[FFmpeg-devel] FFmpeg 2.8

Hendrik Leppkes h.leppkes at gmail.com
Tue Sep 1 02:49:50 CEST 2015


On Mon, Aug 31, 2015 at 11:57 PM, Andreas Cadhalpun
<andreas.cadhalpun at googlemail.com> wrote:
> On 31.08.2015 14:55, compn wrote:
>> On Mon, 31 Aug 2015 14:19:02 +0200
>> Andreas Cadhalpun <andreas.cadhalpun at googlemail.com> wrote:
>>> On 31.08.2015 01:29, Hendrik Leppkes wrote:
>>>> On Mon, Aug 31, 2015 at 12:15 AM, Andreas Cadhalpun
>>>> <andreas.cadhalpun at googlemail.com> wrote:
>>>>> Before the release we should decide what to do with the asf_o
>>>>> demuxer. Since it still crashes often, it's probably best to just
>>>>> remove it.
>>>>>
>>>>> Does someone want to keep it?
>>>>
>>>> Yes, I want to keep it.
>>>
>>> Then you should get the crashes fixed.
>>>
>>>> It works much better for me than the old one.
>>>
>>> At least the old one doesn't crash.
>>
>> if it works for hendrik, then i vote for keeping asf_o.
>
> If Hendrik wants to use it, fine. But then he should also
> take the responsibility and maintain it, i.e. fix the crashes.

So because I want to use a feature, as a downstream user, I have to
maintain it now and take responsibility?
What kind of convoluted logic is that? Do we ask our users to maintain
the features they use now?

>
>> as long as its not default demuxer in ffmpeg, it wont hurt users.
>>
>> maybe we should have a codec_cap_experimental or
>> format_cap_experimental on it or so
>
> Even non-default, experimental demuxers shouldn't crash, as
> that's still a potential security risk.
> I don't think releasing a demuxer with known crashes is
> a good idea.
>
> Best regards,
> Andreas
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel


More information about the ffmpeg-devel mailing list