Konstantin Stefanov Reporter
	Description • 22 years ago

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.3) Gecko/20030312
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.3) Gecko/20030312

I compose a new letter with only latin characters in the body (my default
Character Coding is KOI8-R), then save it as draft. Then I click to that letter
and choose the option "Edit as new". It is opened with coding Western
(ISO-8859-1) Then I add text containing russian characters. When I try to send
the letter, I get the confirmation window which says that there are characters
in the letter that are not in the current character coding, and asks me to
return to composer to change character coding. If I press "OK" to send the
letter anyway, all russian characters are changed to question marks.

Reproducible: Always

Steps to Reproduce:
1. Compose new letter with only latin characters in the body.
2. Save it as draft
3. Select it and choose "edit as new"
4. Add some cyrillic symbols to the body (I suspect that any national symbols
that are not in ISO-8859-1 willdo, but I used cyrillic).
5. Press "Send".
Actual Results:  
You will get a confirmation window.
If you press "OK", the letter will be sent with all cyrillic characters changed
to "?".

Expected Results:  
There is no confirmation window and Mozilla silently changes encoding to the
default (in my case - KOI8-R). If there are characters only from that coding,
the letter is correctly sent with correct encoding.

	Frederic Bezies
	Comment 1 • 22 years ago

*** Bug 197346 has been marked as a duplicate of this bug. ***

	Daniel Wang
	Comment 2 • 22 years ago

> I compose a new letter with only latin characters in the body...
> then save it as draft. Then I [open the msg and it] is opened
> with coding Western (ISO-8859-1)...

that's bug 169761

	Mike Cowperthwaite
	Comment 3 • 21 years ago

(In reply to comment #2)
> > I compose a new letter with only latin characters in the body...
> > then save it as draft. Then I [open the msg and it] is opened
> > with coding Western (ISO-8859-1)...
> 
> that's bug 169761

That's true, it is.  In bug 169761 comment 13, I describe a workaround for this 
problem.

Also: The silent conversion to the ?s, when hitting OK, was expected behavior 
prior to 1.6.  With 1.6, the dialog has changed such that OK means "convert to 
Unicode (UTF-8)"; however, testing this, I'm not seeing that work (which is odd, 
I thought I had seen it work in the past).  With 1.7b-0324, however, the 
automatic conversion appears to work just fine.

*** This bug has been marked as a duplicate of 169761 ***

	Jungshik Shin
	Comment 4 • 21 years ago

(In reply to comment #3)
>  With 1.6, the dialog has changed such that OK means "convert to 
> Unicode (UTF-8)"; however, testing this, I'm not seeing that work (which is odd, 

  It was broken in 1.6, but was fixed during 1.7 alpha cycle.