Hi I am working on a Manga reader for all platforms based off Suwayomi/Tachiyomi. It has instant ocr and 2 tap word lookups on mobile, 1 click on desktop GitHub - KolbyML/Mangatan: A 1 click solution for reading Manga with OCR'ed textboxs, works seamless with Yomitan the app also supports ocr and dictionaries for all languages as well. We have some English/Chinese/Korean learners, but the majority of our users are learning Japanese.
I added Anki support for creating flashcards already for Desktop/Android using AnkiConnect an extension which provides an API for anki.
For iOS I need to use callbacks which is fine, the problem is currently if users look up a word on iOS it will always show the “create flashcard button” which is unideal as creating a flashcard already made then alerting the user it is a duplicate is extremely tedious and bad UX.
So my request is if we could add a callback where I could specify
- the deck
- the notetype
- the field
Then the callback would give me a list of target words/(aka the selected field) which I could use for the duplicate check locally on my end. That way I can show the user if they already have a flashcard for that word or not.
Assuming the callback returns utf 8 and the google says the average japanese word is 2.8 moji, lets use a basic seperator like , 40000 (an assumed worse case, 40k words in japanese is insanely high) target words * 2.8 average moji + 40000 comma’s = 152000 charecters google says ```
For 152,000 Japanese characters in UTF-8 encoding, the size can range from 456,000 to 608,000 bytes, depending on the specific characters used
```
Which I think is more then reasonable for a call which will not be used all the time for syncing the users mined words for duplicate checks.
This feature would increase the UX for making flashcards on iOS immensily and would be greatly appreciated. It would allow for the addNote callback only being called on actually nondupe words, which would greatly decrease the fatigue that builds from mining cards on iOS you have already mined.
Anyways all I can do from here is hope
I am a big fan of Anki and want to provide the best experience possible for my users. I think this addition to the callback API would make a night and day difference.
Thank you for your time