> Probably another option would be to have seperate font files per > locales and pre-compute character width in those font files on a > platform with working wcwidth(). How does it sound?
Hmm. If at all, I suggest an extension to the groff font format to avoid zillions of files for locales... I suspect that the number of character widths affected by the locale are minor. Honestly, right now I'm not aware of a single one which *must* change. Could you provide examples? Regarding CJKV environments, it's a matter of the document style whether, say, the ASCII range gets treated as double-width characters, it's not a necessity. It really depends on the number of affected characters to decide which of the two suggested solution is preferable. > https://code.launchpad.net/~cjwatson/groff/charclass Thanks. I'm bad in working with bazaar. How can I get a diff file to the trunk? Werner
