Dear all, I've just finished updating the ISO 2022 utility, which I've renamed to ``luit''. The new version is available from ftp://ftp.dcs.ed.ac.uk/pub/jec/programs/luit-0.2.tar.gz Many thanks to Tomohiro Kubota for helping me tune the design, and Marc Espi� and Sylvain Baro for help with portability to BSD. This version is almost usable. It builds using Imake, and has a sketchy manual page. The main changes are: - terminal input is now converted (fontenc has been extended to reverse encodings on demand); - window sizes are properly communicated to the child (including when the window changes size at runtime); - compiles and runs on both Linux (glibc) and BSD systems; - initial ISO 2022 assignments are controllable from the command line; charsets are invoked by name, not by ISO 2022 sequence, which makes it possible to use charsets with no ISO 2375 assignment; - by default, ISO 2022 escape sequences are not interpreted; the command-line option `-i' may be used to request that ISO 2022 should be interpreted. The behaviour of fontenc has changed somewhat. The encoding searching function now accepts NULL as the location of the encodings directory, in which case SYSTEM_FONT_ENCODINGS_DIRECTORY is searched; this defaults to $(PROJECTROOT)/lib/X11/fonts/encodings/encodings.dir Suggestions for different locations are welcome. While the program was written to be extensible, this extensibility is not yet visible to the user. A future version will be able to be taught mappings between fontenc and ISO 2022 a plain text file. The same data file will contain mappings between locale names and initial ISO 2022 states. Other than that, I need to implement support for non-ISO 2022 encodings (easy to do for reasonably civilised encodings, such as Microsoft codepages or UTF-8, more tricky for horrors such as Big-5 or UTF-16). Regards, Juliusz - Linux-UTF8: i18n of Linux on all levels Archive: http://mail.nl.linux.org/lists/
