[PATCH 0/3] varobj zero-padded hexadecimal format

[PATCH 0/3] varobj zero-padded hexadecimal format

[Luis Machado]

This set of patches add support for the zero-padded hexadecimal format for
varobj's, defined as "zhexadecimal".  We currently only support regular
non-zero-padded hexadecimal.

Talking with IDE developers, they would like to have this option that is
already available to GDB's print/x commands in the CLI as 'z'.

Luis Machado (3):
  Add zero-padded hexadecimal format support for varobj's
  Add varobj zhexadecimal format tests.
  Documentation changes for the new zhexadecimal format

 gdb/NEWS                                |  5 +++++
 gdb/doc/gdb.texinfo                     |  2 +-
 gdb/mi/mi-cmd-var.c                     |  4 +++-
 gdb/testsuite/gdb.mi/mi-var-display.exp | 25 +++++++++++++++++++++++--
 gdb/varobj.c                            |  5 +++--
 gdb/varobj.h                            |  3 ++-
 6 files changed, 37 insertions(+), 7 deletions(-)

-- 
1.9.1

[PATCH 1/3] Add zero-padded hexadecimal format support for varobj's

[Luis Machado]

This patch changes things so that we can support zero-padded hexadecimal
formats for varobj's. The new type is "zhexadecimal" and maps to GDB's 'z'
print format.

Ideally we would share all the formats we currently support with the MI/varobj
layer, but this requires a bit more work. Meanwhile, adding a new mapping seems
to work.

gdb/ChangeLog:

2015-05-01  Luis Machado  <lgustavo@codesourcery.com>

	* mi/mi-cmd-var.c (mi_parse_format): Handle new "zhexadecimal"
	format.
	* varobj.c (varobj_format_string): Add "zhexadecimal" entry.
	(format_code): Add 'z' entry.
	(varobj_set_display_format): Handle FORMAT_ZHEXADECIMAL.
	* varobj.h (varobj_display_formats) <FORMAT_ZHEXADECIMAL>: New enum
	field.
---
 gdb/mi/mi-cmd-var.c | 4 +++-
 gdb/varobj.c        | 5 +++--
 gdb/varobj.h        | 3 ++-
 3 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/gdb/mi/mi-cmd-var.c b/gdb/mi/mi-cmd-var.c
index ee0bbc6..5554e3c 100644
--- a/gdb/mi/mi-cmd-var.c
+++ b/gdb/mi/mi-cmd-var.c
@@ -233,10 +233,12 @@ mi_parse_format (const char *arg)
 	return FORMAT_HEXADECIMAL;
       else if (strncmp (arg, "octal", len) == 0)
 	return FORMAT_OCTAL;
+      else if (strncmp (arg, "zhexadecimal", len) == 0)
+	return FORMAT_ZHEXADECIMAL;
     }
 
   error (_("Must specify the format as: \"natural\", "
-	   "\"binary\", \"decimal\", \"hexadecimal\", or \"octal\""));
+	   "\"binary\", \"decimal\", \"hexadecimal\", \"octal\" or \"zhexadecimal\""));
 }
 
 void
diff --git a/gdb/varobj.c b/gdb/varobj.c
index b220fd8..2f2939e 100644
--- a/gdb/varobj.c
+++ b/gdb/varobj.c
@@ -50,7 +50,7 @@ show_varobjdebug (struct ui_file *file, int from_tty,
 
 /* String representations of gdb's format codes.  */
 char *varobj_format_string[] =
-  { "natural", "binary", "decimal", "hexadecimal", "octal" };
+  { "natural", "binary", "decimal", "hexadecimal", "octal", "zhexadecimal" };
 
 /* True if we want to allow Python-based pretty-printing.  */
 static int pretty_printing = 0;
@@ -214,7 +214,7 @@ static struct varobj *varobj_add_child (struct varobj *var,
 /* Private data */
 
 /* Mappings of varobj_display_formats enums to gdb's format codes.  */
-static int format_code[] = { 0, 't', 'd', 'x', 'o' };
+static int format_code[] = { 0, 't', 'd', 'x', 'o', 'z' };
 
 /* Header of the list of root variable objects.  */
 static struct varobj_root *rootlist;
@@ -583,6 +583,7 @@ varobj_set_display_format (struct varobj *var,
     case FORMAT_DECIMAL:
     case FORMAT_HEXADECIMAL:
     case FORMAT_OCTAL:
+    case FORMAT_ZHEXADECIMAL:
       var->format = format;
       break;
 
diff --git a/gdb/varobj.h b/gdb/varobj.h
index 8860526..179b2a5 100644
--- a/gdb/varobj.h
+++ b/gdb/varobj.h
@@ -28,7 +28,8 @@ enum varobj_display_formats
     FORMAT_BINARY,		/* Binary display                    */
     FORMAT_DECIMAL,		/* Decimal display                   */
     FORMAT_HEXADECIMAL,		/* Hex display                       */
-    FORMAT_OCTAL		/* Octal display                     */
+    FORMAT_OCTAL,		/* Octal display                     */
+    FORMAT_ZHEXADECIMAL		/* Zero padded hexadecimal	     */
   };
 
 enum varobj_type
-- 
1.9.1

[PATCH 2/3] Add varobj zhexadecimal format tests.

[Luis Machado]

This patch adds a few tests to exercise the new zhexadecimal format for
varobj's. I see 5 FAIL's without the zhexadecimal format support and full
passes with a zhexadecimal format support.

I renamed some of the tests to make them unique.

gdb/testsuite/ChangeLog:

2015-05-01  Luis Machado  <lgustavo@codesourcery.com>

	* gdb.mi/mi-var-display.exp: Add new checks for the zhexadecimal format
	and change test names to make them unique.
---
 gdb/testsuite/gdb.mi/mi-var-display.exp | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/gdb/testsuite/gdb.mi/mi-var-display.exp b/gdb/testsuite/gdb.mi/mi-var-display.exp
index 31bdce4..6aedb53 100644
--- a/gdb/testsuite/gdb.mi/mi-var-display.exp
+++ b/gdb/testsuite/gdb.mi/mi-var-display.exp
@@ -81,6 +81,11 @@ mi_gdb_test "-var-evaluate-expression bar" \
 	"eval variable bar"
 
 # Test: c_variable-6.5
+# Desc: change format of bar to zero-padded hexadecimal
+mi_gdb_test "-var-set-format bar zhexadecimal" \
+	"\\^done,format=\"zhexadecimal\",value=\"0x0.*849\"" \
+	"set format variable bar in zero-padded hexadecimal"
+
 # Desc: change format of bar to hex
 mi_gdb_test "-var-set-format bar hexadecimal" \
 	"\\^done,format=\"hexadecimal\",value=\"0x849\"" \
@@ -241,7 +246,7 @@ mi_gdb_test "-var-set-format weird.func_ptr hexadecimal" \
 
 mi_gdb_test "-var-show-format weird.func_ptr" \
 	"\\^done,format=\"hexadecimal\"" \
-	"show format variable weird.func_ptr"
+	"show format variable weird.func_ptr (hex)"
 
 mi_gdb_test "-var-set-format weird.func_ptr_ptr hexadecimal" \
 	"\\^done,format=\"hexadecimal\",value=\"$hex\"" \
@@ -249,7 +254,23 @@ mi_gdb_test "-var-set-format weird.func_ptr_ptr hexadecimal" \
 
 mi_gdb_test "-var-show-format weird.func_ptr_ptr" \
 	"\\^done,format=\"hexadecimal\"" \
-	"show format variable weird.func_ptr_ptr"
+	"show format variable weird.func_ptr_ptr (hex)"
+
+mi_gdb_test "-var-set-format weird.func_ptr zhexadecimal" \
+	"\\^done,format=\"zhexadecimal\",value=\"$hex\"" \
+	"set format variable weird.func_ptr in zero-padded hex"
+
+mi_gdb_test "-var-show-format weird.func_ptr" \
+	"\\^done,format=\"zhexadecimal\"" \
+	"show format variable weird.func_ptr (zhex)"
+
+mi_gdb_test "-var-set-format weird.func_ptr_ptr zhexadecimal" \
+	"\\^done,format=\"zhexadecimal\",value=\"$hex\"" \
+	"set format variable weird.func_ptr_ptr in zero-padded hex"
+
+mi_gdb_test "-var-show-format weird.func_ptr_ptr" \
+	"\\^done,format=\"zhexadecimal\"" \
+	"show format variable weird.func_ptr_ptr (zhex)"
 
 # Test: c_variable-6.24
 # Desc: format of weird and children
-- 
1.9.1

[PATCH 3/3] Documentation changes for the new zhexadecimal format

[Luis Machado]

This patch updates the documentation to reflect the addition of the new
zhexadecimal format option for -var-set-format and also adds a NEWS entry
for the change.

gdb/ChangeLog:

2015-05-01  Luis Machado  <lgustavo@codesourcery.com>

	* NEWS: Add new note to MI changes.

gdb/doc/ChangeLog:

2015-05-01  Luis Machado  <lgustavo@codesourcery.com>

	* gdb.texinfo (GDB/MI Variable Objects): Update text for
	-var-set-format.
---
 gdb/NEWS            | 5 +++++
 gdb/doc/gdb.texinfo | 2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/gdb/NEWS b/gdb/NEWS
index 651401d..9d4fd03 100644
--- a/gdb/NEWS
+++ b/gdb/NEWS
@@ -142,6 +142,11 @@ qXfer:exec-file:read
 HP/PA running HP-UX           hppa*-*-hpux*
 Itanium running HP-UX         ia64-*-hpux*
 
+* MI changes
+
+  ** The -var-set-format command now accepts the zhexadecimal format.  It
+     outputs data in hexadecimal format with zero-padding on the left.
+
 *** Changes in GDB 7.9
 
 * GDB now supports hardware watchpoints on x86 GNU Hurd.
diff --git a/gdb/doc/gdb.texinfo b/gdb/doc/gdb.texinfo
index 9e2787d..89fd2ca 100644
--- a/gdb/doc/gdb.texinfo
+++ b/gdb/doc/gdb.texinfo
@@ -28927,7 +28927,7 @@ The syntax for the @var{format-spec} is as follows:
 
 @smallexample
  @var{format-spec} @expansion{}
- @{binary | decimal | hexadecimal | octal | natural@}
+ @{binary | decimal | hexadecimal | octal | natural | zhexadecimal@}
 @end smallexample
 
 The natural format is the default format choosen automatically
-- 
1.9.1

Re: [PATCH 3/3] Documentation changes for the new zhexadecimal format

[Eli Zaretskii]

> From: Luis Machado <lgustavo@codesourcery.com>
> Date: Fri, 1 May 2015 11:32:38 -0300
> 
> This patch updates the documentation to reflect the addition of the new
> zhexadecimal format option for -var-set-format and also adds a NEWS entry
> for the change.

OK, but should we perhaps explain what zhexadecimal means?  I didn't
know about it until I read this patch.

Thanks.

Re: [PATCH 3/3] Documentation changes for the new zhexadecimal format

[Luis Machado]

On 05/01/2015 12:07 PM, Eli Zaretskii wrote:
>> From: Luis Machado <lgustavo@codesourcery.com>
>> Date: Fri, 1 May 2015 11:32:38 -0300
>>
>> This patch updates the documentation to reflect the addition of the new
>> zhexadecimal format option for -var-set-format and also adds a NEWS entry
>> for the change.
>
> OK, but should we perhaps explain what zhexadecimal means?  I didn't
> know about it until I read this patch.

Yes. Or maybe have a better (longer?) nomeclature for this option, 
making it more intuitive?

Re: [PATCH 3/3] Documentation changes for the new zhexadecimal format

[Luis Machado]

On 05/01/2015 12:07 PM, Eli Zaretskii wrote:
>> From: Luis Machado <lgustavo@codesourcery.com>
>> Date: Fri, 1 May 2015 11:32:38 -0300
>>
>> This patch updates the documentation to reflect the addition of the new
>> zhexadecimal format option for -var-set-format and also adds a NEWS entry
>> for the change.
>
> OK, but should we perhaps explain what zhexadecimal means?  I didn't
> know about it until I read this patch.

I'm not too sure where that explanation should go. The other formats are 
self-explanatory. What about 'zero-padded-hexadecimal'? Too long?

The 'z' is an attempt to map back to the CLI format specifier /z when 
using the print and x commands. Though that requires some GDB-fu to get 
the first time.

Re: [PATCH 3/3] Documentation changes for the new zhexadecimal format

[Luis Machado]

On 05/06/2015 10:56 AM, Luis Machado wrote:
> On 05/01/2015 12:07 PM, Eli Zaretskii wrote:
>>> From: Luis Machado <lgustavo@codesourcery.com>
>>> Date: Fri, 1 May 2015 11:32:38 -0300
>>>
>>> This patch updates the documentation to reflect the addition of the new
>>> zhexadecimal format option for -var-set-format and also adds a NEWS
>>> entry
>>> for the change.
>>
>> OK, but should we perhaps explain what zhexadecimal means?  I didn't
>> know about it until I read this patch.
>
> I'm not too sure where that explanation should go. The other formats are
> self-explanatory. What about 'zero-padded-hexadecimal'? Too long?
>
> The 'z' is an attempt to map back to the CLI format specifier /z when
> using the print and x commands. Though that requires some GDB-fu to get
> the first time.
>
>

Also, i've been informed by IDE guys that these strings are usually 
translated, so 'zero-padded-hexadecimal' would be correct for 
internationalization purposes so it can be translated to other 
languages, as opposed to 'zhexadecimal'.

Re: [PATCH 3/3] Documentation changes for the new zhexadecimal format

[Eli Zaretskii]

> Date: Wed, 6 May 2015 10:56:32 -0300
> From: Luis Machado <lgustavo@codesourcery.com>
> CC: <gdb-patches@sourceware.org>
> 
> On 05/01/2015 12:07 PM, Eli Zaretskii wrote:
> >> From: Luis Machado <lgustavo@codesourcery.com>
> >> Date: Fri, 1 May 2015 11:32:38 -0300
> >>
> >> This patch updates the documentation to reflect the addition of the new
> >> zhexadecimal format option for -var-set-format and also adds a NEWS entry
> >> for the change.
> >
> > OK, but should we perhaps explain what zhexadecimal means?  I didn't
> > know about it until I read this patch.
> 
> I'm not too sure where that explanation should go.

It should be near the first use of this term.

> What about 'zero-padded-hexadecimal'? Too long?

Yes, and I don't think we should go to such extremes.  Just explaining
the term the first time it is used in the manual should be OK.