new online dictionary

[Vaste]

I would like some feedback on this.

The idea is to look up words based the parts their kanji contain, when copy-paste isn't possible (like anything printed). For example, say I want to lookup 警戒. I see that kanji #1 contains 言 and that kanji #2 contains 廾 and 戈. So I want to search for all words matching this. Query syntax: [言][廾戈]

3 matches for "[言][廾戈]":

• 警戒 【けいかい】: (common) (n,vs) (1) warning; admonition; (2) watch; lookout; vigilance
• 訓戒 【くんかい】: (common) (n,vs) warning; admonition; lesson or a phrase that teaches you not to do something
• 訓誡 【くんかい】: (n,vs) admonishment; warning

Now, am I the only one who've wanted such a dictionary?

(Thank God for URL obfuscation. )

 

[PaulTB]

Are you using kanjidic for the radical information?

 

[JimmySeal]

I really like that dictionary. Maybe the interface could be a bit prettier, but the search feature is awesome. Good work!

 

[Vaste]

Are you using kanjidic for the radical information?

Actually, no. When I wrote it, I expected to use kanjidic, but in the end I only use edict, enamdict, radkfile (here are the radicals) and wordfreq_ck. Possible future plans include adding support for specifying kanji using stroke count (or any other info in kanjidic really) -- then I'd use kanjidic. (Also, such support would be kind of easy to implement, I'm just having trouble coming up with a good and easy syntax for it.)