A test suite is a collection of automated tests evaluated together for a software system or component. Its purpose is to support sustainable growth by catching regressions, enabling refactoring, and preserving development pace without imposing excessive maintenance cost. section 1.0 section 1.6
Build a successful suite by ensuring that it:
- Runs as part of the development cycle, ideally after every code change. section 1.4
- Concentrates on the most important parts of the codebase, especially business logic and the domain model. section 1.4
- Retains tests whose benefits exceed their implementation and maintenance costs. section 1.6
Coverage is not quality
Treat low coverage as a warning, not high coverage as proof. Branch coverage can reveal more completeness than code coverage, but neither accounts for assertions or proves that the suite protects important behavior. Imposing a target percentage can therefore reward superficial tests. section 1.6
Evaluate tests individually and apply judgment rather than relying on an automated quality score. A valuable suite protects observable behavior while avoiding fragile, low-value tests and unnecessary coverage of less important code. section 1.4 section 4.0
