[Ffmpeg-devel] Matroska Patch

Steve Lhomme steve.lhomme
Wed Mar 22 14:16:01 CET 2006

Michael Niedermayer wrote:
> Hi
> On Wed, Mar 22, 2006 at 11:27:16AM +0100, Steve Lhomme wrote:
>> Michael Niedermayer wrote:
>>> what do you do if you are allergic on lets say eggs, and someone knowing 
>>> that
>>> always adds an egg to every piece of food they give you?
>>> hint: we are allergic on compiler bug workarounds
>> And I'm a pragmatic person. I fix things I can fix. I am not going to 
>> fix MSVC, gcc or MinGW. I have better things to do.
> you workaround bugs you dont fix them

Regarding the compiler issues, I turn an error into something that 
compiles (fine). I call that a fix.

>> Now if you don't accept fixes, that's your responsibility not mine.
> good changes get accepted buggy one are rejected, either fix you code
> or fork, broken code will not be accpeted you can call it whatever
> you like, call it critical and clean fixes if you like its still not
> going to be accepted (noone yet succeded in getting knowningly broken
> code accepted)

I suppose if I submited a bug with that background information, it would 
have been totally ridiculous:
"cleanup codec_id mapping (untested)"

Anyway, I'm glad most of the patch made it to CVS. Matroska users will 
thank you.

>> Most of the fixes found in this patch are fixes for *all* existing 
>> Matroska files. I don't think I have to send any. Just open one in 
>> FFMPEG and you'll see. I described all the fixes included in this patch. 
>> So you have #1, #2, #3, #4.
> well,  you dont have to do anything, just if you want a bug fixed or
> patch accepted then you will have to follow the same rules as everyone else
> now if you had a constructive comment about the rules or suggestions
> for their improvement that would be fine too, but you just waste our time

I have nothing against the rules. Just the way code is turned down. Like 
the MPEG2 fix I sent that you know better than me how to fix it, but you 
didn't. I wish you would apply the same strict rules to you than anyone 
else. Committing untested code and turning down tested fixes for 
cosmetic reason is more than questionable.


More information about the ffmpeg-devel mailing list