How to Search via Kanji
If your browser is Japanese-friendly, you can simply cut-and-paste, or use
some kind of FEP to insert EUC-encoded kanji directly into the search text
field of the main dictionary page. If you use Japanese text directly, the
``Japanese Search'' and ``English Search'' selection doesn't matter, as
that only governs if the letters you type should be taken as romaji or
If you know the JIS code for the character, you can insert the character
by writing ``!
####'' in the search string. For example, the word
(``Japan'') is made up of the two characters whose JIS
4b5c. You could search for all entries
that contained `` '' by searching for
!467c!4b5c'' (without the quotes, of course). If you use the
default ``word starting with pattern'', you'll find words beginning with
. Follow this link
to see the main dictionary server with this example set up.
Note that when you insert Japanese text in this way, the ``Japanese
Search'' and ``English Search'' selection doesn't matter, as that only
governs if the letters you type should be taken as romaji or English.
However, if you use ``combo searches'', it does still matter for the
For example, searching for ``
!467c!4b5c AND NOT ENGLISH japan''
would find entries that began with , but didn't have
the English ``japan'' anywhere in them. If the ``Japanese/English''
selector were on ``English'', then you could omit the ``
Finding the JIS Code
Not too many people go around memorizing the entire JIS code, so I've added
a way to look up a character: the kanji lookup page. The head
of each resultant entry shows the character and the ``!'' code (here is a
sample result). Once you find
the codes of the characters you want, you can put them together to form a
Comments appreciated [Return to Change Log]
[Return to Main Page]
[Jump to Index]
(this page's master source last modified 15 years, 0 months ago)
This reply to request 115,720,479 made just for you Sat Nov 29th 2014 12:44am JST [load currently averaging 8690 requests/day over a 517-second sample]