[FFmpeg-cvslog] r14631 - in trunk/tests: ffmpeg.regression.ref regression.sh rotozoom.regression.ref

[Måns Rullgård]

Baptiste Coudurier <baptiste.coudurier at smartjog.com> writes:

> M?ns Rullg?rd wrote:
>> Baptiste Coudurier <baptiste.coudurier at smartjog.com> writes:
>> 
>>> M?ns Rullg?rd wrote:
>>>> Baptiste Coudurier <baptiste.coudurier at smartjog.com> writes:
>>>>
>>>>> M?ns Rullg?rd wrote:
>>>>>> bcoudurier <subversion at mplayerhq.hu> writes:
>>>>>>
>>>>>>> Author: bcoudurier
>>>>>>> Date: Wed Aug  6 01:03:02 2008
>>>>>>> New Revision: 14631
>>>>>>>
>>>>>>> Log:
>>>>>>> add dv 4:1:1 regression test
>>>>>>>
>>>>>>> Modified:
>>>>>>>    trunk/tests/ffmpeg.regression.ref
>>>>>>>    trunk/tests/regression.sh
>>>>>>>    trunk/tests/rotozoom.regression.ref
>>>>>>>
>>>>>>> Modified: trunk/tests/regression.sh
>>>>>>> ==============================================================================
>>>>>>> --- trunk/tests/regression.sh	(original)
>>>>>>> +++ trunk/tests/regression.sh	Wed Aug  6 01:03:02 2008
>>>>>>> @@ -353,6 +353,9 @@ fi
>>>>>>>  if [ -n "$do_dv" ] ; then
>>>>>>>  do_video_encoding dv.dv "-dct int" "-s pal -an"
>>>>>>>  do_video_decoding "" "-s cif"
>>>>>>> +
>>>>>>> +do_video_encoding dv411.dv "-dct int" "-s pal -an -pix_fmt yuv411p"
>>>>>>> +do_video_decoding "" "-s cif"
>>>>>>>  fi
>>>>>> Is there some reason this can't be a separate test, invoked from the
>>>>>> makefile like the rest?
>>>>>>
>>>>> I don't think so, feel free to split each test for each codec (like
>>>>> mpeg2) if you like.
>>>> You should have done that when you added this.  Please do it now.
>>>>
>>> Since when all tests are splitted ? Sorry but no, if you want it, you'll
>>> have to do it. I don't care about it, I only want to ensure that nothing
>>> will break dv 4:1:1.
>> 
>> They've been split ever since I split them half a year ago.
>
> I don't see mpeg2 tests split, you missed it apparently.

Not exactly.  Those if/then/fi blocks were already there.  Besides,
past mistakes are no excuse for new errors.

>> If you'd sent a patch for this, I'd have rejected it for this reason. 
>
> regression.sh has no maintainer.

Your point being?

Are you going to fix this?

-- 
M?ns Rullg?rd
mans at mansr.com