Skip to content

6️⃣ Distinguish different kinds and purposes of "packages"? #70

@mpadge

Description

@mpadge

From @zkamvar via #64:

For the varnish package: this is not a package per-se, but a delivery mechanism for installed components that can be updated independently of the main code. I know that the automated tools cannot detect that, BUT I'm interested to see if you have any ideas of how this can be indicated. For example, adding a DESCRIPTION file is a semi-well-known way to get R to install dependencies for a research compendium, but this would never be considered a package. The same goes for the "learn" "package" ropensci-review-tools.github.io/repometrics-demo/repo.html?repo=learn

And then my comment in response:

6️⃣ This is a big issue, and also related to discussions in #50 about applicability beyond R packages alone. I definitely have some ideas about how atypical packages can be identified, after which flagging them in the dashboard should be relatively straightforward. We have complete treesitter analyses of all packages and functions within those, alongside other code-tagging outputs for languages other than R (including .js/.ts). These data should provide insight into different types and purposes of packages.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions