THIS IS SO AWESOME I LOVE IT!!!
I made a couple small changes:
- I made various optimizations to try and make the search faster (reduced the size of the dictionaries, more aggressively reject obvious mismatches early).
- I manually tweaked the radicals to accept “similar” components interchangeably, so for instance 土 and 十 or 歹 and 夕 will match the same thing. So previously <口十>く wouldn’t find 吐く, now it does.
I hesitated to also consider 火 and 人 as identical but I’m not sure. I think in general 火 is fairly easy to tell apart.
I changed the URL to https://svkt.eu/kankan/ for technical reasons.
Where has this been all my life?!
Hello.
Can you add extra definitions (not in kanjidict)?
Two I know of are:
- 酛 (which is a kokuji, and has three jdict entries, but is not recognized by handwriting keyboards, and the decomposition into く酉元>
- 髎 (used in 肘髎、禾髎、巨髎 (they are names of acupuncture/指圧 points. A decomposition likeく骨羽人ミ〉I suppose) (Unicode also has 翏 and 㐱 that could be used as decomposition input)
This one already works: <酉元> (note that in your comment you used the kana く instead of <).
This one is correctly decomposed in the radical file:
髎 : ノ 个 冂 冖 冫 彡 月 羽 骨
The reason you can’t find it is because there are no entries in jmdict with this kanji, so there’s nothing to be found.
I think I will add kanji with no associated single-kanji vocab manually in the dictionary for this purpose.
I added “loose” kanji from kanjidic to the dictionary (at least those that have an associated meaning, I deemed the rest probably too rare).
Now <骨羽> does find 髎:
I only list the onyomi in katakana and I tag them with [漢字] to distinguish them from real vocab.
an incredible resource, thank you very much! I will use it often
this looks really useful, thank you!
I added a radical picker similar to jisho’s (although I decided to order radicals with the same number of strokes by how common they are which gives a different order, I don’t know if it’s a good idea).
I also tweaked the <> handling: now everything is rewritten to 「」immediately and also for easier typing on mobile (or even desktop) you can just type a comma and it will get rewritten to either 「 or 」 depending on whether there’s an open bracket before on the line.
So now 、列、えば turns into 「列」えば. No more pesky ‘!’.
I’m unable to use it on a mobile phone device (with Samsung browser), if I type real kanji it shows it, but when I try search by components (which is the nicety of your site) it doesn’t trigger the search function.
Could it be possible to add a button to be able to manually trigger it?
Thanks
Weird, it works with Firefox mobile. I’ll try with other mobile browsers. Does sending a newline help maybe?
EDIT: I managed to install the Samsung browser from the play store, I can reproduce the issue. Very odd.
EDIT2: seems to work fine on mobile Chrome too. Even odder.
Can you add to the first post the list of features? I’ve been following the thread but haven’t used it yet, by the time I get to it I think I will have forgotten everything
Edit: oops I can see that you’ve added it directly to the website, neat!
I need to rewrite that blob to be a little more to the point…
Ok I found the issue:
Brilliant.
EDIT: pushed a fix:
if (!Set.prototype.intersection) {
Set.prototype.intersection = function(otherSet) {
const out = new Set();
for (const elem of otherSet) {
if (this.has(elem)) {
out.add(elem);
}
}
return out;
};
}
if (!Set.prototype.difference) {
Set.prototype.difference = function(otherSet) {
const out = new Set();
for (const elem of this) {
if (!otherSet.has(elem)) {
out.add(elem);
}
}
return out;
};
}
It’s a little slower than the native implementation but at least it should work everywhere.
Hey there! First of all, thanks for this great utility!
I tried it out the first days after you published this post and it worked just fine, but starting a few days ago the website (https://svkt.eu/) doesn’t load at all, with the server responding with a 403 Forbidden
(sent by Nginx).
I thought it was a geographical issue (I live in Morocco), tried a U.S VPN but it doesn’t work there either. I use Firefox on Linux, but I’m not sure if that has anything to do with it.
I just noticed the URL change to https://svkt.eu/kankan/
, all good now!
Yeah sorry about that, I should have an index page instead of just nothing at the root. Too much work and too many book clubs…
Don’t have much to add. But I’ve still been using this as my main lookup tool while playing games whenever I come across something I don’t know. It’s a wonderful tool.
I actually don’t use it that much myself because I’m used to looking stuff up in takoboto and it’s better for looking up conjugated verbs etc…
I either need to ask them to integrate the feature or improve this tool to the point where I can make it my main dictionary.
Having some kind of draw mode would be good too, I’m sure there must be open source implementations but I wonder if they’re as good as the one takoboto uses (which I find really good).
I could also use a web worker to speed up the lookups. One reason I don’t use jisho as my native dictionary is that I can’t stand the loading times for every single lookup.