Erik Abele <[EMAIL PROTECTED]> writes: >> That's basically the translator's decision. Though I'd suggest not to >> use >> them, because they blow up the xml source 2 - 4 times (or more). Note >> that >> it's only a *source* issue. Character references are resolved during >> the >> source parsing stage and are inserted as raw iso-2022-jp into the >> transformed >> result. I'm wondering a bit why you want to make such an effort. I can >> imagine (but I don't know) that's even more simple and efficient to >> type in >> the characters (or symbols, however) directly. > > I'm clearly with nd here; I can't see any reason why someone would want > to only use character entities in a XML source which is transformed > anyway and I would strongly suggest to *not* use them exclusively in > the error doc typemaps because of the increased size (they are not > transformed).
+1. The funny things may happen if we insert characters encoded in iso-2022-jp directly to typemap file but we are going to use URI keyword to refer to separate file for Japanese translation, I don't see any need to use character references. -- Yoshiki Hayashi --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
