I sometimes when doing a review like to look at the kanji that make up a vocab, refresh my memory, etc, but it’s a bit tricky going backwards and forwards like this.
I made a small website that lets you search for a kanji, then it displays the radicals, kanji and vocab with a bunch of useful info, all in 1 zoomable/pannable flowchart type thing.
Not sure if it’ll be of any use to anyone!
See here: https://waniexplorer.pages.dev/
It looks pretty clean and straight to the point. I’ll give it a go, thanks for sharing.
This looks really good.
I will give some more feedback when I have used it “in anger”. (Meaning during actual study. Not sure if that is a generally known expression).
When I get things wrong I like to spend some time looking into what the confusion is, so doing this outside the main sesion could be good.
Never heard of this expression.
Programmers (in the UK) use it to mean “in practice, as opposed to in theory”
I’ve never heard this before but I think it’s brilliant. I am not a programmer but I absolutely have use for such a phrase!
Not just programmers, lol
It’s a phrase my dad used to use and I never fully understood. From UK, programmer, dad is not 
Some updates to this coming soon:
Any other features anyone would like to see?
Hey there! Thanks for the app, I think it has lots of potential. I was about to start developing something similar, but discovered this, so I guess I’d be reinventing the wheel.
My particular use case, though, is to be able to look up kanji by its radicals, using romaji. Many times I’m reading something and don’t know the kanji, but I know one or two of the radicals. Ideally the user would input one or more radicals by their name (ie, “ground”, “knife”) and the app would show all the kanji that use those radicals, if any. So far I’ve been making google searches, as in “wanikani ground knife”, but it’d be good to have something quicker and simpler.
An additional feature could be adding information to radicals to be able to break them down. Wanikani has many radicals that are actually made of several other radicals, but this information is not available in the API. An example would be the “old” radical, which consists of the “cross” and “mouth” radicals. But this would require more manual work.
I could add radical names as part of the kanjis search index, it uses AND, so if you searched ground knife, it’d show up all results that contained both those. It would only be in the search list though, not a specific visualisation showing all kanji next to each other. That wouldn’t be too much trouble but would require the UI to be more complex to be able to pick what kind of thing you’re trying to show etc. I’ll have a think though 
For splitting complex radicals in to their parts, as you say, this would be really manual, there are a ton of radicals that are made up of other radicals and I’d have to go through them all and work it out, which I’m probably not going to do But maybe it can be a group effort, the code is open source. github. It would need to be built in to the generation step (which makes requests to wanikani api to get all the subjects, transforms the data, outputs to local files the app uses to build the search index and the bigger documents that are displayed).
Just published a new version with the changes mentioned in previous message!
Yep, there wouldn’t be any need to show the actual data for the kanji, I think that showing them in the search bar would be more than enough! Let me know if you want me to implement that, I might have some time.
Regarding splitting the radicals, it’s something that I might be able to do with some time. If you know the compound radical name it would show it, and if you know the “simpler” radical it should show the same results too.
I’ve just pushed up a bunch more changes:
New url also: https://waniexplorer.pages.dev/ (I updated the other links, but in case you have saved it, the old one won’t be updated again)
This is neat．It’d be nice (for me at least) if one could limit the max WK level of results (so I could restrict it to material I’ve already studied and exclude material that I haven’t).
That’s a great idea, I’ll take a look
Ended up implementing what I needed in a scaled down version - basically radical-name-based search: GitHub - jmle/wkradicals
I’ll make it nicer and deploy it to GH pages in the coming days.
Edit: deployed to WK radical search
Looks handy 
When I get to that feature I’ll take a look at the code
I’ve also just pushed a bunch of updates.
