chore: Add environment variable check script (#126)

# Pull Request

## Description

This change introduces a new Python script to check for required
environment variables before the main execution. The script
`check_environment_variables.py` is added to verify the presence of
essential environment variables like `github_token` and
`INPUT_REPOSITORY_OWNER`.

The Dockerfile is updated to copy the new Python script and its
`__init__.py` file to the root directory. The `run.sh` script now
includes a step to execute this environment variable check before
proceeding with the main execution.

The Justfile is modified to include the new `python_scripts` directory
in the ruff linting and formatting commands, ensuring code quality for
the new additions.

In the `pyproject.toml` file, a new rule is added to ignore the T201
(print statement) warning for files in the `python_scripts` directory,
as printing is intentional for error reporting in the new script.

These changes enhance the robustness of the execution environment by
ensuring that all necessary variables are set before running the main
process.

fixes #125

Dockerfile

@@ -17,6 +17,7 @@ RUN mkdir -p /statistics && \
 
 COPY --chmod=755 run.sh run.sh
 COPY analyser analyser
+COPY __init__.py python_scripts/check_environment_variables.py /
 
 COPY --from=builder requirements.txt requirements.txt
 RUN pip install --no-cache-dir -r requirements.txt

Justfile

@@ -52,6 +52,7 @@ docker-build:
 
 docker-run:
     docker run \
+      --env github_token=${GITHUB_TOKEN} \
       --env INPUT_REPOSITORY_OWNER=JackPlowman \
       --volume "$(pwd)/statistics:/statistics" \
       --volume "$(pwd)/cloned_repositories:/cloned_repositories" \
@@ -68,15 +69,19 @@ ruff-fix:
 
 ruff-lint:
     poetry run ruff check analyser
+    poetry run ruff check python_scripts
 
 ruff-lint-fix:
     poetry run ruff check analyser --fix
+    poetry run ruff check python_scripts --fix
 
 ruff-format:
     poetry run ruff format --check analyser
+    poetry run ruff format --check python_scripts
 
 ruff-format-fix:
     poetry run ruff format analyser
+    poetry run ruff format python_scripts
 
 # ------------------------------------------------------------------------------
 # Other Python Tools

analyser/utils/github_interactions.py

@@ -36,17 +36,10 @@ def retrieve_repositories() -> PaginatedList[Repository]:
     Returns:
         PaginatedList[Repository]: The list of repositories.
     """
-    repository_owner = getenv("INPUT_REPOSITORY_OWNER", "")
-    if repository_owner == "":
-        msg = "repository_owner environment variable is not set."
-        raise ValueError(msg)
-    token = getenv("github_token", "")
-    if token == "":
-        github = Github()
-        logger.debug("Using unauthenticated GitHub API")
-    else:
-        github = Github(token)
-        logger.debug("Using authenticated GitHub API")
+    repository_owner = getenv("INPUT_REPOSITORY_OWNER")
+    token = getenv("INPUT_GITHUB_TOKEN")
+    github = Github(token)
+    logger.debug("Using authenticated GitHub API")
     repositories = github.search_repositories(query=f"user:{repository_owner} archived:false")
     logger.info(
         "Found repositories",

analyser/utils/tests/test_github_interactions.py

@@ -1,7 +1,5 @@
 from unittest.mock import MagicMock, call, patch
 
-import pytest
-
 from analyser.utils.github_interactions import clone_repo, retrieve_repositories
 
 FILE_PATH = "analyser.utils.github_interactions"
@@ -34,24 +32,7 @@ def test_clone_repo_exists(mock_repo: MagicMock, mock_path: MagicMock) -> None:
 
 @patch(f"{FILE_PATH}.Github")
 @patch(f"{FILE_PATH}.getenv")
-def test_retrieve_repositories__unauthenticated(mock_getenv: MagicMock, mock_github: MagicMock) -> None:
-    # Arrange
-    mock_getenv.side_effect = ["Test", ""]
-    full_name = "Test1/Test2"
-    mock_github.return_value.search_repositories.return_value = search_return = MagicMock(
-        totalCount=1, list=[MagicMock(full_name=full_name)]
-    )
-    # Act
-    repositories = retrieve_repositories()
-    # Assert
-    mock_github.assert_called_once_with()
-    mock_getenv.assert_has_calls([call("INPUT_REPOSITORY_OWNER", ""), call("github_token", "")])
-    assert repositories == search_return
-
-
-@patch(f"{FILE_PATH}.Github")
-@patch(f"{FILE_PATH}.getenv")
-def test_retrieve_repositories__authenticated(mock_getenv: MagicMock, mock_github: MagicMock) -> None:
+def test_retrieve_repositories(mock_getenv: MagicMock, mock_github: MagicMock) -> None:
     # Arrange
     token = "TestToken"  # noqa: S105
     mock_getenv.side_effect = ["Test", token]
@@ -63,18 +44,5 @@ def test_retrieve_repositories__authenticated(mock_getenv: MagicMock, mock_githu
     repositories = retrieve_repositories()
     # Assert
     mock_github.assert_called_once_with(token)
-    mock_getenv.assert_has_calls([call("INPUT_REPOSITORY_OWNER", ""), call("github_token", "")])
+    mock_getenv.assert_has_calls([call("INPUT_REPOSITORY_OWNER"), call("INPUT_GITHUB_TOKEN")])
     assert repositories == search_return
-
-
-@patch(f"{FILE_PATH}.Github")
-@patch(f"{FILE_PATH}.getenv")
-def test_retrieve_repositories__no_repository_owner(mock_getenv: MagicMock, mock_github: MagicMock) -> None:
-    # Arrange
-    mock_getenv.return_value = ""
-    # Act
-    with pytest.raises(ValueError, match="repository_owner environment variable is not set."):
-        retrieve_repositories()
-    # Assert
-    mock_github.assert_not_called()
-    mock_getenv.assert_called_once_with("INPUT_REPOSITORY_OWNER", "")

pyproject.toml

@@ -16,9 +16,9 @@ pandas = "2.2.3"
 [tool.poetry.group.dev.dependencies]
 pytest = "8.3.3"
 pytest-cov = "5.0.0"
-ruff = "0.6.8"
-vulture = "2.12"
-check-jsonschema = "0.29.2"
+ruff = "0.6.9"
+vulture = "2.13"
+check-jsonschema = "0.29.3"
 
 [build-system]
 requires = ["poetry-core"]
@@ -70,6 +70,7 @@ dummy-variable-rgx = "(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"
 
 [tool.ruff.lint.per-file-ignores]
 "**test_*.py" = ["S101", "D102", "D103", "SLF001", "PT019"]
+"python_scripts/*.py" = ["T201"]
 
 [tool.ruff.lint.pydocstyle]
 convention = "google" # Use Google docstring convention.

python_scripts/check_environment_variables.py

@@ -0,0 +1,12 @@
+from os import getenv
+from sys import exit
+
+EXPECTED_ENVIRONMENT_VARIABLES = ["INPUT_GITHUB_TOKEN", "INPUT_REPOSITORY_OWNER"]
+
+for variable in EXPECTED_ENVIRONMENT_VARIABLES:
+    if getenv(variable) is None:
+        if "INPUT_" in variable:
+            print(f'Error: Required environment variable {variable.removeprefix("INPUT_")} is not set')
+        else:
+            print(f"Error: Required environment variable {variable} is not set")
+        exit(1)

run.sh

@@ -7,6 +7,9 @@ if [ "$CI" = "true" ]; then
   cd ..
 fi
 
+# Check that the required environment variables are set
+python check_environment_variables.py
+
 # Run the analyser
 python -m analyser
 

sonar-project.properties

@@ -6,5 +6,5 @@ sonar.sources=.
 sonar.python.version=3.12
 sonar.python.coverage.reportPaths=coverage.xml
 
-sonar.coverage.exclusions=**/tests/**
+sonar.coverage.exclusions=**/tests/**,python_scripts/check_environment_variables.py
 sonar.test.exclusions=**/tests/**