Implement testing across the package

[ryantibs]

Now that the package is getting farther along, I think it is about time we implement testing across the package. Especially because I believe (and hope) that more others will be contributing to it in the near future.

@capnrefsmmat You are our resident stickler-for-the-rules-kind-of-guy, did you want to take the lead on this? You know I'm kidding 😛, and I would be grateful if you took this on. I think @qpmnguyen @rafaelcatoia could be willing to help as well, so if you have some ideas/scaffolding then perhaps they would be willing to implement it with you.

[qpmnguyen]

Hello! Happy to help write unit tests!

[ryantibs]

@qpmnguyen @rafaelcatoia Did you guys want to start taking a look at this? You can see the way that testing is implemented in the covidcast R package (for example) and also read up on testing in R in general if you haven't had exposure to this already.

In side conversation with @capnrefsmmat, he said he is busy for a little bit with other responsibilities, but can start taking a look perhaps mid March, so in any case I think it makes sense for Quant & Rafael to just start learning/putting things together in the meantime, if you're still up for it. Thanks!

p.s. in the advanced.Rmd vignette, I have some code chunks with include = FALSE that were basically just some hand-tests I was running to make sure the results were what I expected with sliding. In building tests, you could remove these chunks from that vignette (they're not needed there anymore, they're just in there so I didn't forget them) and incorporate them into the tests you're writing.

[dajmcdon]

@dshemetov @qpmnguyen @rafaelcatoia Is anyone actively working on this? I'm going to re-assign if no.

[dshemetov]

I'm not.

[qpmnguyen]

I am not currently working on this. Feel free to re-assign!

[kenmawer]

I just realized that I misread what tests need to be done; the following do not have tests:
archive.R
growth_rate.R
methods-epi_archive.R (includes epix_slide.R, not to be confused with epi_slide.R)

[brookslogan]

Tests for the above have now been addressed in PRs (e.g., #159, #197); closing as completed. We may want to push on #204 to catch anything else hiding untested.