Version: unknown (using KDE 3.1.4) Installed from: (testing/unstable) Compiler: gcc version 3.3.2 (Debian) OS: Linux (i686) release 2.4.22 When I'm chatting in irc, sometimes a "xml parse error" appears in the chat window. Appearly, when a special characters appears for example : ï â é à è ¤ When I'm tapping a message, the others see Ä in place of é for example. A example (GML as kopete client and gmludo as KSirc) in kopete : An internal Kopete error occurred while parsing a message: XML document could not be parsed! An internal Kopete error occurred while parsing a message: XML document could not be parsed! [12:08:50] <GML>à [12:08:51] <GML>è [12:09:19] <GML>' [12:09:23] <GML># [12:09:33] <GML>_ [12:09:37] <GML>ï [12:09:48] - --FabGM has changed their nickname to FabGM-[Away] [12:09:55] <GML>& [12:09:59] <GML>€ In KSirc : [12:08:36] <gmludo> à [12:08:37] <gmludo> é [12:08:50] <GML> à [12:08:52] <GML> è [12:09:19] <GML> ' [12:09:23] <GML> # [12:09:33] <GML> _ [12:09:38] <GML> ï [12:09:48] FabGM is now known as FabGM-[Away] [12:09:55] <GML> & [12:09:59] <GML> â¬
Similar to Bug #67727, but not entirely the same. Kopete should sanitise the input before feeding libxslt.
*** This bug has been marked as a duplicate of 57129 ***
I'm using a german kopete 0.8 RC1 (KDE 3.2 RC1) but I'm not able to receive special chars 'cause of the same error message...also nobody is able to receive special chars....I hadn't this problem with 0.7.4...
This is how the problem can be reproduced: One needs kopete on _both_ communication sides, trying it with gaim vs kopete works quite fine (kopete can send and receive messages correctly in this combination). The problem occurs afaics only with german special characters. The sending the following short string leads to the problem "ü, ü" (both characters are german lowercase u with dots on it). Various other combinations of punctation and special characters cause the same problem. Just let me know your icq# and I can reproduce the problem on your workstation for debugging ;-)
Wait... The above bug says the problem is in IRC. Yet you just said it is with ICQ. Is it an ICQ or an IRC problem? This is a very important distinction, all the protocols use vastly different mechanisms for encoding and decoding strings. Also note please do not re-open unless you can reproduce it using CVS HEAD, CVS has many encoding fixes that will be in the next Kopete release.