[FFmpeg-cvslog] r21254 - in trunk: Makefile tests/lavf.regression.ref tests/ref tests/ref/lavf tests/ref/lavf/aiff tests/ref/lavf/alaw tests/ref/lavf/asf tests/ref/lavf/au tests/ref/lavf/avi tests/ref/lavf/bmp tes...

Måns Rullgård mans
Sat Jan 16 22:09:20 CET 2010


Reimar D?ffinger <Reimar.Doeffinger at gmx.de> writes:

> On Sat, Jan 16, 2010 at 08:56:04PM +0000, M?ns Rullg?rd wrote:
>> Reimar D?ffinger <Reimar.Doeffinger at gmx.de> writes:
>> > On Sat, Jan 16, 2010 at 09:18:13PM +0100, mru wrote:
>> >> By default, make will abort if any test fails.  To run all tests
>> >> regardless, use make -k.
>> >
>> > I haven't tested it yet, but I suspect that will make FATE far less useful
>> > on some targets.
>> 
>> With this change separate FATE tests can be made for each test.  That
>> will be *more* useful than what we have now.
>
> Yes, though it needs someone to remember to add any new tests to
> FATE as well.  So I think it might make sense to have to run the
> full test as well...

FATE needs maintenance like everything else.  The non-regtest things
there didn't add themselves either.

>> > Can't the default make test be changed so it still runs all tests
>> > and shows the complete diff?
>> 
>> If you run make -k you get a diff for each failed test.  Isn't that
>> good enough?
>
> Yes, I think so, my concern was only about FATE.
> However I guess it is mostly moot, assuming separate tests will be
> added in the reasonable future.

Mike requested this a long time ago.  Let's see what he does.

> (I assume that the make return value will still indicate whether any
> test failed with make -k -- um assuming it did before, I admit I
> never really needed it).

Yes, make will return an error if anything failed.

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



More information about the ffmpeg-cvslog mailing list