Closed Bug 464899 Opened 17 years ago Closed 17 years ago

Search for emails by Cyrilllic pattern in IMAP folder always returns an empty match

Categories

(Thunderbird :: General, defect)

x86
FreeBSD
defect
Not set
major

Tracking

(Not tracked)

RESOLVED DUPLICATE of bug 404255

People

(Reporter: yuri, Unassigned)

Details

User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.9.0.3) Gecko/2008100613 Firefox/3.0.3 Build Identifier: version 2.0.0.17 (20081001) I have some messages with Cyrillic subjects. But when I try to search by subject using some words contained in the subjects of these messages I get an empty result. Reproducible: Always Steps to Reproduce: 1. 2. 3.
I sent a test email to myself with йцуке in the subject, using the nightly TB 3, as well as 2.0.0.17 and search found it just fine. So this WFM.
Sorry, forgot to mention an essential info: this happens for the search in IMAP folder. Search works fine in local folders.
Summary: Search for emails by Cyrilllic pattern always returns an empty match → Search for emails by Cyrilllic pattern in IMAP folder always returns an empty match
Have look into Bug 404255 which is somewhat related
Right, I can confirm that IMAP search fails until the charset is set to UTF8, and then it succeeds. I think we should dup this to 404225. I also think that 404225 should be either invalid or won't fix, since this is a setup issue, but I'm not really the expert in this area.
Search should work out-of-the-box. User shouldn't know anything about UTF8 for such a simple thing like search. I wish the words like UTF never come up in any messages to user.
Interesting how does it do search if charset isn't UTF8, byte for byte? Obviously it should first convert both texts to the common charset (most logically UTF8) and then compare. So this isn't a setup issue, even though setup will fix it somewhat/sometimes. This is a bug in the way how search is performed. And probably this PR should be a duplicate of PR#404255 which should be fixed.
(In reply to comment #6) > Interesting how does it do search if charset isn't UTF8, byte for byte? The search is done on the server. TB transmits a character set field to the server. What the server does is not TB's problem. > > So this isn't a setup issue, even though setup will fix it somewhat/sometimes. > > This is a bug in the way how search is performed. > As far as I can tell, it works correctly once the character set is set. Is that correct? So would it be fair to say that this bug is now a request to automatically detect the character set, rather than force it to be set by the user?
> As far as I can tell, it works correctly once the character set is set. Is that > correct? Yes, it seems to depend on the charset setting. > So would it be fair to say that this bug is now a request to automatically > detect the character set, rather than force it to be set by the user? I believe so. Looks like it's a pitfall with IMAP protocol that TB should avoid by detecting a charset.
No longer depends on: 404255
Status: UNCONFIRMED → RESOLVED
Closed: 17 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.