[Ffmpeg-devel] Matroska Patch

Steve Lhomme steve.lhomme
Wed Mar 22 09:29:01 CET 2006

M?ns Rullg?rd wrote:
> Steve Lhomme <steve.lhomme at free.fr> writes:
>> Michael Niedermayer wrote:
>>>> +            if (track->default_duration)
>>>> +                av_reduce(&st->codec->time_base.num, &st->codec->time_base.den, track->default_duration, 1000000000, 30000);
>>> not acceptable
>> Because ? The idea here is to be able to get 30000/1001 for matching
>> default_duration. The rest of the time the more basic time_base (25/1)
>> is found. That's at least the theory. In practice it doesn't work
>> well...
> If it doesn't work, why are you sending a patch?

I didn't say it doesn't work. It does work most of the time. Which is a 
huge leap forward compared to the current Matroska parser in libavformat.

>> For some reasons MSVC doesn't like a variable called time. It must
>> be defined as a macro somewhere. And changing a name can't be such a
>> bad change.
> Then figure out why.  We don't support msvc, and will not accept
> arbitrary patches to make it work without an explanation.

Maybe it's a bug in MinGW (again). That code is old, I submited a patch 
a long time ago and nothing got accepted.

It's really disappointed to give you code (up to the point you just have 
to click on a button to commit it) and being returned lame comments and 
flame. I'm sorry I'm not using your ace compiler on your ace OS. And 
still, I make softwares and I fix your bugs. If you think you can do 
better, do it. But when someone fixes a bug (or more) you shouldn't spit 
on him like you do...

As far as I know, my matroska code works better than yours. If you don't 
even want me to share it with you I won't. I'll keep on improving 
DrFFMPEG and I will post my patches here. Not because you might do me 
the honour of commiting the code. But because there are other 
developpers more open minded that will use the patch out of your 
official tree. (FFDShow and VLC are good examples of popular softwares 
having successful forks on FFMPEG)


More information about the ffmpeg-devel mailing list