tdd-workflow

Test-driven development in R packages is enforced by writing failing tests with testthat before implementing code, targeting 80%+ coverage, and using testthat Edition 3 patterns to reduce regressions.

Structure unit and integration tests for R data pipelines using testthat Edition 3 patterns, ensuring tests are written before code with 80%+ coverage and stricter expectations for validation paths.

testthat works with snapshot testing for Shiny components in R by using Edition 3 modern expectation helpers to capture and verify UI outputs, enforcing test-first discipline to prevent refactoring regressions.

Code coverage for statistical calculations in R should target 80%+ overall, with 100% coverage expectations specifically for calculation and data validation paths to prevent regressions.

Start writing tests before code when fixing bugs in an R package by using usethis::use_test("your_function") to create a failing test first, then implement the smallest code change needed to pass.

Test-first development for refactoring R code is not suitable when existing test coverage is entirely absent and immediate hotfixes are required, as the workflow mandates writing failing tests before any implementation.