"Ronald S. Bultje" <[email protected]> writes:

> Hi,
>
> 2012/1/8 Måns Rullgård <[email protected]>:
>> Diego Biurrun <[email protected]> writes:
>>> On Tue, Dec 13, 2011 at 12:47:13AM +0000, Måns Rullgård wrote:
>>>> Diego Biurrun <[email protected]> writes:
>>>> > On Tue, Dec 13, 2011 at 12:56:36AM +0100, Luca Barbato wrote:
>>>> >> On 13/12/11 00:52, Måns Rullgård wrote:
>>>> >> >Luca Barbato<[email protected]>  writes:
>>>> >> >>On 13/12/11 00:32, Diego Biurrun wrote:
>>>> >> >>>---
>>>> >> >>>   Makefile           |    4 +++-
>>>> >> >>>   doc/developer.texi |    4 +---
>>>> >> >>>   2 files changed, 4 insertions(+), 4 deletions(-)
>>>> >> >>
>>>> >> >>Seems ok.
>>>> >> >
>>>> >> >But oh so pointless.
>>>> >>
>>>> >> Shall we drop it then?
>>>> >
>>>> > No.  It's a convenient shorthand that will reduce the number of
>>>> > mistakes in the long run if all developers pick up the habit.
>>>>
>>>> People fail to do even the most elementary checking today.  This will
>>>> never change.
>>>
>>> Flaming aside :)
>>>
>>> This issue has come up again with the breakage of the h264-test program.
>>> This just adds one line and can save us from such troubles.
>>
>> Believe me, it will not help with anything whatsoever.  Why would
>> renaming the test cause people to suddenly start running it?
>
> I agree with this; a bunch of test programs are useless if they're not
> run. If they are not tests, but examples, they should be renamed as
> such.
>
> Which are they? Examples or tests? If they are tests, what is the
> expected outcome, what do they test, why aren't they in fate already?

The ones that actually do something verifiably right or wrong are
already in fate.  A few more can probably be massaged into actually
testing something too.

-- 
Måns Rullgård
[email protected]
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to