New ARI warning Wed May 26 01:54:51 UTC 2010

New ARI warning Wed May 26 01:54:51 UTC 2010

[GDB Administrator]

1084a1085
> gdb/python/py-cmd.c:192: gettext: _ markup: All messages should be marked up with _.
gdb/python/py-cmd.c:192:	error ("%s", msg);

Re: New ARI warning Wed May 26 01:54:51 UTC 2010

[Doug Evans]

On Tue, May 25, 2010 at 6:54 PM, GDB Administrator
<gdbadmin@sourceware.org> wrote:
> 1084a1085
>> gdb/python/py-cmd.c:192: gettext: _ markup: All messages should be marked up with _.
> gdb/python/py-cmd.c:192:        error ("%s", msg);
>

There were already a couple of uses of error ("%s", ...) so I figured
it was safe.

OOC, is this why source.c has:

error (("%s"), msg);

?

RE: New ARI warning Wed May 26 01:54:51 UTC 2010

[Pierre Muller]

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Doug Evans
> Envoyé : Wednesday, May 26, 2010 4:48 AM
> À : gdb-patches@sourceware.org
> Objet : Re: New ARI warning Wed May 26 01:54:51 UTC 2010
> 
> On Tue, May 25, 2010 at 6:54 PM, GDB Administrator
> <gdbadmin@sourceware.org> wrote:
> > 1084a1085
> >> gdb/python/py-cmd.c:192: gettext: _ markup: All messages should be
> marked up with _.
> > gdb/python/py-cmd.c:192:        error ("%s", msg);
> >
> 
> There were already a couple of uses of error ("%s", ...) so I figured
> it was safe.

  Of course, translation of "%s" into another language does make sense!
 
> OOC, is this why source.c has:
> 
> error (("%s"), msg);

  I suspect that the braces around "%s" are exactly here for that reason:
they avoid getting this ARI report.
  Does anyone have another idea of the utility of such a wrapping?

  I could probably try to improve the ARI rule 
so that it does not complain if the string is just "%s",
but I would like to be sure that I am not missing some information here.

Pierre Muller
as ARI maintainer.

Re: New ARI warning Wed May 26 01:54:51 UTC 2010

[Joel Brobecker]

> > > gdb/python/py-cmd.c:192:        error ("%s", msg);
> > >
> > 
> > There were already a couple of uses of error ("%s", ...) so I figured
> > it was safe.
> 
>   Of course, translation of "%s" into another language does make sense!
>  
> > OOC, is this why source.c has:
> > 
> > error (("%s"), msg);
> 
>   I suspect that the braces around "%s" are exactly here for that reason:
> they avoid getting this ARI report.
>   Does anyone have another idea of the utility of such a wrapping?
> 
>   I could probably try to improve the ARI rule 
> so that it does not complain if the string is just "%s",
> but I would like to be sure that I am not missing some information here.

I don't think anyone really answered the above. I don't see any issue
either way, and I don't have a preference. ISTM that the simplest would
probably be to use ("%s"), instead of adding an exception for "%s".

-- 
Joel