Closed
Bug 73243
Opened 25 years ago
Closed 25 years ago
Folder default charset is broken for mail header
Categories
(MailNews Core :: Internationalization, defect)
Tracking
(Not tracked)
VERIFIED
FIXED
People
(Reporter: ji, Assigned: jgmyers)
Details
(Keywords: intl)
Attachments
(1 file)
|
926 bytes,
text/plain
|
Details |
***Observed with linux, win32, and Mac 03/21 mtrunk build***
The mails w/o MIME header don't use the folder charset now.
It used to be working even at 03/20 build.
Steps to reproduce:
1. Select a mail w/o MIME header.
2. Change the folder charset to the proper charset and reselect the folder. The
mail w/o MIME header still garbled header on the thread pane and envelope.
Comment 2•25 years ago
|
||
This is about folder default charset, changed the summary.
Reassign to jgmyers, his checked in for bug 65277 on 3/20 might be related.
John, is the default charset supposed to be working now, or you need more
changes to be checked in.
Assignee: nhotta → jgmyers
Keywords: intl
Summary: Charset override is broken for mail header → Folder default charset is broken for mail header
Comment 4•25 years ago
|
||
1051 static PRBool intl_is_legal_utf8(const char *input, unsigned len)
1052 {
1053 int c;
1054
1055 while (len) {
1056 c = (unsigned char)*input++;
1057 len--;
1058 if (c == 0x1B) break;
In case of ISO-2022-JP header with no MIME encoded, it has 0x1B and the function
returns true, then the caller just dup the string without using the default
charset.
Why do you check for 0x1B there?
Keywords: nsbeta1
| Assignee | ||
Comment 5•25 years ago
|
||
That was a transcription error when I pulled the UTF-8 detection code into a
separate routine. Fix:
Index: comi18n.cpp
===================================================================
RCS file: /cvsroot/mozilla/mailnews/mime/src/comi18n.cpp,v
retrieving revision 1.80
diff -u -r1.80 comi18n.cpp
--- comi18n.cpp 2001/03/20 00:47:41 1.80
+++ comi18n.cpp 2001/03/29 02:53:47
@@ -1055,7 +1055,7 @@
while (len) {
c = (unsigned char)*input++;
len--;
- if (c == 0x1B) break;
+ if (c == 0x1B) return PR_FALSE;
if ((c & 0x80) == 0) continue;
if ((c & 0xE0) == 0xC0) {
if (len < 1 || (*input & 0xC0) != 0x80 ||
Comment 6•25 years ago
|
||
r=nhotta
Comment 7•25 years ago
|
||
sr=sspitzer
| Assignee | ||
Comment 8•25 years ago
|
||
Fix checked in.
Status: NEW → RESOLVED
Closed: 25 years ago
Resolution: --- → FIXED
Verified with 04/02 build. It's fixed.
Status: RESOLVED → VERIFIED
Updated•21 years ago
|
Product: MailNews → Core
Updated•17 years ago
|
Product: Core → MailNews Core
You need to log in
before you can comment on or make changes to this bug.
Description
•