deepdep
deepdep copied to clipboard
Visualise and Explore Deep Dependencies of R packages
There should be a way to generate `deepdep` object for given package without depth limit. It would be intuitive to achieve this by using `depth = Inf`, but our current...
When you call: ``` deepdep::deepdep("versionsort") ``` and ``` deepdep::deepdep("stringi") ``` the returned data frames have different columns:  The latter are correct, probably the effect of deleting base R dependencies...
Package weight could be a parameter of `deepdep()` function or a function applied to a `deepdep` object. I don't like parameter bloat, so I'd opt for the latter solution. This...
Best implemented as an S3 method. The call would look like: ```r dd
It would create a deepdep plot and insert it in a separate section at the bottom of `README.md` (or maybe a code chunk in `README.Rmd` instead?).
**Description of a problem related to your idea** I'd like to compare differences between dependencies used in different versions of a package; e.g. differences between deepdep v0.2.0 and deepdep v0.4.1....
The CRAN mirror/repo to use seems to be hardcoded: https://github.com/DominikRafacz/deepdep/blob/389934832e0380e893ef13ddf4434879772302d0/R/get_available_packages.R#L46 Could the default `getOption("repos")` not be used? Or alternatively, something like `contrib.url(getOption("deepdep.repos", "https://cloud.r-project.org/"))` ?
It would probably have to be done by overwriting layer colors; perhaps a separate visualization? It could be helpful when tracking if no license clash occurs.
When passing a character vector of packages to `plot_download()`, the title is incorrect; this is the result of the following call: ```r plot_downloads(c("versionsort", "deepdep")) ```  Clearly, `versionsort` and `deepdep`...
Release 0.4.1 is really sweet but I fear this is a regression: ```r > plot_dependencies(deepdep("data.table")) Error in `[.data.frame`(g, , "labeled") : undefined columns selected > ``` With the default of...