Amazing! My first thought was to develop a small web application to search for words, kanjis and vocabulary and share it with my classmates, but I saw that you only have 500 requests a day and I'm not sure if it's a good idea to use my personal API key and let everyone make requests using it. I would like to hear your opinions;)
Amazing! My first thought was to develop a small web application to search for words, kanjis and vocabulary and share it with my classmates, but I saw that you only have 500 requests a day and I'm not sure if it's a good idea to use my personal API key and let everyone make requests using it. I would like to hear your opinions;)
Couldn't you classmates just create their own accounts?
If it's just about looking up stuff in the dictionary, that then works without logging in.
I've been working on an API client in Rust, and I've found a few inconsistencies with the docs versus the implementation. The only reason I mention it is that it just means a couple more speedbumps in getting started. Is there a preferred way to report issues and provide feedback other than this thread, or am I okay to lay out some technical details and / or nitpicks here?
On the separate topic of 500 requests per day, maybe it makes sense to increase that amount with a Renshuu Pro subscription? Mostly just spitballing, but I'm thinking of the use case where someone wants to do a bulk import of lists from other tools. I like using Akebi as my dictionary when I'm studying, for instance, and I would love to be able to do a bulk import to a Renshuu list from an Akebi list export. Rate limiting and throttling would also be a viable solution.
Amazing! My first thought was to develop a small web application to search for words, kanjis and vocabulary and share it with my classmates, but I saw that you only have 500 requests a day and I'm not sure if it's a good idea to use my personal API key and let everyone make requests using it. I would like to hear your opinions;)
I wouldn't recommend it - that key has write access to your schedule/list contents.
I've been working on an API client in Rust, and I've found a few inconsistencies with the docs versus the implementation. The only reason I mention it is that it just means a couple more speedbumps in getting started. Is there a preferred way to report issues and provide feedback other than this thread, or am I okay to lay out some technical details and / or nitpicks here?
On the separate topic of 500 requests per day, maybe it makes sense to increase that amount with a Renshuu Pro subscription? Mostly just spitballing, but I'm thinking of the use case where someone wants to do a bulk import of lists from other tools. I like using Akebi as my dictionary when I'm studying, for instance, and I would love to be able to do a bulk import to a Renshuu list from an Akebi list export. Rate limiting and throttling would also be a viable solution.
I mentioned this earlier, but this is my first foray into API development, so I am surely making some small (and big) mistakes. Please nitpick away! Everything can go here.
I will most likely have an increased usage for pro. I'd also like to get more info as to how much people think they'd actually use so I can make a better feel for what to set it at (as 500 is very arbitrary.)
I just thought of an endpoint that doesn't exist yet (I think): Adding private sentences (maybe also other stuff like mnemonics) to dictionary entries via API.
For my yomitan-renshuu connector I could configure it so when I add a new word from yomitan, it not only sends the word, but also the sentence it appeared in to my program. I would like to use this to automatically add example sentences.
That is an interesting idea, although there would not be a way to parse the sentence, making it not...super useful. I suppose there could be an option to pass the string through the text analyzer, and take whatever it suggests?
I got another suggestion (via DM) from another user, but would love to get feedback from others as to how to implement it on the endpoint.
Basically, they want to be able to search multiple things at once. For example, search for the words 漢字, 言葉, and 言語, then get separate results for each one.
Technically, this is relatively easy, but in order to keep the endpoint clean, wouldn't it be ideal to have a separate endpoint? Otherwise, the docs/specs on the documentation page would be messy - yes, sometimes the results (number of pages, etc.) are at the top level, but in other situations, they'd be as part of an array of result objects?
There definitely is, although I'll be honest - I have a weird conflict that makes me want to keep the TA from being *too* accessible.
renshuu is a pretty light existence in terms of servers, and helps me keep things cheap for everyone. However, parsing texts in the TA is a TON of server work relative to everything else - I have all of them get processed on a "backend" server that runs various scheduled jobs because it can really slow down things.
Right now, there is still a good amount of growth possible in TA usage before it starts to saturate that server, but I don't want to *rush* off on a route that will require a disproportionate amount of server resources for a relatively minor feature of renshuu.
Adding grammar to lists and schedules doesn't seem to work Getting certain grammar doesn't work (e.g., GET /v1/grammar/42) Grammar search needs paging (no way to get more than 25 results) Words/Kanji/Grammar presence is broken (schedule hasWord is always 0, word lists work though, did not test kanji or grammar lists) Adding words/kanji to schedules works, but deleting does not, probably same cause as the previous point because hasWord is always 0
Thanks - I am on vacation, so I cannot look at most of these quite yet. However, most of the PUT/DELETE/hasWord should be fixed. It actually wasn't really the API's fault - I sharded a schedule database table, and I thought I had caught all of the cases, but not all of them.
If it does work there, please show me the code (api key redacted) that you're using to try and make the call.
hm, it seems to be working fine now. I tried through the swagger UI previously and it was returning an error page only on the /all/{termtype} api. Not sure why but anyways, seems good now!
Hello everyone, I'm working on an android widget with the API, it's still a bit raw but here is the preview of the app and of the widget:
You just open the app, set your api key, and at each app start or every 15 minutes it updates the information in the widget. Clicking the widget brings you to the renshuu app!
I will make this open source when it's ready (now the code is a bit ugly and I'm going to holiday next week), so that it can be integrated directly into renshuu itself in a future
