Convert malformed-yaml action to Python

.github/workflows/test-malformed-yaml.yml

@@ -0,0 +1,29 @@
+name: Test Malformed YAML Action
+on:
+  - pull_request
+
+defaults:
+  run:
+    working-directory: ./malformed-yaml
+
+jobs:
+  test:
+    name: Test Malformed YAML Action
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+      - name: Set up Python 3.12
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - name: Install Python dependencies
+        run: |
+          pip install --upgrade pip
+          pip install --no-cache-dir -r requirements.txt
+          pip install --no-cache-dir -r requirements-dev.txt
+      - name: Run tests with pytest
+        run: |
+          pytest -vv test/

.gitignore

@@ -1 +1,7 @@
-.built-image
+.built-image
+.DS_Store
+__pycache__/
+*.pytest_cache
+*venv
+*.venv
+*.vscode

malformed-yaml/Dockerfile

@@ -1,11 +1,11 @@
-FROM ministryofjustice/cloud-platform-tools:2.1
+FROM python:3.12-slim
 
-# Octokit depends on faraday, and an update to
-# faraday breaks the current version of octokit
-RUN gem install faraday --version 0.9
-RUN gem install octokit --version 4.21.0
+RUN apt-get update && apt-get upgrade -y && apt-get clean && rm -rf /var/lib/apt/lists/*
 
-COPY reject-malformed-yaml.rb /reject-malformed-yaml.rb
-COPY github.rb /github.rb
+WORKDIR /app
 
-ENTRYPOINT ["/reject-malformed-yaml.rb"]
+COPY github_pull_request.py main.py requirements.txt /app/
+
+RUN pip install --no-cache-dir -r requirements.txt
+
+ENTRYPOINT [ "python3", "/app/main.py" ]

malformed-yaml/README.md

@@ -9,7 +9,7 @@ errors before the corresponding PR is merged.
 ## Example
 
 This is an example of malformed YAML, which will raise an error if
-you try to parse it (at least, in ruby):
+you try to parse it (at least, in Ruby and Python):
 
 ```
 desc: Example of a malformed YAML file
@@ -30,13 +30,16 @@ jobs:
   reject-malformed-yaml:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
-      - uses: ministryofjustice/github-actions/malformed-yaml@main
+      - name: Checkout repo
+      - uses: actions/[email protected]
+      - name: Detect malformed YAML files
+      - uses: ministryofjustice/github-actions/[email protected]
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          PR_NUMBER: ${{ github.event.number }}
+
 ```
 
-`GITHUB_TOKEN` is provided automatically by github actions. You do
-not need to do anything extra to make it available. Just use the
+`GITHUB_TOKEN` and `PR_NUMBER` are provided automatically. You do
+not need to do anything extra to make these available. Just use the
 content above, exactly as shown.
-

malformed-yaml/action.yml

@@ -0,0 +1,5 @@
+name: "Malformed YAML"
+description: "Detect malformed YAML files in a PR."
+runs:
+  using: "docker"
+  image: "Dockerfile"

malformed-yaml/github_pull_request.py

@@ -0,0 +1,20 @@
+from github import Github
+
+
+class GitHubPullRequest():
+    """A class to interact with the GitHub API."""
+
+    def __init__(self, github_token, repository_name: str, pr_number: int):
+        self.github_token = github_token
+        self.pr_number = pr_number
+        self.repository_name = repository_name
+        self.client = Github(self.github_token)
+
+    def get_changed_files_from_pr(self) -> list:
+        return [file.filename for file in self.client.get_repo(
+            self.repository_name).get_pull(self.pr_number).get_files()]
+
+    def fail_pr(self, message: str):
+        self.client.get_repo(
+            self.repository_name).get_pull(self.pr_number).create_issue_comment(
+                message)

malformed-yaml/main.py

@@ -0,0 +1,100 @@
+import os
+import logging
+import sys
+import re
+import yaml
+from github_pull_request import GitHubPullRequest as github_pull_request
+
+
+logging.basicConfig(
+    format="%(asctime)s %(levelname)-8s %(message)s",
+    level="INFO",
+    datefmt="%Y-%m-%d %H:%M:%S",
+)
+
+logger = logging.getLogger(__name__)
+
+
+def get_github_env() -> tuple[str, str, str]:
+    """
+    Function to collect the three required GitHub
+    environment variables
+    """
+    token = os.getenv("GITHUB_TOKEN")
+    pr_number = os.getenv("PR_NUMBER")
+    repo = os.getenv("GITHUB_REPOSITORY")
+    if not token:
+        raise ValueError("No GITHUB_TOKEN.")
+    if not pr_number:
+        raise ValueError("No PR_NUMBER.")
+    return token, repo, pr_number
+
+
+def get_changed_yaml_files_from_pr() -> list[str]:
+    """
+    Collect a list of all the new or modified YAML files
+    in a PR, except those in a 'secret/' directory.
+    """
+    token, repository_name, pr = get_github_env()
+    github_pr = github_pull_request(token, repository_name, int(pr))
+    # We assume there must always be some changed or new files in a PR
+    changed_files = github_pr.get_changed_files_from_pr()
+    pattern = re.compile("\\.yml$|\\.yaml$")
+    skip_pattern = re.compile("secret/")
+    return [
+        file for file in changed_files if pattern.search(file) and not skip_pattern.search(file)
+    ]
+
+def get_malformed_yaml_files_and_errors(yaml_files: list[str]) -> list[str]:
+    """
+    Input:
+        yaml_files: List of YAML files to be tested for correct format
+    Output:
+        malformed_yaml_files: List of those YAML files that are malformed and
+        their error messages.
+    """
+    malformed_yaml_files_and_errors= []
+    for y in yaml_files:
+        with open(y) as stream:
+            try:
+                yaml.safe_load(stream)
+            except yaml.YAMLError as exc:
+                malformed_yaml_files_and_errors.append(f"\n{str(y)}:\n{str(exc)}")
+    return malformed_yaml_files_and_errors
+
+def malformed_yaml_files_message(files_and_errors: list):
+    """
+    Compose message to display in the PR.
+    """
+    msg = "😱 The following malformed YAML files and related errors were found:\n"
+    msg += "\n".join(files_and_errors) + "\n\n🥺 Please correct them and resubmit this PR."
+    return msg
+
+def main():
+    """
+    Function to collect the new or modified YAML files from the PR that
+    are malformed, report these to the user, and request changes.
+    """
+    token, repository_name, pr = get_github_env()
+    github_pr = github_pull_request(token, repository_name, int(pr))
+
+
+    changed_yaml_files = get_changed_yaml_files_from_pr()
+    if not changed_yaml_files:
+        logger.info("🫧 No new or modified YAML files.")
+        return False
+
+    malformed_yaml_files_and_errors = get_malformed_yaml_files_and_errors(changed_yaml_files)
+    if malformed_yaml_files_and_errors:
+        msg = malformed_yaml_files_message(malformed_yaml_files_and_errors)
+        github_pr.fail_pr(message=msg)
+        logger.error(msg)
+        return True
+
+    logger.info("🤩 PR YAML files all OK!")
+    return False
+
+
+if __name__ == "__main__":
+    if main():
+        sys.exit(1)