Block level pseudo elements bleed background color to <html>
Created attachment 179138 [details] Patch
Comment on attachment 179138 [details] Patch Attachment 179138 [details] did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/15317012 New failing tests: inspector-protocol/debugger-terminate-dedicated-worker-while-paused.html
(In reply to comment #2) > (From update of attachment 179138 [details]) > Attachment 179138 [details] did not pass chromium-ews (chromium-xvfb): > Output: http://queues.webkit.org/results/15317012 > > New failing tests: > inspector-protocol/debugger-terminate-dedicated-worker-while-paused.html This is definitely not related, seeing as this patch only contains tests. :)
Comment on attachment 179138 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=179138&action=review > LayoutTests/ChangeLog:21 > + This was reported as a Chromium Bug: > + https://code.google.com/p/chromium/issues/detail?id=75544 We usually don't put downstream bug here as they have little signification in WebKit. > LayoutTests/ChangeLog:23 > + Extra space. > LayoutTests/fast/css-generated-content/html-pseudo-background-color.html:8 > + } I think you should also check for html:after. Either in this test (with a different color and the condition for passing would be "there should be no background color" or another one. > LayoutTests/fast/css-generated-content/html-pseudo-background-color.html:11 > +<p> Your bug should say what it tests (ie include a description). Without the ChangeLog, it is difficult to see what could go wrong. Including the bug number and title is also nice.
(In reply to comment #4) > (From update of attachment 179138 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=179138&action=review > ... > > Your bug should say what it tests (ie include a description). Without the ChangeLog, it is difficult to see what could go wrong. So to explain what this bug is about: RenderObject::isRoot checks if document()->documentElement() == m_node and in the old generated content implementation m_node was set to the generatingNode so for html:before or html:after so isRoot() would return true. This meant that when we go through RenderBox::paintBackground we would end up painting the generated content's background color over the whole document. This is fixed by the new generated content implementation that was turned on in r137336 because m_node is now the PseudoElement, not the <html> element.
Created attachment 180271 [details] Patch
Comment on attachment 180271 [details] Patch Clearing flags on attachment: 180271 Committed r138274: <http://trac.webkit.org/changeset/138274>
All reviewed patches have been landed. Closing bug.