roxyglobals

R-CMD-check Codecov test coverage CRAN r-universe latest

Generate utils::globalVariables() from roxygen @autoglobal and @global tags.

Installation

 # Install the released version from CRAN
 install.packages("roxyglobals")
 
 # Install the released version from r-universe
 install.packages("roxyglobals", repos = "https://anthonynorth.r-universe.dev")
 
 # Or the development version from GitHub:
 # install.packages("remotes")
remotes::install_github("anthonynorth/roxyglobals")

Setup

Add roxyglobals to an R package DESCRIPTION via:

 # Add to current R package
roxyglobals::use_roxyglobals()
 
 # Or add to another R package
 # install.packages("withr")
withr::with_dir("path/to/package", roxyglobals::use_roxyglobals())

Config

By default, roxyglobals writes all discovered globals (including duplicates) to R/globals.R. You may configure the filename and global generation behaviour with:

 # write globals to R/generated-globals.R
roxyglobals::options_set_filename("generated-globals.R")
 
 # only emit unique globals
roxyglobals::options_set_unique(TRUE)

Usage

Add @autoglobal to a function roxygen comment block, example:

 #' Summarise responses
 #'
 #' @name summarise_responses
 #' @param responses a data.frame of responses
 #'
 #' @autoglobal
 #' @export
summarise_responses <- function(responses) {
 # station_name, station_type, end_time, start_time need to be added to 
 # utils::globalVariables() to keep R CMD CHECK happy
 responses |>
 dplyr::group_by(station_name, station_type) |>
 dplyr::summarise(
 count_responses = dplyr::n(),
 total_hours = sum(
 as.numeric(end_time - start_time, units = "hours"),
 na.rm = TRUE
 ),
 .groups = "drop"
 )
}

Or @global, example:

 #' Summarise responses
 #'
 #' @name summarise_responses
 #' @param responses a data.frame of responses
 #'
 #' @global station_name station_type end_time start_time
 #' @export
summarise_responses <- function(responses) {
 # station_name, station_type, end_time, start_time need to be added to 
 # utils::globalVariables() to keep R CMD CHECK happy
 responses |>
 dplyr::group_by(station_name, station_type) |>
 dplyr::summarise(
 count_responses = dplyr::n(),
 total_hours = sum(
 as.numeric(end_time - start_time, units = "hours"),
 na.rm = TRUE
 ),
 .groups = "drop"
 )
}

Run devtools::document() to generate utils::globalVariables() in your globals file (default R/globals.R). Example globals file:

 # Generated by roxyglobals: do not edit by hand
 
utils::globalVariables(c(
 "end_time", # <summarise_responses>
 "start_time", # <summarise_responses>
 "station_name", # <summarise_responses>
 "station_type", # <summarise_responses>
 NULL
))