Compute frugal risk metrics for an R package

A lightweight ("frugal") variant of [risk_assess_pkg()] that skips the most expensive parts of a full assessment:

* the test suite is **not** executed (R CMD check is run with `–no-tests`, sourced from [setup_rcmdcheck_args()] with `check_type = "2"`), * unit-test coverage is **not** computed (no call to `test.assessr::get_package_coverage()`), and * remote metadata (CRAN/Bioconductor versions, GitHub stats, downloads, reverse dependencies) is **not** fetched.

What is still produced:

* documentation risk metrics ([doc_riskmetric()]), * R CMD check score (with the frugal flag set described above), * direct & session dependencies, exports, author and license metadata, * a traceability matrix without coverage ([create_traceability_matrix()] with `execute_coverage = FALSE`), * a `risk_analysis` block ([get_risk_analysis()]).

Usage

get_frugal_metrics(package_name, version = NA, repos = getOption("repos"))

Arguments

package_name

Character. Name of the package to assess.

version

Character or `NA`. Specific version to fetch. Default `NA` uses the latest available.

repos

Character. Repository URLs to use when fetching the tarball. Defaults to `getOption("repos")`.

Value

A named list with the same top-level shape as [risk_assess_pkg()]:

`results`

Risk-metric results list (with frugal placeholders).

`covr_list`

Coverage placeholder produced by [init_frugal_covr_list()].

`tm_list`

Traceability matrix (no coverage).

`check_list`

R CMD check output from [run_rcmdcheck()].

`risk_analysis`

Risk levels from [get_risk_analysis()].

Returns `NULL` if the tarball cannot be fetched or the package cannot be unpacked.

See also

[risk_assess_pkg()], [setup_rcmdcheck_args()], [set_up_pkg()], [init_frugal_results()], [init_frugal_covr_list()].

Examples

if (FALSE) { # \dontrun{
frugal <- get_frugal_metrics("here", version = "1.0.1")
} # }