| Title: | Dependency Use Checker for R Packages |
|---|---|
| Description: | A way to check R packages and shiny applications for unused package dependencies, or packages with low usage. Reduces the requirement to have unneccessary dependencies in a project. |
| Authors: | Ashley Baldry [aut, cre] |
| Maintainer: | Ashley Baldry <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.1.0 |
| Built: | 2024-11-23 03:47:39 UTC |
| Source: | https://github.com/ashbaldry/depcheck |
Counting the frequency that a function is called within selected code.
checkFunctionUse(function_name, code, package_name = NULL)checkFunctionUse(function_name, code, package_name = NULL)
function_name |
Name of the function to check against |
code |
A character vector of code chunks to check for function use |
package_name |
Name of the package that |
If package_name is left as NULL, then there is a chance the checks will find times
where the function is used, but is explicitly called from another package.
There are three stages of checking: - If the function name exists anywhere in the code chunks - If it has not been previously assigned in a code chunk - If the function name is not quoted
At each stage, only keep chunks where the function has been called to improve speed
'checkPackageDependencyUse()' checks the DESCRIPTION file for packages mentioned in the Depends
and Imports fields.
Using these packages, it then checks the scripts in the R directory to check for function use of each package.
checkPackageDependencyUse(path = ".", include_suggests = FALSE, verbose = TRUE)checkPackageDependencyUse(path = ".", include_suggests = FALSE, verbose = TRUE)
path |
Path to the package root directory. |
include_suggests |
Logical, should the "Suggests" field also be checked for package dependencies or just the Depends and Imports? |
verbose |
Logical, should informative messages be printed during the dependency evaluation? |
An object of class multi_package_usage, a named list of the dependencies used, each containing a
data.frame of all the functions within the package and how often they are used within the project.
The result will flag any packages that have been rarely used in the project.
checkProjectDependencyUse, checkShinyDependencyUse
## Not run: dependency_use <- checkPackageDependencyUse() summary(dependency_use) ## End(Not run)## Not run: dependency_use <- checkPackageDependencyUse() summary(dependency_use) ## End(Not run)
This will check what functions contained within a specified package are used in code.
checkPackagesUsage(package_names, code, verbose = TRUE) checkPackageUsage(package_name, code)checkPackagesUsage(package_names, code, verbose = TRUE) checkPackageUsage(package_name, code)
package_names, package_name
|
Name(s) of the package to check against |
code |
A character vector of code chunks to check for package use |
verbose |
Logical, should informative messages be printed during the dependency evaluation? |
checkPackageUsage will return a data.frame of class package_usage. When printed it will show
a summary of the package usage.
checkPackagesUsage will return a list of class multi_package_usage. When printed it will show
a summary of all packages mentioned, and flag any rarely used.
'checkProjectDependencyUse()' checks R scripts within a project for package calling through the use of
library, require and requireNamespace. It also finds packages explicitly used via
the pkg::function notation.
Using these packages, it then checks the same R scripts to check for function use of each package.
checkProjectDependencyUse(path = ".", recursive = TRUE, verbose = TRUE)checkProjectDependencyUse(path = ".", recursive = TRUE, verbose = TRUE)
path |
Path to the directories to search for R scripts. By default looks in the current working directory. |
recursive |
Logical. Should the R file search recurse into directories? |
verbose |
Logical, should informative messages be printed during the dependency evaluation? |
An object of class multi_package_usage, a named list of the dependencies used, each containing a
data.frame of all the functions within the package and how often they are used within the project.
The result will flag any packages that have been rarely used in the project.
checkPackageDependencyUse, checkShinyDependencyUse
## Not run: dependency_use <- checkProjectDependencyUse() summary(dependency_use) ## End(Not run)## Not run: dependency_use <- checkProjectDependencyUse() summary(dependency_use) ## End(Not run)
'checkShinyDependencyUse()' checks the global.R and app.R (or ui.R and server.R) scripts within a project, and
any specified directory within the project, for package calling through the use of
library, require and requireNamespace. It also finds packages explicitly used via
the pkg::function notation.
Using these packages, it then checks the same R scripts to check for function use of each package.
checkShinyDependencyUse(path = ".", r_scripts_dir = "R", verbose = TRUE)checkShinyDependencyUse(path = ".", r_scripts_dir = "R", verbose = TRUE)
path |
Path to the application root directory |
r_scripts_dir |
Subdirectories in the shiny application that contain R scripts, write as relative paths to
|
verbose |
Logical, should informative messages be printed during the dependency evaluation? |
An object of class multi_package_usage, a named list of the dependencies used, each containing a
data.frame of all the functions within the package and how often they are used within the project.
The result will flag any packages that have been rarely used in the project.
checkProjectDependencyUse, checkPackageDependencyUse
## Not run: dependency_use <- checkShinyDependencyUse() summary(dependency_use) ## End(Not run)## Not run: dependency_use <- checkShinyDependencyUse() summary(dependency_use) ## End(Not run)
The ability to search for package loading (via library() or requireNamespace), or by direct
referencing through "pkg::function"
extractPackageCalls(code)extractPackageCalls(code)
code |
A character vector of code chunks to check for package calls |
In order to check if packages are included, need to extract all functions from the package.
getPackageFunctions(package_name) getInternalPackageFunctions(package_name)getPackageFunctions(package_name) getInternalPackageFunctions(package_name)
package_name |
Name of the package to find function names for |
Package Usage Printing Options
## S3 method for class 'multi_package_usage' print( x, warn_percent_usage = 0.2, warn_number_usage = 3, ignore_low_usage_packages = character(), ... )## S3 method for class 'multi_package_usage' print( x, warn_percent_usage = 0.2, warn_number_usage = 3, ignore_low_usage_packages = character(), ... )
x |
A |
warn_percent_usage |
Minimum percent of functions to be used within a dependent package. Default is |
warn_number_usage |
Minimum number of functions to be used within a dependent package. Default is |
ignore_low_usage_packages |
A vector of packages to ignore the low usage of, usually when already aware of the low usage, but the dependent package is necessary for the project. |
... |
Not used |
Package usage must be below both thresholds for the warning to appear. With the defaults, if a package has fewer than 5 functions then only 1 function is required to prevent a warning message to appear.
Package Usage Printing Options
## S3 method for class 'package_usage' print(x, warn_percent_usage = 0.2, warn_number_usage = 3, ...)## S3 method for class 'package_usage' print(x, warn_percent_usage = 0.2, warn_number_usage = 3, ...)
x |
A |
warn_percent_usage |
Minimum percent of functions to be used within a dependent package. Default is |
warn_number_usage |
Minimum number of functions to be used within a dependent package. Default is |
... |
Not used |
Package usage must be below both thresholds for the warning to appear. With the defaults, if a package has fewer than 5 functions then only 1 function is required to prevent a warning message to appear.
Reading R files for package dependency checks.
readPackageFiles assumes the selected path is a package, and will check the R subdirectory for files
to read. readDirectoryFiles is a more generic version of readPackageFiles
readPackageRFiles(path = ".") readDirectoryRFiles(path = ".") readRFiles(files, path = NULL)readPackageRFiles(path = ".") readDirectoryRFiles(path = ".") readRFiles(files, path = NULL)
path |
Location of the package or directory to import R files from |
files |
Vector of R file paths to import |
The default directory for readFiles is NULL, to allow the reading of files from multiple directories.
A character vector containing unique code chunks in the specified directory/files.
## Not run: readPackageRFiles() # When running outside of a package readDirectoryRFiles("R") ## End(Not run)## Not run: readPackageRFiles() # When running outside of a package readDirectoryRFiles("R") ## End(Not run)
Package Usage Summary
## S3 method for class 'multi_package_usage' summary( object, warn_percent_usage = 0.2, warn_number_usage = 3, ignore_low_usage_packages = character(), ... )## S3 method for class 'multi_package_usage' summary( object, warn_percent_usage = 0.2, warn_number_usage = 3, ignore_low_usage_packages = character(), ... )
object |
A |
warn_percent_usage |
Minimum percent of functions to be used within a dependent package. Default is |
warn_number_usage |
Minimum number of functions to be used within a dependent package. Default is |
ignore_low_usage_packages |
A vector of packages to ignore the low usage of, usually when already aware of the low usage, but the dependent package is necessary for the project. |
... |
Not used |
Package usage must be below both thresholds for the warning to appear. With the defaults, if a package has fewer than 5 functions then only 1 function is required to prevent a warning message to appear.
Function Usage Summary
## S3 method for class 'package_usage' summary(object, warn_percent_usage = 0.2, warn_number_usage = 3, ...)## S3 method for class 'package_usage' summary(object, warn_percent_usage = 0.2, warn_number_usage = 3, ...)
object |
A |
warn_percent_usage |
Minimum percent of functions to be used within a dependent package. Default is |
warn_number_usage |
Minimum number of functions to be used within a dependent package. Default is |
... |
Not used |
Function usage must be below both thresholds for the warning to appear. With the defaults, if a package has fewer than 5 functions then only 1 function is required to prevent a warning message to appear.
## Not run: package_use <- checkPackageUsage() dependency_use <- checkPackageDependencyUse() summary(dependency_use) ## End(Not run)## Not run: package_use <- checkPackageUsage() dependency_use <- checkPackageDependencyUse() summary(dependency_use) ## End(Not run)