Handling of [, ] and / characters.

[Jim Breen <jimbreen@*********>]

Those of you who are familiar with the EDICT format(s) will know the
basic "漢字 [かんじ] /Kanji/" structure that that I inherited from Mark
Edwards' old (DOS) MOKE ワープロ back in 1990, and have been more
or less stuck with ever since. It has meant all sort of problems with using
the [, ] and / characters as legacy software can get confused when they
are used for other purposes. Of course they are not a problem in the
dictionary database itself, nor in JMdict.

What we are doing now during the generation of the EDICT versions is:
(a) converting [...]  to {...};
(b) converting /  to -.

This means that there is freedom to use those characters within
glosses, etc. in JMdict, but people should be aware that they will
display differently is some applications. Some cases, e.g. W/kg
becoming W-kg, will look a little odd.

Cheers

Jim

-- 
Jim Breen
Adjunct Snr Research Fellow, Japanese Studies Centre, Monash University