| Summary: | Pages encoded with "cp949" are garbled | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Myles C. Maxfield <mmaxfield> | ||||
| Component: | New Bugs | Assignee: | Myles C. Maxfield <mmaxfield> | ||||
| Status: | RESOLVED INVALID | ||||||
| Severity: | Normal | CC: | ap, dino, jonlee, simon.fraser, thorton, webkit-bug-importer | ||||
| Priority: | P2 | Keywords: | InRadar | ||||
| Version: | 528+ (Nightly build) | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Attachments: |
|
||||||
|
Description
Myles C. Maxfield
2015-04-27 14:52:44 PDT
Created attachment 251778 [details]
Patch
Comment on attachment 251778 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=251778&action=review > Source/WebCore/platform/text/TextCodecICU.cpp:178 > + // CFStringConvertEncodingToIANACharSetName(kCFStringEncodingDOSKorean) returns "cp949" instead of "windows-949" > + registrar("cp949", "windows-949"); This change adds support for this alias to web content, which looks like the wrong things to do. Firefox and Chrome don't support it (haven't checked IE), and it's not part of the new standard <https://encoding.spec.whatwg.org>. The only registered alias for cp949 is in Java, and it uses a different codec. What is the user observable issue that you are fixing? We probably need to fix it at a different level. There is already one workaround for this CFString behavior in defaultTextEncodingNameForSystemLanguage() function. Found the description in Radar, I think that we shouldn't do anything here. |