[selectors] :focus-visible parsing and experimental flag
Created attachment 420299 [details] Patch
Created attachment 420307 [details] Patch
Created attachment 420321 [details] Patch
EWSs are green or have unrelated failures, I believe this is ready for review. PTAL.
Comment on attachment 420321 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=420321&action=review > Source/WebCore/css/SelectorCheckerTestFunctions.h:480 > + if (InspectorInstrumentation::forcePseudoState(element, CSSSelector::PseudoClassFocusVisible)) > + return true; > + > + return false; Should just be a return statement, no need for if true else false. > LayoutTests/imported/w3c/ChangeLog:8 > + Update expectations as now :focus-visible is a valid selector. Is this because the feature is on for regression tests? Or do these changes happen with the feature still off? If the latter, I am worried that even these small changes could cause compatibility problems.
Comment on attachment 420321 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=420321&action=review Thanks for the review. >> Source/WebCore/css/SelectorCheckerTestFunctions.h:480 >> + return false; > > Should just be a return statement, no need for if true else false. True, changed that. >> LayoutTests/imported/w3c/ChangeLog:8 >> + Update expectations as now :focus-visible is a valid selector. > > Is this because the feature is on for regression tests? Or do these changes happen with the feature still off? > > If the latter, I am worried that even these small changes could cause compatibility problems. AFAIK the experimental feature flags are enabled by default for tests, but not when you run minibrowser for example.
Created attachment 420504 [details] Patch
Created attachment 420510 [details] Patch
Committed r272983: <https://commits.webkit.org/r272983> All reviewed patches have been landed. Closing bug and clearing flags on attachment 420510 [details].
<rdar://problem/74421155>