I have used Charles proxy in the past to test third party APIs when I have done iOS and android development. You can proxy your requests through Charles (or similar) and intercept the server response and manipulate it before sending it on. I would use it to spike out an idea before investing too much time. It is worth looking into if you haven’t already.
That’s a good shout, shouldn’t be too hard, as you say, to modify the reply, change the subject type, and see how it handles it.
Just tried it out and all seems to be working! (Android 13 mobile) Thanks for your work on this, it’s greatly appreciated!
If you get a chance at some point, I have a small formatting request for the app - on the dashboard the SRS counters are on two lines (apprentice + guru + master, and enlightened + burned). It would be great if the two on the second line were 1.5 times the width of the ones on the first line so as to fill the same width, rather than leaving an awkward gap there. Not very urgent or important but would be nice
On my Android tablet, those SRS counters fit onto a single line, so the equal widths look good.
It’ll be a learn as I go experience with all this, I’ve got some coding experience, but no app experience, hehe. I won’t make anymore changes until I’ve confirmed that the kana support works, at which point I’ll be confident enough to create a thread for it and let all the Flaming Durtle users know there’s something to migrate to ^^
At THAT point, I’ll start taking some feedback for stuff people have been wanting fixed, but it might be a slow process as I get accustomed, haha
And this is where things get complicated, hehe - it’s very hard to account for different screens. I don’t know enough about android layout design to know if it’s possible to have it dynamically adjust based on screen width, but it’s something I’d look at, should it be possible. Again, on the basis of when I’m able to, hehe
I have quite a bit of Android development experience, but very little experience with respect to getting something hosted on Google Play. The one time I went through the process I told myself to avoid it again at all costs. DM me if you have any questions about software or build issues. I have a vested interest in seeing Flaming Durtles continue to burn.
The most amazing aspect of Flaming Durtles is the ability to do reviews offline and then have it sync when you are back online. Not possible via the mobile web experience alone. Everything else is sugar on top of that fundamental feature you can only get with a dedicated app.
I don’t understand why they can’t just make a separate thing from WaniKani and call it something else and that’s where the kana only vocabulary goes. Like to me WaniKani is for learning Kanji and vocabulary that contain those kanji within them.
Adding more and more kana vocabulary would only slow down my process as well.
Yeah that’s what I was thinking too! If they had a separate website for kana only vocabulary I think that would be amazing! Trying to smush kana only vocabulary into WaniKani a platform known for learning kanji and vocabulary that contains those kanji doesn’t make any sense.
I’m happy to say that I’ve been able to work out how to spoof the API responses and so have managed to get the sample kana data confirmed working in the beta version of Smoulding Durtles.
It shows up as a radical right now, but this is purely just because I’ve managed to somehow miss a couple of UI elements. Not sure how, given how comprehensive of a search I did for any permutation of vocabulary or radical, but que sera.
Correctly showing up with meaning, mnemonics, context sentences, etc. and all the other data the API provides:
Once I’ve resolved for the formatting issue I’ll post an updated beta to the github and create a thread for the app - hopefully Scott will get back to me regarding some testing access to I can run it agains the real API, but this at least confirms that the app definitely accepts the data without erroring out
For you maybe, but personally I only started using it because I wanted undo, reorder and other QoL features and I felt like using scripts was too fragile and likely to break.
If WK supported reorder and undo out of the box I don’t think I’d use any third party program. Although I must say that I would be sad to lose Flaming Durtle’s search function which is massively better than the website’s.
Out of curiosity do you know how stock Flaming Durtles would handle these entries? Does it break completely, ignore them or something else?
This honestly feels like a ‘Make people stay here longer/spend more money’ idea as it will now take users longer to learn kanji with kana words now being added. What I don’t understand is why can’t you guys make a separate website that teaches kana vocabulary only so then those will are interested would join there. There you can have not only a WaniKani subscription but then a secondary website just for kana only and have a subscription there.
You and the rest of the WaniKani team not knowing whether this is will make third party apps like flaming durtles and tsurukame usable anymore is not a good sign. User scripts and third party apps are the only reason people use WaniKani. WaniKani just isn’t usable on it’s own which says a lot right there.
Looking through the code, I’m not sure. The database is pretty well structured, but I know that:
- An item with an unknown subject_type would show as a WANIKANI_RADICAL
- Radicals and Kana both share the same assignment types
- Radicals have stroke data and kana do not
Best case, they’d work, but just show up as radicals (which is largely what is happening right now, except that they’re getting correctly filed in the database as kana_vocabulary, not radicals, or they’d not show up at all.
The code contradicts itself in how it defaults out, so radicals are the default fallback for subject_types, but vocabulary is the default fallback for all formatting, so it’s hard to tell.
Ok, I guess it’s lucky if they’re treated as radicals since they’re effectively handled the same way (with meaning but no reading).
If tsurukame (third part iOS app) stops working for me I’ll just stop learning on WK. I’ll only do the burned items and reviews here while continuing my learning journey on MaruMori.io
I definitely don’t want them to make a separate website. I like the idea of having kana only stuff on wanikani so I don’t have to juggle between different websites, which is what I do now. I understand that some people don’t want kana only vocabulary, so maybe an opt out in the settings is indeed the best solution to keep everybody happy. But definitely not more website or additional decks or anything like that. It would compromise the beautiful simplicity of the platform.
I did worry for a second that my beta version is just falling back to radical and I put in all this work for nothing, but nope, it’s in the DB correctly:
This means I’m fairly confident all the features will work as expected and I should be able to fix the card type soon, because I think it’s this section is at fault, the red being the new line I’ve just added in:
I feel it would be better if they make it a separate thing from WK like a separate website or separate section that does not mix in with the kanji reviews. Like you said this website for years has been known for learning kanji so that’s what I’ve come here to do. Learn kanji. Kana vocabulary on the side or different website is okay, but forcing it in with the kanji and kanji vocabulary just makes no sense to me.
Maybe the third party apps would be kind enough in making a separate kana vocabulary track and leave the kanji track online. They have quite a bit of kanji that they could add so I wish they focused on filling that gap first and then do the kana vocabulary, but separately from the kanji section.