From: SourceForge.net <no...@so...> - 2005-02-02 00:05:25
|
Patches item #1112296, was opened at 2005-01-29 17:16 Message generated for change (Comment added) made by thekingant You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300235&aid=1112296&group_id=235 Category: i18n Group: None >Status: Closed >Resolution: Accepted Priority: 5 Submitted By: Magnus Hult (magnushult) Assigned to: Mark Doliner (thekingant) Summary: Patch for [1063680]: Blank ICQ offline messages from Gaim Initial Comment: It seems as if all offline Unicode/UCS-2BE encoded ICQ messages sent from Gaim reach their recipients all empty (at least in ICQ Pro 2003b, Trillian, Licq 1.3.0, Sim 0.9.3, Gaim 1.1.2, probably more). I've done some packet sniffing and it seems that Gaim sends the message and also that the messages reach their destinations. But not a single client (not even Gaim) seems to be able to decode these messages. This patch makes Gaim prefer to use the sender's default encoding rather than Unicode if the recipient is offline. ---------------------------------------------------------------------- >Comment By: Mark Doliner (thekingant) Date: 2005-02-01 19:05 Message: Logged In: YES user_id=20979 I made this change for oldstatus and HEAD. My change is slightly different... I think your patch might leak msg in the case where the the ICQ user is offline. Thanks ---------------------------------------------------------------------- Comment By: Magnus Hult (magnushult) Date: 2005-01-31 06:51 Message: Logged In: YES user_id=829621 After some more research, it seems as if all encoding information is lost in offline icq messages. Therefore, sending unicode/utf-16be offline is probably not very good. It leads to all sorts of ugly errors. A utf-16be encoded "z" (00 7a) (or any other latin character) sent as an offline message is interpreted as just a zero termination (00) in the recipient's end, while for example "<water (Chinese)>z" (6c 34 00 7a) is intepreted as the ascii string "l4" (6c 34, zero terminated). So please do accept this patch. :) ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300235&aid=1112296&group_id=235 |