Bug 191459 – [GStreamer][EME] waitingforkey event should consider decryptors' waiting status

• Component: New Bugs
• Status: RESOLVED FIXED

[Xabier Rodríguez Calvar]

[GStreamer][EME] waitingforkey event should consider decryptors' waiting status

[Xabier Rodríguez Calvar]

Created attachment 354321 [details]
Patch

[Philippe Normand]

Comment on attachment 354321 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=354321&action=review

> Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:1385
> +    bool result { false };
> +    if (m_pipeline) {

Nit: Strange way to assign a variable? 
Also this condition could be transformed to an early return.

> Source/WebCore/platform/graphics/gstreamer/eme/WebKitCommonEncryptionDecryptorGStreamer.cpp:345
> +    return FALSE;

No chaining to parent class?

[Xabier Rodríguez Calvar]

(In reply to Philippe Normand from comment #2)
> Comment on attachment 354321 [details]

Yep to all of it. Thx!

Fixing the EWS as well. Weird that I am not getting that build failure myself...

[Charlie Turner]

Comment on attachment 354321 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=354321&action=review

Nice to get rid of those handlers in the decryptor sink handler :) I guess imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-waiting-for-a-key.https.html is still blocked on the event loop issue, just wondered if this affect the expectations of that test at all.

> Source/WebCore/ChangeLog:10
> +        going with the decryptors because events are reported only once

going on with

> Source/WebCore/ChangeLog:22
> +        Query the pipeline, just a query after after pipeline is built and

after after

> Source/WebCore/ChangeLog:23
> +        manual inspection during build. The optimal would be just the

The query is optimal, but sometimes ...

[Xabier Rodríguez Calvar]

Created attachment 354327 [details]
Patch

[Charlie Turner]

Informal r+ from me FWIW.

[Xabier Rodríguez Calvar]

Created attachment 354541 [details]
Patch

I fixed the concerns raised in this patch and remove the custom API for the iterator codepath as we can still use a query there.

[Xabier Rodríguez Calvar]

Created attachment 354542 [details]
Patch

Removed the WTF:: from boolForPrinting

[Xabier Rodríguez Calvar]

Phil, as you're on holidays, I asked Carlos to r+ this considering that I solved your concerns.

[WebKit Commit Bot]

Comment on attachment 354542 [details]
Patch

Clearing flags on attachment: 354542

Committed r238083: <https://trac.webkit.org/changeset/238083>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/45986842>