[FFmpeg-devel] [PATCH] Some minor fixes of my "patchset" for palettized QuickTime video in Matroska
matsp888 at yahoo.com
Mon Dec 21 17:16:36 CET 2015
On 12/20/2015 10:43 AM, Mats Peterson wrote:
> On 12/17/2015 04:35 PM, Mats Peterson wrote:
>> I know nobody will care whatsoever, but I forgot to change the minimum
>> private data size from 86 to 36 for A_QUICKTIME, and I removed that
>> fprintf() line in qtpalette.c. I'll post it for the hell of it. PERHAPS
>> it can be valuable as a starting point in order to fix this
>> long-standing issue, IN SPITE of my refusal to learn Git. Oh, and it's
>> in zip format this time, since I got complaints from using tar.
>> ffmpeg-devel mailing list
>> ffmpeg-devel at ffmpeg.org
> Yet another update with small corrections of matroskadec.c and mov.c. It
> seems that for 8-bit grayscale Cinepak, the policy has been to "trick"
> the Cinepak decoder into believing the video is rgb24 and not pal8, and
> it has been working fine until I started messing with setting
> bits_per_coded_sample to 8 for 8-bit grayscale Cinepak. I've restored
> the old behaviour of using rgb24.
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
As if that wasn't enough, here is another update. This time it's passing
the FATE tests, unlike the old one. In order to refresh your memories,
this patch fixes the long-standing issue regarding not recognizing
palettized QuickTime video in Matroska. I have created a new function
get_qtpalette() in a new file qtpalette.c, that is shared by both
matroskadec.c and mov.c. And just like before, I'm not really in the
mood to learn Git just to contribute with these simple patches, but I
would be happy if someone could take a look at it, and, if it looks
decent enough, submit proper patches in the right format for Git.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 5259 bytes
Desc: not available
More information about the ffmpeg-devel