11384 2006-10-23 02:43:41 -0700 CSS2/3: Selectors are not re-evaluated after dynamically updating the DOM (Acid3 bug) 2008-02-21 22:19:21 -0800 1 1 1 Unclassified WebKit CSS 420+ Mac OS X 10.4 RESOLVED FIXED http://www.css3.info/selectors-test/ HasReduction P2 Normal --- 17143 11390 17064 1 info webkit-unassigned allan.jensen bdakin eric michiel mrowe webkit oldest_to_newest 51194 0 info 2006-10-23 02:43:41 -0700 After the DOM is modified the CSS selectors should be re-evaluated. Affected selectors: :first-child, :last-child, :only-child, :first-of-type, :last-of-type, :only-of-type, :empty For example: an element that used to be empty would match the :empty selector. If the update added a new child to that element, it should no longer match the :empty selector. Another example: an element that is the only child of its parent would match the following selectors: :first-child, :last-child, :only-child, :first-of-type, :last-of-type and :only-of-type. If we modify the DOM and append another child of the same type, the original element would only match the :first-child selector. 36587 1 mrowe 2007-01-07 02:50:10 -0800 I believe there an existing bug which may already cover this issue, but I can't find it in my quick skim through Bugzilla. 68549 2 18758 webkit 2008-01-29 00:32:11 -0800 Created attachment 18758 Minimal test case I've created minimal test case for this bug based on CSS3 Selectors test suite. 68555 3 webkit 2008-01-29 00:54:36 -0800 (In reply to comment #1) > I believe there an existing bug which may already cover this issue, but I can't > find it in my quick skim through Bugzilla. After searching I've found three reports similar to this bug: bug 12519, bug 12520, bug 16706. 68571 4 webkit 2008-01-29 04:42:37 -0800 *** Bug 16706 has been marked as a duplicate of this bug. *** 68846 5 webkit 2008-01-30 14:37:36 -0800 Found another similar bug 9983. 68851 6 webkit 2008-01-30 15:06:19 -0800 Interesting behavior regarding test case found by Alexey: 1. Open the test case in a new tab in background. 2. Don't activate the background tab unless it loads completely. 3. Switch to the background tab and you see the test case passed. IMO this behavior is caused by the fact that styles are applied when the page becomes visible (not earlier). So the styles are applied on modified DOM tree. The styles don't need to be re-evaluated. 69166 7 hyatt 2008-02-02 02:49:13 -0800 This has been taken care of. 71466 8 webkit 2008-02-21 22:19:21 -0800 *** Bug 9983 has been marked as a duplicate of this bug. *** 18758 2008-01-29 00:32:11 -0800 2008-01-29 00:32:11 -0800 Minimal test case test-case1.html text/html 531 webkit PCFET0NUWVBFIEhUTUwgUFVCTElDICItLy9XM0MvL0RURCBIVE1MIDQuMDEvL0VOIiAiaHR0cDov L3d3dy53My5vcmcvVFIvaHRtbDQvc3RyaWN0LmR0ZCI+CjxodG1sPgo8aGVhZD4KPHRpdGxlPlRl c3QgY2FzZSBmb3IgcmUtZXZhbHVhdGluZyBzZWxlY3RvcnMgYWZ0ZXIgZHluYW1pY2FsbHkgdXBk YXRpbmcgdGhlIERPTTwvdGl0bGU+CjxzdHlsZSB0eXBlPSJ0ZXh0L2NzcyI+CgkjdGVzdDpmaXJz dC1jaGlsZCB7YmFja2dyb3VuZC1jb2xvcjogcmVkO30KPC9zdHlsZT4KPC9oZWFkPgoJCQkJCjxk aXY+Cgk8ZGl2IGlkPSJ0ZXN0Ij5UaGlzIHRleHQgU0hPVUxEIE5PVCBoYXZlIFJFRCBiYWNrZ3Jv dW5kPC9kaXY+CjwvZGl2PgoKPHNjcmlwdCB0eXBlPSJ0ZXh0L2phdmFzY3JpcHQiPgoJdmFyIHRl c3QgPSBkb2N1bWVudC5nZXRFbGVtZW50QnlJZCgndGVzdCcpOwoJdGVzdC5wYXJlbnROb2RlLmlu c2VydEJlZm9yZShkb2N1bWVudC5jcmVhdGVFbGVtZW50KCJkaXYiKSwgdGVzdCk7Cjwvc2NyaXB0 PgoKPC9ib2R5Pgo8L2h0bWw+