Platform/HTML5 parser tasks: Difference between revisions
< Platform
Jump to navigation
Jump to search
Line 10: | Line 10: | ||
**The fix interacts badly with speculative parsing on test cases that use sync XHR to control the test server. | **The fix interacts badly with speculative parsing on test cases that use sync XHR to control the test server. | ||
**Tuning requires try server cycles and could, therefore, take a few days. | **Tuning requires try server cycles and could, therefore, take a few days. | ||
*[https://bugzilla.mozilla.org/show_bug.cgi?id=536097 536097] spinning in nsHtml5TreeBuilder::startTag()? | |||
**This will either be simple or so rare that it's not an alpha blocker after all. | |||
*Drive testsuite failures to zero. (hsivonen but help would be useful) | *Drive testsuite failures to zero. (hsivonen but help would be useful) | ||
**[https://bugzilla.mozilla.org/show_bug.cgi?id=541078 541078] contentEditable reload: docshell/test/navigation/test_bug430624.html fails | **[https://bugzilla.mozilla.org/show_bug.cgi?id=541078 541078] contentEditable reload: docshell/test/navigation/test_bug430624.html fails | ||
***Needs debugging. | ***Needs debugging. | ||
**[https://bugzilla.mozilla.org/show_bug.cgi?id=545662 545662] browser_bookmarksProperties.js times out | **[https://bugzilla.mozilla.org/show_bug.cgi?id=545662 545662] browser_bookmarksProperties.js times out | ||
**[https://bugzilla.mozilla.org/show_bug.cgi?id=546636 546636] [Editor] layout/generic/test/test_backspace_delete.xul fails | **[https://bugzilla.mozilla.org/show_bug.cgi?id=546636 546636] [Editor] layout/generic/test/test_backspace_delete.xul fails | ||
Line 37: | Line 33: | ||
***[https://bugzilla.mozilla.org/show_bug.cgi?id=545664 545664] browser_423515.js fails | ***[https://bugzilla.mozilla.org/show_bug.cgi?id=545664 545664] browser_423515.js fails | ||
***[https://bugzilla.mozilla.org/show_bug.cgi?id=546381 546381] browser_bug537449.js throws and times out | ***[https://bugzilla.mozilla.org/show_bug.cgi?id=546381 546381] browser_bug537449.js throws and times out | ||
**The following are waiting for landings by others | |||
***[https://bugzilla.mozilla.org/show_bug.cgi?id=534458 534458] test failure on css3.info selectors test with :checked{} | |||
****Waiting on [https://bugzilla.mozilla.org/show_bug.cgi?id=502937 lazy frame construction] from Timothy Nikkel. (Seems reviewed and waiting landing.) | |||
***[https://bugzilla.mozilla.org/show_bug.cgi?id=543726 543726] dom/tests/mochitest/ajax/offline/test_foreign.html fails | |||
****honzab has a fix. Waiting for review. | |||
*[https://bugzilla.mozilla.org/show_bug.cgi?id=546635 Update] the html5lib test suite snapshot used by mochitest. (jgriffin) | *[https://bugzilla.mozilla.org/show_bug.cgi?id=546635 Update] the html5lib test suite snapshot used by mochitest. (jgriffin) | ||
*Review the unreviewed parser core code. (sicking?) | *Review the unreviewed parser core code. (sicking?) |
Revision as of 14:11, 3 March 2010
This page outlines what needs to be done to 1) enable the HTML5 parser by default on mozilla-central and 2) ship a release with the HTML5 parser enabled by default.
The HTML5 parser has been suitable for dogfooding for over a year and in the past couple of months there have been virtually no new site compat bugs. Thus, the parser is already alpha quality from a user POV but not from a zero test suite regressions POV.
Things that need to be done before turning the HTML5 parser on by default
- Get reviews on pending patches. (various people)
- Make tp4 not regress. (hsivonen)
- Potential fix in hand.
- The fix interacts badly with speculative parsing on test cases that use sync XHR to control the test server.
- Tuning requires try server cycles and could, therefore, take a few days.
- 536097 spinning in nsHtml5TreeBuilder::startTag()?
- This will either be simple or so rare that it's not an alpha blocker after all.
- Drive testsuite failures to zero. (hsivonen but help would be useful)
- 541078 contentEditable reload: docshell/test/navigation/test_bug430624.html fails
- Needs debugging.
- 545662 browser_bookmarksProperties.js times out
- 546636 [Editor] layout/generic/test/test_backspace_delete.xul fails
- If not dupe of bug 541078, probably needs help from the author of the test case.
- 546641 [Spec] layout/style/test/test_parse_rule.html fails
- Well understood and simple to fix.
- 547674 [Random?] content/xul/templates/tests/test_bug441785.xul fails
- Not understood.
- 547688 toolkit/content/tests/widgets/test_radio.xul fails
- 547933 test_bug451286.xul times out
- 547934 toolkit/content/tests/widgets/test_tooltip.xul times out
- 546648 toolkit/content/tests/widgets/test_mousecapture.xul fails
- Should be relatively simple to debug.
- The following are likely hidden by the tp4 patch.
- The following are waiting for landings by others
- 534458 test failure on css3.info selectors test with :checked{}
- Waiting on lazy frame construction from Timothy Nikkel. (Seems reviewed and waiting landing.)
- 543726 dom/tests/mochitest/ajax/offline/test_foreign.html fails
- honzab has a fix. Waiting for review.
- 534458 test failure on css3.info selectors test with :checked{}
- 541078 contentEditable reload: docshell/test/navigation/test_bug430624.html fails
- Update the html5lib test suite snapshot used by mochitest. (jgriffin)
- Review the unreviewed parser core code. (sicking?)
- Create a "slow tp" test suite for exercising speculative parsing. (bnewman) What if this suite reveals something bad?
- Security review.
Things that need to happen before shipping the HTML5 parser in a final Firefox release
- Fix "[HTML5]" bugs marked P2. (hsivonen)