Hi all. Recently I was translating language strings from English to Chinese and found there is a problem in current language-translation mechanism of texmacs.
In dictionary_rep::translate function, texmacs splits string to be translated firstly by # (this is used in showing footer messages), and then by any characters other than a-z or space. But this mechanism is not so good for CJK languages, because CJK languages usually use different punctuation characters from Western languages, therefore punctuation characters should also be translated.
I think it would be better to split string to be translated only by #.