Skip to main content

Pre-commit integration

Pre-commit blocks commits that introduce rule violations. Ignition Lint provides a pre-commit-compatible hook out of the box; you choose between two install modes.

Add to .pre-commit-config.yaml:

repos:
  - repo: https://github.com/bw-design-group/ignition-lint
    rev: v0.2.4  # use the latest release tag
    hooks:
      - id: ign-lint

Install hooks:

pre-commit install

The hook automatically runs on view.json files. It uses ignition-lint's bundled .ignition-lint-precommit.json (warnings-favored config) by default.

:::note How files reach the hook Pre-commit passes the staged view.json paths to the hook as positional arguments (pass_filenames: true). Do not add --files to the hook's args — --files is for glob mode (a single pattern you run by hand), whereas pre-commit hands the hook an explicit list of files. Ending args with --files makes the first staged file bind to --files and is deprecated; ignition-lint will warn about it. Just omit --files and let pass_filenames do its job. :::

Pros: zero setup, version-pinned via rev:, consistent across the team Cons: clones the full repo (~64 MB) into the pre-commit cache

Option B: Lightweight install (~1 MB)​

Install only the Python package — no test fixtures, docker files, or docs in the cache:

repos:
  - repo: local
    hooks:
      - id: ign-lint
        name: Ignition Lint
        entry: ign-lint
        language: python
        types: [json]
        files: view\.json$
        args: ['--config=rule_config.json']
        pass_filenames: true
        additional_dependencies:
          - 'git+https://github.com/bw-design-group/ignition-lint@v0.2.4'

Pros: ~1 MB cache footprint Cons: more YAML to maintain

Custom configuration​

Both options accept a custom rule_config.json:

repos:
  - repo: https://github.com/bw-design-group/ignition-lint
    rev: v0.2.4
    hooks:
      - id: ign-lint
        args: ['--config=rule_config.json']

A common pattern is one config for pre-commit (warnings-favored, fast) and a separate one for CI (strict, full):

project/
├── .pre-commit-config.yaml         # references pre-commit-config.json
├── pre-commit-config.json          # warning severity, lighter rule set
└── rule_config.json                # full strictness, used in CI

Warnings vs errors​

Default behavior: both warnings and errors block commits.

To allow commits with warnings (block only on errors):

hooks:
  - id: ign-lint
    args: ['--config=rule_config.json', '--ignore-warnings']

This is useful when rolling out a new rule — set it to warning, let teams adapt, then promote to error.

Whitelist for legacy code​

Pair pre-commit with a whitelist to exempt legacy files:

hooks:
  - id: ign-lint
    args:
      - '--config=rule_config.json'
      - '--whitelist=.whitelist.txt'

Generate the whitelist once and commit it:

ign-lint --generate-whitelist "views/legacy/**/*.json"
git add .whitelist.txt
git commit -m "Add whitelist for legacy views (technical debt)"

See Whitelist guide for full details.

Excluding test files​

Test fixtures often contain intentional violations. Exclude them:

hooks:
  - id: ign-lint
    exclude: '^tests/.*|.*test.*\.json$'

Running manually​

# Install hooks (one-time)
pre-commit install

# Run on all files (full repository scan)
pre-commit run --all-files

# Run on staged files only (default behavior)
pre-commit run

# Run only the ignition-lint hook
pre-commit run ign-lint

# Run on specific files
pre-commit run ign-lint --files path/to/view.json

Caveat: --all-files on large repositories​

pre-commit run --all-files passes every matched file as a CLI argument. With hundreds of view files and long paths, this can exceed ARG_MAX on some systems. For full-repository scans, invoke ignition-lint directly:

ign-lint --files "views/**/view.json" --config rule_config.json

The CLI's internal globber sidesteps the argv limit entirely.

Skipping the hook (one-off)​

Sometimes you need to commit despite a known issue:

git commit --no-verify

Use sparingly — --no-verify skips ALL pre-commit hooks, not just ignition-lint.

Setting up a custom configuration​

Sample pre-commit-config.json favoring warnings:

{
  "NamePatternRule": {
    "enabled": true,
    "kwargs": {
      "convention": "PascalCase",
      "severity": "warning"
    }
  },
  "PollingIntervalRule": {
    "enabled": true,
    "kwargs": {
      "minimum_interval": 5000
    }
  },
  "PylintScriptRule": {
    "enabled": true,
    "kwargs": {
      "category_mapping": {
        "F": "error",
        "E": "error",
        "W": "warning",
        "C": "warning",
        "R": "warning"
      }
    }
  }
}

See also​