title |
---|
Community repository |
The R-multiverse Community repository simplifies the distribution and deployment of R package releases, and it puts this power in the hands of package maintainers.
Users can navigate to https://community.r-multiverse.org to learn more about packages in Community.
To install a package, set the repos
argument in install.packages()
.
For example:
^[
Caution: when setting two repositories in the repos
argument,
the same package may be in both.
install.packages()
chooses the most current version,
which could be in either repository.
]
install.packages(
"polars",
repos = c("https://community.r-multiverse.org", getOption("repos"))
)
Anyone can contribute packages to Community via GitHub. After a one-time registration process for a package, the package maintainer directly controls distribution through the release system on GitHub/GitLab. Each new release automatically deploys to the repository without any manual intervention. ^[This maintainer-driven experience was made possible by the incredible infrastructure of rOpenSci's R-universe system.]
A traditional software version cycle begins with the development phase, where bugs are identified and fixed, and ends with the production phase, which promises a seamless user experience. Community aims at a space just before production, commonly referred to as Quality Assurance (QA). This means that:
- Each package release has the full endorsement of its own maintainer. R-multiverse always provides a version that its developer chose to distribute for general use.
- The user is still responsible for judging whether a package is appropriate to use. Some releases may still have issues, as indicated by their check results, publicly available at https://community.r-multiverse.org.
You can deploy a package through a personal or organization-level
universe even if the package also is part of R-multiverse.
In fact, a personal universe may help reproduce and diagnose R CMD check
issues
as you prepare a package for Production.