[FFmpeg-cvslog] r25531 - trunk/libavformat/img2.c

Reimar Döffinger Reimar.Doeffinger
Thu Oct 21 22:40:19 CEST 2010


On Thu, Oct 21, 2010 at 12:45:36PM -0700, Pascal Massimino wrote:
> Michael,
> 
> On Thu, Oct 21, 2010 at 12:14 PM, Michael Niedermayer <michaelni at gmx.at>wrote:
> 
> > On Thu, Oct 21, 2010 at 01:01:56PM -0200, Ramiro Polla wrote:
> > > On Wed, Oct 20, 2010 at 11:59 PM, Pascal Massimino
> > > <pascal.massimino at gmail.com> wrote:
> > > > Ramiro,
> > > > On Wed, Oct 20, 2010 at 6:44 PM, Ramiro Polla <ramiro.polla at gmail.com
> > >wrote:
> > > >> On Wed, Oct 20, 2010 at 10:57 PM, skal <subversion at mplayerhq.hu>
> > wrote:
> > > >> [...]
> > > >> > @@ -413,56 +413,51 @@ static int img_write_packet(AVFormatCont
> > > >> >  /* input */
> > > >> >  #if CONFIG_IMAGE2_DEMUXER
> > > >> >  AVInputFormat image2_demuxer = {
> > > >> > -    "image2",
> > > >> > -    NULL_IF_CONFIG_SMALL("image2 sequence"),
> > > >> > -    sizeof(VideoData),
> > > >> > -    image_probe,
> > > >> > -    img_read_header,
> > > >> > -    img_read_packet,
> > > >> > -    NULL,
> > > >> > -    NULL,
> > > >> > -    NULL,
> > > >> > -    AVFMT_NOFILE,
> > > >> > +    .name = "image2",
> > > >> > +    .long_name = NULL_IF_CONFIG_SMALL("image2 sequence"),
> > > >> > +    .priv_data_size = sizeof(VideoData),
> > > >> > +    .read_probe = read_probe,
> > > >> > +    .read_header = read_header,
> > > >> > +    .read_packet = read_packet,
> > > >> > +    .flags = AVFMT_NOFILE,
> > > >> >  };
> > > >>
> > > >> Sorry for not commenting earlier, but it's quite handy while debugging
> > > >> to have a more descriptive name (img2_read_header for example).
> > > >> Otherwise you'll get only "read_header" which could be from many other
> > > >> demuxers.
> > > >>
> > > >
> > > > as in $attached?
> > >
> > > Yes, like that. Michael is the maintainer though, so you'd have to
> > > wait for his call...
> >
> > I prefer not to prefix all functions by the file name
> > id rather see the debuggers user interface improved ;)
> >
> 
> +1
> 
> something like:
> 
>    (gdb) b read_header:img2.c

Uh, img2.c:read_header works just fine I think.
At least, line numbers work.
It just does not always work so well in backtraces.



More information about the ffmpeg-cvslog mailing list