Reusing workflow configurations

Reusable workflows

Reference information for reusable workflows.

Access to reusable workflows

A reusable workflow can be used by another workflow if any of the following is true:

Both workflows are in the same repository.
The called workflow is stored in a public repository, and your enterprise allows you to use public reusable workflows.
The called workflow is stored in an internal repository and the settings for that repository allow it to be accessed. For more information, see アクションとワークフローを企業と共有する.
The called workflow is stored in a private repository and the settings for that repository allow it to be accessed. For more information, see アクションとワークフローを企業と共有する.

The following table shows the accessibility of reusable workflows to a caller workflow, depending on the visibility of the host repository.

Caller repository	Accessible workflows repositories
`private`	`private`, `internal`, and `public`
`internal`	`internal`, and `public`
`public`	`public`

The Actions permissions on the callers repository's Actions settings page must be configured to allow the use of actions and reusable workflows - see リポジトリの GitHub Actions の設定を管理する.

For internal or private repositories, the Access policy on the Actions settings page of the called workflow's repository must be explicitly configured to allow access from repositories containing caller workflows - see リポジトリの GitHub Actions の設定を管理する.

メモ

セキュリティを強化するため、GitHub Actions はアクションまたは再利用可能なワークフローのリダイレクトをサポートしません。つまり、所有者、アクションのリポジトリの名前、またはアクションの名前が変更されると、そのアクションを以前の名前で使用するすべてのワークフローは失敗します。

Limitations of reusable worklows

You can connect up to four levels of workflows. For more information, see Nesting reusable workflows.
You can call a maximum of 20 unique reusable workflows from a single workflow file. This limit includes any trees of nested reusable workflows that may be called starting from your top-level caller workflow file.

For example, top-level-caller-workflow.yml → called-workflow-1.yml → called-workflow-2.yml counts as 2 reusable workflows.
Any environment variables set in an env context defined at the workflow level in the caller workflow are not propagated to the called workflow. For more information, see 変数に情報を格納する and コンテキストリファレンス.
Similarly, environment variables set in the env context, defined in the called workflow, are not accessible in the env context of the caller workflow. Instead, you must use outputs of the reusable workflow. For more information, see Using outputs from a reusable workflow.
To reuse variables in multiple workflows, set them at the organization, repository, or environment levels and reference them using the vars context. For more information see 変数に情報を格納する and コンテキストリファレンス.
Reusable workflows are called directly within a job, and not from within a job step. You cannot, therefore, use GITHUB_ENV to pass values to job steps in the caller workflow.

Supported keywords for jobs that call a reusable workflow

When you call a reusable workflow, you can only use the following keywords in the job containing the call:

jobs.<job_id>.name
jobs.<job_id>.uses
jobs.<job_id>.with
jobs.<job_id>.with.<input_id>
jobs.<job_id>.secrets
jobs.<job_id>.secrets.<secret_id>
jobs.<job_id>.secrets.inherit
jobs.<job_id>.strategy
jobs.<job_id>.needs
jobs.<job_id>.if
jobs.<job_id>.concurrency
jobs.<job_id>.permissions
メモ
- If jobs.<job_id>.permissions is not specified in the calling job, the called workflow will have the default permissions for the GITHUB_TOKEN. For more information, see GitHub Actions　のワークフロー構文.
- The GITHUB_TOKEN permissions passed from the caller workflow can be only downgraded (not elevated) by the called workflow.
- If you use jobs.<job_id>.concurrency.cancel-in-progress: true, don't use the same value for jobs.<job_id>.concurrency.group in the called and caller workflows as this will cause the workflow that's already running to be cancelled. A called workflow uses the name of its caller workflow in ${{ github.workflow }}, so using this context as the value of jobs.<job_id>.concurrency.group in both caller and called workflows will cause the caller workflow to be cancelled when the called workflow runs.

How reusable workflows use runners

GitHub-hosted runners

The assignment of GitHub-hosted runners is always evaluated using only the caller's context. Billing for GitHub-hosted runners is always associated with the caller. The caller workflow cannot use GitHub-hosted runners from the called repository. For more information, see GitHub ホステッドランナー.

Self-hosted runners

Called workflows that are owned by the same user or organization or enterprise as the caller workflow can access self-hosted runners from the caller's context. This means that a called workflow can access self-hosted runners that are:

In the caller repository
In the caller repository's organization or enterprise, provided that the runner has been made available to the caller repository

Access and permissions for nested workflows

A workflow that contains nested reusable workflows will fail if any of the nested workflows is inaccessible to the initial caller workflow. For more information, see Access to reusable workflows.

GITHUB_TOKEN permissions can only be the same or more restrictive in nested workflows. For example, in the workflow chain A > B > C, if workflow A has package: read token permission, then B and C cannot have package: write permission. For more information, see ワークフローでの認証に GITHUB_TOKEN を使用する.

For information on how to use the API to determine which workflow files were involved in a particular workflow run, see ワークフローを再利用する.

Behavior of reusable workflows when re-running jobs

パブリックリポジトリの再利用可能なワークフローは、SHA、リリースタグ、またはブランチ名を使って参照できます。詳しくは、「ワークフローを再利用する」をご覧ください。

再利用可能なワークフローを使うワークフローを再実行し、参照が SHA ではない場合は、注意すべきいくつかの動作があります。

ワークフロー内のすべてのジョブを再実行すると、指定した参照の再利用可能なワークフローが使われます。ワークフロー内のすべてのジョブの再実行の詳細については、「ワークフローとジョブの再実行」を参照してください。
失敗したジョブまたはワークフロー内の特定のジョブを再実行すると、最初の試行と同じコミット SHA の再利用可能なワークフローが使われます。ワークフローで失敗したジョブを再実行する方法の詳細については、「ワークフローとジョブの再実行」を参照してください。ワークフロー内の特定のジョブの再実行の詳細については、「ワークフローとジョブの再実行」を参照してください。

`github` context

When a reusable workflow is triggered by a caller workflow, the github context is always associated with the caller workflow. The called workflow is automatically granted access to github.token and secrets.GITHUB_TOKEN. For more information about the github context, see コンテキストリファレンス.

Workflow templates

Reference information to use when creating workflow templates for your organization.

Workflow template availability

You can use templates in repositories that match or have more restricted visibility than the template repository.

Workflow templates in a public .github repository are available to all repository types.
Workflow templates in an internal .github repository are only available to internal and private repositories.
Workflow templates in a private .github repository are only available to private repositories.

Because public workflow templates require a public .github repository, they are not available for Enterprise Managed Users.

Granting access for private/internal repositories

If you're using a private or internal .github repository, you need to grant Read access to users or teams who should be able to use the templates.

The `$default-branch` placeholder

If you need to refer to a repository's default branch, you can use the $default-branch placeholder in your workflow template. When a workflow is created the placeholder will be automatically replaced with the name of the repository's default branch.

Example workflow template file

This file named octo-organization-ci.yml demonstrates a basic workflow.

YAML

name: Octo Organization CI
on:
  push:
    branches: [ $default-branch ]
  pull_request:
    branches: [ $default-branch ]
jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Run a one-line script
        run: echo Hello from Octo Organization

name: Octo Organization CI
on:
  push:
    branches: [ $default-branch ]
  pull_request:
    branches: [ $default-branch ]
jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Run a one-line script
        run: echo Hello from Octo Organization

Metadata file requirements

The metadata file must have the same name as the workflow file, but instead of the .yml extension, it must be appended with .properties.json. For example, this file named octo-organization-ci.properties.json contains the metadata for a workflow file named octo-organization-ci.yml:

JSON

{
    "name": "Octo Organization Workflow",
    "description": "Octo Organization CI workflow template.",
    "iconName": "example-icon",
    "categories": [
        "Go"
    ],
    "filePatterns": [
        "package.json$",
        "^Dockerfile",
        ".*\\.md$"
    ]
}

{
    "name": "Octo Organization Workflow",
    "description": "Octo Organization CI workflow template.",
    "iconName": "example-icon",
    "categories": [
        "Go"
    ],
    "filePatterns": [
        "package.json$",
        "^Dockerfile",
        ".*\\.md$"
    ]
}

name - Required. The name of the workflow. This is displayed in the list of available workflows.
description - Required. The description of the workflow. This is displayed in the list of available workflows.
iconName - Optional. Specifies an icon for the workflow that is displayed in the list of workflows. iconName can one of the following types:
- An SVG file that is stored in the workflow-templates directory. To reference a file, the value must be the file name without the file extension. For example, an SVG file named example-icon.svg is referenced as example-icon.
- An icon from GitHub's set of Octicons. To reference an octicon, the value must be octicon <icon name>. For example, octicon smiley.
categories - Optional. Defines the categories that the workflow is shown under. You can use category names from the following lists:
- General category names from the starter-workflows repository.
- Linguist languages from the list in the linguist repository.
- Supported tech stacks from the list in the starter-workflows repository.
filePatterns - Optional. Allows the workflow to be used if the user's repository has a file in its root directory that matches a defined regular expression.

YAML anchors and aliases

You can use YAML anchors and aliases to reduce repetition in your workflows. An anchor (marked with &) identifies a piece of content that you want to reuse, while an alias (marked with *) repeats that content in another location.

For detailed information about anchors and aliases, see Node Anchors and Aliases in the YAML specification.

Here's an example that uses YAML anchors and aliases with environment variables:

jobs:
  job1:
    env: &env_vars # Define the anchor on first use
      NODE_ENV: production
      DATABASE_URL: ${{ secrets.DATABASE_URL }}
    steps:
      - run: echo "Using production settings"

  job2:
    env: *env_vars # Reuse the environment variables
    steps:
      - run: echo "Same environment variables here"

This is equivalent to writing the following YAML without anchors and aliases:

jobs:
  job1:
    env:
      NODE_ENV: production
      DATABASE_URL: ${{ secrets.DATABASE_URL }}
    steps:
      - run: echo "Using production settings"

  job2:
    env:
      NODE_ENV: production
      DATABASE_URL: ${{ secrets.DATABASE_URL }}
    steps:
      - run: echo "Same environment variables here"

You can also use anchors for more complex configurations, such as reusing an entire job configuration:

jobs:
  test: &base_job # Define the anchor on first use
    runs-on: ubuntu-latest
    timeout-minutes: 30
    env:
      NODE_VERSION: '18'
    steps:
      - uses: actions/checkout@v4
      - name: Set up Node.js
        uses: actions/setup-node@v4
        with:
          node-version: ${{ env.NODE_VERSION }}
      - run: npm test

  alt-test: *base_job # Reuse the entire job configuration

この記事の内容