REST API #5

Posts
- Details
- List
- Create
- Update
- Delete
Comments
- Details
- List
- Create
- Update
- Delete
Users
- Details
- List
- Create
- Update
- Delete
Likes
- Details
- List
- Create
- ~~Update~~
- Delete
Boost
- Details
- List
- Create
- ~~Update~~
- Delete

Yes please! :)

Please consider implementing the ActivityPub Client to Server API, if possible.

Most endpoints need authentication, so atm this is blocked by #275

I tested https://docs.joinplu.me/api/, but all request fails.
Is it a bug or just not yet ready?

Are you talking about this embedded widget?

It doesn't work (for now, at least). It just shows how it is. Actually you have to send requests to working Plume instance such as https://fediverse.blog/api/v1/posts?title=Plume+Development+2021-01-25

Yes, I tried to test the api for optional usage from client site with javascript to fill posts by ajax call.

Get posts is possible by

title as full title (path)?
subtile (?)
content (inline search for word(s)?)

Searching title need to be full title (path) hit?

Plume+Development+2021-01-25

Could you give some more examples how to get one or more posts by title, subtitle or content?

Doesn' working

https://fediverse.blog/api/v1/posts?title=plume-development-2021-01-25   # post path?
https://fediverse.blog/api/v1/posts?content=plume+development       # content contains "plume" and "development"
https://fediverse.blog/api/v1/posts?title=plume+development+2021-01-25   # case sensitive search?
https://fediverse.blog/api/v1/posts?content=plume       # content contains "plume"

Searching title need to be full title (path) hit?

Yes, it does.

I read implmentation. According to it, all of title, subtitle and content need to match exactly...

Anyway you can retrieve all contents from GET /posts without any params. (But don't do that againt any instances excluding yours because it occur high loads.)

API doc for GET /posts is wrong. I filed an issue to documentation repo: plume/documentation#122.

Anyway you can retrieve all contents from GET /posts without any params. (But don’t do that againt any instances excluding yours because it occur high loads.)

Because it loads all the posts of the instance without a time or user filter?

So some more filters and default limits would be nice?

content of one user
content of one blog
default limit to post count and with pagination

I think about a way to use plume as backend server for some blogs, loaded by api (limit, pagination) to show inside of a custom html / js frontend.

Because it loads all the posts of the instance without a time or user filter?

Yes, that's the reason.

So some more filters and default limits would be nice?

I think so.

For user activity, you can retrieve Atom feed although it might not be useful for modern development. Its URI is /@/{username}/atom.xml.

I think about a way to use plume as backend server for some blogs, loaded by api (limit, pagination) to show inside of a custom html / js frontend.

That's nice idea! I wish someone had time to implement APIs...

I'm not a developer, just play a little bit with web projects, php and javascript. So unfortunately I can't help much here ... :(
Also started to do some tests with golang and rust, but it's more to learn about than php / js

I found forum / reddit clone lotide: https://sr.ht/~vpzom/lotide/
It's similar to lemmy, but with a swagger compatible api and so a better documentation. So it would be easier to build a custom frontend which looks more like a modern forum for example.

Similar to plume there is pubgate written in python, but development seems to be stopped.

Plume is great and have some pros and cons compared to writefreely. I miss the feature to follow blogs instrad of users here, but writefreely doesn't support image upload.

I think for my blog use case plume is the better one 👍

I didn't mean you should develop API. I just wished. Don't worry.

I'm glad to hear you prefer Plume!