Very minor. There were a few returning keys that were incorrect (page instead of pg, for example), and they were aligned to the api specs. In fact, I think page>pg was the only change.
Also, the definitions that list languages are lined up so English is always "eng" (before, it was sometimes en, sometimes eng).
The rest were type changes (mainly, integers getting returned as strings, or integers being returned correctly, but they are now integers in both the specs and the results.
I’ve been building a Chrome / Firefox extension called YuzuLingo for immersion. It covers burned-in subtitles with on-device OCR, and it has a built-in dictionary popup for video and normal page text.
We’re already exporting to Anki, and the next destination we want is Renshuu. The two things that matter most for us are:
1. Exporting a looked-up word (and the sentence we mined it from) into a renshuu list or schedule the user picks.
2. Pulling known words / study status from renshuu so we can show whether something is already being studied, similar to how other tools highlight known vs unknown words.
If anyone has been using the API for list/schedule adds or known-word style checks, I’d love to hear what worked well and what was awkward. And マイコー, if there’s a preferred way for a mining extension to do those two flows, happy to follow that.
Thanks for keeping this API thread going. Happy studying!
I’ve been building a Chrome / Firefox extension called YuzuLingo for immersion. It covers burned-in subtitles with on-device OCR, and it has a built-in dictionary popup for video and normal page text.
We’re already exporting to Anki, and the next destination we want is Renshuu. The two things that matter most for us are:
1. Exporting a looked-up word (and the sentence we mined it from) into a renshuu list or schedule the user picks.
2. Pulling known words / study status from renshuu so we can show whether something is already being studied, similar to how other tools highlight known vs unknown words.
If anyone has been using the API for list/schedule adds or known-word style checks, I’d love to hear what worked well and what was awkward. And マイコー, if there’s a preferred way for a mining extension to do those two flows, happy to follow that.
Thanks for keeping this API thread going. Happy studying!
I worked on something similar for a while, an OCR app for grabbing words from screen and adding them to Renshuu.
By far the biggest issue I came across was checking if a given word already exists in a list.
The docs for /word/search mention a "presence" field which should easily cover that, but the real endpoint never returns that data.
I know I could've reported this here, but I just kinda gave up on that project sadly.
