RESOLVED FIXED Bug 64542
Move TransformState to platform/graphics and give it the option to transform just a FloatQuad
https://bugs.webkit.org/show_bug.cgi?id=64542
Summary Move TransformState to platform/graphics and give it the option to transform ...
Chris Marrin
Reported 2011-07-14 11:10:35 PDT
Move TransformState to platform/graphics and give it the option to transform just a FloatQuad
Attachments
Patch (46.59 KB, patch)
2011-07-14 15:35 PDT, Chris Marrin
no flags
Patch (46.90 KB, patch)
2011-07-14 15:43 PDT, Chris Marrin
no flags
Archive of layout-test-results from ec2-cr-linux-01 (6.12 MB, application/zip)
2011-07-14 17:09 PDT, WebKit Review Bot
no flags
Patch (81.93 KB, patch)
2011-07-15 11:17 PDT, Chris Marrin
no flags
Archive of layout-test-results from ec2-cr-linux-03 (6.23 MB, application/zip)
2011-07-15 11:36 PDT, WebKit Review Bot
no flags
Revised patch to fix win build problem (82.64 KB, patch)
2011-07-15 12:39 PDT, Chris Marrin
no flags
Patch (81.25 KB, patch)
2011-07-15 13:25 PDT, Chris Marrin
no flags
Patch (76.83 KB, patch)
2011-07-15 13:50 PDT, Chris Marrin
simon.fraser: review+
simon.fraser: commit-queue-
Chris Marrin
Comment 1 2011-07-14 15:35:30 PDT
Chris Marrin
Comment 2 2011-07-14 15:43:29 PDT
Simon Fraser (smfr)
Comment 3 2011-07-14 15:53:59 PDT
Comment on attachment 100878 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=100878&action=review I think this looks good, but would like to see HitTestingTransformState stay in rendering. Is there a way to share the code but just have a RefCounted subclass? > Source/WebCore/platform/graphics/transforms/TransformState.cpp:133 > +// HitTestingTransformState methods I think HitTestingTransformState should either be move back under rendering/, or renamed to RefCountedTransformState or something. > Source/WebCore/platform/graphics/transforms/TransformState.h:103 > - bool m_mapQuad; > + bool m_mapPoint, m_mapQuad; Style guidelines recommend having these on separate lines.
WebKit Review Bot
Comment 4 2011-07-14 17:09:11 PDT
Comment on attachment 100878 [details] Patch Attachment 100878 [details] did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/9064394 New failing tests: fast/repaint/block-selection-gap-in-composited-layer.html editing/selection/image-before-linebreak.html editing/selection/replaced-boundaries-3.html fast/repaint/japanese-rl-selection-repaint.html
WebKit Review Bot
Comment 5 2011-07-14 17:09:21 PDT
Created attachment 100900 [details] Archive of layout-test-results from ec2-cr-linux-01 The attached test failures were seen while running run-webkit-tests on the chromium-ews. Bot: ec2-cr-linux-01 Port: Chromium Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick
Chris Marrin
Comment 6 2011-07-15 11:17:01 PDT
Simon Fraser (smfr)
Comment 7 2011-07-15 11:21:04 PDT
Comment on attachment 101009 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=101009&action=review > Source/WebCore/platform/graphics/transforms/TransformState.cpp:29 > +#include <wtf/PassOwnPtr.h> Don't need this include. > Source/WebCore/platform/graphics/transforms/TransformState.h:2 > + * Copyright (C) 2009 Apple Inc. All rights reserved. +2011 > Source/WebCore/platform/graphics/transforms/TransformState.h:34 > +#include <wtf/PassRefPtr.h> Not needed? > Source/WebCore/platform/graphics/transforms/TransformState.h:36 > +#include <wtf/RefCounted.h> Not needed? > Source/WebCore/rendering/HitTestingTransformState.cpp:2 > + * Copyright (C) 2009 Apple Inc. All rights reserved. Add 2011 too. > Source/WebCore/rendering/HitTestingTransformState.h:40 > +// So there's really no need for a ref counted version. So This class should be tossed and replaced s/tossed/removed :)
WebKit Review Bot
Comment 8 2011-07-15 11:36:50 PDT
Comment on attachment 101009 [details] Patch Attachment 101009 [details] did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/9094360 New failing tests: fast/repaint/block-selection-gap-in-composited-layer.html editing/selection/image-before-linebreak.html editing/selection/replaced-boundaries-3.html fast/repaint/japanese-rl-selection-repaint.html
WebKit Review Bot
Comment 9 2011-07-15 11:36:56 PDT
Created attachment 101014 [details] Archive of layout-test-results from ec2-cr-linux-03 The attached test failures were seen while running run-webkit-tests on the chromium-ews. Bot: ec2-cr-linux-03 Port: Chromium Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick
Chris Marrin
Comment 10 2011-07-15 12:39:04 PDT
Created attachment 101027 [details] Revised patch to fix win build problem
Chris Marrin
Comment 11 2011-07-15 13:25:44 PDT
Chris Marrin
Comment 12 2011-07-15 13:50:14 PDT
Chris Marrin
Comment 13 2011-07-15 14:17:49 PDT
Simon Fraser (smfr)
Comment 14 2011-07-15 14:48:37 PDT
Comment on attachment 101039 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=101039&action=review > Source/WebCore/platform/graphics/transforms/TransformState.cpp:2 > + * Copyright (C) 2009 Apple Inc. All rights reserved. +2011 > Source/WebCore/platform/graphics/transforms/TransformState.h:2 > + * Copyright (C) 2009 Apple Inc. All rights reserved. +2011 > Source/WebCore/platform/graphics/transforms/TransformState.h:34 > +#include <wtf/PassRefPtr.h> Remove. > Source/WebCore/platform/graphics/transforms/TransformState.h:36 > +#include <wtf/RefCounted.h> Remove. > Source/WebCore/platform/graphics/transforms/TransformState.h:103 > + bool m_mapPoint, m_mapQuad; Separate lines please.
Note You need to log in before you can comment on or make changes to this bug.