[JDEV] a problem with i18n in ICQ-t

safiullin at mail.ru safiullin at mail.ru
Mon Jul 9 13:12:01 CDT 2001


 there's an essential problem with i18n in ICQ-t. 

 When non-us/ascii icq-message goes through ICQ-t it's  en(de)coded 
 to(from) UTF8 with assumption of its encoding is windows-1252 (Latin1). 

 But there're other encodings used by ICQ-users worldwide (e.g. cp1251 for
 cyrillic), but there's no way for ICQ-t to determinate which one to use when 
 en(de)code to(from) UTF8, 'cause icq-message doesn't contain "Content-Type"
 header, I guess... ;)

 I can suggest the following  way to resolve that problem. It should be
 written in jabber.xml what charset is used with ICQ-t. And we'll be
 forced to have one publicly available jabber-server for one codepage ;)
 Or that option can be specified when user is registering with ICQ-t,
 if Jabber-transport API allows such tricks.

 Regards, Alex_


More information about the JDev mailing list