helm values schema json plugin

Helm plugin for generating values.schema.json from single or multiple values files. Schema can be enriched by reading annotations from comments. Works only with Helm3 charts.

Installation

helm plugin install https://github.com/losisin/helm-values-schema-json.git

Upgrading

helm plugin update schema

See changelogs:

• Breaking changes
• Full release notes in GitHub Releases

Features

• Add multiple values files and merge them together - default is values.yaml in the current working directory
• Save output with custom name and location - default is values.schema.json in current working directory
• Use preferred schema draft version - default is draft 2020
• Read annotations from comments.
• Read description from helm-docs
• Bundling subschemas referenced in $ref

See docs for more info or checkout example yaml files in testdata.

Integrations

There are several ways to automate schema generation with this plugin. Main reason is that the json schema file can be hard to follow and we as humans tend to forget and update routine tasks. So why not automate it?

GitHub actions

There is GitHub action that I've build using typescript and published on marketplace. You can find it here. Basic usage is as follows:

name: Generate values schema json
on:
  - pull_request
jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v4
      with:
        ref: ${{ github.event.pull_request.head.ref }}
      - name: Generate values schema json
        uses: losisin/helm-values-schema-json-action@v1
        with:
          input: values.yaml

pre-commit hook

With pre-commit, you can ensure your JSON schema is kept up-to-date each time you make a commit.

First install pre-commit and then create or update a .pre-commit-config.yaml in the root of your Git repo with at least the following content:

repos:
  - repo: https://github.com/losisin/helm-values-schema-json
    rev: v1.7.2
    hooks:
      - id: helm-schema
        args: ["--values", "values.yaml"]

Then run:

pre-commit install
pre-commit install-hooks

Further changes to your chart files will cause an update to json schema when you make a commit.

Husky

This is a great tool for adding git hooks to your project. You can find it's documentation here. Here is how you can use it:

"husky": {
  "hooks": {
    "pre-commit": "helm schema --values values.yaml"
  }
},

CI/CD fail-on-diff

You can use this plugin in your CI/CD pipeline to ensure that the schema is always up-to-date. Here is an example for GitLab #82:

schema-check:
  script:
    - cd path/to/helm/chart
    - helm schema -output generated-schema.json
    - CURRENT_SCHEMA=$(cat values.schema.json)
    - GENERATED_SCHEMA=$(cat generated-schema.json)
    - |
      if [ "$CURRENT_SCHEMA" != "$GENERATED_SCHEMA" ]; then
        echo "Schema must be re-generated! Run 'helm schema' in the helm-chart directory" 1>&2
        exit 1
      fi

Usage

$ helm schema -help
Usage:
  helm schema [flags]

Flags:
      --bundle                              Bundle referenced ($ref) subschemas into a single file inside $defs
      --bundle-root string                  Root directory to allow local referenced files to be loaded from (default current working directory)
      --bundle-without-id                   Bundle without using $id to reference bundled schemas, which improves compatibility with e.g the VS Code JSON extension
      --config string                       Config file for setting defaults. (default ".schema.yaml")
      --draft int                           Draft version (4, 6, 7, 2019, or 2020) (default 2020)
  -h, --help                                help for helm
      --indent int                          Indentation spaces (even number) (default 4)
      --k8s-schema-url string               URL template used in $ref: $k8s/... alias (default "https://raw.githubusercontent.com/yannh/kubernetes-json-schema/refs/heads/master/{{ .K8sSchemaVersion }}/")
      --k8s-schema-version string           Version used in the --k8s-schema-url template for $ref: $k8s/... alias
      --no-additional-properties            Default additionalProperties to false for all objects in the schema
  -o, --output string                       Output file path (default "values.schema.json")
      --schema-root.additional-properties   Allow additional properties
      --schema-root.description string      JSON schema description
      --schema-root.id string               JSON schema ID
      --schema-root.ref string              JSON schema URI reference. Relative to current working directory when using "-bundle true".
      --schema-root.title string            JSON schema title
      --use-helm-docs                       Read description from https://github.com/norwoodj/helm-docs comments
  -f, --values strings                      One or more YAML files as inputs. Use comma-separated list or supply flag multiple times (default [values.yaml])

Configuration file

Uses .schema.yaml in the current working directory. Example:

# .schema.yaml

values:
  - values.yaml

draft: 2020
indent: 4
output: values.schema.json

bundle: false
bundleRoot: ""
bundleWithoutID: false

useHelmDocs: false

noAdditionalProperties: false

schemaRoot:
  id: https://example.com/schema
  title: Helm Values Schema
  description: Schema for Helm values
  additionalProperties: true

All options available from CLI can be set in this file. However, do note that the file uses camelCase, while the flags uses kebab-case.

Then, just run the plugin without any arguments:

helm schema

You can override which config file to use with the --config flag:

helm schema --config ./my-helm-schema-config.yaml

CLI

Basic

In most cases you will want to run the plugin with default options:

$ helm schema

This will read values.yaml, set draft version to 2020-12 and save outpout to values.schema.json.

Extended

Multiple values files

Merge multiple values files, set json-schema draft version explicitly and save output to my.schema.json:

values_1.yaml

nodeSelector:
  kubernetes.io/hostname: ""
dummyList:
  - "a"
  - "b"
  - "c"
key1: "asd"
key2: 42
key3: {}
key4: []

custom/path/values_2.yaml

nodeSelector:
  kubernetes.io/hostname: "node1"
deep:
  deep1:
    deep2:
      deep3:
        deep4: "asdf"

Run the following command to merge the yaml files and output json schema:

helm schema --values values_1.yaml,custom/path/values_2.yaml --draft 7 --output my.schema.json

Output will be something like this:

{
    "$schema": "http://json-schema.org/draft-07/schema#",
    "type": "object",
    "properties": {
        "deep": {
            "type": "object",
            "properties": {
                "deep1": {
                    "type": "object",
                    "properties": {
                        "deep2": {
                            "type": "object",
                            "properties": {
                                "deep3": {
                                    "type": "object",
                                    "properties": {
                                        "deep4": {
                                            "type": "string"
                                        }
                                    }
                                }
                            }
                        }
                    }
                }
            }
        },
        "dummyList": {
            "type": "array",
            "items": {
                "type": "string"
            }
        },
        "key1": {
            "type": "string"
        },
        "key2": {
            "type": "integer"
        },
        "key3": {
            "type": "object"
        },
        "key4": {
            "type": "array"
        },
        "nodeSelector": {
            "type": "object",
            "properties": {
                "kubernetes.io/hostname": {
                    "type": "string"
                }
            }
        }
    }
}

Note

When using multiple values files as input, the plugin follows Helm's behavior. This means that if the same yaml keys are present in multiple files, the latter file will take precedence over the former. The same applies to annotations in comments. Therefore, the order of the input files is important.

Root JSON object properties

Adding ID, title and description to the schema:

basic.yaml

image:
  repository: nginx
  tag: latest
  pullPolicy: Always

helm schema --values values.yaml --schema-root.id "https://example.com/schema" --schema-root.ref "schema/product.json" -schema-root.title "My schema" --schema-root.description "This is my schema"

Generated schema will be:

{
    "$id": "https://example.com/schema",
    "$ref": "schema/product.json",
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "additionalProperties": true,
    "description": "This is my schema",
    "properties": {
        "image": {
            "properties": {
                "pullPolicy": {
                    "type": "string"
                },
                "repository": {
                    "type": "string"
                },
                "tag": {
                    "type": "string"
                }
            },
            "type": "object"
        }
    },
    "title": "My schema",
    "type": "object"
}

Issues, Features, Feedback

Your input matters. Feel free to open issues for bugs, feature requests, or any feedback you may have. Check if a similar issue exists before creating a new one, and please use clear titles and explanations to help understand your point better. Your thoughts help me improve this project!

How to Contribute

🌟 Thank you for considering contributing to my project! Your efforts are incredibly valuable. To get started:

1. Fork the repository.
2. Create your feature branch: git checkout -b feature/YourFeature
3. Commit your changes: git commit -am 'Add: YourFeature'
4. Push to the branch: git push origin feature/YourFeature
5. Submit a pull request! 🚀