codecoverage

Code coverage, or test coverage, is a measure of how much of a software program’s source code is executed when a given set of tests runs. It is used to assess the thoroughness of testing, identify untested code paths, and guide test design. Coverage data helps teams evaluate risk and allocate testing effort more effectively.

The most common coverage types are statement (line) coverage, which records whether each executable statement runs; branch (decision) coverage, which checks whether both sides of conditional branches are exercised; path coverage, which attempts to execute all feasible execution paths; and condition or loop coverage, which focuses on boolean subexpressions and iterations. Coverage can be obtained through instrumentation (modifying the program to report executions) or through sampling techniques.

Coverage is usually reported as percentages, often at the file, function, and project levels. Popular tools include JaCoCo and Cobertura for Java, Istanbul/nyc for JavaScript, Coverage.py for Python, gcov for C/C++, and Coverlet for .NET. Reports may show per-file metrics, untested branches, and hotspots that deserve attention.

Limitations should be acknowledged. High coverage does not guarantee correctness, and 100% coverage can be costly or reveal brittle tests that only confirm implementation details. Code may be covered without testing real user scenarios. Effective use combines coverage with other quality measures, focuses on critical features, and may include techniques such as mutation testing and risk-based test design. Integration with CI pipelines is common to track trends over time.