How do I set up testthat for an R package?

Use usethis::use_test() to create test files and infrastructure automatically, or manually configure test directories and run tests with devtools::test(). The README recommends this for easy integration.

testthat vs RUnit: which is better for R testing?

testthat is more modern and widely adopted, with better integration into RStudio and CRAN workflows, while RUnit is older and less actively maintained. testthat's visual feedback and expectation functions make it the preferred choice for most R developers.

How to test error messages and warnings in R with testthat?

Use expect_error() and expect_warning() functions to capture and assert specific conditions in your code, as described in the README's overview of expectation functions.

Can testthat be used for non-package R projects like scripts?

Yes, but it's optimized for package development; for scripts, you might need to manually set up test environments, and the workflow integration with R CMD check is less straightforward.

What are best practices for organizing tests in testthat?

Follow the convention of placing tests in a 'tests/testthat' directory, with each test file matching the source file name, and use descriptive test contexts to group related expectations, as advised in the R Packages book referenced in the README.

How to integrate testthat with continuous integration (CI) pipelines?

Leverage the GitHub Actions workflow shown in the badges, or use R CMD check in CI tools; testthat's compatibility with standard R testing commands makes it easy to automate in environments like Travis CI or GitLab CI.

Open-Awesome

testthat <img class="emoji" alt="heart" src="https://cdn.jsdelivr.net/gh/qinwf/awesome-R@3c66da6e291bcc0520b1649125b0bed750896a9a/heart.png" height="20" align="absmiddle" width="20">

NOASSERTIONRv3.3.2

An R package for unit testing that makes writing and running tests enjoyable and integrated into your workflow.

Visit Website GitHub

930 stars341 forks0 contributors

What is testthat <img class="emoji" alt="heart" src="https://cdn.jsdelivr.net/gh/qinwf/awesome-R@3c66da6e291bcc0520b1649125b0bed750896a9a/heart.png" height="20" align="absmiddle" width="20">?

testthat is an R package for unit testing that provides functions to describe expected behavior of code, including handling errors, warnings, and messages. It solves the problem of painful and tedious testing by making it fun and integrated into existing R workflows, helping developers improve code quality.

Target Audience

R developers and package maintainers who need a robust and widely-adopted testing framework to ensure their code works correctly, especially those building CRAN packages.

Value Proposition

Developers choose testthat because it is the most popular unit testing package for R, offering a comprehensive and enjoyable testing experience with strong community support and seamless integration into R development workflows.

Overview

An R 📦 to make testing 😀

Use Cases

Best For

Writing unit tests for R packages
Ensuring code quality in CRAN submissions
Integrating testing into RStudio workflows
Catching errors, warnings, and messages in R functions
Learning modern testing practices in R
Automating test suites for R projects

Not Ideal For

Projects written in languages other than R, as testthat is specifically designed for R code
Testing R Shiny applications with complex user interface interactions, where browser-based testing tools might be more appropriate
Teams needing built-in advanced mocking or stubbing capabilities without additional dependencies, as testthat requires extra packages for comprehensive isolation

Pros & Cons

Pros

Comprehensive Expectation Functions

Provides easy-to-use functions like expect_error() and expect_warning() to test code behavior, including catching errors, warnings, and messages, as highlighted in the README.

Seamless Workflow Integration

Integrates smoothly with R CMD check and RStudio, allowing tests to be run with commands like devtools::test() or Ctrl+Shift+T in RStudio, as mentioned in the usage section.

Visual Test Feedback

Displays color-coded pass/fail/error status in terminals and RStudio, making test results immediately interpretable and enhancing the testing experience.

Wide Community Adoption

Used by thousands of CRAN packages, ensuring it's a reliable standard with extensive community support and regular updates.

Cons

Initial Setup Complexity

Requires configuration with usethis or manual setup for optimal use, which can be daunting for beginners not familiar with R package development workflows.

Limited Built-in Mocking

Lacks advanced mocking features out-of-the-box; users often need additional packages like mockery for comprehensive test isolation, increasing dependency management.

R-Specific Ecosystem Lock-in

Tied closely to R and its tools, making it less suitable for polyglot projects or environments where cross-language testing frameworks are preferred.

Frequently Asked Questions

Related Projects

devtools <img class="emoji" alt="heart" src="https://cdn.jsdelivr.net/gh/qinwf/awesome-R@3c66da6e291bcc0520b1649125b0bed750896a9a/heart.png" height="20" align="absmiddle" width="20">

Tools to make an R developer's life easier

Stars2,513

Forks763

Last commit1 month ago

lintr

Static Code Analysis for R

Stars1,279

Forks198

Last commit2 days ago

renv <img class="emoji" alt="heart" src="https://cdn.jsdelivr.net/gh/qinwf/awesome-R@3c66da6e291bcc0520b1649125b0bed750896a9a/heart.png" height="20" align="absmiddle" width="20">

renv: Project environments for R.

Stars1,154

Forks165