Skip to main content

Configuration

Environment variables​

  • LINTNET_CONFIG: Configuration file path
  • LINTNET_ERROR_LEVEL: debug|info|warn|error
  • LINTNET_SHOWN_ERROR_LEVEL: debug|info|warn|error
  • LINTNET_OUTPUT_SUCCESS: true|false
  • LINTNET_LOG_LEVEL: trace|debug|info|warn|error|fatal|panic
  • LINTNET_LOG_COLOR: auto|always|never
  • LINTNET_GITHUB_TOKEN: GitHub Access Token for getting Modules
  • LINTNET_ROOT_DIR: Root directory where modules are installed
  • GITHUB_TOKEN: GitHub Access Token for getting Modules

Configuration file path​

lintnet reads a configuration file ^\.?lintnet\.jsonnet$ on the current directory. You can also specify the configuration file path by the command line option --config (-c) and the environment variable LINTNET_CONFIG.

lintnet -c foo.yaml lint

Scaffold a configuration file​

You can scaffold the configuration file by lintnet init command.

lintnet init

Configuration file format​

JSON Schema

The configuration is a function returning an object.

function(param) {
  // targets is a list of lint configuration.
  targets: [
    //
  ],
  // ignored_dirs is a list of ignored directories.
  // ignored_dirs is optional.
  // The default value is [".git", "node_modules"].
  ignored_dirs: [
    ".git",
    "node_modules",
  ],
  // outputs is a list of output configuration.
  // outputs is optional.
  // outputs is used to customize output format.
  outputs: [
    // ...
  ],
}

.targets​

The element of targets is a pair of lint files and data files. data_files is required. Either lint_files or modules is required. Both lint_files and modules can also be used.

{
  id: 'target id', // optional
  base_data_path: '', // optional
  // data_files is a list of glob patterns.
  data_files: [
    'examples/**/*.csv', // relative path from the configuration file
    // ...
  ],
  // lint_files is a list of local lint files.
  lint_files: [
    'main.jsonnet', // relative path from the configuration file
  ],
  // modules is a list of modules.
  modules: [
    'github_archive/github.com/suzuki-shunsuke/example-lintnet-modules/newline.jsonnet@32ca3be646ec5b5861aab72fed30cd71f6eba9bf:v0.1.2',
  ],
}

.targets[].lint_files​

An element of lint_files is either a string or an object. If you want to pass configuration to lint rules, you need to use an object. Otherwise, a string is enough.

[
  'main.jsonnet',
  {
    path: 'examples/lint/filename.jsonnet', // required
    config: {}, // optional. lint rule configuration
  },
],

.targets[].modules​

An element of modules is either a string or an object.

[
  // github_archive/github.com/<repo owner>/<repo name>/<path>@<full commit hash>[:<version>]
  // version is optional.
  'github_archive/github.com/lintnet-modules/ghalint/workflow/action_ref_should_be_full_length_commit_sha/main.jsonnet@00571db321e413d45be457f39e48cd4237399bb7:v0.3.0',
  {
    // path is required
    path: 'github_archive/github.com/lintnet-modules/ghalint/workflow/action_ref_should_be_full_length_commit_sha/main.jsonnet@00571db321e413d45be457f39e48cd4237399bb7:v0.3.0',
    config: {},
  },
  {
    path: 'github_archive/github.com/lintnet-modules/ghalint@00571db321e413d45be457f39e48cd4237399bb7:v0.3.0',
    // You can specify file paths in a module with the attribute files.
    // This style is useful to specify multiple file path patterns in a module and set config parameter by lint rule
    files: [
      'workflow/**/main.jsonnet',
      '!workflow/action_ref_should_be_full_length_commit_sha/main.jsonnet',
      {
        path: 'workflow/action_ref_should_be_full_length_commit_sha/main.jsonnet',
        config: {},
      },
    ],
  },
],

.targets[].base_data_path​

base_data_path is useful to lint files by service directory in a Monorepo.

{
  // data files which are on the same directory as tfaction.yaml.
  base_data_path: '**/tfaction.yaml',
  data_files: [
    // relative path from base_data_path
    // Glob is also available
    '*.tf',
  ],
  // ...
},

In case of linting across multiple files, base_data_path is useful to separate files. In the above case, if **/tfaction.yaml matches foo/tfaction.yaml and bar/tfaction.yaml, foo/*.tf and bar/*.tf are linted separately.

.outputs​

Please see Customize Output.

File paths in configuration files​

The file path separator must be a slash /. If file paths are relative paths, the base must be the configuration file.

Top level argument​

Now the top level argument param is empty. This argument is reserved for future enhancement.

Glob​

data_files and lint_files are lists of patterns matching with data and lint files. Each string is parsed with bmatcuk/doublestar.

Exclude by !​

If each string starts with !, files matching with the pattern are excluded.

e.g. foo/example.jsonnet is excluded

**/*.jsonnet
!foo/example.jsonnet

e.g. foo/example.jsonnet isn't excluded because the later pattern foo/*.jsonnet takes precedence

**/*.jsonnet
!foo/example.jsonnet
foo/*.jsonnet

Excluded directories and files​

lintnet doesn't check .gitignore. lintnet ignores the following directories.

  • .git
  • node_modules

And in lint_files, files *_test.jsonnet are ignored.

See also​