[FFmpeg-devel] [PATCH 7/7] Add check in av_write_header() which validates the rawvideo codec tag.

Michael Niedermayer michaelni
Sun Jun 6 13:01:31 CEST 2010


On Sun, Jun 06, 2010 at 10:52:35AM +0200, Stefano Sabatini wrote:
> On date Sunday 2010-06-06 05:03:57 +0200, Michael Niedermayer encoded:
> > On Sun, Jun 06, 2010 at 02:02:21AM +0200, Stefano Sabatini wrote:
> > > On date Sunday 2010-06-06 01:40:55 +0200, Michael Niedermayer encoded:
> > > > On Sat, Jun 05, 2010 at 09:46:07PM +0200, Stefano Sabatini wrote:
> > > > [...]
> > > > >  const AVCodecTag ff_nut_video_tags[] = {
> > > > > +    { CODEC_ID_RAWVIDEO, 0                         },
> > > > 
> > > > rejected
> > > > 
> > > > i keep saying the same thing and you keep doing the exact opposit.
> > > > i fixed it and ill add some reg test, 
> > > 
> > > It didn't fix neither NUT nor AVI. Please try this:
> > 
> > avi works, we have a regression test for it now
> > nut was what you worked on, if it doesnt work dont blame me.
> > and rgb444 was what you just added, so if that is the only format
> > not working in avi ...
> > 
> > avi worked in the past, and one of the recent patches broke it, and
> > exactly the way i repeatly told you you must be carefull with, namely codec_tag
> > must be 0 in avi for rgb you totally ignored me and ffmpeg generated
> > invalid files as a result. Now this patch might make invalid nut files
> > and you dont even attempt to explain why you would want add such a
> > incorrect codec_tag to nut.
> 
> I already posted the failing command, problem is that NUT/AVI muxers
> are creating files which cannot be played (rather than just fail). It

i already hinted that this is because you messed up.
i fixed rgb444 in avi now, you forgot to add it to a table


> is not a regression as I said multiple times, is a problem I noticed

iam pretty sure codec_tag was correctly 0 in the past for avi, so
this was a regression


[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

I hate to see young programmers poisoned by the kind of thinking
Ulrich Drepper puts forward since it is simply too narrow -- Roman Shaposhnik
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20100606/2d637d16/attachment.pgp>



More information about the ffmpeg-devel mailing list