This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [RFA]: Fix objc/1236 - printing non-debuggable symbols
- From: Adam Fedor <fedor at doc dot com>
- To: Elena Zannoni <ezannoni at redhat dot com>
- Cc: GDB Patches <gdb-patches at sources dot redhat dot com>
- Date: 21 Mar 2004 21:30:09 -0700
- Subject: Re: [RFA]: Fix objc/1236 - printing non-debuggable symbols
- Organization:
- References: <3EE150E9.2080003@doc.com> <16433.23008.848603.345992@localhost.redhat.com> <1077638167.19364.8.camel@localhost.localdomain>
On Tue, 2004-02-24 at 08:56, Adam Fedor wrote:
> On Mon, 2004-02-16 at 17:01, Elena Zannoni wrote:
> > Adam Fedor writes:
> > > Fixes PR objc/1236
> > >
> > > 2003-06-06 Adam Fedor <fedor@gnu.org>
> > >
> > > Fix objc/1236.
> > > * gdb/linespec.c (decode_line_2): Print non-debuggable (minimal)
> > > symbols.
> > > (decode_objc): Update for change in arguments to decode_line_2
> > > (find_methods): Likewise.
> > >
> >
> >
> > sorry for the delay, this got lost in my pile. I see what you are
> > doing, but could you have 2 separate arrays of symbol pointers and of
> > msymbol pointers instead of concatenating the two? This way you would
> > also simplify the find_imps code (no need to swap the symbols and the
> > msymbols) and also the call to decode_line_2 in the c++ code would
> > look less weird, it could just pass NULL's.
> >
>
> Well I looked into doing this, but splitting up the arrays looks pretty
> ugly and more error prone than the other approach. Plus the arrays have
> to be combined eventually in decode_line_2 anyway.
Ping.