Moving this means we'll be able to use this logic in core.
<rdar://problem/83884806>
Created attachment 440266 [details] Patch
Comment on attachment 440266 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=440266&action=review > Source/WebCore/accessibility/mac/WebAccessibilityObjectWrapperMac.mm:-2384 > - // The contents of a tab list are all the children except the tabs. The comment says this, but then the code that follows does the exact opposite thing. I elected to copy the current behavior when I moved this logic to AccessibilityObject.cpp.
Committed r283601 (242553@main): <https://commits.webkit.org/242553@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 440266 [details].
(In reply to Tyler Wilcock from comment #2) > Created attachment 440266 [details] > Patch --- a/Source/WebCore/accessibility/AccessibilityObjectInterface.h +++ a/Source/WebCore/accessibility/AccessibilityObjectInterface.h + virtual void contents(AccessibilityChildrenVector&) = 0; Should be virtual AccessibilityChildrenVector contents() = 0; No need to pass an out parameter. Out parameters are bad and make client code less clear unnecessarily. Also, can we make this method const: virtual AccessibilityChildrenVector contents() const = 0; or some of the function calls in the implementation cannot be const.
(In reply to Tyler Wilcock from comment #2) > Created attachment 440266 [details] > Patch --- a/Source/WebCore/accessibility/AccessibilityObject.cpp +++ a/Source/WebCore/accessibility/AccessibilityObject.cpp + for (RefPtr<AccessibilityObject> child = firstChild(); child; child = child->nextSibling()) { Change RefPtr<AccessibilityObject> to auto*
Handling review comments in: https://bugs.webkit.org/show_bug.cgi?id=231289
> Also, can we make this method const: Unfortunately not, because tabChildren is not const.