Skip to content

Measure the churn/complexity ratio. Higher values mean hotspots where refactorings should happen.

License

Notifications You must be signed in to change notification settings

simonrenoult/code-complexity

Repository files navigation

code-complexity

Measure the churn/complexity score. Higher scores reveal hotspots where refactorings should happen.

Build Status Coverage Status Style Status NPM Version

Quoting Michael Feathers (source here):

Often when we refactor, we look at local areas of code. If we take a wider view, using information from our version control systems, we can get a better sense of the effects of our refactoring efforts.

Note: code-complexity currently measures complexity using either:

  • lines of code count (all languages)
  • cyclomatic complexity (JavaScript/TypeScript)
  • halstead complexity (JavaScript/TypeScript)

Usage

$ npx code-complexity <path-to-git-directory or URL> [options]

Help

    Usage: code-complexity <target> [options]

    Measure the churn/complexity score. Higher values mean hotspots where refactorings should happen.

    Options:
      -V, --version                         output the version number
      --filter <strings>                    list of globs (comma separated) to filter
      -cs, --complexity-strategy [strategy] choose the complexity strategy to analyze your codebase with (allowed values: sloc, cyclomatic, halstead).
      -f, --format [format]                 format results using table, json or csv
      -l, --limit [limit]                   limit the number of files to output
      -i, --since [since]                   limit analysis to commits more recent in age than date
      -u, --until [until]                   limit analysis to commits older in age than date
      -s, --sort [sort]                     sort results (allowed valued: score, churn, complexity or file)
      -d, --directories                     display values for directories instead of files
      -h, --help                            display help for command

    Examples:

    $ code-complexity .
    $ code-complexity https://github.com/simonrenoult/code-complexity
    $ code-complexity foo --limit 3
    $ code-complexity ../foo --sort score
    $ code-complexity /foo/bar --filter 'src/**,!src/front/**'
    $ code-complexity . --limit 10 --sort score
    $ code-complexity . --limit 10 --directories 
    $ code-complexity . --limit 10 --sort score -cs halstead
    $ code-complexity . --since=2021-06-01 --limit 100
    $ code-complexity . --since=2021-04-01 --until=2021-07-01

Output

$ npx code-complexity https://github.com/simonrenoult/code-complexity --sort=score --limit=3

┌──────────────────────────────┬────────────┬───────┬───────┐
│ file                         │ complexity │ churn │ score │
├──────────────────────────────┼────────────┼───────┼───────┤
│ src/cli.ts                   │ 103        │ 8     │ 824   │
├──────────────────────────────┼────────────┼───────┼───────┤
│ test/code-complexity.test.ts │ 107        │ 7     │ 749   │
├──────────────────────────────┼────────────┼───────┼───────┤
│ .idea/workspace.xml          │ 123        │ 6     │ 738   │
└──────────────────────────────┴────────────┴───────┴───────┘

Special thanks

A special thanks to a few contributors that helped me make code-complexity better.

  • Alexander Dormann (alexdo) for fixing the ENOBUFS (and apologies for stealing your code).
  • Scott Brooks (scottamplitude) for initiating the work on complexity strategies

About

Measure the churn/complexity ratio. Higher values mean hotspots where refactorings should happen.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published