Stop using testRunner.overridePreference() to turn on Page Cache, as this does not work for WebKit2 when navigating cross-origin.
Created attachment 379650 [details] WIP Patch
Created attachment 379659 [details] WIP Patch
Created attachment 379661 [details] Patch
Comment on attachment 379661 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=379661&action=review > Tools/DumpRenderTree/TestRunner.cpp:587 > + RELEASE_ASSERT(!JSStringIsEqualToUTF8CString(key.get(), "WebKitUsesPageCachePreferenceKey")); These assertions seem a bit overboard. If it does nothing in WebKitTestRunner, future developers will realize that.
(In reply to Alex Christensen from comment #4) > Comment on attachment 379661 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=379661&action=review > > > Tools/DumpRenderTree/TestRunner.cpp:587 > > + RELEASE_ASSERT(!JSStringIsEqualToUTF8CString(key.get(), "WebKitUsesPageCachePreferenceKey")); > > These assertions seem a bit overboard. If it does nothing in > WebKitTestRunner, future developers will realize that. I don't think it is that obvious, especially if somebody is writing a test for something does not not enter PageCache. Also, a RELEASE_ASSERT() in WKTR really does not hurt anything.
Comment on attachment 379661 [details] Patch Clearing flags on attachment: 379661 Committed r250398: <https://trac.webkit.org/changeset/250398>
All reviewed patches have been landed. Closing bug.
<rdar://problem/55758960>