I would like to automate a bit my flow of creating courses. Are there any plans of creating REST API? Will appreciate it much!
May I ask, what is ‘REST’? I’m probably just ignored :).
Anyway an API sounds good.
If you want to automate the creation of a course, do you know that you can bulk-upload the word list and any attribute fields (such as gender, part of speech or notes). I always keep the course in an excel file and to create a course/add material is as easy as copy/past. I hope this helps for now.
Audio and ‘alternatives’ cannot be uploaded this way - you need to do that manually. Though for alternatives a comma or semi-colon or something can be used; which I always forget because I don’t use that feature.
(I’ve added the words ‘Feature Request’ to the title; it’s a convention on this forum. Memrise uses the feedback for product development, selecting the ones they like.)
‘Representational State Transfer is an architectural style that defines a set of constraints to be used for creating web services. Web services that conform to the REST architectural style, or RESTful web services.’
As I’m thinking it’s actually not so important to do it in the RESTful way. I will be happy with every reasonable API
Yeah, I’m currently doing that, but hope to have some API endpoint when I can attach words and audio also (or at least to have the possibility to do that separately - by calling two endpoints.)
Thanks for your help and making the subject more visible
I found unofficial API https://github.com/carpiediem/memrise-enhancement-suite/wiki/Unofficial-Documentation-for-the-Memrise-API#apicourseget
Can Anybody with Memrise Team say anything about API plans?
Thanks
Adding a vote, with the solemn understanding that the userbase for this feature is likely small and vocal.
carpediem’s work, (cited by dominik137) seems to have been focused on augmenting the Memrise user experience.
My aim is consistent with OP’s - to automate course generation and syncing.
After getting fed up with reversing the existing API (@Memrise team – well done? ), I hacked together some Python that uses Selenium to interact with the Memrise client GUI instead (shudder). Headless Firefox processes will keep my room heated this winter.
Perhaps we can help devs by being more specific about what features we’re looking for. As John_Baite remarks, there is an existing bulk-upload feature. I’d like to use this, but it’s currently lacking two key parts:
- an endpoint to cURL against
- a means to add audio (pretty critical for language learning)
This.
But the specificity thing too. We all know how awful it is to get a non-specific brief that could mean anything from ‘it needs more cowbell’ to ‘nuke the lot and start again from scratch’. Things are more likely to get time allotted to them if it’s obvious how long it’s going to take to achieve whatever’s been requested. Well, they would if it were me, anyway.
Why not look what’s going on behind the scenes when you select a file for upload and do that from code? Sure, it’s not as trivial as using the rest of the API methods, but shouldn’t be too hard to figure out.
Also, Memrise has been pretty much in end-of-life mode (just look around the forums) for a long time, so it would be a bit foolish to expect them to make API improvements at this point.
would love to hear from one of the engineers about what the status is of this request, company doesn’t seem end-of-life imo, maybe i’m wrong?
I’d like to make a simple automation where i can right click a word i don’t know and add it to a course of “incoming” vocab. Looking through API options now, rather not have to duct tape something together like @nkirkby.