#56 J-text character selection and lookup

[mowog]

It would be wonderful if you could add a major function of Dokusha 1.5:
open EUC-encoded text files, in which clicked kanji are looked up,
speedily. If you click on a kanji in the open file, Dokusha searches for
kanji combinations beginning with that character, which makes reading
Japanese much quicker and easier. This particular feature is just
absolutely superb for reading the electronic documents and email my
employer (a Japanese university) gives me!

On the Mac, JEdict has a similar function (you can cut and paste J-text
then click on a kanji to make JEdict seek for all combinations involving
the clicked kanji), but I prefer Dokusha's. It's less "helpful" in that it
doesn't reach to too many kanji; for example, if you want to know "giin"
in "kokkai-giin," you click on "gi". In Dokusha, you'll get "giin", but
JEdict will more likely try to look up "kokkai-giin" because it seems to
choose its lookup strings by scanning outward until it reachs a non-
kanji character.

Dokusha hasn't been updated in a while, and it tends to crash at its
startup on my Palm TX, whereas the latest version of PAdict has been
quite stable. If it could open Japanese-encoded texts too, I'd be even
more delighted!