Closed Bug 102758 Opened 24 years ago Closed 23 years ago

Do not support surrogate correctly in NCR code

Categories

(Core :: Internationalization, defect)

defect
Not set
normal

Tracking

()

VERIFIED WORKSFORME
mozilla0.9.9

People

(Reporter: ftang, Assigned: tetsuroy)

References

Details

(Keywords: intl)

Attachments

(1 obsolete file)

we use 16 to handle NCR, so we won't be able to encode correctly encoded NCR between 0x10000 and 0x10ffff test page http://warp/u/ftang/utf8test/ncr.cgi?p=0x1&r=0x0&c=0x0&face=MS+Sans+Serif should not show any ASCII to us, but we do it right now.
Status: NEW → ASSIGNED
OS: Mac System 8.5 → All
Hardware: Macintosh → All
Target Milestone: --- → mozilla1.0
Keywords: intl
QA Contact: andreasb → ylong
Keywords: mozilla1.0
yokoyama, can you work on it once shanjian add the surrogate support for WinXP?
Assignee: ftang → yokoyama
Status: ASSIGNED → NEW
Target Milestone: mozilla1.0 → mozilla0.9.9
accepting and making 102557 as a depend
Status: NEW → ASSIGNED
Depends on: 102557
we need to cover both html and xml. the code to handle xml may differ from the code handle html. looks like the code for html is in nsHTMLToken.cpp in mozila/htmlparser/src, not sure about xml.
Attached patch patch, need testing. (obsolete) — Splinter Review
Attachment #64346 - Attachment is obsolete: true
Comment on attachment 64346 [details] [diff] [review] patch, need testing. ignore this , it does not work
forget about my comment. I am not sure where should we change. Someone need to set a debugger and trace it back from layout. smontagu, harishd- do you know where so we turn A into 'A' ?
It's done by the expat parser ( mozilla/expat ). Heikki, is that right?
I believe harishd is right, haven't looked or tested but I can't think of how else it could be done.
shanjian: your comment in 102557 >No, it isn't working yet. We have problem in parsing NCR value larger than >0x10000. I have a fix sitting in my local tree. Where is at with the fix? Did it check in?
Yes, so this bug is resolved.
On the test page I see the correct characters where my font (Code2001) has them, but for invalid characters I see ?? instead of ? Is there an open bug on that problem?
frank filed one for that problem.
thanks, shanjian
Status: ASSIGNED → RESOLVED
Closed: 23 years ago
Resolution: --- → WORKSFORME
Verified on 02-11 trunk build on linux and winME, Frank's test page is displayed as "??" now. Mark it as verified.
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: