[Tickets #9842] Re: Search does not work
bugs at horde.org
bugs at horde.org
Wed Apr 13 05:12:59 UTC 2011
DO NOT REPLY TO THIS MESSAGE. THIS EMAIL ADDRESS IS NOT MONITORED.
Ticket URL: http://bugs.horde.org/ticket/9842
------------------------------------------------------------------------------
Ticket | 9842
Updated By | scott at spotman.net
Summary | Search does not work
Queue | IMP
Version | 5.0
Type | Bug
State | Feedback
Priority | 1. Low
Milestone |
Patch |
Owners | Jan Schneider, Michael Slusarz
------------------------------------------------------------------------------
scott at spotman.net (2011-04-13 05:12) wrote:
>> Thanks for looking at the log. Strangely, other clients, including
>> the previous version of horde (I think H3? over a year+ old) appear
>> to work fine with imap search. I am not sure if they are searching
>> in the same manner or not, but definitely performing successful imap
>> search against cyrus 2.4.7, but will see if I can get a log.
>
> Older versions of IMP (and other clients) are probably not sending
> extended search commands.
Ok, making some progress by spying on another client that its working
with, then duplicating results in telnet. here is a paste of my
telnet session:
Working: (no charset)
-------------------------------
. SEARCH RETURN (ALL COUNT) FROM horde
* ESEARCH (TAG ".") ALL
380,766,1051,1300:1301,1303,1927:1928,3102:3103,4880,4900,4921,4923,5722,5749,5757,6051:6052,22753,22778,22782,22788,22796:22799 COUNT
27
. OK Completed (27 msgs in 0.060 secs)
Working w/ charset:
----------------------------------
. SEARCH CHARSET UTF-8 RETURN (ALL COUNT) FROM horde
* ESEARCH (TAG ".") ALL
380,766,1051,1300:1301,1303,1927:1928,3102:3103,4880,4900,4921,4923,5722,5749,5757,6051:6052,22753,22778,22782,22788,22796:22799 COUNT
27
. OK Completed (27 msgs in 0.060 secs)
Broken in imp:
--------------------------------------------------
. SEARCH RETURN (ALL COUNT) CHARSET UTF-8 FROM horde
. BAD Invalid Search criteria
So it appears the order of the CHARSET option. Not sure if a cyrus
specific workaround is needed or if this how its supposed to be.
Every mention of charset in this paper shows it at the beginning:
http://www.faqs.org/rfcs/rfc5182.html
Thanks!
More information about the bugs
mailing list