl_libdir ++ RL_INCLUDEDIR=$ac_cv_rl_includedir ++ AC_MSG_RESULT($ac_cv_rl_version) ++else ++ ++AC_DEFINE_UNQUOTED(RL_READLINE_VERSION, $RL_VERSION, [encoded version of the installed readline library]) ++AC_DEFINE_UNQUOTED(RL_VERSION_MAJOR, $RL_MAJOR, [major version of installed readline library]) ++AC_DEFINE_UNQUOTED(RL_VERSION_MINOR, $RL_MINOR, [minor version of installed readline library]) ++ ++AC_SUBST(RL_VERSION) ++AC_SUBST(RL_MAJOR) ++AC_SUBST(RL_MINOR) ++ ++# set these for use by the caller ++RL_PREFIX=$ac_cv_rl_prefix ++RL_LIBDIR=$ac_cv_rl_libdir ++RL_INCLUDEDIR=$ac_cv_rl_includedir ++ ++AC_MSG_RESULT($ac_cv_rl_version) ++ ++fi ++]) +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/bootstrap.sh cvs/bootstrap.sh +--- bash-2.05b/bootstrap.sh 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/bootstrap.sh 2003-09-01 21:57:32.000000000 +0200 +@@ -0,0 +1,55 @@ ++#!/bin/sh ++# $Id: bootstrap.sh,v 1.4 2003/08/29 15:29:38 snikkt Exp $ ++# Run this to generate all the initial Makefiles, etc. ++ ++# Check how echo works in this /bin/sh ++case `echo -n` in ++-n) _echo_n= _echo_c='\c';; ++*) _echo_n=-n _echo_c=;; ++esac ++ ++srcdir=`dirname $0` ++test -z "$srcdir" && srcdir=. ++ ++(test -f $srcdir/configure.in) || { ++ echo -n "*** Error ***: Directory "\`$srcdir\'" does not look like the" ++ echo " top-level directory" ++ exit 1 ++} ++ ++(echo $_echo_n " + Running aclocal: $_echo_c"; \ ++ aclocal -I . -I debugger ; \ ++ echo "done.") ++rc=$? ++(test -n $rc ) || exit $rc ++ ++(echo $_echo_n " + Running autoconf: $_echo_c"; \ ++ autoconf; \ ++ echo "done.") ++ ++rc=$? ++(test -n $rc ) || exit $rc ++ ++(echo $_echo_n " + Running aclocal in debugger: $_echo_c"; \ ++ cd debugger && aclocal -I .; \ ++ echo "done.") ++rc=$? ++(test -n $rc ) || exit $rc ++ ++(echo $_echo_n " + Running autoconf in debugger: $_echo_c"; \ ++ cd debugger && autoconf; \ ++ echo "done.") ++rc=$? ++(test -n $rc ) || exit $rc ++ ++(echo $_echo_n " + Running automake in debugger: $_echo_c"; \ ++ cd debugger && automake --add-missing; \ ++ echo "done.") ++rc=$? ++(test -n $rc ) || exit $rc ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** ++ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/Makefile.in cvs/builtins/Makefile.in +--- bash-2.05b/builtins/Makefile.in 2002-04-23 15:24:23.000000000 +0200 ++++ cvs/builtins/Makefile.in 2002-09-19 20:49:51.000000000 +0200 +@@ -118,7 +118,7 @@ + $(srcdir)/times.def $(srcdir)/trap.def $(srcdir)/type.def \ + $(srcdir)/ulimit.def $(srcdir)/umask.def $(srcdir)/wait.def \ + $(srcdir)/reserved.def $(srcdir)/pushd.def $(srcdir)/shopt.def \ +- $(srcdir)/printf.def $(srcdir)/complete.def ++ $(srcdir)/printf.def $(srcdir)/complete.def $(srcdir)/caller.def + + STATIC_SOURCE = common.c evalstring.c evalfile.c getopt.c bashgetopt.c \ + getopt.h +@@ -130,7 +130,8 @@ + exit.o fc.o fg_bg.o hash.o help.o history.o jobs.o kill.o let.o \ + pushd.o read.o return.o set.o setattr.o shift.o source.o \ + suspend.o test.o times.o trap.o type.o ulimit.o umask.o \ +- wait.o getopts.o shopt.o printf.o getopt.o bashgetopt.o complete.o ++ wait.o getopts.o shopt.o printf.o getopt.o bashgetopt.o \ ++ complete.o caller.o + + CREATED_FILES = builtext.h builtins.c psize.aux pipesize.h + +@@ -214,48 +215,6 @@ + + $(OFILES): $(MKBUILTINS) ../config.h + +-alias.o: alias.def +-bind.o: bind.def +-break.o: break.def +-builtin.o: builtin.def +-cd.o: cd.def +-colon.o: colon.def +-command.o: command.def +-declare.o: declare.def +-echo.o: echo.def +-enable.o: enable.def +-eval.o: eval.def +-exec.o: exec.def +-exit.o: exit.def +-fc.o: fc.def +-fg_bg.o: fg_bg.def +-hash.o: hash.def +-help.o: help.def +-history.o: history.def +-jobs.o: jobs.def +-kill.o: kill.def +-let.o: let.def +-printf.o: printf.def +-pushd.o: pushd.def +-read.o: read.def +-return.o: return.def +-set.o: set.def +-setattr.o: setattr.def +-shift.o: shift.def +-shopt.o: shopt.def +-source.o: source.def +-suspend.o: suspend.def +-test.o: test.def +-times.o: times.def +-trap.o: trap.def +-type.o: type.def +-ulimit.o: ulimit.def +-umask.o: umask.def +-wait.o: wait.def +-getopts.o: getopts.def +-reserved.o: reserved.def +-complete.o: complete.def +- + # C files + bashgetopt.o: ../config.h $(topdir)/bashansi.h $(BASHINCDIR)/ansi_stdlib.h + bashgetopt.o: $(topdir)/shell.h $(topdir)/syntax.h $(topdir)/bashjmp.h +@@ -550,3 +509,7 @@ + complete.o: ${srcdir}/common.h ${srcdir}/bashgetopt.h + + #bind.o: $(RL_LIBSRC)chardefs.h $(RL_LIBSRC)readline.h $(RL_LIBSRC)keymaps.h ++ ++#;;; Local Variables: *** ++#;;; mode:makefile *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/caller.def cvs/builtins/caller.def +--- bash-2.05b/builtins/caller.def 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/builtins/caller.def 2003-02-15 00:33:44.000000000 +0100 +@@ -0,0 +1,148 @@ ++This file is caller.def, from which is created caller.c. It implements the ++builtin "caller" in Bash. ++ ++Copyright (C) 2002 Rocky Bernstein for Free Software Foundation, Inc. ++ ++This file is part of GNU Bash, the Bourne Again SHell. ++ ++Bash is free software; you can redistribute it and/or modify it under ++the terms of the GNU General Public License as published by the Free ++Software Foundation; either version 2, or (at your option) any later ++version. ++ ++Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++WARRANTY; without even the implied warranty of MERCHANTABILITY or ++FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++for more details. ++ ++You should have received a copy of the GNU General Public License along ++with Bash; see the file COPYING. If not, write to the Free Software ++Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++$PRODUCES caller.c ++ ++$BUILTIN caller ++$FUNCTION caller_builtin ++$SHORT_DOC caller [EXPR] ++ ++Returns the context of the current subroutine call. ++ ++Without EXPR, returns returns "$line $filename". With EXPR, it ++returns "$line $subroutine $filename", and this extra information that ++can be used used to print a stack trace, ++ ++The value of EXPR indicates how many call frames to go back before the ++current one; the top frame number would be 0. ++$END ++ ++#include ++#include ++#include "chartypes.h" ++#include "bashtypes.h" ++ ++#if defined (HAVE_UNISTD_H) ++# ifdef _MINIX ++# include ++# endif ++# include ++#endif ++ ++#include ++ ++#include "../shell.h" ++#include "common.h" ++#include "builtext.h" ++ ++#ifdef LOADABLE_BUILTIN ++# include "builtins.h" ++#endif ++ ++#if !defined (errno) ++extern int errno; ++#endif /* !errno */ ++ ++int ++caller_builtin (list) ++ WORD_LIST *list; ++{ ++ SHELL_VAR *funcname_var = find_variable ("FUNCNAME"); ++ ARRAY *funcname_a = array_cell (funcname_var); ++ SHELL_VAR *bash_source_var = find_variable ("BASH_SOURCE"); ++ ARRAY *bash_source_a = array_cell (bash_source_var); ++ SHELL_VAR *bash_lineno_var = find_variable ("BASH_LINENO"); ++ ARRAY *bash_lineno_a = array_cell (bash_lineno_var); ++ ARRAY_ELEMENT *ae; ++ char *lineno_str=""; ++ char *source_str=""; ++ char *funcname_str=""; ++ ++ intmax_t num; ++ ++ if (!bash_lineno_a || array_empty(bash_lineno_a)) ++ return (EXECUTION_FAILURE); ++ ++ if (!bash_source_a || array_empty(bash_source_a)) ++ return (EXECUTION_FAILURE); ++ ++ /* If there is no argument list, then give short form: ++ line filename. ++ */ ++ if (list == NULL) { ++ lineno_str=array_reference(bash_lineno_a, 0); ++ source_str=array_reference(bash_source_a, 1); ++ printf("%s %s\n", ++ lineno_str ? lineno_str : "NULL" , ++ source_str ? source_str : "NULL"); ++ return (EXECUTION_SUCCESS); ++ } ++ ++ if (legal_number (list->word->word, &num)) { ++ lineno_str=array_reference(bash_lineno_a, num); ++ source_str=array_reference(bash_source_a, num+1); ++ funcname_str=array_reference(funcname_a, num+1); ++ ++ if (!funcname_a || array_empty(funcname_a) ++ || NULL == lineno_str || NULL == source_str || NULL == funcname_str ) ++ return (EXECUTION_FAILURE); ++ ++ printf("%s %s %s\n", lineno_str, funcname_str, source_str); ++ } else { ++ sh_invalidnum (list->word->word); ++ builtin_usage (); ++ return (EXECUTION_FAILURE); ++ } ++ ++ ++ return (EXECUTION_SUCCESS); ++} ++ ++ ++#ifdef LOADABLE_BUILTIN ++static char *caller_doc[] = { ++ "Returns the context of the current subroutine call.", ++ "", ++ "Without EXPR, returns returns \"$line $filename\". With EXPR, it", ++ "returns \"$line $subroutine $filename\"; this extra information,", ++ "for example, can be used used to print a stack trace.", ++ "", ++ "The value of EXPR indicates how many call frames to go back before the", ++ "current one; the top frame number would be 0.", ++ (char *)NULL ++}; ++ ++struct builtin caller_struct = { ++ "caller", ++ caller_builtin, ++ BUILTIN_ENABLED, ++ caller_doc, ++ "pushd [EXPR]", ++ 0 ++}; ++ ++#endif /* LOADABLE_BUILTIN */ ++ ++/* ++ * Local variables: ++ * mode: C ++ * End: ++ */ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/common.c cvs/builtins/common.c +--- bash-2.05b/builtins/common.c 2002-06-28 18:24:31.000000000 +0200 ++++ cvs/builtins/common.c 2002-10-10 14:34:04.000000000 +0200 +@@ -48,6 +48,7 @@ + #include "../builtins.h" + #include "../input.h" + #include "../execute_cmd.h" ++#include "../variables.h" + #include "../trap.h" + #include "bashgetopt.h" + #include "common.h" +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/common.h cvs/builtins/common.h +--- bash-2.05b/builtins/common.h 2002-05-10 18:25:08.000000000 +0200 ++++ cvs/builtins/common.h 2002-10-10 14:34:04.000000000 +0200 +@@ -77,6 +77,7 @@ + + extern char **make_builtin_argv __P((WORD_LIST *, int *)); + extern void remember_args __P((WORD_LIST *, int)); ++extern void forget_args __P((WORD_LIST *)); + + extern int dollar_vars_changed __P((void)); + extern void set_dollar_vars_unchanged __P((void)); +@@ -145,7 +146,7 @@ + extern WORD_LIST *get_directory_stack __P((void)); + + /* Functions from evalstring.c */ +-extern int parse_and_execute __P((char *, const char *, int)); ++extern int parse_and_execute __P((char *, const char *, int, int)); + extern void parse_and_execute_cleanup __P((void)); + + /* Functions from evalfile.c */ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/declare.def cvs/builtins/declare.def +--- bash-2.05b/builtins/declare.def 2002-04-08 19:20:55.000000000 +0200 ++++ cvs/builtins/declare.def 2002-09-13 21:09:09.000000000 +0200 +@@ -32,7 +32,8 @@ + + -a to make NAMEs arrays (if supported) + -f to select from among function names only +- -F to display function names without definitions ++ -F to display function name, line number and source-file name ++ without definitions + -i to make NAMEs have the `integer' attribute + -r to make NAMEs readonly + -t to make NAMEs have the `trace' attribute +@@ -309,9 +310,15 @@ + /* declare -[Ff] name [name...] */ + if (flags_on == att_function && flags_off == 0) + { +- t = nodefs ? var->name +- : named_function_string (name, function_cell (var), 1); +- printf ("%s\n", t); ++ if (nodefs) { ++ FUNCTION_DEF *shell_fn = find_function_def(var->name); ++ ++ printf ("%s %d %s\n", var->name, shell_fn->line, ++ shell_fn->source_file); ++ } else ++ printf ("%s\n", ++ named_function_string (name, ++ function_cell(var), 1)); + } + else /* declare -[fF] -[rx] name [name...] */ + { +@@ -440,3 +447,9 @@ + : ((any_failed == 0) ? EXECUTION_SUCCESS + : EXECUTION_FAILURE)); + } ++ ++/* ++ * Local variables: ++ * mode: C ++ * End: ++ */ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/eval.def cvs/builtins/eval.def +--- bash-2.05b/builtins/eval.def 2002-04-08 19:21:04.000000000 +0200 ++++ cvs/builtins/eval.def 2002-08-25 23:42:10.000000000 +0200 +@@ -49,5 +49,5 @@ + list = loptend; /* skip over possible `--' */ + + /* Note that parse_and_execute () frees the string it is passed. */ +- return (list ? parse_and_execute (string_list (list), "eval", SEVAL_NOHIST) : EXECUTION_SUCCESS); ++ return (list ? parse_and_execute (string_list (list), "eval", SEVAL_NOHIST, 0) : EXECUTION_SUCCESS); + } +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/evalfile.c cvs/builtins/evalfile.c +--- bash-2.05b/builtins/evalfile.c 2002-02-21 21:08:52.000000000 +0100 ++++ cvs/builtins/evalfile.c 2002-10-10 14:34:04.000000000 +0200 +@@ -1,4 +1,4 @@ +-/* Copyright (C) 1996 Free Software Foundation, Inc. ++/* Copyright (C) 1996,2002 Free Software Foundation, Inc. + + This file is part of GNU Bash, the Bourne Again SHell. + +@@ -80,6 +80,17 @@ + size_t file_size; + sh_vmsg_func_t *errfunc; + ++ SHELL_VAR *funcname_var = find_variable ("FUNCNAME"); ++ ARRAY *funcname_a = array_cell (funcname_var); ++ SHELL_VAR *bash_source_var = find_variable ("BASH_SOURCE"); ++ ARRAY *bash_source_a = array_cell (bash_source_var); ++ SHELL_VAR *bash_lineno_var = find_variable ("BASH_LINENO"); ++ ARRAY *bash_lineno_a = array_cell (bash_lineno_var); ++ SHELL_VAR *bash_argv_var = find_variable ("BASH_ARGV"); ++ ARRAY *bash_argv_a = array_cell (bash_argv_var); ++ SHELL_VAR *bash_argc_var = find_variable ("BASH_ARGC"); ++ ARRAY *bash_argc_a = array_cell (bash_argc_var); ++ + USE_VAR(pflags); + + fd = open (filename, O_RDONLY); +@@ -175,6 +186,11 @@ + + return_catch_flag++; + sourcelevel++; ++ array_shift_element(bash_source_a, savestring(filename)); ++ array_shift_element(bash_lineno_a, itos(executing_line_number ())); ++ array_shift_element(bash_argv_a, savestring(filename)); ++ array_shift_element(bash_argc_a, itos(1)); ++ array_shift_element(funcname_a, savestring("source")); + + /* set the flags to be passed to parse_and_execute */ + pflags = (flags & FEVAL_HISTORY) ? 0 : SEVAL_NOHIST; +@@ -191,8 +207,8 @@ + parse_and_execute_cleanup (); + result = return_catch_value; + } +- else +- result = parse_and_execute (string, filename, pflags); ++ else ++ result = parse_and_execute (string, filename, pflags, 1); + + if (flags & FEVAL_UNWINDPROT) + run_unwind_frame ("_evalfile"); +@@ -205,6 +221,11 @@ + COPY_PROCENV (old_return_catch, return_catch); + } + ++ array_dispose_element(array_unshift_element(bash_source_a)); ++ array_dispose_element(array_unshift_element(bash_lineno_a)); ++ array_dispose_element(array_unshift_element(funcname_a)); ++ array_dispose_element(array_unshift_element(bash_argc_a)); ++ array_dispose_element(array_unshift_element(bash_argv_a)); + return ((flags & FEVAL_BUILTIN) ? result : 1); + } + +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/evalstring.c cvs/builtins/evalstring.c +--- bash-2.05b/builtins/evalstring.c 2002-04-04 20:38:50.000000000 +0200 ++++ cvs/builtins/evalstring.c 2002-09-13 00:21:04.000000000 +0200 +@@ -84,13 +84,17 @@ + (flags & SEVAL_INTERACT) -> interactive = 1; + (flags & SEVAL_NOHIST) -> call bash_history_disable () + (flags & SEVAL_NOFREE) -> don't free STRING when finished ++ ++ reset_line_number causes the line number to be reset to 1, or preserves ++ the line number that we were at inside some other script. + */ + + int +-parse_and_execute (string, from_file, flags) ++parse_and_execute (string, from_file, flags, reset_line_number) + char *string; + const char *from_file; + int flags; ++ int reset_line_number; + { + int code, x; + volatile int should_jump_to_top_level, last_result; +@@ -129,7 +133,16 @@ + end_unwind_frame (); + + parse_and_execute_level++; +- push_stream (1); /* reset the line number */ ++ ++ push_stream (reset_line_number); ++ ++ /* If we don't want to reset the line, number we have to subtract ++ one, because we will add one just before executing the next ++ command. (Resetting sets the line_number to 0; the first line ++ number is 1). ++ */ ++ if (!reset_line_number) line_number--; ++ + indirection_level++; + if (flags & (SEVAL_NONINT|SEVAL_INTERACT)) + interactive = (flags & SEVAL_NONINT) ? 0 : 1; +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/fc.def cvs/builtins/fc.def +--- bash-2.05b/builtins/fc.def 2002-03-19 20:05:18.000000000 +0100 ++++ cvs/builtins/fc.def 2002-08-29 05:27:39.000000000 +0200 +@@ -266,7 +266,7 @@ + + fprintf (stderr, "%s\n", command); + fc_replhist (command); /* replace `fc -s' with command */ +- return (parse_and_execute (command, "fc", SEVAL_NOHIST)); ++ return (parse_and_execute (command, "fc", SEVAL_NOHIST, 0)); + } + + /* This is the second form of the command (the list-or-edit-and-rerun +@@ -367,7 +367,7 @@ + command = (char *)xmalloc (3 + strlen (FC_EDIT_COMMAND) + strlen (fn)); + sprintf (command, "%s %s", FC_EDIT_COMMAND, fn); + } +- retval = parse_and_execute (command, "fc", SEVAL_NOHIST); ++ retval = parse_and_execute (command, "fc", SEVAL_NOHIST, 0); + if (retval != EXECUTION_SUCCESS) + { + unlink (fn); +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/history.def cvs/builtins/history.def +--- bash-2.05b/builtins/history.def 2002-03-19 20:06:33.000000000 +0100 ++++ cvs/builtins/history.def 2003-05-24 22:22:20.000000000 +0200 +@@ -34,16 +34,25 @@ + append the contents to the history list instead. `-a' means + to append history lines from this session to the history file. + Argument `-n' means to read all history lines not already read +-from the history file and append them to the history list. If +-FILENAME is given, then that is used as the history file else +-if $HISTFILE has a value, that is used, else ~/.bash_history. +-If the -s option is supplied, the non-option ARGs are appended to +-the history list as a single entry. The -p option means to perform +-history expansion on each ARG and display the result, without storing +-anything in the history list. ++from the history file and append them to the history list. ++ ++If FILENAME is given, then that is used as the history file else if $HISTFILE ++has a value, that is used, else ~/.bash_history. If the -s ++option is supplied, the non-option ARGs are appended to the history ++list as a single entry. The -p option means to perform history ++expansion on each ARG and display the result, without storing anything ++in the history list. ++ ++The format of the timestamp shown in a command history is controlled ++by the $HISTTIMEFORMAT environment variable via strftime; the default ++value is '%a %T ' (abbreviated weekday name and 24-hour time with ++seconds). Note the trailing blank. See strftime(3) for a list of time ++conversion specifiers. ++ + $END + + #include ++#include + + #if defined (HISTORY) + #include "../bashtypes.h" +@@ -214,6 +223,19 @@ + #define histline(i) (hlist[(i)]->line) + #define histdata(i) (hlist[(i)]->data) + ++static char * ++histtime(HIST_ENTRY *hlist, const char *hist_time_format) ++{ ++ static char time_str[100]; ++ time_t t = history_get_time(hlist); ++ ++ if (t != 0) ++ strftime(time_str, 100, hist_time_format, localtime(&t)); ++ else ++ strcpy(time_str, "?? "); ++ return time_str; ++} ++ + static void + display_history (list) + WORD_LIST *list; +@@ -235,6 +257,11 @@ + + if (hlist) + { ++ char *hist_time_format = get_string_value ("HISTTIMEFORMAT"); ++ ++ if (hist_time_format == NULL) ++ hist_time_format = "%a %T "; ++ + for (i = 0; hlist[i]; i++) + ; + +@@ -246,9 +273,9 @@ + while (hlist[i]) + { + QUIT; +- printf ("%5d%c %s\n", i + history_base, ++ printf ("%5d%c %s%s\n", i + history_base, + histdata(i) ? '*' : ' ', +- histline(i)); ++ histtime(hlist[i], hist_time_format), histline(i)); + i++; + } + } +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/return.def cvs/builtins/return.def +--- bash-2.05b/builtins/return.def 2002-04-08 19:23:25.000000000 +0200 ++++ cvs/builtins/return.def 2002-11-02 14:12:05.000000000 +0100 +@@ -54,6 +54,7 @@ + { + return_catch_value = get_exitstat (list); + ++ run_return_trap (); + if (return_catch_flag) + longjmp (return_catch, 1); + else +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/set.def cvs/builtins/set.def +--- bash-2.05b/builtins/set.def 2002-07-10 22:17:20.000000000 +0200 ++++ cvs/builtins/set.def 2002-11-06 04:43:08.000000000 +0100 +@@ -76,7 +76,9 @@ + #if defined (READLINE) + emacs use an emacs-style line editing interface + #endif /* READLINE */ ++ fntrace same as -d + errexit same as -e ++ errtrace same as -E + hashall same as -h + #if defined (BANG_HISTORY) + histexpand same as -H +@@ -172,6 +174,8 @@ + { "emacs", '\0', (int *)NULL, set_edit_mode, get_edit_mode }, + #endif + { "errexit", 'e', (int *)NULL, (setopt_set_func_t *)NULL, (setopt_get_func_t *)NULL }, ++ { "errtrace", 'E', (int *)NULL, (setopt_set_func_t *)NULL, (setopt_get_func_t *)NULL }, ++ { "fntrace", 'd', (int *)NULL, (setopt_set_func_t *)NULL, (setopt_get_func_t *)NULL }, + { "hashall", 'h', (int *)NULL, (setopt_set_func_t *)NULL, (setopt_get_func_t *)NULL }, + #if defined (BANG_HISTORY) + { "histexpand", 'H', (int *)NULL, (setopt_set_func_t *)NULL, (setopt_get_func_t *)NULL }, +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/shift.def cvs/builtins/shift.def +--- bash-2.05b/builtins/shift.def 2002-03-19 20:08:56.000000000 +0100 ++++ cvs/builtins/shift.def 2003-08-11 10:06:30.000000000 +0200 +@@ -68,7 +68,7 @@ + else if (times > number_of_args ()) + { + if (print_shift_error) +- sh_erange (list->word->word, "shift count"); ++ sh_erange (list ? list->word->word : NULL, "shift count"); + return (EXECUTION_FAILURE); + } + +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/builtins/source.def cvs/builtins/source.def +--- bash-2.05b/builtins/source.def 2002-05-10 18:36:56.000000000 +0200 ++++ cvs/builtins/source.def 2003-02-17 22:18:53.000000000 +0100 +@@ -23,16 +23,20 @@ + + $BUILTIN source + $FUNCTION source_builtin +-$SHORT_DOC source filename ++$SHORT_DOC source FILENAME [ARGUMENTS] + Read and execute commands from FILENAME and return. The pathnames + in $PATH are used to find the directory containing FILENAME. ++If any ARGUMENTS are supplied, they become the positional parameters when ++FILENAME is executed. + $END + $BUILTIN . + $DOCNAME dot + $FUNCTION source_builtin +-$SHORT_DOC . filename ++$SHORT_DOC . FILENAME [ARGUMENTS] + Read and execute commands from FILENAME and return. The pathnames + in $PATH are used to find the directory containing FILENAME. ++If any ARGUMENTS are supplied, they become the positional parameters when ++FILENAME is executed. + $END + /* source.c - Implements the `.' and `source' builtins. */ + +@@ -52,10 +56,12 @@ + + #include "../bashansi.h" + ++#include "../flags.h" + #include "../shell.h" + #include "../findcmd.h" + #include "common.h" + #include "bashgetopt.h" ++#include "../trap.h" + + #if !defined (errno) + extern int errno; +@@ -143,9 +149,29 @@ + } + set_dollar_vars_unchanged (); + +- result = source_file (filename); ++ { ++ char *debug_trap = TRAP_STRING(DEBUG_TRAP); + ++ if (debug_trap && !function_trace_mode) ++ { ++ debug_trap = savestring (debug_trap); ++ add_unwind_protect (xfree, debug_trap); ++ add_unwind_protect (set_debug_trap, debug_trap); ++ restore_default_signal (DEBUG_TRAP); ++ } ++ ++ ++ result = source_file (filename); ++ ++ } ++ + run_unwind_frame ("source"); + + return (result); + } ++ ++/* ++ * Local variables: ++ * mode: C ++ * End: ++ */ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/command.h cvs/command.h +--- bash-2.05b/command.h 2002-03-12 21:28:36.000000000 +0100 ++++ cvs/command.h 2002-09-12 12:44:32.000000000 +0200 +@@ -195,6 +195,7 @@ + /* The CASE command. */ + typedef struct case_com { + int flags; /* See description of CMD flags. */ ++ int line; /* The line number of command */ + WORD_DESC *word; /* The thing to test. */ + PATTERN_LIST *clauses; /* The clauses to test against, or NULL. */ + } CASE_COM; +@@ -202,6 +203,7 @@ + /* FOR command. */ + typedef struct for_com { + int flags; /* See description of CMD flags. */ ++ int line; /* The line number of command */ + WORD_DESC *name; /* The variable name to get mapped over. */ + WORD_LIST *map_list; /* The things to map over. This is never NULL. */ + COMMAND *action; /* The action to execute. +@@ -224,6 +226,7 @@ + /* KSH SELECT command. */ + typedef struct select_com { + int flags; /* See description of CMD flags. */ ++ int line; /* The line number of command */ + WORD_DESC *name; /* The variable name to get mapped over. */ + WORD_LIST *map_list; /* The things to map over. This is never NULL. */ + COMMAND *action; /* The action to execute. +@@ -290,6 +293,7 @@ + WORD_DESC *name; /* The name of the function. */ + COMMAND *command; /* The parsed execution tree. */ + int line; /* Line number the function def starts on. */ ++ char *source_file; /* The source file containing function. */ + } FUNCTION_DEF; + + /* A command that is `grouped' allows pipes and redirections to affect all +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/config-bot.h cvs/config-bot.h +--- bash-2.05b/config-bot.h 2002-04-16 23:01:03.000000000 +0200 ++++ cvs/config-bot.h 2003-06-06 06:01:11.000000000 +0200 +@@ -104,7 +104,7 @@ + /* For platforms which support the ISO C amendement 1 functionality we + support user defined character classes. */ + /* Solaris 2.5 has a bug: must be included before . */ +-#if defined (HAVE_WCTYPE_H) && defined (HAVE_WCHAR_H) ++#if defined (HAVE_WCTYPE_H) && defined (HAVE_WCHAR_H) && defined (HAVE_WCWIDTH) + # include + # include + # if defined (HAVE_MBSRTOWCS) /* system is supposed to support XPG5 */ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/config.h.in cvs/config.h.in +--- bash-2.05b/config.h.in 2002-06-25 15:48:52.000000000 +0200 ++++ cvs/config.h.in 2003-08-11 10:06:03.000000000 +0200 +@@ -582,6 +582,9 @@ + /* Define if you have the putenv function. */ + #undef HAVE_PUTENV + ++/* Define if you have the random function. */ ++#undef HAVE_RANDOM ++ + /* Define if you have the readlink function. */ + #undef HAVE_READLINK + +@@ -663,6 +666,9 @@ + /* Define if you have the strsignal function or macro. */ + #undef HAVE_STRSIGNAL + ++/* Define if you have the srandom function. */ ++#undef HAVE_SRANDOM ++ + /* Define if you have the sysconf function. */ + #undef HAVE_SYSCONF + +Only in bash-2.05b: configure +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/configure.in cvs/configure.in +--- bash-2.05b/configure.in 2002-07-16 15:31:25.000000000 +0200 ++++ cvs/configure.in 2003-09-01 21:57:33.000000000 +0200 +@@ -1,11 +1,11 @@ + dnl +-dnl Configure script for bash-2.05 ++dnl Configure script for bash with debugger + dnl +-dnl report bugs to chet@po.cwru.edu ++dnl report bugs to rocky@panix.com + dnl + dnl Process this file with autoconf to produce a configure script. + +-# Copyright (C) 1987-2002 Free Software Foundation, Inc. ++# Copyright (C) 1987-2003 Free Software Foundation, Inc. + + # This program is free software; you can redistribute it and/or modify + # it under the terms of the GNU General Public License as published by +@@ -22,21 +22,28 @@ + # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + # 02111-1307, USA. + +-AC_REVISION([for Bash 2.05b, version 2.144, from autoconf version] AC_ACVERSION)dnl ++define(bashvers, 2.05b) ++define(relstatus, rebash-0.43) + +-AC_INIT(bash, 2.05b, bug-bash@gnu.org) ++AC_INIT([bash],[bashvers-relstatus],[rocky@panix.com]) ++ ++AM_PATH_LISPDIR ++AC_ARG_PROGRAM + + dnl make sure we are using a recent autoconf version +-AC_PREREQ(2.50) ++AC_PREREQ(2.53) + + AC_CONFIG_SRCDIR(shell.h) ++if test x$ac_srcdir = x ; then ++ ac_srcdir=. ++fi + dnl where to find install.sh, config.sub, and config.guess + AC_CONFIG_AUX_DIR(./support) + AC_CONFIG_HEADERS(config.h) + + dnl checks for version info +-BASHVERS=2.05b +-RELSTATUS=release ++BASHVERS=bashvers ++RELSTATUS=relstatus + + dnl defaults for debug settings + case "$RELSTATUS" in +@@ -44,9 +51,8 @@ + *) DEBUG= MALLOC_DEBUG= ;; + esac + +-dnl canonicalize the host and os so we can do some tricky things before +-dnl parsing options +-AC_CANONICAL_HOST ++dnl Allow package selection (for installation), Setting bashdb as default ++AC_BASHDB_PACKAGE([bashdb]) + + dnl configure defaults + opt_bash_malloc=yes +@@ -92,6 +98,9 @@ + *) opt_memscramble=yes ;; + esac + ++dnl variables to configure ++AC_ARG_VAR(DEBUGGER_START_FILE, [location of debugger startup file]) ++ + dnl arguments to configure + dnl packages + AC_ARG_WITH(afs, AC_HELP_STRING([--with-afs], [if you are running AFS]), opt_afs=$withval) +@@ -138,6 +147,34 @@ + prefer_curses=yes + fi + ++if test -z "${DEBUGGER_START_FILE}"; then ++ ac_debugger_start_file=${datadir}/${PACKAGE}/dbg-main.inc ++ DEBUGGER_START_FILE=`( ++ test "x$prefix" = xNONE && prefix="$ac_default_prefix" ++ test "x$exec_prefix" = xNONE && exec_prefix="${prefix}" ++ eval echo "$ac_debugger_start_file" ++ )` ++fi ++ ++dnl We use a diff in regression testing ++AC_PATH_PROG(DIFF, diff, no) ++DIFF_OPTS= ++ ++if test "$DIFF" = no ; then ++ AC_PATH_PROG(DIFF, cmp, no) ++else ++ dnl Try for GNU diff options. ++ # MSDOG output uses \r\n rather than \n in tests ++ for diff_opt in -w --unified ; do ++ if $DIFF $diff_opt ./configure ./configure > /dev/null 2>&1; then ++ AC_MSG_RESULT([adding $diff_opt to diff in regression tests]) ++ DIFF_OPTS="$DIFF_OPTS $diff_opt" ++ fi ++ done ++fi ++AC_SUBST(DIFF) ++AC_SUBST(DIFF_OPTS) ++ + dnl optional shell features in config.h.in + opt_minimal_config=no + +@@ -327,8 +364,7 @@ + AC_MINIX + + dnl test for non-Unix variants +-AC_CYGWIN +-AC_MINGW32 ++AC_CANONICAL_HOST() + + AC_SYS_LARGEFILE + +@@ -345,7 +381,7 @@ + + if test "x$cross_compiling" = "xyes"; then + case "${host}" in +- *-cygwin*) ++ *cygwin*) + cross_cache=${srcdir}/cross-build/cygwin32.cache + SIGNAMES_H='$(srcdir)/cross-build/win32sig.h' + ;; +@@ -394,7 +430,7 @@ + + if test "$opt_static_link" = yes; then + # if we're using gcc, add `-static' to LDFLAGS, except on Solaris >= 2 +- if test -n "$GCC" || test "$ac_cv_prog_gcc" = "yes"; then ++ if test -n "$GCC" || test "$ac_cv_c_compiler_gnu" = "yes"; then + STATIC_LD="-static" + case "$host_os" in + solaris2*) ;; +@@ -609,6 +645,9 @@ + + AC_FUNC_MKTIME + ++dnl checks for random functions ++AC_CHECK_FUNCS(random srandom) ++ + dnl checks for locale functions + AC_CHECK_HEADERS(libintl.h) + AC_CHECK_FUNCS(gettext textdomain bindtextdomain) +@@ -631,7 +670,8 @@ + fi + + dnl this defines SYS_SIGLIST_DECLARED +-AC_DECL_SYS_SIGLIST ++AC_CHECK_DECLS([sys_siglist]) ++ + + dnl network functions -- check for inet_aton again + if test "$ac_cv_func_inet_aton" != 'yes'; then +@@ -945,11 +985,29 @@ + #AC_SUBST(ALLOCA_SOURCE) + #AC_SUBST(ALLOCA_OBJECT) + +-AC_OUTPUT([Makefile builtins/Makefile lib/readline/Makefile lib/glob/Makefile \ ++AC_CONFIG_SUBDIRS(debugger) ++ ++AC_CONFIG_FILES([Makefile builtins/Makefile lib/readline/Makefile lib/glob/Makefile \ + lib/malloc/Makefile lib/sh/Makefile lib/termcap/Makefile \ +- lib/tilde/Makefile doc/Makefile support/Makefile \ +- examples/loadables/Makefile examples/loadables/perl/Makefile], +-[ ++ lib/tilde/Makefile doc/Makefile \ ++ tests/check_common \ ++ support/Makefile pathnames.h \ ++ examples/loadables/Makefile examples/loadables/perl/Makefile]) ++ ++# We don't want TeX do be a requirement for compilation. ++if ! test -e $ac_srcdir/doc/version.texi ;then ++ AC_CONFIG_FILES([ doc/version.texi ]) ++fi ++ ++AC_CONFIG_COMMANDS([default],[[ + # Makefile uses this timestamp file to record whether config.h is up to date. + echo timestamp > stamp-h +-]) ++# Until we get CVS to set these +x ++chmod +x $ac_srcdir/tests/run-* ++# Keep old dates on these so we don't gratuitously have make do work. ++if test -e $ac_srcdir/doc/version.texi ;then ++ echo timestamp > $ac_srcdir/doc/stamp-vti ++ touch -cr $ac_srcdir/configure.in doc/version.texi $ac_srcdir/doc/stamp-vti ++fi ++]],[[]]) ++AC_OUTPUT +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/cvscompile.sh cvs/cvscompile.sh +--- bash-2.05b/cvscompile.sh 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/cvscompile.sh 2002-09-23 23:57:39.000000000 +0200 +@@ -0,0 +1,21 @@ ++#!/bin/sh ++# $Id: cvscompile.sh,v 1.8 2002/09/23 21:57:39 rockyb Exp $ ++# Run this to generate all the initial Makefiles, etc. ++ ++# As reported by Masatake YAMATO, ++ ++# In the Autotools book written by Gary V. Vaughan, et al., ++# bootstrap.sh and autogen.sh are are used for initializing files that ++# are derived and are therefore not checked into CVS. The difference ++# between bootstrap.sh and autogen.sh is running configure at the end ++# of script or not; autogen.sh runs and bootstrap.sh does not. So ++# autogen.sh is found in many projects; bootstrap.sh is good for cross ++# compiling. ++ ++./autogen.sh $@ ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** ++ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/AUTHORS cvs/debugger/AUTHORS +--- bash-2.05b/debugger/AUTHORS 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/AUTHORS 2002-08-30 18:20:26.000000000 +0200 +@@ -0,0 +1,9 @@ ++R. Bernstein (rocky@panix.com) ++ ++with ideas from (via a predecessor bashdb): ++ ++Bill Rosenblatt, ++Michael Loukides, ++Cigy Cyriac, ++Chet Ramey, ++Gary V. Vaughan +\ No newline at end of file +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/CHANGES cvs/debugger/CHANGES +--- bash-2.05b/debugger/CHANGES 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/CHANGES 2003-06-01 03:18:40.000000000 +0200 +@@ -0,0 +1,132 @@ ++------------------------------- ++CHANGES TO THE BASH INTERPRETER ++------------------------------- ++ ++Added timestamped history and customized display via strftime and the ++HISTTIMEFORMAT environment variable. ++ ++Added --debugger option which sources debugger startup script. ++ ++Extended FUNCNAME variable to be an array indicating the call stack of ++function names in effect. The top-level "function" name is "main." ++ ++Extended "declare -F fn" to return the source file name and the line number ++inside that of fn. ++ ++Added dynamic BASH_ARGC and BASH_ARGV arrays which store all of the ++parameters. ARGC gives the number of parameters in a call. BASH_ARGV ++are the parameters in stack like fashion. Last parameter of last call ++is on top, first parameter of initial call is on the bottom (index 0). ++ ++Added dynamic BASH_SOURCE array variable to give the file names associated ++with FUNCNAME. ++ ++Using BASH_SOURCE we now report the right filename when you have an ++evaluation error in a sourced file. ++ ++Added dynamic BASH_LINENO array variable to give the source file line ++numbers names associated with FUNCNAME. ++ ++LINENO: All line numbers are now relative to the beginning of a file, ++not relative to a function name. ++ ++Added dynamic BASH_COMMAND variable which is the command to be ++executed (or is executing) unless the command is a "trap" in which ++case it is the command that will be executed after the trap completes. ++ ++Added dynamic BASH_SUBSHELL variable gives the number of subshells ++that you are nested in. ++ ++Added a set shell option (set -o fntrace or set -d) which ++causes the TRAP DEBUG setting (whether the value is on or off) to ++persist when entering a function. It also does likewise for the ++"source" builtin and command substitution. ++ ++Added a new trap "RETURN" which calls a handler every time a ++function or sourced file is returned. ++ ++Added a new trap "SUBEXIT" which calls a handler every time a ++subshell exits. ++ ++Added caller() builtin function which works like Perl's builtin. ++ ++trap DEBUG will skip the next statement to be executed if the ++handler returns 2 (or sets $? to 2). ++ ++Line number on command substitution `` $() and { } is the line number ++of the source file, and not relative to the beginning of the ++substitution (which is usually 1). For debugging absolute line numbers ++are useful. Even outside of debugging, when reporting errors it's hard ++to see how error messages like these generated from errors.tests in the previous versions of bash (<=2.05): ++ ++./errors.tests: line 1: /bin/sh + 0: syntax error: operand expected (error token is "/bin/sh + 0") ++./errors.tests: line 1: /bin/sh + 0: syntax error: operand expected (error token is "/bin/sh + 0") ++ ++are as helpful than what we get now with absolute line numbers: ++ ++./errors.tests: line 212: /bin/sh + 0: syntax error: operand expected (error token is "/bin/sh + 0") ++./errors.tests: line 213: /bin/sh + 0: syntax error: operand expected (error token is "/bin/sh + 0") ++ ++Line numbers probably a little more accurate on tracing and ++LINENO. ++ ++"for" variables, and "case" conditions and "select" selectors are ++listed on "set -x" tracing and debugging. ++ ++--------- ++CHANGES TO EXTERNAL BASH DEBUGGER AND ITS LIBRARY CODE ++ ++First I've tried to make this look more like gdb and perldb which are ++much richer debuggers. Probably the easiest part was to hack into the ++Emacs GUD (Grand Unified Debugger) code to work with bashdb. Making ++bashdb work with DDD, a debugger GUI front end was likewise ++straightforward (although requiring copious additions to many files - ++the changes were basically, "do this like Perl or do this like GDB"). ++ ++By now (Aug 2002), most of Michael Loukides' and Bill Rosenblatt's ksh ++debugger code (circa 1995 and before) has been replaced. ++ ++The hackery of creating and running a concatenated file with debugging ++routines include (bashdb.pre and bashdb.fns) has been eliminated. Now ++we just source the debugged file or, better, do a funny "source" ++inside bash itself to get $0 correct and the the call stack correct. ++ ++There is now some option processing in bashdb script. In particular ++use -L to tell bashdb where the directory of the debugger code. There ++is more verbose usage reporting. ++ ++There is now an internal source-line array for *each* source file seen ++in debugger execution. The original code always retrieved from a text ++file (and it was presumed that there was only one). Since changes to ++the source after the program is running aren't reflected in the ++execution of the code, reading the source file to retrieve source text ++if that text was modified in the midst of debugging may give the wrong ++source line text. In this respect caching source lines may be more ++accurate. ++ ++There had been a variable which stored breakpoints as a string of line ++numbers, e.g. "3|23|45|". Using "|" as a field separator perhaps was ++convenient since one could then call egrep (without the trailing "|") ++to list a line number. Such was life in days when arrays were not a ++considered to be standard in POSIX-like shells, and bash still has ++provisions for conditionally compiling array support in. However, ++nowadays I think it safe to assume the array datatype standard; it is ++POSIX. ++ ++So I've removed this breakpoint variable and replaced it with an ++array. As for finding the line associated with the breakpoint, I added ++another array to store the source line for that. ++ ++I've added watchpoints as are found in perl5db or gdb. ++ ++With the breakpoint array, I also added breakpoint conditions and ++one-time breakpoints, whether the breakpoint is enabled or disabled, ++all analogous to gdb. ++ ++Signal handling was added, and saving and restoring $?, $1, $2, the ++"set" options. ++ ++With this, I think we've gotten over the main hurdle of getting a ++decent debugger for bash. ++ ++$Id: CHANGES,v 1.14 2003/06/01 01:18:40 rockyb Exp $ +\ No newline at end of file +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/COPYING cvs/debugger/COPYING +--- bash-2.05b/debugger/COPYING 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/COPYING 2002-08-20 18:04:30.000000000 +0200 +@@ -0,0 +1,340 @@ ++ GNU GENERAL PUBLIC LICENSE ++ Version 2, June 1991 ++ ++ Copyright (C) 1989, 1991 Free Software Foundation, Inc. ++ 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA ++ Everyone is permitted to copy and distribute verbatim copies ++ of this license document, but changing it is not allowed. ++ ++ Preamble ++ ++ The licenses for most software are designed to take away your ++freedom to share and change it. By contrast, the GNU General Public ++License is intended to guarantee your freedom to share and change free ++software--to make sure the software is free for all its users. This ++General Public License applies to most of the Free Software ++Foundation's software and to any other program whose authors commit to ++using it. (Some other Free Software Foundation software is covered by ++the GNU Library General Public License instead.) You can apply it to ++your programs, too. ++ ++ When we speak of free software, we are referring to freedom, not ++price. Our General Public Licenses are designed to make sure that you ++have the freedom to distribute copies of free software (and charge for ++this service if you wish), that you receive source code or can get it ++if you want it, that you can change the software or use pieces of it ++in new free programs; and that you know you can do these things. ++ ++ To protect your rights, we need to make restrictions that forbid ++anyone to deny you these rights or to ask you to surrender the rights. ++These restrictions translate to certain responsibilities for you if you ++distribute copies of the software, or if you modify it. ++ ++ For example, if you distribute copies of such a program, whether ++gratis or for a fee, you must give the recipients all the rights that ++you have. You must make sure that they, too, receive or can get the ++source code. And you must show them these terms so they know their ++rights. ++ ++ We protect your rights with two steps: (1) copyright the software, and ++(2) offer you this license which gives you legal permission to copy, ++distribute and/or modify the software. ++ ++ Also, for each author's protection and ours, we want to make certain ++that everyone understands that there is no warranty for this free ++software. If the software is modified by someone else and passed on, we ++want its recipients to know that what they have is not the original, so ++that any problems introduced by others will not reflect on the original ++authors' reputations. ++ ++ Finally, any free program is threatened constantly by software ++patents. We wish to avoid the danger that redistributors of a free ++program will individually obtain patent licenses, in effect making the ++program proprietary. To prevent this, we have made it clear that any ++patent must be licensed for everyone's free use or not licensed at all. ++ ++ The precise terms and conditions for copying, distribution and ++modification follow. ++ ++ GNU GENERAL PUBLIC LICENSE ++ TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION ++ ++ 0. This License applies to any program or other work which contains ++a notice placed by the copyright holder saying it may be distributed ++under the terms of this General Public License. The "Program", below, ++refers to any such program or work, and a "work based on the Program" ++means either the Program or any derivative work under copyright law: ++that is to say, a work containing the Program or a portion of it, ++either verbatim or with modifications and/or translated into another ++language. (Hereinafter, translation is included without limitation in ++the term "modification".) Each licensee is addressed as "you". ++ ++Activities other than copying, distribution and modification are not ++covered by this License; they are outside its scope. The act of ++running the Program is not restricted, and the output from the Program ++is covered only if its contents constitute a work based on the ++Program (independent of having been made by running the Program). ++Whether that is true depends on what the Program does. ++ ++ 1. You may copy and distribute verbatim copies of the Program's ++source code as you receive it, in any medium, provided that you ++conspicuously and appropriately publish on each copy an appropriate ++copyright notice and disclaimer of warranty; keep intact all the ++notices that refer to this License and to the absence of any warranty; ++and give any other recipients of the Program a copy of this License ++along with the Program. ++ ++You may charge a fee for the physical act of transferring a copy, and ++you may at your option offer warranty protection in exchange for a fee. ++ ++ 2. You may modify your copy or copies of the Program or any portion ++of it, thus forming a work based on the Program, and copy and ++distribute such modifications or work under the terms of Section 1 ++above, provided that you also meet all of these conditions: ++ ++ a) You must cause the modified files to carry prominent notices ++ stating that you changed the files and the date of any change. ++ ++ b) You must cause any work that you distribute or publish, that in ++ whole or in part contains or is derived from the Program or any ++ part thereof, to be licensed as a whole at no charge to all third ++ parties under the terms of this License. ++ ++ c) If the modified program normally reads commands interactively ++ when run, you must cause it, when started running for such ++ interactive use in the most ordinary way, to print or display an ++ announcement including an appropriate copyright notice and a ++ notice that there is no warranty (or else, saying that you provide ++ a warranty) and that users may redistribute the program under ++ these conditions, and telling the user how to view a copy of this ++ License. (Exception: if the Program itself is interactive but ++ does not normally print such an announcement, your work based on ++ the Program is not required to print an announcement.) ++ ++These requirements apply to the modified work as a whole. If ++identifiable sections of that work are not derived from the Program, ++and can be reasonably considered independent and separate works in ++themselves, then this License, and its terms, do not apply to those ++sections when you distribute them as separate works. But when you ++distribute the same sections as part of a whole which is a work based ++on the Program, the distribution of the whole must be on the terms of ++this License, whose permissions for other licensees extend to the ++entire whole, and thus to each and every part regardless of who wrote it. ++ ++Thus, it is not the intent of this section to claim rights or contest ++your rights to work written entirely by you; rather, the intent is to ++exercise the right to control the distribution of derivative or ++collective works based on the Program. ++ ++In addition, mere aggregation of another work not based on the Program ++with the Program (or with a work based on the Program) on a volume of ++a storage or distribution medium does not bring the other work under ++the scope of this License. ++ ++ 3. You may copy and distribute the Program (or a work based on it, ++under Section 2) in object code or executable form under the terms of ++Sections 1 and 2 above provided that you also do one of the following: ++ ++ a) Accompany it with the complete corresponding machine-readable ++ source code, which must be distributed under the terms of Sections ++ 1 and 2 above on a medium customarily used for software interchange; or, ++ ++ b) Accompany it with a written offer, valid for at least three ++ years, to give any third party, for a charge no more than your ++ cost of physically performing source distribution, a complete ++ machine-readable copy of the corresponding source code, to be ++ distributed under the terms of Sections 1 and 2 above on a medium ++ customarily used for software interchange; or, ++ ++ c) Accompany it with the information you received as to the offer ++ to distribute corresponding source code. (This alternative is ++ allowed only for noncommercial distribution and only if you ++ received the program in object code or executable form with such ++ an offer, in accord with Subsection b above.) ++ ++The source code for a work means the preferred form of the work for ++making modifications to it. For an executable work, complete source ++code means all the source code for all modules it contains, plus any ++associated interface definition files, plus the scripts used to ++control compilation and installation of the executable. However, as a ++special exception, the source code distributed need not include ++anything that is normally distributed (in either source or binary ++form) with the major components (compiler, kernel, and so on) of the ++operating system on which the executable runs, unless that component ++itself accompanies the executable. ++ ++If distribution of executable or object code is made by offering ++access to copy from a designated place, then offering equivalent ++access to copy the source code from the same place counts as ++distribution of the source code, even though third parties are not ++compelled to copy the source along with the object code. ++ ++ 4. You may not copy, modify, sublicense, or distribute the Program ++except as expressly provided under this License. Any attempt ++otherwise to copy, modify, sublicense or distribute the Program is ++void, and will automatically terminate your rights under this License. ++However, parties who have received copies, or rights, from you under ++this License will not have their licenses terminated so long as such ++parties remain in full compliance. ++ ++ 5. You are not required to accept this License, since you have not ++signed it. However, nothing else grants you permission to modify or ++distribute the Program or its derivative works. These actions are ++prohibited by law if you do not accept this License. Therefore, by ++modifying or distributing the Program (or any work based on the ++Program), you indicate your acceptance of this License to do so, and ++all its terms and conditions for copying, distributing or modifying ++the Program or works based on it. ++ ++ 6. Each time you redistribute the Program (or any work based on the ++Program), the recipient automatically receives a license from the ++original licensor to copy, distribute or modify the Program subject to ++these terms and conditions. You may not impose any further ++restrictions on the recipients' exercise of the rights granted herein. ++You are not responsible for enforcing compliance by third parties to ++this License. ++ ++ 7. If, as a consequence of a court judgment or allegation of patent ++infringement or for any other reason (not limited to patent issues), ++conditions are imposed on you (whether by court order, agreement or ++otherwise) that contradict the conditions of this License, they do not ++excuse you from the conditions of this License. If you cannot ++distribute so as to satisfy simultaneously your obligations under this ++License and any other pertinent obligations, then as a consequence you ++may not distribute the Program at all. For example, if a patent ++license would not permit royalty-free redistribution of the Program by ++all those who receive copies directly or indirectly through you, then ++the only way you could satisfy both it and this License would be to ++refrain entirely from distribution of the Program. ++ ++If any portion of this section is held invalid or unenforceable under ++any particular circumstance, the balance of the section is intended to ++apply and the section as a whole is intended to apply in other ++circumstances. ++ ++It is not the purpose of this section to induce you to infringe any ++patents or other property right claims or to contest validity of any ++such claims; this section has the sole purpose of protecting the ++integrity of the free software distribution system, which is ++implemented by public license practices. Many people have made ++generous contributions to the wide range of software distributed ++through that system in reliance on consistent application of that ++system; it is up to the author/donor to decide if he or she is willing ++to distribute software through any other system and a licensee cannot ++impose that choice. ++ ++This section is intended to make thoroughly clear what is believed to ++be a consequence of the rest of this License. ++ ++ 8. If the distribution and/or use of the Program is restricted in ++certain countries either by patents or by copyrighted interfaces, the ++original copyright holder who places the Program under this License ++may add an explicit geographical distribution limitation excluding ++those countries, so that distribution is permitted only in or among ++countries not thus excluded. In such case, this License incorporates ++the limitation as if written in the body of this License. ++ ++ 9. The Free Software Foundation may publish revised and/or new versions ++of the General Public License from time to time. Such new versions will ++be similar in spirit to the present version, but may differ in detail to ++address new problems or concerns. ++ ++Each version is given a distinguishing version number. If the Program ++specifies a version number of this License which applies to it and "any ++later version", you have the option of following the terms and conditions ++either of that version or of any later version published by the Free ++Software Foundation. If the Program does not specify a version number of ++this License, you may choose any version ever published by the Free Software ++Foundation. ++ ++ 10. If you wish to incorporate parts of the Program into other free ++programs whose distribution conditions are different, write to the author ++to ask for permission. For software which is copyrighted by the Free ++Software Foundation, write to the Free Software Foundation; we sometimes ++make exceptions for this. Our decision will be guided by the two goals ++of preserving the free status of all derivatives of our free software and ++of promoting the sharing and reuse of software generally. ++ ++ NO WARRANTY ++ ++ 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY ++FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN ++OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES ++PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED ++OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF ++MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS ++TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE ++PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, ++REPAIR OR CORRECTION. ++ ++ 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING ++WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR ++REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, ++INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING ++OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED ++TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY ++YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER ++PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE ++POSSIBILITY OF SUCH DAMAGES. ++ ++ END OF TERMS AND CONDITIONS ++ ++ How to Apply These Terms to Your New Programs ++ ++ If you develop a new program, and you want it to be of the greatest ++possible use to the public, the best way to achieve this is to make it ++free software which everyone can redistribute and change under these terms. ++ ++ To do so, attach the following notices to the program. It is safest ++to attach them to the start of each source file to most effectively ++convey the exclusion of warranty; and each file should have at least ++the "copyright" line and a pointer to where the full notice is found. ++ ++ ++ Copyright (C) 19yy ++ ++ This program is free software; you can redistribute it and/or modify ++ it under the terms of the GNU General Public License as published by ++ the Free Software Foundation; either version 2 of the License, or ++ (at your option) any later version. ++ ++ This program is distributed in the hope that it will be useful, ++ but WITHOUT ANY WARRANTY; without even the implied warranty of ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ GNU General Public License for more details. ++ ++ You should have received a copy of the GNU General Public License ++ along with this program; if not, write to the Free Software ++ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA ++ ++ ++Also add information on how to contact you by electronic and paper mail. ++ ++If the program is interactive, make it output a short notice like this ++when it starts in an interactive mode: ++ ++ Gnomovision version 69, Copyright (C) 19yy name of author ++ Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. ++ This is free software, and you are welcome to redistribute it ++ under certain conditions; type `show c' for details. ++ ++The hypothetical commands `show w' and `show c' should show the appropriate ++parts of the General Public License. Of course, the commands you use may ++be called something other than `show w' and `show c'; they could even be ++mouse-clicks or menu items--whatever suits your program. ++ ++You should also get your employer (if you work as a programmer) or your ++school, if any, to sign a "copyright disclaimer" for the program, if ++necessary. Here is a sample; alter the names: ++ ++ Yoyodyne, Inc., hereby disclaims all copyright interest in the program ++ `Gnomovision' (which makes passes at compilers) written by James Hacker. ++ ++ , 1 April 1989 ++ Ty Coon, President of Vice ++ ++This General Public License does not permit incorporating your program into ++proprietary programs. If your program is a subroutine library, you may ++consider it more useful to permit linking proprietary applications with the ++library. If this is what you want to do, use the GNU Library General ++Public License instead of this License. +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/ChangeLog cvs/debugger/ChangeLog +--- bash-2.05b/debugger/ChangeLog 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/ChangeLog 2003-08-02 22:33:22.000000000 +0200 +@@ -0,0 +1,2258 @@ ++2003-08-02 16:20 rockyb ++ ++ * dbg-cmds.inc, dbg-help.inc, doc/bashdb.texi: Doc changes. ++ ++2003-08-02 16:19 rockyb ++ ++ * configure.ac: Time for another release. ++ ++2003-08-02 16:12 rockyb ++ ++ * test/: misc.right, sig.right: Help changed. Not sure why sig ++ keeps changing though. ++ ++2003-08-02 15:54 rockyb ++ ++ * Makefile.am, dbg-file.inc, dbg-help.inc, dbg-io.inc, ++ dbg-list.inc: Add _Dbg_ in front of more debugger variables. In ++ particular: _read_ -> _Dbg_read_ _maxline_ -> _Dbg_maxline_ ++ _source_ -> _Dbg_source_ ++ ++2003-08-02 15:53 rockyb ++ ++ * test/: misc-output.right, misc.right, sig.right: Add test of V ++ command. sig: line number again changed. ++ ++2003-08-02 15:52 rockyb ++ ++ * test/misc.cmd: Add test of V command. ++ ++2003-08-02 15:51 rockyb ++ ++ * dbg-cmds.inc: Routine to list variables and values by pattern ++ possibly inverted. Code courtesy of Mikael Andersson ++ . ++ ++2003-08-02 15:46 rockyb ++ ++ * THANKS: This is long overdue, even if it is meager. ++ ++2003-08-01 06:45 rockyb ++ ++ * dbg-help.inc: Add "help show " and "help info " ++ . ++ ++2003-07-31 23:48 rockyb ++ ++ * dbg-help.inc: Missing a line for "help V" ++ ++2003-06-19 10:49 rockyb ++ ++ * TODO: Simple typo. ++ ++2003-06-13 17:49 rockyb ++ ++ * NEWS, ChangeLog: [no log message] ++ ++2003-06-08 23:19 rockyb ++ ++ * configure.ac: It's = not == in tests ++ ++2003-06-08 10:39 rockyb ++ ++ * configure.ac, test/Makefile.am, test/action.tests, ++ test/bugIFS.tests, test/display.tests, test/finish.tests, ++ test/run-action, test/run-brkpt, test/run-brkpt1, test/run-bugIFS, ++ test/run-debug, test/run-display, test/run-finish, test/run-list, ++ test/run-misc, test/run-multi, test/run-parm, test/run-search, ++ test/run-sig, test/run-skip, test/run-subshell, test/run-tbreak, ++ test/run-watch1, test/run-watch2, test/skip.tests, ++ test/subshell1.tests, test/subshell2.tests, test/subshell3.tests: ++ Debugger regression tests now will use diff --u if possible. Will ++ revert even to cmp if diff is not around. check_common: common ++ routines and setup to simplify/unify regression testing ++ ++2003-06-06 01:01 rockyb ++ ++ * NEWS, test/run-multi: Bash patches 005-007. run-multi: set ++ THIS_SH so it runs under cygwin (and others). ++ ++2003-06-03 14:28 rockyb ++ ++ * test/: Makefile.am, multi1.right, multi1.tests, multi2.right, ++ multi2.tests, multi3.right, multi3.tests, multi4.tests, run-multi: ++ Add multi tests ++ ++2003-06-03 14:27 rockyb ++ ++ * Makefile.am: Add EXTEXT for M$ OS's. ++ ++2003-06-03 13:08 rockyb ++ ++ * test/: Makefile.am, multi.sh, multi1.cmd, multi1.tests, ++ multi2.cmd, multi2.tests, multi3.cmd, multi3.tests, multi4.right, ++ run-multi: Add multi tests ++ ++2003-06-02 05:06 rockyb ++ ++ * configure.ac: [no log message] ++ ++2003-06-02 05:03 rockyb ++ ++ * configure.ac: Create bashdb as executable. ++ ++2003-06-01 11:07 rockyb ++ ++ * Makefile.am: Make sure bash is copied before running "make check" ++ ++2003-06-01 10:35 rockyb ++ ++ * doc/.cvsignore: [no log message] ++ ++2003-06-01 10:35 rockyb ++ ++ * .cvsignore, doc/.cvsignore, doc/fdl.texi: CVS maintainance. ++ ++2003-06-01 10:32 rockyb ++ ++ * ChangeLog, Makefile.am, configure.ac: configure.ac: set unique ++ distribution name Makefile.am: include doc directory. ++ ++2003-06-01 10:31 rockyb ++ ++ * doc/: Makefile.am, bashdb.texi, gpl.texi, version.texi.in: Except ++ Makefile.am, moved from bash doc directory. Makefile.am: pretty ++ standard stuff. ++ ++2003-06-01 10:29 rockyb ++ ++ * doc/.cvsignore: [no log message] ++ ++2003-06-01 10:25 rockyb ++ ++ * NEWS: [no log message] ++ ++2003-06-01 09:23 rockyb ++ ++ * test/: brkpt1.tests, brkpt2.right, brkpt2.tests, brkpt3.tests, ++ debug.tests, list.tests, misc.tests, parm.tests, run-brkpt, ++ run-debug, run-list, run-misc, run-parm, run-search, run-sig, ++ run-tbreak, run-watch2, search.tests, sig.tests, tbreak.tests, ++ watch2.tests: make check and make distcheck now work. ++ ++2003-05-31 22:43 rockyb ++ ++ * Makefile.am: Seems to be needed for make dist to work. ++ ++2003-05-31 22:23 rockyb ++ ++ * configure.ac: configure.in -> configure.ac ++ ++2003-05-31 22:15 rockyb ++ ++ * test/: Makefile.am, run-action, run-brkpt, run-brkpt1, ++ run-bugIFS, run-debug, run-display, run-finish, run-list, run-misc, ++ run-multi, run-parm, run-search, run-sig, run-skip, run-subshell, ++ run-tbreak, run-watch1, run-watch2, skip.tests, subshell1.tests, ++ tbreak.tests: Closer to getting "make distcheck" working. "make ++ check" is currently broken for a couple of tests. ++ ++2003-05-31 21:35 rockyb ++ ++ * test/: brkpt3.tests, debug.tests, display.tests, finish.tests, ++ list.tests, subshell2.tests, subshell3.tests, subshell4.tests, ++ tbreak.tests, watch1.tests, watch2.tests: changes to making "make ++ dist" work ++ ++2003-05-31 21:18 rockyb ++ ++ * CHANGES, Makefile.am, bashdb, bashdb.in, configure.in, ++ dbg-file.inc, test/.cvsignore, test/Makefile.am, test/action.cmd, ++ test/action.right, test/action.tests, test/brkpt1.cmd, ++ test/brkpt1.right, test/brkpt1.tests, test/brkpt2.cmd, ++ test/brkpt2.right, test/brkpt2.tests, test/brkpt3.cmd, ++ test/brkpt3.right, test/brkpt3.tests, test/bugIFS, test/bugIFS.cmd, ++ test/bugIFS.right, test/bugIFS.sh, test/bugIFS.tests, ++ test/dbg-test1.sh, test/dbg-test1.sub, test/dbg-test2.sh, ++ test/debug.cmd, test/debug.right, test/debug.sh, test/debug.tests, ++ test/debug2.cmd, test/display.cmd, test/display.right, ++ test/display.tests, test/finish.cmd, test/finish.right, ++ test/finish.tests, test/list.cmd, test/list.right, test/list.tests, ++ test/misc-output.right, test/misc.cmd, test/misc.right, ++ test/misc.tests, test/multi4.cmd, test/multi4.right, ++ test/multi4.tests, test/parm.cmd, test/parm.right, test/parm.sh, ++ test/parm.tests, test/prof1.cmd, test/prof2.cmd, test/run-action, ++ test/run-all, test/run-brkpt, test/run-brkpt1, test/run-bugIFS, ++ test/run-debug, test/run-display, test/run-finish, test/run-list, ++ test/run-misc, test/run-multi, test/run-parm, test/run-search, ++ test/run-sig, test/run-skip, test/run-subshell, test/run-tbreak, ++ test/run-watch1, test/run-watch2, test/search.cmd, ++ test/search.right, test/search.tests, test/sig.cmd, test/sig.right, ++ test/sig.sh, test/sig.tests, test/skip.cmd, test/skip.right, ++ test/skip.sh, test/skip.tests, test/subshell.sh, ++ test/subshell1.cmd, test/subshell1.right, test/subshell1.tests, ++ test/subshell2.cmd, test/subshell2.right, test/subshell2.tests, ++ test/subshell3.cmd, test/subshell3.right, test/subshell3.tests, ++ test/subshell4.cmd, test/subshell4.right, test/subshell4.tests, ++ test/tbreak.cmd, test/tbreak.right, test/tbreak.tests, ++ test/watch1.cmd, test/watch1.right, test/watch1.tests, ++ test/watch2.cmd, test/watch2.right, test/watch2.tests: ++ configure->configure.ac tests->test bashdb no longer has hard-coded ++ #! some changes to make "make distcheck" work... More changes ++ later.. ++ ++2003-05-27 19:54 rockyb ++ ++ * NEWS: [no log message] ++ ++2003-05-20 14:02 rockyb ++ ++ * dbg-init.inc: More OS-independent (output independent) test if a ++ terminal is attached. ++ ++2003-04-29 20:33 rockyb ++ ++ * NEWS: [no log message] ++ ++2003-04-29 20:22 rockyb ++ ++ * bashdb, dbg-cmds.inc, dbg-fns.inc, dbg-init.inc: Bug: Was using ++ source filename rather than canonicalized (expanded) filename. ++ ++2003-04-07 23:24 rockyb ++ ++ * dbg-cmds.inc: Use gdb undefined command message. ++ ++2003-03-31 10:09 rockyb ++ ++ * NEWS, dbg-brk.inc: Numbers in breakpoint/watchpoint/action ++ numbers are printed in a "%-3d" rather than (I think the nicer) ++ %3d" to match gdb's output. This helps ddd parse breakpoints. ++ ++2003-03-30 22:04 rockyb ++ ++ * NEWS, dbg-brk.inc: Breakpoints are now are origin 1 to match gdb. ++ It also helps ddd. ++ ++2003-03-27 00:34 rockyb ++ ++ * dbg-pre.inc.in: Ooops - changed dbg-pre.inc but should have ++ changed this one instead. ++ ++2003-03-26 09:05 rockyb ++ ++ * dbg-file.inc, dbg-io.inc: Add routine to resolve based on dir and ++ then expand a filename. Hopefully will be more permissive in ++ specified and work more like gdb. ++ ++2003-03-26 09:01 rockyb ++ ++ * dbg-brk.inc: Expand filename only the breakpoint exists. ++ ++2003-03-25 23:30 rockyb ++ ++ * dbg-list.inc: When set annotate=1 expand full filenames when ++ listing source lines. ++ ++2003-03-25 23:29 rockyb ++ ++ * dbg-file.inc: Bug in testing absolute file name. ++ ++2003-03-25 18:46 rockyb ++ ++ * dbg-brk.inc, dbg-file.inc, dbg-list.inc: Canonicalize filenames ++ in list and breakpoint commands and update regression tests to make ++ sure we don't slip back. ++ ++2003-03-21 19:16 rockyb ++ ++ * dbg-file.inc, dbg-set.inc: Small stuff. ++ ++2003-03-21 18:54 rockyb ++ ++ * bashdb, dbg-brk.inc, dbg-file.inc, dbg-fns.inc, dbg-help.inc, ++ dbg-init.inc, dbg-set.inc: Add "show dir" and directory search in ++ filename lookup (such as for breakpoints or line listing). ++ ++2003-03-21 12:59 rockyb ++ ++ * dbg-sig.inc: Signal received message is now gdb's. This ++ simplifies ddd handling. ++ ++2003-03-21 12:56 rockyb ++ ++ * TODO: Current sitiation. ++ ++2003-02-27 05:14 rockyb ++ ++ * dbg-brk.inc, dbg-file.inc, dbg-fns.inc, dbg-help.inc, dbg-io.inc: ++ _filenames -> _Dbg_filenames. Array now has expanded name Add ++ routine to print filename in form appropriate to annotation and ++ basename only ++ ++2003-02-27 02:33 rockyb ++ ++ * NEWS: _bashdb_ -> _Dbg_ ++ ++2003-02-27 02:27 rockyb ++ ++ * bashdb, dbg-brk.inc, dbg-cmds.inc, dbg-file.inc, dbg-fns.inc, ++ dbg-help.inc, dbg-hist.inc, dbg-init.inc, dbg-io.inc, dbg-list.inc, ++ dbg-main.inc, dbg-pre.inc.in, dbg-set-d-vars.inc, dbg-set.inc, ++ dbg-sig-ret.inc, dbg-sig.inc, dbg-stack.inc: All _bashdb_ variables ++ renamed _Dbg_. Might help for debuggers in other POSIX shells. ++ ++2003-02-26 23:34 rockyb ++ ++ * NEWS, dbg-help.inc, dbg-init.inc, dbg-set.inc: Add "set ++ annotate." More better set/show regression tests. ++ ++2003-02-25 11:22 rockyb ++ ++ * dbg-brk.inc: A little closer to gdb. Mesage reads breakpoint ++ already hit 1 time or breakpoint already hit n times ++ ++2003-02-25 10:55 rockyb ++ ++ * dbg-brk.inc: Make breakpoints look more like gdb. It helps ddd ++ parsing. ++ ++2003-02-23 15:40 rockyb ++ ++ * dbg-file.inc: Add tilde to list of translation characters. ++ ++2003-02-19 18:45 rockyb ++ ++ * dbg-brk.inc, dbg-cmds.inc, dbg-io.inc, dbg-sig.inc: Some declare ++ -a's turned into typeset -a on the hope (from David Korn) that this ++ will be legal ksh. ++ ++2003-02-17 19:23 rockyb ++ ++ * .cvsignore: bashdb- -> dbg- ++ ++2003-02-17 19:23 rockyb ++ ++ * Makefile.am, bashdb-sig-ret.inc, dbg-sig-ret.inc: Miscellaneous ++ small changes for new version and bashdb- -> dbg- changes. ++ ++2003-02-17 19:02 rockyb ++ ++ * Makefile.am, TODO, bashdb, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-file.inc, bashdb-fns.inc, bashdb-help.inc, bashdb-hist.inc, ++ bashdb-init.inc, bashdb-io.inc, bashdb-list.inc, bashdb-main.inc, ++ bashdb-set-d-vars.inc, bashdb-set.inc, bashdb-sig.inc, ++ bashdb-stack.inc, dbg-brk.inc, dbg-cmds.inc, dbg-file.inc, ++ dbg-fns.inc, dbg-help.inc, dbg-hist.inc, dbg-init.inc, dbg-io.inc, ++ dbg-list.inc, dbg-main.inc, dbg-pre.inc, dbg-set-d-vars.inc, ++ dbg-set.inc, dbg-sig.inc, dbg-stack.inc, bashdb-pre.inc.in, ++ dbg-pre.inc, dbg-pre.inc.in: bashdb -> dbg to facilitate porting to ++ other POSIX shells. ++ ++2003-02-17 17:02 rockyb ++ ++ * bashdb-fns.inc: Remove SUBEXIT trap. ++ ++2003-02-17 13:15 rockyb ++ ++ * bashdb-brk.inc: I think I've gotten over the hurdle of passing ++ information back. Removing breakpoints in a subshell and having ++ that persist now works. ++ ++2003-02-17 07:34 rockyb ++ ++ * bashdb-brk.inc: Perhaps some cleanup. ++ ++2003-02-17 06:51 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-file.inc, bashdb-sig.inc: ++ State saving bugs: save state of more breakpoint/action variables, ++ step/next variables. ++ ++2003-02-16 20:08 rockyb ++ ++ * NEWS, bashdb-brk.inc, bashdb-file.inc, bashdb-sig.inc: Probably ++ just beating around problems. Perhaps closer. Perhaps not. ++ ++2003-02-16 12:18 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-file.inc, bashdb-fns.inc, ++ bashdb-sig.inc: More state recorded in subshell journal: ++ Enable/disable fntrace options ++ ++ having breakpoints persist inside a subshell now works. (Removing ++ has problems though, as does saving/restoring the step counter) ++ ++2003-02-16 08:34 rockyb ++ ++ * bashdb-brk.inc, bashdb-file.inc, bashdb-fns.inc, bashdb-sig.inc: ++ Back to the future. Back to a single journal file. Can't get ++ SUBEXIT working and this seems to be more reliable (albeit more ++ work/slower). ++ ++2003-02-14 18:47 rockyb ++ ++ * NEWS, TODO, bashdb-sig.inc: Bug in quit 0 n. ++ ++2003-02-14 18:38 rockyb ++ ++ * CHANGES: Note BASH_SUBSHELL. Some typographical mistakes ++ corrected. ++ ++2003-02-14 18:31 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2003-02-14 18:25 rockyb ++ ++ * bashdb-cmds.inc, bashdb-file.inc, bashdb-fns.inc, ++ bashdb-set-d-vars.inc, bashdb-sig.inc: Better journal mechanism: ++ Now have journal per subshell. (Journal is used to pass information ++ between subshell and parent) ++ ++ There were escaping bugs in putting history entries into the ++ journal and retrieving them. ++ ++ Debugger "Print" now works like quote does. We were having ++ substituion problems before, e.g. "This can't happen" (the single ++ quote in the string). ++ ++ bashdb-fns.inc: New fn _bashdb_esq_dq. Save last_xxx state even if ++ we don't enter the debugger. ++ ++2003-02-12 19:55 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-file.inc, bashdb-fns.inc, ++ bashdb-help.inc, bashdb-hist.inc, bashdb-init.inc, bashdb-io.inc, ++ bashdb-list.inc, bashdb-main.inc, bashdb-set.inc, bashdb-sig.inc, ++ bashdb-stack.inc: Try to make more ksh-93 compatible. ++ ++ declare -> typeset (except when using declare -p or declare -a). ++ source -> . ++ ++2003-02-12 01:47 rockyb ++ ++ * bashdb: Small changes to make a little more consistent with ksh93 ++ ++2002-12-31 23:25 rockyb ++ ++ * bashdb-pre.inc.in: Comment more. ++ ++2002-12-31 23:17 rockyb ++ ++ * bashdb-file.inc: Add % to list of translated file characters. ++ ++2002-12-31 23:12 rockyb ++ ++ * bashdb-fns.inc: [no log message] ++ ++2002-12-31 11:57 rockyb ++ ++ * bashdb-fns.inc, bashdb-list.inc, bashdb-sig.inc: Make all bashdb ++ functions begin _bashdb - were were missing a couple of routines. ++ ++2002-12-30 23:23 rockyb ++ ++ * NEWS: [no log message] ++ ++2002-12-30 23:20 rockyb ++ ++ * bashdb-file.inc: Add + to list of translated file2var characters. ++ ++2002-12-25 10:30 rockyb ++ ++ * bashdb-cmds.inc: More state saved. ++ ++2002-12-25 10:20 rockyb ++ ++ * NEWS: Recent stuff. ++ ++2002-12-25 10:18 rockyb ++ ++ * TODO, bashdb-brk.inc, bashdb-cmds.inc, bashdb-file.inc, ++ bashdb-hist.inc, bashdb-set.inc, bashdb-sig.inc: Start the tedious ++ task of having a mechanism for setting variables that in a subshell ++ so that its value appears in its parent shell. ++ ++2002-12-25 00:01 rockyb ++ ++ * TODO: [no log message] ++ ++2002-12-24 23:14 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc: Fix setting breakpoints over ++ restart. Needed to save the _brkpt_filevar and _action_filevar ++ arrays as well as _brkpt_count and _brkpt_cond arrays. ++ ++ If we are in a subshell, we need to save BASHB_RESTART_FILE so it ++ will get set in sup-shells. ++ ++2002-12-24 22:13 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-12-24 22:13 rockyb ++ ++ * bashdb-cmds.inc, bashdb-file.inc, bashdb-sig.inc: Bug/Feature ++ fix: Restart now quits all subshells before issuing its restart ++ (exec) command. ++ ++2002-12-20 21:31 rockyb ++ ++ * bashdb-fns.inc: Small typo. ++ ++2002-12-20 12:22 rockyb ++ ++ * TODO: Update. ++ ++2002-12-20 12:17 rockyb ++ ++ * NEWS: - Bug: if IFS set character in filename: the filename ++ would be truncated. Handle IFS save/restore more pervasively by ++ saving on DEBUG handler entry. ++ ++2002-12-20 12:12 rockyb ++ ++ * bashdb-cmds.inc, bashdb-fns.inc: Save and restore IFS on debug ++ signal handler. It means we can remove custom save/restores ++ elsewhere in the code. ++ ++2002-12-10 04:45 rockyb ++ ++ * bashdb-brk.inc: More bash-oriented way to strip off a suffix. ++ ++2002-12-10 01:05 rockyb ++ ++ * emacs/gud.el: Eval expression command is "x" not "pe" ++ ++2002-12-10 01:04 rockyb ++ ++ * TODO: [no log message] ++ ++2002-12-09 08:16 rockyb ++ ++ * CHANGES: "set -x" now show "for", "case" and "select" headers ( ++ "for x in", "case c in", "select s") ++ ++2002-12-07 22:42 rockyb ++ ++ * ChangeLog, NEWS: Get ready for yet another release. ++ ++2002-12-07 16:54 rockyb ++ ++ * ChangeLog: [no log message] ++ ++2002-12-07 16:49 rockyb ++ ++ * bashdb-main.inc: Make sure we can read the restart file. ++ ++2002-12-07 16:48 rockyb ++ ++ * bashdb-cmds.inc: Forgot to unset flag which says to reread the ++ file. ++ ++2002-12-07 16:43 rockyb ++ ++ * ChangeLog, bashdb, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-main.inc: Preserve breakpoints, watchpoints, debugger set ++ variables on a restart. ++ ++2002-12-07 16:41 rockyb ++ ++ * Makefile.am: Remove compile of bashdb.elc. It's going away ++ anyway. ++ ++2002-12-07 10:24 rockyb ++ ++ * NEWS: As appropriate. ++ ++2002-12-07 10:23 rockyb ++ ++ * ChangeLog: print select head - "select s in" ++ ++2002-12-07 08:35 rockyb ++ ++ * bashdb-io.inc: [no log message] ++ ++2002-12-07 08:18 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-12-07 08:16 rockyb ++ ++ * bashdb-io.inc: bashdb-io.inc (_bashdb_readlin): - When reading ++ large files which takes a long time, print out status every 1000 ++ lines. ++ ++2002-12-07 08:02 rockyb ++ ++ * NEWS: Bug: (parse.y) line number was not updated properly when in ++ a here documented with an unquoted delimter and containing ++ backslash newline. ++ ++ Update and modernize tests and make executable. ++ ++2002-12-06 22:47 rockyb ++ ++ * bashdb-sig-ret.inc: Not used, but it if it were, this would be ++ right. Need to save/restore IFS around a "set -" ++ ++2002-12-06 22:44 rockyb ++ ++ * bashdb-list.inc: Typo. ++ ++2002-12-06 22:43 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-12-06 11:27 rockyb ++ ++ * emacs/: gud.el, gud.el.diff: A typo. Also modified DB when I ++ meant to modify bashdb. ++ ++2002-12-03 19:46 rockyb ++ ++ * bashdb-help.inc: Typo. ++ ++2002-12-03 19:34 rockyb ++ ++ * bashdb-fns.inc: Remove previous variable test. Not needed? ++ ++2002-12-03 19:24 rockyb ++ ++ * bashdb-fns.inc: Bug in set was set of IFS messing up "set" ++ parsing. This time for sure!? ++ ++2002-12-03 18:36 rockyb ++ ++ * bashdb-fns.inc: More precise function comment. ++ ++2002-12-03 18:34 rockyb ++ ++ * bashdb-fns.inc: Bug: if _bashdb_old_set_opts was null. Test now. ++ ++2002-12-02 08:20 rockyb ++ ++ * TODO: Update as appropriate. ++ ++2002-12-02 06:58 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-12-02 06:42 rockyb ++ ++ * bashdb-cmds.inc, bashdb-file.inc, bashdb-init.inc: ++ bashdb-file.inc: source persistent only if it exists. ++ bashdb-init.inc/bashdb-cmds.inc: read persistent, write persistent. ++ ++2002-12-02 06:41 rockyb ++ ++ * bashdb-help.inc: Update doc for "x/examine" ++ ++2002-12-02 06:41 rockyb ++ ++ * bashdb-sig.inc: Hack (but works more often) cleanup of persistent ++ file. ++ ++2002-12-02 06:40 rockyb ++ ++ * CHANGES, NEWS, TODO: As appropriate. ++ ++2002-12-02 04:32 rockyb ++ ++ * CHANGES: Update as appropriate. ++ ++2002-11-30 19:13 rockyb ++ ++ * bashdb-hist.inc: Typo.. "H count" was broken. ++ ++2002-11-30 01:16 masata-y ++ ++ * bashdb: Fix a typo. ++ ++2002-11-26 10:47 rockyb ++ ++ * bashdb-init.inc, bashdb-set-d-vars.inc: More elegant (and I ++ wonder if not more robust) way to set $1 via set. Use "set -- xx ++ yy" rather than "set xx yy" ++ ++2002-11-26 07:23 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-11-26 07:09 rockyb ++ ++ * emacs/gud.el.diff: Set default minibuf-mod-map in bashdb read ++ call from Masatake YAMATO and add link to ++ http://bashdb.sourceforge.net ++ ++2002-11-26 07:04 rockyb ++ ++ * emacs/gud.el: Add HTTP reference. ++ ++2002-11-26 06:34 rockyb ++ ++ * emacs/gud.el: Change from Masatake YAMATO in running bash ++ initially. Uses gud-minibuffer-local-map now. ++ ++2002-11-23 00:17 rockyb ++ ++ * bashdb-sig.inc: Don't wipe temp files unless subshell is 0. ++ Likewise don't enter command shell if subshell is not 0. Don't know ++ that this will make much difference, but I guess it's the right ++ thing to do. ++ ++2002-11-22 22:02 rockyb ++ ++ * bashdb-help.inc: Document help "set showcommand" ++ ++2002-11-22 21:50 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-11-22 21:50 rockyb ++ ++ * bashdb-init.inc: Bug: if no script we were running "set" to dump ++ environment. ++ ++2002-11-22 20:57 rockyb ++ ++ * bashdb-cmds.inc, bashdb-list.inc, bashdb-set.inc, ++ emacs/bashdb.el, emacs/gud.el: We now keep track of the subshell ++ level: dynamic veriable BASH_SUBSHELL and the debugger prints these ++ as ()'s inside of the prompt. ++ ++2002-11-20 12:00 rockyb ++ ++ * TODO: Current state of things and known bugs. ++ ++2002-11-20 10:11 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-11-20 09:55 rockyb ++ ++ * bashdb-init.inc: "bash --debugger script args" was not setting ++ args properly. ++ ++2002-11-19 23:02 rockyb ++ ++ * bashdb-cmds.inc: Changes to have regression tests work with ++ different paths. ++ ++2002-11-19 22:17 rockyb ++ ++ * bashdb: We have a better expand_fn now if we need it. ++ ++2002-11-19 21:44 rockyb ++ ++ * Makefile.am: This is a pain. Get date and autoregen right on ++ bashdb-pre.inc. ++ ++2002-11-19 21:26 rockyb ++ ++ * TODO: Mostly hacks to make a distribution work smoothly. ++ MANIFEST: add DVI's. This reduce TeX dependency, okay? TODO: ++ document bugs configure.in touch files so that we have right the ++ first time. ++ ++2002-11-19 20:26 rockyb ++ ++ * bashdb-cmds.inc: Variable name typo. When rerunning, show full ++ command now. Is messing up a bit too much. ++ ++2002-11-19 20:25 rockyb ++ ++ * bashdb-set.inc: Update help. ++ ++2002-11-19 20:21 rockyb ++ ++ * bashdb: Remove readonly attributes because when we run as a ++ subshell these want to get set again. May have to rethink how to ++ do. ++ ++2002-11-19 20:20 rockyb ++ ++ * Makefile.am: Put touch inside build of bashd-pre.inc to avoid ++ redoing work. ++ ++2002-11-19 20:19 rockyb ++ ++ * bashdb-help.inc: Add set documentation. Put messages in a form ++ that DDD will understand without any hassle. ++ ++2002-11-19 16:01 rockyb ++ ++ * NEWS: Update as appropriate. Gee there are lots of changes, ++ mostly bug fixes. ++ ++2002-11-19 15:32 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, bashdb-help.inc, ++ bashdb-io.inc, bashdb-main.inc: Canonicalize filevars via a file ++ expand routine. Allows multpile to refer to a given file. ++ ++ Add option to have files show only short (basename) names, useful ++ in testing ++ ++ Add set/show of these. ++ ++ Bug in not setting fntrace correctly on debugger exit. ++ ++ Add -c option on bashdb. Analogous to "bash -c string" ++ ++2002-11-19 15:16 rockyb ++ ++ * bashdb-help.inc, bashdb-set.inc: Add set/show options to debug ++ debugger and have filenames get listed as basename only. ++ ++2002-11-19 15:15 rockyb ++ ++ * Makefile.am, bashdb-file.inc: Break out more routines into ++ bashdb-file.inc. Add _bashdb_expand_filename to canonicalize file ++ names. ++ ++2002-11-19 10:36 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc: Fix bug in continue comand. ++ Moved code from bashdb-brk.inc to bashdb-cmds.inc ++ ++2002-11-19 02:58 rockyb ++ ++ * bashdb-brk.inc, bashdb-init.inc, bashdb-list.inc, ++ bashdb-pre.inc.in: Use common routines now. Spelled "Bourne" ++ incorrectly in bashdb-{pre,init}.inc ++ ++2002-11-19 02:55 rockyb ++ ++ * bashdb-fns.inc: Add common routines _bashdb_check_line and ++ _bashdb_get_maxline. Reduces code bulk while making things more ++ reliable. _get_maxline now reads in the file if it hasn't been read ++ in yet. ++ ++2002-11-19 02:53 rockyb ++ ++ * bashdb-io.inc: Test to see if file exists before reading. We ++ could probably do better, but will need more code reorganization ++ for this. ++ ++2002-11-18 22:14 rockyb ++ ++ * bashdb-brk.inc: A little closer for FreeBSD... ++ ++2002-11-18 22:08 rockyb ++ ++ * bashdb-brk.inc, bashdb-stack.inc: Even closer to getting FreeBSD ++ regression tests work okay. ++ ++2002-11-18 21:50 rockyb ++ ++ * bashdb-list.inc: More changes to make FreeBSD regression tests ++ work out -- more to come... ++ ++2002-11-18 21:31 rockyb ++ ++ * bashdb, bashdb-brk.inc, bashdb-init.inc, bashdb-list.inc, ++ bashdb-stack.inc: Add -B option to list only basename in debugger ++ output. Is useful for FreeBSD which list files as foo rather than ++ ./foo. ++ ++2002-11-18 19:40 rockyb ++ ++ * .cvsignore, bashdb-pre.inc.in: Release information included via ++ bashdb-pre.inc (via bashdb-pre.inc.in) ++ ++2002-11-18 19:36 rockyb ++ ++ * Makefile.am, NEWS, bashdb, bashdb-init.inc, bashdb-init.inc.in, ++ bashdb.in: Add release information via new file bashdb-pre.in ++ rather than modify bashdb-init.inc and bashdb ++ ++2002-11-18 09:32 rockyb ++ ++ * NEWS: typo ++ ++2002-11-18 09:30 rockyb ++ ++ * NEWS: Note configure script bug fix. ++ ++2002-11-18 08:57 rockyb ++ ++ * configure.in: Clean up a little. ++ ++2002-11-17 20:26 rockyb ++ ++ * Makefile.am: Changes to make .in targets get recreated. ++ ++2002-11-17 19:44 rockyb ++ ++ * NEWS, TODO: Update to reflect current situation. ++ ++2002-11-17 19:41 rockyb ++ ++ * bashdb-sig.inc: Bug: don't set exit handler inside debug trap ++ handler. Seems to kick to be inhereted on subshells and then we ++ call the exit routine too often. ++ ++ exit_handler more like the other handlers. Understands ++ "stop/nostop, print/noprint stack/nostack". Does more of the same ++ things before entering cmdloop that debug_handler does via a more ++ common debugger enter routine. ++ ++ Saving _curline now done in common debbber enter routine ++ ++2002-11-17 19:38 rockyb ++ ++ * bashdb-cmds.inc: File got trunctated. Like bashdb-help.inc -- ++ Hmm. Something's fishy. ++ ++ Anyway, "debug" with no arguments now uses the last command rather ++ the last statement. This should be correct more often. ++ ++2002-11-17 19:34 rockyb ++ ++ * bashdb-fns.inc, bashdb-main.inc: _curline is now set by ++ BASH_LINENO rather than passing as a parameter to the trap. ++ ++ Add this common thing inside common debugger enter routine. ++ ++2002-11-17 19:30 rockyb ++ ++ * bashdb-help.inc: info signals now shows EXIT in list. ++ ++ Some help text got truncated. ++ ++ Move CVS Id line to end to facilitate debugging. ++ ++2002-11-17 19:26 rockyb ++ ++ * Makefile.am: Add new .in files to distribution ++ ++2002-11-17 19:23 rockyb ++ ++ * bashdb: Derived now from bashdb.in ++ ++2002-11-17 19:23 rockyb ++ ++ * bashdb-init.inc: Derived now from bashdb-init.in ++ ++2002-11-17 19:22 rockyb ++ ++ * bashdb-init.inc.in, bashdb.in: bashdb and bashdb-init.inc are now ++ derived so that we can automatically put in the release number. ++ ++2002-11-16 11:17 rockyb ++ ++ * CHANGES: Update as appropriate ++ ++2002-11-16 09:49 rockyb ++ ++ * bashdb-list.inc: We now print the statement to be executed if it ++ was on the same line/file and the statement part is different from ++ the last time we ran this. ++ ++2002-11-16 09:48 rockyb ++ ++ * emacs/gud.el: Patched version for emacs 21.2 of gud.el with bash ++ support. ++ ++2002-11-16 09:47 rockyb ++ ++ * emacs/gud.el.diff: Patches that I hope will be on the next ++ version of Emacs (assuming the paperwork isn't lost, people don't ++ forget about the patch as a result of the time delay to do the ++ paperwork and so on.) ++ ++2002-11-16 09:45 rockyb ++ ++ * emacs/bashdb.el: Move from parent directory here. May have also ++ added provision for "finish" command. ++ ++2002-11-16 09:40 rockyb ++ ++ * bashdb-main.inc, bashdb-sig.inc: Now save the command that is to ++ be executed next. ++ ++2002-11-16 09:38 rockyb ++ ++ * bashdb-fns.inc: Add common function ++ _bashdb_set_to_return_from_debugger to regularize the things that ++ need to be done on exit. There are more now, since we are saving ++ the last command executed on this line of the source file. ++ ++2002-11-16 09:36 rockyb ++ ++ * bashdb-init.inc: Need to skip more initially for "bashdb -c" ++ setup. ++ ++2002-11-16 09:25 rockyb ++ ++ * bashdb: Add -c option analogous to "bash -c". ++ ++2002-11-14 15:45 rockyb ++ ++ * NEWS, bashdb-cmds.inc: More improvements to "x" command. Use ++ print as a fallback... ++ ++2002-11-14 09:25 rockyb ++ ++ * bashdb-help.inc: configuure.in: get ready for version 0.32 ++ bashdb.texi: minor typo changes debugger/help/sig: Remove "Num" ++ from title as it is no longer there. ++ ++2002-11-14 09:08 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-11-14 08:15 rockyb ++ ++ * bashdb-help.inc, bashdb-sig.inc: A tad more like gdb: no signal ++ numbers listed. No longer trap on SIGWINCH. ++ ++2002-11-13 23:39 rockyb ++ ++ * NEWS, bashdb-help.inc, bashdb-sig.inc: Add help for "handle". ++ Tidy generic signal handler more and make more general (e.g ++ unconditinally turn off set tracing). ++ ++2002-11-13 23:22 rockyb ++ ++ * bashdb-fns.inc, bashdb-sig.inc: Clean up signal handling code a ++ small amount by creating a common routine to set some global ++ variables. ++ ++2002-11-13 19:33 rockyb ++ ++ * bashdb-cmds.inc, bashdb-help.inc, bashdb-sig.inc: Signal handling ++ is more like gdb now. No doubt there're room for improvement, but ++ this is probably a big improvement. ++ ++2002-11-12 15:20 rockyb ++ ++ * NEWS, TODO, bashdb-brk.inc, bashdb-cmds.inc, bashdb-init.inc, ++ bashdb-sig.inc: Some work on signal handling. More is there, more ++ to do. So what else is new? ++ ++2002-11-12 09:30 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-11-12 07:45 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc: Add gdb's ++ "condition" command. ++ ++2002-11-12 07:42 rockyb ++ ++ * bashdb-sig.inc: Oops --- typo in variable name. ++ ++2002-11-12 07:41 rockyb ++ ++ * bashdb-sig.inc: Add provision for debugging debugger. ++ ++2002-11-12 07:40 rockyb ++ ++ * bashdb-io.inc: Add more file translation characters for ", [, and ++ ] and reorder letters to make mnemonic. ++ ++2002-11-11 23:56 rockyb ++ ++ * bashdb-sig.inc: Caught another routine that didn't start _bashdb_ ++ ++2002-11-11 23:45 rockyb ++ ++ * TODO: Cross many things off list. ++ ++2002-11-11 23:45 rockyb ++ ++ * NEWS: Update as appropriate - major changes. ++ ++2002-11-11 23:44 rockyb ++ ++ * Makefile.am: Move bashdb.el into emacs directory. ++ ++2002-11-11 16:45 rockyb ++ ++ * bashdb.el: Moved inside emacs directory now. ++ ++2002-11-10 21:26 rockyb ++ ++ * NEWS: Believe it or not, the C files have gotten a little bit ++ *simpler*. We are closer towards having line numbers recorded ++ accurately in BASH_LINENO (although things are not perfect yet). ++ ++2002-11-10 14:39 rockyb ++ ++ * bashdb-help.inc, bashdb-stack.inc, bashdb.el: bashdb-help: Make ++ more like gdb - facilitates use in ddd. bashdb-stack.inc: Fix bug ++ in "stack_down". (There's another one though lurking) bashdb.el: ++ add key mapping for finish and other keymappings tests/misc: change ++ in "show" output tests/brkpt2: perhaps better although not quite ++ perfect. ++ ++2002-11-10 11:35 rockyb ++ ++ * bashdb-help.inc: Add help for finish command. Invalid help ++ command now reports what gdb does. ++ ++2002-11-09 23:54 rockyb ++ ++ * bashdb-help.inc: Missed a place that should have added _bashdb. ++ ++2002-11-09 20:28 rockyb ++ ++ * bashdb-set.inc: Make show command work like gdb. ++ ++2002-11-09 20:16 rockyb ++ ++ * bashdb-io.inc: Missed a couple more "builtin printf"'s ++ ++2002-11-09 20:15 rockyb ++ ++ * bashdb-io.inc: Specify *builtin* printf. ++ ++2002-11-09 07:38 rockyb ++ ++ * Makefile.am, NEWS, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-hist.inc, bashdb-io.inc, bashdb-main.inc, bashdb-set.inc: ++ echo -> builtin echo Add "info signals" and "info handler" Break ++ out history into new file bashdb-hist.inc. Help command change v ++ -> M ++ ++2002-11-08 21:08 rockyb ++ ++ * bashdb-sig.inc: Previous flie was horribly truncated. ++ ++2002-11-05 22:43 rockyb ++ ++ * bashdb-sig-ret.inc: bashdb.texi: Document finish command. ++ ++ debugger/tests*, MANIFEST, Makefile.am: Add finish command test. ++ ++ *sig: add signal save/restor test ++ ++2002-11-02 15:17 rockyb ++ ++ * NEWS: Today's hacks. ++ ++2002-11-02 14:46 rockyb ++ ++ * Makefile.am, bashdb-list.inc, bashdb-sig-ret.inc, bashdb-sig.inc: ++ Save/restore signals on entering debugger. ++ ++2002-11-02 10:16 rockyb ++ ++ * bashdb-cmds.inc, bashdb-sig.inc: Workaround? for bash bug ++ introduced implementing "trap RETURN" ++ ++2002-11-02 09:15 rockyb ++ ++ * bashdb-sig.inc: [no log message] ++ ++2002-11-02 09:15 rockyb ++ ++ * bashdb-sig.inc: Small comment changes. ++ ++2002-11-02 09:12 rockyb ++ ++ * CHANGES: Reflects better the changes. ++ ++2002-11-02 09:03 rockyb ++ ++ * CHANGES: Update as appropriate. ++ ++2002-11-02 08:12 rockyb ++ ++ * NEWS, bashdb-cmds.inc, bashdb-sig.inc, bashdb-stack.inc: Fix bug ++ where stack trace current line entry was sometimes wrong. No ++ longer mangle line number in stack changing commands. ++ ++ Printing source line on "RETURN" debugger call. ++ ++2002-11-01 07:14 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, bashdb-io.inc, ++ bashdb-list.inc, bashdb-main.inc, bashdb-sig.inc: Add gdb's finish ++ or perl5db's return. Line number on return statement is funny and ++ there are some bugs to make fully work. ++ ++ Now all debugger routines start _bashdb. ++ ++2002-10-31 05:59 rockyb ++ ++ * TODO: [no log message] ++ ++2002-10-30 22:11 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-10-30 22:03 rockyb ++ ++ * bashdb-cmds.inc, bashdb-help.inc: Allow negative numbers in ++ history commands. Get ready for release 0.30. ++ ++2002-10-30 07:13 rockyb ++ ++ * TODO: Cross some things off the list. ++ ++2002-10-29 08:54 rockyb ++ ++ * NEWS, bashdb-cmds.inc, bashdb-help.inc: Make history command more ++ robust and allow for more features. How allows for !n:p. We now ++ give an error message if the history numbers are out of range. ++ Separate history parsing into a subroutine to tidy and facilitate ++ future features. ++ ++2002-10-27 00:05 rockyb ++ ++ * NEWS: Update as appropriate. ++ ++2002-10-27 00:02 rockyb ++ ++ * bashdb-set-d-vars.inc: Better way to set dollar parameters. It ++ handles more than $1..$9 and sets $# correctly too. ++ ++2002-10-26 23:17 rockyb ++ ++ * bashdb-init.inc: bug in bashdb --debugger some local's must be ++ declares. ++ ++2002-10-26 10:27 rockyb ++ ++ * NEWS, bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-init.inc, bashdb-set.inc: Fix/change history. Numbers now ++ match prompt numbers. Can use !n as well as hi n ++ ++ Add GDB's "set editing" command. ++ ++ _bashdb_help_set added to show set commands. Made more like GDB's ++ output. ++ ++2002-10-25 08:15 rockyb ++ ++ * bashdb-cmds.inc: Make H be like perl's. Fix history (hi) command. ++ Don't but some commands like comment and history in the history. ++ Test. ++ ++2002-10-25 08:05 rockyb ++ ++ * bashdb-cmds.inc: Make H be more like perl's. Make hi work. ++ ++2002-10-25 07:07 rockyb ++ ++ * bashdb-set.inc: Tidy up show output. Tests changed largely as a ++ result of size increase in bashdb from last change. run-misc: ++ wasn't giving an error is diff failed on first diff check. ++ ++2002-10-25 06:44 rockyb ++ ++ * NEWS: Latest news... ++ ++2002-10-25 06:41 rockyb ++ ++ * bashdb, bashdb-cmds.inc, bashdb-set.inc: Add "set/show prompt." ++ Make prompt customizable. bashdb: all -t to be a filename. ++ ++2002-10-24 09:33 rockyb ++ ++ * bashdb-set.inc, bashdb-sig.inc: Comment about why slightly more ++ complex loop to set args. ++ ++2002-10-24 09:08 rockyb ++ ++ * NEWS, bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-init.inc, bashdb-io.inc, bashdb-set.inc: - Fix bugs in ++ debugger output when debugged program redirects I/O ++ ++ - Add GDB's "set args" command ++ ++2002-10-23 09:04 rockyb ++ ++ * bashdb: Remove debug statement accidentally committed. ++ ++2002-10-23 08:51 rockyb ++ ++ * bashdb, bashdb-init.inc, bashdb-io.inc: bashdb, bashdb-init.inc: ++ Add -t option to set tty bashdb-io.inc: bug in not testing ++ _bashdb_tty in _bashdb_printf tests: output changes bashdb script ++ has gotten larger. ++ ++2002-10-23 08:49 rockyb ++ ++ * TODO: Update as appropriate. ++ ++2002-10-23 08:26 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-init.inc, bashdb-io.inc, bashdb-set.inc: Change variable ++ $_tty to $_bashdb_tty ++ ++ All "read -e -p" now have input and output redirected. ++ ++ Implement tty command to set tty. ++ ++2002-10-22 06:25 rockyb ++ ++ * bashdb-cmds.inc: Fix bug in printing debugger prompt when ++ debugged program has redirected stderr. Thanks to Masatake YAMATO. ++ ++2002-10-21 10:59 rockyb ++ ++ * bashdb-cmds.inc: Remove trailing blanks. ++ ++2002-10-21 10:48 rockyb ++ ++ * bashdb-io.inc, bashdb-list.inc, bashdb-sig.inc: Just format ++ changes: remove blanks at the ends of some lines. ++ ++2002-10-21 10:26 rockyb ++ ++ * .cvsignore: Allow for different versions of autotools. ++ ++2002-10-21 10:25 rockyb ++ ++ * bashdb-help.inc: Bug in "help return" -- missing some lines ++ there. ++ ++2002-10-19 09:26 rockyb ++ ++ * bashdb-help.inc: Add documentation for new command return. ++ bashdb.texi: had duplicate help screens ++ ++2002-10-19 09:08 rockyb ++ ++ * NEWS: Minor typographical changes. ++ ++2002-10-18 23:24 rockyb ++ ++ * bashdb-cmds.inc: Make so regression test is portable. A hack for ++ now ++ ++2002-10-18 23:16 rockyb ++ ++ * bashdb-cmds.inc: Correct debug test. ++ ++2002-10-18 22:40 rockyb ++ ++ * bashdb-set-d-vars.inc: To set dollar variables ($1, $2, ... $?) ++ before evaluation of a user expression. ++ ++2002-10-18 21:31 rockyb ++ ++ * Makefile.am, bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-io.inc, bashdb-sig.inc: A better way of handling setting ++ dollar variables. ++ ++ Add debug test. ++ ++2002-10-18 08:44 rockyb ++ ++ * bashdb-brk.inc, NEWS: Set $1, $2 before watch-expression ++ evaluation ++ ++2002-10-18 08:10 rockyb ++ ++ * NEWS, bashdb-cmds.inc, bashdb-sig.inc: $? is saved and preserved ++ for use inside "eval" and "print" ++ ++2002-10-18 06:38 rockyb ++ ++ * bashdb-set.inc: Correct show version to add more sources. ++ ++2002-10-18 06:32 rockyb ++ ++ * NEWS, bashdb-cmds.inc, bashdb-sig.inc: Fix bug in skip *count*. ++ ++2002-10-17 08:56 rockyb ++ ++ * NEWS, TODO, bashdb-cmds.inc, bashdb-main.inc: bashdb-main.inc: ++ Bug in changing to _bashdb_libdir bashdb-cmds.inc: number of times ++ nested in debug listed via exported variable ++ BASHDB_LEVEL ++ ++2002-10-17 08:11 rockyb ++ ++ * bashdb-cmds.inc: Remove debugging statments. ++ ++2002-10-17 00:10 rockyb ++ ++ * NEWS, TODO, bashdb, bashdb-cmds.inc: Get "debug" working much ++ better: variables (e.g. $0) are substituted now. "bash ++ --debugger" invocations work since we no longer fork _libdir ++ renamed to _bashdb_libdir (in bashdb). We add "-L ++ _bashdb_libdir" when executing via bashdb script. ++ ++2002-10-16 08:41 rockyb ++ ++ * bashdb-help.inc: Document skip and debug. ++ ++2002-10-16 08:40 rockyb ++ ++ * bashdb-cmds.inc: Make skip work like next/step. ++ ++2002-10-16 08:22 rockyb ++ ++ * NEWS: As appropriate. ++ ++2002-10-15 23:29 rockyb ++ ++ * bashdb-help.inc: Document skip command. ++ ++2002-10-15 22:57 rockyb ++ ++ * bashdb-cmds.inc, bashdb-sig.inc: Add ability to skip execution of ++ a command. Debugger command: skip ++ ++2002-10-11 23:18 rockyb ++ ++ * NEWS: Update to reflect current sitiation ++ ++2002-10-11 23:17 rockyb ++ ++ * bashdb-list.inc: Fix bug in search forward not resetting line to ++ start. ++ ++ Search now lists only one line same as gdb and perl5db do. Update ++ tests as appropriate. ++ ++2002-10-10 21:08 rockyb ++ ++ * TODO: Update to reflect current situtation: parameters now work. ++ ++2002-10-10 20:57 rockyb ++ ++ * bashdb-cmds.inc, bashdb-main.inc, bashdb-sig.inc: Fix bug so that ++ $1 preserve embedded blanks (or more precisely, embedded IFS) that ++ they might have had. ++ ++2002-10-10 08:47 rockyb ++ ++ * bashdb-cmds.inc: Bug in _bashdb_cmd_eval. Was not erasing ++ evalfile before starting if $1.. was empty, so we'd have ++ accumulated output. ++ ++2002-10-10 08:30 rockyb ++ ++ * .cvsignore: Add more good derived stuff (from autotools) ++ ++2002-10-10 08:17 rockyb ++ ++ * bashdb-cmds.inc, bashdb-io.inc, bashdb-sig.inc, bashdb-stack.inc: ++ Add parameters in stack trace. And can now refer to $1 $2 $3 in ++ print and eval commands. ++ ++2002-10-03 08:40 rockyb ++ ++ * bashdb-list.inc: Fix up search command to do the right thing more ++ often. However need to make another pass to match gdb semantics ++ more perfectly. Save last search pattern and can be repeated if no ++ pattern given. ++ ++2002-09-24 13:30 rockyb ++ ++ * Makefile.am, NEWS, bashdb-brk.inc, bashdb-cmds.inc, ++ bashdb-help.inc, bashdb-io.inc, bashdb-list.inc, bashdb-main.inc, ++ bashdb-sig.inc: Add "search" "reverse" (forward/backward searching) ++ and Perl's "a" (action) commands. bashdb-list.inc split off for the ++ debugger growth. ++ ++2002-09-24 01:26 rockyb ++ ++ * bashdb-help.inc: Add tb to help list. ++ ++2002-09-23 18:45 rockyb ++ ++ * NEWS: [no log message] ++ ++2002-09-23 18:19 rockyb ++ ++ * NEWS: [no log message] ++ ++2002-09-23 18:09 rockyb ++ ++ * NEWS: Make mroe readible. ++ ++2002-09-23 17:52 rockyb ++ ++ * bashdb.el: Small typo ++ ++2002-09-23 17:47 rockyb ++ ++ * bashdb.el: Messed up on author and didn't comment properly. ++ ++2002-09-23 17:41 rockyb ++ ++ * MANIFEST: This really isn't used. Autotools rules. (But really we ++ do a make dist in the parent directory.) ++ ++2002-09-23 09:09 rockyb ++ ++ * bashdb-brk.inc, bashdb-io.inc: Change format of display output to ++ identify display number and expression. ++ ++2002-09-22 07:40 rockyb ++ ++ * bashdb-help.inc: Add display regression test. bashdb-help.inc: ++ change lines to linespecs in help. Add missing _bashdb_msg. ++ ++2002-09-21 22:46 rockyb ++ ++ * bashdb.el: Don't know how to spell my own name! ++ ++2002-09-21 22:35 rockyb ++ ++ * bashdb-help.inc: Add help for display. correct en/dis help to ++ watchpoint and display. ++ ++2002-09-21 21:41 rockyb ++ ++ * NEWS, TODO, bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc: Add ++ gdb's auto display commands (display, undisplay, enable/disable ++ display) ++ ++2002-09-21 06:41 rockyb ++ ++ * bashdb-cmds.inc: debug command a tad better -- now allows for ++ command to get run. ++ ++2002-09-21 06:17 rockyb ++ ++ * bashdb-cmds.inc: Okay, now get out some of the bugs in the ++ "debug" command. More work is needed to make this more seamless. ++ (In particular need to skip over execution after debugged execution ++ happens). ++ ++2002-09-21 05:50 rockyb ++ ++ * NEWS, bashdb-cmds.inc, bashdb-sig.inc: Add provision to debug ++ into a script ("debug" command), and leave script without going ++ back into debugger command loop (BASHDB_QUIT_ON_QUIT environment ++ variable). ++ ++2002-09-19 14:49 rockyb ++ ++ * NEWS: Add caller builtin ++ ++2002-09-19 13:01 masata-y ++ ++ * bashdb-help.inc, bashdb.el: bashdb-help.inc: fix typo(tbreak, ++ down), added [count] in up command's help message. bashdb.el: ++ added gud-tbreak. ++ ++2002-09-19 12:51 masata-y ++ ++ * bashdb.el: added simple comments about how to install bashdb.el. ++ provided bashdb. ++ ++2002-09-18 21:37 rockyb ++ ++ * NEWS, bashdb-brk.inc, bashdb-cmds.inc, bashdb-help.inc, ++ bashdb-sig.inc: Add temporary break. ++ ++2002-09-18 02:03 rockyb ++ ++ * bashdb-help.inc: Document "break" command change args help ++ linespec in online help where appropriate. ++ ++2002-09-17 09:01 rockyb ++ ++ * NEWS: typo. ++ ++2002-09-17 08:03 rockyb ++ ++ * NEWS: Last minute additions for version 0.25 - recent doc ++ changes. ++ ++2002-09-17 07:56 rockyb ++ ++ * INSTALL: Remove generic instructions and customize for something ++ specific for the debugger. Should be more helpful, I think. ++ ++2002-09-16 17:37 rockyb ++ ++ * NEWS, bashdb-fns.inc, bashdb-main.inc, bashdb-sig.inc: Move debug ++ trap handler to bashdb-sig.inc and rename it, well, ++ debug_trap_handler. ++ ++ Save restart set parameters and inside set -xv (at least ++ initially). ++ ++2002-09-14 18:26 rockyb ++ ++ * NEWS: Update. ++ ++2002-09-14 18:14 rockyb ++ ++ * bashdb-help.inc: Change info args to match closer gdb. Is now $1, ++ $2 and so on rather than initial program arguments. ++ ++2002-09-14 17:54 rockyb ++ ++ * bashdb-fns.inc, bashdb-main.inc: Now support $1 $2 as ${ARG[1}}, ++ ${ARG[2]}. ++ ++2002-09-14 14:51 rockyb ++ ++ * bashdb-cmds.inc: Missed place in renaming _hi to _bashdb_hi ++ ++2002-09-14 12:19 rockyb ++ ++ * NEWS, TODO, bashdb, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-init.inc, bashdb-sig.inc: Variable name cleanup: more now ++ start with _bashdb ++ ++2002-09-14 12:09 rockyb ++ ++ * bashdb-cmds.inc: Some variable name cleanup and removal. ++ ++2002-09-14 11:40 rockyb ++ ++ * bashdb-cmds.inc: Correct "restart" command when running via "bash ++ -c ..." ++ ++2002-09-14 07:36 rockyb ++ ++ * NEWS, bashdb-init.inc, bashdb-io.inc: Small cleanups. BOGUS file ++ renamed to *BOGUS*. ++ ++2002-09-14 07:21 rockyb ++ ++ * bashdb, bashdb-cmds.inc, bashdb-help.inc, bashdb-init.inc, ++ bashdb-main.inc: Clean up code for debugger source command. Add ++ "info source" ++ ++2002-09-14 01:09 rockyb ++ ++ * bashdb-cmds.inc, bashdb-init.inc: Slightly cleaner code. ++ ++2002-09-14 00:51 rockyb ++ ++ * NEWS, bashdb-cmds.inc, bashdb-init.inc: Command files now nest ++ and pick up where they left off in between script execution. There ++ are still some minor cleanups and a small bug here or there. ++ ++2002-09-13 23:04 rockyb ++ ++ * bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc, bashdb-io.inc: ++ bash --debugger -c "..." now works. (Bug in restarting currently ++ though. restart now picks up full bash name from BASH environment ++ variable. ++ ++ To accomplish the above, environment variable BASH_EXECUTION_STRING ++ was added which contains the contents what was passed after -c. ++ ++2002-09-13 21:24 rockyb ++ ++ * NEWS, bashdb-brk.inc: Fix bug where "watche x > 24" was creating ++ output to file 24 ++ ++2002-09-13 15:09 rockyb ++ ++ * NEWS: Add: Function names now supported by changing the semantics ++ of declare -F. ++ ++2002-09-13 15:06 rockyb ++ ++ * NEWS: Function names now supported. As a result, can now set ++ breakpoints on them. However in non-interactive mode the line ++ number is wrong. It is the caller line. ++ ++2002-09-13 15:03 rockyb ++ ++ * TODO, bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc: Function ++ names now supported. As a result, can now set breakpoints on them. ++ However in non-interactive mode the line number is wrong. It is the ++ caller line. ++ ++ As a result of the above, now extended these commands to include ++ linespecs: continue, break, list. ++ ++2002-09-12 18:18 rockyb ++ ++ * NEWS, bashdb-cmds.inc, bashdb-io.inc: Add list file:line. Rename ++ internal routines from print to list since that's what the command ++ is called. ++ ++2002-09-12 16:17 rockyb ++ ++ * bashdb-cmds.inc, bashdb-fns.inc: Change names of variables so as ++ not to conflict with user variables. Thus more of them are ++ prefaced with "_bashdb_." May have introduced a bug in the renaming ++ though... ++ ++2002-09-12 06:44 rockyb ++ ++ * NEWS, TODO: Now list/stop line of "for" or "select". ++ ++2002-09-11 22:36 rockyb ++ ++ * TODO: Note more line number problems. ++ ++2002-09-11 22:36 rockyb ++ ++ * bashdb-sig.inc: Add cmd-loop loop in exit handler so we in fact ++ don't exit! ++ ++2002-09-11 22:13 rockyb ++ ++ * bashdb-fns.inc: Don't go into debugger command loop when called ++ from the debugger: (the caller is _bashdb_). ++ ++2002-09-11 07:39 rockyb ++ ++ * bashdb-brk.inc, bashdb-fns.inc, bashdb-init.inc: Add break ++ file:line syntax. Add counts on watchpoints and display of this on ++ listing. Display breakpoint number when adding. i ++ ++2002-09-11 07:38 rockyb ++ ++ * NEWS: Version 0.23 changes (so far) ++ ++2002-09-11 00:36 rockyb ++ ++ * bashdb-cmds.inc: Make safe for tampering with IFS. ++ ++2002-09-10 13:25 rockyb ++ ++ * bashdb-cmds.inc: Fix bug in comment without space after #, e.g. ++ #foo. ++ ++2002-09-10 13:13 rockyb ++ ++ * bashdb-cmds.inc: Improve comment pattern so "#foo" works as does ++ "# foo." ++ ++2002-09-10 12:49 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc: Bugs fixed in getting watch ++ command regression tests hammered out; eval now checks for tty, ++ listing watchpoints quotes expressions. ++ ++2002-09-10 11:47 rockyb ++ ++ * TODO: Cross a couple of things off the list. ++ ++2002-09-10 11:02 rockyb ++ ++ * bashdb-brk.inc: Test to see that line breaks are inside lines ++ Remove bug in not setting break condition correctly. Message if no ++ breakpoints have been set. ++ ++2002-09-10 11:00 rockyb ++ ++ * bashdb-cmds.inc: More crud in comments. ++ ++2002-09-10 10:59 rockyb ++ ++ * bashdb-cmds.inc: Removed some crud. ++ ++2002-09-10 10:58 rockyb ++ ++ * bashdb-cmds.inc: Rename to _bashdb_cmd_break to be consistent in ++ naming. ++ ++2002-09-10 09:08 rockyb ++ ++ * Makefile.am, configure.in: Add support to create/run regression ++ tests in tests directory. ++ ++2002-09-09 22:50 rockyb ++ ++ * bashdb-brk.inc, bashdb-cmds.inc: Enable/disable of watchpoints ++ done. delete, enable and disable will allow suffix "w" for ++ watchpoint. ++ ++2002-09-09 20:55 rockyb ++ ++ * bashdb-help.inc: Add help for set command. Minor text changes. ++ ++2002-09-09 19:45 rockyb ++ ++ * NEWS: Add case statement work. ++ ++2002-09-09 19:09 rockyb ++ ++ * TODO: Update as appropriate. ++ ++2002-09-09 19:02 rockyb ++ ++ * TODO: Update as appropriate. ++ ++2002-09-09 12:39 rockyb ++ ++ * Makefile.am, TODO, configure.in: Use Automake AM_LISPDIR rather ++ than trying to roll this code on my own (and failing). Thanks to ++ Masatake YAMATO! ++ ++2002-09-09 12:37 rockyb ++ ++ * NEWS: Add 0.22 features. ++ ++2002-09-08 18:57 rockyb ++ ++ * bashdb-set.inc: Set/show commands ++ ++2002-09-07 14:41 rockyb ++ ++ * bashdb-cmds.inc: Add frame command. Version command is now a ++ separate command. ++ ++2002-09-07 14:35 rockyb ++ ++ * bashdb-stack.inc: Get frame with args command hammered out ++ properly. ++ ++2002-09-07 09:36 rockyb ++ ++ * bashdb-io.inc: Some error correction and error message on list ++ command. ++ ++2002-09-07 09:35 rockyb ++ ++ * bashdb-cmds.inc: Ooops - forgot to allow set/show as valid ++ commands here. Move CVS Id to end. ++ ++2002-09-07 09:09 rockyb ++ ++ * Makefile.am, bashdb-cmds.inc, bashdb-fns.inc, bashdb-help.inc, ++ bashdb-init.inc, bashdb-io.inc, bashdb-main.inc: Start set/show ++ commands. List now takes up where it left off. Move some CVS id ++ lines to the end to facilitate debugging Some minor improvements. ++ ++2002-09-06 03:06 rockyb ++ ++ * bashdb-brk.inc, bashdb-help.inc, bashdb-main.inc: Minor formating ++ changes or typos. ++ ++2002-09-06 01:56 rockyb ++ ++ * Makefile.am: Include new guy bashdb-sig.inc. ++ ++2002-09-05 22:39 rockyb ++ ++ * bashdb-cmds.inc, bashdb-main.inc, bashdb-sig.inc: Add signal ++ handlers for interrupt and exit. Now we catch when the program ++ stopped and report that. ++ ++2002-09-05 22:37 rockyb ++ ++ * TODO: Update as appropriate. ++ ++2002-09-05 10:54 rockyb ++ ++ * bashdb-main.inc: Slightly better? ++ ++2002-09-05 10:53 rockyb ++ ++ * bashdb-main.inc: Test to see if we are using a debugger-enabled ++ bash, and give up if not. ++ ++2002-09-05 08:56 rockyb ++ ++ * bashdb-brk.inc: Add null line + if in break command. ++ ++2002-09-04 22:24 rockyb ++ ++ * TODO: Update. ++ ++2002-09-04 22:08 rockyb ++ ++ * bashdb, bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-help.inc, bashdb-init.inc, bashdb-io.inc: Redo the way ++ breakpoints are handled. Is more like gdb than perltdb. ++ ++ Now have enable/disable and breakpoint conditions. ++ ++ Some other minor changes. Like testing to see if we have an ++ attached tty. ++ ++2002-09-03 13:22 rockyb ++ ++ * bashdb: Change option -s to -q as it is in gdb. ++ ++2002-09-03 12:47 rockyb ++ ++ * bashdb, bashdb-init.inc: Add more options - x: command file run, ++ n: no init file run, -V show version, and -s silient. All ++ compatible with GDB. ++ ++2002-09-03 12:36 rockyb ++ ++ * bashdb-help.inc: Minor change to refer to doc. ++ ++2002-09-03 12:30 rockyb ++ ++ * Makefile.am: Add newly-created file bashdb-io.inc ++ ++2002-09-02 22:37 rocky ++ ++ * TODO, bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-help.inc, bashdb-init.inc, bashdb-io.inc, bashdb-main.inc, ++ bashdb-stack.inc: Major cleanup/expansion and a ++ bug/misunderstanding removed. Now reads in files once. Unset ++ breakpoint code no longer inlined. "info files" now works Stack ++ trace bug when via "info stack" fixed. bashdb-io.inc created. ++ ++2002-09-02 15:29 rocky ++ ++ * bashdb, bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, ++ bashdb-help.inc, bashdb-init.inc: Add info command. In the process, ++ we now allow getting tty and args. ++ ++2002-09-02 00:35 rocky ++ ++ * TODO: Revise as appropriate. ++ ++2002-09-02 00:28 rocky ++ ++ * bashdb-help.inc: Update to reflect current situation. ++ ++2002-09-02 00:16 rocky ++ ++ * bashdb-cmds.inc, bashdb-help.inc, bashdb-stack.inc: Add bt ++ backtrace as an alias for where, or T. Add parameter to specify how ++ many frames to list on bt. ++ ++2002-09-01 19:57 rocky ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-stack.inc: Mostly ++ formatting. Change basic indent from 4 to 2. Stack commands were ++ duplicated in bashdb-cmds.inc, and a file identifier incorrect in ++ bashdb-brk.inc ++ ++2002-09-01 11:05 rocky ++ ++ * bashdb-brk.inc, bashdb-cmds.inc, bashdb-fns.inc, bashdb-help.inc, ++ bashdb-main.inc: Minor: correct function comments at top of file. ++ ++2002-09-01 11:05 rocky ++ ++ * Makefile.am: Break into more smaller files. ++ ++2002-09-01 11:04 rocky ++ ++ * bashdb-stack.inc: Break away from bashdb-cmds.inc ++ ++2002-09-01 10:41 rocky ++ ++ * bashdb-brk.inc: Things with breakpoints are now in a separate ++ file. ++ ++2002-09-01 10:40 rocky ++ ++ * bashdb-help.inc: Things with breakpoints and watchpoints (and ++ later actions) are now their own separate file ++ ++2002-09-01 10:39 rocky ++ ++ * bashdb-help.inc: Help is now it's own separate file. ++ ++2002-09-01 05:32 rocky ++ ++ * bashdb-cmds.inc, bashdb-fns.inc: Change pe to x. On watch, set ++ initial value (and display it). Allow dump of variable. Add ++ internal function _is_var. ++ ++2002-09-01 04:08 rocky ++ ++ * bashdb-fns.inc: Separate watch variables from watch expressions ++ ++2002-09-01 04:07 rocky ++ ++ * bashdb-init.inc: Add array to note if watchpoint is arithmetic or ++ not. ++ ++2002-09-01 04:06 rocky ++ ++ * bashdb-cmds.inc: Separate out watch variable and watch ++ expressions. ++ ++2002-09-01 02:47 rocky ++ ++ * bashdb-init.inc: Set default next/step default value. ++ ++2002-08-31 23:34 rocky ++ ++ * bashdb-cmds.inc, bashdb-init.inc: Make more like gdb. Implement ++ aliases: run, shell. Add Init file startup ~.bashdbinit. Save args ++ on print and pe. ++ ++2002-08-30 20:44 rocky ++ ++ * CHANGES: Note watchpoints, and line number changes. ++ ++2002-08-30 16:17 rocky ++ ++ * bashdb-cmds.inc, bashdb-fns.inc: Small cleanups. Remove function ++ break code which doesn't work anyway. ++ ++2002-08-30 12:46 rocky ++ ++ * bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc: Add ++ watchpoints. Help and commands now allow unique prefix, e.g. del ++ for delete. ++ ++2002-08-30 12:20 rocky ++ ++ * AUTHORS: Again. ++ ++2002-08-30 12:18 rocky ++ ++ * AUTHORS: More precies. ++ ++2002-08-30 00:55 rocky ++ ++ * TODO: Add watchpoints as requested by Bert Caridad. ++ ++2002-08-27 20:56 rocky ++ ++ * bashdb-cmds.inc: Fixed bug Inadvertently added by change \? to ?. ++ (Is now '?') to help emacs script editing. ++ ++2002-08-27 20:52 rocky ++ ++ * bashdb-cmds.inc: Doc continue enhancement. ++ ++2002-08-27 20:51 rocky ++ ++ * bashdb-cmds.inc, bashdb-fns.inc: clearing breakpoints now works ++ via a workaround. ++ ++2002-08-27 20:48 rocky ++ ++ * TODO: Update to reflect current situation. ++ ++2002-08-26 23:53 rocky ++ ++ * TODO: Did "up" and "down" - remove from list. ++ ++2002-08-26 23:52 rocky ++ ++ * bashdb.el: Add back in "up" and "down" which we have (but perldb ++ doesn't). ++ ++2002-08-26 23:51 rocky ++ ++ * bashdb, bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc: "Up ++ [count} and down {count] working. Stack trace is now more like gdb ++ than perl. Probably will move more in that direction, too. ++ ++ Note: There may be some funniness in FUNCNAME BASH_SOURCE and ++ BASH_LINENO. ++ ++2002-08-26 09:22 rocky ++ ++ * bashdb, bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc: S !pat ++ now works. Some alphabetization of commands and more subroutings in ++ bashdb-cmds start _bashdb_cmd_. init Has where we are in stack for ++ getting ready to put in up and down. ++ ++2002-08-26 09:20 rocky ++ ++ * TODO: Update. ++ ++2002-08-26 09:20 rocky ++ ++ * NEWS: More features - note that we're useful even if you don't ++ use the debugger. ++ ++2002-08-26 08:06 rocky ++ ++ * bashdb-cmds.inc: This time, for sure! ++ ++2002-08-26 08:05 rocky ++ ++ * bashdb-cmds.inc: - now lists up to _curline Bug in restart for ++ bash --debugger ++ ++2002-08-26 06:55 rocky ++ ++ * bashdb: Minor changes. ++ ++2002-08-26 05:45 rocky ++ ++ * CHANGES, TODO: Fix off-by-one line number error in command ++ subtitutions $(), `` and {}. ++ ++2002-08-25 23:10 rocky ++ ++ * bashdb-cmds.inc: Help for individual commands. ++ ++2002-08-25 23:09 rocky ++ ++ * TODO: Added emacs lisp install. help on commands and fixed ++ substutution line numbers. ++ ++2002-08-25 23:08 rocky ++ ++ * .cvsignore, Makefile.am, configure.in: batch-compile and install ++ emacs lisp files. (First cut: no doubt there's room for ++ improvement.) ++ ++2002-08-25 20:27 rocky ++ ++ * bashdb-cmds.inc: A couple more. ++ ++2002-08-25 20:17 rocky ++ ++ * bashdb-cmds.inc: Add long-form of commands. ++ ++2002-08-25 20:17 rocky ++ ++ * bashdb: Replace a [ with [[ ++ ++2002-08-25 17:50 rocky ++ ++ * CHANGES: Note line number change in command substitution. ++ ++2002-08-25 12:13 rocky ++ ++ * TODO: Another bug with subshells found. ++ ++2002-08-25 09:58 rocky ++ ++ * TODO, bashdb-cmds.inc: Fix restart in bash --debugger ++ ++2002-08-25 02:18 rocky ++ ++ * bashdb, bashdb-cmds.inc, bashdb-init.inc, bashdb-main.inc: Minor ++ variable name corrections and changes. Mostly to do with version ++ command. Typo in bashdb. ++ ++2002-08-25 02:05 rocky ++ ++ * AUTHORS, Makefile.am, PERMISSION, TODO, bashdb, bashdb-cmds.inc, ++ bashdb-fns.inc, bashdb-init.inc, bashdb-main.inc: Last minute ++ cleanups before initial Savannah CVS import. ++ ++2002-08-24 18:13 rocky ++ ++ * bashdb, bashdb-cmds.inc, bashdb-fns.inc: Mostly small extensions ++ to make this match perl5db: - Add Subroutine list (S) - Add list ++ functions (l func) - Add variable dump (V) - Note where we ++ haven't finished implementing parameters - Fix help ++ ++ bashdb: rename functions to more system-like names (so won't appear ++ in function list) ++ ++2002-08-24 18:08 rocky ++ ++ * TODO, NEWS: Update as appropriate. ++ ++2002-08-24 00:58 rocky ++ ++ * TODO: Fixed "continue" ++ ++2002-08-24 00:55 rocky ++ ++ * bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc: Breakpoints now ++ work! setting, deleting, clearing all, listing. ++ ++2002-08-23 14:15 rocky ++ ++ * bashdb, bashdb-cmds.inc, bashdb-fns.inc: CVS cleanup: add CVS ++ lines ++ ++2002-08-23 14:13 rocky ++ ++ * bashdb-fns.inc: Command "Continue" ('c") now runs silently. ++ ++2002-08-23 07:57 rocky ++ ++ * TODO, bashdb, bashdb-fns.inc, bashdb-init.inc: Get dual execution ++ of "bashdb" and "bash --debugger" with respect to command-line ++ arguments. In bashdb however, $0 is bashdb not the program. ++ ++2002-08-23 07:56 rocky ++ ++ * bashdb.el: Remove looking for "--emacs" option in calling bashdb. ++ This is used in perldb but I don't know if we need it here. It ++ might get added later when we know what this is about and decide ++ it's necessary. ++ ++2002-08-23 00:08 rocky ++ ++ * bashdb, bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc, ++ bashdb-main.inc, bashdb.el: Changes so that we can support "bash ++ --debugger" ++ ++2002-08-22 23:29 rocky ++ ++ * CHANGES, TODO: Update to reflect current state of things. ++ ++2002-08-22 08:13 rocky ++ ++ * bashdb: _cleanup -> _bashdb_cleanup ++ ++2002-08-22 08:03 rocky ++ ++ * bashdb: Bug: not handling -T correctly. Also add emacs mode. ++ ++2002-08-22 07:58 rocky ++ ++ * Makefile.am, bashdb-cmds.inc, bashdb-fns.inc, bashdb-init.inc, ++ bashdb-main.inc: Split bashdb-fns into multiple *.inc files ++ ++2002-08-22 07:56 rocky ++ ++ * .cvsignore: Yet more of the usual ignored files... ++ ++2002-08-22 07:54 rocky ++ ++ * .cvsignore: CVS: Useful file. ++ ++2002-08-21 19:53 rocky ++ ++ * CHANGES, bashdb, bashdb-main.inc: Now source the file to be ++ debugged rather than use that hacky temp file. x ++ ++2002-08-20 12:10 rocky ++ ++ * bashdb, bashdb-main.inc: Bogosity in assuming single-file script ++ removed. Now tracks source file correctly (in conjunction with ++ changes in bash) ++ ++2002-08-20 12:08 rocky ++ ++ * bashdb.el: [no log message] ++ ++2002-08-20 12:07 rocky ++ ++ * bashdb.el: Add GPL ++ ++2002-08-20 12:05 rocky ++ ++ * README: Add CVS id line ++ ++2002-08-20 12:05 rocky ++ ++ * README: No longer a modification but a rewrite. Note that bashdb ++ now requires my patches. ++ ++2002-08-20 12:04 rocky ++ ++ * AUTHORS, CHANGES, COPYING, ChangeLog, INSTALL, Makefile.am, NEWS, ++ TODO, configure.in: Automake'd and autoconf'd ++ ++2002-08-20 11:34 rocky ++ ++ * bashdb-main.inc: Add stack trace. Some typos corrected. ++ ++2002-08-20 11:31 rocky ++ ++ * bashdb-main.inc: Add GPL ++ ++2000-04-03 08:54 rocky ++ ++ * bashdb-main.inc: Typo. Cut over to version control id. ++ ++2000-04-02 23:26 rocky ++ ++ * MANIFEST, PERMISSION, README, bashdb, bashdb-main.inc, bashdb.el: ++ Initial revision ++ ++2000-04-02 23:26 rocky ++ ++ * MANIFEST, PERMISSION, README, bashdb, bashdb-main.inc, bashdb.el: ++ Imported sources ++ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/INSTALL cvs/debugger/INSTALL +--- bash-2.05b/debugger/INSTALL 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/INSTALL 2002-09-17 13:56:07.000000000 +0200 +@@ -0,0 +1,43 @@ ++See also INSTALL in the top-level directory, that is the parent of ++this directory for general instructions on installing the ++debug-enabled version of bash. The instructions here cover only the ++debugger portion. ++ ++First of all, the debugger needs a debugger-enabled version of Bash ++which contains things like the ability to step into functions ++automatically, give tracebacks, and list source-file information. The ++source code to the debugger-enabled version of bash is found in the ++parent directory of the directory that this file is located. (And if ++you have configured and compiled, the binary will be there too.) ++ ++It is possible to try out the debugger without installing it by using ++the bashdb script that is in this directory. To do so you would invoke ++your script as follows assuming you are currently in the directory ++(debugger) that you originally found this file in. ++ ++ ../bash -L . ./bashdb *script-to-be-debugged* *options-to-debugged-program* ++ ++The only downside to this approach is that $0 in will be ``bashdb'' ++(or more likely ``./bashdb'') rather than the name of the script to be ++debugged. If this is a problem, then you will have to install the ++debugger, or modify the script to be debugged to point to the ++debugger-enabled version of bash. For example if your script were in ++this directory (debugger) as well is your current working directory ++(as shown by ``pwd''), then having this at the beginning of your ++script: ++ ++#!../bash --debugger ++ ++might also work. ++ ++For information on the differences between "bash --debugger" and ++bashdb, see Chapter 2 (Getting In and Out) of the bashdb documentation ++(bashdb.info, bashdb.html, or bashdb.texi) ++ ++Before installing please do run "make check" to ensure that all of the ++debugger regression tests work. ++ ++After compiling bash (``make'' in the parent directory of this one) and ++running the regression tests, as root run ``make install.'' That's it! ++ ++$Id: INSTALL,v 1.2 2002/09/17 11:56:07 rockyb Exp $ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/Makefile.am cvs/debugger/Makefile.am +--- bash-2.05b/debugger/Makefile.am 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/Makefile.am 2003-09-01 22:00:46.000000000 +0200 +@@ -0,0 +1,48 @@ ++# Copyright (C) 2002 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++#$Id: Makefile.am,v 1.35 2003/08/29 15:29:41 snikkt Exp $ ++ ++SUBDIRS = test doc emacs ++ ++dist-hook: bash ++ ++bash: $(top_builddir)/../bash$(EXTEXT) ++ cp $(top_builddir)../bash$(EXTEXT) bash$(EXTEXT) ++ ++check-recursive: bash ++ ++man: ++ @( cd $(top_builddir)/doc ; $(MAKE) $(MFLAGS) man ) ++html: ++ @( cd $(top_builddir)/doc ; $(MAKE) $(MFLAGS) html ) ++ ++# Set up the install target ++bin_SCRIPTS = bashdb ++ ++data_DATA = ++pkgdata_DATA = dbg-brk.inc dbg-help.inc dbg-cmds.inc \ ++ dbg-fns.inc dbg-init.inc dbg-io.inc \ ++ dbg-list.inc dbg-main.inc dbg-set.inc \ ++ dbg-sig.inc dbg-stack.inc dbg-set-d-vars.inc \ ++ dbg-hist.inc dbg-pre.inc \ ++ dbg-file.inc ++ ++EXTRA_DIST = dbg-pre.inc.in bashdb.in dbg-main.inc.in \ ++ $(pkgdata_DATA) acinclude.m4 bash THANKS ++ ++#;;; Local Variables: *** ++#;;; mode:makefile *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/NEWS cvs/debugger/NEWS +--- bash-2.05b/debugger/NEWS 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/NEWS 2003-09-08 08:40:24.000000000 +0200 +@@ -0,0 +1,411 @@ ++Version 0.43 ++- Autostuff improvements ++- configurable package and executable names ++- Allow modification the philosophical parts the bashdb reference ++ manual to placate Matthias Klose and presumably Debian folks ++- Bug: a comment line in the history was taken as a timestamp and we ++ dumped core. Non-timestamp comment lines should appear in the history now. ++ ++ ++Version 0.42 ++- Add bashdb manual page. ++- Changes to allow building outside of source tree. ++- Add Debian bash 2.05b patches provided by Matthias Klose. ++ ++Version 0.41 ++ ++- The line number in multi-line assignments had been the line number where ++ the assignment ends, not the beginning. Thus assignments with ++ back-ticked subshells gave the wrong line number. ++- V (info variables) command working thanks to Mikael Andersson ++- Add "help show " and "help info " ++- More global variables start out _Dbg_ now (e.g. source_ -> _Dbg_source_) ++- Add long overdue THANKS ++ ++Version 0.40 ++ ++- "make distcheck" from debugger directory works ++- bashdb doc moved under debugger directory ++- Make portability changes for cygwin and OS's that don't have wcwidth ++- Install public bash 2.05b patches 005-007 ++- use diff -u if that's available on regression tests ++ ++Version 0.39 ++ ++- history now is timestamped. Use environment variable HISTTIMEFORMAT ++ to customize using strftime specifiers. ++- tty detection made less more generic +++ The timestamp patch is was posted to bug-readline@gnu.org; there is no ++ response. ++ ++ ++Version 0.38 ++- Was not finding source file if we did a cd in the debugged script ++ and the source file name was relative. Expand source file names. ++ ++Version 0.37 (since 0.36): ++- As a result of the below ddd's break/clear icon/buttons should work: ++ * Add "set annotate". ++ * Make breakpoint output more like gdb's. make show output more like gdb's ++ * Canonicalize filenames on breakpoints, actions, lists. ++ * add gdb "show dir" and do lookup based on source name and $cdir:$cwd ++- _bashdb_ -> _Dbg_ ++- Add ~ to filename translation characters ++ ++Version 0.36 (since 0.35): ++- Bug: if IFS set character in filename: the filename would be truncated. ++ Handle IFS save/restore more pervasively by saving on DEBUG handler ++ entry. ++ ++- Bug/Feature fix: Restart now quits all subshells before issuing its ++ restart (exec) command. ++ ++- Mechanism for having variables set in a subshell persist in a parent ++ shell worked out. Some of the global debugger settings (like ++ listsize, history) use this. ++ ++- debugger "print" command now uses double quote expansion rules. ++ (So print *** will expand filenames). We were having evaluation problems ++ in printing strings like "This won't work" (the single quote). ++ ++- Bug: setting shell variables with double/single quotes other ++ meta characters now works. ++ ++- Bug: caller() returned mismatched functions for line numbers/filenames. ++ Also now returns null string if no caller. ++ ++- Add + to list of filename translation characters. ++ +++ Patches for this version is posted to bash maintainer; word March 31st ++ 2003, that bash debugger patches have been integrated in a modified ++ form into "mainline bash code". ++ ++Version 0.35 (since 0.34): ++- Bug: if IFS was not blank setting _bashdb_old_set_opts might fail. ++ Save and restore IFS before using "set" ++ ++- Bug: (parse.y) line number was not updated properly when in a here ++ documented with an unquoted delimiter and containing backslash newline. ++ ++- Bug, sort of: now set BASH_COMMAND to for head (for i in x y z ...) ++ in case head (case x in ), and in select head (select s in). ++ ++- Feature: "restart" now saves debugger environment. Environment variable ++ BASHDB_RESTART give the state file to read. ++ ++- Add environment variable BASHDB_INPUT to have a debugger command ++ file get read. This can be a space-separated list of files. Allow ++ -x many times in bashdb. ++ ++- Patches: Install public bash 2.05b patches 001-004. ++ ++- When reading large fileswhich takes a long time (e.g. "configure" ++ scripts), print out status every 1000 lines. ++ ++Version 0.34 (since 0.33): ++- Bug: if no script we were running "set" to dump environment. ++ ++- Bug: "H count" was broken due to typo. ++ ++- We now keep track of the subshell level: dynamic variable ++ BASH_SUBSHELL and the debugger prints these as ()'s inside of the ++ prompt. ++ ++- "quit" leaves nested subshells more properly. Now takes an optional ++ argument to specify how many subshells to leave. ++ ++- Update gud.el (Masatake YAMATO), remove examples/bashdb/bashdb.el which may ++ cause confusion and set expectations for examples/bashdb (via README) ++ ++Version 0.33 (since 0.32): ++- Regression tests work on FreeBSD, OSX (as well as Linux and Solaris) ++ ++- Fix following bugs: ++ ++ * exit handler was getting called each time we exited a subshell. ++ ++ * "bash --debugger script args" was not setting args properly. ++ ++ * (configure) remove --with-debugger-start which are for ++ packages. Is variable DEBUGGER_START_FILE now. Versions are more ++ automatically picked up from a central setting. (More could be done ++ though). Thanks to c_thomps@ecolinux.no-ip.com. ++ ++ * not setting fntrace correctly on debugger exit. ++ ++ * wrong line number on case selector statement. Was esac rather than case. ++ ++ * was getting max line number for some files properly ++ ++ * (makefile) man2html.o not removed on clean ++ ++- Be able to handle EXIT like other signals (stop/nostop stack/nostack). ++ ++- Add provision for finding out which command is going to be executed. ++ Useful on multi-statement lines, conditional tests, for-loop headers a ++ eval (since the statement might be first "eval $foo" and then ++ "echo this is my life." ++ ++- Display command in debugger session we are on the same line/file and this ++ command changed. ++ ++- Canonicalize filevars via a file-expand routine. Allows many ways to ++ refer to the same file. ++ ++- Add option to have files show only short (basename) names, useful in testing ++ ++- Add set/show of these (basename, debugging debugger). ++ ++- Add -c option on bashdb. Analogous to "bash -c string" ++ ++- correct releases automatically put in "bash -v" and debugger M commands. ++ ++- Clean up signal handling code some. No longer need to pass LINENO. ++ Add routine to do some of the common entry/exit things. ++ ++Version 0.32 (since 0.31): ++ ++- Line number reporting improved. ++ ++- Fix bug in using trap line numbers inside traps. BASH_LINENO had ++ been wrong for the parts inside of a trap. ++ ++- Fix bug in case line numbers reporting line number of "esac" rather than case clause ++ ++- Fix bug in "down" command ++ ++- Bug in regression test run-all not picking up right shell sometimes. ++ ++- Make command "x" (display variable) much more useful and smarter ++ and more like perl5db's "x" command: show variable attributes (via ++ declare -p) or function body (via declare -f) evaluates an expression if ++ those fail and if *that* fails do what good ol' "print" would do. ++ ++- Add gdb's: ++ "condition" command. ++ "info break ". ++ "handle" command ++ "info signal" ++ ++- Signal handling more like gdb. ++ ++- bashdb patches for ddd ++ ++- Change help be more like gdb to facilitate ddd support. ++ ++- Improve bashdb.el and patches for gud.el ++ ++- Missing "finish" help info. ++ ++ ++Version 0.31 (since 0.30): ++- Fix bug where stack trace current line entry was sometimes wrong. ++ No longer mangle line number in stack changing commands. ++ ++- Implement gdb "finish" (some line number weirdnesses though.) ++ ++- Don't clobber debugged script's INT or ERR trap signal if set, but ++ do change them inside the debugger. ++ ++- All debugger functions now start _bashdb ++ ++- Expand history commands: "H -n" and "! n" work like perl5db. ++ ++- short command v (versions) changed to M to match perl5db. ++ ++Version 0.30 (since 0.29): ++- Fix bugs in debugger output when debugged program redirects I/O ++ ++- Fix bug in setting $# via the way we set $1, $2 and so on. We now ++ also handle an arbitrary number of parameters. ++ ++- Fix/change history. Numbers now match prompt numbers. Can use !n, !-n as ++ well as hi [-]n. Also ![-]n:p instead of H n. Can limit the number ++ of history items listed on the H command. ++ ++- Add GDB's "set args" command. ++ ++- Add GDB's "set editing" command. ++ ++- Add GDB's set/show prompt and allow for more flexible prompt customization. ++ ++- Add GDB-like tty command and bashdb option -t for this. ++ Unlike GDB though, you can redirect debugger output to a file. ++ ++- Minor tweaks to "help" and "show". "show subtopic" works. ++ ++Version 0.29 (since 0.28): ++ ++- Fix bug in "make uninstall" ++ ++- $? is saved and preserved for use inside "eval", "print", "watch", ++ "display", "break" and "action" expressions. ++ ++- Can also refer to $1, $2, ... in commands mentioned above. ++ ++- debug-level nesting (the number of times we are nested in debugger) ++ is now shown in prompt via new exported variable BASHDB_LEVEL. ++ ++- Implement "skip" command. (Don't run next command.) Integrate this into ++ "debug" command. ++ ++- First cut at gdb-type "return" (premature return from function or ++ sourced program). ++ ++- Document caller() builtin function and "debug" debugger command. ++ ++- "Search" now works like gdb and perl5db. Fix a small bug in search ++ forward: was not resetting line to search location. ++ ++Version 0.28 (since 0.27): ++ ++- Can refer to parameters $1 $2, ... in "print", "eval" and breakpoint ++ conditions. ++ ++- Parameters appear in call stack. To do this: ++ ++- BASH_ARGC and BASH_ARGV arrays added. One is a stack of the count of ++ the parameters and the other a stack of the parameters. ++ ++Version 0.27 (since 0.26): ++- Add forward and backward search. ++ ++- Add Perl "a" and "A" (action) commands. ++ ++- Add builtin caller(). ++ ++Version 0.26 (since 0.25): ++- Document "break" command. ++ ++- Change "line" to "linespec" in online help where ++ appropriate. ++ ++- Add commands "tbreak" (one-time break) and "display" (automatic display). ++ ++- Add provision to debug into a script ("debug" command), and leave ++ script without going back into debugger command loop ++ (BASHDB_QUIT_ON_QUIT environment variable). ++ ++- Document command "tbreak", "enable" and "disable" in manual. ++ ++Version 0.25 (since 0.24): ++ ++- Fix core dump on error (when no source file). ++ ++- Fix bug where "watche x > 24" was creating output to file 24. ++ ++- "restart" now picks up full bash name from BASH environment variable. ++ ++- bash --debugger -c "..." now works. ++ ++ To accomplish the above, environment variable BASH_EXECUTION_STRING ++ was added which contains the contents what was passed after -c. ++ ++- Command files now nest and pick up where they left off in between ++ script execution. ++ ++- save "set" parameters on debugger entry and restore them. Inside the ++ debugger: set +xv. ++ ++- add ARG array for $1, $2 and so on. "info args" now shows these. ++ ++- Revise and update doc for above and (hopefully) better overall ++ organization. ++ ++- Better installation instructions. ++ Revise bash's top-level README and INSTALL's (debugger and top ++ level) to note debugger changes. ++ ++Version 0.24 (since 0.23): ++ ++- Line numbers of initial "for var in..." and "select" listed or stopped. ++ Error messages on those lines should print correctly. ++ ++- Rename debugger variables so as not to conflict with possible user ++ variables (such as "i"). ++ ++- Don't allow stepping into debugger on program exit. ++ loop on command loop in "exit" handler. ++ ++- Function names now supported. breakpoints can be set on them. "list" now ++ allows a function name. ++ ++ To add function name support, the semantics of "declare -F" have ++ changed: the source file and line number is now added in the output. ++ ++ However in non-interactive mode, the line number is wrong. It is the ++ caller line. ++ ++- These commands have been extended to include linespecs ++ "continue", "break" (tbreak), "list". ++ ++ ++Version 0.23 (since 0.22): ++ ++- More work on breakpoints and watchpoints, error checking ++ * Allow breaking on file:line. ++ * Give counts on number of times hit ++ * Allow enabling/disabling/clearing watchpoints via "w" suffix ++- fix bugs ++ handling break/watch conditions ++ eval without a tty ++ "-" command size wrong ++ #comment handling ++ make cmd_loop read safe from tampering with IFS ++- More texinfo manual revisions ++- Separate debugger tests from bash tests ++ ++Version 0.22 (since initial release): ++ ++- More complete texinfo manual ++- Automake improvements to install Emacs lisp (Masatake YAMATO) ++- Set/show started ++- List with no arguments works like gdb and Perl's debugger ++ list regression test added ++- Add frame command ++- Add signal handler for exit so we don't exit debugger when exiting program ++- case statements now indicate line number of "case WORD". ++ ++Initial release - ++ ++Initial release of something that's worthy of being called a debugger! ++ ++In contrast to earlier ksh/bash debuggers, this release has: ++ ++Line numbers and source text are reported correctly ++ - through loops ++ - through conditionals ++ - inside functions ++ - inside sourced files ++ ++A change in source file is now tracked - we don't assume one script ++file nor do we create a fake script file with a funny name and run ++that. ++ ++Emacs support via GUD (grand-unified debugger) ++ ++Step/next (to step inside a function or ignore it) now work ++ - even works on "source" ++ ++Call stack display. ++ - Since "source" is a builtin function that shows up in the call stack ++ - function names, filenames and line numbers appear in call stack ++ ++Command arguments to debugged routine are left intact (if using bash ++--debugger). In particular $0 is the program name. ++ ++Debugging interface similar (and largely a subset of) Perl's debugger ++interface with gdb long mnemonics. (Well, okay: where - but perhaps ++more in the future like "up" and "down"). ++ ++Even without using *our* debugging script, there are changes to bash ++that make other debugging (or writing other debuggers) more ++amenable. In particular, line numbers of command substitution ``, ++$(), and {}, now gives the absolute line number in the file rather ++than relative to the beginning of the substitution. (So error ++reporting you generally see that an unhelpful message that error ++occurred on line 1.) Also support for call stacks with source files ++and line numbers can be helpful for stand-alone debugging and error ++reporting ++ ++$Id: NEWS,v 1.86 2003/09/03 07:54:02 rockyb Exp $ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/README cvs/debugger/README +--- bash-2.05b/debugger/README 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/README 2002-08-20 18:05:49.000000000 +0200 +@@ -0,0 +1,16 @@ ++This is a rewrite of the Korn Shell debugger from Bill Rosenblatt's ++`Learning the Korn Shell', published by O'Reilly and Associates (ISBN ++1-56592-054-6). Michael Loukides and Cigy Cyriac made some additional ++changes. ++ ++The original `kshdb' is available for anonymous FTP with the URL ++ ++ftp://ftp.uu.net/published/oreilly/nutshell/ksh/ksh.tar.Z ++ ++However this code now depends on a number of debugging support ++features that are neither part of the POSIX standard and are probably ++not in many POSIX-like shells. ++ ++R. Bernstein (rocky@panix.com) ++ ++$Id: README,v 1.3 2002/08/20 16:05:49 rocky Exp $ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/THANKS cvs/debugger/THANKS +--- bash-2.05b/debugger/THANKS 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/THANKS 2003-08-11 10:06:34.000000000 +0200 +@@ -0,0 +1,6 @@ ++The following kind people have contributed to this debugger. ++ ++Masatake YAMATO - Emacs support, screenshots ++Mikael Andersson - "V" command ++Matthias Klose - making build system and package ++ more industrial strength. +\ No newline at end of file +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/TODO cvs/debugger/TODO +--- bash-2.05b/debugger/TODO 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/TODO 2003-08-11 10:06:34.000000000 +0200 +@@ -0,0 +1,63 @@ ++For Debian release: ++- bashdb manual page ++- GFDL or GPL for documentation? ++- install lib in /usr/share. Have bashdb fallback to ++ "@DEBUGGER_START_FILE@"? if no -L and no ../lib? ++ ++BUGS ++ ++- source can have parameters but that doesn't appear in ARGC, ARGV. ++ RETURN trap should signal when leaving source'd file. ++ ++- Stack trace on rare occasions can get the wrong line number. ++ ++- Interrupts to program sometimes are delayed. Seems like a bash thing. Why? ++ ++- Darwin compilation problem in siglist ++ ++- Check that all global variables are changed via "journal" ++ ++------------------------------------- ++ ++NECESSARY TO DO: ++ ++ - Documentation: ++ update sample session ? ++ Info for calling from inside script. Developer's handbook? ++ ++FEATURES TO ADD: ++ ++ 0. bash should maintain a list of line numbers that one can ++ set breaks on. ++ ++ 1. perl5db actions ++ ++ *2. debugger to catch error handling. (Does this need bash?) ++ Might be able to somehow cascade onto existing Error routine. ++ Same might be done with SIGDEBUG. ++ ++ 3. gdb signal mechanism - it's partly there ++ ++ 4. "finish" or "return" where frame has been changed from top. ++ ++ Gdb commands that could be added: ++ backtrace -n ++ set history filename ++ set history name ++ set history size ++ ignore ++ ++ ++FUTURE: ++ ++ *More support in bash: list of valid line numbers ++ *Rewrite so debugger lives outside of process ++ - will not be subject to subshell environment discards. ++ - will have its own global state, but it needs to have access to debugger ++ environment ++ - will support remote debugging ++ ++ ++*=Things that bash might help out with. ++------------------------------------- ++$Id: TODO,v 1.49 2003/08/10 23:58:40 rockyb Exp $ +\ No newline at end of file +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/acinclude.m4 cvs/debugger/acinclude.m4 +--- bash-2.05b/debugger/acinclude.m4 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/acinclude.m4 2003-08-28 04:47:10.000000000 +0200 +@@ -0,0 +1,30 @@ ++AC_DEFUN([AC_BASHDB_PACKAGE], [ ++dnl Allow choosing the package name to avoid clashes with ++dnl bash if beeing installed side-by-side ++AC_ARG_VAR( ++ ALT_PACKAGE_NAME, ++ AC_HELP_STRING([],[alternate packagename to use (default is "$1")]) ++) ++if test -z "${ALT_PACKAGE_NAME}"; then ++ ALT_PACKAGE_NAME="$PACKAGE_NAME" ++fi ++ ++dnl define PACKAGE and VERSION. ++PACKAGE=$ALT_PACKAGE_NAME ++VERSION=$PACKAGE_VERSION ++AC_DEFINE_UNQUOTED(PACKAGE,$PACKAGE) ++AC_DEFINE_UNQUOTED(VERSION,$VERSION) ++AC_SUBST(PACKAGE) ++AC_SUBST(VERSION) ++]) ++ ++ ++AC_DEFUN([AC_SUBST_DIR], [ ++ ifelse($2,,,$1="$2") ++ $1=`( ++ test "x$prefix" = xNONE && prefix="$ac_default_prefix" ++ test "x$exec_prefix" = xNONE && exec_prefix="${prefix}" ++ eval echo \""[$]$1"\" ++ )` ++ AC_SUBST($1) ++]) +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/bashdb.in cvs/debugger/bashdb.in +--- bash-2.05b/debugger/bashdb.in 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/bashdb.in 2003-08-30 07:31:04.000000000 +0200 +@@ -0,0 +1,165 @@ ++#!@INTERPRETER_NAME@ ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++# ++# Alternate way to invoke debugger. bash --debugger however is preferred. ++ ++typeset _Dbg_ver=\ ++'$Id: bashdb.in,v 1.4 2003/08/25 16:34:35 snikkt Exp $' ++ ++declare -a _Dbg_script_args="$@" ++ ++# Equivalent to basename $0; the short program name ++typeset _Dbg_pname=${0##*/} ++ ++# Show basename only in location listing. This is needed in regression tests ++typeset -i _Dbg_basename_only=${BASHDB_BASENAME_ONLY:-0} ++ ++typeset _Dbg_main=dbg-main.inc ++typeset _Dbg_libdir=@PKGDATADIR@ ++typeset _Dbg_bindir=$(dirname $0) ++typeset _Dbg_tmpdir=/tmp ++ ++typeset _Dbg_cmd='' # If command string given on command line, this is it. ++ ++ ++_Dbg_usage() { ++ printf "_Dbg_usage: ++ ${_Dbg_pname} [OPTIONS] ++ ++Runs script_file under a (primitive) debugger. ++ ++options: ++ -B basename only on source listings. (Needed in regression tests) ++ -h print this help ++ -n Don't run initialization files ++ -c command Run this passed command of a script ++ -q Quiet. Do not print introductory and quiet messages. ++ -x cmdfile execute commands from cmdfile ++ -L libdir set directory location of library helper file: $_Dbg_main ++ the default directory is: $_Dbg_libdir ++ -T tmpdir set directory location for temporary files: $_Dbg_tmpdir ++ -t tty set debuger terminal ++ -V show version number and no-warranty and exit. ++" 1>&2 ++} ++ ++# What to set for location of helper routines? ++if [[ ! -e $_Dbg_libdir/$_Dbg_main ]] ; then ++ # Use bindir/../share as fallback ++ _Dbg_libdir= ++ if [[ -d $_Dbg_bindir/../share/bashdb ]] ; then ++ _Dbg_libdir=$_Dbg_bindir/../share/bashdb ++ fi ++fi ++ ++while getopts BhnqVc:t:x:L:T: opt; do ++ case $opt in ++ B) _Dbg_basename_only=1 ;; ++ c) _Dbg_cmd="$OPTARG" ;; ++ h) _Dbg_usage; exit 100 ;; ++ n) _Dbg_no_init=1 ;; ++ q) _Dbg_quiet=1 ;; ++ x) BASHDB_INPUT="$BASHDB_INPUT $OPTARG" ;; ++ L) _Dbg_libdir=$OPTARG ;; ++ T) _Dbg_tmpdir=$OPTARG ;; ++ t) ++ if ! $(touch $OPTARG >/dev/null 2>/dev/null); then ++ echo "${_Dbg_pname}: Can't access $OPTARG for writing." ++ elif [[ ! -w $OPTARG ]] ; then ++ echo "${_Dbg_pname}: terminal $OPTARG needs to be writable." ++ else ++ _Dbg_tty=$OPTARG ++ fi ++ ;; ++ V) show_version=1 ;; ++ *) _Dbg_usage; exit 2 ;; ++ esac ++done ++shift $(($OPTIND - 1)) ++ ++[[ $# == 0 && -z $show_version && -z $_Dbg_cmd ]] && { ++ echo "${_Dbg_pname}: Need to give a script name to debug." ++ exit 1 ++} ++ ++if [[ ! -d $_Dbg_libdir ]] && [[ ! -d $_Dbg_libdir ]] ; then ++ echo "${_Dbg_pname}: cannot read $_Dbg_libdir. " \ ++ "Perhaps bashdb is installed wrong." >&2 ++ echo "${_Dbg_pname}: or try using -L (with a different directory)." >&2 ++ exit 1 ++fi ++ ++_source_file=$1 ++shift ++ ++if [[ ! -d $_Dbg_tmpdir ]] && [[ ! -w $_Dbg_tmpdir ]] ; then ++ echo "${_Dbg_pname}: cannot write to temp directory $_Dbg_tmpdir." >&2 ++ echo "${_Dbg_pname}: Use -T try directory location." >&2 ++ exit 1 ++fi ++ ++[[ -r $_Dbg_libdir/$_Dbg_main ]] || { ++ echo "${_Dbg_pname}: cannot read debugger file $_Dbg_libdir/$_Dbg_main." >&2 ++ echo "${_Dbg_pname}: Perhaps bashdb is installed incorrectly." >&2 ++ exit 1 ++} ++ ++# Note that this is called via bashdb rather than "bash --debugger" ++_Dbg_script=1 ++ ++. ${_Dbg_libdir}/dbg-pre.inc ++ ++if [[ -z $_Dbg_quiet ]] ; then ++ echo "Bourne-Again Shell Debugger, release $_Dbg_release" ++ cat <&2 ++ exit 1 ++else ++ typeset -r _Dbg_source_file=$(_Dbg_expand_filename $_source_file) ++fi ++ ++. $_Dbg_libdir/dbg-main.inc ++set -o fntrace ++if [[ -z $_Dbg_cmd ]] ; then ++ . $_source_file ++else ++ eval $_Dbg_cmd ++fi ++ ++ ++# end of bashdb ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** ++ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/configure.ac cvs/debugger/configure.ac +--- bash-2.05b/debugger/configure.ac 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/configure.ac 2003-09-01 22:00:47.000000000 +0200 +@@ -0,0 +1,130 @@ ++dnl ++dnl Configure script for bash-2.05 debugger ++dnl ++dnl report bugs to rocky@panix.com ++dnl ++dnl Process this file with autoconf to produce a configure script. ++ ++# Copyright (C) 2002,2003 Rocky Bernstein ++ ++# This program is free software; you can redistribute it and/or modify ++# it under the terms of the GNU General Public License as published by ++# the Free Software Foundation; either version 2, or (at your option) ++# any later version. ++ ++# This program is distributed in the hope that it will be useful, ++# but WITHOUT ANY WARRANTY; without even the implied warranty of ++# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++# GNU General Public License for more details. ++ ++# You should have received a copy of the GNU General Public License ++# along with this program; if not, write to the Free Software ++# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA ++# 02111-1307, USA. ++ ++dnl TENTATIVE: ++dnl Use same version as in main bash configure.in ++define(bashvers, 2.05b) ++define(relstatus, 0.43) ++ ++AC_INIT([bash],[bashvers-debugger-relstatus],[rocky@panix.com]) ++ ++AM_PATH_LISPDIR ++ ++dnl make sure we are using a recent autoconf version ++AC_PREREQ(2.53) ++ ++AC_ARG_PROGRAM ++AM_INIT_AUTOMAKE([no-define]) ++ ++AC_BASHDB_PACKAGE([bashdb]) ++ ++# Brought over from bash/configure.in to substitute BASHVERS ++# and RELSTATUS in dbg-pre.inc.in and version.texi ++BASHVERS=bashvers ++RELSTATUS=relstatus ++AC_SUBST(BASHVERS) ++AC_SUBST(RELSTATUS) ++ ++ ++AC_CONFIG_SRCDIR(bashdb.in) ++if test x$ac_srcdir = x ; then ++ ac_srcdir=. ++fi ++ ++AC_PATH_PROG(EMACS, emacs, no) ++if test x$EMACS = xno ; then ++ emacs_lisp=no ++fi ++ ++# Create a suitable transform ( without the $ -> $$ escaping added ++# because of $program_transform_name being used in a Makefile ++# This transform is needed because bashbd must be executed by the ++# bashdb-bash regardless if a program transform has taken place ++ac_transform=`echo "$program_transform_name" | sed 's/\\$\\$/\\$/g'` ++ ++# Fully expanded name of bash executable to be substituted into ++# bashdb.This allow us to move this package into any suitable location ++# by using --prefix as an option to configure. ++AC_SUBST_DIR(INTERPRETER_NAME,"${bindir}/"`echo bash | sed "$ac_transform"`) ++ ++# Fully expanded namn of pkgdatadir. This is where arch independent files ++# should be placed according to FHS. Used in bashdb.in and dbg-main.inc.in ++AC_SUBST_DIR(PKGDATADIR,"${datadir}/${PACKAGE}") ++ ++ ++ ++dnl We use a diff in regression testing ++AC_PATH_PROG(DIFF, diff, no) ++DIFF_OPTS= ++ ++if test "$DIFF" = no ; then ++ AC_PATH_PROG(DIFF, cmp, no) ++else ++ dnl Try for GNU diff options. ++ # MSDOG output uses \r\n rather than \n in tests ++ for diff_opt in -w --unified ; do ++ if $DIFF $diff_opt ./configure ./configure > /dev/null 2>&1; then ++ AC_MSG_RESULT([adding $diff_opt to diff in regression tests]) ++ DIFF_OPTS="$DIFF_OPTS $diff_opt" ++ fi ++ done ++fi ++AC_SUBST(DIFF) ++AC_SUBST(DIFF_OPTS) ++ ++AM_CONDITIONAL(INSTALL_EMACS_LISP, test "x$lispdir" != "x") ++ ++#Makefiles ++AC_CONFIG_FILES([Makefile \ ++ doc/Makefile \ ++ emacs/Makefile \ ++ test/Makefile]) ++ ++# Additional files needing substitution of values (not Makefiles). ++AC_CONFIG_FILES([bashdb \ ++ dbg-pre.inc \ ++ dbg-main.inc \ ++ test/check_common ],[chmod +x bashdb]) ++ ++ ++# We don't want TeX do be a requirement for compilation. ++if ! test -e $ac_srcdir/doc/version.texi ;then ++ AC_CONFIG_FILES([ doc/version.texi ]) ++fi ++ ++AC_CONFIG_COMMANDS([default],[[ ++# Keep old dates on these files to prevent rebuilding. ++touch -cr $ac_srcdir/configure.ac doc/bashdb.1 ++touch -cr $ac_srcdir/configure.ac doc/bashdb-man.html ++if test -e $ac_srcdir/doc/version.texi ;then ++ echo timestamp > $ac_srcdir/doc/stamp-vti ++ touch -cr $ac_srcdir/configure.ac $ac_srcdir/doc/version.texi \ ++ $ac_srcdir/doc/stamp-vti \ ++ $ac_srcdir/doc/bashdb.info* ++ ++fi ++]],[[]]) ++ ++AC_OUTPUT ++ +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-brk.inc cvs/debugger/dbg-brk.inc +--- bash-2.05b/debugger/dbg-brk.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-brk.inc 2003-03-31 17:09:19.000000000 +0200 +@@ -0,0 +1,1018 @@ ++# dbg-brk.inc - Bourne Again Shell Debugger Break/Watch/Action routines ++ ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++#================ VARIABLE INITIALIZATIONS ====================# ++ ++declare -r _Dbg_brk_ver=\ ++'$Id: dbg-brk.inc,v 1.11 2003/03/31 15:09:19 rockyb Exp $' ++ ++typeset -ar _Dbg_yn=("n" "y") ++typeset -ar _Dbg_keep=('keep' 'del') ++ ++# action data structures ++typeset -ai _Dbg_action_line=() # Line number of breakpoint ++typeset -a _Dbg_action_file=() # filename of breakpoint ++typeset -ai _Dbg_action_enable=() # 1/0 if enabled or not ++typeset -a _Dbg_action_stmt=() # Statement to eval when line is hit. ++typeset -i _Dbg_action_max=0 # Needed because we can't figure ++ # out what the max index is and arrays ++ # can be sparse ++ ++# Note: we loop over possibly sparse arrays with _brkpt_max by adding one ++# and testing for an entry. Could add yet another array to list only ++# used indices. Bash is kind of primitive. ++ ++# Breakpoint data structures ++typeset -ai _brkpt_line=() # Line number of breakpoint ++typeset -a _brkpt_file=() # filename of breakpoint ++typeset -a _brkpt_enable=() # 1/0 if enabled or not ++typeset -ai _brkpt_count=() # Number of times hit ++typeset -ai _brkpt_onetime=() # Is this a onetime break? ++typeset -a _brkpt_cond=() # Condition to eval true in order to stop. ++typeset -i _brkpt_max=0 # Needed because we can't figure out what ++ # the max index is and arrays can be sparse ++# Note: we loop over possibly sparse arrays with _brkpt_max by adding one ++# and testing for an entry. Could add yet another array to list only ++# used indices. Bash is kind of primitive. ++ ++# Watchpoint data structures ++typeset -a _Dbg_watch_exp=() # Watchpoint expressions ++typeset -a _Dbg_watch_val=() # values of watchpoint expressions ++typeset -ai _Dbg_watch_arith=() # 1 if arithmetic expression or not. ++typeset -ai _Dbg_watch_count=() # Number of times hit ++typeset -ai _Dbg_watch_enable=() # 1/0 if enabled or not ++typeset -i _Dbg_watch_max=0 # Needed because we can't figure out what ++ # the max index is and arrays can be sparse ++ ++typeset -r _watch_pat="${int_pat}[wW]" ++ ++# Display data structures ++typeset -a _Dbg_disp_exp=() # Watchpoint expressions ++typeset -ai _Dbg_disp_enable=() # 1/0 if enabled or not ++typeset -i _Dbg_disp_max=0 # Needed because we can't figure out what ++ # the max index is and arrays can be sparse ++ ++ ++#========================= FUNCTIONS ============================# ++ ++_Dbg_save_breakpoints() { ++ local file ++ for file in ${_Dbg_filenames[@]} ; do ++ local filevar="`_Dbg_file2var $file`" ++ declare -p _brkpt_$filevar >> $_Dbg_statefile 2>/dev/null ++ done ++ declare -p _brkpt_line >> $_Dbg_statefile ++ declare -p _brkpt_file >> $_Dbg_statefile ++ declare -p _brkpt_cond >> $_Dbg_statefile ++ declare -p _brkpt_count >> $_Dbg_statefile ++ declare -p _brkpt_enable >> $_Dbg_statefile ++ declare -p _brkpt_onetime >> $_Dbg_statefile ++ declare -p _brkpt_max >> $_Dbg_statefile ++ ++} ++ ++_Dbg_save_actions() { ++ for file in ${_Dbg_filenames[@]} ; do ++ local filevar="`_Dbg_file2var $file`" ++ declare -p _Dbg_action_$filevar >> $_Dbg_statefile 2>/dev/null ++ done ++ declare -p _Dbg_action_line >> $_Dbg_statefile ++ declare -p _Dbg_action_file >> $_Dbg_statefile ++ declare -p _Dbg_action_enable >> $_Dbg_statefile ++ declare -p _Dbg_action_stmt >> $_Dbg_statefile ++ declare -p _Dbg_action_max >> $_Dbg_statefile ++} ++ ++_Dbg_save_watchpoints() { ++ declare -p _Dbg_watch_exp >> $_Dbg_statefile ++ declare -p _Dbg_watch_val >> $_Dbg_statefile ++ declare -p _Dbg_watch_arith >> $_Dbg_statefile ++ declare -p _Dbg_watch_count >> $_Dbg_statefile ++ declare -p _Dbg_watch_enable >> $_Dbg_statefile ++ declare -p _Dbg_watch_max >> $_Dbg_statefile ++} ++ ++_Dbg_save_display() { ++ declare -p _Dbg_disp_exp >> $_Dbg_statefile ++ declare -p _Dbg_disp_enable >> $_Dbg_statefile ++ declare -p _Dbg_disp_max >> $_Dbg_statefile ++} ++ ++# Start out with general break/watchpoint functions first... ++ ++# Routine to a delete breakpoint/watchpoint by entry numbers. ++_Dbg_cmd_delete() { ++ local -r to_go=$@ ++ local -i i ++ local -i found=0 ++ ++ # set -xv ++ eval "$_seteglob" ++ for del in $to_go ; do ++ case $del in ++ $_watch_pat ) ++ _Dbg_delete_watch_entry ${del:0:${#del}-1} ++ ;; ++ $int_pat ) ++ _Dbg_delete_brkpt_entry $del ++ ((found += $?)) ++ ;; ++ * ) ++ _Dbg_msg "Invalid entry number skipped: $del" ++ esac ++ done ++ eval "$_resteglob" ++ [[ $found != 0 ]] && _Dbg_msg "Removed $found breakpoint(s)." ++ return $found ++ # set +xv ++} ++ ++# Enable/disable breakpoint or watchpoint by entry numbers. ++_Dbg_enable_disable() { ++ if [ -z "$1" ] ; then ++ _Dbg_msg "Expecting a list of breakpoint/watchpoint numbers. Got none." ++ return 1 ++ fi ++ local -i on=$1 ++ local en_dis=$2 ++ shift; shift ++ ++ if [[ $1 = 'display' ]] ; then ++ shift ++ local to_go="$@" ++ local i ++ eval "$_seteglob" ++ for i in $to_go ; do ++ case $i in ++ $int_pat ) ++ _Dbg_enable_disable_display $on $en_dis $i ++ ;; ++ * ) ++ _Dbg_msg "Invalid entry number skipped: $i" ++ esac ++ done ++ eval "$_resteglob" ++ return 0 ++ elif [[ $1 = 'action' ]] ; then ++ shift ++ local to_go="$@" ++ local i ++ eval "$_seteglob" ++ for i in $to_go ; do ++ case $i in ++ $int_pat ) ++ _Dbg_enable_disable_action $on $en_dis $i ++ ;; ++ * ) ++ _Dbg_msg "Invalid entry number skipped: $i" ++ esac ++ done ++ eval "$_resteglob" ++ return 0 ++ fi ++ ++ local to_go="$@" ++ local i ++ eval "$_seteglob" ++ for i in $to_go ; do ++ case $i in ++ $_watch_pat ) ++ _Dbg_enable_disable_watch $on $en_dis ${del:0:${#del}-1} ++ ;; ++ $int_pat ) ++ _Dbg_enable_disable_brkpt $on $en_dis $i ++ ;; ++ * ) ++ _Dbg_msg "Invalid entry number skipped: $i" ++ esac ++ done ++ eval "$_resteglob" ++ return 0 ++} ++ ++_Dbg_cmd_continue() { ++ # set -xv ++ local -r first_arg=$1 ++ ++ [[ -z "$1" ]] && return 0 ++ local -ar word=($(_Dbg_parse_linespec "$first_arg")) ++ if [[ ${#word[@]} == 0 ]] ; then ++ _Dbg_msg "Invalid linespec: $first_arg" ++ return 1 ++ fi ++ local filename=${word[1]} ++ local line_number=${word[0]} ++ local full_filename=`_Dbg_is_file $filename` ++ if [[ -n $full_filename ]] ; then ++ if (( $line_number == 0 )) ; then ++ _Dbg_msg "There is no line 0 to continue at." ++ else ++ _Dbg_check_line $line_number "$full_filename" ++ (( $? == 0 )) && \ ++ _Dbg_set_brkpt "$full_filename" "$line_number" 1 1 ++ return 0 ++ fi ++ else ++ _Dbg_msg "File $filename not found in read-in files." ++ _Dbg_msg "See 'info files' for a list of known files." ++ fi ++ return 1 ++} ++ ++# Enable breakpoint(s)/watchpoint(s) by entry number(s). ++_Dbg_cmd_enable() { ++ _Dbg_enable_disable 1 "enabled" $@ ++} ++ ++# Disable breakpoint(s)/watchpoint(s) by entry number(s). ++_Dbg_cmd_disable() { ++ _Dbg_enable_disable 0 "disabled" $@ ++} ++ ++_Dbg_print_brkpt_count() { ++ local -ir i=$1 ++ if (( _brkpt_count[$i] != 0 )) ; then ++ if (( _brkpt_count[$i] == 1 )) ; then ++ _Dbg_printf "\tbreakpoint already hit 1 time" ++ else ++ _Dbg_printf "\tbreakpoint already hit %d times" ${_brkpt_count[$i]} ++ fi ++ fi ++} ++ ++#======================== BREAKPOINTS ============================# ++ ++# Add breakpoint(s) at given line number of the current file. $1 is ++# the line number or _curline if omitted. $2 is a condition to test ++# for whether to stop. ++ ++_Dbg_cmd_break() { ++ ++ local -i is_temp=$1; ++ shift ++ ++ local n=${1:-$_curline} ++ shift ++ ++ if [[ "$n" == 'if' ]]; then ++ n=$_curline ++ else ++ [[ "$1" == 'if' ]] && shift ++ fi ++ local condition; ++ if [ -z "$1" ] ; then ++ condition=1 ++ else ++ condition="$*" ++ fi ++ ++ local -ar word=($(_Dbg_parse_linespec "$n")) ++ if [[ ${#word[@]} == 0 ]] ; then ++ _Dbg_msg "Invalid linespec $n" ++ return ++ fi ++ ++ local filename=${word[1]} ++ local line_number=${word[0]} ++ local full_filename=`_Dbg_is_file $filename` ++ if [[ -n $full_filename ]] ; then ++ if (( $line_number == 0 )) ; then ++ _Dbg_msg "There is no line 0 to break at." ++ else ++ _Dbg_check_line $line_number "$full_filename" ++ (( $? == 0 )) && \ ++ _Dbg_set_brkpt "$full_filename" "$line_number" $is_temp "$condition" ++ fi ++ else ++ filename=$(_Dbg_adjust_filename "$filename") ++ _Dbg_msg "File $filename not found in read-in files." ++ _Dbg_msg "See 'info files' for a list of known files." ++ fi ++} ++ ++# Set a condition for a given breakpoint $1 is a breakpoint number ++# $2 is a condition. If not given, set "unconditional" or 1. ++# returns 0 if success or 1 if fail. ++_Dbg_cmd_condition() { ++ # set -x ++ local -r n=$1 ++ shift ++ local condition="$@" ++ # set -xv ++ ++ if [[ -z $n ]]; then ++ _Dbg_msg "Argument required (breakpoint number)." ++ return 1 ++ fi ++ ++ eval "$_seteglob" ++ if [[ $n != $int_pat ]]; then ++ eval "$_resteglob" ++ _Dbg_msg "Bad breakpoint number: $n" ++ return 1 ++ fi ++ eval "$_resteglob" ++ ++ if [[ -z ${_brkpt_file[$n]} ]] ; then ++ _Dbg_msg "Breakpoint entry $n is not set. Condition not changed." ++ return 1 ++ fi ++ ++ if [[ -z $condition ]] ; then ++ condition=1 ++ _Dbg_msg "Breakpoint $n now unconditional." ++ fi ++ _brkpt_cond[$n]="$condition" ++ return 0 ++} ++ ++# delete brkpt(s) at given file:line numbers. If no file is given ++# use the current file. ++_Dbg_cmd_clear_brkpt() { ++ # set -x ++ local -r n=${1:-$_curline} ++ ++ local -ar word=($(_Dbg_parse_linespec "$n")) ++ if [[ ${#word[@]} == 0 ]] ; then ++ _Dbg_msg "Invalid linespec $n" ++ return ++ fi ++ ++ local filename=${word[1]} ++ local line_number=${word[0]} ++ local full_filename=`_Dbg_is_file $filename` ++ if [[ -n $full_filename ]] ; then ++ if (( $line_number == 0 )) ; then ++ _Dbg_msg "There is no line 0 to clear." ++ else ++ _Dbg_check_line $line_number "$full_filename" ++ if (( $? == 0 )) ; then ++ _Dbg_unset_brkpt "$full_filename" "$line_number" ++ local -r found=$? ++ if [[ $found != 0 ]] ; then ++ _Dbg_msg "Removed $found breakpoint(s)." ++ else ++ _Dbg_msg "Didn't find any breakpoints to remove at $n." ++ fi ++ fi ++ fi ++ else ++ filename=$(_Dbg_adjust_filename "$filename") ++ _Dbg_msg "File $filename not found in read-in files." ++ _Dbg_msg "See 'info files' for a list of known files." ++ fi ++} ++ ++# list breakpoints and break condition. ++# If $1 is given just list those associated for that line. ++_Dbg_cmd_list_brkpt() { ++ ++ eval "$_seteglob" ++ if [[ -n $1 ]] ; then ++ if [[ $1 != $int_pat ]]; then ++ _Dbg_msg "Bad breakpoint number $1." ++ elif [[ -z ${_brkpt_file[$1]} ]] ; then ++ _Dbg_msg "Breakpoint entry $1 is not set." ++ else ++ local -r -i i=$1 ++ local source_file=${_brkpt_file[$i]} ++ source_file=$(_Dbg_adjust_filename "$source_file") ++ _Dbg_msg "Num Type Disp Enb What" ++ _Dbg_printf "%-3d breakpoint %-4s %-3s %s:%s" $i \ ++ ${_Dbg_keep[${_brkpt_onetime[$i]}]} \ ++ ${_Dbg_yn[${_brkpt_enable[$i]}]} \ ++ $source_file ${_brkpt_line[$i]} ++ if [[ ${_brkpt_cond[$i]} != '1' ]] ; then ++ _Dbg_printf "\tstop only if %s" "${_brkpt_cond[$i]}" ++ fi ++ _Dbg_print_brkpt_count ${_brkpt_count[$i]} ++ fi ++ eval "$_resteglob" ++ return ++ fi ++ ++ if [ ${#_brkpt_line[@]} != 0 ]; then ++ local -i i ++ ++ _Dbg_msg "Num Type Disp Enb What" ++ for (( i=1; (( i <= _brkpt_max )) ; i++ )) ; do ++ local source_file=${_brkpt_file[$i]} ++ if [[ -n ${_brkpt_line[$i]} ]] ; then ++ source_file=$(_Dbg_adjust_filename "$source_file") ++ _Dbg_printf "%-3d breakpoint %-4s %-3s %s:%s" $i \ ++ ${_Dbg_keep[${_brkpt_onetime[$i]}]} \ ++ ${_Dbg_yn[${_brkpt_enable[$i]}]} \ ++ $source_file ${_brkpt_line[$i]} ++ if [[ ${_brkpt_cond[$i]} != '1' ]] ; then ++ _Dbg_printf "\tstop only if %s" "${_brkpt_cond[$i]}" ++ fi ++ if (( _brkpt_count[$i] != 0 )) ; then ++ _Dbg_print_brkpt_count ${_brkpt_count[$i]} ++ fi ++ fi ++ done ++ else ++ _Dbg_msg "No breakpoints have been set." ++ fi ++} ++ ++# clear all brkpts ++_Dbg_clear_all_brkpt() { ++ ++ local -i k ++ for (( k=0; (( k < ${#_Dbg_filenames[@]} )) ; k++ )) ; do ++ local filename=${_filename[$k]} ++ local filevar="`_Dbg_file2var $filename`" ++ local brkpt_a="_brkpt_${filevar}" ++ _Dbg_write_journal_eval "unset ${brkpt_a}[$k]" ++ done ++ _Dbg_write_journal_eval "_brkpt_line=()" ++ _Dbg_write_journal_eval "_brkpt_cond=()" ++ _Dbg_write_journal_eval "_brkpt_file=()" ++ _Dbg_write_journal_eval "_brkpt_enable=()" ++ _Dbg_write_journal_eval "_brkpt_count=()" ++ _Dbg_write_journal_eval "_brkpt_onetime=()" ++} ++ ++# Internal routine to a set breakpoint unconditonally. ++ ++_Dbg_set_brkpt() { ++ local source_file=$1 ++ local -ir line=$2 ++ local -ir is_temp=$3 ++ local -r condition=${4:-1} ++ local -r filevar="`_Dbg_file2var $source_file`" ++ ++ local val_str=`_Dbg_get_assoc_array_entry "_brkpt_$filevar" $line` ++ ++ # Increment brkpt_max here because we are 1-origin ++ ((_brkpt_max++)) ++ ++ if [ -z "$val_str" ] ; then ++ val_str=$_brkpt_max ++ else ++ val_str="$val_str $_brkpt_max" ++ fi ++ ++ _brkpt_line[$_brkpt_max]=$line ++ _brkpt_file[$_brkpt_max]="$source_file" ++ _brkpt_cond[$_brkpt_max]="$condition" ++ _brkpt_onetime[$_brkpt_max]=$is_temp ++ _brkpt_count[$_brkpt_max]=0 ++ _brkpt_enable[$_brkpt_max]=1 ++ ++ local dq_source_file=$(_Dbg_esc_dq "$source_file") ++ local dq_condition=$(_Dbg_esc_dq "$condition") ++ _Dbg_write_journal "_brkpt_line[$_brkpt_max]=$line" ++ _Dbg_write_journal "_brkpt_file[$_brkpt_max]=\"$dq_source_file\"" ++ _Dbg_write_journal "_brkpt_cond[$_brkpt_max]=\"$dq_condition\"" ++ _Dbg_write_journal "_brkpt_onetime[$_brkpt_max]=$is_temp" ++ _Dbg_write_journal "_brkpt_count[$_brkpt_max]=\"0\"" ++ _Dbg_write_journal "_brkpt_enable[$_brkpt_max]=1" ++ ++ _Dbg_set_assoc_array_entry "_brkpt_$filevar" $line $val_str ++ source_file=$(_Dbg_adjust_filename "$source_file") ++ if (( $is_temp == 0 )) ; then ++ _Dbg_msg "Breakpoint $_brkpt_max set in file ${source_file}, line $line." ++ else ++ _Dbg_msg "One-time breakpoint $_brkpt_max set in file ${source_file}, line $line." ++ fi ++ _Dbg_write_journal "_brkpt_max=$_brkpt_max" ++} ++ ++# Internal routine to unset the actual breakpoint arrays ++_Dbg_unset_brkpt_arrays() { ++ local -i del=$1 ++ _Dbg_write_journal_eval "unset _brkpt_line[$del]" ++ _Dbg_write_journal_eval "unset _brkpt_count[$del]" ++ _Dbg_write_journal_eval "unset _brkpt_file[$del]" ++ _Dbg_write_journal_eval "unset _brkpt_enable[$del]" ++ _Dbg_write_journal_eval "unset _brkpt_cond[$del]" ++ _Dbg_write_journal_eval "unset _brkpt_onetime[$del]" ++} ++ ++# Internal routine to delete a breakpoint by file/line. ++_Dbg_unset_brkpt() { ++ local -r filename=$1 ++ local -ir line=$2 ++ local -r filevar="`_Dbg_file2var $filename`" ++ local -r fullname="`_Dbg_expand_filename $filename`" ++ local -i found=0 ++ ++ # set -xv ++ local -r entries=`_Dbg_get_assoc_array_entry "_brkpt_$filevar" $line` ++ local -i del ++ for del in $entries ; do ++ if [[ -z ${_brkpt_file[$del]} ]] ; then ++ _Dbg_msg "No breakpoint found at $filename:$line" ++ continue ++ fi ++ local brkpt_fullname=$(_Dbg_expand_filename ${_brkpt_file[$del]}) ++ if [[ $brkpt_fullname != $fullname ]] ; then ++ _Dbg_msg "Brkpt inconsistency:" \ ++ "$filename[$line] lists ${_brkpt_file[$del]} at entry $del" ++ else ++ _Dbg_unset_brkpt_arrays $del ++ ((found++)) ++ fi ++ done ++ _Dbg_write_journal_eval "unset _brkpt_$filevar[$line]" ++ return $found ++ # set +xv ++} ++ ++# Routine to a delete breakpoint by entry number: $1. ++# Returns whether or not anything was deleted. ++_Dbg_delete_brkpt_entry() { ++ local -r del=$1 ++ local -i i ++ local -i found=0 ++ ++ # set -xv ++ if [[ -z ${_brkpt_file[$del]} ]] ; then ++ _Dbg_msg "Breakpoint entry $del is not set." ++ return 0 ++ fi ++ local filevar="`_Dbg_file2var ${_brkpt_file[$del]}`" ++ local line=${_brkpt_line[$del]} ++ local -r entries=`_Dbg_get_assoc_array_entry "_brkpt_$filevar" $line` ++ local try ++ local -a new_val=() ++ for try in $entries ; do ++ if (( $try == $del )) ; then ++ _Dbg_unset_brkpt_arrays $del ++ found=1 ++ else ++ if [[ -n ${_brkpt_file[$try]} ]] ; then ++ new_val[${#new_val[@]}]=$try ++ fi ++ fi ++ done ++ if [[ ${#new_val[@]} == 0 ]] ; then ++ _Dbg_write_journal_eval "unset _brkpt_$filevar[$line]" ++ else ++ _Dbg_set_assoc_array_entry "_brkpt_$filevar" $line "${new_val[@]}" ++ fi ++ ++ return $found ++ # set +xv ++} ++ ++# Enable/disable breakpoint(s) by entry numbers. ++_Dbg_enable_disable_brkpt() { ++ local -i on=$1 ++ local en_dis=$2 ++ local -i i=$3 ++ if [[ -n "${_brkpt_file[$i]}" ]] ; then ++ if [[ ${_brkpt_enable[$i]} == $on ]] ; then ++ _Dbg_msg "Breakpoint entry $i already $en_dis so nothing done." ++ else ++ _Dbg_write_journal_eval "_brkpt_enable[$i]=$on" ++ _Dbg_msg "Breakpoint entry $i $en_dis." ++ fi ++ else ++ _Dbg_msg "Breakpoint entry $i doesn't exist so nothing done." ++ fi ++} ++ ++#======================== WATCHPOINTS ============================# ++ ++_Dbg_get_watch_exp_eval() { ++ local -i i=$1 ++ local new_val ++ ++ if [[ `eval echo \"${_Dbg_watch_exp[$i]}\"` == "" ]]; then ++ new_val='' ++ elif (( ${_Dbg_watch_arith[$i]} == 1 )) ; then ++ . ${_Dbg_libdir}/dbg-set-d-vars.inc ++ eval let new_val=\"${_Dbg_watch_exp[$i]}\" ++ else ++ . ${_Dbg_libdir}/dbg-set-d-vars.inc ++ eval new_val="${_Dbg_watch_exp[$i]}" ++ fi ++ echo $new_val ++} ++ ++# Enable/disable watchpoint(s) by entry numbers. ++_Dbg_enable_disable_watch() { ++ local -i on=$1 ++ local en_dis=$2 ++ local -i i=$3 ++ if [ -n "${_Dbg_watch_exp[$i]}" ] ; then ++ if [[ ${_Dbg_watch_enable[$i]} == $on ]] ; then ++ _Dbg_msg "Watchpoint entry $i already $en_dis so nothing done." ++ else ++ _Dbg_write_journal_eval "_Dbg_watch_enable[$i]=$on" ++ _Dbg_msg "Watchpoint entry $i $en_dis." ++ fi ++ else ++ _Dbg_msg "Watchpoint entry $i doesn't exist so nothing done." ++ fi ++} ++ ++_Dbg_list_watch() { ++ if [ ${#_Dbg_watch_exp[@]} != 0 ]; then ++ local i=0 j ++ _Dbg_msg "Watch expressions:" ++ for (( i=0; (( i < _Dbg_watch_max )); i++ )) ; do ++ if [ -n "${_Dbg_watch_exp[$i]}" ] ;then ++ _Dbg_printf '%-3d watchpoint %-4s %s' $i \ ++ ${_Dbg_yn[${_Dbg_watch_enable[$i]}]} \ ++ "${_Dbg_watch_exp[$i]}" ++ _Dbg_print_brkpt_count ${_Dbg_watch_count[$i]} ++ fi ++ done ++ else ++ _Dbg_msg "No watch expressions have been set." ++ fi ++} ++ ++_Dbg_delete_watch_entry() { ++ local -i del=$1 ++ ++ if [ -n "${_Dbg_watch_exp[$del]}" ] ; then ++ _Dbg_write_journal_eval "unset _Dbg_watch_exp[$del]" ++ _Dbg_write_journal_eval "unset _Dbg_watch_val[$del]" ++ _Dbg_write_journal_eval "unset _Dbg_watch_enable[$del]" ++ _Dbg_write_journal_eval "unset _Dbg_watch_count[$del]" ++ else ++ _Dbg_msg "Watchpoint entry $del doesn't exist so nothing done." ++ fi ++} ++ ++_Dbg_clear_watch() { ++ if (( $# < 1 )) ; then ++ local _Dbg_prompt_output=${_Dbg_tty:-/dev/null} ++ read $_Dbg_edit -p "Delete all watchpoints? (y/n): " \ ++ <&$_Dbg_input_desc 2>>$_Dbg_prompt_output ++ ++ if [[ $REPLY = [Yy]* ]] ; then ++ _Dbg_write_journal_eval unset _Dbg_watch_exp[@] ++ _Dbg_write_journal_eval unset _Dbg_watch_val[@] ++ _Dbg_write_journal_eval unset _Dbg_watch_enable[@] ++ _Dbg_write_journal_eval unset _Dbg_watch_count[@] ++ _Dbg_msg "All Watchpoints have been cleared" ++ fi ++ return 0 ++ fi ++ ++ eval "$_seteglob" ++ if [[ $1 == $int_pat ]]; then ++ _Dbg_write_journal_eval "unset _Dbg_watch_exp[$1]" ++ _msg "Watchpoint $i has been cleared" ++ else ++ _Dbg_list_watch ++ _basdhb_msg "Please specify a numeric watchpoint number" ++ fi ++ ++ eval "$_resteglob" ++} ++ ++# Set or list watch command ++_Dbg_cmd_watch() { ++ if [ -z "$2" ]; then ++ _Dbg_clear_watch ++ else ++ local -i n=_Dbg_watch_max++ ++ _Dbg_watch_arith[$n]="$1" ++ shift ++ _Dbg_watch_exp[$n]="$1" ++ _Dbg_watch_val[$n]=`_Dbg_get_watch_exp_eval $n` ++ _Dbg_watch_enable[$n]=1 ++ _Dbg_watch_count[$n]=0 ++ _Dbg_printf '%2d: %s==%s arith: %d' $n \ ++ "(${_Dbg_watch_exp[$n]})" ${_Dbg_watch_val[$n]} \ ++ ${_Dbg_watch_arith[$n]} ++ fi ++} ++ ++#======================== ACTIONs ============================# ++ ++# Add actions(s) at given line number of the current file. $1 is ++# the line number or _curline if omitted. $2 is a condition to test ++# for whether to stop. ++ ++_Dbg_cmd_action() { ++ ++ local n=${1:-$_curline} ++ shift ++ ++ local stmt; ++ if [ -z "$1" ] ; then ++ condition=1 ++ else ++ condition="$*" ++ fi ++ ++ local -ar word=($(_Dbg_parse_linespec "$n")) ++ if [[ ${#word[@]} == 0 ]] ; then ++ _Dbg_msg "Invalid linespec $n" ++ return ++ fi ++ ++ local filename=${word[1]} ++ local line_number=${word[0]} ++ local full_filename=`_Dbg_is_file $filename` ++ if [[ -n $full_filename ]] ; then ++ if (( $line_number == 0 )) ; then ++ _Dbg_msg "There is no line 0 to set action at." ++ else ++ _Dbg_check_line $line_number "$full_filename" ++ (( $? == 0 )) && \ ++ _Dbg_set_action "$full_filename" "$line_number" $is_temp "$condition" ++ fi ++ else ++ filename=$(_Dbg_adjust_filename "$filename") ++ _Dbg_msg "File $filename not found in read-in files." ++ _Dbg_msg "See 'info files' for a list of known files." ++ fi ++} ++ ++# clear all actions ++_Dbg_cmd_clear_all_actions() { ++ ++ local _Dbg_prompt_output=${_Dbg_tty:-/dev/null} ++ read $_Dbg_edit -p "Delete all actions? (y/n): " \ ++ <&$_Dbg_input_desc 2>>$_Dbg_prompt_output ++ ++ if [[ $REPLY != [Yy]* ]] ; then ++ return 1 ++ fi ++ local -i k ++ for (( k=0; (( k < ${#_Dbg_filenames[@]} )) ; k++ )) ; do ++ local filename=${_filename[$k]} ++ local filevar="`_Dbg_file2var $filename`" ++ local action_a="_Dbg_action_${filevar}" ++ unset ${action_a}[$k] ++ done ++ _Dbg_write_journal_eval "_Dbg_action_line=()" ++ _Dbg_write_journal_eval "_Dbg_action_stmt=()" ++ _Dbg_write_journal_eval "_Dbg_action_file=()" ++ _Dbg_write_journal_eval "_Dbg_action_enable=()" ++ return 0 ++} ++ ++# delete actions(s) at given file:line numbers. If no file is given ++# use the current file. ++_Dbg_cmd_clear_action() { ++ # set -x ++ local -r n=${1:-$_curline} ++ ++ local -ar word=($(_Dbg_parse_linespec "$n")) ++ if [[ ${#word[@]} == 0 ]] ; then ++ _Dbg_msg "Invalid linespec $n" ++ return ++ fi ++ ++ local filename=${word[1]} ++ local line_number=${word[0]} ++ local full_filename=`_Dbg_is_file $filename` ++ if [[ -n $full_filename ]] ; then ++ if (( $line_number == 0 )) ; then ++ _Dbg_msg "There is no line 0 to clear action at." ++ else ++ _Dbg_check_line $line_number "$full_filename" ++ (( $? == 0 )) && \ ++ _Dbg_unset_action "$full_filename" "$line_number" ++ local -r found=$? ++ if [[ $found != 0 ]] ; then ++ _Dbg_msg "Removed $found action(s)." ++ else ++ _Dbg_msg "Didn't find any actions to remove at $n." ++ fi ++ fi ++ else ++ filename=$(_Dbg_adjust_filename "$filename") ++ _Dbg_msg "File $filename not found in read-in files." ++ _Dbg_msg "See 'info files' for a list of known files." ++ fi ++} ++ ++# list actions ++_Dbg_list_action() { ++ ++ if [ ${#_Dbg_action_line[@]} != 0 ]; then ++ _Dbg_msg "Actions at following places:" ++ local -i i ++ ++ _Dbg_msg "Num Enb Stmt file:line" ++ for (( i=0; (( i < _Dbg_action_max )) ; i++ )) ; do ++ if [[ -n ${_Dbg_action_line[$i]} ]] ; then ++ local source_file=${_Dbg_action_file[$i]} ++ source_file=$(_Dbg_adjust_filename "$source_file") ++ _Dbg_printf "%-3d %3d %-18s %s:%s" $i ${_Dbg_action_enable[$i]} \ ++ "${_Dbg_action_stmt[$i]}" \ ++ $source_file ${_Dbg_action_line[$i]} ++ fi ++ done ++ else ++ _Dbg_msg "No actions have been set." ++ fi ++} ++ ++# Internal routine to a set breakpoint unconditonally. ++ ++_Dbg_set_action() { ++ local source_file=$1 ++ local -ir line=$2 ++ local -r stmt=${3:-1} ++ local -r filevar="`_Dbg_file2var $source_file`" ++ ++ local val_str=`_Dbg_get_assoc_array_entry "_Dbg_action_$filevar" $line` ++ if [ -z "$val_str" ] ; then ++ val_str=$_Dbg_action_max ++ else ++ val_str="$val_str $_Dbg_action_max" ++ fi ++ ++ _Dbg_action_line[$_Dbg_action_max]=$line ++ _Dbg_action_file[$_Dbg_action_max]="$source_file" ++ _Dbg_action_stmt[$_Dbg_action_max]="$stmt" ++ _Dbg_action_enable[$_Dbg_action_max]=1 ++ ++ local dq_source_file=$(_Dbg_esc_dq "$source_file") ++ local dq_stmt=$(_Dbg_esc_dq "stmt") ++ ++ _Dbg_write_journal "_Dbg_action_line[$_Dbg_action_max]=$line" ++ _Dbg_write_journal "_Dbg_action_file[$_Dbg_action_max]=\"$dq_source_file\"" ++ _Dbg_write_journal "_Dbg_action_stmt[$_Dbg_action_max]=\"$dq_stmt\"" ++ _Dbg_write_journal "_Dbg_action_enable[$_Dbg_action_max]=1" ++ ++ _Dbg_set_assoc_array_entry "_Dbg_action_$filevar" $line $val_str ++ source_file=$(_Dbg_adjust_filename "$source_file") ++ _Dbg_msg "Breakpoint $_Dbg_action_max set at ${source_file}:$line." ++ ((_Dbg_action_max++)) ++ _Dbg_write_journal "_Dbg_action_max=$_Dbg_action_max" ++} ++ ++# Internal routine to delete a breakpoint by file/line. ++_Dbg_unset_action() { ++ local -r filename=$1 ++ local -ir line=$2 ++ local -r filevar="`_Dbg_file2var $filename`" ++ local -i found=0 ++ ++ # set -xv ++ local -r entries=`_Dbg_get_assoc_array_entry "_Dbg_action_$filevar" $line` ++ local -i del ++ for del in $entries ; do ++ if [[ -z ${_Dbg_action_file[$del]} ]] ; then ++ _Dbg_msg "No action found at $filename:$line" ++ continue ++ fi ++ if [[ ${_Dbg_action_file[$del]} != $filename ]] ; then ++ _Dbg_msg "action inconsistency:" \ ++ "$filename[$line] lists ${_Dbg_action_file[$del]} at entry $del" ++ else ++ _Dbg_write_journal_eval "unset _Dbg_action_line[$del]" ++ _Dbg_write_journal_eval "unset _Dbg_action_stmt[$del]" ++ _Dbg_write_journal_eval "unset _Dbg_action_file[$del]" ++ _Dbg_write_journal_eval "unset _Dbg_action_enable[$del]" ++ ((found++)) ++ fi ++ done ++ _Dbg_write_journal_eval unset _Dbg_action_$filevar[$line] ++ return $found ++ # set +xv ++} ++ ++# Routine to a delete breakpoint/watchpoint by entry numbers. ++_Dbg_cmd_action_delete() { ++ local -r to_go=$@ ++ local -i i ++ local -i found=0 ++ ++ # set -xv ++ eval "$_seteglob" ++ for del in $to_go ; do ++ case $del in ++ $int_pat ) ++ _Dbg_delete_action_entry $del ++ ((found += $?)) ++ ;; ++ * ) ++ _Dbg_msg "Invalid entry number skipped: $del" ++ esac ++ done ++ eval "$_resteglob" ++ [[ $found != 0 ]] && _Dbg_msg "Removed $found action(s)." ++ return $found ++ # set +xv ++} ++ ++#======================== DISPLAYs ============================# ++ ++# Enable/disable display by entry numbers. ++_Dbg_disp_enable_disable() { ++ if [ -z "$1" ] ; then ++ _Dbg_msg "Expecting a list of display numbers. Got none." ++ return 1 ++ fi ++ local -i on=$1 ++ local en_dis=$2 ++ shift; shift ++ ++ local to_go="$@" ++ local i ++ eval "$_seteglob" ++ for i in $to_go ; do ++ case $i in ++ $int_pat ) ++ _Dbg_enable_disable_display $on $en_dis $i ++ ;; ++ * ) ++ _Dbg_msg "Invalid entry number skipped: $i" ++ esac ++ done ++ eval "$_resteglob" ++ return 0 ++} ++ ++_Dbg_eval_all_display() { ++ local -i i ++ for (( i=0; i < _Dbg_disp_max ; i++ )) ; do ++ if [ -n "${_Dbg_disp_exp[$i]}" ] \ ++ && [[ ${_Dbg_disp_enable[i]} != 0 ]] ; then ++ _Dbg_printf_nocr "%2d (%s): " $i "${_Dbg_disp_exp[i]}" ++ _Dbg_cmd_eval "${_Dbg_disp_exp[i]}" ++ fi ++ done ++} ++ ++# Set display command or list all current display expressions ++_Dbg_cmd_display() { ++ if [[ -z "$@" ]]; then ++ _Dbg_eval_all_display ++ else ++ local -i n=_Dbg_disp_max++ ++ _Dbg_disp_exp[$n]="$@" ++ _Dbg_disp_enable[$n]=1 ++ _Dbg_printf '%2d: %s' $n "${_Dbg_disp_exp[$n]}" ++ fi ++} ++ ++# List display command(s) ++_Dbg_cmd_list_display() { ++ if [ ${#_Dbg_disp_exp[@]} != 0 ]; then ++ local i=0 j ++ _Dbg_msg "Display expressions:" ++ _Dbg_msg "Num Enb Expression " ++ for (( i=0; (( i < _Dbg_disp_max )); i++ )) ; do ++ if [ -n "${_Dbg_disp_exp[$i]}" ] ;then ++ _Dbg_printf '%-3d %3d %s' \ ++ $i ${_Dbg_disp_enable[$i]} "${_Dbg_disp_exp[$i]}" ++ fi ++ done ++ else ++ _Dbg_msg "No display expressions have been set." ++ fi ++} ++ ++# Enable/disable display(s) by entry numbers. ++_Dbg_enable_disable_display() { ++ local -i on=$1 ++ local en_dis=$2 ++ local -i i=$3 ++ if [ -n "${_Dbg_disp_exp[$i]}" ] ; then ++ if [[ ${_Dbg_disp_enable[$i]} == $on ]] ; then ++ _Dbg_msg "Display entry $i already $en_dis so nothing done." ++ else ++ _Dbg_write_journal_eval "_Dbg_disp_enable[$i]=$on" ++ _Dbg_msg "Display entry $i $en_dis." ++ fi ++ else ++ _Dbg_msg "Display entry $i doesn't exist so nothing done." ++ fi ++} ++ ++_Dbg_cmd_undisplay() { ++ local -i del=$1 ++ ++ if [ -n "${_Dbg_disp_exp[$del]}" ] ; then ++ _Dbg_write_journal_eval "unset _Dbg_disp_exp[$del]" ++ _Dbg_write_journal_eval "unset _Dbg_disp_enable[$del]" ++ else ++ _Dbg_msg "Display entry $del doesn't exist so nothing done." ++ fi ++} ++ ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-cmds.inc cvs/debugger/dbg-cmds.inc +--- bash-2.05b/debugger/dbg-cmds.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-cmds.inc 2003-08-06 16:21:55.000000000 +0200 +@@ -0,0 +1,848 @@ ++# dbg-cmds.inc - Bourne Again Shell Debugger Top-level debugger commands ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# Debugger command loop: Come here at to read debugger commands to ++# run. ++ ++# Main-line debugger read/execute command loop ++ ++# ==================== VARIABLES ======================================= ++# input_start_descriptor is the lowest descriptor we use for reading. ++# input_desc is the current descriptor in use. "sourc"ing other ++# command files will increase this descriptor ++ ++typeset -ir _Dbg_input_start_desc=9 ++typeset -i _Dbg_input_desc=_Dbg_input_start_desc-1 # will ++ before use ++ ++# Are we inside the middle of a "skip" command? ++typeset -i _Dbg_inside_skip=0 ++ ++# A variable holding a space is set so it can be used in a "set prompt" command ++# ("read" in the main command loop will remove a trailing space so we need ++# another way to allow a user to enter spaces in the prompt.) ++ ++typeset _Dbg_space=' ' ++ ++# Should we allow editing of debugger commands? ++# The value should either be '-e' or '' ++typeset _Dbg_edit='-e' ++ ++# What do we use for a debugger prompt? Technically we don't need to ++# use the above $bashdb_space in the assignment below, but we put it ++# in to suggest to a user that this is how one gets a spaces into the ++# prompt. ++ ++typeset _Dbg_prompt_str='bashdb${_Dbg_less}${#_Dbg_history[@]}${_Dbg_greater}$_Dbg_space' ++ ++# The arguments in the last "print" command. ++typeset _Dbg_last_print_args='' ++ ++# The arguments in the last "x" command. ++typeset _Dbg_last_x_args='' ++ ++# ===================== FUNCTIONS ======================================= ++ ++# Note: We have to be careful here in naming "local" variables. In contrast ++# to other places in the debugger, because of the read/eval loop, they are ++# in fact seen by those using the debugger. So in contrast to other "local"s ++# in the debugger, we prefer to preface these with _Dbg_. ++_Dbg_cmdloop() { ++ ++ # THIS SHOULD BE DONE IN dbg-sig.inc, but there's a bug in BASH in ++ # trying to change "trap RETURN" inside a "trap RETURN" handler.... ++ # Turn off return trapping. Not strictly necessary, since it *should* be ++ # covered by the _Dbg_ test below if we've named functions correctly. ++ # However turning off the RETURN trap should reduce unnecessary calls. ++ # trap RETURN ++ ++ _Dbg_inside_skip=0 ++ ++ # Evaluate all the display expressions ++ _Dbg_eval_all_display ++ ++ # Loop over all pending open input file descriptors ++ while (( $_Dbg_input_desc >= $_Dbg_input_start_desc )) ; do ++ local _Dbg_cmd; ++ local _Dbg_args; ++ ++ # Set up prompt to show shell level. ++ local _Dbg_greater='' ++ local _Dbg_less='' ++ local -i _Dbg_i=0 ++ for (( _Dbg_i=0 ; _Dbg_i < BASHDB_LEVEL ; _Dbg_i++ )) ; do ++ _Dbg_greater=">$_Dbg_greater" ++ _Dbg_less="<$_Dbg_less" ++ done ++ ++ for (( _Dbg_i=0 ; _Dbg_i < BASH_SUBSHELL ; _Dbg_i++ )) ; do ++ _Dbg_greater=")$_Dbg_greater" ++ _Dbg_less="$_Dbg_less(" ++ done ++ ++ # Loop over debugger commands. But before reading a debugger ++ # command, we need to make sure IFS is set to spaces to ensure our ++ # two variables (command name and rest of the arguments) are set ++ # correctly. Saving the IFS and setting it to the "normal" value ++ # of space should be done in the DEBUG signal handler entry. ++ ++ # Also, we need to make sure the prompt output is ++ # redirected to the debugger terminal. Both of these things may ++ # have been changed by the debugged program for its own ++ # purposes. Furthermore, were we *not* to redirect our stderr ++ # below, we may mess up what the debugged program expects to see ++ # in in stderr by adding our debugger prompt. ++ ++ # if no tty, no prompt ++ local _Dbg_prompt_output=${_Dbg_tty:-/dev/null} ++ ++ eval "local _Dbg_prompt=$_Dbg_prompt_str" ++ while read $_Dbg_edit -p "$_Dbg_prompt" \ ++ _Dbg_cmd _Dbg_args \ ++ <&$_Dbg_input_desc 2>>$_Dbg_prompt_output; do ++ ++ # Set default next, step or skip command ++ if [[ -z $_Dbg_cmd ]]; then ++ _Dbg_cmd=$_Dbg_last_next_step_cmd ++ _Dbg_args=$_Dbg_last_next_step_args ++ fi ++ ++ local dq_cmd=$(_Dbg_esc_dq "$_Dbg_cmd") ++ local dq_args=$(_Dbg_esc_dq "$_Dbg_args") ++ # _Dbg_write_journal_eval doesn't work here. Don't really understand ++ # how to get it to work. So we do this in two steps. ++ _Dbg_write_journal \ ++ "_Dbg_history[${#_Dbg_history[@]}]=\"$dq_cmd $dq_args\"" ++ _Dbg_history[${#_Dbg_history[@]}]="$_Dbg_cmd $_Dbg_args" ++ ++ _Dbg_hi=${#_Dbg_history[@]} ++ ++ local -i _Dbg_redo=1 ++ while (( $_Dbg_redo )) ; do ++ ++ _Dbg_redo=0 ++ ++ case $_Dbg_cmd in ++ ++ # Comment line ++ [#]* ) ++ _Dbg_remove_history_item ++ ;; ++ ++ # List window up to _curline ++ - ) ++ local -i start_line=(_curline+1-$_Dbg_listsize) ++ local -i count=($_Dbg_listsize) ++ if (( start_line <= 0 )) ; then ++ ((count=count+start_line-1)) ++ start_line=1; ++ fi ++ _Dbg_list $_cur_source_file $start_line $count ++ ;; ++ ++ # list current line ++ . ) ++ _Dbg_list $_cur_source_file $_curline 1 ++ ;; ++ ++ # Search forwards for pattern ++ /* ) ++ _Dbg_cmd_search $_Dbg_cmd ++ ;; ++ ++ # Search backwards for pattern ++ [?]* ) ++ _Dbg_cmd_search_back $_Dbg_cmd ++ ;; ++ ++ # Set action to be silently run when a line is hit ++ a ) ++ _Dbg_cmd_action $_Dbg_args ;; ++ ++ # Set breakpoint on a line ++ b | br | bre | brea | break ) ++ _Dbg_cmd_break 0 $_Dbg_args ;; ++ ++ # Continue ++ c | cont | conti |contin |continu | continue ) ++ _Dbg_cmd_continue $_Dbg_args ++ if [[ $? == 0 ]] ; then ++ IFS="$_Dbg_old_IFS"; ++ _Dbg_write_journal_eval \ ++ "_Dbg_old_set_opts=\"$_Dbg_old_set_opts -o fntrace\"" ++ return 0 ++ fi ++ ;; ++ ++ # Breakpoint/Watchpoint Conditions ++ cond | condi |condit |conditi | conditio | condition ) ++ _Dbg_cmd_condition $_Dbg_args ++ ;; ++ ++ # Delete all breakpoints by line number. ++ # Note we use "d" as an alias for "clear" to be compatible ++ # with the Perl5 debugger. ++ d | cl | cle | clea | clea | clear ) ++ _Dbg_cmd_clear_brkpt $_Dbg_args ++ ;; ++ ++ # Delete breakpoints by entry numbers. Note "d" is an alias for ++ # clear. ++ de | del | dele | delet | delete ) ++ _Dbg_cmd_delete $_Dbg_args ++ ;; ++ ++ # Set up a script for debugging into. ++ deb | debu | debug ) ++ _Dbg_cmd_debug $_Dbg_args ++ # Skip over the execute statement which presumably we ran above. ++ _Dbg_cmd_next_step_skip "skip" 1 ++ IFS="$_Dbg_old_IFS"; ++ return 1 ++ ;; ++ ++ # Disable breakpoints ++ di | dis | disa | disab | disabl | disable ) ++ _Dbg_cmd_disable $_Dbg_args ++ ;; ++ ++ # Display expression ++ disp | displ | displa| display ) ++ _Dbg_cmd_display $_Dbg_args ++ ;; ++ ++ # Delete all breakpoints. ++ D | deletea | deleteal | deleteall ) ++ _Dbg_clear_all_brkpt ++ ;; ++ ++ # List stack 1 up ++ do | dow | down ) ++ _Dbg_cmd_down $_Dbg_args ++ ;; ++ ++ # evaluate as bash command ++ e | ev | eva | eval ) ++ _Dbg_cmd_eval $_Dbg_args ++ ;; ++ ++ # evaluate as bash command ++ en | ena | enab | enabl | enable ) ++ _Dbg_cmd_enable $_Dbg_args ++ ;; ++ ++ # ++ fi | fin | fini | finis | finish | r ) ++ ++ trap '_Dbg_debug_trap_handler 1 $LINENO "$@"' RETURN ++ return 0 ++ ;; ++ ++ # View source in file ++ fr | fra | fra | frame ) ++ _Dbg_cmd_frame $_Dbg_args ++ ;; ++ ++ # Set signal handle parameters ++ ha | han | hand | handl | handle ) ++ _Dbg_cmd_handle $_Dbg_args ++ ;; ++ ++ # Info subcommands ++ i | in | inf | info ) ++ _Dbg_cmd_info $_Dbg_args ++ ;; ++ ++ # List line. ++ # print lines in file ++ l | li | lis | list ) ++ _Dbg_cmd_list $_Dbg_args ++ ;; ++ ++ # next/single-step N times (default 1) ++ n | ne | nex | next | s | st | ste | step | sk | ski | skip ) ++ _Dbg_last_next_step_cmd="$_Dbg_cmd" ++ _Dbg_last_next_step_args=$_Dbg_args ++ _Dbg_cmd_next_step_skip $_Dbg_cmd $_Dbg_args ++ if [[ $_Dbg_cmd == sk* ]] ; then ++ _Dbg_inside_skip=1 ++ return 1 ++ else ++ return 0 ++ fi ++ ;; ++ ++ # print globbed or substituted variables ++ p | pr | pri | prin | print ) ++ _Dbg_cmd_print "$_Dbg_args" ++ ;; ++ ++ # quit ++ q | qu | qui | quit ) ++ _Dbg_cmd_quit $_Dbg_args ++ ;; ++ ++ # Search backwards for pattern ++ rev | reve | rever | revers | reverse ) ++ _Dbg_cmd_search_back $_Dbg_args ++ ;; ++ ++ # return from function/source without finishing executions ++ ret | retu | retur | return ) ++ _Dbg_steps=1 ++ _Dbg_write_journal "_Dbg_steps=$_Dbg_steps" ++ IFS="$_Dbg_old_IFS"; ++ return 2 ++ ;; ++ ++ # Run a debugger set command ++ se | set ) ++ _Dbg_cmd_set $_Dbg_args ++ ;; ++ ++ # Search forwards for pattern ++ sea | sear | searc | search | \ ++ for | forw | forwa | forwar | forward ) ++ _Dbg_cmd_search $_Dbg_args ++ ;; ++ ++ # Run a debugger show command ++ sh | sho | show ) ++ _Dbg_cmd_show $_Dbg_args ++ ;; ++ ++ # Run a debugger comamnd file ++ so | sou | sour | sourc | source ) ++ _Dbg_cmd_source $_Dbg_args ++ ;; ++ ++ # toggle execution trace ++ t | to | tog | togg | toggl | toggle ) ++ _Dbg_cmd_trace ++ ;; ++ ++ # Set breakpoint on a line ++ tb | tbr | tbre | tbrea | tbreak ) ++ _Dbg_cmd_break 1 $_Dbg_args ;; ++ ++ # Set the output tty ++ tt | tty ) ++ _Dbg_cmd_tty $_Dbg_args ++ _Dbg_prompt_output=${_Dbg_tty:-/dev/null} ++ ;; ++ ++ # List call stack up ++ u | up ) ++ _Dbg_cmd_up $_Dbg_args ++ ;; ++ ++ # List call stack up ++ un | undi | undis | undisp | undispl | undispla | undisplay ) ++ _Dbg_cmd_undisplay $_Dbg_args ++ ;; ++ ++ # List window around line. ++ w | wi | win | wind | windo | window ) ++ ((_startline=_curline - _Dbg_listsize/2)) ++ (( $_startline <= 0 )) && _startline=1 ++ _Dbg_list $_cur_source_file $_startline ++ ;; ++ ++ # intelligent print of variable, function or expression ++ x | examine ) ++ _Dbg_cmd_x "$_Dbg_args" ++ ;; ++ ++ # List all breakpoints and actions. ++ L ) ++ _Dbg_cmd_list_brkpt ++ _Dbg_list_watch ++ _Dbg_list_action ++ ;; ++ ++ # print help command menu ++ '?' | h | he | hel | help ) ++ _Dbg_cmd_help $_Dbg_args ;; ++ ++ # run shell command. Has to come before ! below. ++ '!!' | sh | she | shell ) ++ eval $_Dbg_args ;; ++ ++ \!* | hi | his | hist | histo | histor | history ) ++ _Dbg_remove_history_item ++ _Dbg_cmd_history_parse $_Dbg_args ++ if (( $history_num >= 0 )) ; then ++ if (( $history_num < ${#_Dbg_history[@]} )) ; then ++ set ${_Dbg_history[$history_num]} ++ _Dbg_cmd=$1 ++ shift ++ _Dbg_args="$@" ++ _Dbg_redo=1; ++ else ++ _Dbg_msg \ ++ "Number $history_num should be less than ${#_Dbg_history[@]}" ++ fi ++ fi ++ ;; ++ ++ # Remove all actions ++ A ) ++ _Dbg_cmd_clear_all_actions $_Dbg_args ++ ;; ++ ++ # Run debugger command history ++ H ) ++ _Dbg_remove_history_item ++ _Dbg_cmd_history_list $_Dbg_args ++ ;; ++ ++ # restart debug session. ++ R | re | res | rest | resta | restar | restart | ru | run ) ++ _Dbg_cmd_restart $_Dbg_args ++ ;; ++ ++ # S List subroutine names ++ S ) ++ _Dbg_cmd_list_subroutines $_Dbg_args ++ ;; ++ ++ # Stack trace ++ T | wh | whe | wher | where | bt| back | backtrace ) ++ _Dbg_cmd_stack_trace 1 $_Dbg_args; ++ ;; ++ ++ # Dump variables ++ V ) ++ _Dbg_cmd_list_variables "$_Dbg_args" ++ ;; ++ ++ # Watch expression ++ We | watche ) ++ _Dbg_cmd_watch 1 "$_Dbg_args" ++ ;; ++ ++ # watch variable ++ W | wa | wat | watch ) ++ local -a a=($_Dbg_args) ++ local first=${a[0]} ++ if [[ $first == '' ]] ; then ++ _Dbg_cmd_watch 0 ++ else ++ if [[ 0 == `_Dbg_is_var $first` ]] ; then ++ _Dbg_msg "Can't set watch: no such variable $first." ++ else ++ unset a first ++ _Dbg_cmd_watch 0 "\$$_Dbg_args" ++ fi ++ fi ++ ;; ++ ++ # Show version information ++ M | ve | ver | vers | versi | versio | version ) ++ _Dbg_cmd_show_versions ++ ;; ++ ++ * ) ++ _Dbg_msg "Undefined command: \"$_Dbg_cmd\". Try \"help\"." ++ _Dbg_remove_history_item ++ ;; ++ esac ++ done ++ ++ IFS=$_Dbg_space_IFS; ++ eval "_Dbg_prompt=$_Dbg_prompt_str" ++ done ++ ++ ((_Dbg_input_desc--)) ++ done ++ ++ # EOF hit. Same as quit without arguments ++ _Dbg_msg "" # Cause since EOF may not have put in. ++ _Dbg_cmd_quit ++} ++ ++# Set up to Debug into another script... ++# TODO: would work better if instead of using $source_line below ++# which might have several statements, we could just pick up the next ++# single statement. ++_Dbg_cmd_debug() { ++ ++ # set -xv ++ local script_cmd=${@:-$_Dbg_bash_command} ++ ++ # We need to expand variables that might be in $script_cmd. ++ # set_Dbg_nested_debug_cmd is set up to to be eval'd below. ++ local set_Dbg_debug_cmd="local _Dbg_debug_cmd=\"$script_cmd\""; ++ ++ [ -z "$BASH" ] && BASH='bash' ++ ++ eval "$_seteglob" ++ # Add appropriate bash debugging options ++ if [[ $_Dbg_script != 1 ]] ; then ++ # Running "bash --debugger", so prepend "bash --debugger" ++ set_Dbg_debug_cmd="local _Dbg_debug_cmd=\"$BASH --debugger $script_cmd\""; ++ elif [[ $0/// == *bashdb/// ]] ; then ++ # Running "bashdb", so prepend "bash bashdb .." ++ set_Dbg_debug_cmd="local _Dbg_debug_cmd=\"$BASH $0 -q -L $_Dbg_libdir $script_cmd\""; ++ fi ++ eval "$_resteglob" ++ eval $set_Dbg_debug_cmd ++ ++ if (( _Dbg_basename_only )) ; then ++ _Dbg_msg "Debugging new script with $script_cmd" ++ else ++ _Dbg_msg "Debugging new script with $_Dbg_debug_cmd" ++ fi ++ local -r old_quit_on_quit=$BASHDB_QUIT_ON_QUIT ++ export BASHDB_QUIT_ON_QUIT=1 ++ export BASHDB_BASENAME_ONLY="$_Dbg_basename_only" ++ ((BASHDB_LEVEL++)) ++ $_Dbg_debug_cmd ++ ((BASHDB_LEVEL--)) ++ export BASHDB_QUIT_ON_QUIT=$old_quit_on_quit ++} ++ ++# V [![pat]] List variables and values for whose variables names which ++# match pat $1. If ! is used, list variables that *don't* match. ++# If pat ($1) is omitted, use * (everything) for the pattern. ++_Dbg_cmd_list_variables() { ++ local _Dbg_old_glob="$GLOBIGNORE" ++ GLOBIGNORE="*" ++ ++ local _Dbg_match="$1" ++ _Dbg_match_inverted=no ++ case ${_Dbg_match} in ++ \!*) ++ _Dbg_match_inverted=yes ++ _Dbg_match=${_Dbg_match#\!} ++ ;; ++ "") ++ _Dbg_match='*' ++ ;; ++ esac ++ local _Dbg_list=`declare -p` ++ local _Dbg_old_ifs=${IFS} ++ IFS=" ++" ++ local _Dbg_temp=${_Dbg_list} ++ _Dbg_list="" ++ local -i i=0 ++ local -a _Dbg_list ++ ++ # GLOBIGNORE protects us against using the result of ++ # a glob expansion, but it doesn't protect us from ++ # actually performing it, and this can bring bash down ++ # with a huge _Dbg_source_ variable being globbed. ++ # So here we disable globbing momentarily ++ set -o noglob ++ for _Dbg_item in ${_Dbg_temp}; do ++ _Dbg_list[${i}]="${_Dbg_item}" ++ i=${i}+1 ++ done ++ set +o noglob ++ IFS=${_Dbg_old_ifs} ++ local _Dbg_item="" ++ local _Dbg_skip=0 ++ local _Dbg_show_cmd="" ++ _Dbg_show_cmd=`echo -e "case \\${_Dbg_item} in \n${_Dbg_match})\n echo yes;;\n*)\necho no;; esac"` ++ ++ for (( i=0; (( i < ${#_Dbg_list[@]} )) ; i++ )) ; do ++ _Dbg_item=${_Dbg_list[$i]} ++ case ${_Dbg_item} in ++ *\ \(\)\ ) ++ _Dbg_skip=1 ++ ;; ++ \}) ++ _Dbg_skip=0 ++ continue ++ esac ++ if [[ _Dbg_skip -eq 1 ]]; then ++ continue ++ fi ++ ++ # Ignore all _Dbg_ variables here because the following ++ # substitutions takes a long while when it encounters ++ # a big _Dbg_source_ ++ case ${_Dbg_item} in ++ _Dbg_*) # Hide/ignore debugger variables. ++ continue; ++ ;; ++ esac ++ ++ _Dbg_item=${_Dbg_item/=/==/} ++ _Dbg_item=${_Dbg_item%%=[^=]*} ++ case ${_Dbg_item} in ++ _=);; ++ *=) ++ _Dbg_item=${_Dbg_item%=} ++ local _Dbg_show=`eval $_Dbg_show_cmd` ++ if [[ "$_Dbg_show" != "$_Dbg_match_inverted" ]]; then ++ if [[ -n ${_Dbg_item} ]]; then ++ local _Dbg_var=`declare -p ${_Dbg_item} 2>/dev/null` ++ if [[ -n "$_Dbg_var" ]]; then ++ # Uncomment the following 3 lines to use literal ++ # linefeeds ++# _Dbg_var=${_Dbg_var//\\\\n/\\n} ++# _Dbg_var=${_Dbg_var// ++#/\n} ++ # Comment the following 3 lines to use literal linefeeds ++ _Dbg_var=${_Dbg_var//\\\\n/\\\\\\n} ++ _Dbg_var=${_Dbg_var// ++/\\n} ++ _Dbg_var=${_Dbg_var#* * } ++ _Dbg_msg ${_Dbg_var} ++ fi ++ fi ++ fi ++ ;; ++ *) ++ ;; ++ esac ++ ++ done ++ GLOBIGNORE=$_Dbg_old_glob ++} ++ ++_Dbg_cmd_eval() { ++ ++ echo ". ${_Dbg_libdir}/dbg-set-d-vars.inc" > $_Dbg_evalfile ++ echo "$@" >> $_Dbg_evalfile ++ if [[ -n $_basdhb_tty ]] ; then ++ . $_Dbg_evalfile >>$_Dbg_tty ++ else ++ . $_Dbg_evalfile ++ fi ++} ++ ++_Dbg_cmd_next_step_skip() { ++ ++ local cmd=$1 ++ local count=${2:-1} ++ # Do we step debug into functions called or not? ++ if [[ $cmd == n* ]] ; then ++ _Dbg_old_set_opts="$_Dbg_old_set_opts +o fntrace" ++ else ++ _Dbg_old_set_opts="$_Dbg_old_set_opts -o fntrace" ++ fi ++ _Dbg_write_journal "_Dbg_old_set_opts=\"$_Dbg_old_set_opts\"" ++ ++ if [[ $count == [0-9]* ]] ; then ++ let _Dbg_steps=${count:-1} ++ else ++ _Dbg_msg "Argument ($count) should be a number or nothing." ++ _Dbg_steps=1 ++ fi ++ _Dbg_write_journal "_Dbg_steps=$_Dbg_steps" ++} ++ ++_Dbg_cmd_print() { ++ local -r _Dbg_expr=${@:-"$_Dbg_last_print_args"} ++ local -r dq_expr=$(_Dbg_esc_dq "$_Dbg_expr") ++ . ${_Dbg_libdir}/dbg-set-d-vars.inc ++ eval "_Dbg_msg $_Dbg_expr" ++ _Dbg_last_print_args="$dq_expr" ++} ++ ++# Restart script in same way with saved arguments (probably the same ++# ones as we were given before). ++_Dbg_cmd_restart() { ++ ++ _Dbg_cleanup; ++ ++ local script_args ++ if (( $# != 0 )) ; then ++ script_args="$@" ++ else ++ script_args="${_Dbg_script_args[@]}" ++ fi ++ ++ local exec_cmd="$0 $script_args"; ++ if [[ $_Dbg_script != 1 ]] ; then ++ [ -z "$BASH" ] && BASH='bash' ++ if [ $_cur_source_file == $_Dbg_bogus_file ] ; then ++ script_args="--debugger -c \"$BASH_EXECUTION_STRING\"" ++ exec_cmd="$BASH --debugger -c \"$BASH_EXECUTION_STRING\""; ++ else ++ exec_cmd="$BASH --debugger $0 $script_args"; ++ fi ++ fi ++ ++ if (( _Dbg_basename_only )) ; then ++ _Dbg_msg "Restarting with: $script_args" ++ else ++ _Dbg_msg "Restarting with: $exec_cmd" ++ fi ++ ++ # If we are in a subshell we need to get out of those levels ++ # first before we restart. The strategy is to write into persistent ++ # storage the restart command, and issue a "quit." The quit should ++ # discover the restart at the last minute and issue the restart. ++ if (( BASH_SUBSHELL > 0 )) ; then ++ _Dbg_msg "Note you are in a subshell. We will need to leave that first." ++ _Dbg_write_journal "BASHDB_RESTART_COMMAND=\"$exec_cmd\"" ++ _Dbg_cmd_quit 0 ++ fi ++ _Dbg_save_state ++ exec $exec_cmd ++} ++ ++# Handle command-file source. If the filename's okay we just increase the ++# input-file descriptor by one and redirect input which will ++# be picked up in next debugger command loop. ++_Dbg_cmd_source() { ++ local filename ++ if [[ -z "$1" ]] ; then ++ _Dbg_msg "Need to give a filename for the source command" ++ return ++ fi ++ _Dbg_glob_filename $1 ++ if [ "$filename" != '/dev/stdin' ] && [[ ! -f $filename ]] ; then ++ _Dbg_msg "Source file $filename does not exist as a regular file." ++ return ++ fi ++ if [[ -r $filename ]] ; then ++ ((_Dbg_input_desc++)) ++ _Dbg_input[$_Dbg_input_desc]=$filename ++ local _Dbg_redirect_cmd="exec $_Dbg_input_desc<$filename" ++ eval $_Dbg_redirect_cmd ++ else ++ _Dbg_msg "Source file $filename is not readable." ++ fi ++} ++ ++# toggle execution trace feature ++_Dbg_cmd_trace() { ++ ((_trace=!$_trace)) ++ ++ _Dbg_msg "Trace = \c" ++ let " $_trace" && _Dbg_msg "on" || _Dbg_msg "off" ++} ++ ++# Set output tty ++_Dbg_cmd_tty() { ++ if [[ -z "$1" ]] ; then ++ _Dbg_msg "Argument required (terminal name for running target process)." ++ return 1 ++ fi ++ if ! $(touch $1 >/dev/null 2>/dev/null); then ++ _Dbg_msg "Can't access $1 for writing." ++ return 1 ++ fi ++ if [[ ! -w $1 ]] ; then ++ _Dbg_msg "tty $1 needs to be writable" ++ return 1 ++ fi ++ _Dbg_tty=$1 ++ return 0 ++} ++ ++_Dbg_cmd_x() { ++ local -r _Dbg_expr=${@:-"$_Dbg_last_x_args"} ++ local _Dbg_result ++ if (( `_Dbg_is_var $_Dbg_expr` )) ; then ++ _Dbg_result=`declare -p $_Dbg_expr` ++ _Dbg_msg "$_Dbg_result" ++ elif (( `_Dbg_is_function $_Dbg_expr` )) ; then ++ _Dbg_result=`declare -f $_Dbg_expr` ++ _Dbg_msg "$_Dbg_result" ++ else ++ local -i _Dbg_rc ++ eval let _Dbg_result=$_Dbg_expr 2>/dev/null; _Dbg_rc=$? ++ if (( $_Dbg_rc != 0 )) ; then ++ _Dbg_cmd_print "$_Dbg_expr" ++ else ++ _Dbg_msg "$_Dbg_result" ++ fi ++ fi ++ _Dbg_last_x_args="$_Dbg_x_args" ++} ++ ++_Dbg_save_state() { ++ _Dbg_statefile=`_Dbg_tempname statefile` ++ echo "" > $_Dbg_statefile ++ _Dbg_save_breakpoints ++ _Dbg_save_actions ++ _Dbg_save_watchpoints ++ _Dbg_save_display ++ _Dbg_save_Dbg_set ++ echo "unset BASHDB_RESTART_FILE" >> $_Dbg_statefile ++ echo "rm $_Dbg_statefile" >> $_Dbg_statefile ++ export BASHDB_RESTART_FILE="$_Dbg_statefile" ++ _Dbg_write_journal "export BASHDB_RESTART_FILE=\"$_Dbg_statefile\"" ++ ++} ++ ++ ++_Dbg_save_Dbg_set() { ++ declare -p _Dbg_basename_only >> $_Dbg_statefile ++ declare -p _Dbg_debug_debugger >> $_Dbg_statefile ++ declare -p _Dbg_edit >> $_Dbg_statefile ++ declare -p _Dbg_listsize >> $_Dbg_statefile ++ declare -p _Dbg_prompt_str >> $_Dbg_statefile ++ declare -p _Dbg_show_command >> $_Dbg_statefile ++} ++ ++_Dbg_restore_state() { ++ local statefile=$1 ++ . $1 ++} ++ ++# ================== INITIALIZATION ===================================== ++ ++_Dbg_cmd_source '/dev/stdin' ++ ++# List of command files to process ++typeset -a _Dbg_input ++ ++# Have we already specified where to read debugger input from? ++# Note: index 0 is only set by bashdb. It is not used otherwise for I/O ++# like those indices >= _Dbg_input_start_desc are. ++if [ -n "$BASHDB_INPUT" ] ; then ++ _Dbg_input=($BASHDB_INPUT) ++ _Dbg_cmd_source ${_Dbg_input[0]} ++ _Dbg_no_init=1 ++fi ++ ++if [[ -z $_Dbg_no_init && -r ~/.bashdbinit ]] ; then ++ _Dbg_cmd_source ~/.bashdbinit ++fi ++ ++# BASHDB_LEVEL is the number of times we are nested inside a debugger ++# by virtue of running "debug" for example. ++if [[ -z "${BASHDB_LEVEL}" ]] ; then ++ export BASHDB_LEVEL=1 ++fi ++ ++# temp file for internal eval'd commands ++typeset _Dbg_evalfile=`_Dbg_tempname eval` ++ ++# File to save information that needs to be passed from a subshell ++# to a parent shell ++typeset _Dbg_journal=`_Dbg_tempname journal` ++if [ ! -f _Dbg_journal ] ; then ++ _Dbg_write_journal "BASHDB_QUIT_LEVELS=0" ++fi ++ ++# This is put at the so we have something at the end to stop at ++# when we debug this. By stopping at the end all of the above functions ++# and variables can be tested. ++_Dbg_cmd_ver='$Id: dbg-cmds.inc,v 1.9 2003/08/06 14:21:55 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-file.inc cvs/debugger/dbg-file.inc +--- bash-2.05b/debugger/dbg-file.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-file.inc 2003-08-02 21:54:47.000000000 +0200 +@@ -0,0 +1,227 @@ ++# Things related to file handling. ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# Directory search patch for unqualified file names ++typeset -a _Dbg_dir=('\$cdir' '\$cwd' ) ++ ++# Directory in which the script is located ++typeset -r _Dbg_cdir=${_Dbg_source_file%/*} ++ ++_Dbg_resolve_expand_filename() { ++ local find_file=$1 ++ ++ if [[ -z "$find_file" ]] ; then ++ _Dbg_msg "Internal debug error: null file to find" ++ echo '' ++ return 1 ++ fi ++ ++ if [[ ${find_file:0:1} == '/' ]] ; then ++ echo "$find_file" ++ return 0 ++ elif [[ ${find_file:0:1} == '.' ]] ; then ++ # Relative file name ++ find_file=$(_Dbg_expand_filename $find_file) ++ echo "$find_file" ++ return 0 ++ else ++ # Resolve file using _Dbg_dir ++ local -i n=${#_Dbg_dir[@]} ++ local -i i ++ for (( i=0 ; i < n; i++ )) ; do ++ local basename="${_Dbg_dir[i]}" ++ if [[ $basename == '$cdir' ]] ; then ++ basename=$_Dbg_cdir ++ elif [[ $basename == '$cwd' ]] ; then ++ basename=$(pwd) ++ fi ++ if [[ -f "$basename/$find_file" ]] ; then ++ echo "$basename/$find_file" ++ return 0 ++ fi ++ done ++ fi ++ echo "" ++ return 1 ++} ++ ++# _Dbg_is_file echoes the full filename if $1 is a filename found in files ++# '' is echo'd if no file found. ++_Dbg_is_file() { ++ local find_file=$1 ++ ++ if [[ -z "$find_file" ]] ; then ++ _Dbg_msg "Internal debug error: null file to find" ++ echo '' ++ return ++ fi ++ ++ if [[ ${find_file:0:1} == '/' ]] ; then ++ # Absolute file name ++ for try_file in ${_Dbg_filenames[@]} ; do ++ if [[ $try_file == $find_file ]] ; then ++ echo "$try_file" ++ return ++ fi ++ done ++ elif [[ ${find_file:0:1} == '.' ]] ; then ++ # Relative file name ++ find_file=$(_Dbg_expand_filename $find_file) ++ for try_file in ${_Dbg_filenames[@]} ; do ++ if [[ $try_file == $find_file ]] ; then ++ echo "$try_file" ++ return ++ fi ++ done ++ else ++ # Resolve file using _Dbg_dir ++ for try_file in ${_Dbg_filenames[@]} ; do ++ local pathname ++ local -i n=${#_Dbg_dir[@]} ++ local -i i ++ for (( i=0 ; i < n; i++ )) ; do ++ local basename="${_Dbg_dir[i]}" ++ if [[ $basename == '$cdir' ]] ; then ++ basename=$_Dbg_cdir ++ elif [[ $basename == '$cwd' ]] ; then ++ basename=$(pwd) ++ fi ++ if [[ "$basename/$find_file" == $try_file ]] ; then ++ echo "$try_file" ++ return ++ fi ++ done ++ done ++ fi ++ echo "" ++} ++ ++# Turn filename $1 into something that is safe to use as a variable name ++_Dbg_file2var() { ++ local filename=$(_Dbg_expand_filename $1) ++ local varname=`builtin echo $filename | tr '=~+%* .?/"[]<>-' 'ETPpABDQSqLRlGM'` ++ builtin echo $varname ++} ++ ++# $1 contains the name you want to glob. return 1 if exists and is ++# readible or 0 if not. ++# The result will be in variable $filename which is assumed to be ++# local'd by the caller ++_Dbg_glob_filename() { ++ local cmd="filename=`expr $1`" ++ eval $cmd ++} ++ ++# Either fill out or strip filename as determined by "basename_only" ++# and annotate settings ++_Dbg_adjust_filename() { ++ local -r filename="$1" ++ if (( _Dbg_annotate == 1 )) ; then ++ echo `_Dbg_resolve_expand_filename $filename` ++ elif ((_Dbg_basename_only)) ; then ++ echo ${filename##*/} ++ else ++ echo $filename ++ fi ++} ++ ++# Return the maximum line in $1 ++_Dbg_get_maxline() { ++ # set -x ++ local -r filename=$1 ++ local -r filevar=`_Dbg_file2var $filename` ++ local is_read=`_Dbg_get_assoc_scalar_entry "_Dbg_read_" $filevar` ++ [ $is_read ] || _Dbg_readin $filename ++ echo `_Dbg_get_assoc_scalar_entry "_Dbg_maxline_" $filevar` ++ # set +x ++} ++ ++# Check that line $2 is not greater than the number of lines in ++# file $1 ++_Dbg_check_line() { ++ local -ir line_number=$1 ++ local filename=$2 ++ local -i max_line=`_Dbg_get_maxline $filename` ++ if (( $line_number > max_line )) ; then ++ (( _Dbg_basename_only )) && filename=${filename##*/} ++ _Dbg_msg "Line $line_number is too large." \ ++ "File $filename has only $max_line lines." ++ return 1 ++ fi ++ return 0 ++} ++ ++# Create temporary file based on $1 ++# file $1 ++_Dbg_tempname() { ++ echo "$_Dbg_tmpdir/bashdb$1$$" ++} ++ ++# append a command into journal file and then run the command. ++_Dbg_write_journal_eval() { ++ _Dbg_write_journal "$*" ++ eval "$*" ++} ++ ++# append a command into journal file and then run the command. ++_Dbg_write_journal_var() { ++ local var_name=$1 ++ local val ++ local val_cmd="$val=\${$var_name}" ++ eval $val_cmd ++ _Dbg_write_journal "${var_name}=${val}" ++} ++ ++_Dbg_write_journal_avar() { ++ local decl_str=$(declare -p $1) ++ local -a decl_a=($decl_str) ++ local -a decl_a2=${decl_a[@]:2} ++ _Dbg_write_journal ${decl_a2[@]} ++} ++ ++# Append a command into journal file. But we only need to do ++# if we are in a subshell. ++_Dbg_write_journal() { ++ if (( $BASH_SUBSHELL != 0 )) ; then ++ echo "$@" >> ${_Dbg_journal} 2>/dev/null ++ fi ++ # return $? ++} ++ ++# Remove all journal files. ++_Dbg_erase_journals() { ++ rm ${_Dbg_journal} 2>/dev/null ++} ++ ++# read in or "source" in journal file which will set variables. ++_Dbg_source_journal() { ++ ++ if [ -r $_Dbg_journal ] ; then ++ . $_Dbg_journal ++ (( BASH_SUBSHELL == 0 )) && _Dbg_erase_journals ++ fi ++} ++ ++# This is put at the so we have something at the end when we debug this. ++typeset -r _Dbg_file_ver=\ ++'$Id: dbg-file.inc,v 1.11 2003/08/02 19:54:47 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-fns.inc cvs/debugger/dbg-fns.inc +--- bash-2.05b/debugger/dbg-fns.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-fns.inc 2003-04-30 02:22:55.000000000 +0200 +@@ -0,0 +1,301 @@ ++# dbg-fns.inc - Bourne Again Shell Debugger Utility Functions ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# Come here via DEBUG trap after each statement in script. ++# This determines if we need to stop and go into the debugger ++# command loop or not. ++ ++# Add escapes to a string $1 so that when it is read back via "$1" ++# it is the same as $1. ++_Dbg_esc_dq() { ++ builtin echo $1 | sed -e 's/[`$\"]/\\\0/g' ++} ++ ++# Set $? to $1 if supplied or the saved entry value of $?. ++_Dbg_set_dol_q () { ++ return ${1:-$_Dbg_debugged_exit_code} ++} ++ ++# Split $2 using $1 as the split character. We accomplish this by ++# temporarily resetting the variable IFS (input field separator). ++# ++# Example: ++# local -a a=($(_Dbg_split ":" "file:line")) ++# a[0] will have file and a{1] will have line. ++ ++_Dbg_split (){ ++ local old_IFS=$IFS ++ local new_ifs=${1:-' '} ++ shift ++ local -r text=$* ++ local -a array ++ IFS="$new_ifs" ++ array=( $text ) ++ echo ${array[@]} ++ IFS=$old_IFS ++} ++ ++# Return value of eval($1$2). Until bash has associative arrays, ++# this is how we simulate such a datatype. In Perl this would be returning ++# the value of $1{$2}. ++ ++_Dbg_get_assoc_scalar_entry() { ++# set -xv ++ local prefix=$1 ++# echo "${BASH_SOURCE[1]}:${BASH_LINENO[1]} ${FUNCNAME[1]}" ++ local entry=$2 ++ local cmd="echo \$${prefix}${entry}" ++ eval $cmd ++# set +xv ++} ++ ++# Set eval($1$2)=$2. Until bash has associative arrays, ++# this is how we simulate such a datatype. In Perl this would be ++# $1{$2}=$3. ++ ++_Dbg_set_assoc_scalar_entry() { ++# set -xv ++ local prefix=$1 ++ local entry=$2 ++ local value=$3 ++ local cmd="${prefix}${entry}=$value" ++ eval $cmd ++ ++ local dq_value=$(_Dbg_esc_dq "$value") ++ _Dbg_write_journal "${prefix}${entry}=\"$dq_value\"" ++ ++# set +xv ++} ++ ++# Return value of eval($1[$2]). If $2 is omitted, use _curline. Until ++# bash has associative arrays which can contain array elements this is ++# how we simulate such a datatype. ++ ++_Dbg_get_assoc_array_entry() { ++ local prefix=$1 ++ local lineno=${2:-$_curline} ++ ++ [[ -z $prefix ]] && _Dbg_msg "Internal debug error (gae) bad prefix" ++ local entry="$prefix[$lineno]" ++ local cmd="echo \"\${$entry}\"" ++ eval $cmd ++} ++ ++# Evaluate eval($1[$2]=$3). If $2 is omitted, use _curline. Until ++# bash has associative arrays cich can contain array elements this is ++# how we simulate such a datatype. ++ ++_Dbg_set_assoc_array_entry() { ++# set -xv ++ local prefix=$1 ++ local lineno=$2 ++ shift; shift ++ local value=$* ++ ++ [[ -z "$prefix" ]] && _Dbg_msg "Internal debug error (sae1) bad prefix" ++ [[ -z "$lineno" ]] && _Dbg_msg "Internal debug error (sae2) bad lineno" ++ [[ -z "$value" ]] && _Dbg_msg "Internal debug error (sae3) bad value" ++ ++ local entry="$prefix[$lineno]" ++ local cmd="$entry=\"$value\"" ++ eval $cmd ++ ++ local dq_value=$(_Dbg_esc_dq "$value") ++ _Dbg_write_journal "$entry=\"$dq_value\"" ++ ++# set +xv ++} ++ ++# _Dbg_is_var echoes 1 if $1 is a defined variable or 0 otherwise. ++_Dbg_is_var() { ++ declare -p $1 >/dev/null 2>&1 ++ if [[ $? != 0 ]] ; then ++ echo 0 ++ else ++ echo 1 ++ fi ++} ++ ++# _Dbg_is_function echoes 1 if $1 is a defined function or 0 otherwise. ++# if $2 is nonzero, system functions, i.e. those whose name starts with ++# an underscore (_), are included in the search. ++_Dbg_is_function() { ++ local needed_fn=$1 ++ if [[ -z $needed_fn ]] ; then ++ echo 0 ++ return; ++ fi ++ local -i include_system=$2 ++ local -a fns_a=(`declare -F`) ++ local -i i ++ ++ # Iterate skipping over consecutive single tokens "declare" and "-F" ++ for (( i=2; (( i < ${#fns_a[@]} )) ; i += 3 )) ; do ++ local fn="${fns_a[$i]}" ++ [[ $fn == _* ]] && (( ! $include_system )) && continue ++ if [[ $needed_fn == $fn ]] ; then ++ echo 1 ++ return ++ fi ++ done ++ echo 0 ++} ++ ++# _get_function echoes a list of all of the functions. ++# if $1 is nonzero, system functions, i.e. those whose name starts with ++# an underscore (_), are included in the search. ++# FIXME add parameter search pattern. ++_Dbg_get_functions() { ++ local -i include_system=${1:-0} ++ local pat=${2:-*} ++ local -a fns_a=(`declare -F`) ++ local -a ret_fns=() ++ local -i i ++ local -i invert=0; ++ ++ if [[ $pat == !* ]] ; then ++ # Remove leading ! ++ pat=$(echo $pat | cut -c2-) ++ invert=1 ++ fi ++ ++ # Iterate skipping over consecutive single tokens "declare" and "-F" ++ for (( i=2; (( i < ${#fns_a[@]} )) ; i += 3 )) ; do ++ local fn="${fns_a[$i]}" ++ [[ $fn == _* ]] && (( ! $include_system )) && continue ++ if [[ $fn == $pat ]] ; then ++ [[ $invert == 0 ]] && ret_fns[${#ret_fns[@]}]=$fn ++ else ++ [[ $invert != 0 ]] && ret_fns[${#ret_fns[@]}]=$fn ++ fi ++ ++ done ++ echo ${ret_fns[@]} ++} ++ ++# Parse linespec in $1 which should be one of ++# int ++# file:line ++# function-num ++# Return tuple (line, filename) ++# We return the line number first since that can't have embedded blanks. ++_Dbg_parse_linespec() { ++ local linespec=$1 ++ eval "$_seteglob" ++ case "$linespec" in ++ ++ # line number only - use _cur_source_file for filename ++ $int_pat ) ++ echo "$linespec $_cur_source_file" ++ ;; ++ ++ # file:line ++ [^:][^:]*[:]$int_pat ) ++ local -a word=($(_Dbg_split ":" $linespec)) ++ echo "${word[1]} ${word[0]}" ++ ;; ++ ++ # Function name or error ++ * ) ++ if [[ 1 == `_Dbg_is_function $linespec $_Dbg_debug_debugger` ]]; then ++ local -a word=(`declare -F $linespec`) ++ echo "${word[1]} ${word[2]}" ++ else ++ echo "" ++ fi ++ ;; ++ esac ++} ++ ++# usage _Dbg_set_ftrace [-u] funcname [funcname...] ++# Sets or unsets a function for stopping by setting ++# the -t or +t property to the function declaration. ++# ++_Dbg_set_ftrace() { ++ local opt=-t tmsg="enabled" func ++ if [[ $1 == -u ]]; then ++ opt=+t ++ tmsg="disabled" ++ shift ++ fi ++ for func; do ++ declare -f $opt $func ++ # _Dbg_msg "Tracing $tmsg for function $func" ++ done ++} ++ ++# Do things for debugger entry. Set some global debugger variables ++# Remove trapping ourselves. ++# We assume that we are nested two calls deep from the point of debug ++# or signal fault. If this isn't the constant 2, then consider adding ++# a parameter to this routine. ++ ++_Dbg_set_debugger_entry() { ++ ++ # Nuke DEBUG trap ++ trap '' DEBUG ++ ++ _cur_fn=${FUNCNAME[2]} ++ let _curline=${BASH_LINENO[1]} ++ ((_curline < 1)) && let _curline=1 ++ ++ _Dbg_old_IFS="$IFS" ++ IFS="$_Dbg_space_IFS"; ++ _cur_source_file=${BASH_SOURCE[2]:-$_Dbg_bogus_file} ++ _cur_source_file="`_Dbg_expand_filename $_cur_source_file`" ++ _cur_filevar="`_Dbg_file2var $_cur_source_file`" ++ _Dbg_stack_pos=$_Dbg_stack_top ++ _Dbg_listline=_curline ++ ++ # echo "-- ${_cur_source_file}:$_curline" ++ ++ PS4=$_cur_source_file ++} ++ ++# Does things to do just before leaving debugger command loop and ++# returning to the debugged program. ++# $1 is 1 if we have called bashdb_set_debugger_entry and saved various ++# variables. ++_Dbg_set_to_return_from_debugger() { ++ _Dbg_rc=$? ++ ++ if (( $1 != 0 )) ; then ++ _Dbg_last_bash_command="$_Dbg_bash_command" ++ _Dbg_last_curline="$_curline" ++ _Dbg_last_source_file="$_cur_source_file" ++ else ++ _Dbg_last_curline==${BASH_LINENO[1]} ++ _Dbg_last_source_file=${BASH_SOURCE[2]:-$_Dbg_bogus_file} ++ _Dbg_last_bash_command="**unsaved _bashdb command**" ++ fi ++ ++ trap '_Dbg_debug_trap_handler 0 "$BASH_COMMAND" "$@"' DEBUG ++ ++ IFS="$_Dbg_space_IFS"; ++ set -$_Dbg_old_set_opts ++ IFS="$_Dbg_old_IFS"; ++} ++ ++# This is put at the so we have something at the end when we debug this. ++typeset -r _Dbg_fns_ver=\ ++'$Id: dbg-fns.inc,v 1.5 2003/04/30 00:22:55 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-help.inc cvs/debugger/dbg-help.inc +--- bash-2.05b/debugger/dbg-help.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-help.inc 2003-08-02 22:20:40.000000000 +0200 +@@ -0,0 +1,713 @@ ++# dbg-help.inc - Bourne Again Shell Debugger Help Routines ++ ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# print help command ++_Dbg_cmd_help() { ++ ++ local -r db_cmd=$1 ++ ++ if [[ -n "$db_cmd" ]] ; then ++ case $db_cmd in ++ !! | sh | she | shell ) ++ _Dbg_msg \ ++"!! cmd [args] Execute shell \"cmd\" \"args\". Alias: shell." ++ return ;; ++ '#' ) ++ _Dbg_msg \ ++"# Comment - ignore line. Maybe useful in command scripts." ++ return ;; ++ . ) ++ _Dbg_msg \ ++". List current window of lines." ++ return ;; ++ / ) ++ _Dbg_msg \ ++"/pat/ Search forward for pat. Trailing / is optional. ++ Long command name: search or forward." ++ return ;; ++ '?'/ ) ++ _Dbg_msg \ ++"?pat? Search backward for pat. Trailing ? is optional. ++ Long command name: rev or reverse" ++ return ;; ++ - ) ++ _Dbg_msg \ ++"- List previous window of lines." ++ return ;; ++ A ) ++ _Dbg_msg \ ++"A Delete all actions" ++ return ;; ++ D | deleteall ) ++ _Dbg_msg \ ++"D Delete all breakpoints" ++ return ;; ++ H ) ++ _Dbg_msg \ ++"H [from [to]] List debugger history. If no arguments given list all history. ++H -count If a single postive integer is given, then list starting from ++![-]num:p that number. If a single negative integer is given list that ++ many history items. If second argument is given then list down ++ to that history number. ++ An alternate form is !n:p or !-n:p where n is an ++ integer. If a minus sign is used, you go back num from the end ++ rather than specify an absolute history number" ++ return ;; ++ L ) ++ _Dbg_msg \ ++"L List all breakpoints." ++ return ;; ++ R | re | res | rest | resta | restar | restart ) ++ _Dbg_msg \ ++"R [args] Attempt to restart the program. ++ The source code is reread and breakpoint information is lost. ++ The command arguments used on the last invocation are used if ++ args is empty. If arguments were given, they are passed to the ++ program. If running via the bashdb script and you want to ++ change arguments you also need to include those arguments ++ to the bashdb script. Long command name: restart. Alias: run." ++ return ;; ++ S ) ++ _Dbg_msg \ ++"S [[!]pattern] List subroutine names [not] matching bash pattern. If no ++ pattern is given, all subroutines are listed. (The pattern ++ is *)." ++ return ;; ++ T | wh | whe | wher | where | bt| back | backtrace ) ++ _Dbg_msg \ ++"T [n] Stack trace of calling functions or sourced files, ++ line numbers, and files. If n is given, list only n lines. ++ Long command name: where or backtrace." ++ return ;; ++ a ) ++ _Dbg_msg \ ++"a [linespec] stmt Perform stmt on reaching linespec." ++ return ;; ++ b | br | bre | brea | break ) ++ _Dbg_msg \ ++"b [linespec] Break on linespec. If no argument is given, us the current ++ line. Long command name: break." ++ return ;; ++ c | cont | conti |contin |continu | continue ) ++ _Dbg_msg \ ++"c [linespec] Continue execution until the next breakpoint or end of ++ program or linespec. Long command name: continue." ++ return ;; ++ cond | condi |condit |conditi | conditio | condition ) ++ _Dbg_msg \ ++"cond N COND Specify breakpoint number N to break only if COND is true. ++N is an integer and COND is an expression to be evaluated whenever ++breakpoint N is reached. Long command name: condition." ++ return ;; ++ d | cl | cle | clea | clea | clear ) ++ _Dbg_msg \ ++"cl [linespec] Clear breakpoint at specified line-spec. If no line given, use ++ the current line. All breakpoints in that line are cleared. ++ Long command name: clear." ++ return ;; ++ de | del | dele | delet | delete ) ++ _Dbg_msg \ ++"d {num}.. Delete the breakpoint entry or entries. ++ Long command name: delete." ++ return ;; ++ deb | debu | debug ) ++ _Dbg_msg \ ++"deb [script] Set up [script] for debugging. If no script is given, take ++ the script name from the command to be executed. ++ Long command name: debug." ++ return ;; ++ di | dis | disa | disab | disabl | disable ) ++ _Dbg_msg \ ++"di {n}... Disable breakpoint entry/entries. Long command name: disable." ++ return ;; ++ disp | displ | displa | display ) ++ _Dbg_msg \ ++"disp {n} Set display expression or list all display expressions. ++ Long command name: display." ++ return ;; ++ do | dow | down ) ++ _Dbg_msg \ ++"down [count] Set file location for printing down the call stack by ++ count. If count is omitted use 1." ++ return ;; ++ e | ev | eva ) ++ _Dbg_msg \ ++"e bash-cmd Evaluate a bash command by sourcing it in a subshell. Long ++ command name: eval." ++ return ;; ++ en | ena | enab | enabl | enable ) ++ _Dbg_msg \ ++"en {n}... Enable breakpoint entry/entries. Long command name: enable." ++ return ;; ++ r | fi | fin| fini | finis | finish ) ++ _Dbg_msg \ ++"r Execute until the current function or source file returns. ++ Long command name: finish." ++ return ;; ++ ha | han | hand | handl | handle ) ++ _Dbg_msg \ ++"ha sig {stop|nostop|stack|nostack|print|noprint} ++ Specify how to handle a signal. Args are signals and ++ actions to apply to those signals. Symbolic signals ++ (e.g. SIGSEGV) are recommended but numeric signals are ++ allowed too. Recognized actions include \"stop\", ++ \"nostop\", \"print\", and \"noprint\". Stop means ++ reenter debugger if this signal happens. Print means ++ print a message if this signal happens. Long command ++ name: handle." ++ return ;; ++ ++ hi | his | hist | histo | histor | history ) ++ _Dbg_msg \ ++"hi n Rerun a debugger command from the debugger history. See also ++![-]n H to list the history. An alternname form is !n or !-n. If ++ the minus sign is used you going back n from the end rather ++ than specifying an absolute history number. ++ Long command name: history." ++ return ;; ++ i | in | inf | info ) ++ _Dbg_help_info $2 ++ return ;; ++ l | li | lis | list ) ++ _Dbg_msg \ ++"l linespec List window lines starting at linespec. ++l min incr List incr lines starting at 'min' linespec. ++l List next window of lines. ++l . Same as above. ++ Long command name: list." ++ return ;; ++ n | ne | nex | next ) ++ _Dbg_msg \ ++"n [count] Single steps possibly skipping execution into subroutine calls ++ and sourced files. Performed [count] times default is 1 time. ++ Long command name: next." ++ return ;; ++ p | pr | pri | prin | print ) ++ _Dbg_msg \ ++"p string Print value of a substituted string via \`echo'. A variable ++ should have leading $ if its value is to be substituted. ++ Long command name: print." ++ return ;; ++ q | qu | qui | quit ) ++ _Dbg_msg \ ++"q [expr] [n] Terminate execution of the program with return code expr. If ++ expr is missing, use 0. If n is given then we terminate only ++ that many subshells or nested shells. Long name: quit." ++ return ;; ++ ret | retu | retur | return ) ++ _Dbg_msg \ ++"ret Skip completion of this function or sourced file. ++ Long name: return." ++ return ;; ++ se | set ) ++ _Dbg_help_set $2 ++ return ;; ++ s | st | ste | step ) ++ _Dbg_msg \ ++"s [count] Single steps possibly executing into subroutine calls ++ and sourced files. Performed [count] times default is 1 time." ++ return ;; ++ sh | sho | show ) ++ _Dbg_help_show $2 ++ return ;; ++ sk | ski | skip ) ++ _Dbg_msg \ ++"sk [count] Skip (don't run) the next [count] command(s) to be executed. ++ Long command name: skip." ++ return ;; ++ so | sou | sour | sourc | source ) ++ _Dbg_msg \ ++"source file Read in debugger command file." ++ return ;; ++ t | to | tog | togg | toggl | toggle ) ++ _Dbg_msg \ ++"t | trace Toggle line-execution tracing. Long command name: toggle." ++ return ;; ++ tb | tbr | tbre | tbrea | tbreak ) ++ _Dbg_msg \ ++"tb [linespec] Set a one-time break on linespec. If no argument is given, ++ us the current line. Long command name: tbreak." ++ return ;; ++ tt | tty ) ++ _Dbg_msg \ ++"tt tty-name Set the output device for debugger output ++ Long command name: tty." ++ return ;; ++ u | up ) ++ _Dbg_msg \ ++"u | up [count] Set file location for printing up the call stack by ++ count. If count is omitted use 1." ++ return ;; ++ v | ve | ver | vers | versi | versio | version ) ++ _Dbg_msg \ ++"M | version Show release version-control IDs of debugger scripts." ++ return ;; ++ w | window ) ++ _Dbg_msg \ ++"w [linespec] List window around line or current linespec. Long command ++ name: window." ++ return ;; ++ x | examine ) ++ _Dbg_msg \ ++"x expr Print value of an expression via \'declare', \`let' and then ++ failing these eval. Single variables and arithmetic expression ++ do not need leading $ for their value is to be substituted. ++ However if neither these, variables need $ to have their value ++ substituted. Long command name: examine" ++ return ;; ++ V ) ++ _Dbg_msg \ ++"V [!][pat] List variables and values for whose variables names which ++ match pat. If ! is used, list variables that *don't* match. ++ If pat is omitted, use * (everything) for the pattern." ++ return ;; ++ We | watche ) ++ _Dbg_msg \ ++"We [arith] Add watchpoint for expression expr. If no expression is given ++ all watchpoints are deleted. Long command name: watche." ++ return ;; ++ W | wa | wat | watch ) ++ _Dbg_msg \ ++"W [var] Add watchpoint for variable var. If no expression is given ++ all watchpoints are deleted. Long command name: watch." ++ return ;; ++ * ) ++ _Dbg_msg "Undefined command: \"$db_cmd\". Try \"help\"." ++ return ;; ++ esac ++ fi ++ ++ _Dbg_msg 'bashdb commands: ++List/search source lines: Control script execution: ++------------------------- ------------------------- ++ l [start|.] [cnt] List cnt lines T [n] Stack trace ++ from line start s [n] Single step [n times] ++ l sub List source code fn n [n] Next, steps over subs ++ - or . List previous/current line / Repeat last n or s ++ w [line] List around line c [linespec] Continue [to linespec] ++ f filename View source in file L List all breakpoints ++ /pat/ Search forward for pat b linespec Set breakpoint ++ ?pat? Search backward for pat del [n].. or D Delete a/all breaks ++ by entry number ++Debugger controls: skip skip execution of cmd ++------------------------- cl linespec Delete breakpoints by ++ H [num] Show last num commands line spec ++ q [exp] or ^D Quit returning exp R [args] Attempt a restart ++ info [cmd] Get info on cmd. u [n] Go up stack by n or 1. ++ !n or hi n Run debugger history n do [n] Go down stack by n or 1. ++ h or ? [cmd] Get help on command W [var] Add watchpoint. If no ++ info [cmd] Get info on cmd no expr, delete all ++ show [cmd] Show settings We [expr] Add Watchpoint arith ++ expr ++ so file read in dbg commands t Toggle trace ++ en/di n enable/disable brkpt, ++ set x y set a debugger variable watchpoint, or display ++ e bash-cmd evaluate a bash command tb linespec Add one-time break ++ disp expr add a display expr a linespec cmd eval "cmd" at linespec ++ M Show module versions A delete all actions ++ x expr evaluate expression ret jump out of fn or source ++ (via declare, let, eval) finish execute until return ++ deb debug into another cond n exp set breakpoint condition ++ shell script ++ !! cmd [args] execute shell command "cmd" with "args" ++ ++Data Examination: also see e, t, x ++------------------------- ++ p variable Print variable ++ V [[!]pat] List variable(s) matching or not (!) matching pattern pat ++ S [[!]pat] List subroutine names [not] matching pattern pat ++ ++Readline command line editing (emacs/vi mode) is available ++For more help, type h or consult online-documentation.' ++ ++} ++ ++_Dbg_help_set() { ++ local -r set_cmd=$1 ++ local label=$2 ++ local -r subcmds="args annotate basename debugger editing listsize prompt showcommand" ++ ++ if [[ -z $set_cmd ]] ; then ++ local thing ++ for thing in $subcmds ; do ++ _Dbg_help_set $thing 1 ++ done ++ return ++ fi ++ ++ case $set_cmd in ++ ar | arg | args ) ++ [[ -n $label ]] && label='set args -- ' ++ _Dbg_msg \ ++"${label}Set argument list to give program being debugged when it is started. ++Follow this command with any number of args, to be passed to the program." ++ return 0 ++ ;; ++ an | ann | anno | annot | annota | annotat | annotate ) ++ if [[ -n $label ]] ; then ++ label='set annotate -- ' ++ else ++ local post_label=' ++0 == normal; 1 == fullname (for use when running under emacs)' ++ fi ++ _Dbg_msg \ ++"${label}Set annotation level.$post_label" ++ return 0 ++ ;; ++ b | ba | bas | base | basen | basena | basenam | basename ) ++ [[ -n $label ]] && label='set basename -- ' ++ local onoff="off." ++ (( $_Dbg_basename_only != 0 )) && onoff='on.' ++ _Dbg_msg \ ++"${label}Set short filenames (the basename) in debug output is" $onoff ++ return 0 ++ ;; ++ d|de|deb|debu|debug|debugg|debugger|debuggi|debuggin|debugging ) ++ local onoff=${1:-'on'} ++ [[ -n $label ]] && label='set debugger -- ' ++ local onoff="off." ++ (( $_Dbg_debug_debugger )) && onoff='on.' ++ _Dbg_msg \ ++"${label}Set debugging the debugger is" $onoff ++ return 0 ++ ;; ++ e | ed | edi | edit | editi | editin | editing ) ++ [[ -n $label ]] && label='set editing -- ' ++ local onoff="off." ++ (( $_Dbg_edit )) && onoff='on.' ++ _Dbg_msg \ ++"${label}Set editing of command lines as they are typed is" $onoff ++ ;; ++ l | li | lis | list | lists | listsi | listsiz | listsize ) ++ [[ -n $label ]] && label='set listsize -- ' ++ _Dbg_msg \ ++"${label}Set number of source lines bashdb will list by default." ++ ;; ++ p | pr | pro | prom | promp | prompt ) ++ [[ -n $label ]] && label='set prompt -- ' ++ _Dbg_msg \ ++"${label}bashdb's prompt is:\n" \ ++" \"$_Dbg_prompt_str\"." ++ return 0 ++ ;; ++ sho|show|showc|showco|showcom|showcomm|showcomma|showcomman|showcommand ) ++ [[ -n $label ]] && label='set showcommand -- ' ++ _Dbg_msg \ ++"${label}Set showing the command to execute is $_Dbg_show_command." ++ return 0 ++ ;; ++ * ) ++ _Dbg_msg \ ++"There is no \"set $set_cmd\" command." ++ esac ++} ++ ++_Dbg_help_show() { ++ local -r show_cmd=$1 ++ local -r subcmds="args annotate basename debugger commands copying debugger directories listsize prompt warranty" ++ ++ if [[ -z $show_cmd ]] ; then ++ local thing ++ for thing in $subcmds ; do ++ _Dbg_help_show $thing 1 ++ done ++ return ++ fi ++ ++ case $show_cmd in ++ ar | arg | args ) ++ _Dbg_msg \ ++"show args -- Show argument list to give program being debugged when it ++ is started" ++ return 0 ++ ;; ++ an | ann | anno | annot | annota | annotat | annotate ) ++ _Dbg_msg \ ++"show annotate -- Show annotation_level" ++ return 0 ++ ;; ++ b | ba | bas | base | basen | basena | basenam | basename ) ++ _Dbg_msg \ ++"show basename -- Show if we are are to short or long filenames" ++ return 0 ++ ;; ++ com | comm | comma | comman | command | commands ) ++ _Dbg_msg \ ++"show commands -- Show the history of commands you typed" ++ ;; ++ cop | copy| copyi | copyin | copying ) ++ _Dbg_msg \ ++"show copying -- Conditions for redistributing copies of debugger" ++ ;; ++ d|de|deb|debu|debug|debugg|debugger|debuggi|debuggin|debugging ) ++ _Dbg_msg \ ++"show debugger -- Show if we are set to debug the debugger" ++ return 0 ++ ;; ++ di|dir|dire|direc|direct|directo|director|directori|directorie|directories) ++ _Dbg_msg \ ++"show directories -- Show if we are set to debug the debugger" ++ ;; ++ l | li | lis | list | lists | listsi | listsiz | listsize ) ++ _Dbg_msg \ ++"show listsize -- Show number of source lines debugger will list by default" ++ ;; ++ p | pr | pro | prom | promp | prompt ) ++ _Dbg_msg \ ++"show prompt -- Show debugger's prompt" ++ return 0 ++ ;; ++ w | wa | war | warr | warra | warran | warrant | warranty ) ++ _Dbg_msg \ ++"show warranty -- Various kinds of warranty you do not have" ++ return 0 ++ ;; ++ * ) ++ _Dbg_msg \ ++ "Undefined show command: \"$show_cmd\". Try \"help show\"." ++ esac ++} ++ ++_Dbg_help_info() { ++ local -r info_cmd=$1 ++ local label=$2 ++ local -r subcmds="args breakpoints display files functions signals source stack terminal variables warranty" ++ ++ if [[ -z $info_cmd ]] ; then ++ local thing ++ _Dbg_msg \ ++"List of info subcommands: ++" ++ for thing in $subcmds ; do ++ _Dbg_help_info $thing 1 ++ done ++ return ++ fi ++ ++ case $info_cmd in ++ ar | arg | args ) ++ _Dbg_msg \ ++"info args -- \$1, \$2, and so on" ++ return 0 ++ ;; ++ b | br | bre | brea | 'break' | breakp | breakpo | breakpoints | \ ++ w | wa | wat | watc | 'watch' | watchp | watchpo | watchpoints ) ++ _Dbg_msg \ ++"info breakpoints -- Status of user-settable breakpoints" ++ return 0 ++ ;; ++ disp | displ | displa | display ) ++ _Dbg_msg \ ++"info display -- Show all display expressions" ++ return 0 ++ ;; ++ fi | file| files | sources ) ++ _Dbg_msg \ ++"info files -- Source files in the program" ++ return 0 ++ ;; ++ fu | fun| func | funct | functi | functio | function | functions ) ++ _Dbg_msg \ ++"info functions -- All function names" ++ return 0 ++ ;; ++ h | ha | han | hand | handl | handle | \ ++ si | sig | sign | signa | signal | signals ) ++ _Dbg_msg \ ++"info signals -- What debugger does when program gets various signals" ++ return 0 ++ ;; ++ so | sou | sourc | source ) ++ _Dbg_msg \ ++"info source -- Information about the current source file" ++ return 0 ++ ;; ++ st | sta | stac | stack ) ++ _Dbg_msg \ ++"info stack -- Backtrace of the stack" ++ return 0 ++ ;; ++ te | ter | term | termi | termin | termina | terminal | tt | tty ) ++ _Dbg_msg \ ++"info terminal -- Print terminal device" ++ return 0 ++ ;; ++ tr|tra|trac|trace|tracep | tracepo | tracepoi | tracepoint | tracepoints ) ++ _Dbg_msg \ ++"info tracepoints -- Status of tracepoints" ++ return 0 ++ ;; ++ v | va | var | vari | varia | variab | variabl | variable | variables ) ++ _Dbg_msg \ ++"info variables -- All global and static variable names" ++ return 0 ++ ;; ++ w | wa | war | warr | warra | warran | warrant | warranty ) ++ _Dbg_msg \ ++"info warranty -- Various kinds of warranty you do not have" ++ return 0 ++ ;; ++ * ) ++ _Dbg_msg \ ++ "Undefined info command: \"$info_cmd\". Try \"help info\"." ++ esac ++} ++_Dbg_cmd_info() { ++ local -r info_cmd=$1 ++ local -ar subcmds=( args breakpoints display files functions source sources \ ++ stack terminal variables watchpoints ) ++ ++ if [[ $info_cmd != '' ]] ; then ++ shift ++ case $info_cmd in ++ a | ar | arg | args ) ++ local -i i ++ for (( i=1 ; i<=${#ARG[@]} ; i++ )) ; do ++ _Dbg_printf "$%d = %s" $i ${ARG[$i]} ++ done ++ return ++ ;; ++ b | br | bre | brea | 'break' | breakp | breakpo | breakpoints | \ ++ w | wa | wat | watc | 'watch' | watchp | watchpo | watchpoints ) ++ _Dbg_cmd_list_brkpt $* ++ _Dbg_list_watch $* ++ return ++ ;; ++ ++ d | di | dis| disp | displ | displa | display ) ++ _Dbg_cmd_list_display $* ++ return ++ ;; ++ ++ fi | file| files | sources ) ++ for file in ${_Dbg_filenames[@]} ; do ++ (( _Dbg_basename_only )) && file=${file##*/} ++ _Dbg_msg "$file" ; ++ done ++ return ++ ;; ++ ++ fu | fun| func | funct | functi | functio | function | functions ) ++ _Dbg_cmd_list_subroutines $* ++ return ++ ;; ++ ++ h | ha | han | hand | handl | handle | \ ++ si | sig | sign | signa | signal | signals ) ++ _Dbg_info_signals ++ return ++ ;; ++ ++ so | sou | sourc | source ) ++ _Dbg_msg "Current script file is $_cur_source_file" ++ local -i max_line=`_Dbg_get_assoc_scalar_entry "_Dbg_maxline_" $_cur_filevar` ++ _Dbg_msg "Contains $max_line lines." ; ++ return ++ ;; ++ ++ st | sta | stac | stack ) ++ _Dbg_cmd_stack_trace 1 $* ++ return ++ ;; ++ te | ter | term | termi | termin | termina | terminal | tt | tty ) ++ _Dbg_msg "tty: $_Dbg_tty" ++ return; ++ ;; ++ v | va | var | vari | varia | variab | variabl | variable | variables ) ++ _Dbg_cmd_list_variables "$1" ++ return ++ ;; ++ w | wa | war | warr | warra | warran | warrant | warranty ) ++ _Dbg_msg " ++ NO WARRANTY ++ ++ 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY ++FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN ++OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES ++PROVIDE THE PROGRAM \"AS IS\" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED ++OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF ++MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS ++TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE ++PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, ++REPAIR OR CORRECTION. ++ ++ 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING ++WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR ++REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, ++INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING ++OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED ++TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY ++YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER ++PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE ++POSSIBILITY OF SUCH DAMAGES. ++" ++ return ++ ;; ++ *) ++ _Dbg_msg "Unknown subcommand: $info_cmd" ++ esac ++ fi ++ _Dbg_msg "Info subcommands are: ${subcmds[@]}" ++} ++ ++# List signal handlers in effect. ++_Dbg_info_signals() { ++ local -i i=0 ++ local signal_name ++ local handler ++ local stop_flag ++ local print_flag ++ ++ _Dbg_msg "Signal Stop Print Stack Value" ++ _Dbg_printf_nocr "%-12s %-6s %-7s %-9s " EXIT \ ++ ${_Dbg_sig_stop[0]:-nostop} ${_Dbg_sig_print[0]:-noprint} \ ++ ${_Dbg_sig_show_stack[0]:-nostack} ++ ++ # This is a horrible hack, but I can't figure out how to get ++ # trap -p 0 into a variable; handler=`trap -p 0` doesn't work. ++ if [[ -n $_Dbg_tty ]] ; then ++ builtin trap -p 0 >>$_Dbg_tty ++ else ++ builtin trap -p 0 ++ fi ++ ++ while [ 1 ] ; do ++ signal_name=`builtin kill -l $i 2>/dev/null` ++ if (( $? != 0 )) ; then ++ break ++ fi ++ handler=`builtin trap -p $i` ++ if [[ -n $handler ]] ; then ++ _Dbg_printf "%-12s %-6s %-7s %-9s %-6s" $signal_name \ ++ ${_Dbg_sig_stop[$i]:-nostop} ${_Dbg_sig_print[$i]:-noprint} \ ++ ${_Dbg_sig_show_stack[$i]:-nostack} "$handler" ++ fi ++ ((i++)) ++ done ++} ++ ++# This is put at the so we have something at the end to stop at ++# when we debug this. By stopping at the end all of the above functions ++# and variables can be tested. ++typeset -r _Dbg_help_ver=\ ++'$Id: dbg-help.inc,v 1.9 2003/08/02 20:20:40 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-hist.inc cvs/debugger/dbg-hist.inc +--- bash-2.05b/debugger/dbg-hist.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-hist.inc 2003-02-27 08:27:46.000000000 +0100 +@@ -0,0 +1,127 @@ ++# dbg-hist.inc - Bourne Again Shell Debugger history routines ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# Remove the last command from the history list. ++_Dbg_remove_history_item() { ++ _Dbg_hi=${#_Dbg_history[@]}-1 ++ unset _Dbg_history[$_Dbg_hi] ++} ++ ++# Set to rerun history item, or print history if command is of the form ++# !n:p. If command is "history" then $1 is number of history item. ++# the history command index to run is returned or $_Dbg_hi if ++# there's nothing to run. ++# Return value in $history_num ++_Dbg_cmd_history_parse() { ++ ++ history_num=$1 ++ ++ _Dbg_hi=${#_Dbg_history[@]} ++ [[ -z $history_num ]] && let history_num=$_Dbg_hi-1 ++ ++ eval "$_seteglob" ++ if [[ $_Dbg_cmd == h* ]] ; then ++ if [[ $history_num != $int_pat ]] ; then ++ if [[ $history_num == -$int_pat ]] ; then ++ history_num=$_Dbg_hi+$history_num ++ else ++ _Dbg_msg "Invalid history number skipped: $history_num" ++ history_num=-1 ++ fi ++ fi ++ else ++ # Handle ! form. May need to parse number out number and modifier ++ case $_Dbg_cmd in ++ \!\-${int_pat}:p ) ++ local -a word1=($(_Dbg_split "!" $_Dbg_cmd)) ++ local -a word2=($(_Dbg_split ":" ${word1[0]})) ++ local -i num=_Dbg_hi+${word2[0]} ++ _Dbg_cmd_history_list $num $num ++ history_num=-1 ++ ;; ++ [!]${int_pat}:p ) ++ local -a word1=($(_Dbg_split "!" $_Dbg_cmd)) ++ local -a word2=($(_Dbg_split ":" ${word1[0]})) ++ _Dbg_cmd_history_list ${word2[0]} ${word2[0]} ++ history_num=-1 ++ ;; ++ \!\-$int_pat ) ++ local -a word=($(_Dbg_split "!" $_Dbg_cmd)) ++ history_num=$_Dbg_hi+${word[0]} ++ ;; ++ \!$int_pat ) ++ local -a word=($(_Dbg_split "!" $_Dbg_cmd)) ++ history_num=${word[0]} ++ ;; ++ '!' ) ++ if [[ $history_num != $int_pat ]] ; then ++ if [[ $history_num == -$int_pat ]] ; then ++ history_num=$_Dbg_hi+$history_num ++ else ++ _Dbg_msg "Invalid history number skipped: $history_num" ++ history_num=-1 ++ fi ++ fi ++ ;; ++ * ) ++ _Dbg_msg "Invalid history number skipped: $_Dbg_cmd" ++ history_num=-1 ++ esac ++ fi ++ eval "$_resteglob" ++} ++ ++# Print debugger history $1 is where to start or highest number if not given. ++# If $1 is negative, it is how many history items. ++# $2 is where to stop or 0 if not given. ++_Dbg_cmd_history_list() { ++ ++ eval "$_seteglob" ++ if [[ $1 != $int_pat ]] && [[ $1 != -$int_pat ]] && [[ -n $1 ]] ; then ++ _Dbg_msg "Invalid history number: $1" ++ eval "$_resteglob" ++ return 1 ++ fi ++ eval "$_resteglob" ++ ++ _Dbg_hi=${#_Dbg_history[@]} ++ local -i n=${1:-$_Dbg_hi-1} ++ local -i stop=${2:0} ++ local -i i ++ ++ # Were we given a count rather than a starting history number? ++ if (( n<0 )) ; then ++ ((stop=_Dbg_hi+n)) ++ ((n=_Dbg_hi-1)) ++ elif (( n > _Dbg_hi-1 )) ; then ++ ((n=_Dbg_hi-1)) ++ fi ++ ++ for (( i=n ; (( i >= stop && i >= 0 )) ; i-- )) ; do ++ _Dbg_msg "${i}: ${_Dbg_history[$i]}" ++ done ++} ++ ++# This is put at the so we have something at the end when we debug this. ++typeset -r _Dbg_hist_ver=\ ++'$Id: dbg-hist.inc,v 1.2 2003/02/27 07:27:46 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: ** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-init.inc cvs/debugger/dbg-init.inc +--- bash-2.05b/debugger/dbg-init.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-init.inc 2003-05-20 20:02:42.000000000 +0200 +@@ -0,0 +1,107 @@ ++# dbg-init.inc - Bourne Again Shell Debugger Global Variablesa ++ ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# Note: the trend now is to move initializations which are generally ++# used in only one sub-part (e.g. variables for break/watch/actions) to ++# the corresponding file. ++ ++typeset _cur_fn # current function of debugged program ++typeset -i _cur_line # current line number of debugged program ++typeset -i _trace=0 # init execution trace flag to on ++typeset -i _Dbg_steps=1 # No. of statements to run after setting trap ++ ++if [[ -n $_Dbg_script ]] ; then ++ _Dbg_steps=4 ++else ++ . ${_Dbg_libdir}/dbg-pre.inc ++ typeset -r _Dbg_source_file=$(_Dbg_expand_filename $0) ++ typeset -i _Dbg_n=$# ++ typeset -i _Dbg_i ++ typeset -i _Dbg_basename_only=${BASHDB_BASENAME_ONLY:-0} ++ declare -a _Dbg_script_args ++ for (( _Dbg_i=0; _Dbg_i<_Dbg_n ; _Dbg_i++ )) ; do ++ _Dbg_script_args[$_Dbg_i]=$1 ++ shift ++ done ++ # Now that we've trashed the script parameters above, restore them. ++ _Dbg_set_str="set --" ++ for (( _Dbg_i=0; _Dbg_i<_Dbg_n ; _Dbg_i++ )) ; do ++ _Dbg_set_str="$_Dbg_set_str \"${_Dbg_script_args[$_Dbg_i]}\"" ++ done ++ eval $_Dbg_set_str ++fi ++ ++# Current next history entry to store into. ++typeset -i _Dbg_hi=0 ++ ++# How many lines in a listing? ++typeset -i _Dbg_listsize=10 ++ ++# Annotation level. ++typeset -i _Dbg_annotate=0 ++ ++typeset -i _Dbg_need_input=1 # True if we need to reassign input. ++ ++typeset last_next_step_cmd='s' # Default is step. ++typeset _Dbg_last_print='' # expression on last print command ++typeset _Dbg_last_printe='' # expression on last print expression command ++ ++ ++# strings to save and restore the setting of `extglob' in debugger functions ++# that need it ++typeset -r _seteglob='local __eopt=-u ; shopt -q extglob && __eopt=-s ; shopt -s extglob' ++typeset -r _resteglob='shopt $__eopt extglob' ++ ++typeset -r int_pat="[0-9]*([0-9])" ++ ++# If called from bashdb script rather than via "bash --debugger", skip ++# over some initial setup commands, like the initial "source" function ++# of debugged shell script. ++ ++# Set tty to use for output. ++if [[ -z $_Dbg_tty ]] ; then ++ typeset _Dbg_tty; ++ _Dbg_tty=$(tty) ++ [[ $? != 0 ]] && _Dbg_tty='' ++fi ++ ++# Equivalent to basename $0 -- the short program name ++typeset _Dbg_pname=${0##*/} ++ ++# $_Dbg_tmpdir could have been set by bashdb script rather than ++# bash --debugger ++[[ -z $_Dbg_tmpdir ]] && declare _Dbg_tmpdir=/tmp ++ ++# Known normal IFS consisting of a space, tab and newline ++typeset -r _Dbg_space_IFS=" ++" ++ ++# If BASHDB_QUIT_LEVELS is set to a positive number, this is the number ++# of levels (subshell or shell nestings) or we should exit out of. ++[ "X$BASHDB_QUIT_LEVELS" = "X" ] && BASHDB_QUIT_LEVELS=0 ++ ++# This is put at the so we have something at the end to stop at ++# when we debug this. By stopping at the end all of the above functions ++# and variables can be tested. ++typeset -r _Dbg_init_ver=\ ++'$Id: dbg-init.inc,v 1.6 2003/05/20 18:02:42 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-io.inc cvs/debugger/dbg-io.inc +--- bash-2.05b/debugger/dbg-io.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-io.inc 2003-08-02 21:54:47.000000000 +0200 +@@ -0,0 +1,143 @@ ++# dbg-io.inc - Bourne Again Shell Debugger Input/Output routines ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# ==================== VARIABLES ======================================= ++ ++# _source_mungedfilename is array which contains source_lines. ++# _read_mungedfilename is array which is 1 filename has been read in. ++ ++# Filename that's used when no source file is around. In particular ++# if bash --debugger -c 'string' was used to invoke us. ++typeset -r _Dbg_bogus_file='*BOGUS*' ++ ++typeset -a _Dbg_filenames # names of all source files read ++typeset _cur_source_file # current source file of debugged program ++typeset _cur_filevar # source file mangled so it can be used in a variable ++typeset -i _curline # current line number of debugged program ++ ++# ===================== FUNCTIONS ======================================= ++ ++# print message to output device ++_Dbg_msg() { ++ if [[ -n $_Dbg_tty ]] ; then ++ builtin echo -e "$@" >>$_Dbg_tty ++ else ++ builtin echo -e "$@" ++ fi ++} ++ ++# print message to output device without a carriage return at the end ++_Dbg_msg_nocr() { ++ if [[ -n $_Dbg_tty ]] ; then ++ builtin echo -n -e "$@" >>$_Dbg_tty ++ else ++ builtin echo -n -e "$@" ++ fi ++} ++ ++# print message to output device ++_Dbg_printf() { ++ local format=$1 ++ shift ++ if [[ -n $_Dbg_tty ]] ; then ++ builtin printf "$format" "$@" >>$_Dbg_tty ++ else ++ builtin printf "$format" "$@" ++ fi ++ _Dbg_msg "" ++} ++ ++# print message to output device without a carriage return at the end ++_Dbg_printf_nocr() { ++ local format=$1 ++ shift ++ if [[ -n $_Dbg_tty ]] ; then ++ builtin printf "$format" "$@" >>$_Dbg_tty ++ else ++ builtin printf "$format" "$@" ++ fi ++} ++ ++# Return text for source line for line $1 of filename $2 in variable ++# $source_line. The hope is that this has been declared "local" in the ++# caller. ++ ++# If $2 is omitted, # use _cur_source_file, if $1 is omitted use _curline. ++_Dbg_get_source_line() { ++ local lineno=${1:-$_curline} ++ local filename=${2:-$_cur_source_file} ++ local filevar=`_Dbg_file2var $filename` ++ local is_read=`_Dbg_get_assoc_scalar_entry "_Dbg_read_" $filevar` ++ [ $is_read ] || _Dbg_readin $filename ++ ++ source_line=`_Dbg_get_assoc_array_entry _Dbg_source_${filevar} $lineno` ++} ++ ++# Read $1 into _source_$1 array. ++# Variable _read_$1 will be set to 1 to note that the file has been read. ++ ++_Dbg_readin() { ++ # set -xv ++ local filename=${1:-$_cur_source_file} ++ ++ local -i i=0 ++ local filevar ++ ++ if [ $filename == $_Dbg_bogus_file ] ; then ++ filevar='ABOGUSA' ++ local cmd="_Dbg_source_${filevar}[1]=\"$BASH_EXECUTION_STRING\"" ++ eval $cmd ++ i=1 ++ else ++ local fullname=$(_Dbg_resolve_expand_filename $filename) ++ filevar=`_Dbg_file2var $filename` ++ if [[ -r $fullname ]] ; then ++ for (( i=1; 1 ; i++ )) ; do ++ local source_entry="_Dbg_source_${filevar}[$i]" ++ local readline_cmd="read -r $source_entry; rc=\$?"; ++ local -i rc=1 ++ if (( i % 1000 == 0 )) ; then ++ (( i==1000 )) && _Dbg_msg_nocr "Reading $filename " ++ _Dbg_msg_nocr "${i}... " ++ fi ++ eval $readline_cmd ++ if [[ $rc != 0 ]] ; then ++ break; ++ fi ++ done < $fullname ++ fi ++ fi ++ ++ (( i >= 1000 )) && _Dbg_msg "done." ++ ++ _Dbg_set_assoc_scalar_entry "_Dbg_read_" $filevar 1 ++ _Dbg_set_assoc_scalar_entry "_Dbg_maxline_" $filevar $i ++ ++ # Add $filename to list of all filenames ++ _Dbg_filenames[${#_Dbg_filenames[@]}]=$fullname; ++ # set +xv ++} ++ ++# This is put at the so we have something at the end when we debug this. ++typeset -r _Dbg_io_ver=\ ++'$Id: dbg-io.inc,v 1.6 2003/08/02 19:54:47 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-list.inc cvs/debugger/dbg-list.inc +--- bash-2.05b/debugger/dbg-list.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-list.inc 2003-08-02 21:54:47.000000000 +0200 +@@ -0,0 +1,230 @@ ++# bashdb-list.inc - Bourne Again Shell Debugger list/search commands ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# List search commands/routines ++ ++# Last search pattern used. ++typeset _Dbg_last_search_pat ++ ++# current line to be listed ++typeset -i _Dbg_listline ++ ++# Print source line in standard format for line $1 of filename $2. If ++# $2 is omitted, use _cur_source_file, if $1 is omitted use _curline. ++ ++_Dbg_print_source_line() { ++ local line_number=${1:-$_curline} ++ local filename=${2:-$_cur_source_file} ++ local source_line ++ _Dbg_get_source_line $line_number $filename ++ filename=$(_Dbg_adjust_filename "$filename") ++ _Dbg_msg "(${filename}:${line_number}): ++${line_number}:\t${source_line}" ++ ++ # If we are at the same place in the file but the command has changed, ++ # then we have multiple commands on the line. So print which one we are ++ # currently at. ++ if [[ $_Dbg_show_command == "on" ]] ; then ++ _Dbg_msg "$_Dbg_bash_command" ++ elif [[ $_Dbg_last_curline == $_curline ]] \ ++ && [[ $_Dbg_last_source_file == $_cur_source_file ]] \ ++ && [[ $_Dbg_last_bash_command != $_Dbg_bash_command \ ++ && $_Dbg_show_command == "auto" ]] ; then ++ _Dbg_msg "$_Dbg_bash_command" ++ fi ++} ++ ++ ++# l [start|.] [cnt] List cnt lines from line start. ++# l sub List source code fn ++_Dbg_cmd_list() { ++ if [[ -n $1 ]] ; then ++ first_arg=$1 ++ shift ++ else ++ first_arg=. ++ fi ++ ++ if [ $first_arg == '.' ] ; then ++ _Dbg_list $_cur_source_file $* ++ return ++ fi ++ ++ local -a word=($(_Dbg_parse_linespec "$first_arg")) ++ if [[ ${#word[@]} == 0 ]] ; then ++ _Dbg_msg "Invalid linespec $n" ++ return ++ fi ++ ++ local -i line_number=${word[0]} ++ local filename=${word[1]} ++ local full_filename=`_Dbg_is_file $filename` ++ if [[ -n $full_filename ]] ; then ++ (( $line_number == 0 )) && line_number=1 ++ _Dbg_check_line $line_number "$full_filename" ++ (( $? == 0 )) && \ ++ _Dbg_list "$full_filename" "$line_number" $* ++ else ++ _Dbg_msg "File $filename not found in read-in files." ++ _Dbg_msg "See 'info files' for a list of known files." ++ fi ++} ++ ++# /search/ ++_Dbg_cmd_search_back() { ++ local delim_search_pat=$1 ++ if [[ -z "$1" ]] ; then ++ _Dbg_msg "Need a search pattern" ++ return 1 ++ fi ++ shift ++ ++ case "$delim_search_pat" in ++ [?] ) ++ ;; ++ [?]* ) ++ local -a word=($(_Dbg_split '?' $delim_search_pat)) ++ _Dbg_last_search_pat=${word[0]} ++ ;; ++ # Error ++ * ) ++ _Dbg_last_search_pat=$delim_search_pat ++ esac ++ local -i i ++ local -i max_line=`_Dbg_get_assoc_scalar_entry "_Dbg_maxline_" $_cur_filevar` ++ for (( i=_Dbg_listline-1; i > 0 ; i-- )) ; do ++ local source_line ++ _Dbg_get_source_line $i ++ eval "$_seteglob" ++ if [[ $source_line == *$_Dbg_last_search_pat* ]] ; then ++ eval "$_resteglob" ++ _Dbg_cmd_list $i 1 ++ _Dbg_listline=$i ++ return 0 ++ fi ++ eval "$_resteglob" ++ done ++ _Dbg_msg "search pattern: $_Dbg_last_search_pat not found." ++ return 1 ++ ++} ++ ++# /search/ ++_Dbg_cmd_search() { ++ local delim_search_pat=${1} ++ if [[ -z "$1" ]] ; then ++ _Dbg_msg "Need a search pattern" ++ return 1 ++ fi ++ shift ++ local search_pat ++ case "$delim_search_pat" in ++ / ) ++ ;; ++ /* ) ++ local -a word=($(_Dbg_split "/" $delim_search_pat)) ++ _Dbg_last_search_pat=${word[0]} ++ ;; ++ * ) ++ _Dbg_last_search_pat=$delim_search_pat ++ esac ++ local -i i ++ local -i max_line=`_Dbg_get_assoc_scalar_entry "_Dbg_maxline_" $_cur_filevar` ++ for (( i=_Dbg_listline+1; i < max_line ; i++ )) ; do ++ local source_line ++ _Dbg_get_source_line $i ++ eval "$_seteglob" ++ if [[ $source_line == *$_Dbg_last_search_pat* ]] ; then ++ eval "$_resteglob" ++ _Dbg_cmd_list $i 1 ++ _Dbg_listline=$i ++ return 0 ++ fi ++ eval "$_resteglob" ++ done ++ _Dbg_msg "search pattern: $_Dbg_last_search_pat not found." ++ return 1 ++ ++} ++ ++# S [[!]pat] List Subroutine names [not] matching a pattern ++# Pass along whether or not to print "system" functions? ++_Dbg_cmd_list_subroutines() { ++ ++ local pat=$1 ++ ++ local -a fns_a=(`_Dbg_get_functions 0 "$pat"`) ++ local -i i ++ for (( i=0; (( i < ${#fns_a[@]} )) ; i++ )) ; do ++ _Dbg_msg ${fns_a[$i]} ++ done ++} ++ ++# list $3 lines starting at line $2 of file $1. If $1 is '', use ++# $_cur_source_file value. If $3 is ommited, print $_Dbg_listsize ++# lines. if $2 is omitted, use global variable $_curline. ++ ++_Dbg_list() { ++ local filename=${1:-$_cur_source_file} ++ ++ if [[ $2 = . ]]; then ++ _Dbg_listline=$_curline ++ elif [[ -n $2 ]] ; then ++ _Dbg_listline=$2 ++ fi ++ (( _Dbg_listline==0 && _Dbg_listline++)) ++ ++ local -ir cnt=${3:-$_Dbg_listsize} ++ local -ir n=$((_Dbg_listline+cnt-1)) ++ ++ local source_line ++ ++ local filevar=`_Dbg_file2var $filename` ++ local is_read ++ is_read=`_Dbg_get_assoc_scalar_entry "_Dbg_read_" $filevar` ++ [ $is_read ] || _Dbg_readin $filename ++ local -i max_line=`_Dbg_get_assoc_scalar_entry "_Dbg_maxline_" $filevar` ++ # echo "debug: -- max_line: $max_line --" ++ ++ if (( _Dbg_listline > max_line )) ; then ++ _Dbg_msg \ ++ "Line number $_Dbg_listline out of range;" \ ++ "$filename has $max_line lines." ++ return 1 ++ fi ++ ++ for (( ; (( _Dbg_listline <= n && _Dbg_listline <= max_line )) \ ++ ; _Dbg_listline++ )) ; do ++ local prefix=" " ++ local source_line ++ _Dbg_get_source_line $_Dbg_listline $filename ++ ++ (( _Dbg_listline == _curline )) \ ++ && [[ $filename == $_cur_source_file ]] && prefix="==>" ++ _Dbg_printf "%3d:%s%s" $_Dbg_listline "$prefix" "$source_line" ++ done ++ (( _Dbg_listline > max_line && _Dbg_listline-- )) ++ return 0 ++} ++ ++# This is put at the so we have something at the end when we debug this. ++_Dbg_list_ver='$Id: dbg-list.inc,v 1.5 2003/08/02 19:54:47 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-main.inc.in cvs/debugger/dbg-main.inc.in +--- bash-2.05b/debugger/dbg-main.inc.in 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-main.inc.in 2003-08-25 18:34:35.000000000 +0200 +@@ -0,0 +1,55 @@ ++# dbg-main.inc - Bourne Again Shell Debugger Main Include ++ ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++typeset -r _Dbg_main_ver=\ ++'$Id: dbg-main.inc.in,v 1.1 2003/08/25 16:34:35 snikkt Exp $' ++ ++# Are we using a debugger-enabled bash? If not let's stop right here. ++if [[ -z "${BASH_SOURCE[0]}" ]] ; then ++ echo "Sorry, you need to use a debugger-enabled version of bash." 2>&1 ++ exit 2 ++fi ++ ++# $_Dbg_libdir could have been set by bashdb script. ++# If not, set if "bash --debugger" ++[ -z $_Dbg_libdir ] && typeset _Dbg_libdir=@PKGDATADIR@ ++ ++. ${_Dbg_libdir}/dbg-init.inc # Initializations - put sooner ++. ${_Dbg_libdir}/dbg-brk.inc # Breakpoints, watchpoints, actions ++. ${_Dbg_libdir}/dbg-file.inc # File functions ++. ${_Dbg_libdir}/dbg-fns.inc # Miscellaneous helper functions ++. ${_Dbg_libdir}/dbg-help.inc # Help routines ++. ${_Dbg_libdir}/dbg-hist.inc # History routines ++. ${_Dbg_libdir}/dbg-io.inc # I/O routines - has to come before ++ # cmds ++. ${_Dbg_libdir}/dbg-cmds.inc # Debugger commands ++. ${_Dbg_libdir}/dbg-list.inc # List/search commands ++. ${_Dbg_libdir}/dbg-stack.inc # Call-stack routines ++. ${_Dbg_libdir}/dbg-set.inc # set/show routines ++. ${_Dbg_libdir}/dbg-sig.inc # signal-handling routines ++ ++[ -n "$BASHDB_RESTART_FILE" ] \ ++ && [ -r "$BASHDB_RESTART_FILE" ] && source $BASHDB_RESTART_FILE ++ ++trap '_Dbg_debug_trap_handler 0 "$BASH_COMMAND" "$@"' DEBUG ++# echo '*** we were here***' # uncomment to test that we were run. ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-pre.inc.in cvs/debugger/dbg-pre.inc.in +--- bash-2.05b/debugger/dbg-pre.inc.in 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-pre.inc.in 2003-03-27 06:34:12.000000000 +0100 +@@ -0,0 +1,79 @@ ++# dbg-pre.inc - Bourne Again Shell Debugger Global Variables ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# We put definiitions common to both the script debugger and bash ++# --debugger. In contrast to other routines this is sourced early -- ++# before most of the bashdb script is run. The other routines are ++# done near the end of the bashdb script. In this way the script can ++# has access to definitions that --debugger has without duplicating code. ++ ++# Note: the trend now is to move initializations which are generally ++# used in only one sub-part (e.g. variables for break/watch/actions) to ++# the corresponding file. ++ ++typeset -r _Dbg_release='bash-@BASHVERS@-@RELSTATUS@' ++ ++# This is put at the so we have something at the end to stop at ++# when we debug this. By stopping at the end all of the above functions ++# and variables can be tested. ++typeset -r _Dbg_pre_ver=\ ++'$Id: dbg-pre.inc.in,v 1.3 2003/03/27 05:34:12 rockyb Exp $' ++ ++# Expand filename given as $1. ++# we echo the expanded name or return $1 unchanged if a bad filename. ++# Return is 0 if good or 1 if bad. ++# File globbing is handled. ++# Note we don't check that the file exists, just that the format is ++# valid; we do check that we can "search" the directory implied in the ++# filename. ++ ++_Dbg_expand_filename() { ++ local -r filename="$1" ++ ++ # Break out basename and dirname ++ local -r basename=${filename##*/} ++ local dirname=${filename%/*} ++ ++ # No slash given in filename? Then use . for dirname ++ [[ $dirname == $basename ]] && dirname="." ++ ++ # Dirname is ''? Then use / for dirname ++ dirname=${dirname:-/} ++ ++ # Handle tilde expansion in dirname ++ local glob_cmd="dirname=`expr $dirname`" ++ eval $glob_cmd 2>/dev/null ++ ++ local long_path; ++ long_path=`(cd $dirname && pwd) 2>/dev/null` ++ if (( $? == 0 )) ; then ++ if [[ $long_path == '/' ]] ; then ++ echo "/$basename" ++ else ++ echo "$long_path/$basename" ++ fi ++ return 0 ++ else ++ echo $filename ++ return 1 ++ fi ++} ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-set-d-vars.inc cvs/debugger/dbg-set-d-vars.inc +--- bash-2.05b/debugger/dbg-set-d-vars.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-set-d-vars.inc 2003-02-27 08:27:46.000000000 +0100 +@@ -0,0 +1,34 @@ ++#$Id: dbg-set-d-vars.inc,v 1.2 2003/02/27 07:27:46 rockyb Exp $ ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# set dollar variables ($1, $2, ... $?) ++# to their values in the debugged environment before we entered the debugger. ++ ++local _Dbg_set_str="set --" ++local -i _Dbg__i ++for (( _Dbg__i=1 ; _Dbg__i<=${#_Dbg_arg[@]}; _Dbg__i++ )) ; do ++ local dq_argi=$(_Dbg_esc_dq "${_Dbg_arg[$_Dbg__i]}") ++ _Dbg_set_str="$_Dbg_set_str \"$dq_argi\"" ++done ++eval $_Dbg_set_str ++ ++_Dbg_set_dol_q ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-set.inc cvs/debugger/dbg-set.inc +--- bash-2.05b/debugger/dbg-set.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-set.inc 2003-03-22 01:16:31.000000000 +0100 +@@ -0,0 +1,525 @@ ++# dbg-set.inc - Bourne Again Shell Debugger Set Routines ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# Sets whether or not to display command to be executed in debugger prompt. ++# If yes, always show. If auto, show only if the same line is to be run ++# but the command is different. ++typeset _Dbg_show_command="auto" ++ ++_Dbg_cmd_set() { ++ local set_cmd=$1 ++ if [[ $set_cmd == '' ]] ; then ++ _Dbg_msg "Argument required (expression to compute)." ++ return; ++ fi ++ shift ++ case $set_cmd in ++ ar | arg | args ) ++ # We use the loop below rather than _Dbg_set_args="(@)" because ++ # we want to preserve embedded blanks in the arguments. ++ _Dbg_script_args=() ++ local -i i ++ local -i n=$# ++ for (( i=0; i 1 )) ; then ++ _Dbg_msg "annotation level must be 0 or 1" ++ else ++ _Dbg_write_journal_eval "_Dbg_annotate=$1" ++ fi ++ else ++ eval "$_resteglob" ++ _Dbg_msg "Integer argument expected; got: $1" ++ return 1 ++ fi ++ eval "$_resteglob" ++ return 0 ++ ;; ++ b | ba | bas | base | basen | basena | basenam | basename ) ++ local onoff=${1:-'off'} ++ case $onoff in ++ on ) ++ _Dbg_write_journal_eval "_Dbg_basename_only=1" ++ ;; ++ off ) ++ _Dbg_write_journal_eval "_Dbg_basename_only=0" ++ ;; ++ * ) ++ _Dbg_msg "\"on\" or \"off\" expected." ++ esac ++ ;; ++ e | ed | edi | edit | editi | editin | editing ) ++ local onoff=${1:-'on'} ++ case $onoff in ++ on ) ++ _Dbg_edit='-e' ++ ;; ++ off ) ++ _Dbg_edit='' ++ ;; ++ * ) ++ _Dbg_msg "\"on\" or \"off\" expected." ++ esac ++ ;; ++ d|de|deb|debu|debug|debugg|debugger|debuggi|debuggin|debugging ) ++ local onoff=${1:-'on'} ++ case $onoff in ++ on ) ++ _Dbg_write_journal_eval "_Dbg_debug_debugger=1" ++ ;; ++ off ) ++ _Dbg_write_journal_eval "_Dbg_debug_debugger=0" ++ ;; ++ * ) ++ _Dbg_msg "\"on\" or \"off\" expected." ++ esac ++ ;; ++ li | lis | list | lists | listsi | listsiz | listsize ) ++ eval "$_seteglob" ++ if [[ $1 == $int_pat ]] ; then ++ _Dbg_write_journal_eval "_Dbg_listsize=$1" ++ else ++ eval "$_resteglob" ++ _Dbg_msg "Integer argument expected; got: $1" ++ return 1 ++ fi ++ eval "$_resteglob" ++ return 0 ++ ;; ++ p | pr | pro | prom | promp | prompt ) ++ _Dbg_prompt_str="$1" ++ ;; ++ sho|show|showc|showco|showcom|showcomm|showcomma|showcomman|showcommand ) ++ case $1 in ++ on | off | auto ) ++ _Dbg_write_journal_eval "_Dbg_show_command=$1" ++ ;; ++ * ) ++ _Dbg_msg "\"on\", \"off\" or \"auto\" expected." ++ esac ++ return 0 ++ ;; ++ *) ++ _Dbg_msg "Unknown subcommand: $set_cmd" ++ return 1 ++ esac ++} ++ ++_Dbg_cmd_show() { ++ local show_cmd=$1 ++ local label=$2 ++ ++ # Warranty, copying and directories are omitted below. ++ local -r subcmds="args basename debugger editing listsize prompt" ++ ++ if [[ -z $show_cmd ]] ; then ++ local thing ++ for thing in $subcmds ; do ++ _Dbg_cmd_show $thing 1 ++ done ++ return ++ fi ++ ++ case $show_cmd in ++ ar | arg | args ) ++ [[ -n $label ]] && label='args: ' ++ _Dbg_msg \ ++"${label}Argument list to give script when debugged program starts is:\n" \ ++" \"${_Dbg_script_args[@]}\"." ++ return 0 ++ ;; ++ an | ann | anno | annot | annota | annotat | annotate ) ++ [[ -n $label ]] && label='annotate: ' ++ _Dbg_msg \ ++"${label}Annotation_level is" \ ++ "$_Dbg_annotate." ++ return 0 ++ ;; ++ b | ba | bas | base | basen | basena | basenam | basename ) ++ [[ -n $label ]] && label='basename: ' ++ local onoff="off." ++ (( $_Dbg_basename_only != 0 )) && onoff='on.' ++ _Dbg_msg \ ++"${label}Show short filenames (the basename) in debug output is" $onoff ++ return 0 ++ ;; ++ com | comm | comma | comman | command | commands ) ++ local -i hi_start=_Dbg_hi-1 ++ local -i hi_stop=hi_start-10 ++ _Dbg_cmd_history_list $hi_start $hi_stop ++ return 0 ++ ;; ++ cop | copy| copyi | copyin | copying ) ++ _Dbg_msg \ ++" ++ GNU GENERAL PUBLIC LICENSE ++ Version 2, June 1991 ++ ++ Copyright (C) 1989, 1991 Free Software Foundation, Inc. ++ 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA ++ Everyone is permitted to copy and distribute verbatim copies ++ of this license document, but changing it is not allowed. ++ ++ Preamble ++ ++ The licenses for most software are designed to take away your ++freedom to share and change it. By contrast, the GNU General Public ++License is intended to guarantee your freedom to share and change free ++software--to make sure the software is free for all its users. This ++General Public License applies to most of the Free Software ++Foundation's software and to any other program whose authors commit to ++using it. (Some other Free Software Foundation software is covered by ++the GNU Library General Public License instead.) You can apply it to ++your programs, too. ++ ++ When we speak of free software, we are referring to freedom, not ++price. Our General Public Licenses are designed to make sure that you ++have the freedom to distribute copies of free software (and charge for ++this service if you wish), that you receive source code or can get it ++if you want it, that you can change the software or use pieces of it ++in new free programs; and that you know you can do these things. ++ ++ To protect your rights, we need to make restrictions that forbid ++anyone to deny you these rights or to ask you to surrender the rights. ++These restrictions translate to certain responsibilities for you if you ++distribute copies of the software, or if you modify it. ++ ++ For example, if you distribute copies of such a program, whether ++gratis or for a fee, you must give the recipients all the rights that ++you have. You must make sure that they, too, receive or can get the ++source code. And you must show them these terms so they know their ++rights. ++ ++ We protect your rights with two steps: (1) copyright the software, and ++(2) offer you this license which gives you legal permission to copy, ++distribute and/or modify the software. ++ ++ Also, for each author's protection and ours, we want to make certain ++that everyone understands that there is no warranty for this free ++software. If the software is modified by someone else and passed on, we ++want its recipients to know that what they have is not the original, so ++that any problems introduced by others will not reflect on the original ++authors' reputations. ++ ++ Finally, any free program is threatened constantly by software ++patents. We wish to avoid the danger that redistributors of a free ++program will individually obtain patent licenses, in effect making the ++program proprietary. To prevent this, we have made it clear that any ++patent must be licensed for everyone's free use or not licensed at all. ++ ++ The precise terms and conditions for copying, distribution and ++modification follow. ++ ++ GNU GENERAL PUBLIC LICENSE ++ TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION ++ ++ 0. This License applies to any program or other work which contains ++a notice placed by the copyright holder saying it may be distributed ++under the terms of this General Public License. The \"Program\", below, ++refers to any such program or work, and a \"work based on the Program\" ++means either the Program or any derivative work under copyright law: ++that is to say, a work containing the Program or a portion of it, ++either verbatim or with modifications and/or translated into another ++language. (Hereinafter, translation is included without limitation in ++the term \"modification\".) Each licensee is addressed as \"you\". ++ ++Activities other than copying, distribution and modification are not ++covered by this License; they are outside its scope. The act of ++running the Program is not restricted, and the output from the Program ++is covered only if its contents constitute a work based on the ++Program (independent of having been made by running the Program). ++Whether that is true depends on what the Program does. ++ ++ 1. You may copy and distribute verbatim copies of the Program's ++source code as you receive it, in any medium, provided that you ++conspicuously and appropriately publish on each copy an appropriate ++copyright notice and disclaimer of warranty; keep intact all the ++notices that refer to this License and to the absence of any warranty; ++and give any other recipients of the Program a copy of this License ++along with the Program. ++ ++You may charge a fee for the physical act of transferring a copy, and ++you may at your option offer warranty protection in exchange for a fee. ++ ++ 2. You may modify your copy or copies of the Program or any portion ++of it, thus forming a work based on the Program, and copy and ++distribute such modifications or work under the terms of Section 1 ++above, provided that you also meet all of these conditions: ++ ++ a) You must cause the modified files to carry prominent notices ++ stating that you changed the files and the date of any change. ++ ++ b) You must cause any work that you distribute or publish, that in ++ whole or in part contains or is derived from the Program or any ++ part thereof, to be licensed as a whole at no charge to all third ++ parties under the terms of this License. ++ ++ c) If the modified program normally reads commands interactively ++ when run, you must cause it, when started running for such ++ interactive use in the most ordinary way, to print or display an ++ announcement including an appropriate copyright notice and a ++ notice that there is no warranty (or else, saying that you provide ++ a warranty) and that users may redistribute the program under ++ these conditions, and telling the user how to view a copy of this ++ License. (Exception: if the Program itself is interactive but ++ does not normally print such an announcement, your work based on ++ the Program is not required to print an announcement.) ++ ++These requirements apply to the modified work as a whole. If ++identifiable sections of that work are not derived from the Program, ++and can be reasonably considered independent and separate works in ++themselves, then this License, and its terms, do not apply to those ++sections when you distribute them as separate works. But when you ++distribute the same sections as part of a whole which is a work based ++on the Program, the distribution of the whole must be on the terms of ++this License, whose permissions for other licensees extend to the ++entire whole, and thus to each and every part regardless of who wrote it. ++ ++Thus, it is not the intent of this section to claim rights or contest ++your rights to work written entirely by you; rather, the intent is to ++exercise the right to control the distribution of derivative or ++collective works based on the Program. ++ ++In addition, mere aggregation of another work not based on the Program ++with the Program (or with a work based on the Program) on a volume of ++a storage or distribution medium does not bring the other work under ++the scope of this License. ++ ++ 3. You may copy and distribute the Program (or a work based on it, ++under Section 2) in object code or executable form under the terms of ++Sections 1 and 2 above provided that you also do one of the following: ++ ++ a) Accompany it with the complete corresponding machine-readable ++ source code, which must be distributed under the terms of Sections ++ 1 and 2 above on a medium customarily used for software interchange; or, ++ ++ b) Accompany it with a written offer, valid for at least three ++ years, to give any third party, for a charge no more than your ++ cost of physically performing source distribution, a complete ++ machine-readable copy of the corresponding source code, to be ++ distributed under the terms of Sections 1 and 2 above on a medium ++ customarily used for software interchange; or, ++ ++ c) Accompany it with the information you received as to the offer ++ to distribute corresponding source code. (This alternative is ++ allowed only for noncommercial distribution and only if you ++ received the program in object code or executable form with such ++ an offer, in accord with Subsection b above.) ++ ++The source code for a work means the preferred form of the work for ++making modifications to it. For an executable work, complete source ++code means all the source code for all modules it contains, plus any ++associated interface definition files, plus the scripts used to ++control compilation and installation of the executable. However, as a ++special exception, the source code distributed need not include ++anything that is normally distributed (in either source or binary ++form) with the major components (compiler, kernel, and so on) of the ++operating system on which the executable runs, unless that component ++itself accompanies the executable. ++ ++If distribution of executable or object code is made by offering ++access to copy from a designated place, then offering equivalent ++access to copy the source code from the same place counts as ++distribution of the source code, even though third parties are not ++compelled to copy the source along with the object code. ++ ++ 4. You may not copy, modify, sublicense, or distribute the Program ++except as expressly provided under this License. Any attempt ++otherwise to copy, modify, sublicense or distribute the Program is ++void, and will automatically terminate your rights under this License. ++However, parties who have received copies, or rights, from you under ++this License will not have their licenses terminated so long as such ++parties remain in full compliance. ++ ++ 5. You are not required to accept this License, since you have not ++signed it. However, nothing else grants you permission to modify or ++distribute the Program or its derivative works. These actions are ++prohibited by law if you do not accept this License. Therefore, by ++modifying or distributing the Program (or any work based on the ++Program), you indicate your acceptance of this License to do so, and ++all its terms and conditions for copying, distributing or modifying ++the Program or works based on it. ++ ++ 6. Each time you redistribute the Program (or any work based on the ++Program), the recipient automatically receives a license from the ++original licensor to copy, distribute or modify the Program subject to ++these terms and conditions. You may not impose any further ++restrictions on the recipients' exercise of the rights granted herein. ++You are not responsible for enforcing compliance by third parties to ++this License. ++ ++ 7. If, as a consequence of a court judgment or allegation of patent ++infringement or for any other reason (not limited to patent issues), ++conditions are imposed on you (whether by court order, agreement or ++otherwise) that contradict the conditions of this License, they do not ++excuse you from the conditions of this License. If you cannot ++distribute so as to satisfy simultaneously your obligations under this ++License and any other pertinent obligations, then as a consequence you ++may not distribute the Program at all. For example, if a patent ++license would not permit royalty-free redistribution of the Program by ++all those who receive copies directly or indirectly through you, then ++the only way you could satisfy both it and this License would be to ++refrain entirely from distribution of the Program. ++ ++If any portion of this section is held invalid or unenforceable under ++any particular circumstance, the balance of the section is intended to ++apply and the section as a whole is intended to apply in other ++circumstances. ++ ++It is not the purpose of this section to induce you to infringe any ++patents or other property right claims or to contest validity of any ++such claims; this section has the sole purpose of protecting the ++integrity of the free software distribution system, which is ++implemented by public license practices. Many people have made ++generous contributions to the wide range of software distributed ++through that system in reliance on consistent application of that ++system; it is up to the author/donor to decide if he or she is willing ++to distribute software through any other system and a licensee cannot ++impose that choice. ++ ++This section is intended to make thoroughly clear what is believed to ++be a consequence of the rest of this License. ++ ++ 8. If the distribution and/or use of the Program is restricted in ++certain countries either by patents or by copyrighted interfaces, the ++original copyright holder who places the Program under this License ++may add an explicit geographical distribution limitation excluding ++those countries, so that distribution is permitted only in or among ++countries not thus excluded. In such case, this License incorporates ++the limitation as if written in the body of this License. ++ ++ 9. The Free Software Foundation may publish revised and/or new versions ++of the General Public License from time to time. Such new versions will ++be similar in spirit to the present version, but may differ in detail to ++address new problems or concerns. ++ ++Each version is given a distinguishing version number. If the Program ++specifies a version number of this License which applies to it and \"any ++later version\", you have the option of following the terms and conditions ++either of that version or of any later version published by the Free ++Software Foundation. If the Program does not specify a version number of ++this License, you may choose any version ever published by the Free Software ++Foundation. ++ ++ 10. If you wish to incorporate parts of the Program into other free ++programs whose distribution conditions are different, write to the author ++to ask for permission. For software which is copyrighted by the Free ++Software Foundation, write to the Free Software Foundation; we sometimes ++make exceptions for this. Our decision will be guided by the two goals ++of preserving the free status of all derivatives of our free software and ++of promoting the sharing and reuse of software generally. ++" ++ return 0 ++ ;; ++ e | ed | edi | edit | editi | editin | editing ) ++ [[ -n $label ]] && label='editing: ' ++ local onoff="on." ++ [[ -z $_Dbg_edit ]] && onoff='off.' ++ _Dbg_msg \ ++"${label}Editing of command lines as they are typed is" $onoff ++ return 0 ++ ;; ++ de|deb|debu|debug|debugg|debugger|debuggi|debuggin|debugging ) ++ local onoff=${1:-'on'} ++ [[ -n $label ]] && label='debugger: ' ++ local onoff="off." ++ (( $_Dbg_debug_debugger )) && onoff='on.' ++ _Dbg_msg \ ++"${label}Allow debugging the debugger is" $onoff ++ return 0 ++ ;; ++ di|dir|dire|direc|direct|directo|director|directori|directorie|directories) ++ local list=${_Dbg_dir[0]} ++ local -i n=${#_Dbg_dir[@]} ++ local -i i ++ for (( i=1 ; i < n; i++ )) ; do ++ list="${list}:${_Dbg_dir[i]}" ++ done ++ ++ _Dbg_msg "Source directories searched: $list" ++ return 0 ++ ;; ++ li | lis | list | lists | listsi | listsiz | listsize ) ++ [[ -n $label ]] && label='listsize: ' ++ _Dbg_msg \ ++"${label}Number of source lines bashdb will list by default is" \ ++ "$_Dbg_listsize." ++ return 0 ++ ;; ++ p | pr | pro | prom | promp | prompt ) ++ [[ -n $label ]] && label='prompt: ' ++ _Dbg_msg \ ++"${label}bashdb's prompt is:\n" \ ++" \"$_Dbg_prompt_str\"." ++ return 0 ++ ;; ++ sho|show|showc|showco|showcom|showcomm|showcomma|showcomman|showcommand ) ++ [[ -n $label ]] && label='showcommand: ' ++ _Dbg_msg \ ++"${label}Show commands in debugger prompt is" \ ++ "$_Dbg_show_command." ++ return 0 ++ ;; ++ v | ve | ver | vers | versi | versio | version ) ++ _Dbg_cmd_show_versions ++ return 0 ++ ;; ++ w | wa | war | warr | warra | warran | warrant | warranty ) ++ _Dbg_cmd_info warranty ++ return 0 ++ ;; ++ *) ++ _Dbg_msg "Don't know how to show $show_cmd." ++ return 1 ++ esac ++} ++ ++_Dbg_cmd_show_versions() ++{ ++ _Dbg_msg "Release => $_Dbg_release" ++ [[ -n $_Dbg_script ]] && ++ _Dbg_msg "'bashdb' => $_Dbg_ver" ++ _Dbg_msg "'brk' => $_Dbg_brk_ver" ++ _Dbg_msg "'cmds' => $_Dbg_cmd_ver" ++ _Dbg_msg "'file' => $_Dbg_file_ver" ++ _Dbg_msg "'fns' => $_Dbg_fns_ver" ++ _Dbg_msg "'help' => $_Dbg_help_ver" ++ _Dbg_msg "'hist' => $_Dbg_hist_ver" ++ _Dbg_msg "'init' => $_Dbg_init_ver" ++ _Dbg_msg "'io' => $_Dbg_io_ver" ++ _Dbg_msg "'list' => $_Dbg_list_ver" ++ _Dbg_msg "'main' => $_Dbg_main_ver" ++ _Dbg_msg "'sig' => $_Dbg_sig_ver" ++ _Dbg_msg "'set' => $_Dbg_set_ver" ++ _Dbg_msg "'stack' => $_Dbg_stack_ver" ++} ++ ++# This is put at the so we have something at the end when we debug this. ++typeset -r _Dbg_set_ver=\ ++'$Id: dbg-set.inc,v 1.5 2003/03/22 00:16:31 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-sig-ret.inc cvs/debugger/dbg-sig-ret.inc +--- bash-2.05b/debugger/dbg-sig-ret.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-sig-ret.inc 2003-02-27 08:27:46.000000000 +0100 +@@ -0,0 +1,52 @@ ++# dbg-sig-ret.inc - Bourne Again Shell Debugger Signal handling routines ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++if [[ $_Dbg_old_ERR_handler != '' ]] ; then ++ eval $_Dbg_old_ERR_handler ++else ++ trap '' ERR ++fi ++ ++## Should make sure we are not using our own return handler... ++#if [[ $_Dbg_old_RETURN_handler != '' ]] ; then ++# eval $_Dbg_old_RETURN_handler ++#else ++# trap '' RETURN ++#fi ++ ++if [[ $_Dbg_old_INT_handler != '' ]] ; then ++ eval $_Dbg_old_INT_handler ++else ++ trap '' INT ++fi ++ ++# if [[ $_Dbg_old_EXIT_handler != '' ]] ; then ++# eval $_Dbg_old_EXIT_handler ++# else ++# trap '' EXIT ++# fi ++ ++local _Dbg_old_IFS="$IFS" ++IFS="$_Dbg_space_IFS"; ++set -$_Dbg_old_set_opts ++IFS="$_Dbg_old_IFS"; ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-sig.inc cvs/debugger/dbg-sig.inc +--- bash-2.05b/debugger/dbg-sig.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-sig.inc 2003-03-21 18:59:57.000000000 +0100 +@@ -0,0 +1,573 @@ ++# dbg-sig.inc - Bourne Again Shell Debugger Signal handling routines ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++# ==================== VARIABLES ======================================= ++ ++# The "set" options in effect ($-) before debugger was invoked. ++typeset _Dbg_old_setopts ++ ++# Return code that debugged program reports ++typeset -i _Dbg_program_exit_code=0 ++ ++# This is set to 1 if you want to debug debugger routines, i.e. routines ++# which start _Dbg_. But you better should know what you are doing ++# if you do this or else you may get into a recursive loop. ++typeset -i _Dbg_debug_debugger=0 ++ ++############################################################ ++## Signal arrays: These are indexed by the signal number. ## ++############################################################ ++ ++# Should we print that a signal was intercepted? ++# Each entry is "print" or "noprint" or null. ++typeset -a _Dbg_sig_print=() ++ ++# Should we reentry the debugger command loop on receiving the signal? ++# Each entry is "stop" or "nostop" or null. ++typeset -a _Dbg_sig_stop=() ++ ++# Should we show a traceback on receiving the signal? ++# Each entry is "stack" or "nostack" or null. ++typeset -a _Dbg_sig_show_stack=() ++# Should pass the signal to the user program?? ++# Each entry is the trap handler with some variables substituted. ++typeset -a _Dbg_sig_passthrough=() ++ ++# ===================== FUNCTIONS ======================================= ++ ++# We come here after before statement is run. This is the function named ++# in trap SIGDEBUG. ++ ++# Note: We have to be careful here in naming "local" variables. In contrast ++# to other places in the debugger, because of the read/eval loop, they are ++# in fact seen by those using the debugger. So in contrast to other "local"s ++# in the debugger, we prefer to preface these with _Dbg_. ++ ++_Dbg_debug_trap_handler() { ++ ++ ### The below is also copied below in _Dbg_sig_handler... ++ ### Should put common stuff into a function. ++ ++ # Consider putting the following line(s) in a routine. ++ # Ditto for the restore environment ++ local -i _Dbg_debugged_exit_code=$? ++ _Dbg_old_set_opts=$- ++ ++ # Turn off line and variable trace listing if were not in our own debug ++ # mode ++ (( !_Dbg_debug_debugger )) && set +x +v ++ ++ # If we are in our own routines -- these start with _bashdb -- then ++ # return. ++ if [[ ${FUNCNAME[1]} == _Dbg_* ]] && (( !_Dbg_debug_debugger )); then ++ _Dbg_set_to_return_from_debugger 0 ++ return 0 ++ fi ++ ++ _Dbg_set_debugger_entry ++ ++ local -i _Dbg_rc=0 ++ ++ # Were we called because of "trap fn ... RETURN"? ++ local -i _Dbg_is_return=$1 ++ shift ++ ++ _Dbg_bash_command=$1 ++ shift ++ ++ # Save values of $1 $2 $3 when debugged program was stopped ++ # We use the loop below rather than _Dbg_set_args="(@)" because ++ # we want to preserve embedded blanks in the arguments. ++ local -i _Dbg_n=${#@} ++ local -i _Dbg_i=1 ++ for (( ; _Dbg_n > 0; _Dbg_n-- )) ; do ++ _Dbg_arg[$_Dbg_i]=$1 ++ ((_Dbg_i++)) ++ shift ++ done ++ unset _Dbg_arg[0] # Get rid of line number; makes array count ++ # correct; also listing all _Dbg_arg works ++ # like $*. ++ ++ # Read in the journal to pick up variable settings that might have ++ # been left from a subshell. ++ _Dbg_source_journal ++ ++ if (( $BASHDB_QUIT_LEVELS > 0 )) ; then ++ _Dbg_cmd_quit $_Dbg_debugged_exit_code ++ fi ++ ++ # if in step mode, decrement counter ++ if ((_Dbg_steps >= 0)) ; then ++ ((_Dbg_steps--)) ++ _Dbg_write_journal "_Dbg_steps=$_Dbg_steps" ++ fi ++ ++ # look for watchpoints. ++ local -i _Dbg_i ++ for (( _Dbg_i=0; _Dbg_i < _Dbg_watch_max ; _Dbg_i++ )) ; do ++ if [ -n "${_Dbg_watch_exp[$_Dbg_i]}" ] \ ++ && [[ ${_Dbg_watch_enable[_Dbg_i]} != 0 ]] ; then ++ local new_val=`_Dbg_get_watch_exp_eval $_Dbg_i` ++ local old_val=${_Dbg_watch_val[$_Dbg_i]} ++ if [ $old_val != $new_val ] ; then ++ ((_Dbg_watch_count[_Dbg_i]++)) ++ _Dbg_msg "Watchpoint $i: ${_Dbg_watch_exp[$_Dbg_i]} changed:" ++ _Dbg_msg " old value: '$old_val'" ++ _Dbg_msg " new value: '$new_val'" ++ _Dbg_print_source_line ++ _Dbg_watch_val[$_Dbg_i]=$new_val ++ _Dbg_cmdloop ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_rc ++ fi ++ fi ++ done ++ ++ # Run applicable action statement ++ local entries=`_Dbg_get_assoc_array_entry "_Dbg_action_$_cur_filevar" $_curline` ++ if [[ $entries != "" ]] ; then ++ local -i _Dbg_i ++ for _Dbg_i in $entries ; do ++ if [[ ${_Dbg_action_enable[_Dbg_i]} != 0 ]] ; then ++ . ${_Dbg_libdir}/dbg-set-d-vars.inc ++ eval "${_Dbg_action_stmt[$_Dbg_i]}" ++ fi ++ done ++ fi ++ ++ # check if breakpoint reached ++ local -r entries=`_Dbg_get_assoc_array_entry "_brkpt_$_cur_filevar" $_curline` ++ if [[ $entries != "" ]] ; then ++ local -i _Dbg_i ++ for _Dbg_i in $entries ; do ++ if [[ ${_brkpt_enable[_Dbg_i]:0} != 0 ]] ; then ++ local -i cond ++ . ${_Dbg_libdir}/dbg-set-d-vars.inc ++ eval let cond=${_brkpt_cond[$_Dbg_i]:0} ++ if [[ $cond != 0 ]] ; then ++ ((_brkpt_count[_Dbg_i]++)) ++ if [[ ${_brkpt_onetime[_Dbg_i]:0} == 1 ]] ; then ++ _Dbg_delete_brkpt_entry $_Dbg_i ++ else ++ _Dbg_msg \ ++ "Breakpoint $_Dbg_i hit (${_brkpt_count[$_Dbg_i]} times)." ++ fi ++ _Dbg_print_source_line ++ _Dbg_cmdloop # enter debugger ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_rc ++ fi ++ fi ++ ++ done ++ fi ++ ++ # next, check if step mode and no. of steps is up ++ if ((_Dbg_steps == 0)); then ++ _Dbg_print_source_line ++ # _Dbg_msg "Stopped at line $_curline" ++ _Dbg_cmdloop # enter debugger ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_rc ++ elif (( _Dbg_is_return == 1 )) ; then ++ # here because a trap RETURN ++ _Dbg_print_source_line ++ _Dbg_cmdloop # enter debugger ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_rc ++ elif (( _Dbg_is_return == 2 )) ; then ++ # here because we are fielding a signal. ++ _Dbg_print_source_line ++ _Dbg_cmdloop # enter debugger ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_rc ++ else ++ ((_trace==1)) && _Dbg_print_source_line ++ fi ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_inside_skip ++} ++ ++# Cleanup routine: erase temp files before exiting. ++_Dbg_cleanup() { ++ rm $_Dbg_evalfile 2>/dev/null ++ _Dbg_erase_journals ++ # This is needed sometimes, and sometimes _Dbg_cleanup2 works. ++ # figure out why... ++ if (( BASHDB_QUIT_LEVELS < 1 )) ; then ++ rm $_Dbg_persistent 2>/dev/null ++ fi ++} ++ ++# Somehow we can't put this in _Dbg_cleanup and have it work. ++# I am not sure why. ++_Dbg_cleanup2() { ++ _Dbg_erase_journals ++ trap - EXIT ++} ++ ++# The real way to leave this program. ++_Dbg_cmd_quit() { ++ local -i return_code=${1:-$_Dbg_program_exit_code} ++ ++ local -i desired_quit_levels=${2:-0} ++ ++ if (( desired_quit_levels == 0 \ ++ || desired_quit_levels > BASH_SUBSHELL+1)) ; then ++ ((desired_quit_levels=BASH_SUBSHELL+1)) ++ fi ++ ++ ((BASHDB_QUIT_LEVELS+=desired_quit_levels)) ++ ++ # Reduce the number of recorded levels that we need to leave by ++ # if BASHDB_QUIT_LEVELS is numeric and greater than 0. ++ ((BASHDB_QUIT_LEVELS--)) ++ ++ ## write this to the next level up can read it. ++ _Dbg_write_journal "BASHDB_QUIT_LEVELS=$BASHDB_QUIT_LEVELS" ++ ++ # Reset signal handlers to their default but only if ++ # we are not in a subshell. ++ if (( BASH_SUBSHELL == 0 )) ; then ++ ++ # If we were told to restart from deep down, restart instead of quit. ++ if [ -n "$BASHDB_RESTART_COMMAND" ] ; then ++ _Dbg_erase_journals ++ _Dbg_save_state ++ exec $BASHDB_RESTART_COMMAND ++ fi ++ ++ _Dbg_cleanup ++ trap - DEBUG ++ # This is a hack we need. I am not sure why. ++ trap "_Dbg_cleanup2" EXIT ++ fi ++ ++ # And just when you thought we'd never get around to it... ++ exit $return_code ++} ++ ++# Save value of handler variable _Dbg_old_$sig ++_Dbg_save_handler() { ++ local -r sig=$1 ++ local old_handler='#unset#'; ++ if [[ `trap -p $sig` ]] ; then ++ old_handler=`trap -p $sig` ++ fi ++ if [[ $old_handler != '#unset#' ]] ; then ++ local -a old_hand_a=($old_handler) ++ old_handler=`_Dbg_subst_handler_var ${old_hand_a[2]}` ++ local -r decl_cmd="typeset -r _Dbg_old_${sig}_handler='$old_handler'" ++ eval $decl_cmd ++ fi ++ ++} ++ ++# Adjust handler variables to take into account the fact that when we ++# call the handler we will have added another call before the user's ++# handler. ++_Dbg_subst_handler_var() { ++ local -i i ++ local result='' ++ for arg in $* ; do ++ case $arg in ++ '$LINENO' ) ++ arg='${BASH_LINENO[0]}' ++ ;; ++ '${BASH_SOURCE[0]}' ) ++ arg='${BASH_SOURCE[1]}' ++ ;; ++ '${FUNCNAME[0]}' ) ++ arg='${FUNCNAME[1]}' ++ ;; ++ '${BASH_LINENO[0]}' ) ++ arg='${BASH_LINENO[1]}' ++ ;; ++ esac ++ if [[ $result == '' ]] ; then ++ result=$arg ++ else ++ result="$result $arg" ++ fi ++ done ++ echo $result ++} ++ ++# Debugger exit handler. Don't really exit - but go back the debugger ++# command loop ++_Dbg_exit_handler() { ++ ++ # Consider putting the following line(s) in a routine. ++ # Ditto for the restore environment ++ local -i _Dbg_debugged_exit_code=$? ++ ++ # Turn off line and variable trace listing. ++ set +x +v ++ ++ if [[ ${_Dbg_sig_print[0]} == "print" ]] ; then ++ # Note: use the same message that gdb does for this. ++ _Dbg_msg "Program received signal EXIT (0)..." ++ if [[ ${_Dbg_sig_show_stack[0]} == "showstack" ]] ; then ++ _Dbg_cmd_stack_trace 0 ++ fi ++ fi ++ ++ if [[ $_Dbg_old_EXIT_handler != '' ]] ; then ++ eval $_Dbg_old_EXIT_handler ++ fi ++ ++ # If we've set the QUIT signal handler not to stop, or we've in the ++ # middle of leaving so many (subshell) levels or we have set to ++ # leave quietly on termination, then do it! ++ ++ if [[ ${_Dbg_sig_stop[0]} != "stop" ]] \ ++ || [[ $BASHDB_QUIT_LEVELS != 0 ]] \ ++ || [ -n "$BASHDB_QUIT_ON_QUIT" ] ; then ++ _Dbg_cmd_quit ++ # We don't return from here. ++ fi ++ ++ # We've tested for all the quitting conditions above. ++ # Even though this is an exit handler, don't exit! ++ ++ local term_msg="normally" ++ if [[ $_Dbg_debugged_exit_code != 0 ]] ; then ++ term_msg="with code $_Dbg_debugged_exit_code" ++ fi ++ ++ # If we tried to exit from inside a subshell, we only want to enter ++ # the command loop if don't have any pending subshells. ++ if (( BASH_SUBSHELL == 0 )) ; then ++ _Dbg_msg \ ++ "Debugged program terminated $term_msg. Use q to quit or R to restart." ++ ++ while [[ 1 ]] ; do ++ _Dbg_cmdloop ++ done ++ fi ++} ++ ++# Generic signal handler for bashdb. We consult global array ++# _Dbg_sig_* for how to handle this signal. ++ ++# Since the command loop may be called we need to be careful about ++# using variable names that would be exposed to the user. ++_Dbg_sig_handler() { ++ ++ # Consider putting the following line(s) in a routine. ++ # Ditto for the restore environment ++ local -i _Dbg_debugged_exit_code=$? ++ _Dbg_old_set_opts=$- ++ ++ # Turn off line and variable trace listing. ++ set +x +v ++ ++ local -i _Dbg_rc=0 ++ ++ # This is the signal number. E.g. 15 is SIGTERM ++ local -r -i _Dbg_signum=$1 ++ ++ if [[ ${_Dbg_sig_print[$_Dbg_signum]} == "print" ]] || \ ++ [[ ${_Dbg_sig_stop[$_Dbg_signum]} == "stop" ]] ; then ++ local -r name=`_Dbg_signum2name $_Dbg_signum` ++ # Note: use the same message that gdb does for this. ++ _Dbg_msg "Program received signal $name ($_Dbg_signum)..." ++ if [[ ${_Dbg_sig_show_stack[$_Dbg_signum]} == "showstack" ]] ; then ++ _Dbg_cmd_stack_trace 0 ++ fi ++ fi ++ if [[ ${_Dbg_sig_stop[$_Dbg_signum]} == "stop" ]] ; then ++ ++ ### The below duplicates what is above in _Dbg_debug_trap handler ++ ### Should put common stuff into a function. ++ ++ shift # signum ++ ++ _Dbg_bash_command=$1 ++ shift ++ ++ # Save values of $1 $2 $3 when debugged program was stopped ++ # We use the loop below rather than _Dbg_set_args="(@)" because ++ # we want to preserve embedded blanks in the arguments. ++ local -i _Dbg_n=${#@} ++ local -i _Dbg_i=1 ++ for (( ; _Dbg_n > 0; _Dbg_n-- )) ; do ++ _Dbg_arg[$_Dbg_i]=$1 ++ ((_Dbg_i++)) ++ shift ++ done ++ unset _Dbg_arg[0] # Get rid of line number; makes array count ++ # correct; also listing all _Dbg_arg works ++ # like $*. ++ ++ _Dbg_set_debugger_entry ++ _Dbg_cmdloop # enter debugger ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_rc ++ ++ elif (( _Dbg_sig_old_handler[_Dbg_signum] )) ; then ++ eval ${_Dbg_sig_old_handler[$_Dbg_signum]} ++ fi ++ _Dbg_set_to_return_from_debugger 1 ++ return $_Dbg_debugged_exit_code ++} ++ ++_Dbg_err_handler() { ++ if [[ $_Dbg_old_ERR_handler != '' ]] ; then ++ eval $_Dbg_old_ERR_handler ++ fi ++ _Dbg_msg "Error occured at ${BASH_SOURCE[1]}:${BASH_LINENO[1]}" ++ _Dbg_cmdloop ++} ++ ++# Echo the name for a given signal number $1. The resulting name ++# is in _Dbg_return ++_Dbg_signum2name() { ++ local -i -r signum=$1; ++ builtin kill -l $signum 2>/dev/null ++ return $? ++} ++ ++# Return the signal number for a given signal name $1. The resulting number ++# is in _Dbg_return ++_Dbg_name2signum() { ++ local -r signame=$1; ++ builtin kill -l $signame 2>/dev/null ++ return $? ++} ++ ++# Process debugger "handle" command. ++_Dbg_cmd_handle() { ++ local sig=$1 ++ local cmd=$2 ++ local -i signum ++ if [[ -z $sig ]] ; then ++ _Dbg_msg "Missing signal name or signal number." ++ return 1 ++ fi ++ ++ eval "$_seteglob" ++ if [[ $sig == $int_pat ]]; then ++ eval "$_resteglob" ++ signame=`_Dbg_signum2name $sig` ++ if (( $? != 0 )) ; then ++ _Dbg_msg "Bad signal number: $sig" ++ return 1 ++ fi ++ signum=sig ++ else ++ eval "$_resteglob" ++ local signum; ++ signum=`_Dbg_name2signum $sig` ++ if (( $? != 0 )) ; then ++ _Dbg_msg "Bad signal name: $sig" ++ return 1 ++ fi ++ fi ++ ++ case $cmd in ++ nop | nopr | nopri | noprin | noprint ) ++ _Dbg_sig_print[signum]='noprint' ++ ;; ++ ++ nosta | nostac | nostack ) ++ _Dbg_sig_show_stack[signum]='nostack' ++ ;; ++ ++ nosto | nostop ) ++ _Dbg_sig_stop[signum]='nostop' ++ ;; ++ ++ p | pr | pri | prin | print ) ++ _Dbg_sig_print[signum]='print' ++ ;; ++ ++ sta | stac | stack ) ++ _Dbg_sig_show_stack[signum]='showstack' ++ ;; ++ ++ sto | stop ) ++ _Dbg_sig_stop[signum]='stop' ++ ;; ++ * ) ++ if (( !cmd )) ; then ++ _Dbg_msg \ ++ "Need to give a command: stop, nostop, stack, nostack, print, noprint" ++ else ++ _Dbg_msg "Invalid handler command $cmd" ++ fi ++ ;; ++ esac ++ ++} ++ ++_Dbg_subexit_handler() { ++ # Read in the journal to pick up variable settings that might have ++ # been left from a subshell. ++ if [[ ${FUNCNAME[1]} == _Dbg_* ]] && (( !_Dbg_debug_debugger )); then ++ return 0 ++ fi ++ _Dbg_source_journal ++ if (( $BASHDB_QUIT_LEVELS > 0 )) ; then ++ _Dbg_cmd_quit $_Dbg_debugged_exit_code ++ fi ++} ++ ++# Set up generic trap handler. Arguments are: ++# NAME print showstack stop passthrough ++_Dbg_init_trap() { ++ local -r name=$1 ++ local -i -r signum=`_Dbg_name2signum $name` ++ ++ _Dbg_sig_print[$signum]=$2; ++ _Dbg_sig_show_stack[$signum]=$3; ++ _Dbg_sig_stop[$signum]=$4; ++ ++ # Work out passthrough later... ++ # if [[ $5 == "pass*" ]] ; then ++ # get existing trap from env. ++ # _Dbg_sig_show_passthrough[$signum]=....; ++ # ++ ++ if (( signum == 0 )) ; then ++ trap '_Dbg_exit_handler "$BASH_COMMAND"' EXIT ++ else ++ local trap_cmd="trap '_Dbg_sig_handler $signum \"\$BASH_COMMAND\" \"\$@\"' $name" ++ eval $trap_cmd ++ fi ++} ++ ++_Dbg_init_trap EXIT "noprint" "nostack" "stop" ++_Dbg_init_trap INT "print" "showstack" "stop" ++_Dbg_init_trap TRAP "print" "showstack" "stop" ++_Dbg_init_trap TERM "print" "showstack" "stop" ++_Dbg_init_trap QUIT "print" "showstack" "stop" ++_Dbg_init_trap TRAP "print" "showstack" "stop" ++_Dbg_init_trap ILL "print" "showstack" "stop" ++ ++# This is put at the so we have something at the end to stop at ++# when we debug this. By stopping at the end all of the above functions ++# and variables can be tested. ++typeset -r _Dbg_sig_ver=\ ++'$Id: dbg-sig.inc,v 1.4 2003/03/21 17:59:57 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/dbg-stack.inc cvs/debugger/dbg-stack.inc +--- bash-2.05b/debugger/dbg-stack.inc 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/dbg-stack.inc 2003-02-27 08:27:46.000000000 +0100 +@@ -0,0 +1,178 @@ ++# dbg-stack.inc - Bourne Again Shell Debugger Call Stack routines ++# ++# Copyright (C) 2002,2003 Rocky Bernstein ++# ++# Bash is free software; you can redistribute it and/or modify it under ++# the terms of the GNU General Public License as published by the Free ++# Software Foundation; either version 2, or (at your option) any later ++# version. ++# ++# Bash is distributed in the hope that it will be useful, but WITHOUT ANY ++# WARRANTY; without even the implied warranty of MERCHANTABILITY or ++# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License ++# for more details. ++# ++# You should have received a copy of the GNU General Public License along ++# with Bash; see the file COPYING. If not, write to the Free Software ++# Foundation, 59 Temple Place, Suite 330, Boston, MA 02111 USA. ++ ++#================ VARIABLE INITIALIZATIONS ====================# ++ ++# The top number items on the FUNCNAME stack are debugging routines ++# Set the index in FUNCNAME that should be reported as index 0 (or top). ++typeset -ir _Dbg_stack_top=2 ++typeset -i _Dbg_stack_pos # Where are we in stack? ++ ++#======================== FUNCTIONS ============================# ++ ++# Move default values down $1 or one in the stack. ++ ++_Dbg_cmd_down() { ++ local -i count=${1:-1} ++ ++ if (( _Dbg_stack_pos-count < _Dbg_stack_top )) ; then ++ _Dbg_msg "Would be beyond bottom-most entry." ++ return ++ fi ++ ((_Dbg_stack_pos -= $count)) ++ local -i j=_Dbg_stack_pos+1 ++ _Dbg_listline=${BASH_LINENO[$_Dbg_stack_pos]} ++ _cur_source_file=${BASH_SOURCE[$j]} ++ _Dbg_print_source_line $_Dbg_listline ++} ++ ++_Dbg_cmd_frame() { ++ eval "$_seteglob" ++ if [[ $1 != '' && $1 != $int_pat ]] ; then ++ _Dbg_msg "Bad integer parameter: $1" ++ eval "$_resteglob" ++ return ++ fi ++ eval "$_resteglob" ++ local -i pos=${1:-0} ++ if (( _Dbg_stack_top+pos+1 >= ${#FUNCNAME[@]} )) ; then ++ _Dbg_msg "Would be beyond top-most entry." ++ return ++ fi ++ _Dbg_stack_pos=$_Dbg_stack_top+$pos ++ _Dbg_cmd_stack_trace 2 1 $pos ++ local -i k=_Dbg_stack_pos+1 ++ _Dbg_listline=${BASH_LINENO[$_Dbg_stack_pos]} ++ _cur_source_file=${BASH_SOURCE[$k]} ++ _Dbg_print_source_line ++} ++ ++# Print a stack trace. ++# $1 is an additional offset correction - this routine is called from two ++# different places and one routine has one more additional call on top. ++# $2 is the maximum number of entries to include. ++# This code assumes the's debugger version of ++# bash where FUNCNAME is an array, not a variable. ++ ++_Dbg_cmd_stack_trace() { ++ ++ local -i n=${#FUNCNAME[@]}-1 ++ ++ eval "$_seteglob" ++ if [[ $1 != $int_pat ]] ; then ++ _Dbg_msg "Bad integer parameter: $1" ++ eval "$_resteglob" ++ return 1 ++ fi ++ if [[ $2 != '' && $2 != $int_pat ]] ; then ++ _Dbg_msg "Bad integer parameter: $2" ++ eval "$_resteglob" ++ return 1 ++ fi ++ eval "$_resteglob" ++ ++ local prefix='##' ++ local -i count=${2:-$n} ++ local -i k=${3:-0} ++ local -i i=_Dbg_stack_top+k+$1 ++ local -i j=i ++ ++ (( j > n )) && return 1 ++ (( i == _Dbg_stack_pos+$1 )) && prefix='->' ++ if (( k == 0 )) ; then ++ local filename=${BASH_SOURCE[$i]} ++ (( _Dbg_basename_only )) && filename=${filename##*/} ++ _Dbg_msg "$prefix$k in file \`$filename' at line $_curline" ++ ++ ((count--)) ; ((k++)) ++ fi ++ ++ # Figure out the index in BASH_ARGV is position "i" (the place where ++ # we start our stack trace from). variable "r" will be that place. ++ ++ local -i q ++ local -i r=0 ++ for (( q=0 ; q 0 )) ; i++ )) ; do ++ local -i arg_count=${BASH_ARGC[$i]} ++ ((j++)) ; ((count--)) ++ prefix='##' ++ (( i == _Dbg_stack_pos )) && prefix='->' ++ if (( i == n )) ; then ++ # main()'s file is the same as the first caller ++ j=i ++ fi ++ ++ _Dbg_msg_nocr "$prefix$k ${FUNCNAME[$i]}(" ++ ++ local parms='' ++ ++ # Print out parameter list. ++ if (( 0 != ${#BASH_ARGC[@]} )) ; then ++ local -i s ++ for (( s=0; s < arg_count; s++ )) ; do ++ if (( s != 0 )) ; then ++ parms="\"${BASH_ARGV[$r]}\", $parms" ++ elif [[ ${FUNCNAME[$i]} == "source" ]] \ ++ && (( _Dbg_basename_only )); then ++ local filename=${BASH_ARGV[$r]} ++ filename=${filename##*/} ++ parms="\"$filename\"" ++ else ++ parms="\"${BASH_ARGV[$r]}\"" ++ fi ++ ((r++)) ++ done ++ fi ++ ++ local filename=${BASH_SOURCE[$j]} ++ (( _Dbg_basename_only )) && filename=${filename##*/} ++ _Dbg_msg "$parms) called from file \`$filename'" \ ++ "at line ${BASH_LINENO[$i]}" ++ ((k++)) ++ done ++ return 0 ++} ++ ++# Move default values up $1 or one in the stack. ++ ++_Dbg_cmd_up() { ++ local -i count=${1:-1} ++ ++ if (( _Dbg_stack_pos+count+1 >= ${#FUNCNAME[@]} )) ; then ++ _Dbg_msg "Would be beyond top-most entry." ++ return ++ fi ++ ((_Dbg_stack_pos += $count)) ++ local -i k=_Dbg_stack_pos+1 ++ _Dbg_listline=${BASH_LINENO[$_Dbg_stack_pos]} ++ _cur_source_file=${BASH_SOURCE[$k]} ++ _Dbg_print_source_line $_Dbg_listline ++} ++ ++# This is put at the so we have something at the end when we debug this. ++_Dbg_stack_ver='$Id: dbg-stack.inc,v 1.2 2003/02/27 07:27:46 rockyb Exp $' ++ ++#;;; Local Variables: *** ++#;;; mode:shell-script *** ++#;;; eval: (sh-set-shell "bash") *** ++#;;; End: *** +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/doc/.#bashdb.texi.1.5 cvs/debugger/doc/.#bashdb.texi.1.5 +--- bash-2.05b/debugger/doc/.#bashdb.texi.1.5 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/doc/.#bashdb.texi.1.5 2003-08-30 08:09:04.000000000 +0200 +@@ -0,0 +1,3754 @@ ++\input texinfo @c -*-texinfo-*- ++@c Copyright 2002, 2003 ++@c Rocky Bernstein for the Free Software Foundation ++@c ++@c TODO: ++@c - add examples for commands ++@c - clean up/improve sample session ++@c - help text is inaccurate and formatted too much to right. ++@c ++ ++@c Sets version and release names and dates. Frees us from changing ++@c this file when a new release comes along. ++ ++@c %**start of header ++@c makeinfo ignores cmds prev to setfilename, so its arg cannot make use ++@c of @set vars. However, you can override filename with makeinfo -o. ++@setfilename bashdb.info ++@c ++@c Name of Bash program. Used in running text. ++@set BASH @acronym{BASH} ++ ++@c ++@c Name of debugger program. Used also for prompt string. ++@set DBG the @value{BASH} debugger ++@set dBGP The @value{BASH} debugger ++@set DDD @acronym{DDD} ++ ++@settitle @value{BASH} Debugger ++@setchapternewpage odd ++@setcontentsaftertitlepage ++ ++@c %**end of header ++ ++@include version.texi ++ ++@c A simple macro for optional variables. ++@macro ovar{varname} ++@r{[}@var{\varname\}@r{]} ++@end macro ++ ++@macro DDD {} ++@acronym{DDD} ++@end macro ++ ++@macro BASH {} ++@acronym{@value{BASH}} ++@end macro ++ ++@macro DBG {} ++@value{DBG} ++@end macro ++ ++@iftex ++@c @smallbook ++@c @cropmarks ++@end iftex ++ ++@finalout ++ ++@c readline appendices use @vindex, @findex and @ftable, ++@c annotate.texi and gdbmi use @findex. ++@c @syncodeindex vr cp ++@c @syncodeindex fn cp ++ ++@c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER. ++ ++@c This is a dir.info fragment to support semi-automated addition of ++@c manuals to an info tree. ++@dircategory Programming & development tools. ++@direntry ++* @value{DBG}: (@value{DBG}). The @sc{bash} debugger. ++@end direntry ++ ++@ifinfo ++This file documents the @sc{bash} debugger @value{BASH}. ++ ++ ++This is the @value{EDITION} Edition, @value{UPDATED-MONTH}, ++of @cite{Debugging with BASHDB: the @sc{gnu} Source-Level Debugger} ++for REBASH ++ ++Copyright (C) 2002,2003 Rocky Bernstein for the Free Software Foundation. ++ ++Permission is granted to copy, distribute and/or modify this document ++under the terms of the GNU Free Documentation License, Version 1.1 or ++@ifset DEBIANHASBECOMEREASONABLE ++@c From Matthias Klose a Debian maintainer on ++@c Sat, 23 Aug 2003 14:24:44 +0200 ++@c ++@c I personally see the invariant sections as the thing in the ++@c GFDL, which hinders me in uploading the package to the archives. ++@c I don't have any problem, if some other Debian developer makes a ++@c bashdb package built from separate sources. ++@c ++@c I am aware that Debian ships other packages containing documentation ++@c covered by the GFDL (and one of them for which I do the packaging as ++@c well), but I won't add a new package, which I maintain. So before an ++@c upload of a bashdb package built from the bash sources either ++@c ++@c ++@c - Debian has a position on the GFDL, which allows inclusion ++@c ++@c - the bashdb manual does not have invariant sections, or is ++@c relicensed, or dual licensed. ++any later version published by the Free Software Foundation; with the ++Invariant Sections being ``Free Software'' and ``Free Software Needs ++Free Documentation'', with the Front-Cover Texts being ``A GNU ++Manual,'' and with the Back-Cover Texts as in (a) below. ++ ++(a) The Free Software Foundation's Back-Cover Text is: ``You have ++freedom to copy and modify this GNU Manual, like GNU software. Copies ++published by the Free Software Foundation raise funds for GNU ++development.'' ++@else ++any later version published by the Free Software Foundation; with no ++Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. ++@end ifset ++@end ifinfo ++ ++@titlepage ++@title Debugging with BASHDB ++@sp 1 ++@subtitle @value{EDITION} Edition, for BASH ++@subtitle @value{UPDATED-MONTH} ++@author Rocky Bernstein ++@page ++@tex ++{\parskip=0pt ++\hfill (Send bugs and comments on bashdb to bug-bashdb\@sourceforge.net.)\par ++\hfill {\it Debugging with BASH}\par ++\hfill \TeX{}info \texinfoversion\par ++} ++@end tex ++ ++@vskip 0pt plus 1filll ++Copyright @copyright{} 2002,2003 Rocky Bernstein for the Free Software ++Foundation. ++ ++Permission is granted to copy, distribute and/or modify this document ++under the terms of the GNU Free Documentation License, Version 1.1 or ++@ifset DEBIANHASBECOMEREASONABLE ++@c From Matthias Klose a Debian maintainer on ++@c Sat, 23 Aug 2003 14:24:44 +0200 ++@c ++@c I personally see the invariant sections as the thing in the ++@c GFDL, which hinders me in uploading the package to the archives. ++@c I don't have any problem, if some other Debian developer makes a ++@c bashdb package built from separate sources. ++@c ++@c I am aware that Debian ships other packages containing documentation ++@c covered by the GFDL (and one of them for which I do the packaging as ++@c well), but I won't add a new package, which I maintain. So before an ++@c upload of a bashdb package built from the bash sources either ++@c ++@c ++@c - Debian has a position on the GFDL, which allows inclusion ++@c ++@c - the bashdb manual does not have invariant sections, or is ++@c relicensed, or dual licensed. ++any later version published by the Free Software Foundation; with the ++Invariant Sections being ``Free Software'' and ``Free Software Needs ++Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,'' ++and with the Back-Cover Texts as in (a) below. ++ ++(a) The Free Software Foundation's Back-Cover Text is: ``You have ++freedom to copy and modify this GNU Manual, like GNU software. Copies ++published by the Free Software Foundation raise funds for GNU ++development.'' ++@else ++any later version published by the Free Software Foundation; with no ++Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. ++@end ifset ++@end titlepage ++@page ++ ++@ifnottex ++@node Top, Summary, (dir), (dir) ++ ++@top Debugging with @DBG ++ ++This file describes @value{DBG}, the @sc{bash} symbolic debugger. ++ ++This is the @value{EDITION} Edition, @value{UPDATED-MONTH}, for REBASH. ++ ++Copyright (C) 2002,2003 Rocky Bernstein ++ ++@menu ++* Summary:: Overview of Debugger with a sample session ++ ++* Invocation:: Getting in and out ++* Running:: Script setup inside the BASH debugger ++* Debugger Command Reference:: BASH debugger command reference ++ ++* Front Ends:: Using @DBG from a front-end user interface ++ ++* BASH Debugger Bugs:: Reporting bugs ++* History and Acknowledgments:: History and Acknowledgments ++ ++Appendices ++* Formatting Documentation:: How to format and print @DBG documentation ++ ++* Installing bashdb:: Installing bashdb ++* Copying:: GNU General Public License says ++ how you can copy and share bashdb ++* GNU Free Documentation License:: The license for this documentation ++ ++Indexes (nodes containing large menus) ++* Function Index:: An item for each function name. ++* Command Index:: An item for each command name. ++* Variable Index:: An item for each documented variable. ++* General Index:: An item for each concept. ++@end menu ++ ++@end ifnottex ++ ++@contents ++ ++@node Summary ++@chapter Summary of the BASH Debugger ++ ++The purpose of a debugger such as @DBG is to allow you to see what is ++going on ``inside'' a bash script while it executes. ++ ++@DBG can do four main kinds of things (plus other things in support of ++these) to help you catch bugs in the act: ++ ++@itemize @bullet ++@item ++Start your script, specifying anything that might affect its behavior. ++ ++@item ++Make your script stop on specified conditions. ++ ++@item ++Examine what has happened, when your script has stopped. ++ ++@item ++Change things in your script, so you can experiment with correcting the ++effects of one bug and go on to learn about another. ++@end itemize ++ ++Although you can use the BASH debugger to debug scripts written in ++BASH, it can also be used just as a front-end for learning more ++about programming in BASH. As an additional aid, the debugger can ++be used within the context of an existing script with its functions ++and variables that have already been initialized; fragments of the ++existing can be experimented with by entering them inside the debugger. ++ ++@menu ++* Sample Session:: A Sample BASH Debugger session ++* Free Software:: Freely redistributable software ++@end menu ++ ++@node Sample Session ++@section A Sample BASH Debugger Session ++ ++You can use this manual at your leisure to read all about @value{DBG}. ++However, a handful of commands are enough to get started using the ++debugger. This chapter illustrates those commands. ++ ++@iftex ++In this sample session, we emphasize user input like this: @b{input}, ++to make it easier to pick out from the surrounding output. ++@end iftex ++ ++Below we will debug a script that contains a function to compute the ++factorial of a number: fact(0) is 1 and fact(n) is n*fact(n-1). ++ ++@smallexample ++$ @b{bashdb -L . fact.sh} ++Bourne-Again Shell Debugger, release bash-@value{BASHRELEASE} ++Copyright 2002,2003 Rocky Bernstein ++This is free software, covered by the GNU General Public License, and you are ++welcome to change it and/or distribute copies of it under certain conditions. ++ ++(fact.sh:9): ++9: echo fact 0 is: `fact 0` ++bashdb<0> @b{-} ++ 1: #!/usr/local/bin/bash ++ 2: fact() @{ ++ 3: ((n==0)) && echo 1 && return ++ 4: ((nm1=n-1)) ++ 5: ((result=n*`fact $nm1`)) ++ 6: return $result ++ 7: @} ++ 8: ++ 9:==> echo fact 0 is: `fact 0` ++bashdb<1> @b{l} ++ 10: echo fact 3 is: $(fact 3) ++@end smallexample ++ ++@noindent ++ ++The command invocation uses the option ``-L .'' Here we assume that ++the @command{bashdb} script and the debugger files are in the same ++location. If you are running from the source code, this will be the ++case. However if bashdb has been installed this probably won't be true ++and here you probably don't need to use ``-L .'' Instead you would ++type simply @code{bashdb fact.sh}. ++ ++The first debugger command we gave @kbd{-}, we listed a window of ++lines @emph{before} where we were executing. Because the window, 10 ++lines, is larger than the number of lines to the top of the file we ++printed only 9 lines here. The next command list starting from the ++current line and again wants to print 10 lines but because there are ++only one remaining line, that is what is printed. ++ ++@smallexample ++bashdb<2> @b{step} ++(fact.sh:9): ++9: echo fact 0 is: `fact 0` ++bashdb<3> @b{@key{RET}} ++2: fact() @{ ++bashdb<4> @b{@key{RET}} ++3: ((n==0)) && echo 1 && return ++bashdb<5> @b{print $n} ++ ++@end smallexample ++ ++Ooops... The variable @kbd{n} isn't initialized. ++ ++The first @kbd{step} command steps the script one instruction. It may ++seem odd that the line printed is exactly the same one as before. What ++has happened though is that we've ``stepped'' into the subshell needed ++to run @kbd{`fact 0`}; we haven't however started running anything ++inside that subshell yet though---the function call. ++ ++The first @kbd{step} command steps the script one instruction; it ++didn't advance the line number at all. That is because we were ++stopping before the command substitution or backtick is to take ++place. The second command we entered was just hitting the return key; ++bashdb remembers that you entered @code{step} previously, so it runs ++the step rather than @kbd{next}, the other alternative when you hit ++@key{RET}. Step one more instruction and we are just before running ++the first statement of the function. ++ ++Next, we print the value of the variable @kbd{n}. Notice we need to add ++a preceding dollar simple to get the substitution or value of n. As we ++will see later, if the @kbd{pe} command were used this would not be ++necessary. ++ ++We now modify the file to add an assignment to local variable @kbd{n} and ++restart. ++ ++@smallexample ++bashdb<6> @b{restart} ++Restarting with: /usr/local/bin/bashdb -L . fact.sh ++(fact.sh:10): ++10: echo fact 0 is: `fact 0` ++bashdb<0> @b{list 1} ++ 1: #!/usr/local/bin/bash ++ 2: fact() @{ ++ 3: local -i n=$@{1:0@} ++ 4: ((n==0)) && echo 1 && return ++ 5: ((nm1=n-1)) ++ 6: ((result=n*`fact $nm1`)) ++ 7: return $result ++ 8: @} ++ 9: ++ 10:==> echo fact 0 is: `fact 0` ++bashdb<1> @b{s 3} ++(fact.sh:3): ++3: local -i n=$@{1:0@} ++bashdb<2> @b{step} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++bashdb<3> @b{print $n} ++print $n ++0 ++@end smallexample ++ ++@noindent ++This time we use the @code{list} debugger command to list the lines in ++the file. From before we know it takes three @code{steps} commands ++before we get into the fact() function, so we add a count onto the ++@code{step} command. Notice we abbreviate @code{step} with @code{s}; ++we could have done likewise and abbreviated @code{list} with @code{l}. ++ ++@smallexample ++bashdb<4> @b{@key{RET}} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++echo 1 ++bashdb<5> @b{@key{RET}} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++return ++@end smallexample ++ ++@noindent ++Again we just use @key{RET} to repeat the last @code{step} ++commands. And again the fact that we are staying on the same line 4 ++means that the next condition in the line is about to be ++executed. Notice that we see the command (@code{echo 1} or ++@code{return}) listed when we stay on the same line which has multiple ++stopping points in it. Given the information above, we know that the ++value echo'ed on return will be 1. ++ ++@smallexample ++bashdb<6> @b{@key{RET}} ++fact 0 is: 1 ++(fact.sh:12): ++12: echo fact 3 is: $(fact 3) ++bashdb<7> @b{break 5} ++Breakpoint 1 set in file fact.sh, line 5. ++bashdb<8> @b{continue} ++@end smallexample ++ ++@noindent ++We saw that we could step with a count into the function ++fact(). However above took another approach: we set a stopping point or ++``breakpoint'' at line 5 to get us a little ways into the fact() ++subroutine. Just before line 5 is to executed, we will get back into ++the debugger. The @code{continue} command just resumes execution until ++the next stopping point which has been set up in some way. ++ ++@smallexample ++(fact.sh:5): ++5: ((nm1=n-1)) ++Breakpoint 1 hit(1 times). ++bashdb<8> @b{x n-1} ++2 ++bashdb<9> @b{s} ++(fact.sh:5): ++6: ((result=n*`fact $nm1`)) ++bashdb<10> @b{c} ++fact.sh: line 6: ((: result=n*: syntax error: operand expected (error token is "*") ++bashdb<7> @b{R} ++Restarting with: bash --debugger fact.sh ++11: echo fact 0 is: `fact 0` ++bashdb<0> @b{l fact} ++ 2: fact () ++ 3: @{ ++ 4: local -i n=$@{1:0@}; ++ 5: (( "n==0" )) && echo 1 && return; ++ 6: (( nm1=n-1 )); ++ 7: ((fact_nm1=`fact $nm1`)) ++ 8: (( "result=n*fact_nm1" )); ++ 9: echo $result ++10: @} ++@end smallexample ++ ++@noindent ++In addition to listing by line numbers, we can also list giving a ++function name. Below, instead of setting a breakpoint at line 5 and ++running ``continue'' as we did above, we try something slightly shorter ++and slightly different. We give the line number on the ``continue'' ++statement. This is a little different in that a one-time break is made ++on line 5. Once that statement is reached the breakpoint is removed. ++ ++@smallexample ++bashdb<1> @b{continue 5} ++One-time breakpoint 1 set in file fact.sh, line 5. ++fact 0 is: 1 ++(fact.sh:5): ++5: ((nm1=n-1)) ++bashdb<2> @b{s} ++6: ((fact_nm1=`fact $nm1`)) ++bashdb<2> @b{s} ++2: fact() @{ ++bashdb<3> @b{T} ++->0 in file `fact.sh' at line 2 ++##1 fact("3") called from file `fact.sh' at line 12 ++##2 source("fact.sh") called from file `/usr/local/bin/bashdb' at line 154 ++##3 main("fact.sh") called from file `/usr/local/bin/bashdb' at line 0 ++bashdb<4> @b{c} ++fact 3 is: 6 ++Debugged program terminated normally. Use q to quit or R to restart. ++@end smallexample ++ ++@noindent ++When we stop at line 5 above, we have already run fact(0) and output ++the correct results. The output from the program ``fact 0 is: 1'' is ++intermixed with the debugger output. The @code{T} command above ++requests call stack output and this confirms that we are not in the ++fact(0) call but in the fact(3) call. There are 4 lines listed in the ++stack trace even though there is just one call from the main ++program. The top line of the trace doesn't really represent a call, ++it's just where we currently are in the program. That last line is an ++artifact of invoking bash from the bashdb script rather than running ++@code{bash --debugger}. ++ ++The last message in the output above @samp{Debugged program exited ++normally.} is from @value{DBG}; it indicates script has finished ++executing. We can end our @value{DDGP} session with the @value{GDBP} ++@code{quit} command. ++ ++Above we did our debugging session on the command line. If you are a ++GNU Emacs user, you can do your debugging inside that. Also there is ++a(nother) GUI interface called DDD that supports @value{DBG}. ++ ++@node Free Software ++@unnumberedsec Free software ++ ++@DBG is @dfn{free software}, protected by the @sc{gnu} ++General Public License ++(GPL). The GPL gives you the freedom to copy or adapt a licensed ++program---but every person getting a copy also gets with it the ++freedom to modify that copy (which means that they must get access to ++the source code), and the freedom to distribute further copies. ++Typical software companies use copyrights to limit your freedoms; the ++Free Software Foundation uses the GPL to preserve these freedoms. ++ ++Fundamentally, the General Public License is a license which says that ++you have these freedoms and that you cannot take these freedoms away ++from anyone else. ++ ++@unnumberedsec Free Software Needs Free Documentation ++ ++The biggest deficiency in the free software community today is not in ++the software---it is the lack of good free documentation that we can ++include with the free software. Many of our most important ++programs do not come with free reference manuals and free introductory ++texts. Documentation is an essential part of any software package; ++when an important free software package does not come with a free ++manual and a free tutorial, that is a major gap. We have many such ++gaps today. ++ ++Consider Perl, for instance. The tutorial manuals that people ++normally use are non-free. How did this come about? Because the ++authors of those manuals published them with restrictive terms---no ++copying, no modification, source files not available---which exclude ++them from the free software world. ++ ++That wasn't the first time this sort of thing happened, and it was far ++from the last. Many times we have heard a GNU user eagerly describe a ++manual that he is writing, his intended contribution to the community, ++only to learn that he had ruined everything by signing a publication ++contract to make it non-free. ++ ++Free documentation, like free software, is a matter of freedom, not ++price. The problem with the non-free manual is not that publishers ++charge a price for printed copies---that in itself is fine. (The Free ++Software Foundation sells printed copies of manuals, too.) The ++problem is the restrictions on the use of the manual. Free manuals ++are available in source code form, and give you permission to copy and ++modify. Non-free manuals do not allow this. ++ ++The criteria of freedom for a free manual are roughly the same as for ++free software. Redistribution (including the normal kinds of ++commercial redistribution) must be permitted, so that the manual can ++accompany every copy of the program, both on-line and on paper. ++ ++Permission for modification of the technical content is crucial too. ++When people modify the software, adding or changing features, if they ++are conscientious they will change the manual too---so they can ++provide accurate and clear documentation for the modified program. A ++manual that leaves you no choice but to write a new manual to document ++a changed version of the program is not really available to our ++community. ++ ++Some kinds of limits on the way modification is handled are ++acceptable. For example, requirements to preserve the original ++author's copyright notice, the distribution terms, or the list of ++authors, are ok. It is also no problem to require modified versions ++to include notice that they were modified. Even entire sections that ++may not be deleted or changed are acceptable, as long as they deal ++with nontechnical topics (like this one). These kinds of restrictions ++are acceptable because they don't obstruct the community's normal use ++of the manual. ++ ++However, it must be possible to modify all the @emph{technical} ++content of the manual, and then distribute the result in all the usual ++media, through all the usual channels. Otherwise, the restrictions ++obstruct the use of the manual, it is not free, and we need another ++manual to replace it. ++ ++Please spread the word about this issue. Our community continues to ++lose manuals to proprietary publishing. If we spread the word that ++free software needs free reference manuals and free tutorials, perhaps ++the next person who wants to contribute by writing documentation will ++realize, before it is too late, that only free manuals contribute to ++the free software community. ++ ++If you are writing documentation, please insist on publishing it under ++the GNU Free Documentation License or another free documentation ++license. Remember that this decision requires your approval---you ++don't have to let the publisher decide. Some commercial publishers ++will use a free license if you insist, but they will not propose the ++option; it is up to you to raise the issue and say firmly that this is ++what you want. If the publisher you are dealing with refuses, please ++try other publishers. If you're not sure whether a proposed license ++is free, write to @email{licensing@@gnu.org}. ++ ++You can encourage commercial publishers to sell more free, copylefted ++manuals and tutorials by buying them, and particularly by buying ++copies from the publishers that paid for their writing or for major ++improvements. Meanwhile, try to avoid buying non-free documentation ++at all. Check the distribution terms of a manual before you buy it, ++and insist that whoever seeks your business must respect your freedom. ++Check the history of the book, and try to reward the publishers that ++have paid or pay the authors to work on it. ++ ++The Free Software Foundation maintains a list of free documentation ++published by other publishers, at ++@url{http://www.fsf.org/doc/other-free-books.html}. ++ ++@node Invocation ++@chapter Getting in and out ++ ++This chapter discusses how to start @value{DBG}, and how to get out of it. ++The essentials are: ++@itemize @bullet ++@item ++type @samp{bash --debugger @emph{script-name}} or @samp{bashdb ++@emph{script-name}} to start @value{DBG}. ++@item ++type @kbd{quit} or @kbd{C-d} inside the debugger to exit. ++@end itemize ++ ++But also note there are two front-ends available as well. One can also ++enter the debugger inside emacs via the command @code{M-x bashdb} ++after loading Emacs' Grand Unified Debugger, @code{gud}. See ++@ref{Emacs,,Using the BASH debugger from @sc{gnu} Emacs}. And there is ++support in a @value{DDD} for bash. ++ ++@menu ++* Starting the BASH debugger:: How to enter the BASH debugger ++* Quitting the BASH debugger:: How to leave the BASH debugger ++@end menu ++ ++@node Starting the BASH debugger ++@section Starting the BASH debugger ++ ++@emph{Note: it is important to use a debugger-enabled bash. You wil ++get an error message if the debugger is run under a version of BASH ++that does not have debugging support.} ++ ++As mentioned above, one can enter @DBG via Emacs or ++DDD. However you don't have to use either of these. And these still ++need a way on their own to get things started. ++ ++There are in fact two @emph{other} ways to start @value{DBG}. The ++first way is to pass the @samp{--debugger} option to bash with the ++name of your script the scripts arguments following that, or with a ++command string (@code{-c}). ++ ++@example ++bash --debugger @var{script} @var{script-arguments...} ++bash --debugger -c @var{command-string}... ++@end example ++ ++This calls a debugger initialization script. It works much like a ++@BASH login profile which may set variables and define ++functions. But this shell profile is customized for debugging and as ++such arranges for itself to get called before each statement is ++executed. Although there are some problems at present in I/O ++redirection that the method described next doesn't have, it is ++expected that over time more features will be enabled in bash when the ++@samp{--debugger} option is in effect. By default, both debugging in ++Emacs via GUD (@ref{Emacs,,Using the BASH debugger under Emacs}) and ++debugging via @value{DDD} work via this method. ++ ++The form @samp{bash --debugger -c ...} can be used to get into the ++debugger without having to give a script name to debug. Sometimes you ++may want to do this just to see how the debugger works: try some ++debugger commands or maybe get online help. If you run @code{ddd ++--bash} without giving a script name, it in fact uses this form. ++ ++In order for the @samp{--debugger} option to work however, you must ++have the debugger scripts installed in a place where @DBG can ++find them. For this reason, in developing @value{DBG}, I use a second ++method more often; it doesn't require the bash debugger to be ++installed. This method uses another script called @code{bashdb} which ++after taking its own options takes the name of the script to debugged ++and the arguments to pass to that script. Using this method, one ++would start the debugger like this: ++ ++@example ++bash @var{path-to-bashdb}/bashdb @var{bashdb-options} @var{script} @var{script-arguments...} ++@end example ++ ++As with the first method, @code{bash} should be a debugger-enabled ++bash. If @code{bashdb} has the path to bash in it at the top (e.g. via ++@code{#!}), and @code{bashdb} can be found in your program-search ++path, then this might be equivalent to the above: ++ ++@example ++bashdb @var{bashdb-options} @var{script} @var{script-arguments...} ++@end example ++ ++There are two or three disadvantages however of running a debugger ++this way. First @code{$0} will have the value @code{bashdb} rather ++than the script you are trying to run. For some scripts this may ++change the behavior of the debugged script. Second a traceback will ++contain additional lines showing the ``source''-ing of the debugged ++script from @code{bashdb}. And third, although this way works better ++than the first method, over time this way may come into disuse. ++ ++An option that you'll probably need to use if bashdb isn't installed ++but run out of the source code directory is @samp{-L} which specifies ++the directory that contains the debugger script files. ++ ++You can further control how bashdb starts up by using command-line ++options. bashdb itself can remind you of the options available. ++ ++@noindent ++Type ++ ++@example ++bashdb -h ++@end example ++ ++@noindent ++to display all available options and briefly describe their use. ++ ++When the bash debugger is invoked either by the @code{bashdb} ++front-end script or @code{bash --debugging}, the first argument that ++does not have an associated option flag for @code{bashdb} or ++@code{bash} (as the case may be) is used as the name a the script file ++to be debugged, and any following options get passed the debugged ++script. ++ ++Options for the @code{bashdb} front-end are shown in the ++following list. ++ ++@menu ++* Options for the bashdb script:: Options you can pass in starting bashdb ++@end menu ++ ++@node Options for the bashdb script ++@subsection Command-line options for @code{bashdb} script ++ ++You can run @DBG in various alternative modes---for example, in ++batch mode or quiet mode. ++ ++@table @code ++@item -h ++@cindex @code{-h} ++This option causes @value{DBG} to print some basic help and exit. ++ ++@item -V ++@cindex @code{-V} ++This option causes @DBG to print its version number and ++no-warranty blurb, and exit. ++ ++@item -n ++@cindex @code{-n} ++Do not execute commands found in any initialization files. Normally, ++@BASH executes the commands in these files after all the command ++options and arguments have been processed. @xref{Command Files,,Command ++files}. ++ ++@item -q ++@cindex @code{-q} ++``Quiet''. Do not print the introductory and copyright messages. These ++messages are also suppressed in batch mode. ++ ++@item -t ++@cindex @code{-t} ++``Terminal output''. Set the file or terminal that you want debugger command ++output to go to. Note that the debugger output is independent of the ++debugged script output. ++ ++@item -L @var{directory} ++@cindex @code{-L} ++Set directory where debugger files reside to @var{directory}. The ++default location is @code{../lib/bashdb} relative to the place that ++the bashdb script is located. For example if bashdb is located in ++@code{/usr/local/bin/bashdb}, the default library location will be ++@code{/usr/local/lib/bashdb} which may or may not exist. If it doesn't ++you'll get an error when you run bashdb. Only if the default location ++is incorrect, should you need to use the @code{-L} option. ++ ++@item -T @var{directory} ++@cindex @code{-T} ++Set directory to use for writing temporary files. ++ ++@end table ++ ++@node Quitting the BASH debugger ++@section Quitting the BASH debugger ++ ++@cindex interrupt ++An interrupt (often @kbd{C-c}) does not exit from @value{DBG}, but ++rather terminates the action of any @DBG command that is in ++progress and returns to @value{DBG} command level. Inside a debugger ++command interpreter, use @code{quit} command (@pxref{Quit, ,Quitting ++the BASH debugger}). ++ ++@node Running ++@chapter Script Setup inside the BASH Debugger ++ ++@menu ++* Starting:: Starting your script ++* Command Files:: Command files ++* Arguments:: Your script's arguments ++* Input/Output:: Your script's input and output ++* Script/Debugger Interaction:: Keeping out of each other's harm ++@end menu ++ ++@need 2000 ++@node Starting ++@section Starting your script ++@cindex starting ++@cindex running ++ ++@table @code ++@kindex restart ++@kindex run @r{(@code{restart})} ++@kindex R @r{(@code{restart})} ++@item restart @ovar{options to debugged script} ++@itemx run ++@itemx R ++Use the @code{restart} command to restart your script under ++@value{DBG}. Without any arguments, the script name and parameters ++from the last invocation are used. @value{dBGP} tries to maintain the ++settings, watchpoints, breakpoints, actions and so on. Internally it ++uses line numbers and filenames to record he position of interesting ++places in your porgram; so if your program changes some or all of ++these numbers may be off. Environment variable ++@code{BASHDB_RESTART_FILE} is and a temporary file are used to signal ++a restart, so you shouldn't uset @code{BASHDB_RESTART_FILE} (or any ++environment variable starting with @code{BASHDB_}. ++ ++@end table ++ ++@node Command Files ++@section Command files ++ ++@cindex command files ++A command file for @DBG is a file of lines that are @DBG ++commands. Comments (lines starting with @kbd{#}) may also be included. ++An empty line in a command file does nothing; it does not mean to repeat ++the last command, as it would from the terminal. ++ ++@cindex init file ++@cindex @file{.bashdbinit} ++@cindex @file{bashdb.ini} ++When you start @value{DBG}, it automatically executes commands from its ++@dfn{init files}, normally called @file{.bashdbinit}@footnote{The DJGPP ++port of @DBG uses the name @file{bashdb.ini} instead, due to the ++limitations of file names imposed by DOS filesystems.}. ++During startup, @DBG does the following: ++ ++@enumerate ++@item ++Reads the init file (if any) in your home directory@footnote{On ++DOS/Windows systems, the home directory is the one pointed to by the ++@code{HOME} environment variable.}. ++ ++@item ++Processes command line options and operands. ++ ++@item ++Reads the init file (if any) in the current working directory. ++ ++@item ++Reads command files specified by the @samp{-x} option. ++@end enumerate ++ ++The init file in your home directory can set options (such as @samp{set ++complaints}) that affect subsequent processing of command line options ++and operands. Init files are not executed if you use the @samp{-x} ++option (@pxref{Options for the bashdb script, ,bashdb script options}). ++ ++@cindex init file name ++On some configurations of @value{DBG}, the init file is known by a ++different name (these are typically environments where a specialized ++form of @DBG may need to coexist with other forms, hence a ++different name for the specialized version's init file). These are the ++environments with special init file names: ++ ++You can also request the execution of a command file with the ++@code{source} command: ++ ++@table @code ++@kindex source ++@item source @var{filename} ++Execute the command file @var{filename}. ++@end table ++ ++The lines in a command file are executed sequentially. They are not ++printed as they are executed. If there is an error, execution ++proceeds to the next command in the file. ++ ++@node Arguments ++@section Your script's arguments ++ ++@cindex arguments (to your script) ++The arguments to your script can be specified by the arguments of the ++@code{restart} command. ++They are passed to a shell, which expands wildcard characters and ++performs redirection of I/O, and thence to your script. ++ ++@code{restart} with no arguments uses the same arguments used by the previous ++@code{restart}, or those set by the @code{set args} command.. ++ ++@table @code ++@kindex set args ++@item set args ++Specify the arguments to be used the next time your program is run. If ++@code{set args} has no arguments, @code{restart} executes your program ++with no arguments. Once you have run your program with arguments, ++using @code{set args} before the next @code{restart} is the only way to run ++it again without arguments. ++ ++@kindex show args ++@item show args ++Show the arguments to give your program when it is started. ++@end table ++ ++@node Input/Output ++@section Your script's input and output ++ ++@cindex redirection ++@cindex I/O ++@cindex terminal ++By default, the script you run under the BASH debugger does input and ++output to the same terminal that BASH uses. Before running the script ++to be debugged, the debugger records the tty that was in effect. All ++of its output is then written to that. However you can change this ++when using the @samp{bashdb} script using the @samp{-t} option. ++ ++ ++@table @code ++@kindex info terminal ++@item info terminal ++Displays information recorded by @DBG about the terminal modes your ++program is using. ++@end table ++ ++@kindex tty ++@cindex controlling terminal ++Another way to specify where your script should do input and output is ++with the @code{tty} command. This command accepts a file name as ++argument, and causes this file to be the default for future @code{restart} ++commands. It also resets the controlling terminal for the child ++process, for future @code{restart} commands. For example, ++ ++@example ++tty /dev/ttyb ++@end example ++ ++@noindent ++directs that processes started with subsequent @code{restart} commands ++default to do input and output on the terminal @file{/dev/ttyb} and have ++that as their controlling terminal. ++ ++An explicit redirection in @code{restart} overrides the @code{tty} command's ++effect on the input/output device, but not its effect on the controlling ++terminal. ++ ++When you use the @code{tty} command or redirect input in the @code{restart} ++command, only the input @emph{for your script} is affected. The input ++for @DBG still comes from your terminal. ++ ++@node Script/Debugger Interaction ++@section Script/Debugger Interaction ++ ++@value{dBGP} and your program live in the same variable space so ++to speak. @BASH does not have a notion of module scoping or lexical ++hiding (yet) and this then imposes some additional care and awareness. ++ ++Most of the variables and functions used inside @DBG start ++@code{_Dbg_}, so please don't use variables or functions with these ++names in your program. ++ ++@emph{Note: there are some other variables that begin with just an ++underscore (@code{_}); over time these will be phased out. But until ++then, avoid those or consult what is used by the debugger. Run ++@samp{bashdb --debugger -c 'declare -p'} to list all the variables in ++use including those used by the debugger.} ++ ++A number of @DBG environment variables are also reserved for ++use; these start with @code{BASHDB_}. For example: ++@env{BASHDB_INPUT}, @env{BASHDB_LEVEL} and, ++@env{BASHDB_QUIT_ON_QUIT} (@pxref{Debug, ,Debug}), ++@env{BASHDB_RESTART_FILE} (@pxref{Starting, ,Starting}), to name a ++few. Finally, there are some @BASH environment dynamic ++variables and these start with @env{BASH_}. For example ++@env{BASH_SUBSHELL} (@pxref{Debug, ,Debug}), @env{BASH_COMMAND} ++(@pxref{Command Display, ,Command Display}), @env{BASH_LINENO}, and ++@env{BASH_SOURCE} to name a few. ++ ++In order to do its work @value{dBGP} sets up a @code{DEBUG} ++trap. Consequently a script shouldn't reset this or the debugger will ++lose control. @value{dBGP} also sets up an @code{EXIT} handler so that ++it can gain control after the script finishes. Another signal ++intercepted is the an interrupt or @code{INT} signal. For more ++information about signal handling, @pxref{Signals, ,Signals} ++ ++@node Debugger Command Reference ++@chapter BASH Debugger Command Reference ++ ++You can abbreviate the long name of @DBG command to the first ++few letters of the command name, if that abbreviation is unambiguous; ++and you can repeat the @code{next} o r@code{step} commands by typing ++just @key{RET}. Some commands which require a parameter, such as ++@code{print} remember the argument that was given to them. ++ ++@menu ++* Command Syntax:: How to give commands to the BASH debugger ++* Help:: How to ask for help (help) ++* Quit:: Leaving the debugger (quit) ++* Information:: Status and Debugger settings (info, show) ++* Stopping:: Stopping and continuing (break, watch, step, cont...) ++* Stack:: Examining the stack (where, up, down, frame) ++* List:: Printing source files (list) ++* Search:: Searching source files (/pat/ ?pat?) ++* Data:: Examining data (print, examine, info variables) ++* Evaluation/Execution:: Arbitrary execution (eval, shell) ++* Auto Display:: Executing expressions on stop (display, undisplay) ++* Controlling bashdb:: Controlling bashdb (set, history) ++@end menu ++ ++@node Command Syntax ++@section Command syntax ++ ++A BASH debugger command is a single line of input. There is no limit on ++how long it can be. It starts with a command name, which is followed by ++arguments whose meaning depends on the command name. For example, the ++command @code{step} accepts an argument which is the number of times to ++step, as in @samp{step 5}. You can also use the @code{step} command ++with no arguments. Some commands do not allow any arguments. ++ ++@cindex repeating next/step commands ++@kindex RET @r{(repeat last command)} ++A blank line as input to @DBG (typing just @key{RET}) means to ++repeat the previous next or step command. ++ ++@kindex # @r{(a comment)} ++@cindex comment ++Any text from a @kbd{#} to the end of the line is a comment; it does ++nothing. This is useful mainly in command files (@pxref{Command ++Files,,Command files}). ++ ++@node Help ++@section Getting help (@samp{help}) ++@cindex online documentation ++ ++Once inside the BASH debugger, you can always ask it for information on its ++commands, using the command @code{help}. ++ ++@table @code ++@kindex h @r{(@code{help})} ++@item help ++@itemx h ++You can use @code{help} (abbreviated @code{h}) with no arguments to ++display a short list of named classes of commands: ++@end table ++ ++@flushleft ++@smallexample ++bashdb<0> @b{help} ++bashdb commands: ++List/search source lines: Control script execution: ++------------------------- ------------------------- ++ l [start|.] [cnt] List cnt lines T [n] Stack trace ++ from line start s [n] Single step [n times] ++ l sub List source code fn n [n] Next, steps over subs ++ - or . List previous/current line / Repeat last n or s ++ w [line] List around line c [linespec] Continue [to linespec] ++ f filename View source in file L List all breakpoints ++ /pat/ Search forward for pat b linespec Set breakpoint ++ ?pat? Search backward for pat del [n].. or D Delete a/all breaks ++ by entry number ++Debugger controls: skip skip execution of cmd ++------------------------- cl linespec Delete breakpoints by ++ H [num] Show last num commands line spec ++ q [exp] or ^D Quit returning exp R [args] Attempt a restart ++ info [cmd] Get info on cmd. u [n] Go up stack by n or 1. ++ !n or hi n Run debugger history n do [n] Go down stack by n or 1. ++ h or ? [cmd] Get help on command W [var] Add watchpoint. If no ++ info [cmd] Get info on cmd no expr, delete all ++ show [cmd] Show settings We [expr] Add Watchpoint arith ++ expr ++ so file read in dbg commands t Toggle trace ++ en/di n enable/disable brkpt, ++ set x y set a debugger variable watchpoint, or display ++ e bash-cmd evaluate a bash command tb linespec Add one-time break ++ disp expr add a display expr a linespec cmd eval "cmd" at linespec ++ M Show module versions A delete all actions ++ x expr evaluate expression ret jump out of fn or source ++ (via declare, let, eval) finish execute until return ++ deb debug into another cond n exp set breakpoint condition ++ shell script ++ !! cmd [args] execute shell command "cmd" with "args" ++ ++Data Examination: also see e, t, x ++------------------------- ++ p variable Print variable ++ V [[!]pat] List variable(s) matching or not (!) matching pattern pat ++ S [[!]pat] List subroutine names [not] matching pattern pat ++ ++Readline command line editing (emacs/vi mode) is available ++For more help, type h or consult online-documentation. ++@end smallexample ++@end flushleft ++@c the above line break eliminates huge line overfull... ++ ++@table @code ++@item help @var{command} ++With a command name as @code{help} argument, the BASH debugger displays ++short information on how to use that command. ++ ++@example ++bashdb<0> @b{help list} ++l linespec List window lines starting at linespec. ++l min incr List incr lines starting at 'min' linespec. ++l List next window of lines. ++l . Same as above. ++ Long command name: list. ++@end example ++ ++In addition to @code{help}, you can use the debugger command ++@code{info} to inquire about the state of your script, or the state of ++@DBG itself. The listings under @code{info} in the Index ++point to all the sub-commands. @xref{Command Index}. ++@end table ++ ++@c @group ++@table @code ++@kindex info ++@kindex i @r{(@code{info})} ++@item info ++This command (abbreviated @code{i}) is for describing the state of ++your program. For example, you can list the arguments given to your ++script with @code{info args}, or list the breakpoints you have set ++with @code{info breakpoints}. You can get a complete list of the ++@code{info} sub-commands with @w{@code{help info}}. ++ ++@example ++bashdb<0> @b{info} ++Info subcommands are: args breakpoints display files source sources stack terminal variables watchpoints ++bashdb<1> @b{info source} ++Current script file is parm.sh ++Contains 34 lines. ++@end example ++@end table ++ ++ ++@node Quit ++@section Quitting the BASH debugger (@samp{quit}) ++ ++@table @code ++@kindex quit @r{[}@var{expression} @ovar{subshell-levels}@r{]} ++@kindex q @r{(@code{quit})} ++@item quit @ovar{expression} ++@item quit @r{[}@var{expression} @ovar{subshell-levels}@r{]} ++@itemx q ++ ++To exit @value{DBG}, use the @code{quit} command (abbreviated ++@code{q}), or type an end-of-file character (usually @kbd{C-d}). If ++you do not supply @var{expression}, @DBG will try to terminate ++normally or with exit code 0. Otherwise it will terminate using the ++result of @var{expression} as the exit code. ++ ++A simple @code{quit} tries to terminate all nested subshells that may ++be in effect. If you are nested a subshell, this is normally ++indicated in a debugger prompt by the number of parentheses that the ++history number is inside --- no parenthesis means there is no subshell ++in effect. The dynamic variable @env{BASH_SUBSHELL} also contains the ++number of subshells in effect. ++ ++If you want only to terminate some number of subshells but not all of ++them, you can give a count of the number of subshells to leave after ++the return-code expression. To leave just one level of subshell ++@code{return} does almost the same thing. (See @pxref{Returning, ++,Returning}) There is a subtle difference between the two though: ++@code{return} will leave you at the beginning of the next statement ++while @code{quit} may leave you at the place the subshell was invoked ++which may be in the middle of another command such as an assingment ++statement or condition test. ++ ++If the environment variable @code{BASHDB_QUIT_ON_QUIT} is set, when the ++program terminates, the debugger will also terminate too. This may be ++useful if you are debugging a script which calls another script and ++you want this inner script just to return to the outer script. ++@end table ++ ++@node Stopping ++@section Stopping and Resuming Execution (@samp{break}, @samp{watch}, @samp{step}, @samp{cont}) ++ ++One important use of a debugger is to stop your program before it ++terminates so that, if your script might run into trouble, you can ++investigate and find out why. However should your script accidently ++continue to termination, @DBG has arranged for it not to leave the ++debugger without your explicit instruction. That way, you can restart ++the program using the same command arguments. ++ ++Inside @value{DBG}, your script may stop for any of several reasons, ++such as a signal, a breakpoint, or reaching a new line after a ++debugger command such as @code{step}. You may then examine and ++change variables, set new breakpoints or remove old ones, and then ++continue execution. ++ ++@menu ++* Breakpoints:: Breakpoints, watchpoints (break, watch, clear) ++* Resuming Execution:: Resuming execution (continue, step, next, skip, debug) ++* Signals:: Signals ++@end menu ++ ++@node Breakpoints ++@subsection Breakpoints, watchpoints ++ ++@cindex breakpoints ++A @dfn{breakpoint} makes your script stop whenever a certain point in ++the program is reached. For each breakpoint, you can add conditions to ++control in finer detail whether your script stops. ++ ++You can set breakpoints with the @code{break} command and its variants ++(@pxref{Set Breaks, ,Setting breakpoints}), to specify the place where ++your script should stop by line number. or function name in the ++debugged script. ++ ++@cindex watchpoints ++@cindex breakpoint on variable modification ++A @dfn{watchpoint} is a special breakpoint that stops your script when ++the value of an expression changes. There is a different command to ++set watchpoints (@pxref{Set Watchpoints, ,Setting watchpoints}). ++ ++But aside from that, you can manage a watchpoint like any other ++breakpoint: you delete enable, and disable both breakpoints and ++watchpoints using the same commands. ++ ++You can arrange to have values from your program displayed automatically ++whenever @value{BASH} stops at a breakpoint. @xref{Auto Display,, ++Automatic display}. ++ ++@cindex breakpoint numbers ++@cindex numbers for breakpoints ++@DBG assigns a number to each breakpoint and watchpoint when ++you create it; these numbers are successive integers starting with ++one. In many of the commands for controlling various features of ++breakpoints you use the breakpoint number to say which breakpoint you ++want to change. Each breakpoint may be @dfn{enabled} or ++@dfn{disabled}; if disabled, it has no effect on your script until you ++enable it again. ++ ++@cindex watchpoints numbers ++@cindex numbers for watchpoints ++Watchpoint numbers however are distiguished from breakpoint numbers by ++virtue of their being suffixed with the either an upper- or lower-case ++`W'. For example, to enable breakpoint entry 0 along with watchpoint ++entry 1 you would write @samp{enable 1 2w}, the ``2w'' refers to the ++watchpoint; ``2W'' would work just as well. ++ ++@ifset FINISHED ++@cindex breakpoint ranges ++@cindex ranges of breakpoints ++Some @DBG commands accept a range of breakpoints on which to ++operate. A breakpoint range is either a single breakpoint number, like ++@samp{5}, or two such numbers, in increasing order, separated by a ++hyphen, like @samp{5-7}. When a breakpoint range is given to a command, ++all breakpoint in that range are operated on. ++@end ifset ++ ++@menu ++* Set Breaks:: Setting breakpoints (break, tbreak) ++* Set Watchpoints:: Setting watchpoints (watch, watche) ++* Delete Breaks:: Deleting breakpoints (delete, clear) ++* Disabling:: Disabling breakpoints (disable, enable) ++* Conditions:: Break conditions (condition) ++@end menu ++ ++@node Set Breaks ++@subsubsection Setting breakpoints (@samp{break} @samp{tbreak}) ++ ++@kindex break ++@kindex b @r{(@code{break})} ++@cindex latest breakpoint ++Breakpoints are set with the @code{break} command (abbreviated ++@code{b}). ++ ++@table @code ++@item break @var{function} ++Set a breakpoint at entry to function @var{function}. ++ ++@item break @var{linenum} ++Set a breakpoint at line @var{linenum} in the current source file. ++The current source file is the last file whose source text was printed. ++The breakpoint will stop your script just before it executes any of the ++code on that line. ++ ++@item break @var{filename}:@var{linenum} ++Set a breakpoint at line @var{linenum} in source file @var{filename}; ++@var{filename} has to be one of the files previously read in and has ++to be specified exactly as the name used when read in. For a list of ++read-in files, use the @samp{info files} command. ++ ++@ifset FINISHED ++@item break ++When called without any arguments, @code{break} sets a breakpoint at ++the next instruction to be executed in the selected stack frame ++(@pxref{Stack, ,Examining the Stack}). In any selected frame but the ++innermost, this makes your script stop as soon as control returns to ++that frame. If you use @code{break} without an argument in the ++innermost frame, @DBG stops the next time it reaches the ++current location; this may be useful inside loops. ++@end ifset ++ ++@item break @dots{} if @var{cond} ++Set a breakpoint with condition @var{cond}; evaluate the expression ++@var{cond} each time the breakpoint is reached, and stop only if the ++value is nonzero---that is, if @var{cond} evaluates as true. The ++expression is evaluated via the @code{let} builtin funtion. ++@samp{@dots{}} stands for one of the possible arguments described ++above (or no argument) specifying where to break. The word ``if'' is ++often optional and is necessary only @samp{@dots{}} is ++omitted. @xref{Conditions, ,Break conditions}, for more information on ++breakpoint conditions. ++ ++Examples: ++@example ++bashdb<0> @b{break fn1} ++Breakpoint 1 set in file parm.sh, line 3. ++bashdb<1> @b{break 28} ++Breakpoint 2 set in file parm.sh, line 28. ++bashdb<2> @b{break parm.sh:29} ++Breakpoint 3 set in file parm.sh, line 29. ++bashdb<3> @b{break 28 if x==5} ++Breakpoint 4 set in file parm.sh, line 28. ++@end example ++ ++@kindex tbreak ++@item tbreak @var{args} ++Set a breakpoint enabled only for one stop. @var{args} are the ++same as for the @code{break} command, and the breakpoint is set in the same ++way, but the breakpoint is automatically deleted after the first time your ++program stops there. @xref{Disabling, ,Disabling breakpoints}. ++ ++@kindex info breakpoints ++@cindex @code{$_} and @code{info breakpoints} ++@item info breakpoints @ovar{n} ++@itemx info break @ovar{n} ++@itemx info watchpoints @ovar{n} ++Print a table of all breakpoints, watchpoints set and not deleted, ++with the following columns for each breakpoint: ++ ++@table @emph ++@item Breakpoint Numbers (@samp{Num}) ++@item Enabled or Disabled (@samp{Enb}) ++Enabled breakpoints are marked with @samp{1}. @samp{0} marks breakpoints ++that are disabled (not enabled). ++@item Count ++The number of times that breakpoint or watchpoint has been hit. ++@item Condition ++The arithmetic expression ++@item File and Line (@samp{file:line}) ++The filename and line number inside that file where of breakpoint in ++the script. The file and line are separated with a colon. ++@end table ++ ++@noindent ++If a breakpoint is conditional, @code{info break} shows the condition on ++the line following the affected breakpoint; breakpoint commands, if any, ++are listed after that. ++ ++@noindent ++@code{info break} displays a count of the number of times the breakpoint ++has been hit. ++ ++@code{info break} with a breakpoint number @var{n} as argument lists ++only that breakpoint. ++ ++Examples: ++@example ++bashdb<4> @b{info break} ++Breakpoints at following places: ++Num Type Disp Enb What ++1 breakpoint keep y parm.sh:3 ++2 breakpoint keep y parm.sh:28 ++3 breakpoint keep y parm.sh:29 ++4 breakpoint keep y parm.sh:28 ++No watch expressions have been set. ++bashdb<5> @b{info break 4} ++Num Type Disp Enb What ++4 breakpoint keep y parm.sh:28 ++No watch expressions have been set. ++@end example ++@end table ++ ++ ++@ifset FINISHED ++This is especially useful in conjunction with the ++@code{ignore} command. You can ignore a large number of breakpoint ++hits, look at the breakpoint info to see how many times the breakpoint ++was hit, and then run again, ignoring one less than that number. This ++will get you quickly to the last hit of that breakpoint. ++@end ifset ++ ++@DBG allows you to set any number of breakpoints at the same place in ++your script. There is nothing silly or meaningless about this. When ++the breakpoints are conditional, this is even useful ++(@pxref{Conditions, ,Break conditions}). ++ ++@node Set Watchpoints ++@subsubsection Setting watchpoints (@samp{watch}, @samp{watche}) ++ ++@cindex setting watchpoints ++You can use a watchpoint to stop execution whenever the value of an ++expression changes, without having to predict a particular place where ++this may happen. As with the @code{print} (@pxref{Data,,Examining ++Data}), the idiosyncracies of a BASH or any POSIX shell derivative ++suggest using two commands. The @code{watch} command is just for a ++single variables; the @code{watche} command uses the builtin ``let'' ++command to evaluate an expression. If the variable you are tracking ++can take a string value, issuing something like @samp{watch foo} will ++not have the desired effect---any string assignment to @code{foo} will ++have a value 0 when it is assigned via ``let.'' ++ ++@table @code ++@kindex watch ++@item watch @var{var} ++Set a watchpoint for a variable. @DBG will break when the ++value of @var{var} changes. In this command do not add a leading ++dollar symbol to @var{var}. ++ ++@item watche @var{expr} ++Set a watchpoint for an expression via the builtin ``let'' command. ++@DBG will break when @var{expr} is written into by the program ++and its value changes. Not that this may not work for tracking ++arbitrary string value changes. For that use @code{watch} described ++earlier. ++@end table ++ ++@node Delete Breaks ++@subsubsection Deleting breakpoints (@samp{clear}, @samp{delete}) ++ ++@cindex clearing breakpoints, watchpoints ++@cindex deleting breakpoints, watchpoints ++It is often necessary to eliminate a breakpoint or watchpoint once it ++has done its job and you no longer want your script to stop there. ++This is called @dfn{deleting} the breakpoint. A breakpoint that has ++been deleted no longer exists; it is forgotten. ++ ++With the @code{clear} command you can delete breakpoints according to ++where they are in your script. With the @code{delete} command you can ++delete individual breakpoints, or watchpoints by specifying their ++breakpoint numbers. @emph{Note: as described below under the ``clear'' ++command, ``d'' is an alias for ``clear'', not ``delete''. } ++ ++It is not necessary to delete a breakpoint to proceed past it. @DBG ++automatically ignores breakpoints on the first instruction to be executed ++when you continue execution. ++ ++@table @code ++@kindex clear ++@kindex d @r{(@code{clear})} ++@item clear ++Delete any breakpoints at the next instruction to be executed in the ++selected stack frame (@pxref{Selection, ,Selecting a frame}). When ++the innermost frame is selected, this is a good way to delete a ++breakpoint where your script just stopped. ++ ++It may seem odd that we have an alias ``d'' for ``clear.'' It so ++happens that Perl's debugger use ``d'' for its delete command and the ++delete concept in Perl's debugger corresponds to ``clear'' in ++GDB. (Perl doesn't have a notion of breakpoint entry numbers). So in ++order to be compatible with both debugger interfaces, ``d'' is used as ++an alias for ``clear.'' Clear? ++ ++@item clear @var{function} ++@itemx clear @var{filename}:@var{function} ++Delete any breakpoints set at entry to the function @var{function}. ++ ++@item clear @var{linenum} ++@itemx d @var{linenum} ++@ifset FINISHED ++@itemx clear @var{filename}:@var{linenum} ++@end ifset ++Delete any breakpoints set at or within the code of the specified line. ++ ++@cindex delete breakpoints ++@kindex delete ++@kindex de @r{(@code{delete})} ++@item delete @ovar{breakpoints} ++Delete the breakpoints, watchpoints specified as arguments. ++ ++If no argument is specified, delete all breakpoints (@DBG asks ++confirmation, unless you have @code{set confirm off}). You can ++abbreviate this command as @code{de}. ++ ++ ++Note that for compatibility with Perl's debugger, @code{d} means ++something else: @code{clear}. ++ ++@end table ++ ++@node Disabling ++@subsubsection Disabling breakpoints (@samp{disable}, @samp{enable}) ++ ++Rather than deleting a breakpoint or watchpoint, you might ++prefer to @dfn{disable} it. This makes the breakpoint inoperative as if ++it had been deleted, but remembers the information on the breakpoint so ++that you can @dfn{enable} it again later. ++ ++You disable and enable breakpoints, watchpoints, and catchpoints with ++the @code{enable} and @code{disable} commands, optionally specifying one ++or more breakpoint numbers as arguments. Use @code{info break} or ++@code{info watch} to print a list of breakpoints, watchpoints, and ++catchpoints if you do not know which numbers to use. ++ ++A breakpoint, watchpoint, or catchpoint can have any of four different ++states of enablement: ++ ++@itemize @bullet ++@item ++Enabled. The breakpoint stops your program. A breakpoint set ++with the @code{break} command starts out in this state. ++@item ++Disabled. The breakpoint has no effect on your program. ++@item ++Enabled once. The breakpoint stops your program, but then becomes ++disabled. ++@item ++Enabled for deletion. The breakpoint stops your program, but ++immediately after it does so it is deleted permanently. A breakpoint ++set with the @code{tbreak} command starts out in this state. ++@end itemize ++ ++You can use the following commands to enable or disable breakpoints, ++watchpoints, and catchpoints: ++ ++@table @code ++@kindex disable breakpoints ++@kindex disable ++@kindex dis @r{(@code{disable})} ++@item disable @ovar{breakpoints} ++Disable the specified breakpoints---or all breakpoints, if none are ++listed. A disabled breakpoint has no effect but is not forgotten. All ++options such as ignore-counts, conditions and commands are remembered in ++case the breakpoint is enabled again later. You may abbreviate ++@code{disable} as @code{dis}. ++ ++@kindex enable breakpoints ++@kindex enable ++@item enable @ovar{breakpoints} ++Enable the specified breakpoints (or all defined breakpoints). They ++become effective once again in stopping your program. ++@end table ++ ++@c FIXME: I think the following ``Except for [...] @code{tbreak}'' is ++@c confusing: tbreak is also initially enabled. ++Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks, ++,Setting breakpoints}), breakpoints that you set are initially enabled; ++subsequently, they become disabled or enabled only when you use one of ++the commands above. (The command @code{until} can set and delete a ++breakpoint of its own, but it does not change the state of your other ++breakpoints; see @ref{Resuming Execution, ,Resuming Execution}.) ++ ++@node Conditions ++@subsubsection Break conditions (@samp{condition}) ++@cindex conditional breakpoints ++@cindex breakpoint conditions ++ ++The simplest sort of breakpoint breaks every time your script reaches ++a specified place. You can also specify a @dfn{condition} for a ++breakpoint. A condition is just a BASH expression. ++ ++Break conditions can be specified when a breakpoint is set, by using ++@samp{if} in the arguments to the @code{break} command. @xref{Set ++Breaks, ,Setting breakpoints}. A breakpoint with a condition ++evaluates the expression each time your script reaches it, and your ++script stops only if the condition is @emph{true}. ++ ++@cindex one-time breakpoints ++There is also a notion of a ``one-time'' breakpoint which gets deleted ++as soon as it is hit, so that that breakpoint is executed once only. ++ ++Conditions are also accepted for watchpoints; you may not need them, ++since a watchpoint is inspecting the value of an expression anyhow---but ++it might be simpler, say, to just set a watchpoint on a variable name, ++and specify a condition that tests whether the new value is an interesting ++one. ++ ++Break conditions can be specified when a breakpoint is set, by using ++@samp{if} in the arguments to the @code{break} command. @xref{Set ++Breaks, ,Setting breakpoints}. They can also be changed at any time ++with the @code{condition} command. ++ ++@ifset FINISHED ++You can also use the @code{if} keyword with the @code{watch} command. ++The @code{catch} command does not recognize the @code{if} keyword; ++@code{condition} is the only way to impose a further condition on a ++catchpoint. ++@end ifset ++ ++@table @code ++@kindex condition ++@item condition @var{bnum} @var{expression} ++Specify @var{expression} as the break condition for breakpoint ++@var{bnum}. After you set a condition, breakpoint @var{bnum} stops ++your program only if the value of @var{expression} is true (nonzero). ++ ++@item condition @var{bnum} ++Remove the condition from breakpoint number @var{bnum}. It becomes ++an ordinary unconditional breakpoint. ++@end table ++ ++@ifset FINISHED ++When you use @code{condition}, @DBG checks @var{expression} ++immediately for syntactic correctness, and to determine whether ++symbols in it have referents in the context of your breakpoint. If ++@var{expression} uses symbols not referenced in the context of the ++breakpoint, @DBG prints an error message: ++ ++@example ++No symbol "foo" in current context. ++@end example ++@end ifset ++ ++@noindent ++@BASH does ++not actually evaluate @var{expression} at the time the @code{condition} ++command (or a command that sets a breakpoint with a condition, like ++@code{break if @dots{}}) is given, however. ++ ++Examples; ++@example ++condition 1 x>5 # Stop on breakpoint 0 only if x>5 is true. ++condition 1 # Change that! Unconditinally stop on breakpoint 1. ++@end example ++ ++@node Resuming Execution ++@subsection Resuming Execution ++ ++@cindex stepping ++@cindex continuing ++@cindex resuming execution ++@dfn{Continuing} means resuming program execution until your script ++completes normally. In contrast, @dfn{stepping} means executing just ++one more ``step'' of your script, where ``step'' may mean either one ++line of source code. Either when continuing or when stepping, ++your script may stop even sooner, due to a breakpoint or a signal. ++ ++@menu ++* Step:: running the next statement (step) ++* Next:: running the next statement skipping over functions (skip) ++* Finish:: running until the return of a function or ``source'' (finish) ++* Skip:: skipping the next statement (skip) ++* Continue:: continuing execution (continue) ++* Debug:: debugging into another program (debug) ++* Returning:: returning ++@end menu ++ ++@node Step ++@subsubsection Step (@samp{step}) ++@table @code ++@kindex step ++@kindex s @r{(@code{step})} ++@item step ++Continue running your script until control reaches a different source ++line, then stop it and return control to @value{DBG}. This command is ++abbreviated @code{s}. ++ ++The @code{step} command only stops at the first instruction of a source ++line. This prevents the multiple stops that could otherwise occur in ++@code{switch} statements, @code{for} loops, etc. @code{step} continues ++to stop if a function that has debugging information is called within ++the line. In other words, @code{step} @emph{steps inside} any functions ++called within the line. ++ ++@item step @ovar{count} ++Continue running as in @code{step}, but do so @var{count} times. If a ++breakpoint is reached, or a signal not related to stepping occurs before ++@var{count} steps, stepping stops right away. ++@end table ++ ++@node Next ++@subsubsection Next (@samp{next}) ++@table @code ++@kindex next ++@kindex n @r{(@code{next})} ++@item next @ovar{count} ++Continue to the next source line in the current (innermost) stack frame. ++This is similar to @code{step}, but function calls that appear within ++the line of code are executed without stopping. Execution stops when ++control reaches a different line of code at the original stack level ++that was executing when you gave the @code{next} command. This command ++is abbreviated @code{n}. ++ ++An argument @var{count} is a repeat count, as for @code{step}. ++@end table ++ ++@node Finish ++@subsubsection Finish (@samp{finish}) ++@table @code ++@kindex finish ++@item finish ++Continue running until just after function returns. @emph{Currently, ++the line shown on a return is the function header, unless the ++@code{return} builtin function is executed in which case it is the ++line number of the @code{return} function.} ++ ++Contrast this with the @code{return} command (@pxref{Returning, ++,Returning from a function}) and the @code{quit} (@pxref{Quitting the ++BASH debugger, ,Quitting the BASH debugger}). ++ ++@end table ++ ++@node Skip ++@subsubsection Skip (@samp{skip}) ++@table @code ++@kindex skip ++@item skip @ovar{count} ++Skip exection of the next source line. ++This may be useful if you have an action that ``fixes'' existing code in ++the script. The @code{debug} command internally uses the @code{skip} command ++to skip over existing non-debugged invocation that was presumably just ++run. ++@end table ++ ++@node Continue ++@subsubsection Continue (@samp{continue}) ++@table @code ++@kindex continue ++@kindex c @r{(@code{continue})} ++@item continue @ovar{linespec} ++@itemx c @ovar{line-number} ++Resume program execution, at the address where your script last ++stopped; any breakpoints set at that address are bypassed. The ++optional argument @var{linespec} allows you to specify a linespec (a ++line number, function, or filename linenumber combination) to set. A ++one-time breakpoint is deleted when that breakpoint is reached. Should ++the program stop before that breakpoint is reached, for example, ++perhaps another breakpoint or watchpoint is reached first, in a ++listing of the breakpoints you will see this entry with the condition ++9999 which indicates a one-time breakpoint. ++@end table ++ ++To resume execution at a different place, you can use @code{return} ++(@pxref{Returning, ,Returning from a function}) to go back to the ++calling function or sourced script. If you are nested inside a ++subshell, @code{quit} with a value for the number of subshells to ++exit also functions like a return. ++ ++A typical technique for using stepping is to set a breakpoint ++(@pxref{Breakpoints, ,Breakpoints; watchpoints}) at the ++beginning of the function or the section of your script where a problem ++is believed to lie, run your script until it stops at that breakpoint, ++and then step through the suspect area, examining the variables that are ++interesting, until you see the problem happen. ++ ++@node Debug ++@subsubsection Debug (@samp{debug}) ++@table @code ++@kindex debug ++@item debug @ovar{script-name} ++Debug into @var{script-name}. If no name is given the current source line ++is used. In either case the options are prepended to cause the ++debugger to run. ++ ++The nesting level of the debugger is saved inside environment variable ++@code{BASHDB_LEVEL}. The debugger prompt indicates the level of nesting ++by enclosing the history in that many nestings of @code{<>} symbols. ++ ++@end table ++ ++@node Returning ++@subsubsection Returning from a function, sourced file, or subshell (@samp{return}) ++ ++@table @code ++@cindex returning from a function, sourced file or subshell ++@kindex return ++@item return ++@itemx return ++You can cancel execution of a function call or a subshell with the ++@code{return} command. ++@end table ++ ++The @code{return} command does not resume execution; it leaves the ++program stopped in the state that would exist if the function had just ++returned. See also the @code{quit} command (@ref{Quit, ,Quitting the ++BASH debugger}). In some situations @code{return} is similar to ++@code{quit}: in particular when the script is @emph{not} currenlty ++inside in a function and the number of subshells in effect is 0, or ++when a subshell count of 1 is given on the @code{quit} command. ++ ++@ifset FINISHED ++In contrast, the @code{finish} command (@pxref{Continuing ++and Stepping, ,Continuing and stepping}) resumes execution until the ++selected stack frame returns naturally. ++@end ifset ++ ++@node Signals ++@subsection Signals ++@cindex signals ++ ++A signal is an asynchronous event that can happen in a program. The ++operating system defines the possible kinds of signals, and gives each ++kind a name and a number. For example, in Unix @code{SIGINT} is the ++signal a program gets when you type an interrupt character (often ++@kbd{C-c}); @code{SIGALRM} occurs when the alarm clock timer goes off ++(which happens only if your program has requested an alarm). ++ ++Some signal handlers are installed and changed for @value{DBG}'s ++normal use: @code{SIGDEBUG} and @code{SIGEXIT}. @code{SIGDEBUG} is ++used by the debugger to potentially stop your program before execution ++of each statement occurs, and @code{SIGEXIT} is used to catch your ++program just before it is set to leave so you have the option of ++restarting the program with the same options (and not leave the ++debugger) or let the program quit. ++ ++Signal handlers that the debugged script might have installed are ++saved and called before the corresponding debugger handler. Thus, the ++debugged program should work roughly in the same fashion as when it is ++not debugged. However there are some call-stack variables which ++inevitably will differ. To try to hedge this a little so the behaviour ++is the same, @value{DBG} will modify arguments to the traps if it ++finds one of the call-stack that change as a result of the debugger ++being in place. In particluar @env{$LINENO} will get replaced with ++@env{$@{BASH_LINENO[0]@}}; also @env{$@{BASH_LINENO[0]@}} and ++@env{$@{BASH_SOURCE[0]@}} get replaced with ++@env{$@{BASH_LINENO[1]@}} and @env{$@{BASH_SOURCE[1]@}} ++respectively. ++ ++The debugger also installs an interrupt handler @code{SIGINT} so that ++errant programs can be interrupted and you can find out where the ++program was when you interrupted it. ++ ++@cindex fatal signals ++Some signals, including @code{SIGALRM}, are a normal part of the ++functioning of your program. Others, such as @code{SIGSEGV}, indicate ++errors; these signals are @dfn{fatal} (they kill your program immediately) if the ++program has not specified in advance some other way to handle the signal. ++@code{SIGINT} does not indicate an error in your program, but it is normally ++fatal so it can carry out the purpose of the interrupt: to kill the program. ++ ++@BASH has the ability to detect any occurrence of a signal in your ++program. You can tell @BASH in advance what to do for each kind of ++signal. ++ ++@cindex handling signals ++Normally, @BASH is set up to let the non-erroneous signals like ++@code{SIGALRM} be silently passed to your program ++(so as not to interfere with their role in the program's functioning) ++but to stop your program immediately whenever an error signal happens. ++You can change these settings with the @code{handle} command. ++ ++@table @code ++@kindex info signals ++@item info signals ++@itemx info handle ++Print a table of all the kinds of signals and how @BASH has been told to ++handle each one. You can use this to see the signal numbers of all ++the defined types of signals. ++ ++@code{info handle} is an alias for @code{info signals}. ++ ++@kindex handle ++@item handle @var{signal} @var{keywords}@dots{} ++Change the way @BASH handles signal @var{signal}. @var{signal} ++can be the number of a signal or its name (with or without the ++@samp{SIG} at the beginning); a list of signal numbers of the form ++@samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the ++known signals. The @var{keywords} say what change to make. ++@end table ++ ++@c @group ++The keywords allowed by the @code{handle} command can be abbreviated. ++Their full names are: ++ ++@table @code ++@item stop ++@BASH should stop your program when this signal happens. This implies ++the @code{print} keyword as well. ++ ++@item nostop ++@BASH should not stop your program when this signal happens. It may ++still print a message telling you that the signal has come in. ++ ++@item print ++@BASH should print a message when this signal happens. ++ ++@item noprint ++@BASH should not mention the occurrence of the signal at all. ++ ++@item stack ++@BASH should print a stack trace when this signal happens. ++ ++@item nostack ++@BASH should not print a stack trace when this signal occurs. ++ ++@ifset FINISHED ++@item pass ++@itemx noignore ++@BASH should allow your program to see this signal; your program ++can handle the signal, or else it may terminate if the signal is fatal ++and not handled. @code{pass} and @code{noignore} are synonyms. ++ ++@item nopass ++@itemx ignore ++@BASH should not allow your program to see this signal. ++@code{nopass} and @code{ignore} are synonyms. ++@end ifset ++@end table ++@c @end group ++ ++@ifset FINISHED ++When a signal stops your program, the signal is not visible to the ++program until you ++continue. Your program sees the signal then, if @code{pass} is in ++effect for the signal in question @emph{at that time}. In other words, ++after @BASH reports a signal, you can use the @code{handle} ++command with @code{pass} or @code{nopass} to control whether your ++program sees that signal when you continue. ++ ++The default is set to @code{nostop}, @code{noprint}, @code{pass} for ++non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and ++@code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the ++erroneous signals. ++ ++You can also use the @code{signal} command to prevent your program from ++seeing a signal, or cause it to see a signal it normally would not see, ++or to give it any signal at any time. For example, if your program stopped ++due to some sort of memory reference error, you might store correct ++values into the erroneous variables and continue, hoping to see more ++execution; but your program would probably terminate immediately as ++a result of the fatal signal once it saw the signal. To prevent this, ++you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your ++program a signal}. ++@end ifset ++ ++@node Information ++@section Status and Debugger Settings (@samp{info}, @samp{show}) ++ ++In addition to @code{help}, you can use the @BASH commands @code{info} ++and @code{show} to inquire about the state of your program, or the state ++of @BASH itself. Each command supports many topics of inquiry; this ++manual introduces each of them in the appropriate context. The listings ++under @code{info} and under @code{show} in the Index point to ++all the sub-commands. @xref{Command Index}. ++ ++@c @group ++@table @code ++@kindex info ++@kindex i @r{(@code{info})} ++@item info ++This command (abbreviated @code{i}) is for describing the state of ++your program. For example, you canlist the current @code{$1}, @code{$2} ++parameters with @code{info args}, or list the breakpoints you have set ++with @code{info breakpoints} or @code{info watchpoints}. You can get ++a complete list of the @code{info} sub-commands with @w{@code{help ++info}}. ++ ++@kindex set ++@item set ++You can assign the result of an expression to an environment variable ++with @code{set}. For example, you can set the number of lines ++@BASH will list by default to 20 with @code{set listsize 20}. ++ ++@kindex show ++@item show ++In contrast to @code{info}, @code{show} is for describing the state of ++@BASH itself. ++You can change most of the things you can @code{show}, by using the ++related command @code{set}; ++ ++The distinction between @code{info} and @code{show} however is a bit ++fuzzy. For example, to list the arguments given to your script use ++@code{show args}; @code{info args} does something different. ++ ++@kindex info set ++To display all the settable parameters and their current ++values, you can use @code{show} with no arguments; you may also use ++@code{info set}. Both commands produce the same display. ++@c FIXME: "info set" violates the rule that "info" is for state of ++@c FIXME...program. Ck w/ GNU: "info set" to be called something else, ++@c FIXME...or change desc of rule---eg "state of prog and debugging session"? ++@end table ++@c @end group ++ ++Here are three miscellaneous @code{show} subcommands, all of which are ++exceptional in lacking corresponding @code{set} commands: ++ ++@table @code ++@kindex show version ++@cindex version number ++@item show version ++Show what version of @BASH is running. You should include this ++information in @BASH bug-reports. If multiple versions of ++@BASH are in use at your site, you may need to determine which ++version of @BASH you are running; as @BASH evolves, new ++commands are introduced, and old ones may wither away. Also, many ++system vendors ship variant versions of @value{BASH}, and there are ++variant versions of @BASH in @sc{gnu}/Linux distributions as well. ++The version number is the same as the one announced when you start ++@value{BASH}. ++ ++@kindex show copying ++@item show copying ++Display information about permission for copying @value{BASH}. ++ ++@kindex show warranty ++@item show warranty ++Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty, ++if your version of @DBG comes with one. ++ ++@end table ++ ++@node Stack ++@section Examining the Stack (@samp{where}, @samp{frame}, @samp{up}, @samp{down}) ++ ++When your script has stopped, one thing you'll probably want to know ++is where it stopped and some idea of how it got there. ++ ++@cindex call stack ++Each time your script performs a function call (either as part of a ++command substitution or not), or `source's a file, information about ++this action is saved. The call stack then is this a history of the ++calls that got you to the point that you are currently stopped at. ++ ++@cindex selected frame ++One of the stack frames is @dfn{selected} by @DBG and many ++@DBG commands refer implicitly to the selected frame. In ++particular, whenever you ask @DBG to list lines without giving ++a line number or location the value is found in the selected frame. ++There are special @DBG commands to select whichever frame you ++are interested in. @xref{Selection, ,Selecting a frame}. ++ ++When your program stops, @BASH automatically selects the ++currently executing frame and describes it briefly, similar to the ++@code{frame} command. ++ ++ ++@menu ++* Frames:: Stack frames ++* Backtrace:: Backtraces (where) ++* Selection:: Selecting a frame (up, down, frame) ++ ++@end menu ++ ++@node Frames ++@subsection Stack frames ++ ++@cindex frame, definition ++@cindex stack frame ++The call stack is divided up into contiguous pieces called @dfn{stack ++frames}, or @dfn{frames} for short; each frame is the data associated ++with one call to one function. The frame contains the line number of ++the caller of the function, the source-file name that the line refers ++to and function name (which could be the built-in name ``source'').. ++ ++@cindex initial frame ++@cindex outermost frame ++@cindex innermost frame ++When your script is started, the stack has only one frame, that of the ++function @code{main}. This is called the @dfn{initial} frame or the ++@dfn{outermost} frame. Each time a function is called, a new frame is ++made. Each time a function returns, the frame for that function invocation ++is eliminated. If a function is recursive, there can be many frames for ++the same function. The frame for the function in which execution is ++actually occurring is called the @dfn{innermost} frame. This is the most ++recently created of all the stack frames that still exist. ++ ++@cindex frame number ++@value{DBG} assigns numbers to all existing stack frames, starting with ++zero for the innermost frame, one for the frame that called it, ++and so on upward. These numbers do not really exist in your script; ++they are assigned by @value{DBG} to give you a way of designating stack ++frames in @value{DBG} commands. ++ ++@node Backtrace ++@subsection Backtraces (@samp{where}) ++ ++@cindex backtraces ++@cindex tracebacks ++@cindex stack traces ++A backtrace is essentially the same as the call stack: a summary of ++how your script got where it is. It shows one line per frame, for ++many frames, starting with the place that you sare stopped at (frame ++zero), followed by its caller (frame one), and on up the stack. ++ ++@table @code ++@kindex backtrace ++@kindex bt @r{(@code{backtrace})} ++@item backtrace ++@itemx bt ++@itemx where ++@itemx T ++Print a backtrace of the entire stack: one line per frame for all ++frames in the stack. ++ ++@item backtrace @var{n} ++@itemx bt @var{n} ++@itemx where @var{n} ++@itemx T @var{n} ++Similar, but print only the innermost @var{n} frames. ++ ++@ifset FINISHED ++@item backtrace -@var{n} ++@itemx bt -@var{n} ++@itemx where -@var{n} ++@itemx T -@var{n} ++Similar, but print only the outermost @var{n} frames. ++@end ifset ++@end table ++ ++@kindex where ++The names @code{where} and @code{T} are additional aliases for ++@code{backtrace}. ++ ++Each line in the backtrace shows the frame number and the function ++name, the source file name and line number, as well as the function name. ++ ++Here is an example of a backtrace taken a program in the ++regression-tests @file{parm.sh}. ++ ++@smallexample ++@group ++% ../bashdb -n -L .. parm.sh ++Bourne-Again Shell Debugger, release @value{BASHRELEASE} ++Copyright 2002,2003 Rocky Bernstein ++This is free software, covered by the GNU General Public License, and you are ++welcome to change it and/or distribute copies of it under certain conditions. ++ ++(./parm.sh:21): ++21: fn1 5 ++bashdb<0> @b{continue fn3} ++One-time breakpoint 1 set in file ./parm.sh, line 17. ++fn2: testing 1 2 3 ++(./parm.sh:17): ++17: fn3() @{ ++bashdb<1> @b{where} ++->0 in file `./parm.sh' at line 14 ++##1 fn3() called from file `./parm.sh' at line 14 ++##2 fn2("testing 1", "2 3") called from file `parm.sh' at line 5 ++##3 fn1("0") called from file `parm.sh' at line 9 ++##4 fn1("1") called from file `parm.sh' at line 9 ++##5 fn1("2") called from file `parm.sh' at line 9 ++##6 fn1("3") called from file `parm.sh' at line 9 ++##7 fn1("4") called from file `parm.sh' at line 9 ++##8 fn1("5") called from file `parm.sh' at line 21 ++##9 source("parm.sh") called from file `bashdb' at line 143 ++##10 main("-n", "-L", "..", "parm.sh") called from file `bashdb' at line 0 ++@end group ++@end smallexample ++ ++@noindent ++The display for ``frame'' zero isn't a frame at all, although it has ++the same information minus a function name; it just indicates that ++your script has stopped at the code for line @code{14} ++of @code{./parm.sh}. ++ ++@node Selection ++@subsection Selecting a frame (@samp{up}, @samp{down}, @samp{frame}) ++ ++Commands for listing source code in your script work on whichever ++stack frame is selected at the moment. Here are the commands for ++selecting a stack frame; all of them finish by printing a brief ++description of the stack frame just selected. ++ ++@table @code ++@kindex up ++@item up @var{n} ++Move @var{n} frames up the stack. For positive numbers @var{n}, this ++advances toward the outermost frame, to higher frame numbers, to frames ++that have existed longer. @var{n} defaults to one. ++ ++@kindex down ++@kindex do @r{(@code{down})} ++@item down @var{n} ++Move @var{n} frames down the stack. For positive numbers @var{n}, this ++advances toward the innermost frame, to lower frame numbers, to frames ++that were created more recently. @var{n} defaults to one. You may ++abbreviate @code{down} as @code{do}. ++@end table ++ ++All of these commands end by printing two lines of output describing the ++frame. The first line shows the frame number, the function name, the ++arguments, and the source file and line number of execution in that ++frame. The second line shows the text of that source line. ++ ++@need 100 ++For example: ++ ++@smallexample ++@group ++bashdb<8> @b{up} ++19: sourced_fn ++bashdb<8> @b{T} ++##0 in file `./bashtest-sourced' at line 8 ++->1 sourced_fn() called from file `bashtest-sourced' at line 19 ++##2 source() called from file `bashdb-test1' at line 23 ++##3 fn2() called from file `bashdb-test1' at line 33 ++##4 fn1() called from file `bashdb-test1' at line 42 ++##5 main() called from file `bashdb-test1' at line 0 ++@end group ++@end smallexample ++ ++After such a printout, the @code{list} command with no arguments ++prints ten lines centered on the point of execution in the frame. ++@xref{List, ,Printing source lines}. ++ ++@table @code ++@kindex frame ++@cindex current stack frame ++@item frame @var{args} ++The @code{frame} command allows you to move from one stack frame to ++another, and to print the stack frame you select. @var{args} is the ++the stack frame number. Without an argument, @code{frame} prints the ++current stack frame. ++@end table ++ ++@node List ++@section Examining Source Files (@samp{list}) ++ ++@value{DBG} can print parts of your script's source. When your ++script stops, @value{DBG} spontaneously prints the line where it ++stopped. Likewise, when you select a stack frame (@pxref{Selection, ++,Selecting a frame}), @value{DBG} prints the line where execution in ++that frame has stopped. You can print other portions of source files ++by explicit command. ++ ++If you use @value{DBG} through its @sc{gnu} Emacs interface, you may ++prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using ++@value{DBG} under @sc{gnu} Emacs}. ++ ++@kindex list ++@kindex l @r{(@code{list})} ++To print lines from a source file, use the @code{list} command ++(abbreviated @code{l}). By default, ten lines are printed. ++There are several ways to specify what part of the file you want to print. ++ ++Here are the forms of the @code{list} command most commonly used: ++ ++@table @code ++@item list @var{linenum} ++@itemx l @var{linenum} ++Print lines centered around line number @var{linenum} in the ++current source file. ++ ++@item list @var{function} ++@itemx l @var{function} ++Print the text of @var{function}. ++ ++@item list ++@itemx l ++Print more lines. If the last lines printed were printed with a ++@code{list} command, this prints lines following the last lines ++printed; however, if the last line printed was a solitary line printed ++as part of displaying a stack frame (@pxref{Stack, ,Examining the ++Stack}), this prints lines centered around that line. ++ ++@item list - ++@itemx l - ++Print lines just before the lines last printed. ++@end table ++ ++By default, @value{DBG} prints ten source lines with any of these forms of ++the @code{list} command. ++You can change this using @code{set listsize}: ++ ++@table @code ++@kindex set listsize ++@item set listsize @var{count} ++Make the @code{list} command display @var{count} source lines (unless ++the @code{list} argument explicitly specifies some other number). ++ ++@kindex show listsize ++@item show listsize ++Display the number of lines that @code{list} prints. ++@end table ++ ++Repeating a @code{list} command with @key{RET} discards the argument, ++so it is equivalent to typing just @code{list}. This is more useful ++than listing the same lines again. An exception is made for an ++argument of @samp{-}; that argument is preserved in repetition so that ++each repetition moves up in the source file. ++ ++@cindex linespec ++In general, the @code{list} command expects you to supply a ++@dfn{linespecs}. Linespecs specify source lines; there are several ways ++of writing them, but the effect is always to specify some source line. ++ ++Here is a complete description of the possible arguments for @code{list}: ++ ++@table @code ++@item list @var{linespec} ++Print lines centered around the line specified by @var{linespec}. ++ ++@item list @var{first} @var{increment} ++Print @var{increment} lines starting from @var{first} ++ ++@item list @var{first} ++Print lines starting with @var{first}. ++ ++@item list - ++Print lines just before the lines last printed. ++ ++@item list . ++Print lines after where the script is stopped. ++ ++@item list ++As described in the preceding table. ++@end table ++ ++Here are the ways of specifying a single source line---all the ++kinds of linespec. ++ ++@table @code ++@item @var{number} ++Specifies line @var{number} of the current source file. ++When a @code{list} command has two linespecs, this refers to ++the same source file as the first linespec. ++ ++@item @var{filename}:@var{number} ++Specifies line @var{number} in the source file @var{filename}. ++ ++@item @var{function} ++Specifies the line that function @var{function} is listed on. ++ ++@ifset FINISHED ++@item @var{filename}:@var{function} ++Specifies the line of function @var{function} in the file ++@var{filename}. You only need the file name with a function name to ++avoid ambiguity when there are identically named functions in ++different source files. ++@end ifset ++@end table ++ ++@node Search ++@section Searching source files (@samp{search}, @samp{reverse}, @samp{/.../}, @samp{?..?}) ++@cindex searching ++@kindex reverse-search ++ ++There are two commands for searching through the current source file for a ++BASH extended pattern-matching expression. ++ ++@table @code ++@kindex search ++@kindex forward ++@item forward @var{bash-pattern} ++@itemx search @var{bash-pattern} ++The command @samp{forward @var{bash-pattern}} checks each line, ++starting with the one following the current line, for a match for ++@var{bash-pattern} which is an extended bash pattern-matching ++expression. It lists the line that is found. You can use the synonym ++@samp{search @var{bash-pattern}} or abbreviate the command name as ++@code{fo} or @code{/@var{pat}/}. ++ ++@item reverse @var{bash-pattern} ++The command @samp{reverse @var{bash-pattern}} checks each line, starting ++with the one before the last line listed and going backward, for a match ++for @var{bash-pattern}. It lists the line that is found. You can abbreviate ++this command as @code{rev} or @code{?@var{bash-pattern}?}. ++@end table ++ ++@node Data ++@section Examining Data (@samp{print}, @samp{examine}, @samp{info variables}) ++ ++@cindex printing data ++@cindex examining data ++@kindex print ++ ++One way to examine string data in your script is with the @code{print} ++command (abbreviated @code{p}). However a more versatile print command ++is @code{x}; it can print variable and function definitions and can do ++arithmetic computations. Finally, the most general method would be ++via @code{eval echo}. ++ ++@table @code ++@kindex print ++@kindex p @r{(@code{print})} ++@item print @var{expr} ++ ++Use @code{print} to dispay strings as you would from @code{echo}. And ++as such, variable names to be substituted have to be preceded with a ++dollar sign. As with echo, filename expansion, e.g. tilde expansion, ++is performed on unquoted strings. So for example if you want to print ++a *, you would write @samp{print "*"}, not @samp{print *}. If you want ++to have the special characters dollars sign appear, use a backslash. ++ ++@item print ++@itemx p ++If you omit @var{expr}, @value{DBG} displays the last expression again. ++ ++@item x @ovar{expr} ++@kindex x ++ ++This is a smarter, more versatile ``print'' command, and although sometimes ++it might not be what you want, and you may want to resort to either ++@code{print} or @code{eval echo...}. ++ ++As with @code{print}, if you omit @var{expr}, @value{DBG} displays ++the last expression again. ++ ++The @code{x} command first checks if @var{expr} is single variable. If ++it is, the definition and value are printed via @value{BASH}'s ++@code{declare -p} command. This will show the variable's attributes ++such as if it is read only or if it is an integer. If the variable is ++an array, that is show and the array values are printed. ++ ++If instead @var{expr} is a function, the function definition is ++printed via @value{BASH}'s @code{declare -f} command. If @var{expr} ++was neither a variable nor an expression, then we try to get a value ++via @code{let}. And if this returns an error, as a last resort we call ++@code{print} and give what it outputs. ++ ++Since @code{let} may be used internally and since (to my thinking) ++@code{let} does funny things, the results may seem odd unless you ++understand the sequence tried above and how @code{let} works. For ++example if the variable @code{foo} has value 5, then @samp{x foo} show ++the definition of foo with value 5, and @samp{x foo+5} prints 10 as ++expected. So far so good. However if @code{foo} is has value ++``string'', @samp{x foo+5} prints 5 because @code{let} has turned the ++string into 0; @samp{p foo+5} will simply print ``foo+5''; if you want ++the value of ``foo'' substituted inside a string, for example you ++expect ``the value of foo is $foo'' to come out ``the value of foo is ++5'', then the right command to use is @code{print} rather than ++@code{x}, making sure you add the dollar onto the beginning of the ++variable. ++ ++@item V @ovar{!}@ovar{pattern} ++@kindex V ++ ++If you want to @emph{all} list variables and there values or a set of ++variables by pattern, use this command. ++ ++@smallexample ++@group ++bashdb<0> @b{V dq*} ++dq_args="dq_*" ++dq_cmd="V" ++@end group ++@end smallexample ++ ++@end table ++ ++@node Evaluation/Execution ++@section Running Arbitrary BASH and Shell commands (@samp{eval}, @samp{shell}) ++ ++The two most general commands and most ``low-level'' are @code{eval} ++and @code{shell}. ++ ++@table @code ++@item eval ++@itemx e ++@kindex e ++ ++In contrast to the commands of the last section the most general way ++to examine data is through @code{eval}. But you do much more with ++this; you can change the values of variables, since, you are just ++evaluating BASH code. ++ ++If you expect output, you should arrange that in the command, such as ++via @code{echo} or @code{printf}. For example, to print the value of ++@var{foo}, you would type @samp{e echo $foo}. This is bit longer than ++@samp{p $foo} or (when possible) @samp{x foo}. However suppose you ++wanted to find out how the builtin test operator @samp{[} works with ++the @samp{-z} test condition. You could use @code{eval} to do this ++such as @samp{e [ -z "$foo"] && echo "yes"}. ++ ++@kindex shell ++@kindex !! @r{(@code{shell})} ++@cindex shell escape ++@item shell @var{command string} ++@itemx !! ++ ++If you need to execute occasional shell commands during your ++debugging session, there is no need to leave or suspend @value{DBG}; you can ++just use the @code{shell} command or its alias @code{!!}. ++ ++Invoke a shell to execute @var{command string}. ++ ++@end table ++ ++@node Auto Display ++@section Automatic display (@samp{display}, @samp{undisplay}) ++@cindex automatic display ++@cindex display of expressions ++ ++If you find that you want to print the value of an expression ++frequently (to see how it changes), you might want to add it to the ++@dfn{automatic display list} so that @value{DBG} evaluates a ++statement each time your program stops. Each expression added to the ++list is given a number to identify it; to remove an expression from ++the list, you specify that number. The automatic display looks like ++this: ++ ++@example ++2 (echo $x): 38 ++@end example ++ ++@noindent ++This display shows item numbers, expressions and their current values. ++ ++@table @code ++@kindex display ++@item display @var{expr} ++Add the expression @var{expr} to the list of expressions to display ++each time your program stops. ++@end table ++ ++@table @code ++@kindex delete display ++@kindex undisplay ++@item undisplay @var{dnums}@dots{} ++@itemx delete display @var{dnums}@dots{} ++Remove item numbers @var{dnums} from the list of expressions to display. ++ ++@code{undisplay} does not repeat if you press @key{RET} after using it. ++(Otherwise you would just get the error @samp{No display number @dots{}}.) ++ ++@kindex disable display ++@item disable display @var{dnums}@dots{} ++Disable the display of item numbers @var{dnums}. A disabled display ++item is not printed automatically, but is not forgotten. It may be ++enabled again later. ++ ++@kindex enable display ++@item enable display @var{dnums}@dots{} ++Enable display of item numbers @var{dnums}. It becomes effective once ++again in auto display of its expression, until you specify otherwise. ++ ++@item display ++Display the current values of the expressions on the list, just as is ++done when your program stops. ++ ++@kindex info display ++@item info display ++Print the list of expressions previously set up to display ++automatically, each one with its item number, but without showing the ++values. This includes disabled expressions, which are marked as such. ++It also includes expressions which would not be displayed right now ++because they refer to automatic variables not currently available. ++@end table ++ ++@node Controlling bashdb ++@section Controlling @DBG (@samp{set}, @samp{history}) ++ ++You can alter the way @BASH interacts with you by using the ++@code{set} command. ++ ++@menu ++* Annotate:: Annotation Level (set annotate) ++* Prompt:: Prompt (set prompt, show prompt) ++* Editing:: Command editing (set editing, show editing) ++* Command Display:: Command display (set showcommand) ++* History:: Command history (history, !, H) ++@end menu ++ ++@node Annotate ++@subsection Annotation Level (@samp{set annoatate}) ++ ++The annotation level controls how much information does @value{DBG} ++print together with its prompt, values of expressions, source lines, ++and other types of output. Level 0 is the normal, level 1 is for use ++when @value{DBG} is run as a subprocess of @sc{gnu} Emacs of @value{DDD}, ++level 2 is the maximum annotation suitable for programs that control ++@value{DBGN}. ++ ++@node Prompt ++@subsection Prompt (@samp{set prompt}, @samp{show prompt}) ++ ++@cindex prompt ++ ++@value{dBGP} indicates its readiness to read a command by printing a ++string called the @dfn{prompt}. This string is normally: ++@example ++bashdb$@{_Dbg_less@}$@{#_Dbg_history[@@]@}$@{_Dbg_greater@}$_Dbg_space ++@end example ++ ++When variables inside the the prompt string are evaluated, the above ++becomes something like @samp{bashdb<5>} if this is the fifth command ++executed or perhaps @samp{bashdb<<2>>} if you have called the debugger ++from inside a debugger session and this is the second command inside ++the debugger session or perhaps @samp{bashdb<(6)>} if you ++entered a subshell after the fifth command. ++ ++You can change the prompt string with the @code{set prompt} command, ++although it is not normally advisable to do so without understanding ++the implications. If you are using the @value{DDD} GUI, it changes the ++changes the prompt and should not do so. In certain other ++circumstances (such as writing a GUI like @value{DDD}), it may be is useful ++to change the prompt. ++ ++@emph{Note:} @code{set prompt} does not add a space for you after the ++prompt you set. This allows you to set a prompt which ends in a space ++or a prompt that does not. Furthermore due to a implementation ++limitation (resulting from a limitation of the bash built-in function ++``read''), to put a space at the end of the prompt use the ++@samp{$_Dbg_space} variable. ++ ++@table @code ++@kindex set prompt ++@item set prompt @var{newprompt} ++Directs @value{DBG} to use @var{newprompt} as its prompt string ++henceforth. ++ ++@emph{Warning: changing the prompt can @value{DDD}'s ability to ++understand when the debugger is waiting for input.} ++ ++@kindex show prompt ++@item show prompt ++Prints a line of the form: @samp{bashdb's prompt is: @var{your-prompt}} ++@end table ++ ++@node Editing ++@subsection Command editing (@samp{set editing}, @samp{show editing}) ++@cindex readline ++@cindex command line editing ++ ++@value{DBG} reads its input commands through bash which uses via the ++@dfn{readline} interface. This @sc{gnu} library provides consistent ++behavior for programs which provide a command line interface to the ++user. Advantages are @sc{gnu} Emacs-style or @dfn{vi}-style inline ++editing of commands, @code{csh}-like history substitution, and a ++storage and recall of command history across debugging sessions. ++ ++You may control the behavior of command line editing in @BASH with the ++command @code{set}. ++ ++@table @code ++@kindex set editing ++@cindex editing ++@item set editing ++@itemx set editing on ++Enable command line editing (enabled by default). ++ ++@item set editing off ++Disable command line editing. ++ ++@kindex show editing ++@item show editing ++Show whether command line editing is enabled. ++@end table ++ ++ ++@node Command Display ++@subsection Command Display (@samp{set showcommand}) ++ ++The debugger normally lists the line number and source line of the for ++the statement to be next executed. Often this line contains one ++expression or one statement and it is clear from this line what's ++going to happen. However @BASH allows many expressions or ++statements to be put on a single source line; some lines ++contain several units of execution. Some examples of this ++behavior are listed below: ++ ++@smallexample ++x=1; y=2; x=3 ++(( x > 5 )) && x=5 ++y=`echo *` ++@end smallexample ++ ++In the first line of the example above, we have three assignment ++statements on a single line. In the second line of the example above ++we have a statement which gets run only if a condition tests true. And ++in the third line of the example above, we have a command that gets ++run and then the output of that is substituted in an assignemnt ++statement. If you were single stepping inside the debugger, each line ++might get listed more than once before each of the actions that might ++get performed. (In the case of the conditional statement, the ++line gets listed only once when the condition is false.) ++ ++In order to assist understanding where you are, the enhanced version ++of @BASH maintains a dynamic variable @env{BASH_COMMAND} that ++contains piece of code next to be run (or is currently being run). The ++debugger has arranged to save this and can display this information ++or not. This is controlled by @code{set showcommand}. ++ ++@table @code ++@kindex set showcommand ++@item set showcommand @r{[}auto | on | off @r{]} ++controls whether or not to show the saved @env{BASH_COMMAND} for the ++command next to be executed. ++@end table ++ ++When the value is @code{auto} the following heuristic is used to ++determine whether or not to display the saved @env{BASH_COMMAND}. If ++the last time you stopped you were at the same place and the command ++string has changed, then show the command. When the value @code{on} is ++used, the debugger always shows @env{BASH_COMMAND} and when ++@code{off} is used, the debugger nevers shows ++@env{BASH_COMMAND}. Note that listing the text of the source line is ++independent of whether or not the command is also listed. ++ ++Some examples: ++@smallexample ++set showcommand auto @b{This is the default} ++set showcommand on @b{Always show the next command to be executed} ++set showcommand off @b{Never show the next command to be executed} ++@end smallexample ++ ++@node History ++@subsection Command history (@samp{H}, @samp{history}, @samp{!}) ++ ++@value{dBGP} can keep track of the commands you type during your ++debugging sessions, so that you can be certain of precisely what ++happened. If the prompt has not been changed (see @ref{Prompt, ++,Prompt}), the history number that will be in use next is by default ++listed in the debugger prompt. Invalid commands and history commands ++are not saved on the history stack. ++ ++@table @code ++@kindex H @r{[}@var{start-number} @ovar{end-number}@r{]} ++@item H @r{[}@var{start-number} @ovar{end-number}@r{]} ++@item H @ovar{-count} ++@itemx !@r{[}-@r{]}@var{n}:p ++ ++You can list what is in the history stack with @code{H}. Debugger ++commands in ths history stack are listed from most recent to least recent. ++If no @var{start-number} is given we start with the most recently ++executed command and end with the first entry in the history stack. ++If @var{start-number} is given, that history number is listed first. If ++@var{end-number} is given, that history number is listed last. If a ++single negative number is given list that many history commands. ++ ++An alternate form is @code{!@emph{n}:p} or @code{!-@emph{n}:p} where ++@emph{n} is an integer. If a minus sign is used, @emph{n} is taken as ++the count to go back from the end rather than as a absolute history ++number. In contrast @code{H}, this form only prints a @emph{single} ++history item. ++ ++Some examples: ++@smallexample ++H @b{List entire history} ++H -2 @b{List the last two history items} ++!-2:p @b{List a single history item starting at the same place as above} ++H 5 @b{List history from history number 5 to the begining (number 0)} ++H 5 0 @b{Same as above} ++H 5 3 @b{List history from history number 5 down to history number 3} ++!5:p @b{List a single history item 5} ++@end smallexample ++ ++@kindex history @r{[}-@r{]}@r{[}@var{n}@r{]} ++@kindex !@r{[}-@r{]}@var{n} @r{(@code{history})} ++@item history @r{[}@r{[}-@r{]}@var{n}@r{]} ++@itemx !@r{[}-@r{]}@var{n} ++ ++Use this command to reexecute a given history number. If no number is ++given, the last debugger command in the history is executed. ++ ++An alternate form is @code{!@emph{n}} or @code{!-@emph{n}} where ++@emph{n} is an integer. ++ ++If a minus sign is used in in either form, @emph{n} is taken as the ++count to go back from the end rather than as a absolute history ++number. ++ ++@end table ++ ++@ifset FINISHED ++Use these commands to manage the @value{DBGN} command ++history facility. ++ ++@table @code ++@cindex history substitution ++@cindex history file ++@kindex set history filename ++@kindex GDBHISTFILE ++@item set history filename @var{fname} ++Set the name of the @BASH command history file to @var{fname}. ++This is the file where @BASH reads an initial command history ++list, and where it writes the command history from this session when it ++exits. You can access this list through history expansion or through ++the history command editing characters listed below. This file defaults ++to the value of the environment variable @code{GDBHISTFILE}, or to ++@file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable ++is not set. ++ ++@cindex history save ++@kindex set history save ++@item set history save ++@itemx set history save on ++Record command history in a file, whose name may be specified with the ++@code{set history filename} command. By default, this option is disabled. ++ ++@item set history save off ++Stop recording command history in a file. ++ ++@cindex history size ++@kindex set history size ++@item set history size @var{size} ++Set the number of commands which @BASH keeps in its history list. ++This defaults to the value of the environment variable ++@code{HISTSIZE}, or to 256 if this variable is not set. ++@end table ++ ++@cindex history expansion ++History expansion assigns special meaning to the character @kbd{!}. ++ ++Since @kbd{!} is also the logical not operator in C, history expansion ++is off by default. If you decide to enable history expansion with the ++@code{set history expansion on} command, you may sometimes need to ++follow @kbd{!} (when it is used as logical not, in an expression) with ++a space or a tab to prevent it from being expanded. The readline ++history facilities do not attempt substitution on the strings ++@kbd{!=} and @kbd{!(}, even when history expansion is enabled. ++ ++The commands to control history expansion are: ++ ++@table @code ++@kindex set history expansion ++@item set history expansion on ++@itemx set history expansion ++Enable history expansion. History expansion is off by default. ++ ++@item set history expansion off ++Disable history expansion. ++ ++The readline code comes with more complete documentation of ++editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs ++or @code{vi} may wish to read it. ++ ++@c @group ++@kindex show history ++@item show history ++@itemx show history filename ++@itemx show history save ++@itemx show history size ++@itemx show history expansion ++These commands display the state of the @BASH history parameters. ++@code{show history} by itself displays all four states. ++@c @end group ++@end table ++ ++@table @code ++@kindex shows ++@item show commands ++Display the last ten commands in the command history. ++ ++@item show commands @var{n} ++Print ten commands centered on command number @var{n}. ++ ++@item show commands + ++Print ten commands just after the commands last printed. ++@end table ++@end ifset ++ ++@node Front Ends ++@chapter Using @value{DBG} from a front-end user interface ++ ++There are two front-ends available that can use @value{DBG} as a back-end ++debugger are @sc{gnu} Emacs and @value{DDD}. ++ ++@menu ++* Emacs:: Using @value{DBG} from @sc{gnu} Emacs ++* DDD:: Using @value{DBG} from @value{DDD} ++@end menu ++ ++ ++@node Emacs ++@section Using @value{DBG} from @sc{gnu} Emacs ++ ++@cindex Emacs ++@cindex @sc{gnu} Emacs ++A special interface allows you to use @sc{gnu} Emacs to view (and ++edit) the source files for the program you are debugging with ++@value{DBG}. ++ ++To use this interface, use the command @kbd{M-x bashdb} in Emacs. ++Give the executable file you want to debug as an argument. This ++command starts @value{DBG} as a subprocess of Emacs, with input and ++output through a newly created Emacs buffer. ++ ++Using @value{DBG} under Emacs is just like using @value{DBG} ++normally except for two things: ++ ++@itemize @bullet ++@item ++All ``terminal'' input and output goes through the Emacs buffer. ++@end itemize ++ ++This applies both to @value{DBG} commands and their output, and to the input ++and output done by the program you are debugging. ++ ++This is useful because it means that you can copy the text of previous ++commands and input them again; you can even use parts of the output ++in this way. ++ ++All the facilities of Emacs' Shell mode are available for interacting ++with your script. In particular, you can send signals the usual ++way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a ++stop. ++ ++@itemize @bullet ++@item ++@value{DBG} displays source code through Emacs. ++@end itemize ++ ++Each time @value{DBG} displays a stack frame, Emacs automatically finds the ++source file for that frame and puts an arrow (@samp{=>}) at the ++left margin of the current line. Emacs uses a separate buffer for ++source display, and splits the screen to show both your @value{DBG} session ++and the source. ++ ++Explicit @value{DBG} @code{list} or search commands still produce output as ++usual, but you probably have no reason to use them from Emacs. ++ ++@quotation ++@emph{Warning:} If the directory where your script resides is not your ++current directory, it can be easy to confuse Emacs about the location of ++the source files, in which case the auxiliary display buffer does not ++appear to show your source. @value{DBG} can find programs by searching your ++environment's @code{PATH} variable, so the @value{DBG} input and output ++session proceeds normally; but Emacs does not get enough information ++back from @value{DBG} to locate the source files in this situation. To ++avoid this problem, either start @value{DBG} mode from the directory where ++your script resides, or specify an absolute file name when prompted for the ++@kbd{M-x gdb} argument. ++ ++A similar confusion can result if you use the @value{DBG} @code{file} command to ++switch to debugging a program in some other location, from an existing ++@value{DBG} buffer in Emacs. ++@end quotation ++ ++By default, @kbd{M-x bashdb} calls the @code{bash --debugger}. If you ++need to call @value{DBG} by a different name (for example, if you ++keep several configurations around, with different names) you can set ++the Emacs variable @code{gud-bashdb-command-name}; for example, ++ ++@example ++(setq gud-bashdb-command-name "bash-debugger") ++@end example ++ ++@noindent ++(preceded by @kbd{M-:} or @kbd{ESC :}, or typed in the @code{*scratch*} buffer, or ++in your @file{.emacs} file) makes Emacs call the program named ++``@code{bash-debugger}'' instead. ++ ++In the @value{DBG} I/O buffer, you can use these special Emacs commands in ++addition to the standard Shell mode commands: ++ ++@table @kbd ++@item C-h m ++Describe the features of Emacs' @value{DBG} Mode. ++ ++@item M-s ++Execute to another source line, like the @value{DBG} @code{step} command; also ++update the display window to show the current file and location. ++ ++@item M-n ++Execute to next source line in this function, skipping all function ++calls, like the @value{DBG} @code{next} command. Then update the display window ++to show the current file and location. ++ ++@item C-c C-f ++Execute until exit from the selected stack frame, like the @value{DBG} ++@code{finish} command. ++ ++@item M-c ++Continue execution of your script, like the @value{DBG} @code{continue} ++command. ++ ++@emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}. ++ ++@item M-u ++Go up the number of frames indicated by the numeric argument ++(@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}), ++like the @value{DBG} @code{up} command. ++ ++@emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}. ++ ++@item M-d ++Go down the number of frames indicated by the numeric argument, like the ++@value{DBG} @code{down} command. ++ ++@emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}. ++ ++@item C-x & ++Read the number where the cursor is positioned, and insert it at the end ++of the @value{DBG} I/O buffer. For example, if you wish to disassemble code ++around an address that was displayed earlier, type @kbd{disassemble}; ++then move the cursor to the address display, and pick up the ++argument for @code{disassemble} by typing @kbd{C-x &}. ++ ++You can customize this further by defining elements of the list ++@code{gdb-print-command}; once it is defined, you can format or ++otherwise process numbers picked up by @kbd{C-x &} before they are ++inserted. A numeric argument to @kbd{C-x &} indicates that you ++wish special formatting, and also acts as an index to pick an element of the ++list. If the list element is a string, the number to be inserted is ++formatted using the Emacs function @code{format}; otherwise the number ++is passed as an argument to the corresponding list element. ++@end table ++ ++In any source file, the Emacs command @kbd{C-x SPC} (@code{gud-break}) ++tells @value{DBG} to set a breakpoint on the source line point is on. ++ ++If you accidentally delete the source-display buffer, an easy way to get ++it back is to type the command @code{frame} in the @value{DBG} buffer, to ++request a frame display; when you run under Emacs, this recreates ++the source buffer if necessary to show you the context of the current ++frame. ++ ++The source files displayed in Emacs are in ordinary Emacs buffers ++which are visiting the source files in the usual way. You can edit ++the files with these buffers if you wish; but keep in mind that @value{DBG} ++communicates with Emacs in terms of line numbers. If you add or ++delete lines from the text, the line numbers that @value{DBG} knows cease ++to correspond properly with the code. ++ ++@node DDD ++@section Using @value{DBG} from @value{DDD} ++@cindex DDD ++ ++@value{DBG} support is rather new in @value{DDD}. As a programming ++language, @value{DBG} is not feature rich: there are no record ++structures or hash tables (yet), no pointers, package variable scoping ++or methods. So much of the data display and visualization features of ++@value{DDD} are disabled. ++ ++As with any scripting or interpreted language (e.g. Perl), one can't ++step by a single machine-language instruction. So the ddd Stepi/Nexti ++commands are disabled. ++ ++Some @value{BASH} settings are essential for @value{DDD} to work ++correctly. These settings with their correct values are: ++ ++@example ++set annotate 1 ++set prompt set prompt bashdb$_Dbg_less$_Dbg_greater$_Dbg_space ++@end example ++ ++@value{DDD} sets these values automatically when invoking ++@value{BASH}; if these values are changed, there may be some ++malfunctions. ++ ++Pay special attention when the prompt has extra angle brackets (a ++nested shell) or has any parenthesis (is in a subshell). Quitting may ++merely exit out of one of these nested (sub)shells rather than leave ++the program. ++ ++@node BASH Debugger Bugs ++@chapter Reporting Bugs ++@cindex bugs ++@cindex reporting bugs ++ ++Your bug reports play an essential role in making the BASH debugger reliable. ++ ++Reporting a bug may help you by bringing a solution to your problem, or it ++may not. But in any case the principal function of a bug report is to help ++the entire community by making the next version of @value{DBG} work better. Bug ++reports are your contribution to the maintenance of @value{DBG}. ++ ++In order for a bug report to serve its purpose, you must include the ++information that enables us to fix the bug. ++ ++@menu ++* Bug Criteria:: Have you found a bug? ++* Bug Reporting:: How to report bugs ++@end menu ++ ++@node Bug Criteria ++@section Have you found a bug? ++@cindex bug criteria ++ ++If you are not sure whether you have found a bug, here are some guidelines: ++ ++@itemize @bullet ++@cindex fatal signal ++@cindex debugger crash ++@cindex crash of debugger ++@item ++If the debugger gets a fatal signal, for any input whatever, that is a ++@value{DBG} bug. Reliable debuggers never crash. ++ ++@cindex error on valid input ++@item ++If @value{DBG} produces an error message for valid input, that is a ++bug. (Note that if you're cross debugging, the problem may also be ++somewhere in the connection to the target.) ++ ++@cindex invalid input ++@item ++If @value{DBG} does not produce an error message for invalid input, ++that is a bug. However, you should note that your idea of ++``invalid input'' might be our idea of ``an extension'' or ``support ++for traditional practice''. ++ ++@item ++If you are an experienced user of debugging tools, your suggestions ++for improvement of @value{DBG} are welcome in any case. ++@end itemize ++ ++@node Bug Reporting ++@section How to report bugs ++@cindex bug reports ++@cindex BASH debugger bugs, reporting ++ ++Submit bug reports for @value{DBG}. Until better mechanism is set up ++the prefered method is to submit them directly using ++@email{mailto:rocky@@panix.com}. There are also a bug tracking system ++and, help, developer and open discussion forums which can be found via ++@uref{http://http://sourceforge.net/forum/?group_id=37260}. ++ ++Although this code shares a lot with BASH, the debugger is of course ++not yet supported by the BASH team, so if there is a debugger problem ++reporting it via bashbug or to @email{bug-bash@@gnu.org} or posted to ++the newsgroup @code{gnu.bash.bug} probably won't help, unless you are ++sure the bug resides in the vanilla bash code and and show that. ++ ++The fundamental principle of reporting bugs usefully is this: ++@strong{report all the facts}. If you are not sure whether to state a ++fact or leave it out, state it! ++ ++Often people omit facts because they think they know what causes the ++problem and assume that some details do not matter. Thus, you might ++assume that the name of the variable you use in an example does not matter. ++Well, probably it does not, but one cannot be sure. Perhaps the bug is a ++stray memory reference which happens to fetch from the location where that ++name is stored in memory; perhaps, if the name were different, the contents ++of that location would fool the debugger into doing the right thing despite ++the bug. Play it safe and give a specific, complete example. That is the ++easiest thing for you to do, and the most helpful. ++ ++Keep in mind that the purpose of a bug report is to enable us to fix the ++bug. It may be that the bug has been reported previously, but neither ++you nor we can know that unless your bug report is complete and ++self-contained. ++ ++Sometimes people give a few sketchy facts and ask, ``Does this ring a ++bell?'' Those bug reports are useless, and we urge everyone to ++@emph{refuse to respond to them} except to chide the sender to report ++bugs properly. ++ ++To enable us to fix the bug, you should include all these things: ++ ++@itemize @bullet ++@item ++The version of @value{DBG}. @value{DBG} announces it if you start ++with no arguments; you can also print it at any time using @code{version} ++command. ++ ++Without this, we will not know whether there is any point in looking for ++the bug in the current version of @value{DBG}. ++ ++@item ++The type of machine you are using, and the operating system name and ++version number. ++ ++@item ++What compiler (and its version) was used to compile BASH---e.g. ++``@value{GCC}--3.1''. ++ ++@item ++The command arguments you gave the compiler to compile your example and ++observe the bug. For example, did you use @samp{-O}? To guarantee ++you will not omit something important, list them all. A copy of the ++Makefile (or the output from make) is sufficient. ++ ++If we were to try to guess the arguments, we would probably guess wrong ++and then we might not encounter the bug. ++ ++@item ++A complete input script, and all necessary source files, that will ++reproduce the bug. ++ ++@item ++A description of what behavior you observe that you believe is ++incorrect. For example, ``It gets a fatal signal.'' ++ ++Of course, if the bug is that @value{DBG} gets a fatal signal, then we ++will certainly notice it. But if the bug is incorrect output, we might ++not notice unless it is glaringly wrong. You might as well not give us ++a chance to make a mistake. ++ ++Even if the problem you experience is a fatal signal, you should still ++say so explicitly. Suppose something strange is going on, such as, your ++copy of @value{DBG} is out of synch, or you have encountered a bug in ++the C library on your system. (This has happened!) Your copy might ++crash and ours would not. If you told us to expect a crash, then when ++ours fails to crash, we would know that the bug was not happening for ++us. If you had not told us to expect a crash, then we would not be able ++to draw any conclusion from our observations. ++ ++@item ++If you wish to suggest changes to the @value{DBG} source, send us context ++diffs. If you even discuss something in the @value{DBG} source, refer to ++it by context, not by line number. ++ ++The line numbers in our development sources will not match those in your ++sources. Your line numbers would convey no useful information to us. ++ ++@end itemize ++ ++Here are some things that are not necessary: ++ ++@itemize @bullet ++@item ++A description of the envelope of the bug. ++ ++Often people who encounter a bug spend a lot of time investigating ++which changes to the input file will make the bug go away and which ++changes will not affect it. ++ ++This is often time consuming and not very useful, because the way we ++will find the bug is by running a single example under the debugger ++with breakpoints, not by pure deduction from a series of examples. ++We recommend that you save your time for something else. ++ ++Of course, if you can find a simpler example to report @emph{instead} ++of the original one, that is a convenience for us. Errors in the ++output will be easier to spot, running under the debugger will take ++less time, and so on. ++ ++However, simplification is not vital; if you do not want to do this, ++report the bug anyway and send us the entire test case you used. ++ ++@item ++A patch for the bug. ++ ++A patch for the bug does help us if it is a good one. But do not omit ++the necessary information, such as the test case, on the assumption that ++a patch is all we need. We might see problems with your patch and decide ++to fix the problem another way, or we might not understand it at all. ++ ++Sometimes with a program as complicated as @value{DBG} it is very hard to ++construct an example that will make the program follow a certain path ++through the code. If you do not send us the example, we will not be able ++to construct one, so we will not be able to verify that the bug is fixed. ++ ++And if we cannot understand what bug you are trying to fix, or why your ++patch should be an improvement, we will not install it. A test case will ++help us to understand. ++ ++@item ++A guess about what the bug is or what it depends on. ++ ++Such guesses are usually wrong. Even we cannot guess right about such ++things without first using the debugger to find the facts. ++@end itemize ++ ++@node History and Acknowledgments ++@chapter History and Acknowledgments ++ ++The suggestion for a debugger for a Bourne-like shell came from the book ++``Learning the Korn Shell'', by Bill Rosenblatt Copyright (C) 1993 by ++O'Reilly and Associates, Inc. Others such as Cigy Cyriac, Chet Ramey, ++Rocky Bernstein, and Gary V. Vaughan expanded and improved on that. ++ ++However Bourne-Shell debuggers rely on a signal mechanism ++(@code{SIGDEBUG}) to call a debugger routine. In the Korn shell as ++well as @sc{bash} in versions prior to 2.05, there was a fundamental ++flaw: the routine that you registered in the trap, got called ++@emph{after} the statement was executed. It takes little imagination ++to realize that this is a bit too late to find and correct errors, ++especially if the offending command happens to do serious damage like ++remove filesystems or reboot a server. As a horrible hack, these ++debuggers added one to the line number that was just executed on the ++wishful thinking that this would then be the line of next statement to ++execute. Sometimes this was correct, but it was too often wrong, such ++as in loops and conditionals, comments, or commands that are continued ++on the next line. ++ ++Another failing of these debuggers was the inability to debug into ++functions or into sourced files, provide a stack trace, dynamically ++skip a statement to be run, unconditionally trace into a function, or ++stop when a subroutine or a sourced file completed. In truth, the crux ++of the problem lay in debugging support in BASH. Given that there was ++limited bash debugging support, it is not surprising that these ++debuggers could not do any of the things listed above and could debug ++only a single source file: lines could be listed only from a single ++text, breakpoints were set into the text which was in fact a copy of ++the script name prepended with debugger routines. ++ ++In version 2.04 of BASH, Rocky Bernstein started hacking on BASH to ++add call-stack information, source file information, allow for ++debugging into functions and for reporting line numbers in functions ++as relative to the file rather than the beginning of a function whose ++origin line number was not accessible from BASH. He started changing ++the user commands in bashdb to be like other more-advanced debuggers, ++in particular @code{perl5db} and @code{gdb}. However he gave up on ++this project when realizing that stopping before a line was crucial. A ++patch for this was nontrivial and wildly changed ++semantics. Furthermore the chance of getting his other patches into ++BASH was was not going to happen in version 2.04. ++ ++In version 2.05, the fundamental necessary change to the semantics of ++@code{SIGDEBUG} trap handling (suggested at least two years earlier) ++was made. Also, version 2.05 changed the line-number reporting in a ++function to be relative to the beginning of the file rather than the ++beginning of a function---sometimes. Rocky then picked up where he ++left off and this then became this debugger. A complete rewrite of the ++debugger, some of which started in 2.04 was undertaken. Debugger ++internals were changed to support multiple file names, save and ++restore the calling environment (such as variables @code{$1} and ++@code{$?}) and install debugger signal handlers. Work was also done on ++the BASH in conjunction with the debugger to save stack trace ++information, provide a means for stopping after a routine finished and ++so on. And a number of changes were made to BASH just to improve the ++accuracy of the line number reporting which is crucial in a debugger. ++ ++This documentation was modified from the GNU Debugger (GDB) Reference ++manual. ++ ++@quotation ++Additions to this section are particularly welcome. If you or your ++friends (or enemies, to be evenhanded) have been unfairly omitted from ++this list, we would like to add your names! ++@end quotation ++ ++The following have contributed directly or indrectly to bashdb: ++ ++Rocky Bernstein (initial full-featured bashdb with stack tracing and ++multi-file support) ++ ++Masatake YAMATO (help to merge Rocky's hack to the official bash source tree) ++ ++Bill Rosenblatt (kshdb), ++Michael Loukides (kshdb), ++Cigy Cyriac (proto bashdb), ++Chet Ramey (proto bashdb), ++and ++Gary V. Vaughan (proto bashdb). ++ ++Authors of per5ldb: ++ ++Ray Lischner, ++Johan Vromans, ++and ++Ilya Zakharevich. ++ ++Authors of GDB: ++ ++Richard Stallman, ++Andrew Cagney, ++Jim Blandy, ++Jason Molenda, ++Stan Shebs, ++Fred Fish, ++Stu Grossman, ++John Gilmore, ++Jim Kingdon, ++and ++Randy Smith (to name just a few). ++ ++Authors of GUD: ++ ++Eric S. Raymond. ++ ++@c The readline documentation is distributed with the readline code ++@c and consists of the two following files: ++@c rluser.texinfo ++@c inc-hist.texinfo ++@c Use -I with makeinfo to point to the appropriate directory, ++@c environment var TEXINPUTS with TeX. ++@c @include rluser.texinfo ++@c @include hsuser.texinfo ++ ++ ++@node Formatting Documentation ++@appendix Formatting Documentation ++ ++@ifset FINISHED ++@cindex @value{DBG} reference card ++@cindex reference card ++The @value{DBG} 4 release includes an already-formatted reference card, ready ++for printing with PostScript or Ghostscript, in the @file{gdb} ++subdirectory of the main source directory@footnote{In ++@file{gdb-@value{BASHRELEASE}/gdb/refcard.ps} of the version @value{BASHRELEASE} ++release.}. If you can use PostScript or Ghostscript with your printer, ++you can print the reference card immediately with @file{refcard.ps}. ++ ++The release also includes the source for the reference card. You ++can format it, using @TeX{}, by typing: ++ ++@example ++make refcard.dvi ++@end example ++ ++The @value{DBG} reference card is designed to print in @dfn{landscape} ++mode on US ``letter'' size paper; ++that is, on a sheet 11 inches wide by 8.5 inches ++high. You will need to specify this form of printing as an option to ++your @sc{dvi} output program. ++ ++@cindex documentation ++ ++All the documentation for @value{DBG} comes as part of the machine-readable ++distribution. The documentation is written in Texinfo format, which is ++a documentation system that uses a single source file to produce both ++on-line information and a printed manual. You can use one of the Info ++formatting commands to create the on-line version of the documentation ++and @TeX{} (or @code{texi2roff}) to typeset the printed version. ++ ++@value{DBG} includes an already formatted copy of the on-line Info ++version of this manual in the @file{gdb} subdirectory. The main Info ++file is @file{gdb-@value{BASHRELEASE}/gdb/gdb.info}, and it refers to ++subordinate files matching @samp{gdb.info*} in the same directory. If ++necessary, you can print out these files, or read them with any editor; ++but they are easier to read using the @code{info} subsystem in @sc{gnu} ++Emacs or the standalone @code{info} program, available as part of the ++@sc{gnu} Texinfo distribution. ++ ++If you want to format these Info files yourself, you need one of the ++Info formatting programs, such as @code{texinfo-format-buffer} or ++@code{makeinfo}. ++ ++If you have @code{makeinfo} installed, and are in the top level ++@value{DBG} source directory (@file{gdb-@value{BASHRELEASE}}, in the case of ++version @value{BASHRELEASE}), you can make the Info file by typing: ++ ++@example ++cd gdb ++make gdb.info ++@end example ++ ++If you want to typeset and print copies of this manual, you need @TeX{}, ++a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the ++Texinfo definitions file. ++ ++@TeX{} is a typesetting program; it does not print files directly, but ++produces output files called @sc{dvi} files. To print a typeset ++document, you need a program to print @sc{dvi} files. If your system ++has @TeX{} installed, chances are it has such a program. The precise ++command to use depends on your system; @kbd{lpr -d} is common; another ++(for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may ++require a file name without any extension or a @samp{.dvi} extension. ++ ++@TeX{} also requires a macro definitions file called ++@file{texinfo.tex}. This file tells @TeX{} how to typeset a document ++written in Texinfo format. On its own, @TeX{} cannot either read or ++typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB ++and is located in the @file{gdb-@var{version-number}/texinfo} ++directory. ++ ++If you have @TeX{} and a @sc{dvi} printer program installed, you can ++typeset and print this manual. First switch to the the @file{gdb} ++subdirectory of the main source directory (for example, to ++@file{gdb-@value{BASHRELEASE}/gdb}) and type: ++ ++@example ++make gdb.dvi ++@end example ++ ++Then give @file{gdb.dvi} to your @sc{dvi} printing program. ++@end ifset ++ ++@node Installing bashdb ++@appendix Installing @value{DBG} ++@cindex configuring @value{DBG} ++@cindex installation ++ ++@ifset FINISHED ++@value{DBG} comes with a @code{configure} script that automates the process ++of preparing @value{DBG} for installation; you can then use @code{make} to ++build the @code{gdb} program. ++@iftex ++@c irrelevant in info file; it's as current as the code it lives with. ++@footnote{If you have a more recent version of @value{DBG} than @value{BASHRELEASE}, ++look at the @file{README} file in the sources; we may have improved the ++installation procedures since publishing this manual.} ++@end iftex ++ ++The @value{DBG} distribution includes all the source code you need for ++@value{DBG} in a single directory, whose name is usually composed by ++appending the version number to @samp{gdb}. ++ ++For example, the @value{DBG} version @value{BASHRELEASE} distribution is in the ++@file{gdb-@value{BASHRELEASE}} directory. That directory contains: ++ ++@table @code ++@item gdb-@value{BASHRELEASE}/configure @r{(and supporting files)} ++script for configuring @value{DBG} and all its supporting libraries ++ ++@item gdb-@value{BASHRELEASE}/gdb ++the source specific to @value{DBG} itself ++ ++@item gdb-@value{BASHRELEASE}/bfd ++source for the Binary File Descriptor library ++ ++@item gdb-@value{BASHRELEASE}/include ++@sc{gnu} include files ++ ++@item gdb-@value{BASHRELEASE}/libiberty ++source for the @samp{-liberty} free software library ++ ++@item gdb-@value{BASHRELEASE}/opcodes ++source for the library of opcode tables and disassemblers ++ ++@item gdb-@value{BASHRELEASE}/readline ++source for the @sc{gnu} command-line interface ++ ++@item gdb-@value{BASHRELEASE}/glob ++source for the @sc{gnu} filename pattern-matching subroutine ++ ++@item gdb-@value{BASHRELEASE}/mmalloc ++source for the @sc{gnu} memory-mapped malloc package ++@end table ++ ++The simplest way to configure and build @value{DBG} is to run @code{configure} ++from the @file{gdb-@var{version-number}} source directory, which in ++this example is the @file{gdb-@value{BASHRELEASE}} directory. ++ ++First switch to the @file{gdb-@var{version-number}} source directory ++if you are not already in it; then run @code{configure}. Pass the ++identifier for the platform on which @value{DBG} will run as an ++argument. ++ ++For example: ++ ++@example ++cd gdb-@value{BASHRELEASE} ++./configure @var{host} ++make ++@end example ++ ++@noindent ++where @var{host} is an identifier such as @samp{sun4} or ++@samp{decstation}, that identifies the platform where @value{DBG} will run. ++(You can often leave off @var{host}; @code{configure} tries to guess the ++correct value by examining your system.) ++ ++Running @samp{configure @var{host}} and then running @code{make} builds the ++@file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty} ++libraries, then @code{gdb} itself. The configured source files, and the ++binaries, are left in the corresponding source directories. ++ ++@need 750 ++@code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your ++system does not recognize this automatically when you run a different ++shell, you may need to run @code{sh} on it explicitly: ++ ++@example ++sh configure @var{host} ++@end example ++ ++If you run @code{configure} from a directory that contains source ++directories for multiple libraries or programs, such as the ++@file{gdb-@value{BASHRELEASE}} source directory for version @value{BASHRELEASE}, @code{configure} ++creates configuration files for every directory level underneath (unless ++you tell it not to, with the @samp{--norecursion} option). ++ ++You can run the @code{configure} script from any of the ++subordinate directories in the @value{DBG} distribution if you only want to ++configure that subdirectory, but be sure to specify a path to it. ++ ++For example, with version @value{BASHRELEASE}, type the following to configure only ++the @code{bfd} subdirectory: ++ ++@example ++@group ++cd gdb-@value{BASHRELEASE}/bfd ++../configure @var{host} ++@end group ++@end example ++ ++You can install @code{@value{DBG}} anywhere; it has no hardwired paths. ++However, you should make sure that the shell on your path (named by ++the @samp{SHELL} environment variable) is publicly readable. Remember ++that @value{DBG} uses the shell to start your script---some systems refuse to ++let @value{DBG} debug child processes whose programs are not readable. ++ ++@menu ++* Separate Objdir:: Compiling @value{DBG} in another directory ++* Config Names:: Specifying names for hosts and targets ++* Configure Options:: Summary of options for configure ++@end menu ++ ++@node Separate Objdir ++@section Compiling @value{DBG} in another directory ++ ++If you want to run @value{DBG} versions for several host or target machines, ++you need a different @code{gdb} compiled for each combination of ++host and target. @code{configure} is designed to make this easy by ++allowing you to generate each configuration in a separate subdirectory, ++rather than in the source directory. If your @code{make} program ++handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running ++@code{make} in each of these directories builds the @code{gdb} ++program specified there. ++ ++To build @code{gdb} in a separate directory, run @code{configure} ++with the @samp{--srcdir} option to specify where to find the source. ++(You also need to specify a path to find @code{configure} ++itself from your working directory. If the path to @code{configure} ++would be the same as the argument to @samp{--srcdir}, you can leave out ++the @samp{--srcdir} option; it is assumed.) ++ ++For example, with version @value{BASHRELEASE}, you can build @value{DBG} in a ++separate directory for a Sun 4 like this: ++ ++@example ++@group ++cd gdb-@value{BASHRELEASE} ++mkdir ../gdb-sun4 ++cd ../gdb-sun4 ++../gdb-@value{BASHRELEASE}/configure sun4 ++make ++@end group ++@end example ++ ++When @code{configure} builds a configuration using a remote source ++directory, it creates a tree for the binaries with the same structure ++(and using the same names) as the tree under the source directory. In ++the example, you'd find the Sun 4 library @file{libiberty.a} in the ++directory @file{gdb-sun4/libiberty}, and @value{DBG} itself in ++@file{gdb-sun4/gdb}. ++ ++One popular reason to build several @value{DBG} configurations in separate ++directories is to configure @value{DBG} for cross-compiling (where ++@value{DBG} runs on one machine---the @dfn{host}---while debugging ++programs that run on another machine---the @dfn{target}). ++You specify a cross-debugging target by ++giving the @samp{--target=@var{target}} option to @code{configure}. ++ ++When you run @code{make} to build a program or library, you must run ++it in a configured directory---whatever directory you were in when you ++called @code{configure} (or one of its subdirectories). ++ ++The @code{Makefile} that @code{configure} generates in each source ++directory also runs recursively. If you type @code{make} in a source ++directory such as @file{gdb-@value{BASHRELEASE}} (or in a separate configured ++directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{BASHRELEASE}}), you ++will build all the required libraries, and then build GDB. ++ ++When you have multiple hosts or targets configured in separate ++directories, you can run @code{make} on them in parallel (for example, ++if they are NFS-mounted on each of the hosts); they will not interfere ++with each other. ++ ++@node Config Names ++@section Specifying names for hosts and targets ++ ++The specifications used for hosts and targets in the @code{configure} ++script are based on a three-part naming scheme, but some short predefined ++aliases are also supported. The full naming scheme encodes three pieces ++of information in the following pattern: ++ ++@example ++@var{architecture}-@var{vendor}-@var{os} ++@end example ++ ++For example, you can use the alias @code{sun4} as a @var{host} argument, ++or as the value for @var{target} in a @code{--target=@var{target}} ++option. The equivalent full name is @samp{sparc-sun-sunos4}. ++ ++The @code{configure} script accompanying @value{DBG} does not provide ++any query facility to list all supported host and target names or ++aliases. @code{configure} calls the Bourne shell script ++@code{config.sub} to map abbreviations to full names; you can read the ++script, if you wish, or you can use it to test your guesses on ++abbreviations---for example: ++ ++@smallexample ++% sh config.sub i386-linux ++i386-pc-linux-gnu ++% sh config.sub alpha-linux ++alpha-unknown-linux-gnu ++% sh config.sub hp9k700 ++hppa1.1-hp-hpux ++% sh config.sub sun4 ++sparc-sun-sunos4.1.1 ++% sh config.sub sun3 ++m68k-sun-sunos4.1.1 ++% sh config.sub i986v ++Invalid configuration `i986v': machine `i986v' not recognized ++@end smallexample ++ ++@noindent ++@code{config.sub} is also distributed in the @value{DBG} source ++directory (@file{gdb-@value{BASHRELEASE}}, for version @value{BASHRELEASE}). ++ ++@node Configure Options ++@section @code{configure} options ++ ++Here is a summary of the @code{configure} options and arguments that ++are most often useful for building @value{DBG}. @code{configure} also has ++several other options not listed here. @inforef{What Configure ++Does,,configure.info}, for a full explanation of @code{configure}. ++ ++@example ++configure @r{[}--help@r{]} ++ @r{[}--prefix=@var{dir}@r{]} ++ @r{[}--exec-prefix=@var{dir}@r{]} ++ @r{[}--srcdir=@var{dirname}@r{]} ++ @r{[}--norecursion@r{]} @r{[}--rm@r{]} ++ @r{[}--target=@var{target}@r{]} ++ @var{host} ++@end example ++ ++@noindent ++You may introduce options with a single @samp{-} rather than ++@samp{--} if you prefer; but you may abbreviate option names if you use ++@samp{--}. ++ ++@table @code ++@item --help ++Display a quick summary of how to invoke @code{configure}. ++ ++@item --prefix=@var{dir} ++Configure the source to install programs and files under directory ++@file{@var{dir}}. ++ ++@item --exec-prefix=@var{dir} ++Configure the source to install programs under directory ++@file{@var{dir}}. ++ ++@c avoid splitting the warning from the explanation: ++@need 2000 ++@item --srcdir=@var{dirname} ++@strong{Warning: using this option requires @sc{gnu} @code{make}, or another ++@code{make} that implements the @code{VPATH} feature.}@* ++Use this option to make configurations in directories separate from the ++@value{DBG} source directories. Among other things, you can use this to ++build (or maintain) several configurations simultaneously, in separate ++directories. @code{configure} writes configuration specific files in ++the current directory, but arranges for them to use the source in the ++directory @var{dirname}. @code{configure} creates directories under ++the working directory in parallel to the source directories below ++@var{dirname}. ++ ++@item --norecursion ++Configure only the directory level where @code{configure} is executed; do not ++propagate configuration to subdirectories. ++ ++@item --target=@var{target} ++Configure @value{DBG} for cross-debugging programs running on the specified ++@var{target}. Without this option, @value{DBG} is configured to debug ++programs that run on the same machine (@var{host}) as @value{DBG} itself. ++ ++There is no convenient way to generate a list of all available targets. ++ ++@item @var{host} @dots{} ++Configure @value{DBG} to run on the specified @var{host}. ++ ++There is no convenient way to generate a list of all available hosts. ++@end table ++ ++There are many other options available as well, but they are generally ++needed for special purposes only. ++@end ifset ++ ++@include gpl.texi ++ ++@include fdl.texi ++ ++@node Function Index ++@unnumbered Function Index ++@printindex fn ++ ++@node Command Index ++@unnumbered Command Index ++@printindex ky ++ ++@node Variable Index ++@unnumbered Variable Index ++@printindex vr ++ ++@node General Index ++@unnumbered General Index ++@printindex cp ++ ++@tex ++% I think something like @colophon should be in texinfo. In the ++% meantime: ++\long\def\colophon{\hbox to0pt{}\vfill ++\centerline{The body of this manual is set in} ++\centerline{\fontname\tenrm,} ++\centerline{with headings in {\bf\fontname\tenbf}} ++\centerline{and examples in {\tt\fontname\tentt}.} ++\centerline{{\it\fontname\tenit\/},} ++\centerline{{\bf\fontname\tenbf}, and} ++\centerline{{\sl\fontname\tensl\/}} ++\centerline{are used for emphasis.}\vfill} ++\page\colophon ++% Blame: doc@cygnus.com, 1991. ++@end tex ++ ++@bye +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/doc/Makefile.am cvs/debugger/doc/Makefile.am +--- bash-2.05b/debugger/doc/Makefile.am 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/doc/Makefile.am 2003-09-08 08:40:26.000000000 +0200 +@@ -0,0 +1,62 @@ ++############################################################################## ++# $Id: Makefile.am,v 1.5 2003/09/02 00:26:24 rockyb Exp $ ++# Copyright (C) 2003 Rocky Bernstein ++# This program is free software; you can redistribute it and/or modify ++# it under the terms of the GNU General Public License as published by ++# the Free Software Foundation; either version 2 of the License, or ++# (at your option) any later version. ++# ++# This program is distributed in the hope that it will be useful, ++# but WITHOUT ANY WARRANTY; without even the implied warranty of ++# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++# GNU General Public License for more details. ++# ++# You should have received a copy of the GNU General Public License ++# along with this program; if not, write to the Free Software ++# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA ++############################################################################## ++ ++EXT=1 ++man1_MANS = bashdb.$(EXT) ++ ++EXTRA_DIST = $(man1_MANS) bashdb-man.pod bashdb-man.html ++ ++info_TEXINFOS = bashdb.texi ++bashdb_TEXINFOS = fdl.texi gpl.texi ++ ++all: $(INFO_DEPS) $(man1_MANS) html ++ ++pdf: bashdb.pdf ++ ++txt: bashdb.txt ++ ++ps: bashdb.ps ++ ++man: $(man1_MANS) ++ ++html: bashdb.html bashdb-man.html ++ ++bashdb.html: bashdb.texi ++ texi2html $(srcdir)/bashdb.texi ++ ++bashdb-man.html: bashdb-man.pod ++ pod2html --infile=$(srcdir)/bashdb-man.pod --outfile=$@ ++ ++$(man1_MANS): bashdb-man.pod ++ pod2man --section=$(EXT) $(srcdir)/bashdb-man.pod >$@ ++ ++%.ps.gz: %.ps ++ gzip -9c $< > $@ ++ ++.texi.pdf: ++ texi2pdf $< ++ ++.texi.html: ++ texi2html $< ++ ++.texi.txt: ++ makeinfo --no-headers $< > $@ ++ ++all-formats: pdf dvi txt ps html ++ ++MOSTLYCLEANFILES = bashdb.tgs bashdb.ps.gz bashdb.pdf bashdb.html bashdb_toc.html bashdb_foot.html $(man1_MANS) bashdb-man.html +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/doc/bashdb-man.pod cvs/debugger/doc/bashdb-man.pod +--- bash-2.05b/debugger/doc/bashdb-man.pod 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/doc/bashdb-man.pod 2003-08-12 15:56:11.000000000 +0200 +@@ -0,0 +1,192 @@ ++=pod ++ ++=head1 NAME ++ ++bashdb - bash debugger script ++ ++=head1 SYNOPSIS ++ ++B [I] I ++ ++B [I] -c I ++ ++B [I...] I ++ ++=head1 DESCRIPTION ++ ++C is a bash script to which arranges for another bash script ++to be debugged. The debugger has a similar command interface as L ++or Perl's L debugger. The way this script arranges debugging ++to occur is by including (or actually "source"-ing) some debug-support ++code and then sourcing the given script or command string. ++ ++One problem with sourcing a debugged script is that the program name ++stored in $0 will be C rather than the name of the script to ++be debugged. The debugged script will appear in a call stack not as ++the top item but as the item below C. If this is of concern, ++use the last form given above, C I. ++ ++A downside of invoking bash with the C<--debugger> option is any of ++the options below that are not bash options don't work, and those that ++I bash options have the bash meaning rather than the bash ++debugger meaning. For example, C<-n> in bash means don't run a bash ++script but just syntax check it which is different from what is listed ++below. ++ ++=head1 OPTIONS ++ ++=over 4 ++ ++=item -h ++ ++Print a usage message on standard error and exit with a return code ++of 100. ++ ++Z<> ++ ++=item -B ++ ++In places where a filename appears in debugger output give just the ++basename only. This is needed in for regression testing. Using this ++option is equivalent to issuing: ++ ++ set basename on ++ ++inside the debugger. ++ ++Z<> ++ ++=item -n ++ ++Normally the debugger will read debugger commands in C<~/.bashdbinit> ++if that file exists before accepting user interaction. ++C<.bashdbinit> is analogus to Perl's C<.perldb> or GNU gdb's ++C<.gdbinit>: a user might want to create such a debugger profile to ++add various user-specific customizations. ++ ++Using the C<-n> option this initialization file will not be read. This ++is useful in regression testing or in tracking down a problem with ++one's C<.bashdbinit> profile. ++ ++Z<> ++ ++=item -c I ++ ++Instead of specifying the name of a bash script file, one can give an ++execution string that is to be debugged. Use this option to do that. ++ ++If you invoke the debugger via C, the filename that will ++appear in source listing or in a call stack trace will be the artifical name ++*BOGUS*. ++ ++Z<> ++ ++=item -q ++ ++Do not print introductory version and copyright information. This is ++again useful in regression testing where we don't want to include a ++changeable copyright date in the regression-test matching. ++ ++Z<> ++ ++=item -x I ++ ++Run the debugger commands I before accepting user ++input. These commands are read however after any C<.bashdbinit> ++commands. Again this is useful running regression-testing debug ++scripts. ++ ++Z<> ++ ++=item -L I ++ ++The debugger needs to source or include a number of functions and ++these reside in a library. If this option is not given the default location ++of library is relative to the installed bashdb script: C<../lib/bashdb>. ++ ++Z<> ++ ++=item -T I ++ ++The debugger needs to make use of some temporary filesystem storage to ++save persistent information across a subshell return or in order to ++evaluate an expression. The default directory is C but you can ++use this option to set the directory where debugger temporary files ++will be created. ++ ++Z<> ++ ++=item -t I ++ ++Debugger output usually goes to a terminal rather than stdout or stdin ++which the debugged program may use. Determination of the tty or ++pseudo-tty is normally done automatically. However if you want to ++control where the debugger output goes, use this option. ++ ++Z<> ++ ++=item -V ++ ++Show version number and no-warranty and exit with return code 1. ++ ++=back ++ ++=head1 BUGS ++ ++The C script and C<--debugger> option assume a patched version ++of bash. That is you can't debug bash scripts using the standard-issue ++version 2.05 bash or earlier versions. If you try to run the bashdb ++script on such as shell, may get the message: ++ ++ Sorry, you need to use a debugger-enabled version of bash. ++ ++This is not a bug in the debugger so much as a bug in bash itself or ++the lack of debugging support thereof. ++ ++Debugging can be slow especially on large bash scripts. Scripts ++created by GNU autoconf are at a minimum hundreds of lines and it is ++not uncommon for them to be tens of thousands of lines. ++ ++Part of the reason of the debugger slowness is that the debugger has ++to intercept every line and check to see if some action is to be taken ++for this and this is all in bash code. A better and faster ++architecture would be for the debugger to register a list of ++conditions or stopping places inside the bash code itself and have it ++arrange to call the debugger only when a condition requiring the ++debugger arises. Checks would be faster as this would be done in C ++code and access to internal structures would make this moe ++efficient. Did I mention the lack of debug support in bash (and other ++POSIX shells)? ++ ++Another place you may find slowness is in initial startup of such ++large debugger scripts. The source code has to be read into internal ++arrays and this apparently takes time. ++ ++=head1 SEE ALSO ++ ++L. There also an extensive debugger reference manual. ++ ++=head1 AUTHOR ++ ++The current version is maintained (or not) by C. ++ ++=head1 COPYRIGHT ++ ++ Copyright (C) 2003 Rocky Bernstein, email: rocky@panix.com. ++ This program is free software; you can redistribute it and/or modify ++ it under the terms of the GNU General Public License as published by ++ the Free Software Foundation; either version 2 of the License, or ++ (at your option) any later version. ++ ++ This program is distributed in the hope that it will be useful, ++ but WITHOUT ANY WARRANTY; without even the implied warranty of ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ GNU General Public License for more details. ++ ++ You should have received a copy of the GNU General Public License ++ along with this program; if not, write to the Free Software ++ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. ++ ++I<$Id: bashdb-man.pod,v 1.2 2003/08/12 13:56:11 rockyb Exp $> ++ ++=cut +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/doc/bashdb.texi cvs/debugger/doc/bashdb.texi +--- bash-2.05b/debugger/doc/bashdb.texi 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/doc/bashdb.texi 2003-09-08 08:40:27.000000000 +0200 +@@ -0,0 +1,3759 @@ ++\input texinfo @c -*-texinfo-*- ++@c Copyright 2002, 2003 ++@c Rocky Bernstein for the Free Software Foundation ++@c ++@c TODO: ++@c - add examples for commands ++@c - clean up/improve sample session ++@c - help text is inaccurate and formatted too much to right. ++@c ++ ++@c Sets version and release names and dates. Frees us from changing ++@c this file when a new release comes along. ++ ++@c %**start of header ++@c makeinfo ignores cmds prev to setfilename, so its arg cannot make use ++@c of @set vars. However, you can override filename with makeinfo -o. ++@setfilename bashdb.info ++@c ++@c Name of Bash program. Used in running text. ++@set BASH @acronym{BASH} ++ ++@c ++@c Name of debugger program. Used also for prompt string. ++@set DBG the @value{BASH} debugger ++@set dBGP The @value{BASH} debugger ++@set DDD @acronym{DDD} ++ ++@settitle @value{BASH} Debugger ++@setchapternewpage odd ++@setcontentsaftertitlepage ++ ++@c %**end of header ++ ++@include version.texi ++ ++@c A simple macro for optional variables. ++@macro ovar{varname} ++@r{[}@var{\varname\}@r{]} ++@end macro ++ ++@macro DDD {} ++@acronym{DDD} ++@end macro ++ ++@macro BASH {} ++@acronym{@value{BASH}} ++@end macro ++ ++@macro DBG {} ++@value{DBG} ++@end macro ++ ++@iftex ++@c @smallbook ++@c @cropmarks ++@end iftex ++ ++@finalout ++ ++@c readline appendices use @vindex, @findex and @ftable, ++@c annotate.texi and gdbmi use @findex. ++@c @syncodeindex vr cp ++@c @syncodeindex fn cp ++ ++@c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER. ++ ++@c This is a dir.info fragment to support semi-automated addition of ++@c manuals to an info tree. ++@dircategory Programming & development tools. ++@direntry ++* @value{DBG}: (@value{DBG}). The @sc{bash} debugger. ++@end direntry ++ ++@ifinfo ++This file documents the @sc{bash} debugger @value{BASH}. ++ ++ ++This is the @value{EDITION} Edition, @value{UPDATED-MONTH}, ++of @cite{Debugging with BASHDB: the @sc{gnu} Source-Level Debugger} ++for REBASH ++ ++Copyright (C) 2002,2003 Rocky Bernstein for the Free Software Foundation. ++ ++Permission is granted to copy, distribute and/or modify this document ++under the terms of the GNU Free Documentation License, Version 1.1 or ++@ifset DEBIANHASBECOMEREASONABLE ++@c From Matthias Klose a Debian maintainer on ++@c Sat, 23 Aug 2003 14:24:44 +0200 ++@c ++@c I personally see the invariant sections as the thing in the ++@c GFDL, which hinders me in uploading the package to the archives. ++@c I don't have any problem, if some other Debian developer makes a ++@c bashdb package built from separate sources. ++@c ++@c I am aware that Debian ships other packages containing documentation ++@c covered by the GFDL (and one of them for which I do the packaging as ++@c well), but I won't add a new package, which I maintain. So before an ++@c upload of a bashdb package built from the bash sources either ++@c ++@c ++@c - Debian has a position on the GFDL, which allows inclusion ++@c ++@c - the bashdb manual does not have invariant sections, or is ++@c relicensed, or dual licensed. ++@c ++any later version published by the Free Software Foundation; with the ++Invariant Sections being ``Free Software'' and ``Free Software Needs ++Free Documentation'', with the Front-Cover Texts being ``A GNU ++Manual,'' and with the Back-Cover Texts as in (a) below. ++ ++(a) The Free Software Foundation's Back-Cover Text is: ``You have ++freedom to copy and modify this GNU Manual, like GNU software. Copies ++published by the Free Software Foundation raise funds for GNU ++development.'' ++@end ifset ++@ifclear DEBIANHASBECOMEREASONABLE ++any later version published by the Free Software Foundation; with no ++Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. ++@end ifclear ++@end ifinfo ++ ++@titlepage ++@title Debugging with BASHDB ++@sp 1 ++@subtitle @value{EDITION} Edition, for BASH ++@subtitle @value{UPDATED-MONTH} ++@author Rocky Bernstein ++@page ++@tex ++{\parskip=0pt ++\hfill (Send bugs and comments on bashdb to bug-bashdb\@sourceforge.net.)\par ++\hfill {\it Debugging with BASH}\par ++\hfill \TeX{}info \texinfoversion\par ++} ++@end tex ++ ++@vskip 0pt plus 1filll ++Copyright @copyright{} 2002,2003 Rocky Bernstein for the Free Software ++Foundation. ++ ++Permission is granted to copy, distribute and/or modify this document ++under the terms of the GNU Free Documentation License, Version 1.1 or ++@ifset DEBIANHASBECOMEREASONABLE ++@c From Matthias Klose a Debian maintainer on ++@c Sat, 23 Aug 2003 14:24:44 +0200 ++@c ++@c I personally see the invariant sections as the thing in the ++@c GFDL, which hinders me in uploading the package to the archives. ++@c I don't have any problem, if some other Debian developer makes a ++@c bashdb package built from separate sources. ++@c ++@c I am aware that Debian ships other packages containing documentation ++@c covered by the GFDL (and one of them for which I do the packaging as ++@c well), but I won't add a new package, which I maintain. So before an ++@c upload of a bashdb package built from the bash sources either ++@c ++@c ++@c - Debian has a position on the GFDL, which allows inclusion ++@c ++@c - the bashdb manual does not have invariant sections, or is ++@c relicensed, or dual licensed. ++@c ++@c ++any later version published by the Free Software Foundation; with the ++Invariant Sections being ``Free Software'' and ``Free Software Needs ++Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,'' ++and with the Back-Cover Texts as in (a) below. ++ ++(a) The Free Software Foundation's Back-Cover Text is: ``You have ++freedom to copy and modify this GNU Manual, like GNU software. Copies ++published by the Free Software Foundation raise funds for GNU ++development.'' ++@end ifset ++@ifclear DEBIANHASBECOMEREASONABLE ++any later version published by the Free Software Foundation; with no ++Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. ++@end ifclear ++@end titlepage ++@page ++ ++@ifnottex ++@node Top, Summary, (dir), (dir) ++ ++@top Debugging with @DBG ++ ++This file describes @value{DBG}, the @sc{bash} symbolic debugger. ++ ++This is the @value{EDITION} Edition, @value{UPDATED-MONTH}, for REBASH. ++ ++Copyright (C) 2002,2003 Rocky Bernstein ++ ++@menu ++* Summary:: Overview of Debugger with a sample session ++ ++* Invocation:: Getting in and out ++* Running:: Script setup inside the BASH debugger ++* Debugger Command Reference:: BASH debugger command reference ++ ++* Front Ends:: Using @DBG from a front-end user interface ++ ++* BASH Debugger Bugs:: Reporting bugs ++* History and Acknowledgments:: History and Acknowledgments ++ ++Appendices ++* Formatting Documentation:: How to format and print @DBG documentation ++ ++* Installing bashdb:: Installing bashdb ++* Copying:: GNU General Public License says ++ how you can copy and share bashdb ++* GNU Free Documentation License:: The license for this documentation ++ ++Indexes (nodes containing large menus) ++* Function Index:: An item for each function name. ++* Command Index:: An item for each command name. ++* Variable Index:: An item for each documented variable. ++* General Index:: An item for each concept. ++@end menu ++ ++@end ifnottex ++ ++@contents ++ ++@node Summary ++@chapter Summary of the BASH Debugger ++ ++The purpose of a debugger such as @DBG is to allow you to see what is ++going on ``inside'' a bash script while it executes. ++ ++@DBG can do four main kinds of things (plus other things in support of ++these) to help you catch bugs in the act: ++ ++@itemize @bullet ++@item ++Start your script, specifying anything that might affect its behavior. ++ ++@item ++Make your script stop on specified conditions. ++ ++@item ++Examine what has happened, when your script has stopped. ++ ++@item ++Change things in your script, so you can experiment with correcting the ++effects of one bug and go on to learn about another. ++@end itemize ++ ++Although you can use the BASH debugger to debug scripts written in ++BASH, it can also be used just as a front-end for learning more ++about programming in BASH. As an additional aid, the debugger can ++be used within the context of an existing script with its functions ++and variables that have already been initialized; fragments of the ++existing can be experimented with by entering them inside the debugger. ++ ++@menu ++* Sample Session:: A Sample BASH Debugger session ++* Free Software:: Freely redistributable software ++@end menu ++ ++@node Sample Session ++@section A Sample BASH Debugger Session ++ ++You can use this manual at your leisure to read all about @value{DBG}. ++However, a handful of commands are enough to get started using the ++debugger. This chapter illustrates those commands. ++ ++@iftex ++In this sample session, we emphasize user input like this: @b{input}, ++to make it easier to pick out from the surrounding output. ++@end iftex ++ ++Below we will debug a script that contains a function to compute the ++factorial of a number: fact(0) is 1 and fact(n) is n*fact(n-1). ++ ++@smallexample ++$ @b{bashdb -L . fact.sh} ++Bourne-Again Shell Debugger, release bash-@value{BASHRELEASE} ++Copyright 2002,2003 Rocky Bernstein ++This is free software, covered by the GNU General Public License, and you are ++welcome to change it and/or distribute copies of it under certain conditions. ++ ++(fact.sh:9): ++9: echo fact 0 is: `fact 0` ++bashdb<0> @b{-} ++ 1: #!/usr/local/bin/bash ++ 2: fact() @{ ++ 3: ((n==0)) && echo 1 && return ++ 4: ((nm1=n-1)) ++ 5: ((result=n*`fact $nm1`)) ++ 6: return $result ++ 7: @} ++ 8: ++ 9:==> echo fact 0 is: `fact 0` ++bashdb<1> @b{l} ++ 10: echo fact 3 is: $(fact 3) ++@end smallexample ++ ++@noindent ++ ++The command invocation uses the option ``-L .'' Here we assume that ++the @command{bashdb} script and the debugger files are in the same ++location. If you are running from the source code, this will be the ++case. However if bashdb has been installed this probably won't be true ++and here you probably don't need to use ``-L .'' Instead you would ++type simply @code{bashdb fact.sh}. ++ ++The first debugger command we gave @kbd{-}, we listed a window of ++lines @emph{before} where we were executing. Because the window, 10 ++lines, is larger than the number of lines to the top of the file we ++printed only 9 lines here. The next command list starting from the ++current line and again wants to print 10 lines but because there are ++only one remaining line, that is what is printed. ++ ++@smallexample ++bashdb<2> @b{step} ++(fact.sh:9): ++9: echo fact 0 is: `fact 0` ++bashdb<3> @b{@key{RET}} ++2: fact() @{ ++bashdb<4> @b{@key{RET}} ++3: ((n==0)) && echo 1 && return ++bashdb<5> @b{print $n} ++ ++@end smallexample ++ ++Ooops... The variable @kbd{n} isn't initialized. ++ ++The first @kbd{step} command steps the script one instruction. It may ++seem odd that the line printed is exactly the same one as before. What ++has happened though is that we've ``stepped'' into the subshell needed ++to run @kbd{`fact 0`}; we haven't however started running anything ++inside that subshell yet though---the function call. ++ ++The first @kbd{step} command steps the script one instruction; it ++didn't advance the line number at all. That is because we were ++stopping before the command substitution or backtick is to take ++place. The second command we entered was just hitting the return key; ++bashdb remembers that you entered @code{step} previously, so it runs ++the step rather than @kbd{next}, the other alternative when you hit ++@key{RET}. Step one more instruction and we are just before running ++the first statement of the function. ++ ++Next, we print the value of the variable @kbd{n}. Notice we need to add ++a preceding dollar simple to get the substitution or value of n. As we ++will see later, if the @kbd{pe} command were used this would not be ++necessary. ++ ++We now modify the file to add an assignment to local variable @kbd{n} and ++restart. ++ ++@smallexample ++bashdb<6> @b{restart} ++Restarting with: /usr/local/bin/bashdb -L . fact.sh ++(fact.sh:10): ++10: echo fact 0 is: `fact 0` ++bashdb<0> @b{list 1} ++ 1: #!/usr/local/bin/bash ++ 2: fact() @{ ++ 3: local -i n=$@{1:0@} ++ 4: ((n==0)) && echo 1 && return ++ 5: ((nm1=n-1)) ++ 6: ((result=n*`fact $nm1`)) ++ 7: return $result ++ 8: @} ++ 9: ++ 10:==> echo fact 0 is: `fact 0` ++bashdb<1> @b{s 3} ++(fact.sh:3): ++3: local -i n=$@{1:0@} ++bashdb<2> @b{step} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++bashdb<3> @b{print $n} ++print $n ++0 ++@end smallexample ++ ++@noindent ++This time we use the @code{list} debugger command to list the lines in ++the file. From before we know it takes three @code{steps} commands ++before we get into the fact() function, so we add a count onto the ++@code{step} command. Notice we abbreviate @code{step} with @code{s}; ++we could have done likewise and abbreviated @code{list} with @code{l}. ++ ++@smallexample ++bashdb<4> @b{@key{RET}} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++echo 1 ++bashdb<5> @b{@key{RET}} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++return ++@end smallexample ++ ++@noindent ++Again we just use @key{RET} to repeat the last @code{step} ++commands. And again the fact that we are staying on the same line 4 ++means that the next condition in the line is about to be ++executed. Notice that we see the command (@code{echo 1} or ++@code{return}) listed when we stay on the same line which has multiple ++stopping points in it. Given the information above, we know that the ++value echo'ed on return will be 1. ++ ++@smallexample ++bashdb<6> @b{@key{RET}} ++fact 0 is: 1 ++(fact.sh:12): ++12: echo fact 3 is: $(fact 3) ++bashdb<7> @b{break 5} ++Breakpoint 1 set in file fact.sh, line 5. ++bashdb<8> @b{continue} ++@end smallexample ++ ++@noindent ++We saw that we could step with a count into the function ++fact(). However above took another approach: we set a stopping point or ++``breakpoint'' at line 5 to get us a little ways into the fact() ++subroutine. Just before line 5 is to executed, we will get back into ++the debugger. The @code{continue} command just resumes execution until ++the next stopping point which has been set up in some way. ++ ++@smallexample ++(fact.sh:5): ++5: ((nm1=n-1)) ++Breakpoint 1 hit(1 times). ++bashdb<8> @b{x n-1} ++2 ++bashdb<9> @b{s} ++(fact.sh:5): ++6: ((result=n*`fact $nm1`)) ++bashdb<10> @b{c} ++fact.sh: line 6: ((: result=n*: syntax error: operand expected (error token is "*") ++bashdb<7> @b{R} ++Restarting with: bash --debugger fact.sh ++11: echo fact 0 is: `fact 0` ++bashdb<0> @b{l fact} ++ 2: fact () ++ 3: @{ ++ 4: local -i n=$@{1:0@}; ++ 5: (( "n==0" )) && echo 1 && return; ++ 6: (( nm1=n-1 )); ++ 7: ((fact_nm1=`fact $nm1`)) ++ 8: (( "result=n*fact_nm1" )); ++ 9: echo $result ++10: @} ++@end smallexample ++ ++@noindent ++In addition to listing by line numbers, we can also list giving a ++function name. Below, instead of setting a breakpoint at line 5 and ++running ``continue'' as we did above, we try something slightly shorter ++and slightly different. We give the line number on the ``continue'' ++statement. This is a little different in that a one-time break is made ++on line 5. Once that statement is reached the breakpoint is removed. ++ ++@smallexample ++bashdb<1> @b{continue 5} ++One-time breakpoint 1 set in file fact.sh, line 5. ++fact 0 is: 1 ++(fact.sh:5): ++5: ((nm1=n-1)) ++bashdb<2> @b{s} ++6: ((fact_nm1=`fact $nm1`)) ++bashdb<2> @b{s} ++2: fact() @{ ++bashdb<3> @b{T} ++->0 in file `fact.sh' at line 2 ++##1 fact("3") called from file `fact.sh' at line 12 ++##2 source("fact.sh") called from file `/usr/local/bin/bashdb' at line 154 ++##3 main("fact.sh") called from file `/usr/local/bin/bashdb' at line 0 ++bashdb<4> @b{c} ++fact 3 is: 6 ++Debugged program terminated normally. Use q to quit or R to restart. ++@end smallexample ++ ++@noindent ++When we stop at line 5 above, we have already run fact(0) and output ++the correct results. The output from the program ``fact 0 is: 1'' is ++intermixed with the debugger output. The @code{T} command above ++requests call stack output and this confirms that we are not in the ++fact(0) call but in the fact(3) call. There are 4 lines listed in the ++stack trace even though there is just one call from the main ++program. The top line of the trace doesn't really represent a call, ++it's just where we currently are in the program. That last line is an ++artifact of invoking bash from the bashdb script rather than running ++@code{bash --debugger}. ++ ++The last message in the output above @samp{Debugged program exited ++normally.} is from @value{DBG}; it indicates script has finished ++executing. We can end our @value{DDGP} session with the @value{GDBP} ++@code{quit} command. ++ ++Above we did our debugging session on the command line. If you are a ++GNU Emacs user, you can do your debugging inside that. Also there is ++a(nother) GUI interface called DDD that supports @value{DBG}. ++ ++@node Free Software ++@unnumberedsec Free software ++ ++@DBG is @dfn{free software}, protected by the @sc{gnu} ++General Public License ++(GPL). The GPL gives you the freedom to copy or adapt a licensed ++program---but every person getting a copy also gets with it the ++freedom to modify that copy (which means that they must get access to ++the source code), and the freedom to distribute further copies. ++Typical software companies use copyrights to limit your freedoms; the ++Free Software Foundation uses the GPL to preserve these freedoms. ++ ++Fundamentally, the General Public License is a license which says that ++you have these freedoms and that you cannot take these freedoms away ++from anyone else. ++ ++@unnumberedsec Free Software Needs Free Documentation ++ ++The biggest deficiency in the free software community today is not in ++the software---it is the lack of good free documentation that we can ++include with the free software. Many of our most important ++programs do not come with free reference manuals and free introductory ++texts. Documentation is an essential part of any software package; ++when an important free software package does not come with a free ++manual and a free tutorial, that is a major gap. We have many such ++gaps today. ++ ++Consider Perl, for instance. The tutorial manuals that people ++normally use are non-free. How did this come about? Because the ++authors of those manuals published them with restrictive terms---no ++copying, no modification, source files not available---which exclude ++them from the free software world. ++ ++That wasn't the first time this sort of thing happened, and it was far ++from the last. Many times we have heard a GNU user eagerly describe a ++manual that he is writing, his intended contribution to the community, ++only to learn that he had ruined everything by signing a publication ++contract to make it non-free. ++ ++Free documentation, like free software, is a matter of freedom, not ++price. The problem with the non-free manual is not that publishers ++charge a price for printed copies---that in itself is fine. (The Free ++Software Foundation sells printed copies of manuals, too.) The ++problem is the restrictions on the use of the manual. Free manuals ++are available in source code form, and give you permission to copy and ++modify. Non-free manuals do not allow this. ++ ++The criteria of freedom for a free manual are roughly the same as for ++free software. Redistribution (including the normal kinds of ++commercial redistribution) must be permitted, so that the manual can ++accompany every copy of the program, both on-line and on paper. ++ ++Permission for modification of the technical content is crucial too. ++When people modify the software, adding or changing features, if they ++are conscientious they will change the manual too---so they can ++provide accurate and clear documentation for the modified program. A ++manual that leaves you no choice but to write a new manual to document ++a changed version of the program is not really available to our ++community. ++ ++Some kinds of limits on the way modification is handled are ++acceptable. For example, requirements to preserve the original ++author's copyright notice, the distribution terms, or the list of ++authors, are ok. It is also no problem to require modified versions ++to include notice that they were modified. Even entire sections that ++may not be deleted or changed are acceptable, as long as they deal ++with nontechnical topics (like this one). These kinds of restrictions ++are acceptable because they don't obstruct the community's normal use ++of the manual. ++ ++However, it must be possible to modify all the @emph{technical} ++content of the manual, and then distribute the result in all the usual ++media, through all the usual channels. Otherwise, the restrictions ++obstruct the use of the manual, it is not free, and we need another ++manual to replace it. ++ ++Please spread the word about this issue. Our community continues to ++lose manuals to proprietary publishing. If we spread the word that ++free software needs free reference manuals and free tutorials, perhaps ++the next person who wants to contribute by writing documentation will ++realize, before it is too late, that only free manuals contribute to ++the free software community. ++ ++If you are writing documentation, please insist on publishing it under ++the GNU Free Documentation License or another free documentation ++license. Remember that this decision requires your approval---you ++don't have to let the publisher decide. Some commercial publishers ++will use a free license if you insist, but they will not propose the ++option; it is up to you to raise the issue and say firmly that this is ++what you want. If the publisher you are dealing with refuses, please ++try other publishers. If you're not sure whether a proposed license ++is free, write to @email{licensing@@gnu.org}. ++ ++You can encourage commercial publishers to sell more free, copylefted ++manuals and tutorials by buying them, and particularly by buying ++copies from the publishers that paid for their writing or for major ++improvements. Meanwhile, try to avoid buying non-free documentation ++at all. Check the distribution terms of a manual before you buy it, ++and insist that whoever seeks your business must respect your freedom. ++Check the history of the book, and try to reward the publishers that ++have paid or pay the authors to work on it. ++ ++The Free Software Foundation maintains a list of free documentation ++published by other publishers, at ++@url{http://www.fsf.org/doc/other-free-books.html}. ++ ++@node Invocation ++@chapter Getting in and out ++ ++This chapter discusses how to start @value{DBG}, and how to get out of it. ++The essentials are: ++@itemize @bullet ++@item ++type @samp{bash --debugger @emph{script-name}} or @samp{bashdb ++@emph{script-name}} to start @value{DBG}. ++@item ++type @kbd{quit} or @kbd{C-d} inside the debugger to exit. ++@end itemize ++ ++But also note there are two front-ends available as well. One can also ++enter the debugger inside emacs via the command @code{M-x bashdb} ++after loading Emacs' Grand Unified Debugger, @code{gud}. See ++@ref{Emacs,,Using the BASH debugger from @sc{gnu} Emacs}. And there is ++support in a @value{DDD} for bash. ++ ++@menu ++* Starting the BASH debugger:: How to enter the BASH debugger ++* Quitting the BASH debugger:: How to leave the BASH debugger ++@end menu ++ ++@node Starting the BASH debugger ++@section Starting the BASH debugger ++ ++@emph{Note: it is important to use a debugger-enabled bash. You wil ++get an error message if the debugger is run under a version of BASH ++that does not have debugging support.} ++ ++As mentioned above, one can enter @DBG via Emacs or ++DDD. However you don't have to use either of these. And these still ++need a way on their own to get things started. ++ ++There are in fact two @emph{other} ways to start @value{DBG}. The ++first way is to pass the @samp{--debugger} option to bash with the ++name of your script the scripts arguments following that, or with a ++command string (@code{-c}). ++ ++@example ++bash --debugger @var{script} @var{script-arguments...} ++bash --debugger -c @var{command-string}... ++@end example ++ ++This calls a debugger initialization script. It works much like a ++@BASH login profile which may set variables and define ++functions. But this shell profile is customized for debugging and as ++such arranges for itself to get called before each statement is ++executed. Although there are some problems at present in I/O ++redirection that the method described next doesn't have, it is ++expected that over time more features will be enabled in bash when the ++@samp{--debugger} option is in effect. By default, both debugging in ++Emacs via GUD (@ref{Emacs,,Using the BASH debugger under Emacs}) and ++debugging via @value{DDD} work via this method. ++ ++The form @samp{bash --debugger -c ...} can be used to get into the ++debugger without having to give a script name to debug. Sometimes you ++may want to do this just to see how the debugger works: try some ++debugger commands or maybe get online help. If you run @code{ddd ++--bash} without giving a script name, it in fact uses this form. ++ ++In order for the @samp{--debugger} option to work however, you must ++have the debugger scripts installed in a place where @DBG can ++find them. For this reason, in developing @value{DBG}, I use a second ++method more often; it doesn't require the bash debugger to be ++installed. This method uses another script called @code{bashdb} which ++after taking its own options takes the name of the script to debugged ++and the arguments to pass to that script. Using this method, one ++would start the debugger like this: ++ ++@example ++bash @var{path-to-bashdb}/bashdb @var{bashdb-options} @var{script} @var{script-arguments...} ++@end example ++ ++As with the first method, @code{bash} should be a debugger-enabled ++bash. If @code{bashdb} has the path to bash in it at the top (e.g. via ++@code{#!}), and @code{bashdb} can be found in your program-search ++path, then this might be equivalent to the above: ++ ++@example ++bashdb @var{bashdb-options} @var{script} @var{script-arguments...} ++@end example ++ ++There are two or three disadvantages however of running a debugger ++this way. First @code{$0} will have the value @code{bashdb} rather ++than the script you are trying to run. For some scripts this may ++change the behavior of the debugged script. Second a traceback will ++contain additional lines showing the ``source''-ing of the debugged ++script from @code{bashdb}. And third, although this way works better ++than the first method, over time this way may come into disuse. ++ ++An option that you'll probably need to use if bashdb isn't installed ++but run out of the source code directory is @samp{-L} which specifies ++the directory that contains the debugger script files. ++ ++You can further control how bashdb starts up by using command-line ++options. bashdb itself can remind you of the options available. ++ ++@noindent ++Type ++ ++@example ++bashdb -h ++@end example ++ ++@noindent ++to display all available options and briefly describe their use. ++ ++When the bash debugger is invoked either by the @code{bashdb} ++front-end script or @code{bash --debugging}, the first argument that ++does not have an associated option flag for @code{bashdb} or ++@code{bash} (as the case may be) is used as the name a the script file ++to be debugged, and any following options get passed the debugged ++script. ++ ++Options for the @code{bashdb} front-end are shown in the ++following list. ++ ++@menu ++* Options for the bashdb script:: Options you can pass in starting bashdb ++@end menu ++ ++@node Options for the bashdb script ++@subsection Command-line options for @code{bashdb} script ++ ++You can run @DBG in various alternative modes---for example, in ++batch mode or quiet mode. ++ ++@table @code ++@item -h ++@cindex @code{-h} ++This option causes @value{DBG} to print some basic help and exit. ++ ++@item -V ++@cindex @code{-V} ++This option causes @DBG to print its version number and ++no-warranty blurb, and exit. ++ ++@item -n ++@cindex @code{-n} ++Do not execute commands found in any initialization files. Normally, ++@BASH executes the commands in these files after all the command ++options and arguments have been processed. @xref{Command Files,,Command ++files}. ++ ++@item -q ++@cindex @code{-q} ++``Quiet''. Do not print the introductory and copyright messages. These ++messages are also suppressed in batch mode. ++ ++@item -t ++@cindex @code{-t} ++``Terminal output''. Set the file or terminal that you want debugger command ++output to go to. Note that the debugger output is independent of the ++debugged script output. ++ ++@item -L @var{directory} ++@cindex @code{-L} ++Set directory where debugger files reside to @var{directory}. The ++default location is @code{../lib/bashdb} relative to the place that ++the bashdb script is located. For example if bashdb is located in ++@code{/usr/local/bin/bashdb}, the default library location will be ++@code{/usr/local/lib/bashdb} which may or may not exist. If it doesn't ++you'll get an error when you run bashdb. Only if the default location ++is incorrect, should you need to use the @code{-L} option. ++ ++@item -T @var{directory} ++@cindex @code{-T} ++Set directory to use for writing temporary files. ++ ++@end table ++ ++@node Quitting the BASH debugger ++@section Quitting the BASH debugger ++ ++@cindex interrupt ++An interrupt (often @kbd{C-c}) does not exit from @value{DBG}, but ++rather terminates the action of any @DBG command that is in ++progress and returns to @value{DBG} command level. Inside a debugger ++command interpreter, use @code{quit} command (@pxref{Quit, ,Quitting ++the BASH debugger}). ++ ++@node Running ++@chapter Script Setup inside the BASH Debugger ++ ++@menu ++* Starting:: Starting your script ++* Command Files:: Command files ++* Arguments:: Your script's arguments ++* Input/Output:: Your script's input and output ++* Script/Debugger Interaction:: Keeping out of each other's harm ++@end menu ++ ++@need 2000 ++@node Starting ++@section Starting your script ++@cindex starting ++@cindex running ++ ++@table @code ++@kindex restart ++@kindex run @r{(@code{restart})} ++@kindex R @r{(@code{restart})} ++@item restart @ovar{options to debugged script} ++@itemx run ++@itemx R ++Use the @code{restart} command to restart your script under ++@value{DBG}. Without any arguments, the script name and parameters ++from the last invocation are used. @value{dBGP} tries to maintain the ++settings, watchpoints, breakpoints, actions and so on. Internally it ++uses line numbers and filenames to record he position of interesting ++places in your porgram; so if your program changes some or all of ++these numbers may be off. Environment variable ++@code{BASHDB_RESTART_FILE} is and a temporary file are used to signal ++a restart, so you shouldn't uset @code{BASHDB_RESTART_FILE} (or any ++environment variable starting with @code{BASHDB_}. ++ ++@end table ++ ++@node Command Files ++@section Command files ++ ++@cindex command files ++A command file for @DBG is a file of lines that are @DBG ++commands. Comments (lines starting with @kbd{#}) may also be included. ++An empty line in a command file does nothing; it does not mean to repeat ++the last command, as it would from the terminal. ++ ++@cindex init file ++@cindex @file{.bashdbinit} ++@cindex @file{bashdb.ini} ++When you start @value{DBG}, it automatically executes commands from its ++@dfn{init files}, normally called @file{.bashdbinit}@footnote{The DJGPP ++port of @DBG uses the name @file{bashdb.ini} instead, due to the ++limitations of file names imposed by DOS filesystems.}. ++During startup, @DBG does the following: ++ ++@enumerate ++@item ++Reads the init file (if any) in your home directory@footnote{On ++DOS/Windows systems, the home directory is the one pointed to by the ++@code{HOME} environment variable.}. ++ ++@item ++Processes command line options and operands. ++ ++@item ++Reads the init file (if any) in the current working directory. ++ ++@item ++Reads command files specified by the @samp{-x} option. ++@end enumerate ++ ++The init file in your home directory can set options (such as @samp{set ++complaints}) that affect subsequent processing of command line options ++and operands. Init files are not executed if you use the @samp{-x} ++option (@pxref{Options for the bashdb script, ,bashdb script options}). ++ ++@cindex init file name ++On some configurations of @value{DBG}, the init file is known by a ++different name (these are typically environments where a specialized ++form of @DBG may need to coexist with other forms, hence a ++different name for the specialized version's init file). These are the ++environments with special init file names: ++ ++You can also request the execution of a command file with the ++@code{source} command: ++ ++@table @code ++@kindex source ++@item source @var{filename} ++Execute the command file @var{filename}. ++@end table ++ ++The lines in a command file are executed sequentially. They are not ++printed as they are executed. If there is an error, execution ++proceeds to the next command in the file. ++ ++@node Arguments ++@section Your script's arguments ++ ++@cindex arguments (to your script) ++The arguments to your script can be specified by the arguments of the ++@code{restart} command. ++They are passed to a shell, which expands wildcard characters and ++performs redirection of I/O, and thence to your script. ++ ++@code{restart} with no arguments uses the same arguments used by the previous ++@code{restart}, or those set by the @code{set args} command.. ++ ++@table @code ++@kindex set args ++@item set args ++Specify the arguments to be used the next time your program is run. If ++@code{set args} has no arguments, @code{restart} executes your program ++with no arguments. Once you have run your program with arguments, ++using @code{set args} before the next @code{restart} is the only way to run ++it again without arguments. ++ ++@kindex show args ++@item show args ++Show the arguments to give your program when it is started. ++@end table ++ ++@node Input/Output ++@section Your script's input and output ++ ++@cindex redirection ++@cindex I/O ++@cindex terminal ++By default, the script you run under the BASH debugger does input and ++output to the same terminal that BASH uses. Before running the script ++to be debugged, the debugger records the tty that was in effect. All ++of its output is then written to that. However you can change this ++when using the @samp{bashdb} script using the @samp{-t} option. ++ ++ ++@table @code ++@kindex info terminal ++@item info terminal ++Displays information recorded by @DBG about the terminal modes your ++program is using. ++@end table ++ ++@kindex tty ++@cindex controlling terminal ++Another way to specify where your script should do input and output is ++with the @code{tty} command. This command accepts a file name as ++argument, and causes this file to be the default for future @code{restart} ++commands. It also resets the controlling terminal for the child ++process, for future @code{restart} commands. For example, ++ ++@example ++tty /dev/ttyb ++@end example ++ ++@noindent ++directs that processes started with subsequent @code{restart} commands ++default to do input and output on the terminal @file{/dev/ttyb} and have ++that as their controlling terminal. ++ ++An explicit redirection in @code{restart} overrides the @code{tty} command's ++effect on the input/output device, but not its effect on the controlling ++terminal. ++ ++When you use the @code{tty} command or redirect input in the @code{restart} ++command, only the input @emph{for your script} is affected. The input ++for @DBG still comes from your terminal. ++ ++@node Script/Debugger Interaction ++@section Script/Debugger Interaction ++ ++@value{dBGP} and your program live in the same variable space so ++to speak. @BASH does not have a notion of module scoping or lexical ++hiding (yet) and this then imposes some additional care and awareness. ++ ++Most of the variables and functions used inside @DBG start ++@code{_Dbg_}, so please don't use variables or functions with these ++names in your program. ++ ++@emph{Note: there are some other variables that begin with just an ++underscore (@code{_}); over time these will be phased out. But until ++then, avoid those or consult what is used by the debugger. Run ++@samp{bashdb --debugger -c 'declare -p'} to list all the variables in ++use including those used by the debugger.} ++ ++A number of @DBG environment variables are also reserved for ++use; these start with @code{BASHDB_}. For example: ++@env{BASHDB_INPUT}, @env{BASHDB_LEVEL} and, ++@env{BASHDB_QUIT_ON_QUIT} (@pxref{Debug, ,Debug}), ++@env{BASHDB_RESTART_FILE} (@pxref{Starting, ,Starting}), to name a ++few. Finally, there are some @BASH environment dynamic ++variables and these start with @env{BASH_}. For example ++@env{BASH_SUBSHELL} (@pxref{Debug, ,Debug}), @env{BASH_COMMAND} ++(@pxref{Command Display, ,Command Display}), @env{BASH_LINENO}, and ++@env{BASH_SOURCE} to name a few. ++ ++In order to do its work @value{dBGP} sets up a @code{DEBUG} ++trap. Consequently a script shouldn't reset this or the debugger will ++lose control. @value{dBGP} also sets up an @code{EXIT} handler so that ++it can gain control after the script finishes. Another signal ++intercepted is the an interrupt or @code{INT} signal. For more ++information about signal handling, @pxref{Signals, ,Signals} ++ ++@node Debugger Command Reference ++@chapter BASH Debugger Command Reference ++ ++You can abbreviate the long name of @DBG command to the first ++few letters of the command name, if that abbreviation is unambiguous; ++and you can repeat the @code{next} o r@code{step} commands by typing ++just @key{RET}. Some commands which require a parameter, such as ++@code{print} remember the argument that was given to them. ++ ++@menu ++* Command Syntax:: How to give commands to the BASH debugger ++* Help:: How to ask for help (help) ++* Quit:: Leaving the debugger (quit) ++* Information:: Status and Debugger settings (info, show) ++* Stopping:: Stopping and continuing (break, watch, step, cont...) ++* Stack:: Examining the stack (where, up, down, frame) ++* List:: Printing source files (list) ++* Search:: Searching source files (/pat/ ?pat?) ++* Data:: Examining data (print, examine, info variables) ++* Evaluation/Execution:: Arbitrary execution (eval, shell) ++* Auto Display:: Executing expressions on stop (display, undisplay) ++* Controlling bashdb:: Controlling bashdb (set, history) ++@end menu ++ ++@node Command Syntax ++@section Command syntax ++ ++A BASH debugger command is a single line of input. There is no limit on ++how long it can be. It starts with a command name, which is followed by ++arguments whose meaning depends on the command name. For example, the ++command @code{step} accepts an argument which is the number of times to ++step, as in @samp{step 5}. You can also use the @code{step} command ++with no arguments. Some commands do not allow any arguments. ++ ++@cindex repeating next/step commands ++@kindex RET @r{(repeat last command)} ++A blank line as input to @DBG (typing just @key{RET}) means to ++repeat the previous next or step command. ++ ++@kindex # @r{(a comment)} ++@cindex comment ++Any text from a @kbd{#} to the end of the line is a comment; it does ++nothing. This is useful mainly in command files (@pxref{Command ++Files,,Command files}). ++ ++@node Help ++@section Getting help (@samp{help}) ++@cindex online documentation ++ ++Once inside the BASH debugger, you can always ask it for information on its ++commands, using the command @code{help}. ++ ++@table @code ++@kindex h @r{(@code{help})} ++@item help ++@itemx h ++You can use @code{help} (abbreviated @code{h}) with no arguments to ++display a short list of named classes of commands: ++@end table ++ ++@flushleft ++@smallexample ++bashdb<0> @b{help} ++bashdb commands: ++List/search source lines: Control script execution: ++------------------------- ------------------------- ++ l [start|.] [cnt] List cnt lines T [n] Stack trace ++ from line start s [n] Single step [n times] ++ l sub List source code fn n [n] Next, steps over subs ++ - or . List previous/current line / Repeat last n or s ++ w [line] List around line c [linespec] Continue [to linespec] ++ f filename View source in file L List all breakpoints ++ /pat/ Search forward for pat b linespec Set breakpoint ++ ?pat? Search backward for pat del [n].. or D Delete a/all breaks ++ by entry number ++Debugger controls: skip skip execution of cmd ++------------------------- cl linespec Delete breakpoints by ++ H [num] Show last num commands line spec ++ q [exp] or ^D Quit returning exp R [args] Attempt a restart ++ info [cmd] Get info on cmd. u [n] Go up stack by n or 1. ++ !n or hi n Run debugger history n do [n] Go down stack by n or 1. ++ h or ? [cmd] Get help on command W [var] Add watchpoint. If no ++ info [cmd] Get info on cmd no expr, delete all ++ show [cmd] Show settings We [expr] Add Watchpoint arith ++ expr ++ so file read in dbg commands t Toggle trace ++ en/di n enable/disable brkpt, ++ set x y set a debugger variable watchpoint, or display ++ e bash-cmd evaluate a bash command tb linespec Add one-time break ++ disp expr add a display expr a linespec cmd eval "cmd" at linespec ++ M Show module versions A delete all actions ++ x expr evaluate expression ret jump out of fn or source ++ (via declare, let, eval) finish execute until return ++ deb debug into another cond n exp set breakpoint condition ++ shell script ++ !! cmd [args] execute shell command "cmd" with "args" ++ ++Data Examination: also see e, t, x ++------------------------- ++ p variable Print variable ++ V [[!]pat] List variable(s) matching or not (!) matching pattern pat ++ S [[!]pat] List subroutine names [not] matching pattern pat ++ ++Readline command line editing (emacs/vi mode) is available ++For more help, type h or consult online-documentation. ++@end smallexample ++@end flushleft ++@c the above line break eliminates huge line overfull... ++ ++@table @code ++@item help @var{command} ++With a command name as @code{help} argument, the BASH debugger displays ++short information on how to use that command. ++ ++@example ++bashdb<0> @b{help list} ++l linespec List window lines starting at linespec. ++l min incr List incr lines starting at 'min' linespec. ++l List next window of lines. ++l . Same as above. ++ Long command name: list. ++@end example ++ ++In addition to @code{help}, you can use the debugger command ++@code{info} to inquire about the state of your script, or the state of ++@DBG itself. The listings under @code{info} in the Index ++point to all the sub-commands. @xref{Command Index}. ++@end table ++ ++@c @group ++@table @code ++@kindex info ++@kindex i @r{(@code{info})} ++@item info ++This command (abbreviated @code{i}) is for describing the state of ++your program. For example, you can list the arguments given to your ++script with @code{info args}, or list the breakpoints you have set ++with @code{info breakpoints}. You can get a complete list of the ++@code{info} sub-commands with @w{@code{help info}}. ++ ++@example ++bashdb<0> @b{info} ++Info subcommands are: args breakpoints display files source sources stack terminal variables watchpoints ++bashdb<1> @b{info source} ++Current script file is parm.sh ++Contains 34 lines. ++@end example ++@end table ++ ++ ++@node Quit ++@section Quitting the BASH debugger (@samp{quit}) ++ ++@table @code ++@kindex quit @r{[}@var{expression} @ovar{subshell-levels}@r{]} ++@kindex q @r{(@code{quit})} ++@item quit @ovar{expression} ++@item quit @r{[}@var{expression} @ovar{subshell-levels}@r{]} ++@itemx q ++ ++To exit @value{DBG}, use the @code{quit} command (abbreviated ++@code{q}), or type an end-of-file character (usually @kbd{C-d}). If ++you do not supply @var{expression}, @DBG will try to terminate ++normally or with exit code 0. Otherwise it will terminate using the ++result of @var{expression} as the exit code. ++ ++A simple @code{quit} tries to terminate all nested subshells that may ++be in effect. If you are nested a subshell, this is normally ++indicated in a debugger prompt by the number of parentheses that the ++history number is inside --- no parenthesis means there is no subshell ++in effect. The dynamic variable @env{BASH_SUBSHELL} also contains the ++number of subshells in effect. ++ ++If you want only to terminate some number of subshells but not all of ++them, you can give a count of the number of subshells to leave after ++the return-code expression. To leave just one level of subshell ++@code{return} does almost the same thing. (See @pxref{Returning, ++,Returning}) There is a subtle difference between the two though: ++@code{return} will leave you at the beginning of the next statement ++while @code{quit} may leave you at the place the subshell was invoked ++which may be in the middle of another command such as an assingment ++statement or condition test. ++ ++If the environment variable @code{BASHDB_QUIT_ON_QUIT} is set, when the ++program terminates, the debugger will also terminate too. This may be ++useful if you are debugging a script which calls another script and ++you want this inner script just to return to the outer script. ++@end table ++ ++@node Stopping ++@section Stopping and Resuming Execution (@samp{break}, @samp{watch}, @samp{step}, @samp{cont}) ++ ++One important use of a debugger is to stop your program before it ++terminates so that, if your script might run into trouble, you can ++investigate and find out why. However should your script accidently ++continue to termination, @DBG has arranged for it not to leave the ++debugger without your explicit instruction. That way, you can restart ++the program using the same command arguments. ++ ++Inside @value{DBG}, your script may stop for any of several reasons, ++such as a signal, a breakpoint, or reaching a new line after a ++debugger command such as @code{step}. You may then examine and ++change variables, set new breakpoints or remove old ones, and then ++continue execution. ++ ++@menu ++* Breakpoints:: Breakpoints, watchpoints (break, watch, clear) ++* Resuming Execution:: Resuming execution (continue, step, next, skip, debug) ++* Signals:: Signals ++@end menu ++ ++@node Breakpoints ++@subsection Breakpoints, watchpoints ++ ++@cindex breakpoints ++A @dfn{breakpoint} makes your script stop whenever a certain point in ++the program is reached. For each breakpoint, you can add conditions to ++control in finer detail whether your script stops. ++ ++You can set breakpoints with the @code{break} command and its variants ++(@pxref{Set Breaks, ,Setting breakpoints}), to specify the place where ++your script should stop by line number. or function name in the ++debugged script. ++ ++@cindex watchpoints ++@cindex breakpoint on variable modification ++A @dfn{watchpoint} is a special breakpoint that stops your script when ++the value of an expression changes. There is a different command to ++set watchpoints (@pxref{Set Watchpoints, ,Setting watchpoints}). ++ ++But aside from that, you can manage a watchpoint like any other ++breakpoint: you delete enable, and disable both breakpoints and ++watchpoints using the same commands. ++ ++You can arrange to have values from your program displayed automatically ++whenever @value{BASH} stops at a breakpoint. @xref{Auto Display,, ++Automatic display}. ++ ++@cindex breakpoint numbers ++@cindex numbers for breakpoints ++@DBG assigns a number to each breakpoint and watchpoint when ++you create it; these numbers are successive integers starting with ++one. In many of the commands for controlling various features of ++breakpoints you use the breakpoint number to say which breakpoint you ++want to change. Each breakpoint may be @dfn{enabled} or ++@dfn{disabled}; if disabled, it has no effect on your script until you ++enable it again. ++ ++@cindex watchpoints numbers ++@cindex numbers for watchpoints ++Watchpoint numbers however are distiguished from breakpoint numbers by ++virtue of their being suffixed with the either an upper- or lower-case ++`W'. For example, to enable breakpoint entry 0 along with watchpoint ++entry 1 you would write @samp{enable 1 2w}, the ``2w'' refers to the ++watchpoint; ``2W'' would work just as well. ++ ++@ifset FINISHED ++@cindex breakpoint ranges ++@cindex ranges of breakpoints ++Some @DBG commands accept a range of breakpoints on which to ++operate. A breakpoint range is either a single breakpoint number, like ++@samp{5}, or two such numbers, in increasing order, separated by a ++hyphen, like @samp{5-7}. When a breakpoint range is given to a command, ++all breakpoint in that range are operated on. ++@end ifset ++ ++@menu ++* Set Breaks:: Setting breakpoints (break, tbreak) ++* Set Watchpoints:: Setting watchpoints (watch, watche) ++* Delete Breaks:: Deleting breakpoints (delete, clear) ++* Disabling:: Disabling breakpoints (disable, enable) ++* Conditions:: Break conditions (condition) ++@end menu ++ ++@node Set Breaks ++@subsubsection Setting breakpoints (@samp{break} @samp{tbreak}) ++ ++@kindex break ++@kindex b @r{(@code{break})} ++@cindex latest breakpoint ++Breakpoints are set with the @code{break} command (abbreviated ++@code{b}). ++ ++@table @code ++@item break @var{function} ++Set a breakpoint at entry to function @var{function}. ++ ++@item break @var{linenum} ++Set a breakpoint at line @var{linenum} in the current source file. ++The current source file is the last file whose source text was printed. ++The breakpoint will stop your script just before it executes any of the ++code on that line. ++ ++@item break @var{filename}:@var{linenum} ++Set a breakpoint at line @var{linenum} in source file @var{filename}; ++@var{filename} has to be one of the files previously read in and has ++to be specified exactly as the name used when read in. For a list of ++read-in files, use the @samp{info files} command. ++ ++@ifset FINISHED ++@item break ++When called without any arguments, @code{break} sets a breakpoint at ++the next instruction to be executed in the selected stack frame ++(@pxref{Stack, ,Examining the Stack}). In any selected frame but the ++innermost, this makes your script stop as soon as control returns to ++that frame. If you use @code{break} without an argument in the ++innermost frame, @DBG stops the next time it reaches the ++current location; this may be useful inside loops. ++@end ifset ++ ++@item break @dots{} if @var{cond} ++Set a breakpoint with condition @var{cond}; evaluate the expression ++@var{cond} each time the breakpoint is reached, and stop only if the ++value is nonzero---that is, if @var{cond} evaluates as true. The ++expression is evaluated via the @code{let} builtin funtion. ++@samp{@dots{}} stands for one of the possible arguments described ++above (or no argument) specifying where to break. The word ``if'' is ++often optional and is necessary only @samp{@dots{}} is ++omitted. @xref{Conditions, ,Break conditions}, for more information on ++breakpoint conditions. ++ ++Examples: ++@example ++bashdb<0> @b{break fn1} ++Breakpoint 1 set in file parm.sh, line 3. ++bashdb<1> @b{break 28} ++Breakpoint 2 set in file parm.sh, line 28. ++bashdb<2> @b{break parm.sh:29} ++Breakpoint 3 set in file parm.sh, line 29. ++bashdb<3> @b{break 28 if x==5} ++Breakpoint 4 set in file parm.sh, line 28. ++@end example ++ ++@kindex tbreak ++@item tbreak @var{args} ++Set a breakpoint enabled only for one stop. @var{args} are the ++same as for the @code{break} command, and the breakpoint is set in the same ++way, but the breakpoint is automatically deleted after the first time your ++program stops there. @xref{Disabling, ,Disabling breakpoints}. ++ ++@kindex info breakpoints ++@cindex @code{$_} and @code{info breakpoints} ++@item info breakpoints @ovar{n} ++@itemx info break @ovar{n} ++@itemx info watchpoints @ovar{n} ++Print a table of all breakpoints, watchpoints set and not deleted, ++with the following columns for each breakpoint: ++ ++@table @emph ++@item Breakpoint Numbers (@samp{Num}) ++@item Enabled or Disabled (@samp{Enb}) ++Enabled breakpoints are marked with @samp{1}. @samp{0} marks breakpoints ++that are disabled (not enabled). ++@item Count ++The number of times that breakpoint or watchpoint has been hit. ++@item Condition ++The arithmetic expression ++@item File and Line (@samp{file:line}) ++The filename and line number inside that file where of breakpoint in ++the script. The file and line are separated with a colon. ++@end table ++ ++@noindent ++If a breakpoint is conditional, @code{info break} shows the condition on ++the line following the affected breakpoint; breakpoint commands, if any, ++are listed after that. ++ ++@noindent ++@code{info break} displays a count of the number of times the breakpoint ++has been hit. ++ ++@code{info break} with a breakpoint number @var{n} as argument lists ++only that breakpoint. ++ ++Examples: ++@example ++bashdb<4> @b{info break} ++Breakpoints at following places: ++Num Type Disp Enb What ++1 breakpoint keep y parm.sh:3 ++2 breakpoint keep y parm.sh:28 ++3 breakpoint keep y parm.sh:29 ++4 breakpoint keep y parm.sh:28 ++No watch expressions have been set. ++bashdb<5> @b{info break 4} ++Num Type Disp Enb What ++4 breakpoint keep y parm.sh:28 ++No watch expressions have been set. ++@end example ++@end table ++ ++ ++@ifset FINISHED ++This is especially useful in conjunction with the ++@code{ignore} command. You can ignore a large number of breakpoint ++hits, look at the breakpoint info to see how many times the breakpoint ++was hit, and then run again, ignoring one less than that number. This ++will get you quickly to the last hit of that breakpoint. ++@end ifset ++ ++@DBG allows you to set any number of breakpoints at the same place in ++your script. There is nothing silly or meaningless about this. When ++the breakpoints are conditional, this is even useful ++(@pxref{Conditions, ,Break conditions}). ++ ++@node Set Watchpoints ++@subsubsection Setting watchpoints (@samp{watch}, @samp{watche}) ++ ++@cindex setting watchpoints ++You can use a watchpoint to stop execution whenever the value of an ++expression changes, without having to predict a particular place where ++this may happen. As with the @code{print} (@pxref{Data,,Examining ++Data}), the idiosyncracies of a BASH or any POSIX shell derivative ++suggest using two commands. The @code{watch} command is just for a ++single variables; the @code{watche} command uses the builtin ``let'' ++command to evaluate an expression. If the variable you are tracking ++can take a string value, issuing something like @samp{watch foo} will ++not have the desired effect---any string assignment to @code{foo} will ++have a value 0 when it is assigned via ``let.'' ++ ++@table @code ++@kindex watch ++@item watch @var{var} ++Set a watchpoint for a variable. @DBG will break when the ++value of @var{var} changes. In this command do not add a leading ++dollar symbol to @var{var}. ++ ++@item watche @var{expr} ++Set a watchpoint for an expression via the builtin ``let'' command. ++@DBG will break when @var{expr} is written into by the program ++and its value changes. Not that this may not work for tracking ++arbitrary string value changes. For that use @code{watch} described ++earlier. ++@end table ++ ++@node Delete Breaks ++@subsubsection Deleting breakpoints (@samp{clear}, @samp{delete}) ++ ++@cindex clearing breakpoints, watchpoints ++@cindex deleting breakpoints, watchpoints ++It is often necessary to eliminate a breakpoint or watchpoint once it ++has done its job and you no longer want your script to stop there. ++This is called @dfn{deleting} the breakpoint. A breakpoint that has ++been deleted no longer exists; it is forgotten. ++ ++With the @code{clear} command you can delete breakpoints according to ++where they are in your script. With the @code{delete} command you can ++delete individual breakpoints, or watchpoints by specifying their ++breakpoint numbers. @emph{Note: as described below under the ``clear'' ++command, ``d'' is an alias for ``clear'', not ``delete''. } ++ ++It is not necessary to delete a breakpoint to proceed past it. @DBG ++automatically ignores breakpoints on the first instruction to be executed ++when you continue execution. ++ ++@table @code ++@kindex clear ++@kindex d @r{(@code{clear})} ++@item clear ++Delete any breakpoints at the next instruction to be executed in the ++selected stack frame (@pxref{Selection, ,Selecting a frame}). When ++the innermost frame is selected, this is a good way to delete a ++breakpoint where your script just stopped. ++ ++It may seem odd that we have an alias ``d'' for ``clear.'' It so ++happens that Perl's debugger use ``d'' for its delete command and the ++delete concept in Perl's debugger corresponds to ``clear'' in ++GDB. (Perl doesn't have a notion of breakpoint entry numbers). So in ++order to be compatible with both debugger interfaces, ``d'' is used as ++an alias for ``clear.'' Clear? ++ ++@item clear @var{function} ++@itemx clear @var{filename}:@var{function} ++Delete any breakpoints set at entry to the function @var{function}. ++ ++@item clear @var{linenum} ++@itemx d @var{linenum} ++@ifset FINISHED ++@itemx clear @var{filename}:@var{linenum} ++@end ifset ++Delete any breakpoints set at or within the code of the specified line. ++ ++@cindex delete breakpoints ++@kindex delete ++@kindex de @r{(@code{delete})} ++@item delete @ovar{breakpoints} ++Delete the breakpoints, watchpoints specified as arguments. ++ ++If no argument is specified, delete all breakpoints (@DBG asks ++confirmation, unless you have @code{set confirm off}). You can ++abbreviate this command as @code{de}. ++ ++ ++Note that for compatibility with Perl's debugger, @code{d} means ++something else: @code{clear}. ++ ++@end table ++ ++@node Disabling ++@subsubsection Disabling breakpoints (@samp{disable}, @samp{enable}) ++ ++Rather than deleting a breakpoint or watchpoint, you might ++prefer to @dfn{disable} it. This makes the breakpoint inoperative as if ++it had been deleted, but remembers the information on the breakpoint so ++that you can @dfn{enable} it again later. ++ ++You disable and enable breakpoints, watchpoints, and catchpoints with ++the @code{enable} and @code{disable} commands, optionally specifying one ++or more breakpoint numbers as arguments. Use @code{info break} or ++@code{info watch} to print a list of breakpoints, watchpoints, and ++catchpoints if you do not know which numbers to use. ++ ++A breakpoint, watchpoint, or catchpoint can have any of four different ++states of enablement: ++ ++@itemize @bullet ++@item ++Enabled. The breakpoint stops your program. A breakpoint set ++with the @code{break} command starts out in this state. ++@item ++Disabled. The breakpoint has no effect on your program. ++@item ++Enabled once. The breakpoint stops your program, but then becomes ++disabled. ++@item ++Enabled for deletion. The breakpoint stops your program, but ++immediately after it does so it is deleted permanently. A breakpoint ++set with the @code{tbreak} command starts out in this state. ++@end itemize ++ ++You can use the following commands to enable or disable breakpoints, ++watchpoints, and catchpoints: ++ ++@table @code ++@kindex disable breakpoints ++@kindex disable ++@kindex dis @r{(@code{disable})} ++@item disable @ovar{breakpoints} ++Disable the specified breakpoints---or all breakpoints, if none are ++listed. A disabled breakpoint has no effect but is not forgotten. All ++options such as ignore-counts, conditions and commands are remembered in ++case the breakpoint is enabled again later. You may abbreviate ++@code{disable} as @code{dis}. ++ ++@kindex enable breakpoints ++@kindex enable ++@item enable @ovar{breakpoints} ++Enable the specified breakpoints (or all defined breakpoints). They ++become effective once again in stopping your program. ++@end table ++ ++@c FIXME: I think the following ``Except for [...] @code{tbreak}'' is ++@c confusing: tbreak is also initially enabled. ++Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks, ++,Setting breakpoints}), breakpoints that you set are initially enabled; ++subsequently, they become disabled or enabled only when you use one of ++the commands above. (The command @code{until} can set and delete a ++breakpoint of its own, but it does not change the state of your other ++breakpoints; see @ref{Resuming Execution, ,Resuming Execution}.) ++ ++@node Conditions ++@subsubsection Break conditions (@samp{condition}) ++@cindex conditional breakpoints ++@cindex breakpoint conditions ++ ++The simplest sort of breakpoint breaks every time your script reaches ++a specified place. You can also specify a @dfn{condition} for a ++breakpoint. A condition is just a BASH expression. ++ ++Break conditions can be specified when a breakpoint is set, by using ++@samp{if} in the arguments to the @code{break} command. @xref{Set ++Breaks, ,Setting breakpoints}. A breakpoint with a condition ++evaluates the expression each time your script reaches it, and your ++script stops only if the condition is @emph{true}. ++ ++@cindex one-time breakpoints ++There is also a notion of a ``one-time'' breakpoint which gets deleted ++as soon as it is hit, so that that breakpoint is executed once only. ++ ++Conditions are also accepted for watchpoints; you may not need them, ++since a watchpoint is inspecting the value of an expression anyhow---but ++it might be simpler, say, to just set a watchpoint on a variable name, ++and specify a condition that tests whether the new value is an interesting ++one. ++ ++Break conditions can be specified when a breakpoint is set, by using ++@samp{if} in the arguments to the @code{break} command. @xref{Set ++Breaks, ,Setting breakpoints}. They can also be changed at any time ++with the @code{condition} command. ++ ++@ifset FINISHED ++You can also use the @code{if} keyword with the @code{watch} command. ++The @code{catch} command does not recognize the @code{if} keyword; ++@code{condition} is the only way to impose a further condition on a ++catchpoint. ++@end ifset ++ ++@table @code ++@kindex condition ++@item condition @var{bnum} @var{expression} ++Specify @var{expression} as the break condition for breakpoint ++@var{bnum}. After you set a condition, breakpoint @var{bnum} stops ++your program only if the value of @var{expression} is true (nonzero). ++ ++@item condition @var{bnum} ++Remove the condition from breakpoint number @var{bnum}. It becomes ++an ordinary unconditional breakpoint. ++@end table ++ ++@ifset FINISHED ++When you use @code{condition}, @DBG checks @var{expression} ++immediately for syntactic correctness, and to determine whether ++symbols in it have referents in the context of your breakpoint. If ++@var{expression} uses symbols not referenced in the context of the ++breakpoint, @DBG prints an error message: ++ ++@example ++No symbol "foo" in current context. ++@end example ++@end ifset ++ ++@noindent ++@BASH does ++not actually evaluate @var{expression} at the time the @code{condition} ++command (or a command that sets a breakpoint with a condition, like ++@code{break if @dots{}}) is given, however. ++ ++Examples; ++@example ++condition 1 x>5 # Stop on breakpoint 0 only if x>5 is true. ++condition 1 # Change that! Unconditinally stop on breakpoint 1. ++@end example ++ ++@node Resuming Execution ++@subsection Resuming Execution ++ ++@cindex stepping ++@cindex continuing ++@cindex resuming execution ++@dfn{Continuing} means resuming program execution until your script ++completes normally. In contrast, @dfn{stepping} means executing just ++one more ``step'' of your script, where ``step'' may mean either one ++line of source code. Either when continuing or when stepping, ++your script may stop even sooner, due to a breakpoint or a signal. ++ ++@menu ++* Step:: running the next statement (step) ++* Next:: running the next statement skipping over functions (skip) ++* Finish:: running until the return of a function or ``source'' (finish) ++* Skip:: skipping the next statement (skip) ++* Continue:: continuing execution (continue) ++* Debug:: debugging into another program (debug) ++* Returning:: returning ++@end menu ++ ++@node Step ++@subsubsection Step (@samp{step}) ++@table @code ++@kindex step ++@kindex s @r{(@code{step})} ++@item step ++Continue running your script until control reaches a different source ++line, then stop it and return control to @value{DBG}. This command is ++abbreviated @code{s}. ++ ++The @code{step} command only stops at the first instruction of a source ++line. This prevents the multiple stops that could otherwise occur in ++@code{switch} statements, @code{for} loops, etc. @code{step} continues ++to stop if a function that has debugging information is called within ++the line. In other words, @code{step} @emph{steps inside} any functions ++called within the line. ++ ++@item step @ovar{count} ++Continue running as in @code{step}, but do so @var{count} times. If a ++breakpoint is reached, or a signal not related to stepping occurs before ++@var{count} steps, stepping stops right away. ++@end table ++ ++@node Next ++@subsubsection Next (@samp{next}) ++@table @code ++@kindex next ++@kindex n @r{(@code{next})} ++@item next @ovar{count} ++Continue to the next source line in the current (innermost) stack frame. ++This is similar to @code{step}, but function calls that appear within ++the line of code are executed without stopping. Execution stops when ++control reaches a different line of code at the original stack level ++that was executing when you gave the @code{next} command. This command ++is abbreviated @code{n}. ++ ++An argument @var{count} is a repeat count, as for @code{step}. ++@end table ++ ++@node Finish ++@subsubsection Finish (@samp{finish}) ++@table @code ++@kindex finish ++@item finish ++Continue running until just after function returns. @emph{Currently, ++the line shown on a return is the function header, unless the ++@code{return} builtin function is executed in which case it is the ++line number of the @code{return} function.} ++ ++Contrast this with the @code{return} command (@pxref{Returning, ++,Returning from a function}) and the @code{quit} (@pxref{Quitting the ++BASH debugger, ,Quitting the BASH debugger}). ++ ++@end table ++ ++@node Skip ++@subsubsection Skip (@samp{skip}) ++@table @code ++@kindex skip ++@item skip @ovar{count} ++Skip exection of the next source line. ++This may be useful if you have an action that ``fixes'' existing code in ++the script. The @code{debug} command internally uses the @code{skip} command ++to skip over existing non-debugged invocation that was presumably just ++run. ++@end table ++ ++@node Continue ++@subsubsection Continue (@samp{continue}) ++@table @code ++@kindex continue ++@kindex c @r{(@code{continue})} ++@item continue @ovar{linespec} ++@itemx c @ovar{line-number} ++Resume program execution, at the address where your script last ++stopped; any breakpoints set at that address are bypassed. The ++optional argument @var{linespec} allows you to specify a linespec (a ++line number, function, or filename linenumber combination) to set. A ++one-time breakpoint is deleted when that breakpoint is reached. Should ++the program stop before that breakpoint is reached, for example, ++perhaps another breakpoint or watchpoint is reached first, in a ++listing of the breakpoints you will see this entry with the condition ++9999 which indicates a one-time breakpoint. ++@end table ++ ++To resume execution at a different place, you can use @code{return} ++(@pxref{Returning, ,Returning from a function}) to go back to the ++calling function or sourced script. If you are nested inside a ++subshell, @code{quit} with a value for the number of subshells to ++exit also functions like a return. ++ ++A typical technique for using stepping is to set a breakpoint ++(@pxref{Breakpoints, ,Breakpoints; watchpoints}) at the ++beginning of the function or the section of your script where a problem ++is believed to lie, run your script until it stops at that breakpoint, ++and then step through the suspect area, examining the variables that are ++interesting, until you see the problem happen. ++ ++@node Debug ++@subsubsection Debug (@samp{debug}) ++@table @code ++@kindex debug ++@item debug @ovar{script-name} ++Debug into @var{script-name}. If no name is given the current source line ++is used. In either case the options are prepended to cause the ++debugger to run. ++ ++The nesting level of the debugger is saved inside environment variable ++@code{BASHDB_LEVEL}. The debugger prompt indicates the level of nesting ++by enclosing the history in that many nestings of @code{<>} symbols. ++ ++@end table ++ ++@node Returning ++@subsubsection Returning from a function, sourced file, or subshell (@samp{return}) ++ ++@table @code ++@cindex returning from a function, sourced file or subshell ++@kindex return ++@item return ++@itemx return ++You can cancel execution of a function call or a subshell with the ++@code{return} command. ++@end table ++ ++The @code{return} command does not resume execution; it leaves the ++program stopped in the state that would exist if the function had just ++returned. See also the @code{quit} command (@ref{Quit, ,Quitting the ++BASH debugger}). In some situations @code{return} is similar to ++@code{quit}: in particular when the script is @emph{not} currenlty ++inside in a function and the number of subshells in effect is 0, or ++when a subshell count of 1 is given on the @code{quit} command. ++ ++@ifset FINISHED ++In contrast, the @code{finish} command (@pxref{Continuing ++and Stepping, ,Continuing and stepping}) resumes execution until the ++selected stack frame returns naturally. ++@end ifset ++ ++@node Signals ++@subsection Signals ++@cindex signals ++ ++A signal is an asynchronous event that can happen in a program. The ++operating system defines the possible kinds of signals, and gives each ++kind a name and a number. For example, in Unix @code{SIGINT} is the ++signal a program gets when you type an interrupt character (often ++@kbd{C-c}); @code{SIGALRM} occurs when the alarm clock timer goes off ++(which happens only if your program has requested an alarm). ++ ++Some signal handlers are installed and changed for @value{DBG}'s ++normal use: @code{SIGDEBUG} and @code{SIGEXIT}. @code{SIGDEBUG} is ++used by the debugger to potentially stop your program before execution ++of each statement occurs, and @code{SIGEXIT} is used to catch your ++program just before it is set to leave so you have the option of ++restarting the program with the same options (and not leave the ++debugger) or let the program quit. ++ ++Signal handlers that the debugged script might have installed are ++saved and called before the corresponding debugger handler. Thus, the ++debugged program should work roughly in the same fashion as when it is ++not debugged. However there are some call-stack variables which ++inevitably will differ. To try to hedge this a little so the behaviour ++is the same, @value{DBG} will modify arguments to the traps if it ++finds one of the call-stack that change as a result of the debugger ++being in place. In particluar @env{$LINENO} will get replaced with ++@env{$@{BASH_LINENO[0]@}}; also @env{$@{BASH_LINENO[0]@}} and ++@env{$@{BASH_SOURCE[0]@}} get replaced with ++@env{$@{BASH_LINENO[1]@}} and @env{$@{BASH_SOURCE[1]@}} ++respectively. ++ ++The debugger also installs an interrupt handler @code{SIGINT} so that ++errant programs can be interrupted and you can find out where the ++program was when you interrupted it. ++ ++@cindex fatal signals ++Some signals, including @code{SIGALRM}, are a normal part of the ++functioning of your program. Others, such as @code{SIGSEGV}, indicate ++errors; these signals are @dfn{fatal} (they kill your program immediately) if the ++program has not specified in advance some other way to handle the signal. ++@code{SIGINT} does not indicate an error in your program, but it is normally ++fatal so it can carry out the purpose of the interrupt: to kill the program. ++ ++@BASH has the ability to detect any occurrence of a signal in your ++program. You can tell @BASH in advance what to do for each kind of ++signal. ++ ++@cindex handling signals ++Normally, @BASH is set up to let the non-erroneous signals like ++@code{SIGALRM} be silently passed to your program ++(so as not to interfere with their role in the program's functioning) ++but to stop your program immediately whenever an error signal happens. ++You can change these settings with the @code{handle} command. ++ ++@table @code ++@kindex info signals ++@item info signals ++@itemx info handle ++Print a table of all the kinds of signals and how @BASH has been told to ++handle each one. You can use this to see the signal numbers of all ++the defined types of signals. ++ ++@code{info handle} is an alias for @code{info signals}. ++ ++@kindex handle ++@item handle @var{signal} @var{keywords}@dots{} ++Change the way @BASH handles signal @var{signal}. @var{signal} ++can be the number of a signal or its name (with or without the ++@samp{SIG} at the beginning); a list of signal numbers of the form ++@samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the ++known signals. The @var{keywords} say what change to make. ++@end table ++ ++@c @group ++The keywords allowed by the @code{handle} command can be abbreviated. ++Their full names are: ++ ++@table @code ++@item stop ++@BASH should stop your program when this signal happens. This implies ++the @code{print} keyword as well. ++ ++@item nostop ++@BASH should not stop your program when this signal happens. It may ++still print a message telling you that the signal has come in. ++ ++@item print ++@BASH should print a message when this signal happens. ++ ++@item noprint ++@BASH should not mention the occurrence of the signal at all. ++ ++@item stack ++@BASH should print a stack trace when this signal happens. ++ ++@item nostack ++@BASH should not print a stack trace when this signal occurs. ++ ++@ifset FINISHED ++@item pass ++@itemx noignore ++@BASH should allow your program to see this signal; your program ++can handle the signal, or else it may terminate if the signal is fatal ++and not handled. @code{pass} and @code{noignore} are synonyms. ++ ++@item nopass ++@itemx ignore ++@BASH should not allow your program to see this signal. ++@code{nopass} and @code{ignore} are synonyms. ++@end ifset ++@end table ++@c @end group ++ ++@ifset FINISHED ++When a signal stops your program, the signal is not visible to the ++program until you ++continue. Your program sees the signal then, if @code{pass} is in ++effect for the signal in question @emph{at that time}. In other words, ++after @BASH reports a signal, you can use the @code{handle} ++command with @code{pass} or @code{nopass} to control whether your ++program sees that signal when you continue. ++ ++The default is set to @code{nostop}, @code{noprint}, @code{pass} for ++non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and ++@code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the ++erroneous signals. ++ ++You can also use the @code{signal} command to prevent your program from ++seeing a signal, or cause it to see a signal it normally would not see, ++or to give it any signal at any time. For example, if your program stopped ++due to some sort of memory reference error, you might store correct ++values into the erroneous variables and continue, hoping to see more ++execution; but your program would probably terminate immediately as ++a result of the fatal signal once it saw the signal. To prevent this, ++you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your ++program a signal}. ++@end ifset ++ ++@node Information ++@section Status and Debugger Settings (@samp{info}, @samp{show}) ++ ++In addition to @code{help}, you can use the @BASH commands @code{info} ++and @code{show} to inquire about the state of your program, or the state ++of @BASH itself. Each command supports many topics of inquiry; this ++manual introduces each of them in the appropriate context. The listings ++under @code{info} and under @code{show} in the Index point to ++all the sub-commands. @xref{Command Index}. ++ ++@c @group ++@table @code ++@kindex info ++@kindex i @r{(@code{info})} ++@item info ++This command (abbreviated @code{i}) is for describing the state of ++your program. For example, you canlist the current @code{$1}, @code{$2} ++parameters with @code{info args}, or list the breakpoints you have set ++with @code{info breakpoints} or @code{info watchpoints}. You can get ++a complete list of the @code{info} sub-commands with @w{@code{help ++info}}. ++ ++@kindex set ++@item set ++You can assign the result of an expression to an environment variable ++with @code{set}. For example, you can set the number of lines ++@BASH will list by default to 20 with @code{set listsize 20}. ++ ++@kindex show ++@item show ++In contrast to @code{info}, @code{show} is for describing the state of ++@BASH itself. ++You can change most of the things you can @code{show}, by using the ++related command @code{set}; ++ ++The distinction between @code{info} and @code{show} however is a bit ++fuzzy. For example, to list the arguments given to your script use ++@code{show args}; @code{info args} does something different. ++ ++@kindex info set ++To display all the settable parameters and their current ++values, you can use @code{show} with no arguments; you may also use ++@code{info set}. Both commands produce the same display. ++@c FIXME: "info set" violates the rule that "info" is for state of ++@c FIXME...program. Ck w/ GNU: "info set" to be called something else, ++@c FIXME...or change desc of rule---eg "state of prog and debugging session"? ++@end table ++@c @end group ++ ++Here are three miscellaneous @code{show} subcommands, all of which are ++exceptional in lacking corresponding @code{set} commands: ++ ++@table @code ++@kindex show version ++@cindex version number ++@item show version ++Show what version of @BASH is running. You should include this ++information in @BASH bug-reports. If multiple versions of ++@BASH are in use at your site, you may need to determine which ++version of @BASH you are running; as @BASH evolves, new ++commands are introduced, and old ones may wither away. Also, many ++system vendors ship variant versions of @value{BASH}, and there are ++variant versions of @BASH in @sc{gnu}/Linux distributions as well. ++The version number is the same as the one announced when you start ++@value{BASH}. ++ ++@kindex show copying ++@item show copying ++Display information about permission for copying @value{BASH}. ++ ++@kindex show warranty ++@item show warranty ++Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty, ++if your version of @DBG comes with one. ++ ++@end table ++ ++@node Stack ++@section Examining the Stack (@samp{where}, @samp{frame}, @samp{up}, @samp{down}) ++ ++When your script has stopped, one thing you'll probably want to know ++is where it stopped and some idea of how it got there. ++ ++@cindex call stack ++Each time your script performs a function call (either as part of a ++command substitution or not), or `source's a file, information about ++this action is saved. The call stack then is this a history of the ++calls that got you to the point that you are currently stopped at. ++ ++@cindex selected frame ++One of the stack frames is @dfn{selected} by @DBG and many ++@DBG commands refer implicitly to the selected frame. In ++particular, whenever you ask @DBG to list lines without giving ++a line number or location the value is found in the selected frame. ++There are special @DBG commands to select whichever frame you ++are interested in. @xref{Selection, ,Selecting a frame}. ++ ++When your program stops, @BASH automatically selects the ++currently executing frame and describes it briefly, similar to the ++@code{frame} command. ++ ++ ++@menu ++* Frames:: Stack frames ++* Backtrace:: Backtraces (where) ++* Selection:: Selecting a frame (up, down, frame) ++ ++@end menu ++ ++@node Frames ++@subsection Stack frames ++ ++@cindex frame, definition ++@cindex stack frame ++The call stack is divided up into contiguous pieces called @dfn{stack ++frames}, or @dfn{frames} for short; each frame is the data associated ++with one call to one function. The frame contains the line number of ++the caller of the function, the source-file name that the line refers ++to and function name (which could be the built-in name ``source'').. ++ ++@cindex initial frame ++@cindex outermost frame ++@cindex innermost frame ++When your script is started, the stack has only one frame, that of the ++function @code{main}. This is called the @dfn{initial} frame or the ++@dfn{outermost} frame. Each time a function is called, a new frame is ++made. Each time a function returns, the frame for that function invocation ++is eliminated. If a function is recursive, there can be many frames for ++the same function. The frame for the function in which execution is ++actually occurring is called the @dfn{innermost} frame. This is the most ++recently created of all the stack frames that still exist. ++ ++@cindex frame number ++@value{DBG} assigns numbers to all existing stack frames, starting with ++zero for the innermost frame, one for the frame that called it, ++and so on upward. These numbers do not really exist in your script; ++they are assigned by @value{DBG} to give you a way of designating stack ++frames in @value{DBG} commands. ++ ++@node Backtrace ++@subsection Backtraces (@samp{where}) ++ ++@cindex backtraces ++@cindex tracebacks ++@cindex stack traces ++A backtrace is essentially the same as the call stack: a summary of ++how your script got where it is. It shows one line per frame, for ++many frames, starting with the place that you sare stopped at (frame ++zero), followed by its caller (frame one), and on up the stack. ++ ++@table @code ++@kindex backtrace ++@kindex bt @r{(@code{backtrace})} ++@item backtrace ++@itemx bt ++@itemx where ++@itemx T ++Print a backtrace of the entire stack: one line per frame for all ++frames in the stack. ++ ++@item backtrace @var{n} ++@itemx bt @var{n} ++@itemx where @var{n} ++@itemx T @var{n} ++Similar, but print only the innermost @var{n} frames. ++ ++@ifset FINISHED ++@item backtrace -@var{n} ++@itemx bt -@var{n} ++@itemx where -@var{n} ++@itemx T -@var{n} ++Similar, but print only the outermost @var{n} frames. ++@end ifset ++@end table ++ ++@kindex where ++The names @code{where} and @code{T} are additional aliases for ++@code{backtrace}. ++ ++Each line in the backtrace shows the frame number and the function ++name, the source file name and line number, as well as the function name. ++ ++Here is an example of a backtrace taken a program in the ++regression-tests @file{parm.sh}. ++ ++@smallexample ++@group ++% ../bashdb -n -L .. parm.sh ++Bourne-Again Shell Debugger, release @value{BASHRELEASE} ++Copyright 2002,2003 Rocky Bernstein ++This is free software, covered by the GNU General Public License, and you are ++welcome to change it and/or distribute copies of it under certain conditions. ++ ++(./parm.sh:21): ++21: fn1 5 ++bashdb<0> @b{continue fn3} ++One-time breakpoint 1 set in file ./parm.sh, line 17. ++fn2: testing 1 2 3 ++(./parm.sh:17): ++17: fn3() @{ ++bashdb<1> @b{where} ++->0 in file `./parm.sh' at line 14 ++##1 fn3() called from file `./parm.sh' at line 14 ++##2 fn2("testing 1", "2 3") called from file `parm.sh' at line 5 ++##3 fn1("0") called from file `parm.sh' at line 9 ++##4 fn1("1") called from file `parm.sh' at line 9 ++##5 fn1("2") called from file `parm.sh' at line 9 ++##6 fn1("3") called from file `parm.sh' at line 9 ++##7 fn1("4") called from file `parm.sh' at line 9 ++##8 fn1("5") called from file `parm.sh' at line 21 ++##9 source("parm.sh") called from file `bashdb' at line 143 ++##10 main("-n", "-L", "..", "parm.sh") called from file `bashdb' at line 0 ++@end group ++@end smallexample ++ ++@noindent ++The display for ``frame'' zero isn't a frame at all, although it has ++the same information minus a function name; it just indicates that ++your script has stopped at the code for line @code{14} ++of @code{./parm.sh}. ++ ++@node Selection ++@subsection Selecting a frame (@samp{up}, @samp{down}, @samp{frame}) ++ ++Commands for listing source code in your script work on whichever ++stack frame is selected at the moment. Here are the commands for ++selecting a stack frame; all of them finish by printing a brief ++description of the stack frame just selected. ++ ++@table @code ++@kindex up ++@item up @var{n} ++Move @var{n} frames up the stack. For positive numbers @var{n}, this ++advances toward the outermost frame, to higher frame numbers, to frames ++that have existed longer. @var{n} defaults to one. ++ ++@kindex down ++@kindex do @r{(@code{down})} ++@item down @var{n} ++Move @var{n} frames down the stack. For positive numbers @var{n}, this ++advances toward the innermost frame, to lower frame numbers, to frames ++that were created more recently. @var{n} defaults to one. You may ++abbreviate @code{down} as @code{do}. ++@end table ++ ++All of these commands end by printing two lines of output describing the ++frame. The first line shows the frame number, the function name, the ++arguments, and the source file and line number of execution in that ++frame. The second line shows the text of that source line. ++ ++@need 100 ++For example: ++ ++@smallexample ++@group ++bashdb<8> @b{up} ++19: sourced_fn ++bashdb<8> @b{T} ++##0 in file `./bashtest-sourced' at line 8 ++->1 sourced_fn() called from file `bashtest-sourced' at line 19 ++##2 source() called from file `bashdb-test1' at line 23 ++##3 fn2() called from file `bashdb-test1' at line 33 ++##4 fn1() called from file `bashdb-test1' at line 42 ++##5 main() called from file `bashdb-test1' at line 0 ++@end group ++@end smallexample ++ ++After such a printout, the @code{list} command with no arguments ++prints ten lines centered on the point of execution in the frame. ++@xref{List, ,Printing source lines}. ++ ++@table @code ++@kindex frame ++@cindex current stack frame ++@item frame @var{args} ++The @code{frame} command allows you to move from one stack frame to ++another, and to print the stack frame you select. @var{args} is the ++the stack frame number. Without an argument, @code{frame} prints the ++current stack frame. ++@end table ++ ++@node List ++@section Examining Source Files (@samp{list}) ++ ++@value{DBG} can print parts of your script's source. When your ++script stops, @value{DBG} spontaneously prints the line where it ++stopped. Likewise, when you select a stack frame (@pxref{Selection, ++,Selecting a frame}), @value{DBG} prints the line where execution in ++that frame has stopped. You can print other portions of source files ++by explicit command. ++ ++If you use @value{DBG} through its @sc{gnu} Emacs interface, you may ++prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using ++@value{DBG} under @sc{gnu} Emacs}. ++ ++@kindex list ++@kindex l @r{(@code{list})} ++To print lines from a source file, use the @code{list} command ++(abbreviated @code{l}). By default, ten lines are printed. ++There are several ways to specify what part of the file you want to print. ++ ++Here are the forms of the @code{list} command most commonly used: ++ ++@table @code ++@item list @var{linenum} ++@itemx l @var{linenum} ++Print lines centered around line number @var{linenum} in the ++current source file. ++ ++@item list @var{function} ++@itemx l @var{function} ++Print the text of @var{function}. ++ ++@item list ++@itemx l ++Print more lines. If the last lines printed were printed with a ++@code{list} command, this prints lines following the last lines ++printed; however, if the last line printed was a solitary line printed ++as part of displaying a stack frame (@pxref{Stack, ,Examining the ++Stack}), this prints lines centered around that line. ++ ++@item list - ++@itemx l - ++Print lines just before the lines last printed. ++@end table ++ ++By default, @value{DBG} prints ten source lines with any of these forms of ++the @code{list} command. ++You can change this using @code{set listsize}: ++ ++@table @code ++@kindex set listsize ++@item set listsize @var{count} ++Make the @code{list} command display @var{count} source lines (unless ++the @code{list} argument explicitly specifies some other number). ++ ++@kindex show listsize ++@item show listsize ++Display the number of lines that @code{list} prints. ++@end table ++ ++Repeating a @code{list} command with @key{RET} discards the argument, ++so it is equivalent to typing just @code{list}. This is more useful ++than listing the same lines again. An exception is made for an ++argument of @samp{-}; that argument is preserved in repetition so that ++each repetition moves up in the source file. ++ ++@cindex linespec ++In general, the @code{list} command expects you to supply a ++@dfn{linespecs}. Linespecs specify source lines; there are several ways ++of writing them, but the effect is always to specify some source line. ++ ++Here is a complete description of the possible arguments for @code{list}: ++ ++@table @code ++@item list @var{linespec} ++Print lines centered around the line specified by @var{linespec}. ++ ++@item list @var{first} @var{increment} ++Print @var{increment} lines starting from @var{first} ++ ++@item list @var{first} ++Print lines starting with @var{first}. ++ ++@item list - ++Print lines just before the lines last printed. ++ ++@item list . ++Print lines after where the script is stopped. ++ ++@item list ++As described in the preceding table. ++@end table ++ ++Here are the ways of specifying a single source line---all the ++kinds of linespec. ++ ++@table @code ++@item @var{number} ++Specifies line @var{number} of the current source file. ++When a @code{list} command has two linespecs, this refers to ++the same source file as the first linespec. ++ ++@item @var{filename}:@var{number} ++Specifies line @var{number} in the source file @var{filename}. ++ ++@item @var{function} ++Specifies the line that function @var{function} is listed on. ++ ++@ifset FINISHED ++@item @var{filename}:@var{function} ++Specifies the line of function @var{function} in the file ++@var{filename}. You only need the file name with a function name to ++avoid ambiguity when there are identically named functions in ++different source files. ++@end ifset ++@end table ++ ++@node Search ++@section Searching source files (@samp{search}, @samp{reverse}, @samp{/.../}, @samp{?..?}) ++@cindex searching ++@kindex reverse-search ++ ++There are two commands for searching through the current source file for a ++BASH extended pattern-matching expression. ++ ++@table @code ++@kindex search ++@kindex forward ++@item forward @var{bash-pattern} ++@itemx search @var{bash-pattern} ++The command @samp{forward @var{bash-pattern}} checks each line, ++starting with the one following the current line, for a match for ++@var{bash-pattern} which is an extended bash pattern-matching ++expression. It lists the line that is found. You can use the synonym ++@samp{search @var{bash-pattern}} or abbreviate the command name as ++@code{fo} or @code{/@var{pat}/}. ++ ++@item reverse @var{bash-pattern} ++The command @samp{reverse @var{bash-pattern}} checks each line, starting ++with the one before the last line listed and going backward, for a match ++for @var{bash-pattern}. It lists the line that is found. You can abbreviate ++this command as @code{rev} or @code{?@var{bash-pattern}?}. ++@end table ++ ++@node Data ++@section Examining Data (@samp{print}, @samp{examine}, @samp{info variables}) ++ ++@cindex printing data ++@cindex examining data ++@kindex print ++ ++One way to examine string data in your script is with the @code{print} ++command (abbreviated @code{p}). However a more versatile print command ++is @code{x}; it can print variable and function definitions and can do ++arithmetic computations. Finally, the most general method would be ++via @code{eval echo}. ++ ++@table @code ++@kindex print ++@kindex p @r{(@code{print})} ++@item print @var{expr} ++ ++Use @code{print} to dispay strings as you would from @code{echo}. And ++as such, variable names to be substituted have to be preceded with a ++dollar sign. As with echo, filename expansion, e.g. tilde expansion, ++is performed on unquoted strings. So for example if you want to print ++a *, you would write @samp{print "*"}, not @samp{print *}. If you want ++to have the special characters dollars sign appear, use a backslash. ++ ++@item print ++@itemx p ++If you omit @var{expr}, @value{DBG} displays the last expression again. ++ ++@item x @ovar{expr} ++@kindex x ++ ++This is a smarter, more versatile ``print'' command, and although sometimes ++it might not be what you want, and you may want to resort to either ++@code{print} or @code{eval echo...}. ++ ++As with @code{print}, if you omit @var{expr}, @value{DBG} displays ++the last expression again. ++ ++The @code{x} command first checks if @var{expr} is single variable. If ++it is, the definition and value are printed via @value{BASH}'s ++@code{declare -p} command. This will show the variable's attributes ++such as if it is read only or if it is an integer. If the variable is ++an array, that is show and the array values are printed. ++ ++If instead @var{expr} is a function, the function definition is ++printed via @value{BASH}'s @code{declare -f} command. If @var{expr} ++was neither a variable nor an expression, then we try to get a value ++via @code{let}. And if this returns an error, as a last resort we call ++@code{print} and give what it outputs. ++ ++Since @code{let} may be used internally and since (to my thinking) ++@code{let} does funny things, the results may seem odd unless you ++understand the sequence tried above and how @code{let} works. For ++example if the variable @code{foo} has value 5, then @samp{x foo} show ++the definition of foo with value 5, and @samp{x foo+5} prints 10 as ++expected. So far so good. However if @code{foo} is has value ++``string'', @samp{x foo+5} prints 5 because @code{let} has turned the ++string into 0; @samp{p foo+5} will simply print ``foo+5''; if you want ++the value of ``foo'' substituted inside a string, for example you ++expect ``the value of foo is $foo'' to come out ``the value of foo is ++5'', then the right command to use is @code{print} rather than ++@code{x}, making sure you add the dollar onto the beginning of the ++variable. ++ ++@item V @ovar{!}@ovar{pattern} ++@kindex V ++ ++If you want to @emph{all} list variables and there values or a set of ++variables by pattern, use this command. ++ ++@smallexample ++@group ++bashdb<0> @b{V dq*} ++dq_args="dq_*" ++dq_cmd="V" ++@end group ++@end smallexample ++ ++@end table ++ ++@node Evaluation/Execution ++@section Running Arbitrary BASH and Shell commands (@samp{eval}, @samp{shell}) ++ ++The two most general commands and most ``low-level'' are @code{eval} ++and @code{shell}. ++ ++@table @code ++@item eval ++@itemx e ++@kindex e ++ ++In contrast to the commands of the last section the most general way ++to examine data is through @code{eval}. But you do much more with ++this; you can change the values of variables, since, you are just ++evaluating BASH code. ++ ++If you expect output, you should arrange that in the command, such as ++via @code{echo} or @code{printf}. For example, to print the value of ++@var{foo}, you would type @samp{e echo $foo}. This is bit longer than ++@samp{p $foo} or (when possible) @samp{x foo}. However suppose you ++wanted to find out how the builtin test operator @samp{[} works with ++the @samp{-z} test condition. You could use @code{eval} to do this ++such as @samp{e [ -z "$foo"] && echo "yes"}. ++ ++@kindex shell ++@kindex !! @r{(@code{shell})} ++@cindex shell escape ++@item shell @var{command string} ++@itemx !! ++ ++If you need to execute occasional shell commands during your ++debugging session, there is no need to leave or suspend @value{DBG}; you can ++just use the @code{shell} command or its alias @code{!!}. ++ ++Invoke a shell to execute @var{command string}. ++ ++@end table ++ ++@node Auto Display ++@section Automatic display (@samp{display}, @samp{undisplay}) ++@cindex automatic display ++@cindex display of expressions ++ ++If you find that you want to print the value of an expression ++frequently (to see how it changes), you might want to add it to the ++@dfn{automatic display list} so that @value{DBG} evaluates a ++statement each time your program stops. Each expression added to the ++list is given a number to identify it; to remove an expression from ++the list, you specify that number. The automatic display looks like ++this: ++ ++@example ++2 (echo $x): 38 ++@end example ++ ++@noindent ++This display shows item numbers, expressions and their current values. ++ ++@table @code ++@kindex display ++@item display @var{expr} ++Add the expression @var{expr} to the list of expressions to display ++each time your program stops. ++@end table ++ ++@table @code ++@kindex delete display ++@kindex undisplay ++@item undisplay @var{dnums}@dots{} ++@itemx delete display @var{dnums}@dots{} ++Remove item numbers @var{dnums} from the list of expressions to display. ++ ++@code{undisplay} does not repeat if you press @key{RET} after using it. ++(Otherwise you would just get the error @samp{No display number @dots{}}.) ++ ++@kindex disable display ++@item disable display @var{dnums}@dots{} ++Disable the display of item numbers @var{dnums}. A disabled display ++item is not printed automatically, but is not forgotten. It may be ++enabled again later. ++ ++@kindex enable display ++@item enable display @var{dnums}@dots{} ++Enable display of item numbers @var{dnums}. It becomes effective once ++again in auto display of its expression, until you specify otherwise. ++ ++@item display ++Display the current values of the expressions on the list, just as is ++done when your program stops. ++ ++@kindex info display ++@item info display ++Print the list of expressions previously set up to display ++automatically, each one with its item number, but without showing the ++values. This includes disabled expressions, which are marked as such. ++It also includes expressions which would not be displayed right now ++because they refer to automatic variables not currently available. ++@end table ++ ++@node Controlling bashdb ++@section Controlling @DBG (@samp{set}, @samp{history}) ++ ++You can alter the way @BASH interacts with you by using the ++@code{set} command. ++ ++@menu ++* Annotate:: Annotation Level (set annotate) ++* Prompt:: Prompt (set prompt, show prompt) ++* Editing:: Command editing (set editing, show editing) ++* Command Display:: Command display (set showcommand) ++* History:: Command history (history, !, H) ++@end menu ++ ++@node Annotate ++@subsection Annotation Level (@samp{set annoatate}) ++ ++The annotation level controls how much information does @value{DBG} ++print together with its prompt, values of expressions, source lines, ++and other types of output. Level 0 is the normal, level 1 is for use ++when @value{DBG} is run as a subprocess of @sc{gnu} Emacs of @value{DDD}, ++level 2 is the maximum annotation suitable for programs that control ++@value{DBGN}. ++ ++@node Prompt ++@subsection Prompt (@samp{set prompt}, @samp{show prompt}) ++ ++@cindex prompt ++ ++@value{dBGP} indicates its readiness to read a command by printing a ++string called the @dfn{prompt}. This string is normally: ++@example ++bashdb$@{_Dbg_less@}$@{#_Dbg_history[@@]@}$@{_Dbg_greater@}$_Dbg_space ++@end example ++ ++When variables inside the the prompt string are evaluated, the above ++becomes something like @samp{bashdb<5>} if this is the fifth command ++executed or perhaps @samp{bashdb<<2>>} if you have called the debugger ++from inside a debugger session and this is the second command inside ++the debugger session or perhaps @samp{bashdb<(6)>} if you ++entered a subshell after the fifth command. ++ ++You can change the prompt string with the @code{set prompt} command, ++although it is not normally advisable to do so without understanding ++the implications. If you are using the @value{DDD} GUI, it changes the ++changes the prompt and should not do so. In certain other ++circumstances (such as writing a GUI like @value{DDD}), it may be is useful ++to change the prompt. ++ ++@emph{Note:} @code{set prompt} does not add a space for you after the ++prompt you set. This allows you to set a prompt which ends in a space ++or a prompt that does not. Furthermore due to a implementation ++limitation (resulting from a limitation of the bash built-in function ++``read''), to put a space at the end of the prompt use the ++@samp{$_Dbg_space} variable. ++ ++@table @code ++@kindex set prompt ++@item set prompt @var{newprompt} ++Directs @value{DBG} to use @var{newprompt} as its prompt string ++henceforth. ++ ++@emph{Warning: changing the prompt can @value{DDD}'s ability to ++understand when the debugger is waiting for input.} ++ ++@kindex show prompt ++@item show prompt ++Prints a line of the form: @samp{bashdb's prompt is: @var{your-prompt}} ++@end table ++ ++@node Editing ++@subsection Command editing (@samp{set editing}, @samp{show editing}) ++@cindex readline ++@cindex command line editing ++ ++@value{DBG} reads its input commands through bash which uses via the ++@dfn{readline} interface. This @sc{gnu} library provides consistent ++behavior for programs which provide a command line interface to the ++user. Advantages are @sc{gnu} Emacs-style or @dfn{vi}-style inline ++editing of commands, @code{csh}-like history substitution, and a ++storage and recall of command history across debugging sessions. ++ ++You may control the behavior of command line editing in @BASH with the ++command @code{set}. ++ ++@table @code ++@kindex set editing ++@cindex editing ++@item set editing ++@itemx set editing on ++Enable command line editing (enabled by default). ++ ++@item set editing off ++Disable command line editing. ++ ++@kindex show editing ++@item show editing ++Show whether command line editing is enabled. ++@end table ++ ++ ++@node Command Display ++@subsection Command Display (@samp{set showcommand}) ++ ++The debugger normally lists the line number and source line of the for ++the statement to be next executed. Often this line contains one ++expression or one statement and it is clear from this line what's ++going to happen. However @BASH allows many expressions or ++statements to be put on a single source line; some lines ++contain several units of execution. Some examples of this ++behavior are listed below: ++ ++@smallexample ++x=1; y=2; x=3 ++(( x > 5 )) && x=5 ++y=`echo *` ++@end smallexample ++ ++In the first line of the example above, we have three assignment ++statements on a single line. In the second line of the example above ++we have a statement which gets run only if a condition tests true. And ++in the third line of the example above, we have a command that gets ++run and then the output of that is substituted in an assignemnt ++statement. If you were single stepping inside the debugger, each line ++might get listed more than once before each of the actions that might ++get performed. (In the case of the conditional statement, the ++line gets listed only once when the condition is false.) ++ ++In order to assist understanding where you are, the enhanced version ++of @BASH maintains a dynamic variable @env{BASH_COMMAND} that ++contains piece of code next to be run (or is currently being run). The ++debugger has arranged to save this and can display this information ++or not. This is controlled by @code{set showcommand}. ++ ++@table @code ++@kindex set showcommand ++@item set showcommand @r{[}auto | on | off @r{]} ++controls whether or not to show the saved @env{BASH_COMMAND} for the ++command next to be executed. ++@end table ++ ++When the value is @code{auto} the following heuristic is used to ++determine whether or not to display the saved @env{BASH_COMMAND}. If ++the last time you stopped you were at the same place and the command ++string has changed, then show the command. When the value @code{on} is ++used, the debugger always shows @env{BASH_COMMAND} and when ++@code{off} is used, the debugger nevers shows ++@env{BASH_COMMAND}. Note that listing the text of the source line is ++independent of whether or not the command is also listed. ++ ++Some examples: ++@smallexample ++set showcommand auto @b{This is the default} ++set showcommand on @b{Always show the next command to be executed} ++set showcommand off @b{Never show the next command to be executed} ++@end smallexample ++ ++@node History ++@subsection Command history (@samp{H}, @samp{history}, @samp{!}) ++ ++@value{dBGP} can keep track of the commands you type during your ++debugging sessions, so that you can be certain of precisely what ++happened. If the prompt has not been changed (see @ref{Prompt, ++,Prompt}), the history number that will be in use next is by default ++listed in the debugger prompt. Invalid commands and history commands ++are not saved on the history stack. ++ ++@table @code ++@kindex H @r{[}@var{start-number} @ovar{end-number}@r{]} ++@item H @r{[}@var{start-number} @ovar{end-number}@r{]} ++@item H @ovar{-count} ++@itemx !@r{[}-@r{]}@var{n}:p ++ ++You can list what is in the history stack with @code{H}. Debugger ++commands in ths history stack are listed from most recent to least recent. ++If no @var{start-number} is given we start with the most recently ++executed command and end with the first entry in the history stack. ++If @var{start-number} is given, that history number is listed first. If ++@var{end-number} is given, that history number is listed last. If a ++single negative number is given list that many history commands. ++ ++An alternate form is @code{!@emph{n}:p} or @code{!-@emph{n}:p} where ++@emph{n} is an integer. If a minus sign is used, @emph{n} is taken as ++the count to go back from the end rather than as a absolute history ++number. In contrast @code{H}, this form only prints a @emph{single} ++history item. ++ ++Some examples: ++@smallexample ++H @b{List entire history} ++H -2 @b{List the last two history items} ++!-2:p @b{List a single history item starting at the same place as above} ++H 5 @b{List history from history number 5 to the begining (number 0)} ++H 5 0 @b{Same as above} ++H 5 3 @b{List history from history number 5 down to history number 3} ++!5:p @b{List a single history item 5} ++@end smallexample ++ ++@kindex history @r{[}-@r{]}@r{[}@var{n}@r{]} ++@kindex !@r{[}-@r{]}@var{n} @r{(@code{history})} ++@item history @r{[}@r{[}-@r{]}@var{n}@r{]} ++@itemx !@r{[}-@r{]}@var{n} ++ ++Use this command to reexecute a given history number. If no number is ++given, the last debugger command in the history is executed. ++ ++An alternate form is @code{!@emph{n}} or @code{!-@emph{n}} where ++@emph{n} is an integer. ++ ++If a minus sign is used in in either form, @emph{n} is taken as the ++count to go back from the end rather than as a absolute history ++number. ++ ++@end table ++ ++@ifset FINISHED ++Use these commands to manage the @value{DBGN} command ++history facility. ++ ++@table @code ++@cindex history substitution ++@cindex history file ++@kindex set history filename ++@kindex GDBHISTFILE ++@item set history filename @var{fname} ++Set the name of the @BASH command history file to @var{fname}. ++This is the file where @BASH reads an initial command history ++list, and where it writes the command history from this session when it ++exits. You can access this list through history expansion or through ++the history command editing characters listed below. This file defaults ++to the value of the environment variable @code{GDBHISTFILE}, or to ++@file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable ++is not set. ++ ++@cindex history save ++@kindex set history save ++@item set history save ++@itemx set history save on ++Record command history in a file, whose name may be specified with the ++@code{set history filename} command. By default, this option is disabled. ++ ++@item set history save off ++Stop recording command history in a file. ++ ++@cindex history size ++@kindex set history size ++@item set history size @var{size} ++Set the number of commands which @BASH keeps in its history list. ++This defaults to the value of the environment variable ++@code{HISTSIZE}, or to 256 if this variable is not set. ++@end table ++ ++@cindex history expansion ++History expansion assigns special meaning to the character @kbd{!}. ++ ++Since @kbd{!} is also the logical not operator in C, history expansion ++is off by default. If you decide to enable history expansion with the ++@code{set history expansion on} command, you may sometimes need to ++follow @kbd{!} (when it is used as logical not, in an expression) with ++a space or a tab to prevent it from being expanded. The readline ++history facilities do not attempt substitution on the strings ++@kbd{!=} and @kbd{!(}, even when history expansion is enabled. ++ ++The commands to control history expansion are: ++ ++@table @code ++@kindex set history expansion ++@item set history expansion on ++@itemx set history expansion ++Enable history expansion. History expansion is off by default. ++ ++@item set history expansion off ++Disable history expansion. ++ ++The readline code comes with more complete documentation of ++editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs ++or @code{vi} may wish to read it. ++ ++@c @group ++@kindex show history ++@item show history ++@itemx show history filename ++@itemx show history save ++@itemx show history size ++@itemx show history expansion ++These commands display the state of the @BASH history parameters. ++@code{show history} by itself displays all four states. ++@c @end group ++@end table ++ ++@table @code ++@kindex shows ++@item show commands ++Display the last ten commands in the command history. ++ ++@item show commands @var{n} ++Print ten commands centered on command number @var{n}. ++ ++@item show commands + ++Print ten commands just after the commands last printed. ++@end table ++@end ifset ++ ++@node Front Ends ++@chapter Using @value{DBG} from a front-end user interface ++ ++There are two front-ends available that can use @value{DBG} as a back-end ++debugger are @sc{gnu} Emacs and @value{DDD}. ++ ++@menu ++* Emacs:: Using @value{DBG} from @sc{gnu} Emacs ++* DDD:: Using @value{DBG} from @value{DDD} ++@end menu ++ ++ ++@node Emacs ++@section Using @value{DBG} from @sc{gnu} Emacs ++ ++@cindex Emacs ++@cindex @sc{gnu} Emacs ++A special interface allows you to use @sc{gnu} Emacs to view (and ++edit) the source files for the program you are debugging with ++@value{DBG}. ++ ++To use this interface, use the command @kbd{M-x bashdb} in Emacs. ++Give the executable file you want to debug as an argument. This ++command starts @value{DBG} as a subprocess of Emacs, with input and ++output through a newly created Emacs buffer. ++ ++Using @value{DBG} under Emacs is just like using @value{DBG} ++normally except for two things: ++ ++@itemize @bullet ++@item ++All ``terminal'' input and output goes through the Emacs buffer. ++@end itemize ++ ++This applies both to @value{DBG} commands and their output, and to the input ++and output done by the program you are debugging. ++ ++This is useful because it means that you can copy the text of previous ++commands and input them again; you can even use parts of the output ++in this way. ++ ++All the facilities of Emacs' Shell mode are available for interacting ++with your script. In particular, you can send signals the usual ++way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a ++stop. ++ ++@itemize @bullet ++@item ++@value{DBG} displays source code through Emacs. ++@end itemize ++ ++Each time @value{DBG} displays a stack frame, Emacs automatically finds the ++source file for that frame and puts an arrow (@samp{=>}) at the ++left margin of the current line. Emacs uses a separate buffer for ++source display, and splits the screen to show both your @value{DBG} session ++and the source. ++ ++Explicit @value{DBG} @code{list} or search commands still produce output as ++usual, but you probably have no reason to use them from Emacs. ++ ++@quotation ++@emph{Warning:} If the directory where your script resides is not your ++current directory, it can be easy to confuse Emacs about the location of ++the source files, in which case the auxiliary display buffer does not ++appear to show your source. @value{DBG} can find programs by searching your ++environment's @code{PATH} variable, so the @value{DBG} input and output ++session proceeds normally; but Emacs does not get enough information ++back from @value{DBG} to locate the source files in this situation. To ++avoid this problem, either start @value{DBG} mode from the directory where ++your script resides, or specify an absolute file name when prompted for the ++@kbd{M-x gdb} argument. ++ ++A similar confusion can result if you use the @value{DBG} @code{file} command to ++switch to debugging a program in some other location, from an existing ++@value{DBG} buffer in Emacs. ++@end quotation ++ ++By default, @kbd{M-x bashdb} calls the @code{bash --debugger}. If you ++need to call @value{DBG} by a different name (for example, if you ++keep several configurations around, with different names) you can set ++the Emacs variable @code{gud-bashdb-command-name}; for example, ++ ++@example ++(setq gud-bashdb-command-name "bash-debugger") ++@end example ++ ++@noindent ++(preceded by @kbd{M-:} or @kbd{ESC :}, or typed in the @code{*scratch*} buffer, or ++in your @file{.emacs} file) makes Emacs call the program named ++``@code{bash-debugger}'' instead. ++ ++In the @value{DBG} I/O buffer, you can use these special Emacs commands in ++addition to the standard Shell mode commands: ++ ++@table @kbd ++@item C-h m ++Describe the features of Emacs' @value{DBG} Mode. ++ ++@item M-s ++Execute to another source line, like the @value{DBG} @code{step} command; also ++update the display window to show the current file and location. ++ ++@item M-n ++Execute to next source line in this function, skipping all function ++calls, like the @value{DBG} @code{next} command. Then update the display window ++to show the current file and location. ++ ++@item C-c C-f ++Execute until exit from the selected stack frame, like the @value{DBG} ++@code{finish} command. ++ ++@item M-c ++Continue execution of your script, like the @value{DBG} @code{continue} ++command. ++ ++@emph{Warning:} In Emacs v19, this command is @kbd{C-c C-p}. ++ ++@item M-u ++Go up the number of frames indicated by the numeric argument ++(@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}), ++like the @value{DBG} @code{up} command. ++ ++@emph{Warning:} In Emacs v19, this command is @kbd{C-c C-u}. ++ ++@item M-d ++Go down the number of frames indicated by the numeric argument, like the ++@value{DBG} @code{down} command. ++ ++@emph{Warning:} In Emacs v19, this command is @kbd{C-c C-d}. ++ ++@item C-x & ++Read the number where the cursor is positioned, and insert it at the end ++of the @value{DBG} I/O buffer. For example, if you wish to disassemble code ++around an address that was displayed earlier, type @kbd{disassemble}; ++then move the cursor to the address display, and pick up the ++argument for @code{disassemble} by typing @kbd{C-x &}. ++ ++You can customize this further by defining elements of the list ++@code{gdb-print-command}; once it is defined, you can format or ++otherwise process numbers picked up by @kbd{C-x &} before they are ++inserted. A numeric argument to @kbd{C-x &} indicates that you ++wish special formatting, and also acts as an index to pick an element of the ++list. If the list element is a string, the number to be inserted is ++formatted using the Emacs function @code{format}; otherwise the number ++is passed as an argument to the corresponding list element. ++@end table ++ ++In any source file, the Emacs command @kbd{C-x SPC} (@code{gud-break}) ++tells @value{DBG} to set a breakpoint on the source line point is on. ++ ++If you accidentally delete the source-display buffer, an easy way to get ++it back is to type the command @code{frame} in the @value{DBG} buffer, to ++request a frame display; when you run under Emacs, this recreates ++the source buffer if necessary to show you the context of the current ++frame. ++ ++The source files displayed in Emacs are in ordinary Emacs buffers ++which are visiting the source files in the usual way. You can edit ++the files with these buffers if you wish; but keep in mind that @value{DBG} ++communicates with Emacs in terms of line numbers. If you add or ++delete lines from the text, the line numbers that @value{DBG} knows cease ++to correspond properly with the code. ++ ++@node DDD ++@section Using @value{DBG} from @value{DDD} ++@cindex DDD ++ ++@value{DBG} support is rather new in @value{DDD}. As a programming ++language, @value{DBG} is not feature rich: there are no record ++structures or hash tables (yet), no pointers, package variable scoping ++or methods. So much of the data display and visualization features of ++@value{DDD} are disabled. ++ ++As with any scripting or interpreted language (e.g. Perl), one can't ++step by a single machine-language instruction. So the ddd Stepi/Nexti ++commands are disabled. ++ ++Some @value{BASH} settings are essential for @value{DDD} to work ++correctly. These settings with their correct values are: ++ ++@example ++set annotate 1 ++set prompt set prompt bashdb$_Dbg_less$_Dbg_greater$_Dbg_space ++@end example ++ ++@value{DDD} sets these values automatically when invoking ++@value{BASH}; if these values are changed, there may be some ++malfunctions. ++ ++Pay special attention when the prompt has extra angle brackets (a ++nested shell) or has any parenthesis (is in a subshell). Quitting may ++merely exit out of one of these nested (sub)shells rather than leave ++the program. ++ ++@node BASH Debugger Bugs ++@chapter Reporting Bugs ++@cindex bugs ++@cindex reporting bugs ++ ++Your bug reports play an essential role in making the BASH debugger reliable. ++ ++Reporting a bug may help you by bringing a solution to your problem, or it ++may not. But in any case the principal function of a bug report is to help ++the entire community by making the next version of @value{DBG} work better. Bug ++reports are your contribution to the maintenance of @value{DBG}. ++ ++In order for a bug report to serve its purpose, you must include the ++information that enables us to fix the bug. ++ ++@menu ++* Bug Criteria:: Have you found a bug? ++* Bug Reporting:: How to report bugs ++@end menu ++ ++@node Bug Criteria ++@section Have you found a bug? ++@cindex bug criteria ++ ++If you are not sure whether you have found a bug, here are some guidelines: ++ ++@itemize @bullet ++@cindex fatal signal ++@cindex debugger crash ++@cindex crash of debugger ++@item ++If the debugger gets a fatal signal, for any input whatever, that is a ++@value{DBG} bug. Reliable debuggers never crash. ++ ++@cindex error on valid input ++@item ++If @value{DBG} produces an error message for valid input, that is a ++bug. (Note that if you're cross debugging, the problem may also be ++somewhere in the connection to the target.) ++ ++@cindex invalid input ++@item ++If @value{DBG} does not produce an error message for invalid input, ++that is a bug. However, you should note that your idea of ++``invalid input'' might be our idea of ``an extension'' or ``support ++for traditional practice''. ++ ++@item ++If you are an experienced user of debugging tools, your suggestions ++for improvement of @value{DBG} are welcome in any case. ++@end itemize ++ ++@node Bug Reporting ++@section How to report bugs ++@cindex bug reports ++@cindex BASH debugger bugs, reporting ++ ++Submit bug reports for @value{DBG}. Until better mechanism is set up ++the prefered method is to submit them directly using ++@email{mailto:rocky@@panix.com}. There are also a bug tracking system ++and, help, developer and open discussion forums which can be found via ++@uref{http://http://sourceforge.net/forum/?group_id=37260}. ++ ++Although this code shares a lot with BASH, the debugger is of course ++not yet supported by the BASH team, so if there is a debugger problem ++reporting it via bashbug or to @email{bug-bash@@gnu.org} or posted to ++the newsgroup @code{gnu.bash.bug} probably won't help, unless you are ++sure the bug resides in the vanilla bash code and and show that. ++ ++The fundamental principle of reporting bugs usefully is this: ++@strong{report all the facts}. If you are not sure whether to state a ++fact or leave it out, state it! ++ ++Often people omit facts because they think they know what causes the ++problem and assume that some details do not matter. Thus, you might ++assume that the name of the variable you use in an example does not matter. ++Well, probably it does not, but one cannot be sure. Perhaps the bug is a ++stray memory reference which happens to fetch from the location where that ++name is stored in memory; perhaps, if the name were different, the contents ++of that location would fool the debugger into doing the right thing despite ++the bug. Play it safe and give a specific, complete example. That is the ++easiest thing for you to do, and the most helpful. ++ ++Keep in mind that the purpose of a bug report is to enable us to fix the ++bug. It may be that the bug has been reported previously, but neither ++you nor we can know that unless your bug report is complete and ++self-contained. ++ ++Sometimes people give a few sketchy facts and ask, ``Does this ring a ++bell?'' Those bug reports are useless, and we urge everyone to ++@emph{refuse to respond to them} except to chide the sender to report ++bugs properly. ++ ++To enable us to fix the bug, you should include all these things: ++ ++@itemize @bullet ++@item ++The version of @value{DBG}. @value{DBG} announces it if you start ++with no arguments; you can also print it at any time using @code{version} ++command. ++ ++Without this, we will not know whether there is any point in looking for ++the bug in the current version of @value{DBG}. ++ ++@item ++The type of machine you are using, and the operating system name and ++version number. ++ ++@item ++What compiler (and its version) was used to compile BASH---e.g. ++``@value{GCC}--3.1''. ++ ++@item ++The command arguments you gave the compiler to compile your example and ++observe the bug. For example, did you use @samp{-O}? To guarantee ++you will not omit something important, list them all. A copy of the ++Makefile (or the output from make) is sufficient. ++ ++If we were to try to guess the arguments, we would probably guess wrong ++and then we might not encounter the bug. ++ ++@item ++A complete input script, and all necessary source files, that will ++reproduce the bug. ++ ++@item ++A description of what behavior you observe that you believe is ++incorrect. For example, ``It gets a fatal signal.'' ++ ++Of course, if the bug is that @value{DBG} gets a fatal signal, then we ++will certainly notice it. But if the bug is incorrect output, we might ++not notice unless it is glaringly wrong. You might as well not give us ++a chance to make a mistake. ++ ++Even if the problem you experience is a fatal signal, you should still ++say so explicitly. Suppose something strange is going on, such as, your ++copy of @value{DBG} is out of synch, or you have encountered a bug in ++the C library on your system. (This has happened!) Your copy might ++crash and ours would not. If you told us to expect a crash, then when ++ours fails to crash, we would know that the bug was not happening for ++us. If you had not told us to expect a crash, then we would not be able ++to draw any conclusion from our observations. ++ ++@item ++If you wish to suggest changes to the @value{DBG} source, send us context ++diffs. If you even discuss something in the @value{DBG} source, refer to ++it by context, not by line number. ++ ++The line numbers in our development sources will not match those in your ++sources. Your line numbers would convey no useful information to us. ++ ++@end itemize ++ ++Here are some things that are not necessary: ++ ++@itemize @bullet ++@item ++A description of the envelope of the bug. ++ ++Often people who encounter a bug spend a lot of time investigating ++which changes to the input file will make the bug go away and which ++changes will not affect it. ++ ++This is often time consuming and not very useful, because the way we ++will find the bug is by running a single example under the debugger ++with breakpoints, not by pure deduction from a series of examples. ++We recommend that you save your time for something else. ++ ++Of course, if you can find a simpler example to report @emph{instead} ++of the original one, that is a convenience for us. Errors in the ++output will be easier to spot, running under the debugger will take ++less time, and so on. ++ ++However, simplification is not vital; if you do not want to do this, ++report the bug anyway and send us the entire test case you used. ++ ++@item ++A patch for the bug. ++ ++A patch for the bug does help us if it is a good one. But do not omit ++the necessary information, such as the test case, on the assumption that ++a patch is all we need. We might see problems with your patch and decide ++to fix the problem another way, or we might not understand it at all. ++ ++Sometimes with a program as complicated as @value{DBG} it is very hard to ++construct an example that will make the program follow a certain path ++through the code. If you do not send us the example, we will not be able ++to construct one, so we will not be able to verify that the bug is fixed. ++ ++And if we cannot understand what bug you are trying to fix, or why your ++patch should be an improvement, we will not install it. A test case will ++help us to understand. ++ ++@item ++A guess about what the bug is or what it depends on. ++ ++Such guesses are usually wrong. Even we cannot guess right about such ++things without first using the debugger to find the facts. ++@end itemize ++ ++@node History and Acknowledgments ++@chapter History and Acknowledgments ++ ++The suggestion for a debugger for a Bourne-like shell came from the book ++``Learning the Korn Shell'', by Bill Rosenblatt Copyright (C) 1993 by ++O'Reilly and Associates, Inc. Others such as Cigy Cyriac, Chet Ramey, ++Rocky Bernstein, and Gary V. Vaughan expanded and improved on that. ++ ++However Bourne-Shell debuggers rely on a signal mechanism ++(@code{SIGDEBUG}) to call a debugger routine. In the Korn shell as ++well as @sc{bash} in versions prior to 2.05, there was a fundamental ++flaw: the routine that you registered in the trap, got called ++@emph{after} the statement was executed. It takes little imagination ++to realize that this is a bit too late to find and correct errors, ++especially if the offending command happens to do serious damage like ++remove filesystems or reboot a server. As a horrible hack, these ++debuggers added one to the line number that was just executed on the ++wishful thinking that this would then be the line of next statement to ++execute. Sometimes this was correct, but it was too often wrong, such ++as in loops and conditionals, comments, or commands that are continued ++on the next line. ++ ++Another failing of these debuggers was the inability to debug into ++functions or into sourced files, provide a stack trace, dynamically ++skip a statement to be run, unconditionally trace into a function, or ++stop when a subroutine or a sourced file completed. In truth, the crux ++of the problem lay in debugging support in BASH. Given that there was ++limited bash debugging support, it is not surprising that these ++debuggers could not do any of the things listed above and could debug ++only a single source file: lines could be listed only from a single ++text, breakpoints were set into the text which was in fact a copy of ++the script name prepended with debugger routines. ++ ++In version 2.04 of BASH, Rocky Bernstein started hacking on BASH to ++add call-stack information, source file information, allow for ++debugging into functions and for reporting line numbers in functions ++as relative to the file rather than the beginning of a function whose ++origin line number was not accessible from BASH. He started changing ++the user commands in bashdb to be like other more-advanced debuggers, ++in particular @code{perl5db} and @code{gdb}. However he gave up on ++this project when realizing that stopping before a line was crucial. A ++patch for this was nontrivial and wildly changed ++semantics. Furthermore the chance of getting his other patches into ++BASH was was not going to happen in version 2.04. ++ ++In version 2.05, the fundamental necessary change to the semantics of ++@code{SIGDEBUG} trap handling (suggested at least two years earlier) ++was made. Also, version 2.05 changed the line-number reporting in a ++function to be relative to the beginning of the file rather than the ++beginning of a function---sometimes. Rocky then picked up where he ++left off and this then became this debugger. A complete rewrite of the ++debugger, some of which started in 2.04 was undertaken. Debugger ++internals were changed to support multiple file names, save and ++restore the calling environment (such as variables @code{$1} and ++@code{$?}) and install debugger signal handlers. Work was also done on ++the BASH in conjunction with the debugger to save stack trace ++information, provide a means for stopping after a routine finished and ++so on. And a number of changes were made to BASH just to improve the ++accuracy of the line number reporting which is crucial in a debugger. ++ ++This documentation was modified from the GNU Debugger (GDB) Reference ++manual. ++ ++@quotation ++Additions to this section are particularly welcome. If you or your ++friends (or enemies, to be evenhanded) have been unfairly omitted from ++this list, we would like to add your names! ++@end quotation ++ ++The following have contributed directly or indrectly to bashdb: ++ ++Rocky Bernstein (initial full-featured bashdb with stack tracing and ++multi-file support) ++ ++Masatake YAMATO (help to merge Rocky's hack to the official bash source tree) ++ ++Bill Rosenblatt (kshdb), ++Michael Loukides (kshdb), ++Cigy Cyriac (proto bashdb), ++Chet Ramey (proto bashdb), ++and ++Gary V. Vaughan (proto bashdb). ++ ++Authors of per5ldb: ++ ++Ray Lischner, ++Johan Vromans, ++and ++Ilya Zakharevich. ++ ++Authors of GDB: ++ ++Richard Stallman, ++Andrew Cagney, ++Jim Blandy, ++Jason Molenda, ++Stan Shebs, ++Fred Fish, ++Stu Grossman, ++John Gilmore, ++Jim Kingdon, ++and ++Randy Smith (to name just a few). ++ ++Authors of GUD: ++ ++Eric S. Raymond. ++ ++@c The readline documentation is distributed with the readline code ++@c and consists of the two following files: ++@c rluser.texinfo ++@c inc-hist.texinfo ++@c Use -I with makeinfo to point to the appropriate directory, ++@c environment var TEXINPUTS with TeX. ++@c @include rluser.texinfo ++@c @include hsuser.texinfo ++ ++ ++@node Formatting Documentation ++@appendix Formatting Documentation ++ ++@ifset FINISHED ++@cindex @value{DBG} reference card ++@cindex reference card ++The @value{DBG} 4 release includes an already-formatted reference card, ready ++for printing with PostScript or Ghostscript, in the @file{gdb} ++subdirectory of the main source directory@footnote{In ++@file{gdb-@value{BASHRELEASE}/gdb/refcard.ps} of the version @value{BASHRELEASE} ++release.}. If you can use PostScript or Ghostscript with your printer, ++you can print the reference card immediately with @file{refcard.ps}. ++ ++The release also includes the source for the reference card. You ++can format it, using @TeX{}, by typing: ++ ++@example ++make refcard.dvi ++@end example ++ ++The @value{DBG} reference card is designed to print in @dfn{landscape} ++mode on US ``letter'' size paper; ++that is, on a sheet 11 inches wide by 8.5 inches ++high. You will need to specify this form of printing as an option to ++your @sc{dvi} output program. ++ ++@cindex documentation ++ ++All the documentation for @value{DBG} comes as part of the machine-readable ++distribution. The documentation is written in Texinfo format, which is ++a documentation system that uses a single source file to produce both ++on-line information and a printed manual. You can use one of the Info ++formatting commands to create the on-line version of the documentation ++and @TeX{} (or @code{texi2roff}) to typeset the printed version. ++ ++@value{DBG} includes an already formatted copy of the on-line Info ++version of this manual in the @file{gdb} subdirectory. The main Info ++file is @file{gdb-@value{BASHRELEASE}/gdb/gdb.info}, and it refers to ++subordinate files matching @samp{gdb.info*} in the same directory. If ++necessary, you can print out these files, or read them with any editor; ++but they are easier to read using the @code{info} subsystem in @sc{gnu} ++Emacs or the standalone @code{info} program, available as part of the ++@sc{gnu} Texinfo distribution. ++ ++If you want to format these Info files yourself, you need one of the ++Info formatting programs, such as @code{texinfo-format-buffer} or ++@code{makeinfo}. ++ ++If you have @code{makeinfo} installed, and are in the top level ++@value{DBG} source directory (@file{gdb-@value{BASHRELEASE}}, in the case of ++version @value{BASHRELEASE}), you can make the Info file by typing: ++ ++@example ++cd gdb ++make gdb.info ++@end example ++ ++If you want to typeset and print copies of this manual, you need @TeX{}, ++a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the ++Texinfo definitions file. ++ ++@TeX{} is a typesetting program; it does not print files directly, but ++produces output files called @sc{dvi} files. To print a typeset ++document, you need a program to print @sc{dvi} files. If your system ++has @TeX{} installed, chances are it has such a program. The precise ++command to use depends on your system; @kbd{lpr -d} is common; another ++(for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may ++require a file name without any extension or a @samp{.dvi} extension. ++ ++@TeX{} also requires a macro definitions file called ++@file{texinfo.tex}. This file tells @TeX{} how to typeset a document ++written in Texinfo format. On its own, @TeX{} cannot either read or ++typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB ++and is located in the @file{gdb-@var{version-number}/texinfo} ++directory. ++ ++If you have @TeX{} and a @sc{dvi} printer program installed, you can ++typeset and print this manual. First switch to the the @file{gdb} ++subdirectory of the main source directory (for example, to ++@file{gdb-@value{BASHRELEASE}/gdb}) and type: ++ ++@example ++make gdb.dvi ++@end example ++ ++Then give @file{gdb.dvi} to your @sc{dvi} printing program. ++@end ifset ++ ++@node Installing bashdb ++@appendix Installing @value{DBG} ++@cindex configuring @value{DBG} ++@cindex installation ++ ++@ifset FINISHED ++@value{DBG} comes with a @code{configure} script that automates the process ++of preparing @value{DBG} for installation; you can then use @code{make} to ++build the @code{gdb} program. ++@iftex ++@c irrelevant in info file; it's as current as the code it lives with. ++@footnote{If you have a more recent version of @value{DBG} than @value{BASHRELEASE}, ++look at the @file{README} file in the sources; we may have improved the ++installation procedures since publishing this manual.} ++@end iftex ++ ++The @value{DBG} distribution includes all the source code you need for ++@value{DBG} in a single directory, whose name is usually composed by ++appending the version number to @samp{gdb}. ++ ++For example, the @value{DBG} version @value{BASHRELEASE} distribution is in the ++@file{gdb-@value{BASHRELEASE}} directory. That directory contains: ++ ++@table @code ++@item gdb-@value{BASHRELEASE}/configure @r{(and supporting files)} ++script for configuring @value{DBG} and all its supporting libraries ++ ++@item gdb-@value{BASHRELEASE}/gdb ++the source specific to @value{DBG} itself ++ ++@item gdb-@value{BASHRELEASE}/bfd ++source for the Binary File Descriptor library ++ ++@item gdb-@value{BASHRELEASE}/include ++@sc{gnu} include files ++ ++@item gdb-@value{BASHRELEASE}/libiberty ++source for the @samp{-liberty} free software library ++ ++@item gdb-@value{BASHRELEASE}/opcodes ++source for the library of opcode tables and disassemblers ++ ++@item gdb-@value{BASHRELEASE}/readline ++source for the @sc{gnu} command-line interface ++ ++@item gdb-@value{BASHRELEASE}/glob ++source for the @sc{gnu} filename pattern-matching subroutine ++ ++@item gdb-@value{BASHRELEASE}/mmalloc ++source for the @sc{gnu} memory-mapped malloc package ++@end table ++ ++The simplest way to configure and build @value{DBG} is to run @code{configure} ++from the @file{gdb-@var{version-number}} source directory, which in ++this example is the @file{gdb-@value{BASHRELEASE}} directory. ++ ++First switch to the @file{gdb-@var{version-number}} source directory ++if you are not already in it; then run @code{configure}. Pass the ++identifier for the platform on which @value{DBG} will run as an ++argument. ++ ++For example: ++ ++@example ++cd gdb-@value{BASHRELEASE} ++./configure @var{host} ++make ++@end example ++ ++@noindent ++where @var{host} is an identifier such as @samp{sun4} or ++@samp{decstation}, that identifies the platform where @value{DBG} will run. ++(You can often leave off @var{host}; @code{configure} tries to guess the ++correct value by examining your system.) ++ ++Running @samp{configure @var{host}} and then running @code{make} builds the ++@file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty} ++libraries, then @code{gdb} itself. The configured source files, and the ++binaries, are left in the corresponding source directories. ++ ++@need 750 ++@code{configure} is a Bourne-shell (@code{/bin/sh}) script; if your ++system does not recognize this automatically when you run a different ++shell, you may need to run @code{sh} on it explicitly: ++ ++@example ++sh configure @var{host} ++@end example ++ ++If you run @code{configure} from a directory that contains source ++directories for multiple libraries or programs, such as the ++@file{gdb-@value{BASHRELEASE}} source directory for version @value{BASHRELEASE}, @code{configure} ++creates configuration files for every directory level underneath (unless ++you tell it not to, with the @samp{--norecursion} option). ++ ++You can run the @code{configure} script from any of the ++subordinate directories in the @value{DBG} distribution if you only want to ++configure that subdirectory, but be sure to specify a path to it. ++ ++For example, with version @value{BASHRELEASE}, type the following to configure only ++the @code{bfd} subdirectory: ++ ++@example ++@group ++cd gdb-@value{BASHRELEASE}/bfd ++../configure @var{host} ++@end group ++@end example ++ ++You can install @code{@value{DBG}} anywhere; it has no hardwired paths. ++However, you should make sure that the shell on your path (named by ++the @samp{SHELL} environment variable) is publicly readable. Remember ++that @value{DBG} uses the shell to start your script---some systems refuse to ++let @value{DBG} debug child processes whose programs are not readable. ++ ++@menu ++* Separate Objdir:: Compiling @value{DBG} in another directory ++* Config Names:: Specifying names for hosts and targets ++* Configure Options:: Summary of options for configure ++@end menu ++ ++@node Separate Objdir ++@section Compiling @value{DBG} in another directory ++ ++If you want to run @value{DBG} versions for several host or target machines, ++you need a different @code{gdb} compiled for each combination of ++host and target. @code{configure} is designed to make this easy by ++allowing you to generate each configuration in a separate subdirectory, ++rather than in the source directory. If your @code{make} program ++handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running ++@code{make} in each of these directories builds the @code{gdb} ++program specified there. ++ ++To build @code{gdb} in a separate directory, run @code{configure} ++with the @samp{--srcdir} option to specify where to find the source. ++(You also need to specify a path to find @code{configure} ++itself from your working directory. If the path to @code{configure} ++would be the same as the argument to @samp{--srcdir}, you can leave out ++the @samp{--srcdir} option; it is assumed.) ++ ++For example, with version @value{BASHRELEASE}, you can build @value{DBG} in a ++separate directory for a Sun 4 like this: ++ ++@example ++@group ++cd gdb-@value{BASHRELEASE} ++mkdir ../gdb-sun4 ++cd ../gdb-sun4 ++../gdb-@value{BASHRELEASE}/configure sun4 ++make ++@end group ++@end example ++ ++When @code{configure} builds a configuration using a remote source ++directory, it creates a tree for the binaries with the same structure ++(and using the same names) as the tree under the source directory. In ++the example, you'd find the Sun 4 library @file{libiberty.a} in the ++directory @file{gdb-sun4/libiberty}, and @value{DBG} itself in ++@file{gdb-sun4/gdb}. ++ ++One popular reason to build several @value{DBG} configurations in separate ++directories is to configure @value{DBG} for cross-compiling (where ++@value{DBG} runs on one machine---the @dfn{host}---while debugging ++programs that run on another machine---the @dfn{target}). ++You specify a cross-debugging target by ++giving the @samp{--target=@var{target}} option to @code{configure}. ++ ++When you run @code{make} to build a program or library, you must run ++it in a configured directory---whatever directory you were in when you ++called @code{configure} (or one of its subdirectories). ++ ++The @code{Makefile} that @code{configure} generates in each source ++directory also runs recursively. If you type @code{make} in a source ++directory such as @file{gdb-@value{BASHRELEASE}} (or in a separate configured ++directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{BASHRELEASE}}), you ++will build all the required libraries, and then build GDB. ++ ++When you have multiple hosts or targets configured in separate ++directories, you can run @code{make} on them in parallel (for example, ++if they are NFS-mounted on each of the hosts); they will not interfere ++with each other. ++ ++@node Config Names ++@section Specifying names for hosts and targets ++ ++The specifications used for hosts and targets in the @code{configure} ++script are based on a three-part naming scheme, but some short predefined ++aliases are also supported. The full naming scheme encodes three pieces ++of information in the following pattern: ++ ++@example ++@var{architecture}-@var{vendor}-@var{os} ++@end example ++ ++For example, you can use the alias @code{sun4} as a @var{host} argument, ++or as the value for @var{target} in a @code{--target=@var{target}} ++option. The equivalent full name is @samp{sparc-sun-sunos4}. ++ ++The @code{configure} script accompanying @value{DBG} does not provide ++any query facility to list all supported host and target names or ++aliases. @code{configure} calls the Bourne shell script ++@code{config.sub} to map abbreviations to full names; you can read the ++script, if you wish, or you can use it to test your guesses on ++abbreviations---for example: ++ ++@smallexample ++% sh config.sub i386-linux ++i386-pc-linux-gnu ++% sh config.sub alpha-linux ++alpha-unknown-linux-gnu ++% sh config.sub hp9k700 ++hppa1.1-hp-hpux ++% sh config.sub sun4 ++sparc-sun-sunos4.1.1 ++% sh config.sub sun3 ++m68k-sun-sunos4.1.1 ++% sh config.sub i986v ++Invalid configuration `i986v': machine `i986v' not recognized ++@end smallexample ++ ++@noindent ++@code{config.sub} is also distributed in the @value{DBG} source ++directory (@file{gdb-@value{BASHRELEASE}}, for version @value{BASHRELEASE}). ++ ++@node Configure Options ++@section @code{configure} options ++ ++Here is a summary of the @code{configure} options and arguments that ++are most often useful for building @value{DBG}. @code{configure} also has ++several other options not listed here. @inforef{What Configure ++Does,,configure.info}, for a full explanation of @code{configure}. ++ ++@example ++configure @r{[}--help@r{]} ++ @r{[}--prefix=@var{dir}@r{]} ++ @r{[}--exec-prefix=@var{dir}@r{]} ++ @r{[}--srcdir=@var{dirname}@r{]} ++ @r{[}--norecursion@r{]} @r{[}--rm@r{]} ++ @r{[}--target=@var{target}@r{]} ++ @var{host} ++@end example ++ ++@noindent ++You may introduce options with a single @samp{-} rather than ++@samp{--} if you prefer; but you may abbreviate option names if you use ++@samp{--}. ++ ++@table @code ++@item --help ++Display a quick summary of how to invoke @code{configure}. ++ ++@item --prefix=@var{dir} ++Configure the source to install programs and files under directory ++@file{@var{dir}}. ++ ++@item --exec-prefix=@var{dir} ++Configure the source to install programs under directory ++@file{@var{dir}}. ++ ++@c avoid splitting the warning from the explanation: ++@need 2000 ++@item --srcdir=@var{dirname} ++@strong{Warning: using this option requires @sc{gnu} @code{make}, or another ++@code{make} that implements the @code{VPATH} feature.}@* ++Use this option to make configurations in directories separate from the ++@value{DBG} source directories. Among other things, you can use this to ++build (or maintain) several configurations simultaneously, in separate ++directories. @code{configure} writes configuration specific files in ++the current directory, but arranges for them to use the source in the ++directory @var{dirname}. @code{configure} creates directories under ++the working directory in parallel to the source directories below ++@var{dirname}. ++ ++@item --norecursion ++Configure only the directory level where @code{configure} is executed; do not ++propagate configuration to subdirectories. ++ ++@item --target=@var{target} ++Configure @value{DBG} for cross-debugging programs running on the specified ++@var{target}. Without this option, @value{DBG} is configured to debug ++programs that run on the same machine (@var{host}) as @value{DBG} itself. ++ ++There is no convenient way to generate a list of all available targets. ++ ++@item @var{host} @dots{} ++Configure @value{DBG} to run on the specified @var{host}. ++ ++There is no convenient way to generate a list of all available hosts. ++@end table ++ ++There are many other options available as well, but they are generally ++needed for special purposes only. ++@end ifset ++ ++@include gpl.texi ++ ++@include fdl.texi ++ ++@node Function Index ++@unnumbered Function Index ++@printindex fn ++ ++@node Command Index ++@unnumbered Command Index ++@printindex ky ++ ++@node Variable Index ++@unnumbered Variable Index ++@printindex vr ++ ++@node General Index ++@unnumbered General Index ++@printindex cp ++ ++@tex ++% I think something like @colophon should be in texinfo. In the ++% meantime: ++\long\def\colophon{\hbox to0pt{}\vfill ++\centerline{The body of this manual is set in} ++\centerline{\fontname\tenrm,} ++\centerline{with headings in {\bf\fontname\tenbf}} ++\centerline{and examples in {\tt\fontname\tentt}.} ++\centerline{{\it\fontname\tenit\/},} ++\centerline{{\bf\fontname\tenbf}, and} ++\centerline{{\sl\fontname\tensl\/}} ++\centerline{are used for emphasis.}\vfill} ++\page\colophon ++% Blame: doc@cygnus.com, 1991. ++@end tex ++ ++@bye +diff -ur --unidirectional-new-file --exclude CVS --exclude=.cvsignore bash-2.05b/debugger/doc/bashdb.texi~ cvs/debugger/doc/bashdb.texi~ +--- bash-2.05b/debugger/doc/bashdb.texi~ 1970-01-01 01:00:00.000000000 +0100 ++++ cvs/debugger/doc/bashdb.texi~ 2003-08-30 07:31:23.000000000 +0200 +@@ -0,0 +1,3752 @@ ++\input texinfo @c -*-texinfo-*- ++@c Copyright 2002, 2003 ++@c Rocky Bernstein for the Free Software Foundation ++@c ++@c TODO: ++@c - add examples for commands ++@c - clean up/improve sample session ++@c - help text is inaccurate and formatted too much to right. ++@c ++ ++@c Sets version and release names and dates. Frees us from changing ++@c this file when a new release comes along. ++ ++@c %**start of header ++@c makeinfo ignores cmds prev to setfilename, so its arg cannot make use ++@c of @set vars. However, you can override filename with makeinfo -o. ++@setfilename bashdb.info ++@c ++@c Name of Bash program. Used in running text. ++@set BASH @acronym{BASH} ++ ++@c ++@c Name of debugger program. Used also for prompt string. ++@set DBG the @value{BASH} debugger ++@set dBGP The @value{BASH} debugger ++@set DDD @acronym{DDD} ++ ++@settitle @value{BASH} Debugger ++@setchapternewpage odd ++@setcontentsaftertitlepage ++ ++@c %**end of header ++ ++@include version.texi ++ ++@c A simple macro for optional variables. ++@macro ovar{varname} ++@r{[}@var{\varname\}@r{]} ++@end macro ++ ++@macro DDD {} ++@acronym{DDD} ++@end macro ++ ++@macro BASH {} ++@acronym{@value{BASH}} ++@end macro ++ ++@macro DBG {} ++@value{DBG} ++@end macro ++ ++@iftex ++@c @smallbook ++@c @cropmarks ++@end iftex ++ ++@finalout ++ ++@c readline appendices use @vindex, @findex and @ftable, ++@c annotate.texi and gdbmi use @findex. ++@c @syncodeindex vr cp ++@c @syncodeindex fn cp ++ ++@c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER. ++ ++@c This is a dir.info fragment to support semi-automated addition of ++@c manuals to an info tree. ++@dircategory Programming & development tools. ++@direntry ++* @value{DBG}: (@value{DBG}). The @sc{bash} debugger. ++@end direntry ++ ++@ifinfo ++This file documents the @sc{bash} debugger @value{BASH}. ++ ++ ++This is the @value{EDITION} Edition, @value{UPDATED-MONTH}, ++of @cite{Debugging with BASHDB: the @sc{gnu} Source-Level Debugger} ++for REBASH ++ ++Copyright (C) 2002,2003 Rocky Bernstein for the Free Software Foundation. ++ ++Permission is granted to copy, distribute and/or modify this document ++under the terms of the GNU Free Documentation License, Version 1.1 or ++@ifset DEBIANHASBECOMEREASONABLE ++@c From Matthias Klose a Debian maintainer on ++@c Sat, 23 Aug 2003 14:24:44 +0200 ++@c ++@c I personally see the invariant sections as the thing in the ++@c GFDL, which hinders me in uploading the package to the archives. ++@c I don't have any problem, if some other Debian developer makes a ++@c bashdb package built from separate sources. ++@c ++@c I am aware that Debian ships other packages containing documentation ++@c covered by the GFDL (and one of them for which I do the packaging as ++@c well), but I won't add a new package, which I maintain. So before an ++@c upload of a bashdb package built from the bash sources either ++@c ++@c ++@c - Debian has a position on the GFDL, which allows inclusion ++@c ++@c - the bashdb manual does not have invariant sections, or is ++@c relicensed, or dual licensed. ++any later version published by the Free Software Foundation; with the ++Invariant Sections being ``Free Software'' and ``Free Software Needs ++Free Documentation'', with the Front-Cover Texts being ``A GNU ++Manual,'' and with the Back-Cover Texts as in (a) below. ++ ++(a) The Free Software Foundation's Back-Cover Text is: ``You have ++freedom to copy and modify this GNU Manual, like GNU software. Copies ++published by the Free Software Foundation raise funds for GNU ++development.'' ++@else ++any later version published by the Free Software Foundation. ++@end ifset ++@end ifinfo ++ ++@titlepage ++@title Debugging with BASHDB ++@sp 1 ++@subtitle @value{EDITION} Edition, for BASH ++@subtitle @value{UPDATED-MONTH} ++@author Rocky Bernstein ++@page ++@tex ++{\parskip=0pt ++\hfill (Send bugs and comments on bashdb to bug-bashdb\@sourceforge.net.)\par ++\hfill {\it Debugging with BASH}\par ++\hfill \TeX{}info \texinfoversion\par ++} ++@end tex ++ ++@vskip 0pt plus 1filll ++Copyright @copyright{} 2002,2003 Rocky Bernstein for the Free Software ++Foundation. ++ ++Permission is granted to copy, distribute and/or modify this document ++under the terms of the GNU Free Documentation License, Version 1.1 or ++@ifset DEBIANHASBECOMEREASONABLE ++@c From Matthias Klose a Debian maintainer on ++@c Sat, 23 Aug 2003 14:24:44 +0200 ++@c ++@c I personally see the invariant sections as the thing in the ++@c GFDL, which hinders me in uploading the package to the archives. ++@c I don't have any problem, if some other Debian developer makes a ++@c bashdb package built from separate sources. ++@c ++@c I am aware that Debian ships other packages containing documentation ++@c covered by the GFDL (and one of them for which I do the packaging as ++@c well), but I won't add a new package, which I maintain. So before an ++@c upload of a bashdb package built from the bash sources either ++@c ++@c ++@c - Debian has a position on the GFDL, which allows inclusion ++@c ++@c - the bashdb manual does not have invariant sections, or is ++@c relicensed, or dual licensed. ++any later version published by the Free Software Foundation; with the ++Invariant Sections being ``Free Software'' and ``Free Software Needs ++Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,'' ++and with the Back-Cover Texts as in (a) below. ++ ++(a) The Free Software Foundation's Back-Cover Text is: ``You have ++freedom to copy and modify this GNU Manual, like GNU software. Copies ++published by the Free Software Foundation raise funds for GNU ++development.'' ++else ++any later version published by the Free Software Foundation. ++@end ifset ++@end titlepage ++@page ++ ++@ifnottex ++@node Top, Summary, (dir), (dir) ++ ++@top Debugging with @DBG ++ ++This file describes @value{DBG}, the @sc{bash} symbolic debugger. ++ ++This is the @value{EDITION} Edition, @value{UPDATED-MONTH}, for REBASH. ++ ++Copyright (C) 2002,2003 Rocky Bernstein ++ ++@menu ++* Summary:: Overview of Debugger with a sample session ++ ++* Invocation:: Getting in and out ++* Running:: Script setup inside the BASH debugger ++* Debugger Command Reference:: BASH debugger command reference ++ ++* Front Ends:: Using @DBG from a front-end user interface ++ ++* BASH Debugger Bugs:: Reporting bugs ++* History and Acknowledgments:: History and Acknowledgments ++ ++Appendices ++* Formatting Documentation:: How to format and print @DBG documentation ++ ++* Installing bashdb:: Installing bashdb ++* Copying:: GNU General Public License says ++ how you can copy and share bashdb ++* GNU Free Documentation License:: The license for this documentation ++ ++Indexes (nodes containing large menus) ++* Function Index:: An item for each function name. ++* Command Index:: An item for each command name. ++* Variable Index:: An item for each documented variable. ++* General Index:: An item for each concept. ++@end menu ++ ++@end ifnottex ++ ++@contents ++ ++@node Summary ++@chapter Summary of the BASH Debugger ++ ++The purpose of a debugger such as @DBG is to allow you to see what is ++going on ``inside'' a bash script while it executes. ++ ++@DBG can do four main kinds of things (plus other things in support of ++these) to help you catch bugs in the act: ++ ++@itemize @bullet ++@item ++Start your script, specifying anything that might affect its behavior. ++ ++@item ++Make your script stop on specified conditions. ++ ++@item ++Examine what has happened, when your script has stopped. ++ ++@item ++Change things in your script, so you can experiment with correcting the ++effects of one bug and go on to learn about another. ++@end itemize ++ ++Although you can use the BASH debugger to debug scripts written in ++BASH, it can also be used just as a front-end for learning more ++about programming in BASH. As an additional aid, the debugger can ++be used within the context of an existing script with its functions ++and variables that have already been initialized; fragments of the ++existing can be experimented with by entering them inside the debugger. ++ ++@menu ++* Sample Session:: A Sample BASH Debugger session ++* Free Software:: Freely redistributable software ++@end menu ++ ++@node Sample Session ++@section A Sample BASH Debugger Session ++ ++You can use this manual at your leisure to read all about @value{DBG}. ++However, a handful of commands are enough to get started using the ++debugger. This chapter illustrates those commands. ++ ++@iftex ++In this sample session, we emphasize user input like this: @b{input}, ++to make it easier to pick out from the surrounding output. ++@end iftex ++ ++Below we will debug a script that contains a function to compute the ++factorial of a number: fact(0) is 1 and fact(n) is n*fact(n-1). ++ ++@smallexample ++$ @b{bashdb -L . fact.sh} ++Bourne-Again Shell Debugger, release bash-@value{BASHRELEASE} ++Copyright 2002,2003 Rocky Bernstein ++This is free software, covered by the GNU General Public License, and you are ++welcome to change it and/or distribute copies of it under certain conditions. ++ ++(fact.sh:9): ++9: echo fact 0 is: `fact 0` ++bashdb<0> @b{-} ++ 1: #!/usr/local/bin/bash ++ 2: fact() @{ ++ 3: ((n==0)) && echo 1 && return ++ 4: ((nm1=n-1)) ++ 5: ((result=n*`fact $nm1`)) ++ 6: return $result ++ 7: @} ++ 8: ++ 9:==> echo fact 0 is: `fact 0` ++bashdb<1> @b{l} ++ 10: echo fact 3 is: $(fact 3) ++@end smallexample ++ ++@noindent ++ ++The command invocation uses the option ``-L .'' Here we assume that ++the @command{bashdb} script and the debugger files are in the same ++location. If you are running from the source code, this will be the ++case. However if bashdb has been installed this probably won't be true ++and here you probably don't need to use ``-L .'' Instead you would ++type simply @code{bashdb fact.sh}. ++ ++The first debugger command we gave @kbd{-}, we listed a window of ++lines @emph{before} where we were executing. Because the window, 10 ++lines, is larger than the number of lines to the top of the file we ++printed only 9 lines here. The next command list starting from the ++current line and again wants to print 10 lines but because there are ++only one remaining line, that is what is printed. ++ ++@smallexample ++bashdb<2> @b{step} ++(fact.sh:9): ++9: echo fact 0 is: `fact 0` ++bashdb<3> @b{@key{RET}} ++2: fact() @{ ++bashdb<4> @b{@key{RET}} ++3: ((n==0)) && echo 1 && return ++bashdb<5> @b{print $n} ++ ++@end smallexample ++ ++Ooops... The variable @kbd{n} isn't initialized. ++ ++The first @kbd{step} command steps the script one instruction. It may ++seem odd that the line printed is exactly the same one as before. What ++has happened though is that we've ``stepped'' into the subshell needed ++to run @kbd{`fact 0`}; we haven't however started running anything ++inside that subshell yet though---the function call. ++ ++The first @kbd{step} command steps the script one instruction; it ++didn't advance the line number at all. That is because we were ++stopping before the command substitution or backtick is to take ++place. The second command we entered was just hitting the return key; ++bashdb remembers that you entered @code{step} previously, so it runs ++the step rather than @kbd{next}, the other alternative when you hit ++@key{RET}. Step one more instruction and we are just before running ++the first statement of the function. ++ ++Next, we print the value of the variable @kbd{n}. Notice we need to add ++a preceding dollar simple to get the substitution or value of n. As we ++will see later, if the @kbd{pe} command were used this would not be ++necessary. ++ ++We now modify the file to add an assignment to local variable @kbd{n} and ++restart. ++ ++@smallexample ++bashdb<6> @b{restart} ++Restarting with: /usr/local/bin/bashdb -L . fact.sh ++(fact.sh:10): ++10: echo fact 0 is: `fact 0` ++bashdb<0> @b{list 1} ++ 1: #!/usr/local/bin/bash ++ 2: fact() @{ ++ 3: local -i n=$@{1:0@} ++ 4: ((n==0)) && echo 1 && return ++ 5: ((nm1=n-1)) ++ 6: ((result=n*`fact $nm1`)) ++ 7: return $result ++ 8: @} ++ 9: ++ 10:==> echo fact 0 is: `fact 0` ++bashdb<1> @b{s 3} ++(fact.sh:3): ++3: local -i n=$@{1:0@} ++bashdb<2> @b{step} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++bashdb<3> @b{print $n} ++print $n ++0 ++@end smallexample ++ ++@noindent ++This time we use the @code{list} debugger command to list the lines in ++the file. From before we know it takes three @code{steps} commands ++before we get into the fact() function, so we add a count onto the ++@code{step} command. Notice we abbreviate @code{step} with @code{s}; ++we could have done likewise and abbreviated @code{list} with @code{l}. ++ ++@smallexample ++bashdb<4> @b{@key{RET}} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++echo 1 ++bashdb<5> @b{@key{RET}} ++(fact.sh:4): ++4: ((n==0)) && echo 1 && return ++return ++@end smallexample ++ ++@noindent ++Again we just use @key{RET} to repeat the last @code{step} ++commands. And again the fact that we are staying on the same line 4 ++means that the next condition in the line is about to be ++executed. Notice that we see the command (@code{echo 1} or ++@code{return}) listed when we stay on the same line which has multiple ++stopping points in it. Given the information above, we know that the ++value echo'ed on return will be 1. ++ ++@smallexample ++bashdb<6> @b{@key{RET}} ++fact 0 is: 1 ++(fact.sh:12): ++12: echo fact 3 is: $(fact 3) ++bashdb<7> @b{break 5} ++Breakpoint 1 set in file fact.sh, line 5. ++bashdb<8> @b{continue} ++@end smallexample ++ ++@noindent ++We saw that we could step with a count into the function ++fact(). However above took another approach: we set a stopping point or ++``breakpoint'' at line 5 to get us a little ways into the fact() ++subroutine. Just before line 5 is to executed, we will get back into ++the debugger. The @code{continue} command just resumes execution until ++the next stopping point which has been set up in some way. ++ ++@smallexample ++(fact.sh:5): ++5: ((nm1=n-1)) ++Breakpoint 1 hit(1 times). ++bashdb<8> @b{x n-1} ++2 ++bashdb<9> @b{s} ++(fact.sh:5): ++6: ((result=n*`fact $nm1`)) ++bashdb<10> @b{c} ++fact.sh: line 6: ((: result=n*: syntax error: operand expected (error token is "*") ++bashdb<7> @b{R} ++Restarting with: bash --debugger fact.sh ++11: echo fact 0 is: `fact 0` ++bashdb<0> @b{l fact} ++ 2: fact () ++ 3: @{ ++ 4: local -i n=$@{1:0@}; ++ 5: (( "n==0" )) && echo 1 && return; ++ 6: (( nm1=n-1 )); ++ 7: ((fact_nm1=`fact $nm1`)) ++ 8: (( "result=n*fact_nm1" )); ++ 9: echo $result ++10: @} ++@end smallexample ++ ++@noindent ++In addition to listing by line numbers, we can also list giving a ++function name. Below, instead of setting a breakpoint at line 5 and ++running ``continue'' as we did above, we try something slightly shorter ++and slightly different. We give the line number on the ``continue'' ++statement. This is a little different in that a one-time break is made ++on line 5. Once that statement is reached the breakpoint is removed. ++ ++@smallexample ++bashdb<1> @b{continue 5} ++One-time breakpoint 1 set in file fact.sh, line 5. ++fact 0 is: 1 ++(fact.sh:5): ++5: ((nm1=n-1)) ++bashdb<2> @b{s} ++6: ((fact_nm1=`fact $nm1`)) ++bashdb<2> @b{s} ++2: fact() @{ ++bashdb<3> @b{T} ++->0 in file `fact.sh' at line 2 ++##1 fact("3") called from file `fact.sh' at line 12 ++##2 source("fact.sh") called from file `/usr/local/bin/bashdb' at line 154 ++##3 main("fact.sh") called from file `/usr/local/bin/bashdb' at line 0 ++bashdb<4> @b{c} ++fact 3 is: 6 ++Debugged program terminated normally. Use q to quit or R to restart. ++@end smallexample ++ ++@noindent ++When we stop at line 5 above, we have already run fact(0) and output ++the correct results. The output from the program ``fact 0 is: 1'' is ++intermixed with the debugger output. The @code{T} command above ++requests call stack output and this confirms that we are not in the ++fact(0) call but in the fact(3) call. There are 4 lines listed in the ++stack trace even though there is just one call from the main ++program. The top line of the trace doesn't really represent a call, ++it's just where we currently are in the program. That last line is an ++artifact of invoking bash from the bashdb script rather than running ++@code{bash --debugger}. ++ ++The last message in the output above @samp{Debugged program exited ++normally.} is from @value{DBG}; it indicates script has finished ++executing. We can end our @value{DDGP} session with the @value{GDBP} ++@code{quit} command. ++ ++Above we did our debugging session on the command line. If you are a ++GNU Emacs user, you can do your debugging inside that. Also there is ++a(nother) GUI interface called DDD that supports @value{DBG}. ++ ++@node Free Software ++@unnumberedsec Free software ++ ++@DBG is @dfn{free software}, protected by the @sc{gnu} ++General Public License ++(GPL). The GPL gives you the freedom to copy or adapt a licensed ++program---but every person getting a copy also gets with it the ++freedom to modify that copy (which means that they must get access to ++the source code), and the freedom to distribute further copies. ++Typical software companies use copyrights to limit your freedoms; the ++Free Software Foundation uses the GPL to preserve these freedoms. ++ ++Fundamentally, the General Public License is a license which says that ++you have these freedoms and that you cannot take these freedoms away ++from anyone else. ++ ++@unnumberedsec Free Software Needs Free Documentation ++ ++The biggest deficiency in the free software community today is not in ++the software---it is the lack of good free documentation that we can ++include with the free software. Many of our most important ++programs do not come with free reference manuals and free introductory ++texts. Documentation is an essential part of any software package; ++when an important free software package does not come with a free ++manual and a free tutorial, that is a major gap. We have many such ++gaps today. ++ ++Consider Perl, for instance. The tutorial manuals that people ++normally use are non-free. How did this come about? Because the ++authors of those manuals published them with restrictive terms---no ++copying, no modification, source files not available---which exclude ++them from the free software world. ++ ++That wasn't the first time this sort of thing happened, and it was far ++from the last. Many times we have heard a GNU user eagerly describe a ++manual that he is writing, his intended contribution to the community, ++only to learn that he had ruined everything by signing a publication ++contract to make it non-free. ++ ++Free documentation, like free software, is a matter of freedom, not ++price. The problem with the non-free manual is not that publishers ++charge a price for printed copies---that in itself is fine. (The Free ++Software Foundation sells printed copies of manuals, too.) The ++problem is the restrictions on the use of the manual. Free manuals ++are available in source code form, and give you permission to copy and ++modify. Non-free manuals do not allow this. ++ ++The criteria of freedom for a free manual are roughly the same as for ++free software. Redistribution (including the normal kinds of ++commercial redistribution) must be permitted, so that the manual can ++accompany every copy of the program, both on-line and on paper. ++ ++Permission for modification of the technical content is crucial too. ++When people modify the software, adding or changing features, if they ++are conscientious they will change the manual too---so they can ++provide accurate and clear documentation for the modified program. A ++manual that leaves you no choice but to write a new manual to document ++a changed version of the program is not really available to our ++community. ++ ++Some kinds of limits on the way modification is handled are ++acceptable. For example, requirements to preserve the original ++author's copyright notice, the distribution terms, or the list of ++authors, are ok. It is also no problem to require modified versions ++to include notice that they were modified. Even entire sections that ++may not be deleted or changed are acceptable, as long as they deal ++with nontechnical topics (like this one). These kinds of restrictions ++are acceptable because they don't obstruct the community's normal use ++of the manual. ++ ++However, it must be possible to modify all the @emph{technical} ++content of the manual, and then distribute the result in all the usual ++media, through all the usual channels. Otherwise, the restrictions ++obstruct the use of the manual, it is not free, and we need another ++manual to replace it. ++ ++Please spread the word about this issue. Our community continues to ++lose manuals to proprietary publishing. If we spread the word that ++free software needs free reference manuals and free tutorials, perhaps ++the next person who wants to contribute by writing documentation will ++realize, before it is too late, that only free manuals contribute to ++the free software community. ++ ++If you are writing documentation, please insist on publishing it under ++the GNU Free Documentation License or another free documentation ++license. Remember that this decision requires your approval---you ++don't have to let the publisher decide. Some commercial publishers ++will use a free license if you insist, but they will not propose the ++option; it is up to you to raise the issue and say firmly that this is ++what you want. If the publisher you are dealing with refuses, please ++try other publishers. If you're not sure whether a proposed license ++is free, write to @email{licensing@@gnu.org}. ++ ++You can encourage commercial publishers to sell more free, copylefted ++manuals and tutorials by buying them, and particularly by buying ++copies from the publishers that paid for their writing or for major ++improvements. Meanwhile, try to avoid buying non-free documentation ++at all. Check the distribution terms of a manual before you buy it, ++and insist that whoever seeks your business must respect your freedom. ++Check the history of the book, and try to reward the publishers that ++have paid or pay the authors to work on it. ++ ++The Free Software Foundation maintains a list of free documentation ++published by other publishers, at ++@url{http://www.fsf.org/doc/other-free-books.html}. ++ ++@node Invocation ++@chapter Getting in and out ++ ++This chapter discusses how to start @value{DBG}, and how to get out of it. ++The essentials are: ++@itemize @bullet ++@item ++type @samp{bash --debugger @emph{script-name}} or @samp{bashdb ++@emph{script-name}} to start @value{DBG}. ++@item ++type @kbd{quit} or @kbd{C-d} inside the debugger to exit. ++@end itemize ++ ++But also note there are two front-ends available as well. One can also ++enter the debugger inside emacs via the command @code{M-x bashdb} ++after loading Emacs' Grand Unified Debugger, @code{gud}. See ++@ref{Emacs,,Using the BASH debugger from @sc{gnu} Emacs}. And there is ++support in a @value{DDD} for bash. ++ ++@menu ++* Starting the BASH debugger:: How to enter the BASH debugger ++* Quitting the BASH debugger:: How to leave the BASH debugger ++@end menu ++ ++@node Starting the BASH debugger ++@section Starting the BASH debugger ++ ++@emph{Note: it is important to use a debugger-enabled bash. You wil ++get an error message if the debugger is run under a version of BASH ++that does not have debugging support.} ++ ++As mentioned above, one can enter @DBG via Emacs or ++DDD. However you don't have to use either of these. And these still ++need a way on their own to get things started. ++ ++There are in fact two @emph{other} ways to start @value{DBG}. The ++first way is to pass the @samp{--debugger} option to bash with the ++name of your script the scripts arguments following that, or with a ++command string (@code{-c}). ++ ++@example ++bash --debugger @var{script} @var{script-arguments...} ++bash --debugger -c @var{command-string}... ++@end example ++ ++This calls a debugger initialization script. It works much like a ++@BASH login profile which may set variables and define ++functions. But this shell profile is customized for debugging and as ++such arranges for itself to get called before each statement is ++executed. Although there are some problems at present in I/O ++redirection that the method described next doesn't have, it is ++expected that over time more features will be enabled in bash when the ++@samp{--debugger} option is in effect. By default, both debugging in ++Emacs via GUD (@ref{Emacs,,Using the BASH debugger under Emacs}) and ++debugging via @value{DDD} work via this method. ++ ++The form @samp{bash --debugger -c ...} can be used to get into the ++debugger without having to give a script name to debug. Sometimes you ++may want to do this just to see how the debugger works: try some ++debugger commands or maybe get online help. If you run @code{ddd ++--bash} without giving a script name, it in fact uses this form. ++ ++In order for the @samp{--debugger} option to work however, you must ++have the debugger scripts installed in a place where @DBG can ++find them. For this reason, in developing @value{DBG}, I use a second ++method more often; it doesn't require the bash debugger to be ++installed. This method uses another script called @code{bashdb} which ++after taking its own options takes the name of the script to debugged ++and the arguments to pass to that script. Using this method, one ++would start the debugger like this: ++ ++@example ++bash @var{path-to-bashdb}/bashdb @var{bashdb-options} @var{script} @var{script-arguments...} ++@end example ++ ++As with the first method, @code{bash} should be a debugger-enabled ++bash. If @code{bashdb} has the path to bash in it at the top (e.g. via ++@code{#!}), and @code{bashdb} can be found in your program-search ++path, then this might be equivalent to the above: ++ ++@example ++bashdb @var{bashdb-options} @var{script} @var{script-arguments...} ++@end example ++ ++There are two or three disadvantages however of running a debugger ++this way. First @code{$0} will have the value @code{bashdb} rather ++than the script you are trying to run. For some scripts this may ++change the behavior of the debugged script. Second a traceback will ++contain additional lines showing the ``source''-ing of the debugged ++script from @code{bashdb}. And third, although this way works better ++than the first method, over time this way may come into disuse. ++ ++An option that you'll probably need to use if bashdb isn't installed ++but run out of the source code directory is @samp{-L} which specifies ++the directory that contains the debugger script files. ++ ++You can further control how bashdb starts up by using command-line ++options. bashdb itself can remind you of the options available. ++ ++@noindent ++Type ++ ++@example ++bashdb -h ++@end example ++ ++@noindent ++to display all available options and briefly describe their use. ++ ++When the bash debugger is invoked either by the @code{bashdb} ++front-end script or @code{bash --debugging}, the first argument that ++does not have an associated option flag for @code{bashdb} or ++@code{bash} (as the case may be) is used as the name a the script file ++to be debugged, and any following options get passed the debugged ++script. ++ ++Options for the @code{bashdb} front-end are shown in the ++following list. ++ ++@menu ++* Options for the bashdb script:: Options you can pass in starting bashdb ++@end menu ++ ++@node Options for the bashdb script ++@subsection Command-line options for @code{bashdb} script ++ ++You can run @DBG in various alternative modes---for example, in ++batch mode or quiet mode. ++ ++@table @code ++@item -h ++@cindex @code{-h} ++This option causes @value{DBG} to print some basic help and exit. ++ ++@item -V ++@cindex @code{-V} ++This option causes @DBG to print its version number and ++no-warranty blurb, and exit. ++ ++@item -n ++@cindex @code{-n} ++Do not execute commands found in any initialization files. Normally, ++@BASH executes the commands in these files after all the command ++options and arguments have been processed. @xref{Command Files,,Command ++files}. ++ ++@item -q ++@cindex @code{-q} ++``Quiet''. Do not print the introductory and copyright messages. These ++messages are also suppressed in batch mode. ++ ++@item -t ++@cindex @code{-t} ++``Terminal output''. Set the file or terminal that you want debugger command ++output to go to. Note that the debugger output is independent of the ++debugged script output. ++ ++@item -L @var{directory} ++@cindex @code{-L} ++Set directory where debugger files reside to @var{directory}. The ++default location is @code{../lib/bashdb} relative to the place that ++the bashdb script is located. For example if bashdb is located in ++@code{/usr/local/bin/bashdb}, the default library location will be ++@code{/usr/local/lib/bashdb} which may or may not exist. If it doesn't ++you'll get an error when you run bashdb. Only if the default location ++is incorrect, should you need to use the @code{-L} option. ++ ++@item -T @var{directory} ++@cindex @code{-T} ++Set directory to use for writing temporary files. ++ ++@end table ++ ++@node Quitting the BASH debugger ++@section Quitting the BASH debugger ++ ++@cindex interrupt ++An interrupt (often @kbd{C-c}) does not exit from @value{DBG}, but ++rather terminates the action of any @DBG command that is in ++progress and returns to @value{DBG} command level. Inside a debugger ++command interpreter, use @code{quit} command (@pxref{Quit, ,Quitting ++the BASH debugger}). ++ ++@node Running ++@chapter Script Setup inside the BASH Debugger ++ ++@menu ++* Starting:: Starting your script ++* Command Files:: Command files ++* Arguments:: Your script's arguments ++* Input/Output:: Your script's input and output ++* Script/Debugger Interaction:: Keeping out of each other's harm ++@end menu ++ ++@need 2000 ++@node Starting ++@section Starting your script ++@cindex starting ++@cindex running ++ ++@table @code ++@kindex restart ++@kindex run @r{(@code{restart})} ++@kindex R @r{(@code{restart})} ++@item restart @ovar{options to debugged script} ++@itemx run ++@itemx R ++Use the @code{restart} command to restart your script under ++@value{DBG}. Without any arguments, the script name and parameters ++from the last invocation are used. @value{dBGP} tries to maintain the ++settings, watchpoints, breakpoints, actions and so on. Internally it ++uses line numbers and filenames to record he position of interesting ++places in your porgram; so if your program changes some or all of ++these numbers may be off. Environment variable ++@code{BASHDB_RESTART_FILE} is and a temporary file are used to signal ++a restart, so you shouldn't uset @code{BASHDB_RESTART_FILE} (or any ++environment variable starting with @code{BASHDB_}. ++ ++@end table ++ ++@node Command Files ++@section Command files ++ ++@cindex command files ++A command file for @DBG is a file of lines that are @DBG ++commands. Comments (lines starting with @kbd{#}) may also be included. ++An empty line in a command file does nothing; it does not mean to repeat ++the last command, as it would from the terminal. ++ ++@cindex init file ++@cindex @file{.bashdbinit} ++@cindex @file{bashdb.ini} ++When you start @value{DBG}, it automatically executes commands from its ++@dfn{init files}, normally called @file{.bashdbinit}@footnote{The DJGPP ++port of @DBG uses the name @file{bashdb.ini} instead, due to the ++limitations of file names imposed by DOS filesystems.}. ++During startup, @DBG does the following: ++ ++@enumerate ++@item ++Reads the init file (if any) in your home directory@footnote{On ++DOS/Windows systems, the home directory is the one pointed to by the ++@code{HOME} environment variable.}. ++ ++@item ++Processes command line options and operands. ++ ++@item ++Reads the init file (if any) in the current working directory. ++ ++@item ++Reads command files specified by the @samp{-x} option. ++@end enumerate ++ ++The init file in your home directory can set options (such as @samp{set ++complaints}) that affect subsequent processing of command line options ++and operands. Init files are not executed if you use the @samp{-x} ++option (@pxref{Options for the bashdb script, ,bashdb script options}). ++ ++@cindex init file name ++On some configurations of @value{DBG}, the init file is known by a ++different name (these are typically environments where a specialized ++form of @DBG may need to coexist with other forms, hence a ++different name for the specialized version's init file). These are the ++environments with special init file names: ++ ++You can also request the execution of a command file with the ++@code{source} command: ++ ++@table @code ++@kindex source ++@item source @var{filename} ++Execute the command file @var{filename}. ++@end table ++ ++The lines in a command file are executed sequentially. They are not ++printed as they are executed. If there is an error, execution ++proceeds to the next command in the file. ++ ++@node Arguments ++@section Your script's arguments ++ ++@cindex arguments (to your script) ++The arguments to your script can be specified by the arguments of the ++@code{restart} command. ++They are passed to a shell, which expands wildcard characters and ++performs redirection of I/O, and thence to your script. ++ ++@code{restart} with no arguments uses the same arguments used by the previous ++@code{restart}, or those set by the @code{set args} command.. ++ ++@table @code ++@kindex set args ++@item set args ++Specify the arguments to be used the next time your program is run. If ++@code{set args} has no arguments, @code{restart} executes your program ++with no arguments. Once you have run your program with arguments, ++using @code{set args} before the next @code{restart} is the only way to run ++it again without arguments. ++ ++@kindex show args ++@item show args ++Show the arguments to give your program when it is started. ++@end table ++ ++@node Input/Output ++@section Your script's input and output ++ ++@cindex redirection ++@cindex I/O ++@cindex terminal ++By default, the script you run under the BASH debugger does input and ++output to the same terminal that BASH uses. Before running the script ++to be debugged, the debugger records the tty that was in effect. All ++of its output is then written to that. However you can change this ++when using the @samp{bashdb} script using the @samp{-t} option. ++ ++ ++@table @code ++@kindex info terminal ++@item info terminal ++Displays information recorded by @DBG about the terminal modes your ++program is using. ++@end table ++ ++@kindex tty ++@cindex controlling terminal ++Another way to specify where your script should do input and output is ++with the @code{tty} command. This command accepts a file name as ++argument, and causes this file to be the default for future @code{restart} ++commands. It also resets the controlling terminal for the child ++process, for future @code{restart} commands. For example, ++ ++@example ++tty /dev/ttyb ++@end example ++ ++@noindent ++directs that processes started with subsequent @code{restart} commands ++default to do input and output on the terminal @file{/dev/ttyb} and have ++that as their controlling terminal. ++ ++An explicit redirection in @code{restart} overrides the @code{tty} command's ++effect on the input/output device, but not its effect on the controlling ++terminal. ++ ++When you use the @code{tty} command or redirect input in the @code{restart} ++command, only the input @emph{for your script} is affected. The input ++for @DBG still comes from your terminal. ++ ++@node Script/Debugger Interaction ++@section Script/Debugger Interaction ++ ++@value{dBGP} and your program live in the same variable space so ++to speak. @BASH does not have a notion of module scoping or lexical ++hiding (yet) and this then imposes some additional care and awareness. ++ ++Most of the variables and functions used inside @DBG start ++@code{_Dbg_}, so please don't use variables or functions with these ++names in your program. ++ ++@emph{Note: there are some other variables that begin with just an ++underscore (@code{_}); over time these will be phased out. But until ++then, avoid those or consult what is used by the debugger. Run ++@samp{bashdb --debugger -c 'declare -p'} to list all the variables in ++use including those used by the debugger.} ++ ++A number of @DBG environment variables are also reserved for ++use; these start with @code{BASHDB_}. For example: ++@env{BASHDB_INPUT}, @env{BASHDB_LEVEL} and, ++@env{BASHDB_QUIT_ON_QUIT} (@pxref{Debug, ,Debug}), ++@env{BASHDB_RESTART_FILE} (@pxref{Starting, ,Starting}), to name a ++few. Finally, there are some @BASH environment dynamic ++variables and these start with @env{BASH_}. For example ++@env{BASH_SUBSHELL} (@pxref{Debug, ,Debug}), @env{BASH_COMMAND} ++(@pxref{Command Display, ,Command Display}), @env{BASH_LINENO}, and ++@env{BASH_SOURCE} to name a few. ++ ++In order to do its work @value{dBGP} sets up a @code{DEBUG} ++trap. Consequently a script shouldn't reset this or the debugger will ++lose control. @value{dBGP} also sets up an @code{EXIT} handler so that ++it can gain control after the script finishes. Another signal ++intercepted is the an interrupt or @code{INT} signal. For more ++information about signal handling, @pxref{Signals, ,Signals} ++ ++@node Debugger Command Reference ++@chapter BASH Debugger Command Reference ++ ++You can abbreviate the long name of @DBG command to the first ++few letters of the command name, if that abbreviation is unambiguous; ++and you can repeat the @code{next} o r@code{step} commands by typing ++just @key{RET}. Some commands which require a parameter, such as ++@code{print} remember the argument that was given to them. ++ ++@menu ++* Command Syntax:: How to give commands to the BASH debugger ++* Help:: How to ask for help (help) ++* Quit:: Leaving the debugger (quit) ++* Information:: Status and Debugger settings (info, show) ++* Stopping:: Stopping and continuing (break, watch, step, cont...) ++* Stack:: Examining the stack (where, up, down, frame) ++* List:: Printing source files (list) ++* Search:: Searching source files (/pat/ ?pat?) ++* Data:: Examining data (print, examine, info variables) ++* Evaluation/Execution:: Arbitrary execution (eval, shell) ++* Auto Display:: Executing expressions on stop (display, undisplay) ++* Controlling bashdb:: Controlling bashdb (set, history) ++@end menu ++ ++@node Command Syntax ++@section Command syntax ++ ++A BASH debugger command is a single line of input. There is no limit on ++how long it can be. It starts with a command name, which is followed by ++arguments whose meaning depends on the command name. For example, the ++command @code{step} accepts an argument which is the number of times to ++step, as in @samp{step 5}. You can also use the @code{step} command ++with no arguments. Some commands do not allow any arguments. ++ ++@cindex repeating next/step commands ++@kindex RET @r{(repeat last command)} ++A blank line as input to @DBG (typing just @key{RET}) means to ++repeat the previous next or step command. ++ ++@kindex # @r{(a comment)} ++@cindex comment ++Any text from a @kbd{#} to the end of the line is a comment; it does ++nothing. This is useful mainly in command files (@pxref{Command ++Files,,Command files}). ++ ++@node Help ++@section Getting help (@samp{help}) ++@cindex online documentation ++ ++Once inside the BASH debugger, you can always ask it for information on its ++commands, using the command @code{help}. ++ ++@table @code ++@kindex h @r{(@code{help})} ++@item help ++@itemx h ++You can use @code{help} (abbreviated @code{h}) with no arguments to ++display a short list of named classes of commands: ++@end table ++ ++@flushleft ++@smallexample ++bashdb<0> @b{help} ++bashdb commands: ++List/search source lines: Control script execution: ++------------------------- ------------------------- ++ l [start|.] [cnt] List cnt lines T [n] Stack trace ++ from line start s [n] Single step [n times] ++ l sub List source code fn n [n] Next, steps over subs ++ - or . List previous/current line / Repeat last n or s ++ w [line] List around line c [linespec] Continue [to linespec] ++ f filename View source in file L List all breakpoints ++ /pat/ Search forward for pat b linespec Set breakpoint ++ ?pat? Search backward for pat del [n].. or D Delete a/all breaks ++ by entry number ++Debugger controls: skip skip execution of cmd ++------------------------- cl linespec Delete breakpoints by ++ H [num] Show last num commands line spec ++ q [exp] or ^D Quit returning exp R [args] Attempt a restart ++ info [cmd] Get info on cmd. u [n] Go up stack by n or 1. ++ !n or hi n Run debugger history n do [n] Go down stack by n or 1. ++ h or ? [cmd] Get help on command W [var] Add watchpoint. If no ++ info [cmd] Get info on cmd no expr, delete all ++ show [cmd] Show settings We [expr] Add Watchpoint arith ++ expr ++ so file read in dbg commands t Toggle trace ++ en/di n enable/disable brkpt, ++ set x y set a debugger variable watchpoint, or display ++ e bash-cmd evaluate a bash command tb linespec Add one-time break ++ disp expr add a display expr a linespec cmd eval "cmd" at linespec ++ M Show module versions A delete all actions ++ x expr evaluate expression ret jump out of fn or source ++ (via declare, let, eval) finish execute until return ++ deb debug into another cond n exp set breakpoint condition ++ shell script ++ !! cmd [args] execute shell command "cmd" with "args" ++ ++Data Examination: also see e, t, x ++------------------------- ++ p variable Print variable ++ V [[!]pat] List variable(s) matching or not (!) matching pattern pat ++ S [[!]pat] List subroutine names [not] matching pattern pat ++ ++Readline command line editing (emacs/vi mode) is available ++For more help, type h or consult online-documentation. ++@end smallexample ++@end flushleft ++@c the above line break eliminates huge line overfull... ++ ++@table @code ++@item help @var{command} ++With a command name as @code{help} argument, the BASH debugger displays ++short information on how to use that command. ++ ++@example ++bashdb<0> @b{help list} ++l linespec List window lines starting at linespec. ++l min incr List incr lines starting at 'min' linespec. ++l List next window of lines. ++l . Same as above. ++ Long command name: list. ++@end example ++ ++In addition to @code{help}, you can use the debugger command ++@code{info} to inquire about the state of your script, or the state of ++@DBG itself. The listings under @code{info} in the Index ++point to all the sub-commands. @xref{Command Index}. ++@end table ++ ++@c @group ++@table @code ++@kindex info ++@kindex i @r{(@code{info})} ++@item info ++This command (abbreviated @code{i}) is for describing the state of ++your program. For example, you can list the arguments given to your ++script with @code{info args}, or list the breakpoints you have set ++with @code{info breakpoints}. You can get a complete list of the ++@code{info} sub-commands with @w{@code{help info}}. ++ ++@example ++bashdb<0> @b{info} ++Info subcommands are: args breakpoints display files source sources stack terminal variables watchpoints ++bashdb<1> @b{info source} ++Current script file is parm.sh ++Contains 34 lines. ++@end example ++@end table ++ ++ ++@node Quit ++@section Quitting the BASH debugger (@samp{quit}) ++ ++@table @code ++@kindex quit @r{[}@var{expression} @ovar{subshell-levels}@r{]} ++@kindex q @r{(@code{quit})} ++@item quit @ovar{expression} ++@item quit @r{[}@var{expression} @ovar{subshell-levels}@r{]} ++@itemx q ++ ++To exit @value{DBG}, use the @code{quit} command (abbreviated ++@code{q}), or type an end-of-file character (usually @kbd{C-d}). If ++you do not supply @var{expression}, @DBG will try to terminate ++normally or with exit code 0. Otherwise it will terminate using the ++result of @var{expression} as the exit code. ++ ++A simple @code{quit} tries to terminate all nested subshells that may ++be in effect. If you are nested a subshell, this is normally ++indicated in a debugger prompt by the number of parentheses that the ++history number is inside --- no parenthesis means there is no subshell ++in effect. The dynamic variable @env{BASH_SUBSHELL} also contains the ++number of subshells in effect. ++ ++If you want only to terminate some number of subshells but not all of ++them, you can give a count of the number of subshells to leave after ++the return-code expression. To leave just one level of subshell ++@code{return} does almost the same thing. (See @pxref{Returning, ++,Returning}) There is a subtle difference between the two though: ++@code{return} will leave you at the beginning of the next statement ++while @code{quit} may leave you at the place the subshell was invoked ++which may be in the middle of another command such as an assingment ++statement or condition test. ++ ++If the environment variable @code{BASHDB_QUIT_ON_QUIT} is set, when the ++program terminates, the debugger will also terminate too. This may be ++useful if you are debugging a script which calls another script and ++you want this inner script just to return to the outer script. ++@end table ++ ++@node Stopping ++@section Stopping and Resuming Execution (@samp{break}, @samp{watch}, @samp{step}, @samp{cont}) ++ ++One important use of a debugger is to stop your program before it ++terminates so that, if your script might run into trouble, you can ++investigate and find out why. However should your script accidently ++continue to termination, @DBG has arranged for it not to leave the ++debugger without your explicit instruction. That way, you can restart ++the program using the same command arguments. ++ ++Inside @value{DBG}, your script may stop for any of several reasons, ++such as a signal, a breakpoint, or reaching a new line after a ++debugger command such as @code{step}. You may then examine and ++change variables, set new breakpoints or remove old ones, and then ++continue execution. ++ ++@menu ++* Breakpoints:: Breakpoints, watchpoints (break, watch, clear) ++* Resuming Execution:: Resuming execution (continue, step, next, skip, debug) ++* Signals:: Signals ++@end menu ++ ++@node Breakpoints ++@subsection Breakpoints, watchpoints ++ ++@cindex breakpoints ++A @dfn{breakpoint} makes your script stop whenever a certain point in ++the program is reached. For each breakpoint, you can add conditions to ++control in finer detail whether your script stops. ++ ++You can set breakpoints with the @code{break} command and its variants ++(@pxref{Set Breaks, ,Setting breakpoints}), to specify the place where ++your script should stop by line number. or function name in the ++debugged script. ++ ++@cindex watchpoints ++@cindex breakpoint on variable modification ++A @dfn{watchpoint} is a special breakpoint that stops your script when ++the value of an expression changes. There is a different command to ++set watchpoints (@pxref{Set Watchpoints, ,Setting watchpoints}). ++ ++But aside from that, you can manage a watchpoint like any other ++breakpoint: you delete enable, and disable both breakpoints and ++watchpoints using the same commands. ++ ++You can arrange to have values from your program displayed automatically ++whenever @value{BASH} stops at a breakpoint. @xref{Auto Display,, ++Automatic display}. ++ ++@cindex breakpoint numbers ++@cindex numbers for breakpoints ++@DBG assigns a number to each breakpoint and watchpoint when ++you create it; these numbers are successive integers starting with ++one. In many of the commands for controlling various features of ++breakpoints you use the breakpoint number to say which breakpoint you ++want to change. Each breakpoint may be @dfn{enabled} or ++@dfn{disabled}; if disabled, it has no effect on your script until you ++enable it again. ++ ++@cindex watchpoints numbers ++@cindex numbers for watchpoints ++Watchpoint numbers however are distiguished from breakpoint numbers by ++virtue of their being suffixed with the either an upper- or lower-case ++`W'. For example, to enable breakpoint entry 0 along with watchpoint ++entry 1 you would write @samp{enable 1 2w}, the ``2w'' refers to the ++watchpoint; ``2W'' would work just as well. ++ ++@ifset FINISHED ++@cindex breakpoint ranges ++@cindex ranges of breakpoints ++Some @DBG commands accept a range of breakpoints on which to ++operate. A breakpoint range is either a single breakpoint number, like ++@samp{5}, or two such numbers, in increasing order, separated by a ++hyphen, like @samp{5-7}. When a breakpoint range is given to a command, ++all breakpoint in that range are operated on. ++@end ifset ++ ++@menu ++* Set Breaks:: Setting breakpoints (break, tbreak) ++* Set Watchpoints:: Setting watchpoints (watch, watche) ++* Delete Breaks:: Deleting breakpoints (delete, clear) ++* Disabling:: Disabling breakpoints (disable, enable) ++* Conditions:: Break conditions (condition) ++@end menu ++ ++@node Set Breaks ++@subsubsection Setting breakpoints (@samp{break} @samp{tbreak}) ++ ++@kindex break ++@kindex b @r{(@code{break})} ++@cindex latest breakpoint ++Breakpoints are set with the @code{break} command (abbreviated ++@code{b}). ++ ++@table @code ++@item break @var{function} ++Set a breakpoint at entry to function @var{function}. ++ ++@item break @var{linenum} ++Set a breakpoint at line @var{linenum} in the current source file. ++The current source file is the last file whose source text was printed. ++The breakpoint will stop your script just before it executes any of the ++code on that line. ++ ++@item break @var{filename}:@var{linenum} ++Set a breakpoint at line @var{linenum} in source file @var{filename}; ++@var{filename} has to be one of the files previously read in and has ++to be specified exactly as the name used when read in. For a list of ++read-in files, use the @samp{info files} command. ++ ++@ifset FINISHED ++@item break ++When called without any arguments, @code{break} sets a breakpoint at ++the next instruction to be executed in the selected stack frame ++(@pxref{Stack, ,Examining the Stack}). In any selected frame but the ++innermost, this makes your script stop as soon as control returns to ++that frame. If you use @code{break} without an argument in the ++innermost frame, @DBG stops the next time it reaches the ++current location; this may be useful inside loops. ++@end ifset ++ ++@item break @dots{} if @var{cond} ++Set a breakpoint with condition @var{cond}; evaluate the expression ++@var{cond} each time the breakpoint is reached, and stop only if the ++value is nonzero---that is, if @var{cond} evaluates as true. The ++expression is evaluated via the @code{let} builtin funtion. ++@samp{@dots{}} stands for one of the possible arguments described ++above (or no argument) specifying where to break. The word ``if'' is ++often optional and is necessary only @samp{@dots{}} is ++omitted. @xref{Conditions, ,Break conditions}, for more information on ++breakpoint conditions. ++ ++Examples: ++@example ++bashdb<0> @b{break fn1} ++Breakpoint 1 set in file parm.sh, line 3. ++bashdb<1> @b{break 28} ++Breakpoint 2 set in file parm.sh, line 28. ++bashdb<2> @b{break parm.sh:29} ++Breakpoint 3 set in file parm.sh, line 29. ++bashdb<3> @b{break 28 if x==5} ++Breakpoint 4 set in file parm.sh, line 28. ++@end example ++ ++@kindex tbreak ++@item tbreak @var{args} ++Set a breakpoint enabled only for one stop. @var{args} are the ++same as for the @code{break} command, and the breakpoint is set in the same ++way, but the breakpoint is automatically deleted after the first time your ++program stops there. @xref{Disabling, ,Disabling breakpoints}. ++ ++@kindex info breakpoints ++@cindex @code{$_} and @code{info breakpoints} ++@item info breakpoints @ovar{n} ++@itemx info break @ovar{n} ++@itemx info watchpoints @ovar{n} ++Print a table of all breakpoints, watchpoints set and not deleted, ++with the following columns for each breakpoint: ++ ++@table @emph ++@item Breakpoint Numbers (@samp{Num}) ++@item Enabled or Disabled (@samp{Enb}) ++Enabled breakpoints are marked with @samp{1}. @samp{0} marks breakpoints ++that are disabled (not enabled). ++@item Count ++The number of times that breakpoint or watchpoint has been hit. ++@item Condition ++The arithmetic expression ++@item File and Line (@samp{file:line}) ++The filename and line number inside that file where of breakpoint in ++the script. The file and line are separated with a colon. ++@end table ++ ++@noindent ++If a breakpoint is conditional, @code{info break} shows the condition on ++the line following the affected breakpoint; breakpoint commands, if any, ++are listed after that. ++ ++@noindent ++@code{info break} displays a count of the number of times the breakpoint ++has been hit. ++ ++@code{info break} with a breakpoint number @var{n} as argument lists ++only that breakpoint. ++ ++Examples: ++@example ++bashdb<4> @b{info break} ++Breakpoints at following places: ++Num Type Disp Enb What ++1 breakpoint keep y parm.sh:3 ++2 breakpoint keep y parm.sh:28 ++3 breakpoint keep y parm.sh:29 ++4 breakpoint keep y parm.sh:28 ++No watch expressions have been set. ++bashdb<5> @b{info break 4} ++Num Type Disp Enb What ++4 breakpoint keep y parm.sh:28 ++No watch expressions have been set. ++@end example ++@end table ++ ++ ++@ifset FINISHED ++This is especially useful in conjunction with the ++@code{ignore} command. You can ignore a large number of breakpoint ++hits, look at the breakpoint info to see how many times the breakpoint ++was hit, and then run again, ignoring one less than that number. This ++will get you quickly to the last hit of that breakpoint. ++@end ifset ++ ++@DBG allows you to set any number of breakpoints at the same place in ++your script. There is nothing silly or meaningless about this. When ++the breakpoints are conditional, this is even useful ++(@pxref{Conditions, ,Break conditions}). ++ ++@node Set Watchpoints ++@subsubsection Setting watchpoints (@samp{watch}, @samp{watche}) ++ ++@cindex setting watchpoints ++You can use a watchpoint to stop execution whenever the value of an ++expression changes, without having to predict a particular place where ++this may happen. As with the @code{print} (@pxref{Data,,Examining ++Data}), the idiosyncracies of a BASH or any POSIX shell derivative ++suggest using two commands. The @code{watch} command is just for a ++single variables; the @code{watche} command uses the builtin ``let'' ++command to evaluate an expression. If the variable you are tracking ++can take a string value, issuing something like @samp{watch foo} will ++not have the desired effect---any string assignment to @code{foo} will ++have a value 0 when it is assigned via ``let.'' ++ ++@table @code ++@kindex watch ++@item watch @var{var} ++Set a watchpoint for a variable. @DBG will break when the ++value of @var{var} changes. In this command do not add a leading ++dollar symbol to @var{var}. ++ ++@item watche @var{expr} ++Set a watchpoint for an expression via the builtin ``let'' command. ++@DBG will break when @var{expr} is written into by the program ++and its value changes. Not that this may not work for tracking ++arbitrary string value changes. For that use @code{watch} described ++earlier. ++@end table ++ ++@node Delete Breaks ++@subsubsection Deleting breakpoints (@samp{clear}, @samp{delete}) ++ ++@cindex clearing breakpoints, watchpoints ++@cindex deleting breakpoints, watchpoints ++It is often necessary to eliminate a breakpoint or watchpoint once it ++has done its job and you no longer want your script to stop there. ++This is called @dfn{deleting} the breakpoint. A breakpoint that has ++been deleted no longer exists; it is forgotten. ++ ++With the @code{clear} command you can delete breakpoints according to ++where they are in your script. With the @code{delete} command you can ++delete individual breakpoints, or watchpoints by specifying their ++breakpoint numbers. @emph{Note: as described below under the ``clear'' ++command, ``d'' is an alias for ``clear'', not ``delete''. } ++ ++It is not necessary to delete a breakpoint to proceed past it. @DBG ++automatically ignores breakpoints on the first instruction to be executed ++when you continue execution. ++ ++@table @code ++@kindex clear ++@kindex d @r{(@code{clear})} ++@item clear ++Delete any breakpoints at the next instruction to be executed in the ++selected stack frame (@pxref{Selection, ,Selecting a frame}). When ++the innermost frame is selected, this is a good way to delete a ++breakpoint where your script just stopped. ++ ++It may seem odd that we have an alias ``d'' for ``clear.'' It so ++happens that Perl's debugger use ``d'' for its delete command and the ++delete concept in Perl's debugger corresponds to ``clear'' in ++GDB. (Perl doesn't have a notion of breakpoint entry numbers). So in ++order to be compatible with both debugger interfaces, ``d'' is used as ++an alias for ``clear.'' Clear? ++ ++@item clear @var{function} ++@itemx clear @var{filename}:@var{function} ++Delete any breakpoints set at entry to the function @var{function}. ++ ++@item clear @var{linenum} ++@itemx d @var{linenum} ++@ifset FINISHED ++@itemx clear @var{filename}:@var{linenum} ++@end ifset ++Delete any breakpoints set at or within the code of the specified line. ++ ++@cindex delete breakpoints ++@kindex delete ++@kindex de @r{(@code{delete})} ++@item delete @ovar{breakpoints} ++Delete the breakpoints, watchpoints specified as arguments. ++ ++If no argument is specified, delete all breakpoints (@DBG asks ++confirmation, unless you have @code{set confirm off}). You can ++abbreviate this command as @code{de}. ++ ++ ++Note that for compatibility with Perl's debugger, @code{d} means ++something else: @code{clear}. ++ ++@end table ++ ++@node Disabling ++@subsubsection Disabling breakpoints (@samp{disable}, @samp{enable}) ++ ++Rather than deleting a breakpoint or watchpoint, you might ++prefer to @dfn{disable} it. This makes the breakpoint inoperative as if ++it had been deleted, but remembers the information on the breakpoint so ++that you can @dfn{enable} it again later. ++ ++You disable and enable breakpoints, watchpoints, and catchpoints with ++the @code{enable} and @code{disable} commands, optionally specifying one ++or more breakpoint numbers as arguments. Use @code{info break} or ++@code{info watch} to print a list of breakpoints, watchpoints, and ++catchpoints if you do not know which numbers to use. ++ ++A breakpoint, watchpoint, or catchpoint can have any of four different ++states of enablement: ++ ++@itemize @bullet ++@item ++Enabled. The breakpoint stops your program. A breakpoint set ++with the @code{break} command starts out in this state. ++@item ++Disabled. The breakpoint has no effect on your program. ++@item ++Enabled once. The breakpoint stops your program, but then becomes ++disabled. ++@item ++Enabled for deletion. The breakpoint stops your program, but ++immediately after it does so it is deleted permanently. A breakpoint ++set with the @code{tbreak} command starts out in this state. ++@end itemize ++ ++You can use the following commands to enable or disable breakpoints, ++watchpoints, and catchpoints: ++ ++@table @code ++@kindex disable breakpoints ++@kindex disable ++@kindex dis @r{(@code{disable})} ++@item disable @ovar{breakpoints} ++Disable the specified breakpoints---or all breakpoints, if none are ++listed. A disabled breakpoint has no effect but is not forgotten. All ++options such as ignore-counts, conditions and commands are remembered in ++case the breakpoint is enabled again later. You may abbreviate ++@code{disable} as @code{dis}. ++ ++@kindex enable breakpoints ++@kindex enable ++@item enable @ovar{breakpoints} ++Enable the specified breakpoints (or all defined breakpoints). They ++become effective once again in stopping your program. ++@end table ++ ++@c FIXME: I think the following ``Except for [...] @code{tbreak}'' is ++@c confusing: tbreak is also initially enabled. ++Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks, ++,Setting breakpoints}), breakpoints that you set are initially enabled; ++subsequently, they become disabled or enabled only when you use one of ++the commands above. (The command @code{until} can set and delete a ++breakpoint of its own, but it does not change the state of your other ++breakpoints; see @ref{Resuming Execution, ,Resuming Execution}.) ++ ++@node Conditions ++@subsubsection Break conditions (@samp{condition}) ++@cindex conditional breakpoints ++@cindex breakpoint conditions ++ ++The simplest sort of breakpoint breaks every time your script reaches ++a specified place. You can also specify a @dfn{condition} for a ++breakpoint. A condition is just a BASH expression. ++ ++Break conditions can be specified when a breakpoint is set, by using ++@samp{if} in the arguments to the @code{break} command. @xref{Set ++Breaks, ,Setting breakpoints}. A breakpoint with a condition ++evaluates the expression each time your script reaches it, and your ++script stops only if the condition is @emph{true}. ++ ++@cindex one-time breakpoints ++There is also a notion of a ``one-time'' breakpoint which gets deleted ++as soon as it is hit, so that that breakpoint is executed once only. ++ ++Conditions are also accepted for watchpoints; you may not need them, ++since a watchpoint is inspecting the value of an expression anyhow---but ++it might be simpler, say, to just set a watchpoint on a variable name, ++and specify a condition that tests whether the new value is an interesting ++one. ++ ++Break conditions can be specified when a breakpoint is set, by using ++@samp{if} in the arguments to the @code{break} command. @xref{Set ++Breaks, ,Setting breakpoints}. They can also be changed at any time ++with the @code{condition} command. ++ ++@ifset FINISHED ++You can also use the @code{if} keyword with the @code{watch} command. ++The @code{catch} command does not recognize the @code{if} keyword; ++@code{condition} is the only way to impose a further condition on a ++catchpoint. ++@end ifset ++ ++@table @code ++@kindex condition ++@item condition @var{bnum} @var{expression} ++Specify @var{expression} as the break condition for breakpoint ++@var{bnum}. After you set a condition, breakpoint @var{bnum} stops ++your program only if the value of @var{expression} is true (nonzero). ++ ++@item condition @var{bnum} ++Remove the condition from breakpoint number @var{bnum}. It becomes ++an ordinary unconditional breakpoint. ++@end table ++ ++@ifset FINISHED ++When you use @code{condition}, @DBG checks @var{expression} ++immediately for syntactic correctness, and to determine whether ++symbols in it have referents in the context of your breakpoint. If ++@var{expression} uses symbols not referenced in the context of the ++breakpoint, @DBG prints an error message: ++ ++@example ++No symbol "foo" in current context. ++@end example ++@end ifset ++ ++@noindent ++@BASH does ++not actually evaluate @var{expression} at the time the @code{condition} ++command (or a command that sets a breakpoint with a condition, like ++@code{break if @dots{}}) is given, however. ++ ++Examples; ++@example ++condition 1 x>5 # Stop on breakpoint 0 only if x>5 is true. ++condition 1 # Change that! Unconditinally stop on breakpoint 1. ++@end example ++ ++@node Resuming Execution ++@subsection Resuming Execution ++ ++@cindex stepping ++@cindex continuing ++@cindex resuming execution ++@dfn{Continuing} means resuming program execution until your script ++completes normally. In contrast, @dfn{stepping} means executing just ++one more ``step'' of your script, where ``step'' may mean either one ++line of source code. Either when continuing or when stepping, ++your script may stop even sooner, due to a breakpoint or a signal. ++ ++@menu ++* Step:: running the next statement (step) ++* Next:: running the next statement skipping over functions (skip) ++* Finish:: running until the return of a function or ``source'' (finish) ++* Skip:: skipping the next statement (skip) ++* Continue:: continuing execution (continue) ++* Debug:: debugging into another program (debug) ++* Returning:: returning ++@end menu ++ ++@node Step ++@subsubsection Step (@samp{step}) ++@table @code ++@kindex step ++@kindex s @r{(@code{step})} ++@item step ++Continue running your script until control reaches a different source ++line, then stop it and return control to @value{DBG}. This command is ++abbreviated @code{s}. ++ ++The @code{step} command only stops at the first instruction of a source ++line. This prevents the multiple stops that could otherwise occur in ++@code{switch} statements, @code{for} loops, etc. @code{step} continues ++to stop if a function that has debugging information is called within ++the line. In other words, @code{step} @emph{steps inside} any functions ++called within the line. ++ ++@item step @ovar{count} ++Continue running as in @code{step}, but do so @var{count} times. If a ++breakpoint is reached, or a signal not related to stepping occurs before ++@var{count} steps, stepping stops right away. ++@end table ++ ++@node Next ++@subsubsection Next (@samp{next}) ++@table @code ++@kindex next ++@kindex n @r{(@code{next})} ++@item next @ovar{count} ++Continue to the next source line in the current (innermost) stack frame. ++This is similar to @code{step}, but function calls that appear within ++the line of code are executed without stopping. Execution stops when ++control reaches a different line of code at the original stack level ++that was executing when you gave the @code{next} command. This command ++is abbreviated @code{n}. ++ ++An argument @var{count} is a repeat count, as for @code{step}. ++@end table ++ ++@node Finish ++@subsubsection Finish (@samp{finish}) ++@table @code ++@kindex finish ++@item finish ++Continue running until just after function returns. @emph{Currently, ++the line shown on a return is the function header, unless the ++@code{return} builtin function is executed in which case it is the ++line number of the @code{return} function.} ++ ++Contrast this with the @code{return} command (@pxref{Returning, ++,Returning from a function}) and the @code{quit} (@pxref{Quitting the ++BASH debugger, ,Quitting the BASH debugger}). ++ ++@end table ++ ++@node Skip ++@subsubsection Skip (@samp{skip}) ++@table @code ++@kindex skip ++@item skip @ovar{count} ++Skip exection of the next source line. ++This may be useful if you have an action that ``fixes'' existing code in ++the script. The @code{debug} command internally uses the @code{skip} command ++to skip over existing non-debugged invocation that was presumably just ++run. ++@end table ++ ++@node Continue ++@subsubsection Continue (@samp{continue}) ++@table @code ++@kindex continue ++@kindex c @r{(@code{continue})} ++@item continue @ovar{linespec} ++@itemx c @ovar{line-number} ++Resume program execution, at the address where your script last ++stopped; any breakpoints set at that address are bypassed. The ++optional argument @var{linespec} allows you to specify a linespec (a ++line number, function, or filename linenumber combination) to set. A ++one-time breakpoint is deleted when that breakpoint is reached. Should ++the program stop before that breakpoint is reached, for example, ++perhaps another breakpoint or watchpoint is reached first, in a ++listing of the breakpoints you will see this entry with the condition ++9999 which indicates a one-time breakpoint. ++@end table ++ ++To resume execution at a different place, you can use @code{return} ++(@pxref{Returning, ,Returning from a function}) to go back to the ++calling function or sourced script. If you are nested inside a ++subshell, @code{quit} with a value for the number of subshells to ++exit also functions like a return. ++ ++A typical technique for using stepping is to set a breakpoint ++(@pxref{Breakpoints, ,Breakpoints; watchpoints}) at the ++beginning of the function or the section of your script where a problem ++is believed to lie, run your script until it stops at that breakpoint, ++and then step through the suspect area, examining the variables that are ++interesting, until you see the problem happen. ++ ++@node Debug ++@subsubsection Debug (@samp{debug}) ++@table @code ++@kindex debug ++@item debug @ovar{script-name} ++Debug into @var{script-name}. If no name is given the current source line ++is used. In either case the options are prepended to cause the ++debugger to run. ++ ++The nesting level of the debugger is saved inside environment variable ++@code{BASHDB_LEVEL}. The debugger prompt indicates the level of nesting ++by enclosing the history in that many nestings of @code{<>} symbols. ++ ++@end table ++ ++@node Returning ++@subsubsection Returning from a function, sourced file, or subshell (@samp{return}) ++ ++@table @code ++@cindex returning from a function, sourced file or subshell ++@kindex return ++@item return ++@itemx return ++You can cancel execution of a function call or a subshell with the ++@code{return} command. ++@end table ++ ++The @code{return} command does not resume execution; it leaves the ++program stopped in the state that would exist if the function had just ++returned. See also the @code{quit} command (@ref{Quit, ,Quitting the ++BASH debugger}). In some situations @code{return} is similar to ++@code{quit}: in particular when the script is @emph{not} currenlty ++inside in a function and the number of subshells in effect is 0, or ++when a subshell count of 1 is given on the @code{quit} command. ++ ++@ifset FINISHED ++In contrast, the @code{finish} command (@pxref{Continuing ++and Stepping, ,Continuing and stepping}) resumes execution until the ++selected stack frame returns naturally. ++@end ifset ++ ++@node Signals ++@subsection Signals ++@cindex signals ++ ++A signal is an asynchronous event that can happen in a program. The ++operating system defines the possible kinds of signals, and gives each ++kind a name and a number. For example, in Unix @code{SIGINT} is the ++signal a program gets when you type an interrupt character (often ++@kbd{C-c}); @code{SIGALRM} occurs when the alarm clock timer goes off ++(which happens only if your program has requested an alarm). ++ ++Some signal handlers are installed and changed for @value{DBG}'s ++normal use: @code{SIGDEBUG} and @code{SIGEXIT}. @code{SIGDEBUG} is ++used by the debugger to potentially stop your program before execution ++of each statement occurs, and @code{SIGEXIT} is used to catch your ++program just before it is set to leave so you have the option of ++restarting the program with the same options (and not leave the ++debugger) or let the program quit. ++ ++Signal handlers that the debugged script might have installed are ++saved and called before the corresponding debugger handler. Thus, the ++debugged program should work roughly in the same fashion as when it is ++not debugged. However there are some call-stack variables which ++inevitably will differ. To try to hedge this a little so the behaviour ++is the same, @value{DBG} will modify arguments to the traps if it ++finds one of the call-stack that change as a result of the debugger ++being in place. In particluar @env{$LINENO} will get replaced with ++@env{$@{BASH_LINENO[0]@}}; also @env{$@{BASH_LINENO[0]@}} and ++@env{$@{BASH_SOURCE[0]@}} get replaced with ++@env{$@{BASH_LINENO[1]@}} and @env{$@{BASH_SOURCE[1]@}} ++respectively. ++ ++The debugger also installs an interrupt handler @code{SIGINT} so that ++errant programs can be interrupted and you can find out where the ++program was when you interrupted it. ++ ++@cindex fatal signals ++Some signals, including @code{SIGALRM}, are a normal part of the ++functioning of your program. Others, such as @code{SIGSEGV}, indicate ++errors; these signals are @dfn{fatal} (they kill your program immediately) if the ++program has not specified in advance some other way to handle the signal. ++@code{SIGINT} does not indicate an error in your program, but it is normally ++fatal so it can carry out the purpose of the interrupt: to kill the program. ++ ++@BASH has the ability to detect any occurrence of a signal in your ++program. You can tell @BASH in advance what to do for each kind of ++signal. ++ ++@cindex handling signals ++Normally, @BASH is set up to let the non-erroneous signals like ++@code{SIGALRM} be silently passed to your program ++(so as not to interfere with their role in the program's functioning) ++but to stop your program immediately whenever an error signal happens. ++You can change these settings with the @code{handle} command. ++ ++@table @code ++@kindex info signals ++@item info signals ++@itemx info handle ++Print a table of all the kinds of signals and how @BASH has been told to ++handle each one. You can use this to see the signal numbers of all ++the defined types of signals. ++ ++@code{info handle} is an alias for @code{info signals}. ++ ++@kindex handle ++@item handle @var{signal} @var{keywords}@dots{} ++Change the way @BASH handles signal @var{signal}. @var{signal} ++can be the number of a signal or its name (with or without the ++@samp{SIG} at the beginning); a list of signal numbers of the form ++@samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the ++known signals. The @var{keywords} say what change to make. ++@end table ++ ++@c @group ++The keywords allowed by the @code{handle} command can be abbreviated. ++Their full names are: ++ ++@table @code ++@item stop ++@BASH should stop your program when this signal happens. This implies ++the @code{print} keyword as well. ++ ++@item nostop ++@BASH should not stop your program when this signal happens. It may ++still print a message telling you that the signal has come in. ++ ++@item print ++@BASH should print a message when this signal happens. ++ ++@item noprint ++@BASH should not mention the occurrence of the signal at all. ++ ++@item stack ++@BASH should print a stack trace when this signal happens. ++ ++@item nostack ++@BASH should not print a stack trace when this signal occurs. ++ ++@ifset FINISHED ++@item pass ++@itemx noignore ++@BASH should allow your program to see this signal; your program ++can handle the signal, or else it may terminate if the signal is fatal ++and not handled. @code{pass} and @code{noignore} are synonyms. ++ ++@item nopass ++@itemx ignore ++@BASH should not allow your program to see this signal. ++@code{nopass} and @code{ignore} are synonyms. ++@end ifset ++@end table ++@c @end group ++ ++@ifset FINISHED ++When a signal stops your program, the signal is not visible to the ++program until you ++continue. Your program sees the signal then, if @code{pass} is in ++effect for the signal in question @emph{at that time}. In other words, ++after @BASH reports a signal, you can use the @code{handle} ++command with @code{pass} or @code{nopass} to control whether your ++program sees that signal when you continue. ++ ++The default is set to @code{nostop}, @code{noprint}, @code{pass} for ++non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and ++@code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the ++erroneous signals. ++ ++You can also use the @code{signal} command to prevent your program from ++seeing a signal, or cause it to see a signal it normally would not see, ++or to give it any signal at any time. For example, if your program stopped ++due to some sort of memory reference error, you might store correct ++values into the erroneous variables and continue, hoping to see more ++execution; but your program would probably terminate immediately as ++a result of the fatal signal once it saw the signal. To prevent this, ++you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your ++program a signal}. ++@end ifset ++ ++@node Information ++@section Status and Debugger Settings (@samp{info}, @samp{show}) ++ ++In addition to @code{help}, you can use the @BASH commands @code{info} ++and @code{show} to inquire about the state of your program, or the state ++of @BASH itself. Each command supports many topics of inquiry; this ++manual introduces each of them in the appropriate context. The listings ++under @code{info} and under @code{show} in the Index point to ++all the sub-commands. @xref{Command Index}. ++ ++@c @group ++@table @code ++@kindex info ++@kindex i @r{(@code{info})} ++@item info ++This command (abbreviated @code{i}) is for describing the state of ++your program. For example, you canlist the current @code{$1}, @code{$2} ++parameters with @code{info args}, or list the breakpoints you have set ++with @code{info breakpoints} or @code{info watchpoints}. You can get ++a complete list of the @code{info} sub-commands with @w{@code{help ++info}}. ++ ++@kindex set ++@item set ++You can assign the result of an expression to an environment variable ++with @code{set}. For example, you can set the number of lines ++@BASH will list by default to 20 with @code{set listsize 20}. ++ ++@kindex show ++@item show ++In contrast to @code{info}, @code{show} is for describing the state of ++@BASH itself. ++You can change most of the things you can @code{show}, by using the ++related command @code{set}; ++ ++The distinction between @code{info} and @code{show} however is a bit ++fuzzy. For example, to list the arguments given to your script use ++@code{show args}; @code{info args} does something different. ++ ++@kindex info set ++To display all the settable parameters and their current ++values, you can use @code{show} with no arguments; you may also use ++@code{info set}. Both commands produce the same display. ++@c FIXME: "info set" violates the rule that "info" is for state of ++@c FIXME...program. Ck w/ GNU: "info set" to be called something else, ++@c FIXME...or change desc of rule---eg "state of prog and debugging session"? ++@end table ++@c @end group ++ ++Here are three miscellaneous @code{show} subcommands, all of which are ++exceptional in lacking corresponding @code{set} commands: ++ ++@table @code ++@kindex show version ++@cindex version number ++@item show version ++Show what version of @BASH is running. You should include this ++information in @BASH bug-reports. If multiple versions of ++@BASH are in use at your site, you may need to determine which ++version of @BASH you are running; as @BASH evolves, new ++commands are introduced, and old ones may wither away. Also, many ++system vendors ship variant versions of @value{BASH}, and there are ++variant versions of @BASH in @sc{gnu}/Linux distributions as well. ++The version number is the same as the one announced when you start ++@value{BASH}. ++ ++@kindex show copying ++@item show copying ++Display information about permission for copying @value{BASH}. ++ ++@kindex show warranty ++@item show warranty ++Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty, ++if your version of @DBG comes with one. ++ ++@end table ++ ++@node Stack ++@section Examining the Stack (@samp{where}, @samp{frame}, @samp{up}, @samp{down}) ++ ++When your script has stopped, one thing you'll probably want to know ++is where it stopped and some idea of how it got there. ++ ++@cindex call stack ++Each time your script performs a function call (either as part of a ++command substitution or not), or `source's a file, information about ++this action is saved. The call stack then is this a history of the ++calls that got you to the point that you are currently stopped at. ++ ++@cindex selected frame ++One of the stack frames is @dfn{selected} by @DBG and many ++@DBG commands refer implicitly to the selected frame. In ++particular, whenever you ask @DBG to list lines without giving ++a line number or location the value is found in the selected frame. ++There are special @DBG commands to select whichever frame you ++are interested in. @xref{Selection, ,Selecting a frame}. ++ ++When your program stops, @BASH automatically selects the ++currently executing frame and describes it briefly, similar to the ++@code{frame} command. ++ ++ ++@menu ++* Frames:: Stack frames ++* Backtrace:: Backtraces (where) ++* Selection:: Selecting a frame (up, down, frame) ++ ++@end menu ++ ++@node Frames ++@subsection Stack frames ++ ++@cindex frame, definition ++@cindex stack frame ++The call stack is divided up into contiguous pieces called @dfn{stack ++frames}, or @dfn{frames} for short; each frame is the data associated ++with one call to one function. The frame contains the line number of ++the caller of the function, the source-file name that the line refers ++to and function name (which could be the built-in name ``source'').. ++ ++@cindex initial frame ++@cindex outermost frame ++@cindex innermost frame ++When your script is started, the stack has only one frame, that of the ++function @code{main}. This is called the @dfn{initial} frame or the ++@dfn{outermost} frame. Each time a function is called, a new frame is ++made. Each time a function returns, the frame for that function invocation ++is eliminated. If a function is recursive, there can be many frames for ++the same function. The frame for the function in which execution is ++actually occurring is called the @dfn{innermost} frame. This is the most ++recently created of all the stack frames that still exist. ++ ++@cindex frame number ++@value{DBG} assigns numbers to all existing stack frames, starting with ++zero for the innermost frame, one for the frame that called it, ++and so on upward. These numbers do not really exist in your script; ++they are assigned by @value{DBG} to give you a way of designating stack ++frames in @value{DBG} commands. ++ ++@node Backtrace ++@subsection Backtraces (@samp{where}) ++ ++@cindex backtraces ++@cindex tracebacks ++@cindex stack traces ++A backtrace is essentially the same as the call stack: a summary of ++how your script got where it is. It shows one line per frame, for ++many frames, starting with the place that you sare stopped at (frame ++zero), followed by its caller (frame one), and on up the stack. ++ ++@table @code ++@kindex backtrace ++@kindex bt @r{(@code{backtrace})} ++@item backtrace ++@itemx bt ++@itemx where ++@itemx T ++Print a backtrace of the entire stack: one line per frame for all ++frames in the stack. ++ ++@item backtrace @var{n} ++@itemx bt @var{n} ++@itemx where @var{n} ++@itemx T @var{n} ++Similar, but print only the innermost @var{n} frames. ++ ++@ifset FINISHED ++@item backtrace -@var{n} ++@itemx bt -@var{n} ++@itemx where -@var{n} ++@itemx T -@var{n} ++Similar, but print only the outermost @var{n} frames. ++@end ifset ++@end table ++ ++@kindex where ++The names @code{where} and @code{T} are additional aliases for ++@code{backtrace}. ++ ++Each line in the backtrace shows the frame number and the function ++name, the source file name and line number, as well as the function name. ++ ++Here is an example of a backtrace taken a program in the ++regression-tests @file{parm.sh}. ++ ++@smallexample ++@group ++% ../bashdb -n -L .. parm.sh ++Bourne-Again Shell Debugger, release @value{BASHRELEASE} ++Copyright 2002,2003 Rocky Bernstein ++This is free software, covered by the GNU General Public License, and you are ++welcome to change it and/or distribute copies of it under certain conditions. ++ ++(./parm.sh:21): ++21: fn1 5 ++bashdb<0> @b{continue fn3} ++One-time breakpoint 1 set in file ./parm.sh, line 17. ++fn2: testing 1 2 3 ++(./parm.sh:17): ++17: fn3() @{ ++bashdb<1> @b{where} ++->0 in file `./parm.sh' at line 14 ++##1 fn3() called from file `./parm.sh' at line 14 ++##2 fn2("testing 1", "2 3") called from file `parm.sh' at line 5 ++##3 fn1("0") called from file `parm.sh' at line 9 ++##4 fn1("1") called from file `parm.sh' at line 9 ++##5 fn1("2") called from file `parm.sh' at line 9 ++##6 fn1("3") called from file `parm.sh' at line 9 ++##7 fn1("4") called from file `parm.sh' at line 9 ++##8 fn1("5") called from file `parm.sh' at line 21 ++##9 source("parm.sh") called from file `bashdb' at line 143 ++##10 main("-n", "-L", "..", "parm.sh") called from file `bashdb' at line 0 ++@end group ++@end smallexample ++ ++@noindent ++The display for ``frame'' zero isn't a frame at all, although it has ++the same information minus a function name; it just indicates that ++your script has stopped at the code for line @code{14} ++of @code{./parm.sh}. ++ ++@node Selection ++@subsection Selecting a frame (@samp{up}, @samp{down}, @samp{frame}) ++ ++Commands for listing source code in your script work on whichever ++stack frame is selected at the moment. Here are the commands for ++selecting a stack frame; all of them finish by printing a brief ++description of the stack frame just selected. ++ ++@table @code ++@kindex up ++@item up @var{n} ++Move @var{n} frames up the stack. For positive numbers @var{n}, this ++advances toward the outermost frame, to higher frame numbers, to frames ++that have existed longer. @var{n} defaults to one. ++ ++@kindex down ++@kindex do @r{(@code{down})} ++@item down @var{n} ++Move @var{n} frames down the stack. For positive numbers @var{n}, this ++advances toward the innermost frame, to lower frame numbers, to frames ++that were created more recently. @var{n} defaults to one. You may ++abbreviate @code{down} as @code{do}. ++@end table ++ ++All of these commands end by printing two lines of output describing the ++frame. The first line shows the frame number, the function name, the ++arguments, and the source file and line number of execution in that ++frame. The second line shows the text of that source line. ++ ++@need 100 ++For example: ++ ++@smallexample ++@group ++bashdb<8> @b{up} ++19: sourced_fn ++bashdb<8> @b{T} ++##0 in file `./bashtest-sourced' at line 8 ++->1 sourced_fn() called from file `bashtest-sourced' at line 19 ++##2 source() called from file `bashdb-test1' at line 23 ++##3 fn2() called from file `bashdb-test1' at line 33 ++##4 fn1() called from file `bashdb-test1' at line 42 ++##5 main() called from file `bashdb-test1' at line 0 ++@end group ++@end smallexample ++ ++After such a printout, the @code{list} command with no arguments ++prints ten lines centered on the point of execution in the frame. ++@xref{List, ,Printing source lines}. ++ ++@table @code ++@kindex frame ++@cindex current stack frame ++@item frame @var{args} ++The @code{frame} command allows you to move from one stack frame to ++another, and to print the stack frame you select. @var{args} is the ++the stack frame number. Without an argument, @code{frame} prints the ++current stack frame. ++@end table ++ ++@node List ++@section Examining Source Files (@samp{list}) ++ ++@value{DBG} can print parts of your script's source. When your ++script stops, @value{DBG} spontaneously prints the line where it ++stopped. Likewise, when you select a stack frame (@pxref{Selection, ++,Selecting a frame}), @value{DBG} prints the line where execution in ++that frame has stopped. You can print other portions of source files ++by explicit command. ++ ++If you use @value{DBG} through its @sc{gnu} Emacs interface, you may ++prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using ++@value{DBG} under @sc{gnu} Emacs}. ++ ++@kindex list ++@kindex l @r{(@code{list})} ++To print lines from a source file, use the @code{list} command ++(abbreviated @code{l}). By default, ten lines are printed. ++There are several ways to specify what part of the file you want to print. ++ ++Here are the forms of the @code{list} command most commonly used: ++ ++@table @code ++@item list @var{linenum} ++@itemx l @var{linenum} ++Print lines centered around line number @var{linenum} in the ++current source file. ++ ++@item list @var{function} ++@itemx l @var{function} ++Print the text of @var{function}. ++ ++@item list ++@itemx l ++Print more lines. If the last lines printed were printed with a ++@code{list} command, this prints lines following the last lines ++printed; however, if the last line printed was a solitary line printed ++as part of displaying a stack frame (@pxref{Stack, ,Examining the ++Stack}), this prints lines centered around that line. ++ ++@item list - ++@itemx l - ++Print lines just before the lines last printed. ++@end table ++ ++By default, @value{DBG} prints ten source lines with any of these forms of ++the @code{list} command. ++You can change this using @code{set listsize}: ++ ++@table @code ++@kindex set listsize ++@item set listsize @var{count} ++Make the @code{list} command display @var{count} source lines (unless ++the @code{list} argument explicitly specifies some other number). ++ ++@kindex show listsize ++@item show listsize ++Display the number of lines that @code{list} prints. ++@end table ++ ++Repeating a @code{list} command with @key{RET} discards the argument, ++so it is equivalent to typing just @code{list}. This is more useful ++than listing the same lines again. An exception is made for an ++argument of @samp{-}; that argument is preserved in repetition so that ++each repetition moves up in the source file. ++ ++@cindex linespec ++In general, the @code{list} command expects you to supply a ++@dfn{linespecs}. Linespecs specify source lines; there are several ways ++of writing them, but the effect is always to specify some source line. ++ ++Here is a complete description of the possible arguments for @code{list}: ++ ++@table @code ++@item list @var{linespec} ++Print lines centered around the line specified by @var{linespec}. ++ ++@item list @var{first} @var{increment} ++Print @var{increment} lines starting from @var{first} ++ ++@item list @var{first} ++Print lines starting with @var{first}. ++ ++@item list - ++Print lines just before the lines last printed. ++ ++@item list . ++Print lines after where the script is stopped. ++ ++@item list ++As described in the preceding table. ++@end table ++ ++Here are the ways of specifying a single source line---all the ++kinds of linespec. ++ ++@table @code ++@item @var{number} ++Specifies line @var{number} of the current source file. ++When a @code{list} command has two linespecs, this refers to ++the same source file as the first linespec. ++ ++@item @var{filename}:@var{number} ++Specifies line @var{number} in the source file @var{filename}. ++ ++@item @var{function} ++Specifies the line that function @var{function} is listed on. ++ ++@ifset FINISHED ++@item @var{filename}:@var{function} ++Specifies the line of function @var{function} in the file ++@var{filename}. You only need the file name with a function name to ++avoid ambiguity when there are identically named functions in ++different source files. ++@end ifset ++@end table ++ ++@node Search ++@section Searching source files (@samp{search}, @samp{reverse}, @samp{/.../}, @samp{?..?}) ++@cindex searching ++@kindex reverse-search ++ ++There are two commands for searching through the current source file for a ++BASH extended pattern-matching expression. ++ ++@table @code ++@kindex search ++@kindex forward ++@item forward @var{bash-pattern} ++@itemx search @var{bash-pattern} ++The command @samp{forward @var{bash-pattern}} checks each line, ++starting with the one following the current line, for a match for ++@var{bash-pattern} which is an extended bash pattern-matching ++expression. It lists the line that is found. You can use the synonym ++@samp{search @var{bash-pattern}} or abbreviate the command name as ++@code{fo} or @code{/@var{pat}/}. ++ ++@item reverse @var{bash-pattern} ++The command @samp{reverse @var{bash-pattern}} checks each line, starting ++with the one before the last line listed and going backward, for a match ++for @var{bash-pattern}. It lists the line that is found. You can abbreviate ++this command as @code{rev} or @code{?@var{bash-pattern}?}. ++@end table ++ ++@node Data ++@section Examining Data (@samp{print}, @samp{examine}, @samp{info variables}) ++ ++@cindex printing data ++@cindex examining data ++@kindex print ++ ++One way to examine string data in your script is with the @code{print} ++command (abbreviated @code{p}). However a more versatile print command ++is @code{x}; it can print variable and function definitions and can do ++arithmetic computations. Finally, the most general method would be ++via @code{eval echo}. ++ ++@table @code ++@kindex print ++@kindex p @r{(@code{print})} ++@item print @var{expr} ++ ++Use @code{print} to dispay strings as you would from @code{echo}. And ++as such, variable names to be substituted have to be preceded with a ++dollar sign. As with echo, filename expansion, e.g. tilde expansion, ++is performed on unquoted strings. So for example if you want to print ++a *, you would write @samp{print "*"}, not @samp{print *}. If you want ++to have the special characters dollars sign appear, use a backslash. ++ ++@item print ++@itemx p ++If you omit @var{expr}, @value{DBG} displays the last expression again. ++ ++@item x @ovar{expr} ++@kindex x ++ ++This is a smarter, more versatile ``print'' command, and although sometimes ++it might not be what you want, and you may want to resort to either ++@code{print} or @code{eval echo...}. ++ ++As with @code{print}, if you omit @var{expr}, @value{DBG} displays ++the last expression again. ++ ++The @code{x} command first checks if @var{expr} is single variable. If ++it is, the definition and value are printed via @value{BASH}'s ++@code{declare -p} command. This will show the variable's attributes ++such as if it is read only or if it is an integer. If the variable is ++an array, that is show and the array values are printed. ++ ++If instead @var{expr} is a function, the function definition is ++printed via @value{BASH}'s @code{declare -f} command. If @var{expr} ++was neither a variable nor an expression, then we try to get a value ++via @code{let}. And if this returns an error, as a last resort we call ++@code{print} and give what it outputs. ++ ++Since @code{let} may be used internally and since (to my thinking) ++@code{let} does funny things, the results may seem odd unless you ++understand the sequence tried above and how @code{let} works. For ++example if the variable @code{foo} has value 5, then @samp{x foo} show ++the definition of foo with value 5, and @samp{x foo+5} prints 10 as ++expected. So far so good. However if @code{foo} is has value ++``string'', @samp{x foo+5} prints 5 because @code{let} has turned the ++string into 0; @samp{p foo+5} will simply print ``foo+5''; if you want ++the value of ``foo'' substituted inside a string, for example you ++expect ``the value of foo is $foo'' to come out ``the value of foo is ++5'', then the right command to use is @code{print} rather than ++@code{x}, making sure you add the dollar onto the beginning of the ++variable. ++ ++@item V @ovar{!}@ovar{pattern} ++@kindex V ++ ++If you want to @emph{all} list variables and there values or a set of ++variables by pattern, use this command. ++ ++@smallexample ++@group ++bashdb<0> @b{V dq*} ++dq_args="dq_*" ++dq_cmd="V" ++@end group ++@end smallexample ++ ++@end table ++ ++@node Evaluation/Execution ++@section Running Arbitrary BASH and Shell commands (@samp{eval}, @samp{shell}) ++ ++The two most general commands and most ``low-level'' are @code{eval} ++and @code{shell}. ++ ++@table @code ++@item eval ++@itemx e ++@kindex e ++ ++In contrast to the commands of the last section the most general way ++to examine data is through @code{eval}. But you do much more with ++this; you can change the values of variables, since, you are just ++evaluating BASH code. ++ ++If you expect output, you should arrange that in the command, such as ++via @code{echo} or @code{printf}. For example, to print the value of ++@var{foo}, you would type @samp{e echo $foo}. This is bit longer than ++@samp{p $foo} or (when possible) @samp{x foo}. However suppose you ++wanted to find out how the builtin test operator @samp{[} works with ++the @samp{-z} test condition. You could use @code{eval} to do this ++such as @samp{e [ -z "$foo"] && echo "yes"}. ++ ++@kindex shell ++@kindex !! @r{(@code{shell})} ++@cindex shell escape ++@item shell @var{command string} ++@itemx !! ++ ++If you need to execute occasional shell commands during your ++debugging session, there is no need to leave or suspend @value{DBG}; you can ++just use the @code{shell} command or its alias @code{!!}. ++ ++Invoke a shell to execute @var{command string}. ++ ++@end table ++ ++@node Auto Display ++@section Automatic display (@samp{display}, @samp{undisplay}) ++@cindex automatic display ++@cindex display of expressions ++ ++If you find that you want to print the value of an expression ++frequently (to see how it changes), you might want to add it to the ++@dfn{automatic display list} so that @value{DBG} evaluates a ++statement each time your program stops. Each expression added to the ++list is given a number to identify it; to remove an expression from ++the list, you specify that number. The automatic display looks like ++this: ++ ++@example ++2 (echo $x): 38 ++@end example ++ ++@noindent ++This display shows item numbers, expressions and their current values. ++ ++@table @code ++@kindex display ++@item display @var{expr} ++Add the expression @var{expr} to the list of expressions to display ++each time your program stops. ++@end table ++ ++@table @code ++@kindex delete display ++@kindex undisplay ++@item undisplay @var{dnums}@dots{} ++@itemx delete display @var{dnums}@dots{} ++Remove item numbers @var{dnums} from the list of expressions to display. ++ ++@code{undisplay} does not repeat if you press @key{RET} after using it. ++(Otherwise you would just get the error @samp{No display number @dots{}}.) ++ ++@kindex disable display ++@item disable display @var{dnums}@dots{} ++Disable the display of item numbers @var{dnums}. A disabled display ++item is not printed automatically, but is not forgotten. It may be ++enabled again later. ++ ++@kindex enable display ++@item enable display @var{dnums}@dots{} ++Enable display of item numbers @var{dnums}. It becomes effective once ++again in auto display of its expression, until you specify otherwise. ++ ++@item display ++Display the current values of the expressions on the list, just as is ++done when your program stops. ++ ++@kindex info display ++@item info display ++Print the list of expressions previously set up to display ++automatically, each one with its item number, but without showing the ++values. This includes disabled expressions, which are marked as such. ++It also includes expressions which would not be displayed right now ++because they refer to automatic variables not currently available. ++@end table ++ ++@node Controlling bashdb ++@section Controlling @DBG (@samp{set}, @samp{history}) ++ ++You can alter the way @BASH interacts with you by using the ++@code{set} command. ++ ++@menu ++* Annotate:: Annotation Level (set annotate) ++* Prompt:: Prompt (set prompt, show prompt) ++* Editing:: Command editing (set editing, show editing) ++* Command Display:: Command display (set showcommand) ++* History:: Command history (history, !, H) ++@end menu ++ ++@node Annotate ++@subsection Annotation Level (@samp{set annoatate}) ++ ++The annotation level controls how much information does @value{DBG} ++print together with its prompt, values of expressions, source lines, ++and other types of output. Level 0 is the normal, level 1 is for use ++when @value{DBG} is run as a subprocess of @sc{gnu} Emacs of @value{DDD}, ++level 2 is the maximum annotation suitable for programs that control ++@value{DBGN}. ++ ++@node Prompt ++@subsection Prompt (@samp{set prompt}, @samp{show prompt}) ++ ++@cindex prompt ++ ++@value{dBGP} indicates its readiness to read a command by printing a ++string called the @dfn{prompt}. This string is normally: ++@example ++bashdb$@{_Dbg_less@}$@{#_Dbg_history[@@]@}$@{_Dbg_greater@}$_Dbg_space ++@end example ++ ++When variables inside the the prompt string are evaluated, the above ++becomes something like @samp{bashdb<5>} if this is the fifth command ++executed or perhaps @samp{bashdb<<2>>} if you have called the debugger ++from inside a debugger session and this is the second command inside ++the debugger session or perhaps @samp{bashdb<(6)>} if you ++entered a subshell after the fifth command. ++ ++You can change the prompt string with the @code{set prompt} command, ++although it is not normally advisable to do so without understanding ++the implications. If you are using the @value{DDD} GUI, it changes the ++changes the prompt and should not do so. In certain other ++circumstances (such as writing a GUI like @value{DDD}), it may be is useful ++to change the prompt. ++ ++@emph{Note:} @code{set prompt} does not add a space for you after the ++prompt you set. This allows you to set a prompt which ends in a space ++or a prompt that does not. Furthermore due to a implementation ++limitation (resulting from a limitation of the bash built-in function ++``read''), to put a space at the end of the prompt use the ++@samp{$_Dbg_space} variable. ++ ++@table @code ++@kindex set prompt ++@item set prompt @var{newprompt} ++Directs @value{DBG} to use @var{newprompt} as its prompt string ++henceforth. ++ ++@emph{Warning: changing the prompt can @value{DDD}'s ability to ++understand when the debugger is waiting for input.} ++ ++@kindex show prompt ++@item show prompt ++Prints a line of the form: @samp{bashdb's prompt is: @var{your-prompt}} ++@end table ++ ++@node Editing ++@subsection Command editing (@samp{set editing}, @samp{show editing}) ++@cindex readline ++@cindex command line editing ++ ++@value{DBG} reads its input commands through bash which uses via the ++@dfn{readline} interface. This @sc{gnu} library provides consistent ++behavior for programs which provide a command line interface to the ++user. Advantages are @sc{gnu} Emacs-style or @dfn{vi}-style inline ++editing of commands, @code{csh}-like history substitution, and a ++storage and recall of command history across debugging sessions. ++ ++You may control the behavior of command line editing in @BASH with the ++command @code{set}. ++ ++@table @code ++@kindex set editing ++@cindex editing ++@item set editing ++@itemx set editing on ++Enable command line editing (enabled by default). ++ ++@item set editing off ++Disable command line editing. ++ ++@kindex show editing ++@item show editing ++Show whether command line editing is enabled. ++@end table ++ ++ ++@node Command Display ++@subsection Command Display (@samp{set showcommand}) ++ ++The debugger normally lists the line number and source line of the for ++the statement to be next executed. Often this line contains one ++expression or one statement and it is clear from this line what's ++going to happen. However @BASH allows many expressions or ++statements to be put on a single source line; some lines ++contain several units of execution. Some examples of this ++behavior are listed below: ++ ++@smallexample ++x=1; y=2; x=3 ++(( x > 5 )) && x=5 ++y=`echo *` ++@end smallexample ++ ++In the first line of the example above, we have three assignment ++statements on a single line. In the second line of the example above ++we have a statement which gets run only if a condition tests true. And ++in the third line of the example above, we have a command that gets ++run and then the output of that is substituted in an assignemnt ++statement. If you were single stepping inside the debugger, each line ++might get listed more than once before each of the actions that might ++get performed. (In the case of the conditional statement, the ++line gets listed only once when the condition is false.) ++ ++In order to assist understanding where you are, the enhanced version ++of @BASH maintains a dynamic variable @env{BASH_COMMAND} that ++contains piece of code next to be run (or is currently being run). The ++debugger has arranged to save this and can display this information ++or not. This is controlled by @code{set showcommand}. ++ ++@table @code ++@kindex set showcommand ++@item set showcommand @r{[}auto | on | off @r{]} ++controls whether or not to show the saved @env{BASH_COMMAND} for the ++command next to be executed. ++@end table ++ ++When the value is @code{auto} the following heuristic is used to ++determine whether or not to display the saved @env{BASH_COMMAND}. If ++the last time you stopped you were at the same place and the command ++string has changed, then show the command. When the value @code{on} is ++used, the debugger always shows @env{BASH_COMMAND} and when ++@code{off} is used, the debugger nevers shows ++@env{BASH_COMMAND}. Note that listing the text of the source line is ++independent of whether or not the command is also listed. ++ ++Some examples: ++@smallexample ++set showcommand auto @b{This is the default} ++set showcommand on @b{Always show the next command to be executed} ++set showcommand off @b{Never show the next command to be executed} ++@end smallexample ++ ++@node History ++@subsection Command history (@samp{H}, @samp{history}, @samp{!}) ++ ++@value{dBGP} can keep track of the commands you type during your ++debugging sessions, so that you can be certain of precisely what ++happened. If the prompt has not been changed (see @ref{Prompt, ++,Prompt}), the history number that will be in use next is by default ++listed in the debugger prompt. Invalid commands and history commands ++are not saved on the history stack. ++ ++@table @code ++@kindex H @r{[}@var{start-number} @ovar{end-number}@r{]} ++@item H @r{[}@var{start-number} @ovar{end-number}@r{]} ++@item H @ovar{-count} ++@itemx !@r{[}-@r{]}@var{n}:p ++ ++You can list what is in the history stack with @code{H}. Debugger ++commands in ths history stack are listed from most recent to least recent. ++If no @var{start-number} is given we start with the most recently ++executed command and end with the first entry in the history stack. ++If @var{start-number} is given, that history number is listed first. If ++@var{end-number} is given, that history number is listed last. If a ++single negative number is given list that many history commands. ++ ++An alternate form is @code{!@emph{n}:p} or @code{!-@emph{n}:p} where ++@emph{n} is an integer. If a minus sign is used, @emph{n} is taken as ++the count to go back from the end rather than as a absolute history ++number. In contrast @code{H}, this form only prints a @emph{single} ++history item. ++ ++Some examples: ++@smallexample ++H @b{List entire history} ++H -2 @b{List the last two history items} ++!-2:p @b{List a single history item starting at the same place as above} ++H 5 @b{List history from history number 5 to the begining (number 0)} ++H 5 0 @b{Same as above} ++H 5 3 @b{List history from history number 5 down to history number 3} ++!5:p @b{List a single history item 5} ++@end smallexample ++ ++@kindex history @r{[}-@r{]}@r{[}@var{n}@r{]} ++@kindex !@r{[}-@r{]}@var{n} @r{(@code{history})} ++@item history @r{[}@r{[}-@r{]}@var{n}@r{]} ++@itemx !@r{[}-@r{]}@var{n} ++ ++Use this command to reexecute a given history number. If no number is ++given, the last debugger command in the history is executed. ++ ++An alternate form is @code{!@emph{n}} or @code{!-@emph{n}} where ++@emph{n} is an integer. ++ ++If a minus sign is used in in either form, @emph{n} is taken as the ++count to go back from the end rather than as a absolute history ++number. ++ ++@end table ++ ++@ifset FINISHED ++Use these commands to manage the @value{DBGN} command ++history facility. ++ ++@table @code ++@cindex history substitution ++@cindex history file ++@kindex set history filename ++@kindex GDBHISTFILE ++@item set history filename @var{fname} ++Set the name of the @BASH command history file to @var{fname}. ++This is the file where @BASH reads an initial command history ++list, and where it writes the command history from this session when it ++exits. You can access this list through history expansion or through ++the history command editing characters listed below. This file defaults ++to the value of the environment variable @code{GDBHISTFILE}, or to ++@file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable ++is not set. ++ ++@cindex history save ++@kindex set history save ++@item set history save ++@itemx set history save on ++Record command history in a file, whose name may be specified with the ++@code{set history filename} command. By default, this option is disabled. ++ ++@item set history save off ++Stop recording command history in a file. ++ ++@cindex history size ++@kindex set history size ++@item set history size @var{size} ++Set the number of commands which @BASH keeps in its history list. ++This defaults to the value of the environment variable ++@code{HISTSIZE}, or to 256 if this variable is not set. ++@end table ++ ++@cindex history expansion ++History expansion assigns special meaning to the character @kbd{!}. ++ ++Since @kbd{!} is also the logical not operator in C, history expansion ++is off by default. If you decide to enable history expansion with the ++@code{set history expansion on} command, you may sometimes need to ++follow @kbd{!} (when it is used as logical not, in an expression) with ++a space or a tab to prevent it from being expanded. The readline ++history facilities do not attempt substitution on the strings ++@kbd{!=} and @kbd{!(}, even when history expansion is enabled. ++ ++The commands to control history expansion are: ++ ++@table @code ++@kindex set history expansion ++@item set history expansion on ++@itemx set history expansion ++Enable history expansion. History expansion is off by default. ++ ++@item set history expansion off ++Disable history expansion. ++ ++The readline code comes with more complete documentation of ++editing and history expansion features. Users unfamiliar with @sc{gnu} Emacs ++or @code{vi} may wish to read it. ++ ++@c @group ++@kindex show history ++@item show history ++@itemx show history filename ++@itemx show history save ++@itemx show history size ++@itemx show history expansion ++These commands display the state of the @BASH history parameters. ++@code{show history} by itself displays all four states. ++@c @end group ++@end table ++ ++@table @code ++@kindex shows ++@item sh