On Sun, Jul 20, 2014 at 1:59 PM, Justin Ruggles
<justin.rugg...@gmail.com> wrote:
> On 07/20/2014 01:58 PM, Reinhard Tartler wrote:
>>
>> On Sun, Jul 20, 2014 at 1:50 PM, Justin Ruggles
>> <justin.rugg...@gmail.com> wrote:
>>>
>>> On 07/19/2014 01:22 PM, Reinhard Tartler wrote:
>>>>
>>>> On Sat, Jul 19, 2014 at 12:26 PM, Diego Biurrun <di...@biurrun.de>
>>>> wrote:
>>>>>
>>>>> On Sat, Jul 19, 2014 at 12:23:58PM -0400, Reinhard Tartler wrote:
>>>>>>
>>>>>> For release branches, we currently don't have any FATE Farm. It would
>>>>>> be great to have something like that, but as far as I understand, it
>>>>>> is a lot of effort to setup FATE for release branches, which is really
>>>>>> a shame.
>>>>>
>>>>> Huh? It's just a matter of changing a Git URL..
>>>>
>>>> I wasn't aware that it's that easy, please go ahead and change fate to
>>>> also cover the release branches.
>>>>
>>> We also need to be careful about removing or changing existing tests in
>>> ways
>>> that will break release branches. An alternative would be to copy all of
>>> the
>>> reference samples to a separate dir when a release branch is made and use
>>> that instead.
>>
>> That's what we've been doing since release/0.7:
>>
>>>> rsync -vaLW rsync://fate-suite.libav.org
>>
>> samples r/o access to Libav's samples collection
>> fate-suite r/o access to Libav's fate-suite
>> fate-suite-10 r/o access to Libav's fate-suite for Libav 10
>> fate-suite-9 r/o access to Libav's fate-suite for Libav 9
>> fate-suite-0.8 r/o access to Libav's fate-suite for Libav 0.8
>> fate-suite-0.7 r/o access to Libav's fate-suite for Libav 0.7
>>
>> We also adjust the target "fate-rsync" in tests/Makefile for release
>> branches.
>>
>>
>>
>
> Oh, nice. :)
Nevertheless, it would be nice if we could establish a rule of never
changing an existing file, but only adding fixed samples instead. Then
we wouldn't need to branch in the first place.
If disk space was a concern, we could explore the use of git-annex for
storing the fate data.
--
regards,
Reinhard
_______________________________________________
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
