Summary: | HTMLTokenizer should ASSERT that it never emits a null character | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Eric Seidel (no email) <eric> | ||||
Component: | New Bugs | Assignee: | Eric Seidel (no email) <eric> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | abarth, commit-queue | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | Other | ||||||
OS: | OS X 10.5 | ||||||
Bug Depends on: | |||||||
Bug Blocks: | 41123 | ||||||
Attachments: |
|
Description
Eric Seidel (no email)
2010-07-01 01:11:55 PDT
Created attachment 60214 [details]
Patch
I still think it's a bad bad idea in HTML5 to replace nulls with FFFD's. The ASSERTs here are mostly to guard against us ever emitting \0 which we've repurposed to mean EndOfFile. You should definitely take your complaints up with Hixie. If we do allow emitting \0 at some point then we'll need to use a different character to mean EOF. Also, one side-effect of allowing \0 in the rendered content is that diff treats files with \0 in them as binary (which bit us in bug 41436). Comment on attachment 60214 [details]
Patch
@ap: I don't mean to ignore your comments. I certainly find them valuable, but that's an spec-level issue, not something that should block us ASSERTing that the implementation is doing what we intend it to do at the moment.
Comment on attachment 60214 [details] Patch Clearing flags on attachment: 60214 Committed r62352: <http://trac.webkit.org/changeset/62352> All reviewed patches have been landed. Closing bug. |