Create a dependency graph for all Bioconductor packages.

The function takes in a 'binary_repo' which is a CRAN style google bucket. It creates a package dependency graph in the form of a 'list()' while excluding R 'base' packages. The 'binary_repo' needs to be a public google bucket. If you need to create a new google bucket in a CRAN style structure, see 'gcloud_create_cran_bucket()'. If a newly created bucket is passed into the function, it will create a full package dependency structure for all Biconductor packages.

pkg_dependencies(
  version,
  build = c("_software", "_update"),
  binary_repo = character(),
  exclude = character()
)

Arguments

version

character() Bioconductor version number.

build

character() One of '_software' (rebuild all packages in the 'BioCsoft' repository) or '_update' (existing binary packages in `binary_repo` for which newer versions are available in 'BioCsoft', and packages in 'BioCsoft' that are not available in `binary_repo`).

binary_repo

character() vector of the binary repository in the form eg. "anvil-rstudio-bioconductor/0.99/3.11"

exclude

character() vector of packages to exclude

Value

'pkg_dependencies()' returns a list of Bioconductor packages with the dependencies of the package. If the 'binary_repo' given has a pre-populated set of packages then only the packages that need to updated are in the list.

See also

'gcloud_create_cran_bucket'

Examples

if (FALSE) {
## First way, give it a pre-existing binary repository
## hosted as a google bucket.
deps <- pkg_dependencies(
    binary_repo = "anvil-rstudio-bioconductor/0.99/3.11"
)

## Second way, create a new bucket with no packages in it.
gcloud_create_cran_bucket(
    "gs://my-new-binary-bucket", "1.0",
    "3.11", secret = "/home/mysecret.json",
     public = TRUE
)
deps_new <- pkg_dependencies(
    "_software",
    binary_repo = "my-new-binary-bucket/1.0/3.11"
)
}