Add cff

[c-martinez]

I'm using this PR within my repo to have a code review before pushing upstream.

As far as I understand, the workflow should be that you create a package, and add a CFF as follows:

usethis::create_package('.')
usethis::use_cff()

And the output should be something like this:

#> ✔ Writing 'citation.cff'.
# ☐ Edit 'citation.cff'. (see <https://book.the-turing-way.org/communication/citable/citable-cffinit.html> and <https://citation-file-format.github.io/cff-initializer-javascript/#/>)

Things I am not very sure about:

• Can we fill some information in the created CITATION.cff file? It looks to me as if some fields could be pre-filled for the user, for example license, author name, release date?
• Are tests sufficient?
• Is documentation sufficient?
• Do we need to generate documentation via roxygen2 somehow? (see I saw man/*.Rd files mention not to edit by hand, but I have no idea how this works).

[c-martinez]

Invited @PabRod as collaborator, once he accepts, I can assign him as reviewer.

[PabRod]

I booked some time to take a look at this on Tue 30th.

[PabRod]

Very good. Are you sure this is the first time you use R?

[PabRod]

I like the parallel with use-citation() (available in R/citation.R). It looks good, but I am slightly confused about their not-so-straightforward workflow.

[PabRod]

@c-martinez:

• Do we need to generate documentation via roxygen2 somehow? (see I saw man/*.Rd files mention not to edit by hand, but I have no idea how this works).

Documentation can be generated from the Build/More menu. In R jargon, this is know as roxygenizing.

It is usually a good idea to set it to roxygenize on every Install and restart from the Build/More/Configure build tools/Documentation with Roxygen/Configure... submenu.

Forgetting to roxygenize is a very common mistake. Indeed, there were functions in main that were not roxygenized.

[PabRod]

• Can we fill some information in the created CITATION.cff file? It looks to me as if some fields could be pre-filled for the user, for example license, author name, release date?

I suggest keeping it as it is. I like how it mirrors use_citation(). If you disagree, please mark it again as to do.

[PabRod]

• Are tests sufficient?

I'd say yes, but for some reason yours is not running. The whole battery of tests shows many fails, and it stops before reaching this one. I couldn't manage to run this test only, and I don't fully understand why. My guess: it is related with their workflow.

Update: I was wrong. After updating my R to the latest version, only one test fails. It's not the cff one!

[PabRod]

• Is documentation sufficient?

It is definitely not insufficient, but let's:

• Draft a pull request explaining why cff is important
  • and consider editing the in-code documentation afterwards

[c-martinez]

I didn't see you edited your comment about tests. It looks like you can run a single test: https://testthat.r-lib.org/reference/test_file.html but I somehow don't get that to work (clearly need to learn how to use R studio properly).

Anyway, if you are satisfied with this review, I will merge and draft a PR.

[c-martinez]

I've created a PR, now let see if they have any feedback. Thanks for your help!