Closed Bug 149724 Opened 23 years ago Closed 23 years ago

Auto-detect Chinese doesn't work

Categories

(Core :: Internationalization, defect)

Product:
Core
Component:
Internationalization
Platform:
x86
Windows XP
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
VERIFIED FIXED

People

(Reporter: amyy, Assigned: shanjian)

References

(
URL
)

Details

(Keywords: intl, regression, Whiteboard: [adt2 RTM] [ETA Needed])

Attachments

(1 file)

patch
1.75 KB, patch
ftang
: review+
jst
: superreview+
jud
: approval+
Details | Diff | Splinter Review
Build: 06-05 1.0.0 branch build on WinXP-SC. By load a non-meta charset Chinese page http://www.f10.com.cn, when set to auto-detect to Chinese is not working.
This is a regression from N6.2x, -> nsbeta1.
Keywords: intl, nsbeta1, regression
QA Contact: ruixu → ylong
I just noticed there is bug 149699 filed for mail.
Actually the regression is after N7.0PR1 - it works fine on N7.0PR1. Is this a regression by fix bug 132006?
->shanjian
Assignee: yokoyama → shanjian
*** Bug 149699 has been marked as a duplicate of this bug. ***
Attached patch patch — — Splinter Review
This is caused by the addition of gb18030. Big5 encoding covers gb2312, and frank used frequency analysis to pick the right one. His sampling code only handle euc like encoding. GB18030 is not one of them. When there is non-euc like candidate left, sampling will not be run. When this happens because of gb18030, I eliminated from considering list as what we did for ucs2. The patch works because, 1, gb18030 is rarely used. 2, when gb18030 is there, if we still have more than 2 cadidates, most likely gb2312 will be there as well. 3, If the encoding is gb18030, and gb2312 is not there, most likely we will end up with state ITSME for gb18030.
Status: NEW → ASSIGNED
frank, could you review?
taked with ftang, nsbeta1+
Keywords: nsbeta1 → nsbeta1+
should autodetect chinese work for 1.0.1?
Blocks: 143047
Whiteboard: [adt2 RTM] [ETA Needed]
it is turned on by default in the chinese build
Blocks: 141008
Comment on attachment 87113 [details] [diff] [review] patch r=ftang
Attachment #87113 - Flags: review+
Comment on attachment 87113 [details] [diff] [review] patch sr=jst
Attachment #87113 - Flags: superreview+
fix checked into trunk.
Status: ASSIGNED → RESOLVED
Closed: 23 years ago
Resolution: --- → FIXED
adt and drivers, please approve. we need this fix to ship good simplified chinese release. The sc release use this detector by default the fix only impact several converters which won't risk us build
Keywords: adt1.0.1, approval, mozilla1.0.1
Verified fixed on 06-26 trunk build / WinXP-SC.
Status: RESOLVED → VERIFIED
Keywords: adt1.0.1 → adt1.0.1+
adding adt1.0.1+. Please get drivers approval before checking into the branch.
Attachment #87113 - Flags: approval+
please checkin to the 1.0.1 branch. once there, remove the "mozilla1.0.1+" keyword and add the "fixed1.0.1" keyword.
Keywords: mozilla1.0.1 → mozilla1.0.1+
fix checked into branch.
Keywords: mozilla1.0.1+ → fixed1.0.1
Verified fixed on 06-28 1.0.0 branch build/WinXP-SC.
Keywords: fixed1.0.1 → verified1.0.1
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: