RESOLVED FIXED181977
[Cairo] Use GraphicsContextImplCairo for ImageBuffer context
https://bugs.webkit.org/show_bug.cgi?id=181977
Summary [Cairo] Use GraphicsContextImplCairo for ImageBuffer context
Zan Dobersek
Reported 2018-01-23 01:37:47 PST
[Cairo] Use GraphicsContextImplCairo for ImageBuffer context
Attachments
WIP (48.81 KB, patch)
2018-01-23 01:38 PST, Zan Dobersek
no flags
Patch (51.75 KB, patch)
2018-01-23 23:44 PST, Zan Dobersek
no flags
Zan Dobersek
Comment 1 2018-01-23 01:38:30 PST
Zan Dobersek
Comment 2 2018-01-23 23:44:25 PST
Zan Dobersek
Comment 3 2018-01-23 23:48:46 PST
(In reply to Zan Dobersek from comment #2) > Created attachment 332128 [details] > Patch Given this changes code in GraphicsContext, GraphicsContextImpl and DisplayList::Recorder, I'd prefer if Simon or Myles can review this.
Carlos Garcia Campos
Comment 4 2018-01-24 00:04:16 PST
Comment on attachment 332128 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=332128&action=review > Source/WebCore/platform/graphics/displaylists/DisplayListRecorder.h:60 > + bool hasPlatformContext() const override { return false; } > + PlatformGraphicsContext* platformContext() const override { return nullptr; } I wonder why we have hasPlatformContext, since platformContext() returns a pointer we could simply check its return value no?
Carlos Garcia Campos
Comment 5 2018-01-24 00:07:20 PST
(In reply to Zan Dobersek from comment #3) > (In reply to Zan Dobersek from comment #2) > > Created attachment 332128 [details] > > Patch > > Given this changes code in GraphicsContext, GraphicsContextImpl and > DisplayList::Recorder, I'd prefer if Simon or Myles can review this. I was already reviewing this when you added this comment. Changes look sane to me, but I agree it would be better if graphics guys double check it.
Simon Fraser (smfr)
Comment 6 2018-01-24 13:54:25 PST
Comment on attachment 332128 [details] Patch I like this. At some point I'd like the CG drawing to be a graphicsContextImpl too.
Zan Dobersek
Comment 7 2018-01-24 23:16:33 PST
Comment on attachment 332128 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=332128&action=review >> Source/WebCore/platform/graphics/displaylists/DisplayListRecorder.h:60 >> + PlatformGraphicsContext* platformContext() const override { return nullptr; } > > I wonder why we have hasPlatformContext, since platformContext() returns a pointer we could simply check its return value no? This was done to exactly mirror the GraphicsContext methods. It might be possible to just deduce this from the platformContext() value like you propose, but I'd wait on that until more implementations (e.g. GraphicsContextImplCG) see whether the extra method is necessary.
Zan Dobersek
Comment 8 2018-01-24 23:18:26 PST
Comment on attachment 332128 [details] Patch Clearing flags on attachment: 332128 Committed r227594: <https://trac.webkit.org/changeset/227594>
Zan Dobersek
Comment 9 2018-01-24 23:18:30 PST
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 10 2018-01-24 23:19:18 PST
Note You need to log in before you can comment on or make changes to this bug.