[rfa] add DMGL_PARAMS (was: new demangler)

[rfa] add DMGL_PARAMS (was: new demangler)

[David Carlton]

On Tue, 16 Dec 2003 00:30:22 -0500, Daniel Jacobowitz <drow@mvista.com> said:
> On Mon, Dec 15, 2003 at 03:38:28PM -0800, David Carlton wrote:

>> Okay; I'll just stick in the DMGL_PARAMS for now, then - my tests
>> start passing again with that change.  (Or I could rewrite the code
>> that parses the output to not look for parentheses, I suppose.)

> Do the former for now.

The latter actually should be easy enough given the parsing
functionality that's already in cp-support.c, but I have no objection
to the former, which I present here.  Is it okay for mainline?  Tested
on i686-pc-linux-gnu, GCC 3.2, DWARF 2; no regressions.  (No tests,
but it's necessary for my current patch awaiting approval to work
properly.)

David Carlton
carlton@kealia.com

2003-12-16  David Carlton  <carlton@kealia.com>

	* cp-support.c (class_name_from_physname): Add DMGL_PARAMS.
	(method_name_from_physname): Ditto.

Index: cp-support.c
===================================================================
RCS file: /cvs/src/src/gdb/cp-support.c,v
retrieving revision 1.10
diff -u -p -r1.10 cp-support.c
--- cp-support.c	5 Dec 2003 04:25:09 -0000	1.10
+++ cp-support.c	16 Dec 2003 16:52:10 -0000
@@ -145,7 +145,7 @@ class_name_from_physname (const char *ph
   char *ret = NULL;
   const char *end;
   int depth = 0;
-  char *demangled_name = cplus_demangle (physname, DMGL_ANSI);
+  char *demangled_name = cplus_demangle (physname, DMGL_ANSI | DMGL_PARAMS);
 
   if (demangled_name == NULL)
     return NULL;
@@ -170,7 +170,7 @@ method_name_from_physname (const char *p
   char *ret = NULL;
   const char *end;
   int depth = 0;
-  char *demangled_name = cplus_demangle (physname, DMGL_ANSI);
+  char *demangled_name = cplus_demangle (physname, DMGL_ANSI | DMGL_PARAMS);
 
   if (demangled_name == NULL)
     return NULL;

Re: [rfa] add DMGL_PARAMS (was: new demangler)

[Daniel Jacobowitz]

On Tue, Dec 16, 2003 at 08:54:57AM -0800, David Carlton wrote:
> On Tue, 16 Dec 2003 00:30:22 -0500, Daniel Jacobowitz <drow@mvista.com> said:
> > On Mon, Dec 15, 2003 at 03:38:28PM -0800, David Carlton wrote:
> 
> >> Okay; I'll just stick in the DMGL_PARAMS for now, then - my tests
> >> start passing again with that change.  (Or I could rewrite the code
> >> that parses the output to not look for parentheses, I suppose.)
> 
> > Do the former for now.
> 
> The latter actually should be easy enough given the parsing
> functionality that's already in cp-support.c, but I have no objection
> to the former, which I present here.  Is it okay for mainline?  Tested
> on i686-pc-linux-gnu, GCC 3.2, DWARF 2; no regressions.  (No tests,
> but it's necessary for my current patch awaiting approval to work
> properly.)
> 
> David Carlton
> carlton@kealia.com
> 
> 2003-12-16  David Carlton  <carlton@kealia.com>
> 
> 	* cp-support.c (class_name_from_physname): Add DMGL_PARAMS.
> 	(method_name_from_physname): Ditto.

Yes, this is fine.  Thanks!

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

Re: [rfa] add DMGL_PARAMS

[David Carlton]

On Tue, 16 Dec 2003 14:23:07 -0500, Daniel Jacobowitz <drow@mvista.com> said:

>> 2003-12-16  David Carlton  <carlton@kealia.com>
>> 
>> * cp-support.c (class_name_from_physname): Add DMGL_PARAMS.
>> (method_name_from_physname): Ditto.

> Yes, this is fine.  Thanks!

Thanks, committed.

David Carlton
carlton@kealia.com

Re: [rfa] add DMGL_PARAMS (was: new demangler)

[Michael Elizabeth Chastain]

Looks okay to me.

Be aware that dwarf 2 and stabs+ have had differences in this area.  For
example, when is_ctor_or_dtor in the demangler was broken, dwarf 2 was
not affected, but stabs+ was.  So there's always a chance that I'll come
back in a few days or a week and say "you broke stabs+"!

Michael C

===

2003-12-16  David Carlton  <carlton@kealia.com>

	* cp-support.c (class_name_from_physname): Add DMGL_PARAMS.
	(method_name_from_physname): Ditto.