Platform/HTML5 parser tasks: Difference between revisions

From MozillaWiki
Jump to navigation Jump to search
 
(153 intermediate revisions by 5 users not shown)
Line 1: Line 1:
This page outlines what needs to happen to 1) enable the HTML5 parser by default on mozilla-central and 2) ship a release with the HTML5 parser enabled by default.  
This page outlines what needs to be done to 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 shipping the HTML5 parser in a final Firefox release ==


== Things that need to happen before turning the HTML5 parser on by default ==
* Fix "[HTML5]" bugs marked [https://bugzilla.mozilla.org/buglist.cgi?priority=P2;short_desc=%5bHTML5%5d;field0-0-0=short_desc;query_format=advanced;bug_status=NEW;bug_status=ASSIGNED;short_desc_type=allwordssubstr;type0-0-0=notsubstring;value0-0-0=%5bPatch%5d P2]. (hsivonen)
** [https://bugzilla.mozilla.org/show_bug.cgi?id=569528 569528] <h5> inside <button> inside <p> implicitly closes the <p>
** [https://bugzilla.mozilla.org/show_bug.cgi?id=579846 579846] HTML5 parser regresses cache-control support in meta pragmas
** [https://bugzilla.mozilla.org/show_bug.cgi?id=579867 579867] Implement July 2010 spec changes to SVG in HTML
*** Related [https://bugzilla.mozilla.org/show_bug.cgi?id=577416 577416] [https://bugzilla.mozilla.org/show_bug.cgi?id=574884 574884] [https://bugzilla.mozilla.org/show_bug.cgi?id=577418 577418]
** [https://bugzilla.mozilla.org/show_bug.cgi?id=564098 564098] "ASSERTION: Script global object not ready"
** [https://bugzilla.mozilla.org/show_bug.cgi?id=577508 577508] Hide old document when new one pops a dialog
*Reviews on patches
** [https://bugzilla.mozilla.org/show_bug.cgi?id=574884 574884] <math></html> crashes (sicking)
** [https://bugzilla.mozilla.org/show_bug.cgi?id=570309 570309] -- > closing comments breaks CNN Money (sicking)
** [https://bugzilla.mozilla.org/show_bug.cgi?id=575175 575175] Deal with Unicode decoder consuming -1 bytes (smontagu)
*Security review.
** Talked to dveditz about how to do the security review for the HTML5 parser, and it's not clear what the best format for the review would be. But we did decide that it's not something we need to do before turning the parser on by default, but we need to do it before we ship it. --jst
* [https://bugzilla.mozilla.org/show_bug.cgi?id=544994 544994]Create a "slow tp" test suite for exercising speculative parsing. (who owns this now?)
** ''What if this suite reveals something bad?''


*Land 13 or so reviewed patches in hsivonen's queue (trivial once the tree is green). (hsivonen)
==Suspended items that may re-enter the list of release blockers==
*Get reviews on pending [https://bugzilla.mozilla.org/buglist.cgi?emailassigned_to1=1;query_format=advanced;short_desc=%5bHTML5%5d;field0-0-0=setters.login_name;bug_status=ASSIGNED;short_desc_type=allwordssubstr;email1=hsivonen%40iki.fi;type0-0-0=equals;value0-0-0=hsivonen%40iki.fi;emailtype1=exact patches]. (various people)
*Make tp4 not regress. (hsivonen)
*Drive testsuite failures to zero. (hsivonen but help would be useful)
*Either change form submission code to deal with IE-style isindex (hsivonen but help would be useful) or change the parser to use a Gecko-style isindex for now (hsivonen).
*Update the html5lib test suite snapshot used by mochitest. (jgriffin)
*Review the unreviewed parser core code. (sicking?)
**Some more code documentation on the Java side would be useful. (hsivonen)
*Create a "slow tp" test suite for exercising speculative parsing. (bnewman) ''What if this suite reveals something bad?''


== Things that need to happen before shipping the HTML5 parser is a final Firefox release ==
* Suspended bugs that were previously P2
** [https://bugzilla.mozilla.org/show_bug.cgi?id=482917 482917] Make HTML5 parsing not regress stream data availability to extensions
***There's now a patch but no way to test it.
***Suspended, because there are no known extensions that use the API in question and are compatible with Firefox 3.6.
***''Time estimate'': Unknown. (Need to find an extension that actually uses this. The alternative is writing an extension just for testing.)


* Fix "[HTML5]" bugs marked [https://bugzilla.mozilla.org/buglist.cgi?priority=P2;short_desc=%5bHTML5%5d;field0-0-0=short_desc;query_format=advanced;bug_status=NEW;bug_status=ASSIGNED;short_desc_type=allwordssubstr;type0-0-0=notsubstring;value0-0-0=%5bPatch%5d P2]. (hsivonen)
==Notable evangelism bugs==
 
* Bugs that need site-side fixes
** [https://bugzilla.mozilla.org/show_bug.cgi?id=540480 540480] Hootsuite.com dashboard not loading with HTML5 parser turned on
***Site contacted, contact acknowledged (hsivonen)
** [https://bugzilla.mozilla.org/show_bug.cgi?id=553795 553795] Blank page on weer.nl (Windows only)
***''No owner''
***Bad browser sniffing leads to document.write without an insertion point
** [https://bugzilla.mozilla.org/show_bug.cgi?id=565689 565689] </script> in string literal breaks out of script early on wsecu.org
***Site contacted (mbrubeck)
***''Bank''
** [https://bugzilla.mozilla.org/show_bug.cgi?id=565938 565938] Blank page on LinkedIn Recruiter
***Site rep CCed on the bug.
***Bad browser sniffing leads to document.write without an insertion point

Latest revision as of 15:44, 19 July 2010

This page outlines what needs to be done to ship a release with the HTML5 parser enabled by default.

Things that need to be done before shipping the HTML5 parser in a final Firefox release

  • Fix "[HTML5]" bugs marked P2. (hsivonen)
    • 569528 <h5> inside <button> inside <p> implicitly closes the <p>
    • 579846 HTML5 parser regresses cache-control support in meta pragmas
    • 579867 Implement July 2010 spec changes to SVG in HTML
    • 564098 "ASSERTION: Script global object not ready"
    • 577508 Hide old document when new one pops a dialog
  • Reviews on patches
    • 574884 <math></html> crashes (sicking)
    • 570309 -- > closing comments breaks CNN Money (sicking)
    • 575175 Deal with Unicode decoder consuming -1 bytes (smontagu)
  • Security review.
    • Talked to dveditz about how to do the security review for the HTML5 parser, and it's not clear what the best format for the review would be. But we did decide that it's not something we need to do before turning the parser on by default, but we need to do it before we ship it. --jst
  • 544994Create a "slow tp" test suite for exercising speculative parsing. (who owns this now?)
    • What if this suite reveals something bad?

Suspended items that may re-enter the list of release blockers

  • Suspended bugs that were previously P2
    • 482917 Make HTML5 parsing not regress stream data availability to extensions
      • There's now a patch but no way to test it.
      • Suspended, because there are no known extensions that use the API in question and are compatible with Firefox 3.6.
      • Time estimate: Unknown. (Need to find an extension that actually uses this. The alternative is writing an extension just for testing.)

Notable evangelism bugs

  • Bugs that need site-side fixes
    • 540480 Hootsuite.com dashboard not loading with HTML5 parser turned on
      • Site contacted, contact acknowledged (hsivonen)
    • 553795 Blank page on weer.nl (Windows only)
      • No owner
      • Bad browser sniffing leads to document.write without an insertion point
    • 565689 </script> in string literal breaks out of script early on wsecu.org
      • Site contacted (mbrubeck)
      • Bank
    • 565938 Blank page on LinkedIn Recruiter
      • Site rep CCed on the bug.
      • Bad browser sniffing leads to document.write without an insertion point