Hi, Currently vresion of AbiWord exports documents with single-byte chartype (i.e. each character of the document's text is output as 1 byte). For proper i18n the correct way is using another chartype - that outputs 2 bytes (lsb first) of character's unicode value or to produce multybyte apporximation of unicode character (if its value is > 0x7f) as string (currently AbiWord uses printf pattern "&#x%x;" for writing - but Word2000 doesn't understand such representation at all). So, could you please answer to either of my questions (and quickly if possible since it's midnight begins here): 1) What is the correct printf pattern for representation of unicode characters when chartype is the one correspong to byte-per-character data? Does it exist at all? If yes, I would be very happy since only only line will need to be tweaked in AbiWord exporter, and you won't have to answer my 2nd question. 2) If there is no printf pattern for representation of unicode characters, how to mark current data as taking 2 bytes in output stream? I guess something in FIB structure should be modified after it was initialized, or in the stucture corresponging to current block. Could you please tell me what member of which structure and to which value should be changed after its initialization? I grepped over wv for two hours and didn't find the answer myself. I'm looking forward to your reply. Thank you in advance. Best regards, -Vlad
