Rcmdcheck fixes

[ahl27]

in response to requests in ahl27/biostrings-isc-proposal-2024#10 (comment)

• formatting fixes (4 space indents, K&R style control flow / functions)
• new tests for MIndexList
• documentation for MIndexList
• documentation fixes for matchPDict
• small other fixes to resolve some BiocCheck warnings

[ahl27]

@hpages i'm going to wait for you to take a look on this one after previous misses, would appreciate a second set of eyes.

[hpages]

Thanks @ahl27.

PR looks good. Thanks for adding a man page and dedicated unit test file for the MIndexList class. Note that the name of the latter should be test-MIndexList-class.R (test-<name-of-corresponding-dot-R-file>).

I wonder what the ..R.check folder is doing in the PR. Looks like maybe you ran R CMD check from within the package source tree? Please remove.

Also not sure what the "All checks have failed error" is about. The exact error is:

cat: ./test_status.md: No such file or directory
Error: Process completed with exit code 1.

but I don't understand it.

I'm actually not sure that we need the test-coverage-nocodecov workflow in the first place. We don't have strict guidelines for GitHub workflows, but the recommendation for Bioconductor packages is to install a workflow that reproduces the standard R CMD build + R CMD check steps that developers normally run locally before pushing anything. Note that these are also the steps run by our daily builds here. One widely used workflow amongst Bioconductor packages is the R-CMD-check-bioc workflow from the biocthis package (FWIW I use some lightly edited version of it in the igblastr repo).

H.

[ahl27]

oops, thanks for the catch -- my computer has been acting up a little recently and my workflows haven't working like I expect. Removed the unnecessary file and updated naming.

For the automatic workflow, I can disable it -- it used to run all the testthat tests, R CMD CHECK, BiocCheck, and code coverage analysis to make sure that new PRs were passing tests and increasing code coverage...but at some point that workflow seems to have broken.

I can turn it off in a separate commit.

[ahl27]

Automatic workflow has been disabled; future PRs shouldn't have this error (unfortunately it doesn't seem to disappear from this PR, but going forward at least it's fixed). I can either fix the workflow in the future or remove it entirely later.

friendly ping @hpages , I'll plan to merge this week so that there aren't outstanding rcmdcheck/bioccheck issues.

[hpages]

Thanks for the ping. I didn't forget. I just need a couple more days. Thanks again!