Bug 5620

Summary:

Should only honor encoding from <meta> in HTML

Product:

WebKit

Reporter:

Alexey Proskuryakov <ap>

Component:

DOM

Assignee:

Alexey Proskuryakov <ap>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

cdumez, ian, troyb

Priority:

Version:

420+

Hardware:

Mac

OS:

OS X 10.4

Attachments:

Description	Flags
test case	none
proposed fix	none
same patch ignoring whitespace	none
revised patch	none
same patch, ignoring whitespace	none
revised patch	none

Alexey Proskuryakov

Reported 2005-11-03 13:32:47 PST

When a document is served as application/xhtml+xml, Firefox doesn't honor an encoding specified in a <meta> (probably, true for everything except for text/html). However, Decoder::decode() has no logic like this. I don't have a reference to any standard right now (haven't looked yet).

Attachments
test case (233 bytes, application/xhtml+xml) 2005-11-03 13:33 PST, Alexey Proskuryakov	no flags	Details
proposed fix (91.24 KB, patch) 2006-08-26 14:35 PDT, Alexey Proskuryakov	no flags	Details Formatted Diff Diff
same patch ignoring whitespace (46.58 KB, patch) 2006-08-27 00:53 PDT, Alexey Proskuryakov	no flags	Details Formatted Diff Diff
revised patch (51.48 KB, patch) 2006-08-28 11:15 PDT, Alexey Proskuryakov	no flags	Details Formatted Diff Diff
same patch, ignoring whitespace (45.02 KB, patch) 2006-08-28 11:16 PDT, Alexey Proskuryakov	no flags	Details Formatted Diff Diff
revised patch (43.69 KB, patch) 2006-08-31 21:36 PDT, Alexey Proskuryakov	no flags	Details Formatted Diff Diff
Show Obsolete (4) View All Add attachment proposed patch, testcase, etc.

Alexey Proskuryakov

Comment 1 2005-11-03 13:33:51 PST

Created attachment 4586 [details] test case This test case will render properly until a fix to 5548 is committed :)

Alexey Proskuryakov

Comment 2 2006-08-26 14:35:04 PDT

Created attachment 10244 [details] proposed fix Also fixes bug 9783 and bug 10155.

Eric Seidel (no email)

Comment 3 2006-08-26 16:47:05 PDT

Comment on attachment 10244 [details] proposed fix You might also try using svn --dif-cmd diff -x uPw to ignore whitespace. That might result in an even-easier-to-read diff. I didn't give it a full indepth review, but in general the changes look great. If I were to review this, I'd want to discuss some of the pseudo-xml cases with you over irc.

Alexey Proskuryakov

Comment 4 2006-08-27 00:53:37 PDT

Created attachment 10251 [details] same patch ignoring whitespace

Alexey Proskuryakov

Comment 5 2006-08-27 13:34:55 PDT

Comment on attachment 10244 [details] proposed fix Mitz has suggested a way not to dump pixels in CSS test cases. Also, after looking at the current XMLHttpRequest draft, I think that leaving the CSS case without incremental loading support is imprudent (currently, we only look inside XML content, so it doesn't matter, but that's not what the XHR spec says).

Alexey Proskuryakov

Comment 6 2006-08-28 11:15:44 PDT

Created attachment 10278 [details] revised patch This doesn't implement CSSCharsetRule creation, which is somewhat tricky, because WinIE doesn't seem to create those. Fortunately, IE and Firefox have different properties for accessing rules (.rules vs. .cssRules), so it might be possible to figure something out (I'll file a followup bug).

Alexey Proskuryakov

Comment 7 2006-08-28 11:16:15 PDT

Created attachment 10279 [details] same patch, ignoring whitespace

Alexey Proskuryakov

Comment 8 2006-08-31 21:36:37 PDT

Created attachment 10345 [details] revised patch Incorporating suggestions made by Eric on IRC.

Alexey Proskuryakov

Comment 9 2006-09-01 12:29:42 PDT

Comment on attachment 10345 [details] revised patch Was reviewed by Eric.

Alexey Proskuryakov

Comment 10 2006-09-01 12:29:58 PDT

Committed revision 16175.

Lucas Forschler

Comment 11 2019-02-06 09:04:05 PST

Mass moving XML DOM bugs to the "DOM" Component.

Note You need to log in before you can comment on or make changes to this bug.