[webkit-reviews] review denied: [Bug 22116] Revise manual charset alias in TextCodecICU : [Attachment 25689] patch
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Wed Dec 3 01:22:50 PST 2008
Alexey Proskuryakov <ap at webkit.org> has denied Jungshik Shin
<jshin at chromium.org>'s request for review:
Bug 22116: Revise manual charset alias in TextCodecICU
https://bugs.webkit.org/show_bug.cgi?id=22116
Attachment 25689: patch
https://bugs.webkit.org/attachment.cgi?id=25689&action=review
------- Additional Comments from Alexey Proskuryakov <ap at webkit.org>
Thank you, I like this patch a lot!
> + - Uses windows-949 and windows-874 instead of windows-949-2000 and
windows-874
A typo, should be windows-874-2000.
The reason why I used windows-949-2000 is that windows-949 is ambiguous in ICU,
see <http://demo.icu-project.org/icu-bin/convexp?s=IANA&s=MIME&s=ALL>. Does
anything guarantee that the correct codec will be picked for windows-949 family
encodings with your change? I _think_ that windows-949-2000 is the correct one,
as in matching WinIE better, but I haven't verified that assumption. This is
not an issue for windows-874.
On a tangential note, HTML5 uses the windows-949 name:
<http://www.whatwg.org/specs/web-apps/current-work/#character-encoding-requirem
ents>.
> + - Maps 'dos-874' to 'windows-874'
Could you please mention the rationale, in this bug and/or in ChangeLog? Is
this done for compatibility with other browsers? Do both Firefox and IE support
this name? Do we have a reason to think that actual pages ever use it? I don't
have anything against this change, just want it fully documented.
> + - Maps 'EUC-CN' to 'GBK' and replaces it with 'GBK' when it's used
as the canonical name
Is this a change in behavior? According to ICU converter explorer, EUC-CN is
already mapped to GB2312, which we upgrade to GBK.
> + - Adds 'x-uhc' as an alias for 'windows-949'
Could you please mention the rationale?
r- because of the concern about ambiguity of windows-949, but please feel free
to mark for review again if it's actually OK for some reason that I missed.
More information about the webkit-reviews
mailing list