WaniKani to Anki Exporter - http://wanikanitoanki.com

Does anyone have any idea about my problem? I am assuming it has something to do with “mapping” but honestly I just don’t get it :frowning:

I hadn’t visited this thread in a long time, so sorry about the non-responses over the past half year.

@korinkite: I agree with your assessment. I mainly designed this for myself mostly to deal with the critical items, but I added the rest more of a convenience in case others wanted to use it as well. Having options is a good thing :slight_smile:

itoen90: Sorry I hadn’t seen you running into problems before. Can you shoot me an email (dennis@dennmart.com) with some more information to see if I can help you out, or figure out if there’s a bug somewhere that I need to fix?

Can you post the code on github with appropriate open licences please?
That way anyone can pick this up and carry it on if you decide not to continue.

I responded to this via email already, but in case anyone else is curious, the code for this site has been open-sourced for a few months now at https://github.com/dennmart/wanikani-to-anki. Pull requests welcome!

this is great, now i have an easy way to review my burned items on android. Thanks a bunch!

yagmot said...One issue: downloading vocabulary for a specific level often includes vocabulary not seen until the next level.
Can you provide a concrete example?
The vocab are from the wanikani api https://www.wanikani.com/api#vocabulary-list and this only returns the vocab for the levels specified.

Question: is there a URL scheme we can use to download a specific file? (maybe this is possible with the WK API, I haven't looked into it yet.) Basically I'm looking to build a script that will download the radicals, kanji, and vocab files each time reach a new level.
If you mean the API data then that is what the wanikani api provides. See https://www.wanikani.com/api
This only provides the raw data, it does not provide the details that you find on the web pages.
There is no way to obtain those web pages without your username/password for WaniKani, the API_KEY is not enough.

The way I've seen other people do it is via a script you load into your browser (mostly Chrome extensions).
See /t/Is-WaniKani-not-hard-enough-for-you-listening-comprehension/1148/1

It uses jquery to locate the audio on the page and store it into Chrome local storage.

Whenever I try to import, not all of my current radicals/kanji/vocab are making it into anki.  Also radicals are ending up in my kanji deck and vice versa.  Am I doing something incorrectly?

Hold off for a day or so and my pull request should land on the application with instructions on how to create an Anki deck for WaniKani with pictures!

The changes also mean all the exports have the same format, so they can all go into the one deck if you want.

Since I’m only focusing on “critical items” and maybe my current levels radicals and kanji so I can get some additional repetition in the apprentice stage, I’m not sure what you are trying to do.

But the import process lets you choose which Deck you want to import into.
The actual import file knows nothing about your Anki Decks, so if they are going into the wrong deck that is something you are doing wrong in Anki.

I was scrolling through the API and Third-Party Apps tag, thinking there’s no way I could find additional tools created by awesome community members like you, since there are already so many great ones. I thought, “do I even have any wanikani needs that haven’t yet been fulfilled?” and then I stumbled upon this post… I haven’t tried it yet, but I know for a fact that it’s going to be incredibly useful. Thank you so much! You’re awesome!

Dennmart has  upload my changes, try it out.
I’d be grateful for people to try the Docs and setup the Anki decks to make sure they are undertandable and give you the Cards you need.

Hey, I know this thread is a little old, but I’m looking for sentences to study that contain my want-kani level kanji. Would it be possible to add a “sentences” download that pulls the sentences from the lessons?

The short answer is “no”.
The vocabulary api doesn’t have that option, so the only way to obtain it is to scrape WaniKani itself.
Which means you need to be logged into the site and running a third party script to do the scraping.

I don’t know if someone has already written something like that.
It’s a bit like scaping the audio - and Viet has closed threads discussing that.
I think they don’t mind if you doing it, but allowing it on mass seems a no-no.

Ah, I see. Ok, I’ll keep looking around. Maybe there are enough sentences on textfugu with the same vocab to be helpful. Thanks!

I’m using Anki on Android, and it only imports .akpg format. Is there a way to convert the text files to that format? Or is there an option in Ankidroid that I haven’t seen that can import text format? LOL, thanks for this, and Ai & Heiwa!

Don’t know. Not easily I think is the answer.

https://ankidroid.org/docs/manual.html#importing says you are stuck with .apkg formats.
Your options are
1) Import via the Desktop and sync to ankiweb (that’s what I do)
2) Borrow a Computer, Install Anki, follow the docs and then export the deck as .apkg, somehow transfer it to the phone (via email?) and then import it.

Since 2 looks like a lot of work, I’d stick with 1.

I have used the downloaded vocabulary file to programmatically find all the verbs (based on a heuristic filter on meaning), check the godan/ichidan status (based on the reading and on a list of eru/iru godan verbs), and dump the -masu and -te form.
1: Would this be a nice add-on? then I can submit the code.
2: The WK vocabulary pages tell you if a word is godan or ichidan, but this information was not in the download. Is that because the API does not provide that information?

 Anyway, I greatly enjoyed playing around with the data!

You can see everything the api provides here https://www.wanikani.com/api
Its fairly limited, most of the good stuff is locked up on wanikani :slight_smile:

Have you seen whether this is already available from Jim Breen’s EDICT files http://www.edrdg.org/jmdict/edict.html, the documentation link has a link to the DTD. I dont know what the file actually contains but since places like jisho are based on EDICT and that have noun, verbs, etc that might already be done.

I haven't read through the whole thing, but when I import it to anki all the backs of the cards say vocabulary and not a definition or pronucation or anything. How do I fix this?

Nvm I figured it out. Thank you.

kelth said... Nvm I figured it out. Thank you.
Can you explain what you did wrong, and then what you did to fix it.
Otherwise you aren't leaving useful information to future people who will read this thread and potentially have the same problem.