fix: Resolve Windows test failures and CI improvements

- Fixed Windows test failures by removing /dev/null redirects
- Fixed path validation to work with Windows backslashes
- Added GITHUB_TOKEN permissions to CI workflow
- Updated CONTRIBUTING.md with pytest test instructions
- Applied ruff linting and formatting to all test files

Author: dguido

.github/workflows/ci.yml

@@ -14,6 +14,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   tests:
     name: CI (Python ${{ matrix.python }} on ${{ matrix.os }})

CONTRIBUTING.md

@@ -71,13 +71,23 @@ pylint solc_select --rcfile pyproject.toml
 
 ## Running Tests
 
-Tests are platform-specific and can be run locally:
+Tests use pytest and can be run locally:
 
 ```bash
-# Run tests for your platform
-bash scripts/test_linux.sh    # On Linux
-bash scripts/test_macos.sh    # On macOS  
-bash scripts/test_windows.sh  # On Windows
+# Install development dependencies (includes pytest)
+pip install -e ".[dev]"
+
+# Run all tests
+pytest tests/
+
+# Run specific test file
+pytest tests/test_compiler_versions.py
+
+# Run tests with verbose output
+pytest tests/ -v
+
+# Skip slow tests (upgrade test) and platform boundary tests
+pytest tests/ -k "not version_boundaries" -m "not slow"
 
 # Tests also run automatically in GitHub Actions on all platforms
 ```

tests/__init__.py

@@ -1 +1 @@
-"""Test suite for solc-select."""
+"""Test suite for solc-select."""

tests/conftest.py

@@ -29,20 +29,20 @@ def solc_select_path() -> Path:
 def backup_current_version(solc_select_path: Path) -> Generator[None, None, None]:
     """
     Backup and restore the current solc version.
-    
+
     This fixture ensures that each test starts with a clean state
     and doesn't affect the user's current solc configuration.
     """
     global_version_file = solc_select_path / "global-version"
     backup_file = None
-    
+
     # Backup current version if it exists
     if global_version_file.exists():
         backup_file = global_version_file.with_suffix(".backup")
         shutil.copy2(global_version_file, backup_file)
-    
+
     yield
-    
+
     # Restore original version
     if backup_file and backup_file.exists():
         shutil.copy2(backup_file, global_version_file)
@@ -56,27 +56,27 @@ def backup_current_version(solc_select_path: Path) -> Generator[None, None, None
 def clean_artifacts(solc_select_path: Path) -> Generator[None, None, None]:
     """
     Clean up test artifacts after each test.
-    
+
     This is used for tests that need complete isolation and
     should start with no installed versions.
     """
     artifacts_dir = solc_select_path / "artifacts"
     backup_dir = None
-    
+
     # Backup existing artifacts if they exist
     if artifacts_dir.exists():
         backup_dir = artifacts_dir.with_suffix(".backup")
         if backup_dir.exists():
             shutil.rmtree(backup_dir)
         shutil.copytree(artifacts_dir, backup_dir)
         shutil.rmtree(artifacts_dir)
-    
+
     yield
-    
+
     # Clean up test artifacts
     if artifacts_dir.exists():
         shutil.rmtree(artifacts_dir)
-    
+
     # Restore original artifacts if they existed
     if backup_dir and backup_dir.exists():
         shutil.copytree(backup_dir, artifacts_dir)
@@ -87,35 +87,38 @@ def clean_artifacts(solc_select_path: Path) -> Generator[None, None, None]:
 def run_command():
     """
     Execute shell commands and return output.
-    
+
     This fixture provides a conservative way to run commands,
     exactly as the bash tests did.
     """
-    def _run(cmd: str, check: bool = True, capture_stderr: bool = True) -> subprocess.CompletedProcess:
+
+    def _run(
+        cmd: str, check: bool = True, capture_stderr: bool = True
+    ) -> subprocess.CompletedProcess:
         """
         Run a shell command and return the result.
-        
+
         Args:
             cmd: Command to run
             check: Whether to raise on non-zero exit code
             capture_stderr: Whether to capture stderr
-        
+
         Returns:
             CompletedProcess instance with stdout, stderr, and returncode
         """
         stderr_setting = subprocess.STDOUT if capture_stderr else subprocess.PIPE
-        
+
         result = subprocess.run(
             cmd,
             shell=True,
             capture_output=False,
             stdout=subprocess.PIPE,
             stderr=stderr_setting,
             text=True,
-            check=check
+            check=check,
         )
         return result
-    
+
     return _run
 
 
@@ -129,33 +132,23 @@ def test_contracts_dir() -> Path:
 def ensure_solc_select_installed():
     """
     Ensure solc-select is installed in development mode.
-    
+
     This runs once per test session to ensure the current
     development version is being tested.
     """
     # Check if solc-select is available
-    result = subprocess.run(
-        ["solc-select", "--help"],
-        capture_output=True,
-        text=True
-    )
-    
+    result = subprocess.run(["solc-select", "--help"], capture_output=True, text=True)
+
     if result.returncode != 0:
         pytest.exit("solc-select is not installed. Please run: pip install -e .")
 
 
 # Platform markers for conditional test execution
 def pytest_configure(config):
     """Register custom markers."""
-    config.addinivalue_line(
-        "markers", "linux: mark test to run only on Linux"
-    )
-    config.addinivalue_line(
-        "markers", "macos: mark test to run only on macOS"
-    )
-    config.addinivalue_line(
-        "markers", "windows: mark test to run only on Windows"
-    )
+    config.addinivalue_line("markers", "linux: mark test to run only on Linux")
+    config.addinivalue_line("markers", "macos: mark test to run only on macOS")
+    config.addinivalue_line("markers", "windows: mark test to run only on Windows")
 
 
 def pytest_runtest_setup(item):
@@ -165,4 +158,4 @@ def pytest_runtest_setup(item):
     if "macos" in item.keywords and sys.platform != "darwin":
         pytest.skip("Test requires macOS")
     if "windows" in item.keywords and sys.platform != "win32":
-        pytest.skip("Test requires Windows")
+        pytest.skip("Test requires Windows")