[FFmpeg-devel] [PATCH] Fix Ogg data_offset computation.

[Aaron Colwell]

Ok. Hopefully this is the last round. I've attached a new
patch. It is essentially the code that Reimar gave me,
except an additional check was added to only set the
header field to 0 on streams that have the value set to > 0.

The new check preserves existing behavior for Skeleton
streams that have header set to -1 when this code executes.
The -1 value appears to be some sort of signal to the code in
ogg_read_header() that NULLs out the codec pointer for the
Skeleton stream.

I've verified that this change works in the original ogg file I've
been investigating as well as a few other ones ("just to be sure").

Aaron

On Mon, Nov 22, 2010 at 12:42 PM, Reimar D?ffinger <Reimar.Doeffinger at gmx.de
> wrote:

> On Mon, Nov 22, 2010 at 12:26:41PM -0800, Aaron Colwell wrote:
> > I see what your saying. I think you might be right. I'll test this and
> > see if it works out for me. Thanks for the help. I appreciate it.
>
> You're welcome, even though I already now feel it impacting my sanity...
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at mplayerhq.hu
> https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ogg_data_offset.patch
Type: text/x-patch
Size: 899 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20101122/05187f86/attachment.bin>