Laura Bąkała

Honestly, as mentioned in #39, valid version numbers can be almost anything. While there are guidelines for semantical versioning, not all packages follow them and, ultimately, the package's author has...

Ahh, didn't know that, interesting! I'm on board with dropping that, should I implement this check?

I saw you using `assertthat`, so I reused their function `is.string()` with added (also theirs) `noNA()` check. Also, dropped the `check_version` parameter, since the check itself has to pass now...

Checking the failing tests... - [x] `cran_package_history()` fails because v1.2.10 of `igraph` was published and the sorting would place it between v1.2.1 and v1.2.2; I know it's related to #39,...

Okay, I believe I might have fixed failing tests. I went as far as submitting [versionsort](https://github.com/ErdaradunGaztea/versionsort) to CRAN, but for now I copied version sorting code into `R/versort_utils.R`.

Hey, checked out `cranly` and we would definitely want to include it in our vignette. I like the idea of package dependence index, though I couldn't find any explanation, what...

Uhh, this one is worrying since it turned out it doesn't fail until it's plotted to a device. Precisely, saving above to a variable doesn't raise an exception, displaying it...

It's basically two-line fix and I pushed it to [fix/no-deps](https://github.com/DominikRafacz/deepdep/tree/fix/no-deps) branch, if you'd need it. And it's true about too many plot options, the way they interact with each other...

(right, `usethis` suggests calling `use_dev_version()` after publishing a new release, so it's 0.4.1.9000 now)

(sorry for not answering earlier, I saw it just today) Yes, it very much could (unless there's something I don't know of), good catch! Could you provide an example URL...