Code Metrics - CLI

Computes complexity in TypeScript / JavaScript files.

Complexity calculation

The steps of the calculation:

• create an AST from the input source file
• walk through each and every node of it
• depending on the type of the node and the configuration associated with it create a new entry about the node. This entry contains everything necessary for further use (e.g. a textual representation for the node, complexity increment, child nodes etc.)
• show the sum of complexity of child nodes for methods and the maximum of child nodes for classes

Please note that it is not a standard metric, but it is a close approximation of Cyclomatic complexity.

Please also note that it is possible to balance the complexity calculation for the project / team / personal taste by adjusting the relevant configuration entries.

For example if one prefers guard clauses, and is ok with all the branches in switch statements then the following could be applied:

// codemetrics.config.js
// @ts-check
/** @type {import("codemetrics-cli").MetricsConfiguration}*/
var config = {};
config.ReturnStatement = 0;
config.CaseClause = 0;
config.DefaultClause = 0;

module.exports = config;

Configuration

The computation can be configured by default via the codemetrics.config.js file which is looked up automatically in the current and in the parent directories or by specifying the config file (with a relative or absolute path) via the --config parameter.

The file must have a single export, the configuration itself for which typescript type definitions are also provided. (See the example above)

Parameters

  -t, --threshold <number>  Minimum complexity to be shown
  -c, --config <string>     Path to the config file
  -l, --lint                Exit with non-zero on issues
  -d  --deep                Show complexity results from top level nodes, rather than an aggregation per file
  -p, --pattern <string>    Glob pattern

You can also use codemetrics-cli --help to get the list of the possible options.

Contributions

• Add lint option to fail output when threshold not met - by @kopach
• Introduce deep complexity comparison (per top level function) - by @kopach