Bug 185731 – [MSE][GStreamer] Add patch to jhbuild: matroskademux: Allow Matroska headers to be read more than once

• Component: WebKitGTK
• Status: RESOLVED FIXED

[Alicia Boya García]

Upstream bug in GStreamer: https://bugzilla.gnome.org/show_bug.cgi?id=793333

This fixes YTTV 35. AppendMultipleInitOpusAudio (and any other use
case where two WebM initialization segments are appended on a row).

[Alicia Boya García]

Created attachment 340589 [details]
Patch

[Michael Catanzaro]

Comment on attachment 340589 [details]
Patch

I don't see enough activity on the upstream Bugzilla (nothing's happened since February) to justify pulling in this patch now. We need to be really careful about pulling in downstream patches that haven't landed upstream yet. Let's get the patch landed upstream first.

[Xabier Rodríguez Calvar]

(In reply to Michael Catanzaro from comment #2)
> I don't see enough activity on the upstream Bugzilla (nothing's happened
> since February) to justify pulling in this patch now. We need to be really
> careful about pulling in downstream patches that haven't landed upstream
> yet. Let's get the patch landed upstream first.

This is not the first patch that is needed for WebKit that is taking long to land. GStreamer community is very busy. Besides this, MSE is not activated by default in GTK+ so harm is done there.

Anyway, we'll try to get this patch landed in GStreamer this week and if that does not happen, I'll r+ this myself because it is needed even when GStreamer upstream does not support it.

[Michael Catanzaro]

We discussed this last week on our internal chat and decided that Philippe and Thibault would review any pending patches and ensure they are committed upstream, and then Alicia can add them to our JHBuild moduleset once that's been handled.

[Xabier Rodríguez Calvar]

(In reply to Michael Catanzaro from comment #4)
> We discussed this last week on our internal chat and decided that Philippe
> and Thibault would review any pending patches and ensure they are committed
> upstream, and then Alicia can add them to our JHBuild moduleset once that's
> been handled.

IMHO we shouldn't remove them in any case. Those patches are needed and will be handled sooner or later (I know it is later than sooner but you know we do our best, no need to put more pressure ;) )

[Michael Catanzaro]

(In reply to Xabier Rodríguez Calvar from comment #5)
> IMHO we shouldn't remove them in any case. Those patches are needed and will
> be handled sooner or later (I know it is later than sooner but you know we
> do our best, no need to put more pressure ;) )

Better late than never: all our current patches are now upstream, so our GStreamer patch situation is now in a perfect state, as far as I'm concerned!

Alicia, feel free to add any other patches you need once they've first landed upstream, and complain very loudly if you're having trouble getting reviews for that. The ultimate goal is for your work to land, after all. I think we'll all be happier about the situation once that happens.

[Michael Catanzaro]

(In reply to Michael Catanzaro from comment #6)
> Better late than never: all our current patches are now upstream, so our
> GStreamer patch situation is now in a perfect state, as far as I'm concerned!

(Based on Thibault's comment https://bugs.webkit.org/show_bug.cgi?id=185689#c13).

[Alicia Boya García]

Created attachment 350631 [details]
Patch

[Alicia Boya García]

This is finally upstream, I hope there are no objections anymore for adding it to jhbuild.

[Xabier Rodríguez Calvar]

Please, there was something weird when landing this patch because this bug was not closed automatically.

[Alicia Boya García]

Committed r236398: <https://trac.webkit.org/changeset/236398/webkit>

[Enrique Ocaña]

*** Bug 179241 has been marked as a duplicate of this bug. ***