Codedekking

Codedekking is the measure of how much of a program's source code is executed during automated tests. It is typically expressed as a percentage and can be reported at different granularity levels, such as per statement, per branch, or per module. The primary purpose of codedekking is to reveal parts of the codebase that are not exercised by tests, helping teams assess the thoroughness of their test suites and identify gaps that may hide defects.

Common types include statement (line) coverage, which tracks whether each executable line runs; branch coverage, which checks whether each outcome of a conditional has been executed; path coverage, which explores all feasible execution paths; condition coverage, which focuses on boolean subexpressions; and loop coverage, which exercises loops with various counts. Many projects report multiple metrics to obtain a fuller picture.

Codedekking is usually measured by instrumentation or by sampling during test execution. Tools emit a report showing the percentage of code covered and often highlight uncovered lines. Popular tools include coverage.py for Python, JaCoCo and Cobertura for Java, Istanbul (nyc) for JavaScript, and Coverlet for .NET, among others. Integrations with continuous integration pipelines enable monitoring of coverage trends over time.

Limitations and caveats: high codedekking does not guarantee correct software; untested code may be dead or irrelevant, and tests can be poorly designed or superficial. Coverage should be used alongside other quality measures, including mutation testing, static analysis, and manual reviews, to assess overall software quality.