openbsd-ports/japanese/README

45 lines
1.9 KiB
Plaintext

The japanese tools are somewhat ackward to use and difficult to setup
for the time being. Here is some useful information.
* japanese and locale
OpenBSD does not have any true japanese locale support for the time being.
Startup errors for kterm (`can't set locale for ja...') are quite normal.
Manual pages for, e.g., jvim do install under /usr/local/man/ja_JP.EUC/,
as they are written in Japanese.
For the time being, you will have to fix your /etc/man.conf to see them,
so that the _default setup reads:
_default /usr/{share,X11R6,X386,X11,X11R4,contrib,gnu,local}/{man,man/old,man/ja_JP_EUC}/
* is kterm working ?
Once kterm is built, the distribution holds an uuencoded file (DEMO.kt.uu)
that you should be able to cat after uudecoding.
Note that the choice of fonts is reduced when you need to display japanese
or corean characters.
* jless vs. less
Normally, jless should be highly compatible with less, to the point where
it doesn't display japanese before you set JLESSCHARSET in your
environment. iso8 is the sanest setting.
* the jvim puzzle
jvim depends on several pieces to work correctly:
- kterm for the display, jvim uses ONLY EUC mode,
- Wnn for the dictionary conversion,
- onew for the interface between Wnn and jvim.
as japanese includes thousands of characters, the only reasonable method
for inputting these is to use a dictionary: you enter your text
phonetically, then the automated dictionary makes a guess at the conversion,
and you confirm the right choice. Wnn is the dictionary server.
It needs to be started as root (this will probably be fixed in the future),
it is called /usr/local/bin/Wnn4/jserver.
To handle conversions, jvim adds another set of modes to the usual vim
modes.
ctrl-space, ctrl-@, or ctrl-\ is used to toggle from normal insert mode to
japanese inserts. If Wnn does not work, you can still enter
katakana/hiragana, but you will need Wnn to convert them to kanji.