NLOC

NLOC, or Number of Lines of Code, is a software size metric used to estimate the size of a codebase by counting lines of source code. Counting rules vary: some methods exclude blank lines and comments, while others count them; some count only executable statements, and others count every line that contains code. Because counting conventions differ across languages and projects, NLOC is not directly comparable without a defined counting standard.

In practice, NLOC supports rough sizing for planning, effort estimation, budgeting, and historical benchmarking. It is often used alongside language-aware variants (such as SLOC) and normalized measures. It should be interpreted as a proxy for effort or scale rather than a direct indicator of quality or functionality.

Limitations include its sensitivity to language verbosity and coding style, and its inability to reflect code complexity, architecture, or maintainability. Generated code or boilerplate can inflate NLOC, while highly abstracted designs may reduce lines without reducing effort. Therefore, NLOC should be used with clear counting rules and in combination with other metrics.

Counting approaches may differ by granularity (project-wide, module, file) and by whether comments and blank lines are included. Tools such as cloc, SLOCCount, and language-aware analyzers automate counting and can export standardized reports. It is best practice to document the counting methodology when reporting NLOC figures.

Related metrics include cyclomatic complexity, function points, and code churn, which, together with NLOC, provide a more complete view of software size, complexity, and change rate.