On Tue, 2003-09-30 at 05:11, Stephen J. Turnbull wrote: > > > Ville> A semi-unrelated question: is UTF-8 ok in .el files? > > No. The current standard (adopted in the days of Steve Baur) says > that .el files must be 100% ASCII to be under xemacs-packages or in > core, and Mule .els are ISO-2022-JP. Ville> Ok, thanks. I suppose that applies to all content therein, Ville> including comments? Yes. The problem is that since each file has only one coding system, it's quite possible to really confuse Mule with cut and paste. Unfortunately, our Mule is not very safe when confused; it has a habit of saving files in such a way that data is corrupted. In some cases, it would be something like ISO 8859 something -> ISO 2022, and the data would be recoverable, but non-ISO-2022 capable apps (including non-Mule XEmacs) would drop core etc. In others, it's even worse, random substitution of ASCII TILDE or Japanese GETA MARK for random characters not understood by Mule's current coding system. I'm working on the safety parts, but it's not easy, especially with the big changes Ben made between 21.4 and 21.5. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software.