application specifies the encoding or we will try to extract it from the request object (sent by the browser) or use Pythons default

About extracting from request: After playing with Preferences/Languages in Mozilla, I found out: If Mozilla is set to ISO-8859-2: HTTP_ACCEPT_CHARSET='ISO-8859-2, utf-8;q=0.66, *;q=0.66' If Mozilla is set to Win-1250: HTTP_ACCEPT_CHARSET='windows-1250, utf-8;q=0.66, *;q=0.66' So yes, you should be able to extract this from the browser.

All this applies to TXNG 1.1 but *not* 1.0. 1.0 will not handle unicode very well. I know about the issues but I will not solve them for 1.0. 1.1 will fully support unicode ;-)

Thank you for explaining this. In short, it looks that TXNG 1.1 will be able to do everything I asked in my previous email. -- Milos Prudek