Follow up of bug 65140, our behavior was reverted to our previous behavior, which does not match HTML5. This was done to avoid more regressions. However to keep test coverage, we have kept some of the developped test cases in the trunk. Some of them are failing currently as they are made to match the existing spec (HTML5 and CSS OM). They also should ensure that we don't get those regression back.
What are the failing tests? I don't see anything obvious - are they skipped now?
(In reply to comment #1) > What are the failing tests? Here is the list of test landed with failing baselines. They match FF and Opera behavior (IE does not support link.sheet so those tests really don't match it): * fast/css/stylesheet-enable-first-alternate-on-load-link.html * fast/css/stylesheet-enable-first-alternate-on-load-sheet.html * fast/css/stylesheet-enable-second-alternate-link.html * fast/css/stylesheet-enable-second-alternate-on-load-link.html * fast/css/stylesheet-enable-second-alternate-on-load-sheet.html * http/tests/css/link-css-disabled-value-with-slow-loading-sheet-in-error.html * http/tests/css/link-css-disabled-value-with-slow-loading-sheet.html We regressed some part of those tests: * fast/css/link-disabled-attr-expected.txt * fast/dom/HTMLLinkElement/disabled-attribute-expected.txt * fast/dom/boolean-attribute-reflection-expected.txt (note: regressing here means with respect to the current HTML5 which needs to be changed to accomodate what web-authors are expecting) > I don't see anything obvious - are they skipped now? They are not. They were landed with the appropriate (ie wrong) baselines so that they do pass. This was done to at least keep the coverage until someone comes around and fixes the situation.
FYI I raised the issue we had in our implementation to the WHATWG some time ago - http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2011-October/033392.html Here is the interesting part of how Firefox implements the spec (from the last email in the thread): "So for what it's worth, Gecko implemented the current behavior of creating the stylesheet immediately as soon as we know the <link> is linking to a stylesheet in https://bugzilla.mozilla.org/show_bug.cgi?id=107567 One of the considerations there was in fact allowing pages to change disabled state without having to wait for the sheet to load. That includes things like selection of alternate stylesheet sets working correctly even if not all the alternate sheets have finished loading and so forth..." Following FF here should ensure that we don't see the regressions we have seen in bug 65140 (basically exactly what is mentioned at the beginning of the 2nd pasted paragraph). This should also make us pass most of the test cases.