Mailing List ArchiveSupport open source code!
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]tlug: More generic than Linux: X
- To: tlug@example.com
- Subject: tlug: More generic than Linux: X
- From: "Stephen J. Turnbull" <turnbull@example.com>
- Date: Mon, 25 Jan 1999 18:27:17 +0900 (JST)
- Content-Transfer-Encoding: 7bit
- Content-Type: text/plain; charset=us-ascii
- In-Reply-To: <19990125170028.A36113@example.com>
- References: <19990125170028.A36113@example.com>
- Reply-To: tlug@example.com
- Sender: owner-tlug@example.com
>>>>> "Scott" == Scott Perlman <perlman@example.com> writes: Scott> How in zee world does X negotiate codesets and fonts? X11 itself doesn't. There's zero magic in the X server. The app or a toolkit is responsible for squirting out strings that are correctly encoded. When you see a -*-JISX0208*-0 font requested, that means that X will display it correctly if and only if straight 7-bit JIS code (no escapes, no GR codes, please) of correct endianness is sent to the server. The standard toolkits (maybe it's as low as libXmu, I don't think it's libX11) provide X Compound Text, as does Motif. Motif also provides a more complex tagged Text structure. X Compound Text is a member of the ISO-2022 family and allows 8 bit codes, but the shifts are different from the EUC conventions. None of the standard Japanese coding systems are compatible with Compound Text (although EUC <-> Compound text is trivial, just add a few appropriate escape sequences). The tagged text structure is ugly and not accessible to the user; it's not relevant to your problem. So you must look at the application level. Kterms (and probably rxvts, exterms, and cxterms) can switch encodings on the fly using a Ctrl-MouseButton to access menus. Unless your LANG variable is Japanese (ja_JP* and possibly locally defined aliases), X?Emacs/Mule assumes X compound text, and your EUC and possibly Shift-JIS text will be mojibake. Mule is pretty good at guessing otherwise. Japanese less includes appropriate guessing magic, I believe. Try starting a kterm with `kterm -km euc &', then in that kterm `cat $file_containing_Japanese | nkf -e'. That should get you correct Japanese, regardless of your LANG setting(s). If it doesn't, you have big problems (or one of the rare files that nkf can't guess). I just realized that your distinction was GUI vs CLI. Does that mean that kterm (rxvt, whatever) works, but other programs don't? Which ones? Mule, Netscape, Japanized window manager, ...? What's not OK, text display or GUI button labels or both? Did it suddenly stop working, or did it never work? Scott> Can anyone give me a pointer as to where this is explained? Scott> (Free preferred, but a good book, is also very good) OReilly Associates, Xlib Programmer's Guide (Vol 1). Also Vol 0 (on the X protocol). For X. I don't think there's much relevant content in it but FWIW (haff, kaff, blush) http://turnbull.sk.tsukuba.ac.jp/Tools/I18N/LJ-I18N.html has a bunch of references. Unfortunately, for Japanese, you're pretty much SOL because every Japanese program does it a little bit differently. The lang_COUNTRY.Encoding style locale specification wasn't really intended for the mess we've got in Japan. There isn't much standardization for how to handle this stuff. If Ken Lunde's book (revised version) is on the shelves, that's the best bet. (Also OReilly, I think it's called Chinese, Japanese, Korean, Vietnamese Info Processing, commonly referred to as Son-of-UJIP.) UJIP goes into excruciating detail about Japanese codes, but doesn't help much with the operating environment. I don't know if CJKV has more of that. Many books about Japanese environments have cookbook approaches to setting your various variables, that may give a hint. None of them are much help on debugging if you've followed the recipe and it don't work, though. "Use the Source, Luke." -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 __________________________________________________________________________ __________________________________________________________________________ What are those two straight lines for? "Free software rules." ------------------------------------------------------------------- Next Technical Meeting: February 13 (Sat), 12:30 place: Temple Univ. ** presentation: XEmacs, by Steven Baur and Martin Buchholz Next Nomikai: March 19 (Fri), 19:30 Tengu TokyoEkiMae 03-3275-3691 ------------------------------------------------------------------- more info: http://tlug.linux.or.jp Sponsor: PHT
- Follow-Ups:
- tlug: Ken Lundes book
- From: Darren Cook <darren@example.com>
- References:
- tlug: More generic than Linux: X
- From: Scott Perlman <perlman@example.com>
Home | Main Index | Thread Index
- Prev by Date: tlug: More generic than Linux: X
- Next by Date: tlug: Ken Lundes book
- Prev by thread: tlug: More generic than Linux: X
- Next by thread: tlug: Ken Lundes book
- Index(es):
Home Page Mailing List Linux and Japan TLUG Members Links