> > - Joliet use Unicode by default. Every character is represented in > > UTF-16, so characters above U+FFFF can't be stored properly when > > done the windows way. > > Doesn't it use surrogate pairs? I thought that was "the windows way".
If the Joliet specification calls for UTF-16 (which, by the way, is used at many other places besides Windows; e.g. MacOS, Epoc, ...) then characters above U+FFFF should be represented as UTF-16 code pairs. If the Joliet specification calls for UCS-2, then you can still represent characters above U+FFFF as UTF-16 code pairs, since that does not interfere with UCS-2 per se (and the Joliet spec. should be updated). (B.t.w. wasn't Joliet something prior to the latest ISO spec. on CD file system formatting? Any up-to-date references? http://bmrc.berkeley.edu/people/chaffee/jolspec.html#unicode seems to specify UCS-2, and Unicode 1(!), which is really bad, for Korean in particular; I just hope I stumbled on a really out-of-date document. I should keep track of the CD formats, but haven't. Sorry.) Kind regards /kent k -- Linux-UTF8: i18n of Linux on all levels Archive: http://mail.nl.linux.org/linux-utf8/
