[PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

[-- Attachment #1: Type: text/plain, Size: 189 bytes --]

Hi,

I am creating this new thread as the earlier threads went very long.
This patch contains code change, test-cases and documentation as well.
Please review this.

Thanks,
Abhijit Halder

[-- Attachment #2: gdb-pipe-command.patch --]
[-- Type: text/x-patch, Size: 12975 bytes --]

diff -rup src/gdb/ChangeLog dst/gdb/ChangeLog
--- src/gdb/ChangeLog	2011-08-14 07:27:51.502233004 +0530
+++ dst/gdb/ChangeLog	2011-08-24 22:16:45.326146038 +0530
@@ -1,3 +1,13 @@
+2011-08-29  Abhijit Halder  <abhijit.k.halder@gmail.com>
+
+	Implementation of pipe to make I/O communication
+	between gdb and shell.
+
+	* NEWS: Add news item for the new command "pipe".
+	* pipe.c: New file.
+	* Makefile.in (SFILES): Add pipe.c.
+	(COMMON_OBS): Add pipe.o.
+
 2011-08-12  Doug Evans  <dje@google.com>
 
 	* NEWS: Mention new "type" attribute of python gdb.Symbol objects.
diff -rup src/gdb/doc/ChangeLog dst/gdb/doc/ChangeLog
--- src/gdb/doc/ChangeLog	2011-08-14 07:27:53.158233001 +0530
+++ dst/gdb/doc/ChangeLog	2011-08-24 22:37:01.722146068 +0530
@@ -1,3 +1,7 @@
+2011-08-29  Abhijit Halder  <abhijit.k.halder@gmail.com>
+
+	* gdb.texinfo (pipe command): New node.
+
 2011-08-12  Doug Evans  <dje@google.com>
 
 	* gdb.texinfo (Symbols In Python): Document symbol.type.
diff -rup src/gdb/doc/gdb.texinfo dst/gdb/doc/gdb.texinfo
--- src/gdb/doc/gdb.texinfo	2011-08-14 07:27:53.242233001 +0530
+++ dst/gdb/doc/gdb.texinfo	2011-08-25 00:02:42.166145990 +0530
@@ -156,6 +156,7 @@ software in general.  We will miss him.
 * GDB/MI::                      @value{GDBN}'s Machine Interface.
 * Annotations::                 @value{GDBN}'s annotation interface.
 * JIT Interface::               Using the JIT debugging interface.
+* Output Redirection::          Redirecting @value{GDBN}'s command output to shell
 
 * GDB Bugs::                    Reporting bugs in @value{GDBN}
 
@@ -30958,6 +30959,24 @@ Set @code{action_flag} to @code{JIT_UNRE
 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
 and the JIT will leak the memory used for the associated symbol files.
 
+@node Output Redirection
+@chapter Redirecting @value{GDBN}'s Command Output to the Shell
+
+@value{GDBN}'s command output can be redirected to the shell using @code{pipe}
+command for further processing.  This enables inline processing of the output of
+a particular @value{GDBN} command without explicitly invoking the @code{shell}
+command.  Further, there is no need to store the output of the @value{GDBN}
+command in any file to pass it to the shell.  Even if the environment variable
+@code{SHELL} is set, the default shell (@file{/bin/sh} on Unix systems,
+@file{COMMAND.COM} on MS-DOS, etc.) is used.  The @code{pipe} command expects
+in its argument a delimiter string (without having any whitespace), followed by
+a @value{GDBN} command, followed by the same delimiter string and finally a
+shell command.  The delimiter should not contain any leading `-'.
+
+@smallexample
+(@value{GDBP}) @b{pipe | thread apply all bt | cat >./analyze_threads.log} 
+@end smallexample
+
 @node GDB Bugs
 @chapter Reporting Bugs in @value{GDBN}
 @cindex bugs in @value{GDBN}
diff -rup src/gdb/Makefile.in dst/gdb/Makefile.in
--- src/gdb/Makefile.in	2011-07-27 23:55:26.000000000 +0530
+++ dst/gdb/Makefile.in	2011-08-14 07:31:15.802233004 +0530
@@ -713,7 +713,7 @@ SFILES = ada-exp.y ada-lang.c ada-typepr
 	objc-exp.y objc-lang.c \
 	objfiles.c osabi.c observer.c osdata.c \
 	opencl-lang.c \
-	p-exp.y p-lang.c p-typeprint.c p-valprint.c parse.c printcmd.c \
+	p-exp.y p-lang.c p-typeprint.c p-valprint.c parse.c pipe.c printcmd.c \
 	proc-service.list progspace.c \
 	prologue-value.c psymtab.c \
 	regcache.c reggroups.c remote.c remote-fileio.c reverse.c \
@@ -870,7 +870,7 @@ COMMON_OBS = $(DEPFILES) $(CONFIG_OBS) $
 	mi-common.o \
 	event-loop.o event-top.o inf-loop.o completer.o \
 	gdbarch.o arch-utils.o gdbtypes.o osabi.o copying.o \
-	memattr.o mem-break.o target.o parse.o language.o buildsym.o \
+	memattr.o mem-break.o target.o parse.o pipe.o language.o buildsym.o \
 	findcmd.o \
 	std-regs.o \
 	signals.o \
diff -rup src/gdb/NEWS dst/gdb/NEWS
--- src/gdb/NEWS	2011-08-14 07:27:51.582233000 +0530
+++ dst/gdb/NEWS	2011-08-14 07:50:38.562234498 +0530
@@ -64,6 +64,9 @@
 
 * New commands "info macros", and "info definitions" have been added.
 
+* New command "pipe" has been added to make GDB command output available to the
+  shell for processing. 
+
 * Changed commands
 
 watch EXPRESSION mask MASK_VALUE
diff -rup src/gdb/pipe.c dst/gdb/pipe.c
--- src/gdb/pipe.c	2011-07-29 15:15:26.078048517 +0530
+++ dst/gdb/pipe.c	2011-08-23 22:38:53.320364947 +0530
@@ -0,0 +1,227 @@
+/* Everything about pipe, for GDB.
+
+   Copyright (C) 2011 Free Software Foundation, Inc.
+
+   This file is part of GDB.
+
+   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 3 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, see <http://www.gnu.org/licenses/>.  */
+
+#include "defs.h"
+#include <ctype.h>
+#include "gdb_string.h"
+#include "ui-file.h"
+#include "ui-out.h"
+#include "cli/cli-utils.h"
+#include "gdbcmd.h"
+#include "libiberty.h"
+#include "exceptions.h"
+
+#if defined(__MINGW32__)
+# define SHELL "cmd.exe"
+# define OPTION_TO_SHELL "/c"
+#else
+# define SHELL "/bin/sh"
+# define OPTION_TO_SHELL "-c"
+#endif
+
+/* Structure to encapsulate all entities associated with pipe.  */
+
+struct pipe_obj
+{
+  /* The delimiter to separate out gdb-command and shell-command.  This can be
+     any arbitrary string without containing any whitespace.  There should not
+     be any leading '-' in the delimiter.  */
+  char *dlim;
+
+  /* The gdb-command.  */
+  char *gdb_cmd;
+
+  /* The shell-command.  */
+  char *shell_cmd;
+
+  /* The gdb-side stream pointer to the pipe.  */
+  FILE *handle;
+
+  /* The pex object used to create pipeline between gdb and shell.  */
+  struct pex_obj *pex;
+};
+
+/* Destruct pipe object referenced by ARG.  */
+
+static void
+destruct_pipe (void *arg)
+{
+  struct pipe_obj *pipe = (struct pipe_obj *) arg;
+
+  if (pipe->handle != NULL)
+    fclose (pipe->handle);
+
+  if (pipe->pex != NULL)
+    {
+      int status;
+
+      /* Wait till the process on the other side of the pipe completes its
+	 job before closing its file descriptors.  */
+      pex_get_status (pipe->pex, 1, &status);
+
+      if (!WIFEXITED (status))
+	warning (_("Execution of shell-command `%s' may not be completed"),
+		 pipe->shell_cmd);
+
+      pex_free (pipe->pex);
+    }
+
+  xfree (pipe->dlim);
+  xfree (pipe);
+}
+
+/* Construct a pipe object by parsing argument ARG to the pipe command.  */
+
+static struct pipe_obj *
+construct_pipe (const char *arg)
+{
+  char *p, *t;
+  struct pipe_obj *pipe = NULL;
+  struct cleanup *cleanup;
+
+  if (arg == NULL)
+    error (_("No argument is specified"));
+
+  pipe = XCNEW (struct pipe_obj);
+  cleanup = make_cleanup (destruct_pipe, pipe);
+
+  p = xstrdup (arg);
+  pipe->dlim = p;
+
+  if (*pipe->dlim == '-')
+    error (_("Delimiter pattern should not start with '-'"));
+
+  t = skip_to_space (p);
+  p = skip_spaces (t);
+
+  if (*p == '\0')
+    error (_("No gdb-command is specified"));
+
+  *t = '\0';
+  pipe->gdb_cmd = p;
+
+  for (;;)
+    {
+      t = skip_to_space (p);
+
+      if (*t == '\0')
+	error (_("No shell-command is specified"));
+
+      /* Check whether the token separated by whitespace matches with
+	 delimiter.  */ 
+      if (memcmp (p, pipe->dlim, (t - p)) == 0
+	  && pipe->dlim[t - p] == '\0')
+	{
+	  *p = '\0';
+	  pipe->shell_cmd = skip_spaces (t);
+	  break;
+	}
+
+       p = skip_spaces (t);
+    }
+
+  discard_cleanups (cleanup);
+  return pipe;
+}
+
+/* Run execute_command for PIPE and FROM_TTY.  Write output to the pipe, do not
+   display it to the screen.  */
+
+static void
+execute_command_to_pipe (struct pipe_obj *pipe, int from_tty)
+{
+  char *argv[4];
+  struct cleanup *cleanup;
+  struct ui_file *fp;
+  int status;
+  const char *errmsg;
+  volatile struct gdb_exception exception;
+
+  argv[0] = SHELL;
+  argv[1] = OPTION_TO_SHELL;
+  argv[2] = pipe->shell_cmd;
+  argv[3] = NULL;
+
+  pipe->pex = pex_init (PEX_USE_PIPES, argv[0], NULL);
+  pipe->handle = pex_input_pipe (pipe->pex, 0);
+
+  if (pipe->handle == NULL)
+    error (_("Failed to create pipe"));
+
+  errmsg = pex_run (pipe->pex, PEX_LAST, argv[0], argv, NULL, NULL, &status);
+
+  if (errmsg != NULL)
+    error (_("Failed to execute shell-command `%s' on pipe: %s %s"),
+	   pipe->shell_cmd, errmsg, safe_strerror (status));
+
+  /* GDB_STDOUT should be better already restored during these
+     restoration callbacks.  */
+  cleanup = set_batch_flag_and_make_cleanup_restore_page_info ();
+  fp = stdio_fileopen (pipe->handle);
+  make_cleanup_ui_file_delete (fp);
+  make_cleanup_restore_ui_file (&gdb_stdout);
+  make_cleanup_restore_ui_file (&gdb_stderr);
+  make_cleanup_restore_ui_file (&gdb_stdlog);
+  make_cleanup_restore_ui_file (&gdb_stdtarg);
+  make_cleanup_restore_ui_file (&gdb_stdtargerr);
+
+  if (ui_out_redirect (current_uiout, fp) < 0)
+    warning (_("Current output protocol does not support redirection"));
+  else 
+    make_cleanup_ui_out_redirect_pop (current_uiout);
+
+  gdb_stdout = fp;
+  gdb_stderr = fp;
+  gdb_stdlog = fp;
+  gdb_stdtarg = fp;
+  gdb_stdtargerr = fp;
+
+  TRY_CATCH (exception, RETURN_MASK_ERROR)
+    {
+      execute_command (pipe->gdb_cmd, from_tty);
+    }
+
+  if (exception.reason < 0)
+    exception_print (gdb_stderr, exception);
+
+  do_cleanups (cleanup);
+}
+
+/* Execute the pipe command with argument ARG and FROM_TTY.  */
+
+static void
+pipe_command (char *arg, int from_tty)
+{
+  struct pipe_obj *pipe = construct_pipe (arg);
+  struct cleanup *cleanup = make_cleanup (destruct_pipe, pipe);
+
+  execute_command_to_pipe (pipe, from_tty);
+  do_cleanups (cleanup);
+}
+
+/* Module initialization.  */
+
+void
+_initialize_pipe (void)
+{
+  add_cmd ("pipe", no_class, pipe_command, _("\
+Create pipe to pass gdb-command output to the shell for processing.\n\
+Arguments are a delimiter, followed by a gdb-command, then the same delimiter \
+again and finally a shell-command.  The delimiter can be any string \
+containing no whitespace and should not have any leading '-'."),
+	   &cmdlist);
+}
diff -rup src/gdb/testsuite/ChangeLog dst/gdb/testsuite/ChangeLog
--- src/gdb/testsuite/ChangeLog	2011-08-14 07:27:53.706232999 +0530
+++ dst/gdb/testsuite/ChangeLog	2011-08-24 22:25:47.670145996 +0530
@@ -1,3 +1,7 @@
+2011-08-29  Abhijit Halder  <abhijit.k.halder@gmail.com>
+
+	* gdb.base/pipe.exp: New file.
+
 2011-08-12  Doug Evans  <dje@google.com>
 
 	* gdb.python/py-symbol.exp: Add test for symbol.type.
diff -rup src/gdb/testsuite/gdb.base/pipe.exp dst/gdb/testsuite/gdb.base/pipe.exp
--- src/gdb/testsuite/gdb.base/pipe.exp	2011-08-16 22:37:45.969351119 +0530
+++ dst/gdb/testsuite/gdb.base/pipe.exp	2011-08-23 21:51:20.408367391 +0530
@@ -0,0 +1,52 @@
+# Copyright 2011 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
+# the Free Software Foundation; either version 3 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, see <http://www.gnu.org/licenses/>.
+
+#
+# test gdb pipe commands
+#
+
+set test "pipe"
+set tempfile ${objdir}/${subdir}/${test}.x
+
+gdb_exit
+gdb_start
+
+gdb_test "pipe" "No argument is specified"
+gdb_test "pipe |" "No gdb-command is specified"
+gdb_test "pipe -x" "Delimiter pattern should not start with '-'"
+gdb_test "pipe | print 'x'" "No shell-command is specified"
+gdb_test "pipe | print 'x' |< cat" "No shell-command is specified"
+gdb_test "pipe |< print 'x' | cat" "No shell-command is specified"
+gdb_test "pipe | print 'x' >| cat" "No shell-command is specified"
+gdb_test "pipe >| print 'x' | cat" "No shell-command is specified"
+gdb_test "pipe | print 'x' | cat" " = 120 'x'"
+gdb_test "pipe | print 'x' | cat | cat" " = 120 'x'"
+
+file delete $tempfile
+gdb_test_no_output "pipe | p 'x' | cat >$tempfile"
+if ![file exists $tempfile] then {
+    perror "$tempfile does not exist."
+    return 0
+} else {
+    set fp [open $tempfile r]
+    set fdata [read $fp]
+    close $fp
+    regsub -all {\$[0-9]+} $fdata {} pattern
+    if ![string match $pattern " = 120 'x'\n"] then {
+        fail $test
+    } else {
+        pass $test
+    }
+}

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Fri, 26 Aug 2011 17:13:28 +0530
> From: Abhijit Halder <abhijit.k.halder@gmail.com>
> Cc: Eli Zaretskii <eliz@gnu.org>, Sergio Durigan Junior <sergiodj@redhat.com>, 
> 	Jan Kratochvil <jan.kratochvil@redhat.com>, Tom Tromey <tromey@redhat.com>, 
> 	Pedro Alves <pedro@codesourcery.com>
> 
> I am creating this new thread as the earlier threads went very long.
> This patch contains code change, test-cases and documentation as well.

Thanks.

> +@node Output Redirection
> +@chapter Redirecting @value{GDBN}'s Command Output to the Shell

I'd rather not make this a new chapter.  I think a subsection of the
Input/Output node is a much better place.

> +@value{GDBN}'s command output can be redirected to the shell using @code{pipe}
                                                   ^^^^^^^^^^^^
"to a shell command", I think.

>                                        Even if the environment variable
> +@code{SHELL} is set, the default shell (@file{/bin/sh} on Unix systems,
> +@file{COMMAND.COM} on MS-DOS, etc.) is used.

This might be seen by some as a

>                                           The @code{pipe} command expects
> +in its argument a delimiter string (without having any whitespace), followed by
> +a @value{GDBN} command, followed by the same delimiter string and finally a
> +shell command.  The delimiter should not contain any leading `-'.
> +
> +@smallexample
> +(@value{GDBP}) @b{pipe | thread apply all bt | cat >./analyze_threads.log} 
> +@end smallexample

Please explain which part is the "delimiter" here.  Since you are
using the "normal" shell pipe character `|', the above explanation
about the delimiter string will raise a few brows.  A word about why
you need something beyond the usual `|' would also be nice.

In addition, I think we should explain which parts of the command are
passed to the shell verbatim and which parts are processed by GDB.  If
the last part ("cat >./analyze_threads.log") is passed to the shell,
then it must use the shell syntax (e.g., backslashes on MS-DOS/Windows).

Finally, can we have a more useful example than just redirecting the
output to a file?  Such redirection can be achieved without this
command at all.

Please also remove the @b markup, it is not used anywhere in the
examples we have in the manual.

> +* New command "pipe" has been added to make GDB command output available to the
> +  shell for processing. 

"available to shell commands"

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

On Fri, Aug 26, 2011 at 5:48 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> Date: Fri, 26 Aug 2011 17:13:28 +0530
>> From: Abhijit Halder <abhijit.k.halder@gmail.com>
>> Cc: Eli Zaretskii <eliz@gnu.org>, Sergio Durigan Junior <sergiodj@redhat.com>,
>>       Jan Kratochvil <jan.kratochvil@redhat.com>, Tom Tromey <tromey@redhat.com>,
>>       Pedro Alves <pedro@codesourcery.com>
>>
>> I am creating this new thread as the earlier threads went very long.
>> This patch contains code change, test-cases and documentation as well.
>
> Thanks.
>
>> +@node Output Redirection
>> +@chapter Redirecting @value{GDBN}'s Command Output to the Shell
>
> I'd rather not make this a new chapter.  I think a subsection of the
> Input/Output node is a much better place.
>
Actually I could not find such section. Can someone please direct me
to the appropriate section?

>> +@value{GDBN}'s command output can be redirected to the shell using @code{pipe}
>                                                   ^^^^^^^^^^^^
> "to a shell command", I think.
Yes. I will correct this.

>
>>                                        Even if the environment variable
>> +@code{SHELL} is set, the default shell (@file{/bin/sh} on Unix systems,
>> +@file{COMMAND.COM} on MS-DOS, etc.) is used.
>
> This might be seen by some as a
>
Sorry I am getting this. Please give me some more context.

>>                                           The @code{pipe} command expects
>> +in its argument a delimiter string (without having any whitespace), followed by
>> +a @value{GDBN} command, followed by the same delimiter string and finally a
>> +shell command.  The delimiter should not contain any leading `-'.
>> +
>> +@smallexample
>> +(@value{GDBP}) @b{pipe | thread apply all bt | cat >./analyze_threads.log}
>> +@end smallexample
>
> Please explain which part is the "delimiter" here.  Since you are
> using the "normal" shell pipe character `|', the above explanation
> about the delimiter string will raise a few brows.  A word about why
> you need something beyond the usual `|' would also be nice.
>
Sure.

> In addition, I think we should explain which parts of the command are
> passed to the shell verbatim and which parts are processed by GDB.  If
> the last part ("cat >./analyze_threads.log") is passed to the shell,
> then it must use the shell syntax (e.g., backslashes on MS-DOS/Windows).
>
Sure.

> Finally, can we have a more useful example than just redirecting the
> output to a file?  Such redirection can be achieved without this
> command at all.
>
Sure. I will try to provide some useful application.

> Please also remove the @b markup, it is not used anywhere in the
> examples we have in the manual.
>
Ok, I think when we need to show the command output along with the
command itself, the only the command has to be within @b markup.

>> +* New command "pipe" has been added to make GDB command output available to the
>> +  shell for processing.
>
> "available to shell commands"
>

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Sun, 28 Aug 2011 14:21:43 +0530
> From: Abhijit Halder <abhijit.k.halder@gmail.com>
> Cc: gdb-patches@sourceware.org, sergiodj@redhat.com, jan.kratochvil@redhat.com, 	tromey@redhat.com, pedro@codesourcery.com
> 
> >>                                        Even if the environment variable
> >> +@code{SHELL} is set, the default shell (@file{/bin/sh} on Unix systems,
> >> +@file{COMMAND.COM} on MS-DOS, etc.) is used.
> >
> > This might be seen by some as a
> >
> Sorry I am getting this. Please give me some more context.

It was supposed to be deleted, please ignore this part.

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Sun, 28 Aug 2011 14:21:43 +0530
> From: Abhijit Halder <abhijit.k.halder@gmail.com>
> Cc: gdb-patches@sourceware.org, sergiodj@redhat.com, jan.kratochvil@redhat.com, 
> 	tromey@redhat.com, pedro@codesourcery.com
> 
> >> +@node Output Redirection
> >> +@chapter Redirecting @value{GDBN}'s Command Output to the Shell
> >
> > I'd rather not make this a new chapter.  I think a subsection of the
> > Input/Output node is a much better place.
> >
> Actually I could not find such section. Can someone please direct me
> to the appropriate section?

Search for this:

  @node Input/Output
  @section Your Program's Input and Output

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Sun, 28 Aug 2011 05:50:49 -0400
> From: Eli Zaretskii <eliz@gnu.org>
> CC: gdb-patches@sourceware.org, sergiodj@redhat.com,	jan.kratochvil@redhat.com, tromey@redhat.com,	pedro@codesourcery.com
> Reply-to: Eli Zaretskii <eliz@gnu.org>
> 
> > Date: Sun, 28 Aug 2011 14:21:43 +0530
> > From: Abhijit Halder <abhijit.k.halder@gmail.com>
> > Cc: gdb-patches@sourceware.org, sergiodj@redhat.com, jan.kratochvil@redhat.com, 
> > 	tromey@redhat.com, pedro@codesourcery.com
> > 
> > >> +@node Output Redirection
> > >> +@chapter Redirecting @value{GDBN}'s Command Output to the Shell
> > >
> > > I'd rather not make this a new chapter.  I think a subsection of the
> > > Input/Output node is a much better place.
> > >
> > Actually I could not find such section. Can someone please direct me
> > to the appropriate section?
> 
> Search for this:
> 
>   @node Input/Output
>   @section Your Program's Input and Output

An even better place is this:

  @node Shell Commands
  @section Shell Commands

Sorry that I didn't think about this before.

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

[-- Attachment #1: Type: text/plain, Size: 1286 bytes --]

On Sun, Aug 28, 2011 at 3:41 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> Date: Sun, 28 Aug 2011 05:50:49 -0400
>> From: Eli Zaretskii <eliz@gnu.org>
>> CC: gdb-patches@sourceware.org, sergiodj@redhat.com,  jan.kratochvil@redhat.com, tromey@redhat.com,   pedro@codesourcery.com
>> Reply-to: Eli Zaretskii <eliz@gnu.org>
>>
>> > Date: Sun, 28 Aug 2011 14:21:43 +0530
>> > From: Abhijit Halder <abhijit.k.halder@gmail.com>
>> > Cc: gdb-patches@sourceware.org, sergiodj@redhat.com, jan.kratochvil@redhat.com,
>> >     tromey@redhat.com, pedro@codesourcery.com
>> >
>> > >> +@node Output Redirection
>> > >> +@chapter Redirecting @value{GDBN}'s Command Output to the Shell
>> > >
>> > > I'd rather not make this a new chapter.  I think a subsection of the
>> > > Input/Output node is a much better place.
>> > >
>> > Actually I could not find such section. Can someone please direct me
>> > to the appropriate section?
>>
>> Search for this:
>>
>>   @node Input/Output
>>   @section Your Program's Input and Output
>
> An even better place is this:
>
>  @node Shell Commands
>  @section Shell Commands
>
> Sorry that I didn't think about this before.
>

Hi,
I have made the corrections in the doc section. Please review the same.

Thanks,
Abhijit Halder

[-- Attachment #2: gdb-pipe-command-doc.patch --]
[-- Type: text/x-patch, Size: 2818 bytes --]

diff -rup src/gdb/doc/ChangeLog dst/gdb/doc/ChangeLog
--- src/gdb/doc/ChangeLog	2011-08-27 20:35:38.459934029 +0530
+++ dst/gdb/doc/ChangeLog	2011-08-28 16:24:53.024714073 +0530
@@ -1,3 +1,7 @@
+2011-08-29  Abhijit Halder  <abhijit.k.halder@gmail.com>
+
+	* gdb.texinfo (pipe command): New node.
+
 2011-08-25  Andrew Oakley  <andrew@ado.is-a-geek.net>
 
 	* gdb.texinfo (Types In Python): Document 'bitpos' for enums.
diff -rup src/gdb/doc/gdb.texinfo dst/gdb/doc/gdb.texinfo
--- src/gdb/doc/gdb.texinfo	2011-08-26 21:11:39.938354007 +0530
+++ dst/gdb/doc/gdb.texinfo	2011-08-30 18:39:03.200191037 +0530
@@ -1367,6 +1367,53 @@ Execute the @code{make} program with the
 arguments.  This is equivalent to @samp{shell make @var{make-args}}.
 @end table
 
+If you want to process the output of a @value{GDBN} command using some shell
+command or some script, that can be done by using the command @code{pipe}.  You
+don't even need to get a shell for that.  The same could be achieved by
+enabling logging and redirecting the output of @value{GDBN} command to some
+file and then processing the file content by invoking a shell using
+@code{shell} command.  But this incurs higher costs as the explicit invokation
+of shell involves execution of its startup scripts and locating the file
+containing @value{GDBN}'s command output involves directory lookup; even though
+OS may optimize these processes by its cache mechanism.
+
+@table @code
+@item pipe @var{args}
+The @var{args} consists of a delimiter followed by a @value{GDBN} command,
+followed by the same delimiter and finally a shell command.  The delimiter is a
+string of arbitrary length, containing no whitespace and no leading '-', acts
+as a separator between a @value{GDBN} command and a shell command.  The shell
+command should be in compliance with the syntax of the default shell.
+@end table
+
+@smallexample
+(@value{GDBP}) @b{ptype dd_tbl}
+type = struct dd @{
+    int dd_handle;
+    const char *dd_name;
+    int dd_major;
+    int dd_minor;
+    void *dd_code;
+    void *dd_data;
+@} [1024]
+(@value{GDBP}) @b{pipe <br> print dd_tbl <br> sed 's/@}/\n/g' | grep @
+"\.test_dd" | tr ',' '\n'}
+
+  @{dd_handle = 10
+  dd_name = 0x8048538 ".test_dd"
+  dd_major = 100
+  dd_minor = 0
+  dd_code = 0xcc
+  dd_data = 0x80
+@end smallexample
+
+In the above example @samp{@var{<br>}} acts as a delimiter.  The output of
+@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
+grep ".test_dd" | tr ',' '\n'} for processing.
+
+In the given example the output of @value{GDBN} command is huge and not well
+formated.  The use of shell commands like ``sed'', ``tr'' and ``grep'' ease
+the searching of desired pattern and hence ease debugging.
+
 @node Logging Output
 @section Logging Output
 @cindex logging @value{GDBN} output

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Pedro Alves]

On Tuesday 30 August 2011 14:58:02, Abhijit Halder wrote:
> +@table @code
> +@item pipe @var{args}

I suggest writing this as something like:

@item pipe @var{delim} @var{gdbcmd} @var{delim} @var{shellcmd}

and then adjust the text a bit to refer to the variables by name.
The syntax is more immediately obvious this way.

> +The @var{args} consists of a delimiter followed by a @value{GDBN} command,
> +followed by the same delimiter and finally a shell command.  The delimiter is a
> +string of arbitrary length, containing no whitespace and no leading '-', acts
> +as a separator between a @value{GDBN} command and a shell command.  The shell
> +command should be in compliance with the syntax of the default shell.
> +@end table

-- 
Pedro Alves

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

On Tue, Aug 30, 2011 at 7:57 PM, Pedro Alves <pedro@codesourcery.com> wrote:
> On Tuesday 30 August 2011 14:58:02, Abhijit Halder wrote:
>> +@table @code
>> +@item pipe @var{args}
>
> I suggest writing this as something like:
>
> @item pipe @var{delim} @var{gdbcmd} @var{delim} @var{shellcmd}
>
> and then adjust the text a bit to refer to the variables by name.
> The syntax is more immediately obvious this way.
>
>> +The @var{args} consists of a delimiter followed by a @value{GDBN} command,
>> +followed by the same delimiter and finally a shell command.  The delimiter is a
>> +string of arbitrary length, containing no whitespace and no leading '-', acts
>> +as a separator between a @value{GDBN} command and a shell command.  The shell
>> +command should be in compliance with the syntax of the default shell.
>> +@end table
>
> --
> Pedro Alves
>

Sure. I am going to change this.

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

On Tue, Aug 30, 2011 at 8:01 PM, Abhijit Halder
<abhijit.k.halder@gmail.com> wrote:
> On Tue, Aug 30, 2011 at 7:57 PM, Pedro Alves <pedro@codesourcery.com> wrote:
>> On Tuesday 30 August 2011 14:58:02, Abhijit Halder wrote:
>>> +@table @code
>>> +@item pipe @var{args}
>>
>> I suggest writing this as something like:
>>
>> @item pipe @var{delim} @var{gdbcmd} @var{delim} @var{shellcmd}
>>
>> and then adjust the text a bit to refer to the variables by name.
>> The syntax is more immediately obvious this way.
>>
>>> +The @var{args} consists of a delimiter followed by a @value{GDBN} command,
>>> +followed by the same delimiter and finally a shell command.  The delimiter is a
>>> +string of arbitrary length, containing no whitespace and no leading '-', acts
>>> +as a separator between a @value{GDBN} command and a shell command.  The shell
>>> +command should be in compliance with the syntax of the default shell.
>>> +@end table
>>
>> --
>> Pedro Alves
>>
>
> Sure. I am going to change this.
>

Hi all,

In gdb.texinfo file I have seen @cindex and @kindex markup. Someone
please tell me the significance of them.
Secondly, the example I have used is from a sample code written by me
just for this purpose. This should not have any licensing issue I
hope.

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

[-- Attachment #1: Type: text/plain, Size: 1491 bytes --]

On Tue, Aug 30, 2011 at 8:18 PM, Abhijit Halder
<abhijit.k.halder@gmail.com> wrote:
> On Tue, Aug 30, 2011 at 8:01 PM, Abhijit Halder
> <abhijit.k.halder@gmail.com> wrote:
>> On Tue, Aug 30, 2011 at 7:57 PM, Pedro Alves <pedro@codesourcery.com> wrote:
>>> On Tuesday 30 August 2011 14:58:02, Abhijit Halder wrote:
>>>> +@table @code
>>>> +@item pipe @var{args}
>>>
>>> I suggest writing this as something like:
>>>
>>> @item pipe @var{delim} @var{gdbcmd} @var{delim} @var{shellcmd}
>>>
>>> and then adjust the text a bit to refer to the variables by name.
>>> The syntax is more immediately obvious this way.
>>>
>>>> +The @var{args} consists of a delimiter followed by a @value{GDBN} command,
>>>> +followed by the same delimiter and finally a shell command.  The delimiter is a
>>>> +string of arbitrary length, containing no whitespace and no leading '-', acts
>>>> +as a separator between a @value{GDBN} command and a shell command.  The shell
>>>> +command should be in compliance with the syntax of the default shell.
>>>> +@end table
>>>
>>> --
>>> Pedro Alves
>>>
>>
>> Sure. I am going to change this.
>>
>
> Hi all,
>
> In gdb.texinfo file I have seen @cindex and @kindex markup. Someone
> please tell me the significance of them.
> Secondly, the example I have used is from a sample code written by me
> just for this purpose. This should not have any licensing issue I
> hope.
>

Changes as suggested by Pedro.

Regards,
Abhijit Halder

[-- Attachment #2: gdb-pipe-command-doc.patch --]
[-- Type: text/x-patch, Size: 2745 bytes --]

diff -rup src/gdb/doc/ChangeLog dst/gdb/doc/ChangeLog
--- src/gdb/doc/ChangeLog	2011-08-27 20:35:38.459934029 +0530
+++ dst/gdb/doc/ChangeLog	2011-08-28 16:24:53.024714073 +0530
@@ -1,3 +1,7 @@
+2011-08-29  Abhijit Halder  <abhijit.k.halder@gmail.com>
+
+	* gdb.texinfo (pipe command): New node.
+
 2011-08-25  Andrew Oakley  <andrew@ado.is-a-geek.net>
 
 	* gdb.texinfo (Types In Python): Document 'bitpos' for enums.
diff -rup src/gdb/doc/gdb.texinfo dst/gdb/doc/gdb.texinfo
--- src/gdb/doc/gdb.texinfo	2011-08-26 21:11:39.938354007 +0530
+++ dst/gdb/doc/gdb.texinfo	2011-08-30 20:09:24.120190599 +0530
@@ -1367,6 +1367,53 @@ Execute the @code{make} program with the
 arguments.  This is equivalent to @samp{shell make @var{make-args}}.
 @end table
 
+If you want to process the output of a @value{GDBN} command using some shell
+command or some script, that can be done by using the command @code{pipe}.  You
+don't even need to get a shell for that.  The same could be achieved by
+enabling logging and redirecting the output of @value{GDBN} command to some
+file and then processing the file content by invoking a shell using
+@code{shell} command.  But this incurs higher costs as the explicit invokation
+of shell involves execution of its startup scripts and locating the file
+containing @value{GDBN}'s command output involves directory lookup; even though
+OS may optimize these processes by its cache mechanism.
+
+@table @code
+@item pipe @var{dlim} @var{gdbcmd} @var{dlim} @var{shellcmd}
+@var{dlim} is a string of arbitrary length, containing no whitespace and no
+leading '-', acts as a separator between a @value{GDBN} command @var{gdbcmd}
+and a shell command @var{shellcmd}.  The shell command should be in compliance
+with the syntax of the default shell.
+@end table
+
+@smallexample
+(@value{GDBP}) @b{ptype dd_tbl}
+type = struct dd @{
+    int dd_handle;
+    const char *dd_name;
+    int dd_major;
+    int dd_minor;
+    void *dd_code;
+    void *dd_data;
+@} [1024]
+(@value{GDBP}) @b{pipe <br> print dd_tbl <br> sed 's/@}/\n/g' | grep @
+"\.test_dd" | tr ',' '\n'}
+
+  @{dd_handle = 10
+  dd_name = 0x8048538 ".test_dd"
+  dd_major = 100
+  dd_minor = 0
+  dd_code = 0xcc
+  dd_data = 0x80
+@end smallexample
+
+In the above example @samp{@var{<br>}} acts as a delimiter.  The output of
+@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
+grep ".test_dd" | tr ',' '\n'} for processing.
+
+In the given example the output of @value{GDBN} command is huge and not well
+formated.  The use of shell commands like ``sed'', ``tr'' and ``grep'' ease
+the searching of desired pattern and hence ease debugging.
+
 @node Logging Output
 @section Logging Output
 @cindex logging @value{GDBN} output

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Tue, 30 Aug 2011 22:08:22 +0530
> From: Abhijit Halder <abhijit.k.halder@gmail.com>
> Cc: Eli Zaretskii <eliz@gnu.org>, gdb-patches@sourceware.org, sergiodj@redhat.com, 
> 	jan.kratochvil@redhat.com, tromey@redhat.com
> 
> +	* gdb.texinfo (pipe command): New node.

You didn't write a new node, you just added text to the existing node.
So you should state the name of that node in the parentheses.

> +If you want to process the output of a @value{GDBN} command using some shell
> +command or some script, that can be done by using the command @code{pipe}.  You
> +don't even need to get a shell for that.  The same could be achieved by
> +enabling logging and redirecting the output of @value{GDBN} command to some
> +file and then processing the file content by invoking a shell using
> +@code{shell} command.  But this incurs higher costs as the explicit invokation
> +of shell involves execution of its startup scripts and locating the file
> +containing @value{GDBN}'s command output involves directory lookup; even though
> +OS may optimize these processes by its cache mechanism.

I would drop this part, and leave only the first sentence.  We don't
need to justify commands in the manual.

> +@table @code
> +@item pipe @var{dlim} @var{gdbcmd} @var{dlim} @var{shellcmd}

You need a "@kindex pipe" here, to index this command as we do with
all other commands.

> +@var{dlim} is a string of arbitrary length, containing no whitespace and no
> +leading '-', acts as a separator between a @value{GDBN} command @var{gdbcmd}
           ^^^
@samp{-}

> +In the above example @samp{@var{<br>}} acts as a delimiter.  The output of

Just @samp{<br>}, without @var.

> +@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
> +grep ".test_dd" | tr ',' '\n'} for processing.

I would suggest to add that the example does not use `|' as a
delimiter to allow the shell command to use it in its usual shell
semantics.

> +In the given example the output of @value{GDBN} command is huge and not well
> +formated.  The use of shell commands like ``sed'', ``tr'' and ``grep'' ease

Please use @command{sed}, @command{tr}, etc.

Thanks.

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Tue, 30 Aug 2011 20:18:41 +0530
> From: Abhijit Halder <abhijit.k.halder@gmail.com>
> Cc: Eli Zaretskii <eliz@gnu.org>, gdb-patches@sourceware.org, sergiodj@redhat.com, 
> 	jan.kratochvil@redhat.com, tromey@redhat.com
> 
> In gdb.texinfo file I have seen @cindex and @kindex markup. Someone
> please tell me the significance of them.

Each one adds an entry to a different index: @cindex to the Concept
Index section, @kindex to the Command Index.

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

[-- Attachment #1: Type: text/plain, Size: 2687 bytes --]

On Tue, Aug 30, 2011 at 11:36 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> Date: Tue, 30 Aug 2011 22:08:22 +0530
>> From: Abhijit Halder <abhijit.k.halder@gmail.com>
>> Cc: Eli Zaretskii <eliz@gnu.org>, gdb-patches@sourceware.org, sergiodj@redhat.com,
>>       jan.kratochvil@redhat.com, tromey@redhat.com
>>
>> +     * gdb.texinfo (pipe command): New node.
>
> You didn't write a new node, you just added text to the existing node.
> So you should state the name of that node in the parentheses.
>
Sorry, I missed editing this part.

>> +If you want to process the output of a @value{GDBN} command using some shell
>> +command or some script, that can be done by using the command @code{pipe}.  You
>> +don't even need to get a shell for that.  The same could be achieved by
>> +enabling logging and redirecting the output of @value{GDBN} command to some
>> +file and then processing the file content by invoking a shell using
>> +@code{shell} command.  But this incurs higher costs as the explicit invokation
>> +of shell involves execution of its startup scripts and locating the file
>> +containing @value{GDBN}'s command output involves directory lookup; even though
>> +OS may optimize these processes by its cache mechanism.
>
> I would drop this part, and leave only the first sentence.  We don't
> need to justify commands in the manual.
>
>> +@table @code
>> +@item pipe @var{dlim} @var{gdbcmd} @var{dlim} @var{shellcmd}
>
> You need a "@kindex pipe" here, to index this command as we do with
> all other commands.
>
>> +@var{dlim} is a string of arbitrary length, containing no whitespace and no
>> +leading '-', acts as a separator between a @value{GDBN} command @var{gdbcmd}
>           ^^^
> @samp{-}
>
>> +In the above example @samp{@var{<br>}} acts as a delimiter.  The output of
>
> Just @samp{<br>}, without @var.
>
>> +@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
>> +grep ".test_dd" | tr ',' '\n'} for processing.
>
> I would suggest to add that the example does not use `|' as a
> delimiter to allow the shell command to use it in its usual shell
> semantics.
>
In the above example one can use `|' as the delimiter. it will not
break anything syntactically but will not be readable much, hence I
avoided using `|' as delimiter. Do we need to mention that here?

>
>> +In the given example the output of @value{GDBN} command is huge and not well
>> +formated.  The use of shell commands like ``sed'', ``tr'' and ``grep'' ease
>
> Please use @command{sed}, @command{tr}, etc.
>
> Thanks.
>

I made the corrections as suggested. Please review this:

Thanks,
Abhijit Halder

[-- Attachment #2: gdb-pipe-command-doc.patch --]
[-- Type: text/x-patch, Size: 2291 bytes --]

diff -rup src/gdb/doc/ChangeLog dst/gdb/doc/ChangeLog
--- src/gdb/doc/ChangeLog	2011-08-27 20:35:38.459934029 +0530
+++ dst/gdb/doc/ChangeLog	2011-08-31 13:33:59.188118016 +0530
@@ -1,3 +1,7 @@
+2011-08-29  Abhijit Halder  <abhijit.k.halder@gmail.com>
+
+	* gdb.texinfo (Shell Commands): Add pipe command.
+
 2011-08-25  Andrew Oakley  <andrew@ado.is-a-geek.net>
 
 	* gdb.texinfo (Types In Python): Document 'bitpos' for enums.
diff -rup src/gdb/doc/gdb.texinfo dst/gdb/doc/gdb.texinfo
--- src/gdb/doc/gdb.texinfo	2011-08-26 21:11:39.938354007 +0530
+++ dst/gdb/doc/gdb.texinfo	2011-08-31 13:43:45.836117888 +0530
@@ -1367,6 +1367,48 @@ Execute the @code{make} program with the
 arguments.  This is equivalent to @samp{shell make @var{make-args}}.
 @end table
 
+If you want to process the output of a @value{GDBN} command using some shell
+command or some script, that can be done by using the command @code{pipe}.
+
+@table @code
+@kindex pipe
+@item pipe @var{dlim} @var{gdbcmd} @var{dlim} @var{shellcmd}
+@var{dlim} is a string of arbitrary length, containing no whitespace and no
+leading @samp{-}, acts as a separator between a @value{GDBN} command
+@var{gdbcmd} and a shell command @var{shellcmd}.  The shell command should be
+in compliance with the syntax of the default shell.
+@end table
+
+@smallexample
+(@value{GDBP}) @b{ptype dd_tbl}
+type = struct dd @{
+    int dd_handle;
+    const char *dd_name;
+    int dd_major;
+    int dd_minor;
+    void *dd_code;
+    void *dd_data;
+@} [1024]
+(@value{GDBP}) @b{pipe <br> print dd_tbl <br> sed 's/@}/\n/g' | grep @
+"\.test_dd" | tr ',' '\n'}
+
+  @{dd_handle = 10
+  dd_name = 0x8048538 ".test_dd"
+  dd_major = 100
+  dd_minor = 0
+  dd_code = 0xcc
+  dd_data = 0x80
+@end smallexample
+
+In the above example @samp{<br>} acts as a delimiter.  The output of
+@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
+grep ".test_dd" | tr ',' '\n'} for processing.
+
+In the given example the output of @value{GDBN} command @samp{print dd_tbl} is
+huge and not well formated.  The use of shell commands like @command{sed},
+@command{tr} and @command{grep} ease the searching of desired pattern and hence
+ease debugging.
+
 @node Logging Output
 @section Logging Output
 @cindex logging @value{GDBN} output

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Wed, 31 Aug 2011 14:13:49 +0530
> From: Abhijit Halder <abhijit.k.halder@gmail.com>
> Cc: pedro@codesourcery.com, gdb-patches@sourceware.org, sergiodj@redhat.com, 
> 	jan.kratochvil@redhat.com, tromey@redhat.com
> 
> >> +@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
> >> +grep ".test_dd" | tr ',' '\n'} for processing.
> >
> > I would suggest to add that the example does not use `|' as a
> > delimiter to allow the shell command to use it in its usual shell
> > semantics.
> >
> In the above example one can use `|' as the delimiter. it will not
> break anything syntactically but will not be readable much, hence I
> avoided using `|' as delimiter. Do we need to mention that here?

We don't _have_ to mention that, but I think it's a good idea, because
the natural choice of the delimiter is the normal pipe character.

> I made the corrections as suggested. Please review this:

OK, thanks.

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Abhijit Halder]

[-- Attachment #1: Type: text/plain, Size: 1116 bytes --]

On Wed, Aug 31, 2011 at 4:30 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> Date: Wed, 31 Aug 2011 14:13:49 +0530
>> From: Abhijit Halder <abhijit.k.halder@gmail.com>
>> Cc: pedro@codesourcery.com, gdb-patches@sourceware.org, sergiodj@redhat.com,
>>       jan.kratochvil@redhat.com, tromey@redhat.com
>>
>> >> +@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
>> >> +grep ".test_dd" | tr ',' '\n'} for processing.
>> >
>> > I would suggest to add that the example does not use `|' as a
>> > delimiter to allow the shell command to use it in its usual shell
>> > semantics.
>> >
>> In the above example one can use `|' as the delimiter. it will not
>> break anything syntactically but will not be readable much, hence I
>> avoided using `|' as delimiter. Do we need to mention that here?
>
> We don't _have_ to mention that, but I think it's a good idea, because
> the natural choice of the delimiter is the normal pipe character.
>
Okay I got it. Made that change, please comment on it.

>> I made the corrections as suggested. Please review this:
>
> OK, thanks.
>

[-- Attachment #2: gdb-pipe-command-doc.patch --]
[-- Type: text/x-patch, Size: 2554 bytes --]

diff -rup src/gdb/doc/ChangeLog dst/gdb/doc/ChangeLog
--- src/gdb/doc/ChangeLog	2011-08-27 20:35:38.459934029 +0530
+++ dst/gdb/doc/ChangeLog	2011-08-31 13:33:59.188118016 +0530
@@ -1,3 +1,7 @@
+2011-08-29  Abhijit Halder  <abhijit.k.halder@gmail.com>
+
+	* gdb.texinfo (Shell Commands): Add pipe command.
+
 2011-08-25  Andrew Oakley  <andrew@ado.is-a-geek.net>
 
 	* gdb.texinfo (Types In Python): Document 'bitpos' for enums.
diff -rup src/gdb/doc/gdb.texinfo dst/gdb/doc/gdb.texinfo
--- src/gdb/doc/gdb.texinfo	2011-08-26 21:11:39.938354007 +0530
+++ dst/gdb/doc/gdb.texinfo	2011-08-31 18:04:11.348118042 +0530
@@ -1367,6 +1367,52 @@ Execute the @code{make} program with the
 arguments.  This is equivalent to @samp{shell make @var{make-args}}.
 @end table
 
+If you want to process the output of a @value{GDBN} command using some shell
+command or some script, that can be done by using the command @code{pipe}.
+
+@table @code
+@kindex pipe
+@item pipe @var{dlim} @var{gdbcmd} @var{dlim} @var{shellcmd}
+@var{dlim} is a string of arbitrary length, containing no whitespace and no
+leading @samp{-}, acts as a separator between a @value{GDBN} command
+@var{gdbcmd} and a shell command @var{shellcmd}.  The shell command should be
+in compliance with the syntax of the default shell.  The @var{dlim} pattern
+should not appear in @var{gdbcomd} otherwise the parsing will fail; although it
+may appear in @var{shellcmd}. 
+@end table
+
+@smallexample
+(@value{GDBP}) @b{ptype dd_tbl}
+type = struct dd @{
+    int dd_handle;
+    const char *dd_name;
+    int dd_major;
+    int dd_minor;
+    void *dd_code;
+    void *dd_data;
+@} [1024]
+(@value{GDBP}) @b{pipe <br> print dd_tbl <br> sed 's/@}/\n/g' | grep @
+"\.test_dd" | tr ',' '\n'}
+
+  @{dd_handle = 10
+  dd_name = 0x8048538 ".test_dd"
+  dd_major = 100
+  dd_minor = 0
+  dd_code = 0xcc
+  dd_data = 0x80
+@end smallexample
+
+In the above example @samp{<br>} acts as a delimiter.  The output of
+@samp{print dd_tbl} is passed to the shell command @samp{sed 's/@}/\n/g' | @
+grep ".test_dd" | tr ',' '\n'} for processing.  @samp{|} could be one natural
+choise of using as the delimiter.  In the above we avoided the same for
+readability purpose.
+
+In the given example the output of @value{GDBN} command @samp{print dd_tbl} is
+huge and not well formated.  The use of shell commands like @command{sed},
+@command{tr} and @command{grep} ease the searching of desired pattern and hence
+ease debugging.
+
 @node Logging Output
 @section Logging Output
 @cindex logging @value{GDBN} output

Re: [PATCH][TEST-CASE][DOC] Implementation of pipe to pass GDB's command output to the shell.

[Eli Zaretskii]

> Date: Wed, 31 Aug 2011 18:12:31 +0530
> From: Abhijit Halder <abhijit.k.halder@gmail.com>
> Cc: pedro@codesourcery.com, gdb-patches@sourceware.org, sergiodj@redhat.com, 
> 	jan.kratochvil@redhat.com, tromey@redhat.com
> 
> Okay I got it. Made that change, please comment on it.

This version is fine with me, thanks.