Just a random idea- but,
Would anyone be interested in having a .net based blazor front end for lemmy?
Sadly, there isn’t yet a c# nuget package, but, that wouldn’t be hard to fix.
The API itself, is pretty well documented.
And, building custom front-ends is encouraged as well.
https://join-lemmy.org/docs/en/contributors/04-api.html#creating-a-custom-frontend
You must log in or register to comment.
Figuring out how to actually put a package on nuget, but I have fully implemented the API
https://github.com/SlyCedix/dotNETLemmy/tree/main
https://slycedix.github.io/dotNETLemmy/I intend to write a mobile app using Avalonia UI using it
Do note, it’s mostly untested so there’s a good chance a handful of the datatypes won’t serialize or deserialize correctly, but that’s an easy fix once found
edit: Uploaded to nuget, currently being validated
Just, gonna add-
If I end up building the front-end for this, is gonna end up looking like a bootstrap example website!!!
That’s a good idea. I would like to experiment with some ideas where one could have an RSS reader together with a lemmy client. But the ideas are too raw and I’m not yet sure if it’s feasible.
Maybe I misunderstood, but it sounds like you are looking for “team mates” to develop a lemmy client. Or was it just a suggestion to others?
I think that would be great! I started working on a Azure Functions bot and started a DI dotnet client that I put on nuget https://github.com/ydinkov/Lemmy.Net
It’s super wip right now, but Ill be adding more models, tests and error handling.
Just out of interest, why not use refit for something like this? You could get a basic client going much faster - unless there’s some specific behaviour you want for the client (which would require painful customisation with refit).
I’ve come to be quite a fan of refit recently purely for how quick you can get a basic, working client
I like refit too! It’s super nice, but very quickly i realised the API does some weird undocumented stuff and isn’t very consistently implemented. I feel more comfortable doing the boilerplate my myself (chatgpt helps) while i figure out the funk.
Ah, that’s fair - haven’t encountered anything weird with it yet. I can imagine with chatgpt it’s become much easier though - I need to start giving AI tools a try. Do you have a particular set up for using chatgpt with rider or anything?
No setup at all, just strip-mine the sourcecode for the existing TS client, have it generate curl requests. Run the curls, ask it to create POCOs for the results, and adjust.
@pwshguy@programming.dev wrote a pwsh module for lemmy and can attest to the troubles =D
