File: Jisho0.1.zip 
Author: Pascal Goguey (pascal@sakura.email.ne.jp) 
Release: 0.1    January 8th, 1999
Compatibility: R4 PPC, Intel
Location: Utilities/misc
Description: 

Jisho is a Japanese -> English dictionary.
Well, I should rather say that it is an interface to an existing
Japanese->English file. It is a quick & dirty implementation that
I wrote because I needed a quick lookup. In fact, it does not more
than a quick lookup. Although it is an J->E dictionary, it is possible
to use it as a E->J translator with the option "contains".

The data file:
Jisho's data file is named Edict (English Dictionary). It is the result
of a project conducted by Jim Breen, Monash U, Australia. Please
read edict.doc at least once. The file is not written in a "natural"
dictionary sequence. A regular japanese dictionary is sorted by
syllablials (a, i, u, e, o, followed by their combinations with the
consonnants in the following order k, g, s, z, t, d, n, h, b, p, m, y, r. 
Basically Edict is not sorted. However, as it is easier to use a sorted
file (e.g. lookup next entry or previous entry), I have sorted the
UTF-converted edict and put back its header at the beginning.

If you want to have an up to date file, please look once in a while
at ftp.monash.edu.au/pub/nihongo/edict.zip. The file edict.zip is
about 1 MBytes. It is updated a couple of times a year. In the early
days (around 92-93, there were 2 or 3 updates every months), but
now it's a lot less than that.


How to use it:
1. Font
Before you use it: make sure you have a japanese font installed. In
the whole program, I didn't care about the font at all except for
the entry counter, and I don't know why it works, but for sure,
you should have one japanese font installed.

2. Location
As it is a first implementation (a couple of hours of codng only),
it's not so smart. Tou have to leave all in the same directory exactly
as decompressed. You can put the folder anywhere you want, but
you should keep the dictionary with the exec code. When you run
it the first time, it will create an index table This should take a
couple of seconds.

3. Run:
- Double-click (what a surprise). Jisho will appear immediately or
after a few seconds if it has to create its table of contents.
- Enter something in the caption;
- Press return. The translation will be displayed in the text area.
	Then, if you want to find subsequent matches, press the magnify
(search) button. Once you found what uou were looking for, you can
also use the arrow buttons to look around that entry.
File menu: it is a fake menu (see paragraph bugs).
Options:
There are 4 options to search for an entry:
	- Exact match
	- Begin with
	- Ends with (don't ask me why this one!!!)
	- Contains.
By using it, I noticed that the option "contains" is the most useful one.
But Begin with or Exact match are faster.

4. Edict update:
- Download a new version of edict (ftp.monash.edu.au/pub/nihongo/edict.zip)
- Download also the edict.doc and read it once foe the license stuff.
- Decompress and convert it to UTF-8. I used to have a converter, but I can't
  find it anywhere. jconv, if I remember well.
- Remove the header, sort the file, and put the header back.
- Delete your old edict.toc
- Run Jisho. The new edict.toc will be created, and that's it.

5 Bugs:
Not so many, but it does not so much... However:
- When Raphael Moll tried this software, he had a CRC error when
decompressing edict.utf.zip. This time, it is not exactly the same file,
so it may not happen. If it does happen, forget edict do what is
explained at paragraph 4 (edict update).
- The menu "file->Create user file" is a fake menu. It might disappear or
might be implemented in a later version to allow the user to create his
own file in the same format.