Bug 134528 – Improve handling of primary offscreen plugins

• Status: RESOLVED
• Resolution: FIXED
• Component: Plug-ins

[Roger Fong]

I noticed two issues with the handling of the primary off screen plugin case.
The first is that the determination of whether or not the plugin started off screen was happening too late, not at the initialization of the plugin as it should have been.
The second issue is that the plugin process type was being set to PluginProcessTypeSnapshot for these potential offscreen plugins. 
This caused them to be muted. The process type should instead be switched to PluginProcessTypeNormal. 
Note that setting the display state of the plugin to DisplaySnapshot also causes the plugin to be muted.
Alternatively, in the case where we do want to display a snapshot I could also recreate and reinitialize the plugin and force the plugin process type to be PluginProcessTypeSnapshot.

<rdar://problem/17471864>

[Roger Fong]

Created attachment 234223 [details]
patch

[Roger Fong]

This is a followup to https://bugs.webkit.org/show_bug.cgi?id=133667

[Dean Jackson]

Comment on attachment 234223 [details]
patch

Can we test this?

PS. GTK failed. Didn't look :)

[Roger Fong]

(In reply to comment #3)
> (From update of attachment 234223 [details])
> Can we test this?
Hmmm I suppose i can write a test which pulls in a fairly largely sized plugin from offscreen and have it play audio.
So I'd need a regular layout test result and an audio result...do audio results even work these days?

> 
> PS. GTK failed. Didn't look :)
Meh, unrelated failures.

[WebKit Commit Bot]

Comment on attachment 234223 [details]
patch

Clearing flags on attachment: 234223

Committed r170743: <http://trac.webkit.org/changeset/170743>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Carlos Alberto Lopez Perez]

(In reply to comment #4)
> > PS. GTK failed. Didn't look :)
> Meh, unrelated failures.

unrelated failures??? This just broke the GTK build: http://build.webkit.org/builders/GTK%20Linux%2064-bit%20Release%20%28Build%29/builds/48974/steps/compile-webkit/logs/stdio

[Carlos Alberto Lopez Perez]

(In reply to comment #7)
> (In reply to comment #4)
> > > PS. GTK failed. Didn't look :)
> > Meh, unrelated failures.
> 
> unrelated failures??? This just broke the GTK build: http://build.webkit.org/builders/GTK%20Linux%2064-bit%20Release%20%28Build%29/builds/48974/steps/compile-webkit/logs/stdio

It also broke the EFL build. 

Reported here: https://bugs.webkit.org/show_bug.cgi?id=134585

[Roger Fong]

(In reply to comment #8)
> (In reply to comment #7)
> > (In reply to comment #4)
> > > > PS. GTK failed. Didn't look :)
> > > Meh, unrelated failures.
> > 
> > unrelated failures??? This just broke the GTK build: http://build.webkit.org/builders/GTK%20Linux%2064-bit%20Release%20%28Build%29/builds/48974/steps/compile-webkit/logs/stdio
> 
> It also broke the EFL build. 
> 
> Reported here: https://bugs.webkit.org/show_bug.cgi?id=134585

Sorry bout that, I was looking at what I thought was GTK test results which would've implied GTK built successfully.