186997 2018-06-25 02:51:07 -0700 [Web Animations] Ensure animations are updated prior to requestAnimationFrame callbacks 2018-06-25 13:35:47 -0700 1 1 1 Unclassified WebKit Animations WebKit Nightly Build All All RESOLVED FIXED https://bugs.webkit.org/show_bug.cgi?id=183826 InRadar P2 Normal --- 1 graouts webkit-unassigned ddkilzer dino realdawei ross.kirsling webkit-bug-importer oldest_to_newest 1436022 0 graouts 2018-06-25 02:51:07 -0700 The HTML5 event loop mandates that animations are updated prior to requestAnimationFrame callbacks being fired. We need to implement this if only to reduce test flakiness since some tests use a rAF callback to check that animations have been updated. This is needed to fix https://bugs.webkit.org/show_bug.cgi?id=183826. 1436023 1 webkit-bug-importer 2018-06-25 02:51:35 -0700 <rdar://problem/41419414> 1436024 2 webkit-bug-importer 2018-06-25 02:51:42 -0700 <rdar://problem/41419416> 1436025 3 graouts 2018-06-25 02:54:42 -0700 Committed r233140: <https://trac.webkit.org/changeset/233140> 1436036 4 ddkilzer 2018-06-25 05:43:43 -0700 (In reply to Antoine Quint from comment #3) > Committed r233140: <https://trac.webkit.org/changeset/233140> Attempt to fix Windows build failure: Committed r233144: <https://trac.webkit.org/changeset/233144> 1436082 5 realdawei 2018-06-25 09:25:32 -0700 (In reply to David Kilzer (:ddkilzer) from comment #4) > (In reply to Antoine Quint from comment #3) > > Committed r233140: <https://trac.webkit.org/changeset/233140> > > Attempt to fix Windows build failure: > > Committed r233144: <https://trac.webkit.org/changeset/233144> unfortunately Windows is still failing to build: Sample error: https://build.webkit.org/builders/Apple%20High%20Sierra%20Release%20WK2%20(Tests)/builds/5228/steps/layout-test/logs/stdio C:\cygwin\home\buildbot\slave\win-release\build\Source\WebCore\dom/Document.cpp(2448): error C2027: use of undefined type 'WebCore::DocumentAnimationScheduler' (compiling source file C:\cygwin\home\buildbot\slave\win-release\build\WebKitBuild\Release\DerivedSources\WebCore\unified-sources\UnifiedSource206.cpp) [C:\cygwin\home\buildbot\slave\win-release\build\WebKitBuild\Release\Source\WebCore\WebCore.vcxproj] c:\cygwin\home\buildbot\slave\win-release\build\source\webcore\dom\Document.h(83): note: see declaration of 'WebCore::DocumentAnimationScheduler' (compiling source file C:\cygwin\home\buildbot\slave\win-release\build\WebKitBuild\Release\DerivedSources\WebCore\unified-sources\UnifiedSource206.cpp) 1436099 6 ross.kirsling 2018-06-25 10:15:09 -0700 The first error before r233144 was > WebCore\dom/Document.cpp(2448): error C2027: use of undefined type 'WebCore::DocumentAnimationScheduler' The first error after r233144 is now > WebCore\animation/KeyframeEffectReadOnly.cpp(707): error C2027: use of undefined type 'WebCore::FrameView' 1436110 7 ross.kirsling 2018-06-25 10:32:45 -0700 Looks like DocumentAnimationScheduler really is an undefined type on Windows: - https://github.com/WebKit/webkit/blob/master/Source/WTF/wtf/Platform.h#L1138-L1140 - https://github.com/WebKit/webkit/blob/master/Source/WebCore/animation/DocumentAnimationScheduler.h#L28 So I think r233144 could be reversed in favor of three extra #ifs in Document.cpp (unless we wanted to make a stub implementation). 1436179 8 ddkilzer 2018-06-25 12:24:29 -0700 (In reply to Ross Kirsling from comment #7) > Looks like DocumentAnimationScheduler really is an undefined type on Windows: > - > https://github.com/WebKit/webkit/blob/master/Source/WTF/wtf/Platform.h#L1138- > L1140 > - > https://github.com/WebKit/webkit/blob/master/Source/WebCore/animation/ > DocumentAnimationScheduler.h#L28 > > So I think r233144 could be reversed in favor of three extra #ifs in > Document.cpp (unless we wanted to make a stub implementation). Thanks! Just landed: Committed r233163: <https://trac.webkit.org/changeset/233163> 1436221 9 ddkilzer 2018-06-25 13:34:28 -0700 (In reply to David Kilzer (:ddkilzer) from comment #8) > (In reply to Ross Kirsling from comment #7) > > Looks like DocumentAnimationScheduler really is an undefined type on Windows: > > - > > https://github.com/WebKit/webkit/blob/master/Source/WTF/wtf/Platform.h#L1138- > > L1140 > > - > > https://github.com/WebKit/webkit/blob/master/Source/WebCore/animation/ > > DocumentAnimationScheduler.h#L28 > > > > So I think r233144 could be reversed in favor of three extra #ifs in > > Document.cpp (unless we wanted to make a stub implementation). > > Thanks! Just landed: > > Committed r233163: <https://trac.webkit.org/changeset/233163> And: Committed r233170: < https://trac.webkit.org/changeset/233170> 1436224 10 ross.kirsling 2018-06-25 13:35:47 -0700 (In reply to David Kilzer (:ddkilzer) from comment #9) > (In reply to David Kilzer (:ddkilzer) from comment #8) > > (In reply to Ross Kirsling from comment #7) > > > Looks like DocumentAnimationScheduler really is an undefined type on Windows: > > > - > > > https://github.com/WebKit/webkit/blob/master/Source/WTF/wtf/Platform.h#L1138- > > > L1140 > > > - > > > https://github.com/WebKit/webkit/blob/master/Source/WebCore/animation/ > > > DocumentAnimationScheduler.h#L28 > > > > > > So I think r233144 could be reversed in favor of three extra #ifs in > > > Document.cpp (unless we wanted to make a stub implementation). > > > > Thanks! Just landed: > > > > Committed r233163: <https://trac.webkit.org/changeset/233163> > > And: > > Committed r233170: < https://trac.webkit.org/changeset/233170> You beat me to the punch. :) Apparently KeyframeEffectReadOnly.cpp had a legitimate separate issue, but builds should go green after this one.