In RDocTaskForce/testextra: Extract Test Blocks

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

testextra

The goal of testextra is to facilitate extraction of tests embedded in source code.

Installation

You will be able to install the released version of testextra from CRAN once it is released with:

install.packages("testextra")

Until then or if you wish to get the latest version prior to release you may install directly from GitHub with:

remotes::install_github("RDocTaskForce/testextra")

Including Tests in Source Code.

To include tests in source code put your tests following the function definition nested in an if(FALSE){...} block and tag it with a #@testing block tag.

#' Hello World Example
hello_world <- function(){
    message("hello world!")
}
if(FALSE){#@testing
    expect_message(hello_world(), "hello world!")
}

Assuming the preceding code is in a package file ./R/hello_world.R running the extract_tests() command will create a file ./tests/testthat/test-hello_world.R with the contents as below.

#! This file was automatically produced by the testextra package.
#! Changes will be overwritten.

context('tests extracted from file `hello_world.R`')
#line 5 "./R/hello_world.R"
test_that('hello_world', {#@testing
    expect_message(hello_world(), "hello world!")
})

When run, if there are error messages, the line given will be the line and file from the original source code.

Combination Functions

The functions test() and test_file() provided in testextra will both extract tests from source files, run said tests, and output the results. test() operates on a package as a whole or a subset of a package by
setting the filter argument, see the help file for details. test_file() is intended to work with the RStudio GUI. It takes the currently selected file, extracts tests and runs the tests. This way a tests may be run only for the current file being evaluated.

Both test() and test_file() are available through the add-ins, and made accessible through the menu of RStudio.

Other Helpers

The 'testextra` package provides a number of useful testing functions to use when testing code.

Inheritance

• all_inherit() - tests if all elements of a list are of the given class or classes.
• are() - similar to all_inherit(), however uses the is() mechanism which is more appropriate for S4 classes.
• is_exactly() - Tests if an object is a class, but disallows inheritance.
• all_are_exactly() - The is_exactly() test mapped over a list of objects.

Strings

• is_nonempty_string() - similar to asssertthat::is.string() but also ensures that the provided string is not missing (NA) and not empty ("")
• is_optional_string() - same as is_nonempty_string() except does allow a character vector of length 0.

Validity

• is_valid() - Performs validObject() in a manner that is compatible with
  validate_that(), assert_that(), or see_if() from the assertthat package.
• are_valid() - is_valid() over a list, which when used with the functions listed above gives the indices of objects that are not valid.
• expect_valid() - Check validity which is to be used with the testthat framework.

Namespaces

When testing dynamic class creation and modification, it is often necessary to have a package environment other that the package environment in which the creation functions are defined. For this purpose, testextra provides these namespace manipulation functions.

• new_namespace_env() - Create a namespace environment. *Similar functionality exists in pkgload, but is not exposed and registers the namespace by default, which new_namespace_env() does not.
• new_pkg_environment() - Create a package environment. Technically a namespace does not have to be a package environment, however that is essentially always the case. This function does allow for registration of the environment as a namespace but does not do so by default.
• register_namespace - Explicitly register a previously created namespace.
• is_namespace_registered - Check if a namespace is registered.

Others

A few other helpers that do not fit into one of the above categories.

• catch_condition() - Evaluates code and captures any signals that may be raised. Useful for capturing and subsequently running multiple tests on the error captured, as an alternative to expect_error(), expect_warning(), and expect_message() from the testthat package.
• class0 - retrieve the class of an object as a single string. Separates elements by a '/' if there are more than one. Same functionality as the knitr::klass() function.
• is_valid_regex - Check if a regular expression is valid, not that it does what is intended just that it is valid.

Documentation

The testextra package is developed by the R Documentation Task Force, an R Consortium Infrastructure Steering Committee working group.

RDocTaskForce/testextra documentation built on Feb. 5, 2020, 11:05 p.m.