Bug 191490 - RealtimeOutgoing A/V sources should observe their sources only if having a sink
Summary: RealtimeOutgoing A/V sources should observe their sources only if having a sink
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebRTC (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: youenn fablet
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2018-11-09 15:12 PST by youenn fablet
Modified: 2018-11-12 11:50 PST (History)
4 users (show)

See Also:

Attachments
Patch (18.84 KB, patch)
2018-11-09 15:37 PST, youenn fablet 		no flags Details | Formatted Diff | Diff
Patch (19.91 KB, patch)
2018-11-09 16:42 PST, youenn fablet 		no flags Details | Formatted Diff | Diff
Patch (20.19 KB, patch)
2018-11-09 20:46 PST, youenn fablet 		no flags Details | Formatted Diff | Diff
Patch (21.02 KB, patch)
2018-11-09 21:21 PST, youenn fablet 		no flags Details | Formatted Diff | Diff
Patch for landing (22.01 KB, patch)
2018-11-12 11:00 PST, youenn fablet 		no flags Details | Formatted Diff | Diff
Show Obsolete (4) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description youenn fablet 2018-11-09 15:12:00 PST 
This is useless otherwise
Comment 1 youenn fablet 2018-11-09 15:37:55 PST 
Created attachment 354395 [details]
Patch
Comment 2 youenn fablet 2018-11-09 16:42:16 PST 
Created attachment 354413 [details]
Patch
Comment 3 youenn fablet 2018-11-09 20:46:27 PST 
Created attachment 354434 [details]
Patch
Comment 4 youenn fablet 2018-11-09 21:21:04 PST 
Created attachment 354438 [details]
Patch
Comment 5 Eric Carlson 2018-11-12 08:29:15 PST 
Comment on attachment 354438 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=354438&action=review

> Source/WebCore/platform/mediastream/RealtimeOutgoingAudioSource.cpp:65
> +    observeSource();

As we discussed, this shouldn't be necessary because you also call it in AddSink.

> Source/WebCore/platform/mediastream/RealtimeOutgoingVideoSource.cpp:73
> +    observeSource();

Ditto.

> Source/WebCore/platform/mediastream/RealtimeOutgoingVideoSource.cpp:81
>      m_videoSource->removeObserver(*this);

Nit: why does RealtimeOutgoingAudioSource have unobserveSource and RealtimeOutgoingVideoSource just calls removeObserver?
Comment 6 youenn fablet 2018-11-12 10:27:50 PST 
Thanks for the review.

(In reply to Eric Carlson from comment #5)
> Comment on attachment 354438 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=354438&action=review
> 
> > Source/WebCore/platform/mediastream/RealtimeOutgoingAudioSource.cpp:65
> > +    observeSource();
> 
> As we discussed, this shouldn't be necessary because you also call it in
> AddSink.

Yes, I will lock the sinks so that we observe/unobserve appropriately.

> > Source/WebCore/platform/mediastream/RealtimeOutgoingVideoSource.cpp:73
> > +    observeSource();
> 
> Ditto.
> 
> > Source/WebCore/platform/mediastream/RealtimeOutgoingVideoSource.cpp:81
> >      m_videoSource->removeObserver(*this);
> 
> Nit: why does RealtimeOutgoingAudioSource have unobserveSource and
> RealtimeOutgoingVideoSource just calls removeObserver?

I will add unobserveSource.
Comment 7 youenn fablet 2018-11-12 11:00:36 PST 
Created attachment 354571 [details]
Patch for landing
Comment 8 WebKit Commit Bot 2018-11-12 11:49:20 PST 
Comment on attachment 354571 [details]
Patch for landing

Clearing flags on attachment: 354571

Committed r238102: <https://trac.webkit.org/changeset/238102>
Comment 9 WebKit Commit Bot 2018-11-12 11:49:22 PST 
All reviewed patches have been landed.  Closing bug.
Comment 10 Radar WebKit Bug Importer 2018-11-12 11:50:38 PST 
<rdar://problem/45998054>