Re: utf-8 questions [was: Re: fontlock-speed]

Stef> yup, attached is the file. Ah, Markus's torture test. I've got that around somewhere. Stef> yup, but even in the gnome-terminal with the correct fonts Stef> it could not do it... Right. That's because there is no way to represent the missing characters _in the buffer_ in default Mule code. It could be faked, somebody who knows CCL could probably code it up in about 30 minutes. But there are only three people who know CCL and I haven't time to become #4 at the moment. :-þ So when you read the Unicode into the buffer, all of the unrepresentable characters turn into ? permanently as far as XEmacs is concerned. I doubt you'd get a better result in GNU Emacs, they operate under the same constraints with Mule that we do. You could try UTF-2000 XEmacs, http://www.m17n.org/utf-2000/ (Japanese, but there should be links to CVS and FTP in plain text). But it's a bloated pig (last I looked about 30MB binary) and not very stable. And it's based on XEmacs 21.2.35. Old.... -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Don't ask how you can "do" free software business; ask what your business can "do for" free software.