Testing taking way too long

[ryansurf]

The automated tests are taking way too long in the ci pipeline. This is probably because actual API calls are being made in the tests (ex: to test the GPT, we send a request & wait for a response from the service 🤦). We should utilize mocking/dependency injection instead

Im planning on working on this and writing it down as a reminder, but if anybody else wants to take a shot, feel free! If i havent assigned myself i havent started yet

[DanieSimonlLowe]

I Will have a go at this if that is okay. (I saw on discord someone else is already doing this, so I won't do it unless I was mistaken )

[ryansurf]

Hey @DanieSimonlLowe, go for it! Let me know if you have any questions

[christe6-osu]

I'd love to help out with this if there is still work to be done.

[christe6-osu]

I was able to create a mock test for the Timeout situation for default_location but the other tests require some more research.

For openmeteo, I was able to dig into the source and find how they are converting json responses into a flat buffer. I was thinking that a helper function could be created that would do the same thing. This could then be used to create the expected response and allow mocking for the weatherapi calls that many of the tests use. Does that sound like a good approach?

As for the gpt tests, I am having a harder time finding what exactly is being returned by client.chat.completions.create. If you have any insight there, please let me know.