Bug 39259
| Summary: | MASTER: WebKit needs an HTML5 Tokenizer | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Eric Seidel (no email) <eric> |
| Component: | WebCore Misc. | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED FIXED | ||
| Severity: | Normal | CC: | abarth, annevk, ap, aroben, arv, bfulgham, eric.carlson, jchaffraix, jeffschiller, joepeck, joone, krit, michelangelo, mike, mjs, Ms2ger, paulirish, peter, thakis, tonyg, wangtao.adi, webmaster |
| Priority: | P2 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | All | ||
| OS: | All | ||
| Bug Depends on: | 39260, 39261, 39262, 39263, 39264, 39265, 39267, 39272, 39273, 39274, 39275, 39298, 39311, 39333, 39334, 39335, 39342, 39345, 39349, 39350, 39520, 39525, 39533, 39534, 39537, 39571, 39623, 39630, 39638, 39641, 39642, 39649, 39653, 39660, 39716, 39755, 39796, 39802, 39809, 39815, 39823, 39828, 39873, 39891, 39939, 39981, 39982, 39984, 39985, 40048, 40172, 40199, 40200, 40262, 40264, 40265, 40268, 40274, 40282, 40283, 40284, 40311, 40347, 40354, 40356, 40384, 40385, 40388, 40391, 40393, 40404, 40406, 40416, 40417, 40428, 40454, 40458, 40485, 40491, 40557, 40592, 40596, 40638, 40645, 40649, 40650, 40657, 40658, 40659, 40664, 40726, 40729, 40739, 40764, 40802, 40946 | ||
| Bug Blocks: | 30388, 32934 | ||
Eric Seidel (no email)
MASTER: WebKit needs an HTML5 Tokenizer
Adam and I are going to work on adding one this week. This is a master bug which we will relate all of the various bugs to.
We're going to start by forking from PreloadScanner. We will be in #webkit and happy to discuss our design thoughts.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Adam Barth
Henri mentioned that Mozilla has 6 open evangelism related to their HTML5 parser:
http://www.w3.org/Bugs/Public/show_bug.cgi?id=9767#c3
Adam Barth
Here's the spreadsheet we're using to triage the remaining LayoutTest failures:
https://spreadsheets.google.com/ccc?key=0AppchfQ5mBrEdDFJUW5DOGNsdmtvZkN0ZmIzMjdaT0E&hl=en
Currently, there are around 150 failures. However, for some of them, our new behavior is correct w.r.t. the spec but the tests have their expectations set to the old parser behavior.
Adam Barth
This is now enabled by default. There are still some dependent bugs to finish up, but we can close the master. We'll probably open a new master to deal with regressions.
Adam Barth
Here's the master bug for tracking compatibility regressions due to the new tokenizer:
https://bugs.webkit.org/show_bug.cgi?id=41115
Eric Seidel (no email)
*** Bug 40634 has been marked as a duplicate of this bug. ***