curiouslycurious / gw2api

A wrapper of the Guild Wars 2 API written entirely in Rust

curious

• 🦀 Rust zealot 🦀
• Linux _minimalist
• Has too many projects to work on...

Other Places

• sr.ht - the superior git

gw2api's Issues

Fix codecov updates

Currently codecov does not update upon a succesful CI build. Needs investigation.

Use Chrono for timestamp strings

Need to look into the Chrono crate and see how well-made it is, but if it is fairly good there shouldn't be a problem with replacing timestamp strings in e.g v1/wvw/matches.rs with a Chrono implementation.

Wiki Errors

There are a few errors or cases of incompleteness on the official wiki for the API. I'll try to list them as I find them in this issue.

v1/map_floor

Missing objects:

• list of god_shrines
• list of adventures
• list of training_points
• list of adventures

Fields missing:

• skill_challenges:
  • expac
  • idx
• sectors:
  • bounds
• tasks:
  • bounds

Replace reqwest with another HTTP client

There are a few issues with reqwest, as pointed out in: https://medium.com/@shnatsel/smoke-testing-rust-http-clients-b8f2ee5db4e6

I could use one of the simpler ones, since the GW2 API only fetches resources via GETs and has no need for more advanced functionality. This would also significantly decrease the compiled size of the project since the reqwest project for some reason is gigantic.

Then it would be up to the end-user to multi-thread any functionality, which is more desirable, since they then have more control over when a response is handled.

Finish v2 endpoints

This will probably be an easier task than the v1 endpoints as they are more consistent and not as oddly defined, and are often more concise.

Upload crate to crates.io

Need to decide if the current name: gw2api is good enough or if I should message some of the other unfinished wrappers if I can grab their name instead. Will have to think about this until v1 endpoint is finished. Once that is done I should publish 0.2.

Flesh out the documentation

Currently the documentation is decent, but somewhat barebones.

I don't need a code example for every little thing, but some descriptions here and there will go a long way.

Implement better error handling

Currently error handling is very bare bones. After switching to another request library the wrapper should be able to handle if a request times out or the endpoint responds with a 404 and not just panic.

Further error handling should be done on the deserialization from json to some struct object, since any error message from a failed deserialization is ususally non-sensical to debug.

Add a simple example to the lib.rs and README.md

A simple example should be made to showcase the ease-of-use and flexibility of the wrapper. Currently you have to either read test code or try to figure stuff out from the unfinished documentation and cross-reference with the wiki.