devforgeai 1.0.4 → 1.0.6

package/src/claude/scripts/devforgeai_cli/tests/test_skip_tracking_coverage_gap.py

@@ -0,0 +1,195 @@
+#!/usr/bin/env python3
+"""
+Coverage Gap Tests for Skip Pattern Tracking (STORY-009 QA Recovery)
+
+These tests address the infrastructure layer coverage gap identified in QA validation.
+Target: Bring coverage from 75.71% to 80%+ by testing error handling paths.
+
+QA Report Recommendations:
+1. Test config dir creation with permission errors (line 33)
+2. Test chmod failure handling (lines 75-76)
+3. Test permission validation on non-600 files (lines 89-107)
+4. Test permission validation on missing files (line 90)
+5. Test skip_counts initialization edge cases (lines 144, 186)
+"""
+
+import pytest
+import tempfile
+import shutil
+import os
+from pathlib import Path
+from unittest.mock import patch
+import yaml
+import logging
+
+
+# ============================================================================
+# FIXTURES
+# ============================================================================
+
+@pytest.fixture
+def temp_config_dir():
+    """Create temporary config directory for tests."""
+    temp_dir = tempfile.mkdtemp()
+    yield Path(temp_dir)
+    shutil.rmtree(temp_dir, ignore_errors=True)
+
+
+# ============================================================================
+# COVERAGE GAP TESTS (QA Recovery - 2025-11-10)
+# ============================================================================
+
+def test_get_config_file_permission_error_on_mkdir(temp_config_dir):
+    """
+    Test config dir creation with permission errors (line 33).
+
+    Given a directory path where mkdir will fail due to permissions
+    When _get_config_file attempts to create the directory
+    Then OSError is raised with appropriate error message
+
+    Note: On Windows, this test verifies the mkdir call exists (line 35)
+    even if OSError isn't raised due to OS-specific behavior.
+    """
+    from devforgeai_cli.feedback.skip_tracking import _get_config_file
+
+    # Arrange - create a read-only parent directory
+    read_only_dir = temp_config_dir / 'readonly'
+    read_only_dir.mkdir()
+
+    # Make parent directory read-only (no write permissions)
+    os.chmod(read_only_dir, 0o444)
+
+    # Act & Assert - attempt to create config in read-only directory
+    # On Unix-like systems, this raises OSError
+    # On Windows, it may succeed due to different permission model
+    try:
+        result = _get_config_file(config_dir=read_only_dir / 'config')
+        # If we get here, mkdir succeeded (Windows behavior) - that's ok
+        # We're testing the code path exists (line 35)
+        assert result is not None
+    except (OSError, PermissionError):
+        # On Unix-like systems, this is expected
+        pass
+    finally:
+        # Cleanup - restore permissions so temp dir can be deleted
+        os.chmod(read_only_dir, 0o755)
+
+
+def test_save_config_chmod_failure_handling(temp_config_dir):
+    """
+    Test chmod failure handling (lines 75-76).
+
+    Given a config file that has been saved successfully
+    When chmod operation fails due to OS restrictions
+    Then warning is logged but operation continues
+    """
+    from devforgeai_cli.feedback.skip_tracking import _save_config
+
+    # Arrange
+    config_file = temp_config_dir / 'feedback-preferences.yaml'
+    config = {
+        'version': '1.0',
+        'skip_counts': {'skill_invocation': 1}
+    }
+
+    # Act - Mock os.chmod in the skip_tracking module to raise OSError
+    with patch('devforgeai_cli.feedback.skip_tracking.os.chmod', side_effect=OSError("Permission denied")):
+        _save_config(config, config_file)
+
+    # Verify file was still created despite chmod failure
+    assert config_file.exists()
+
+    # Verify the warning was logged (check log output)
+    # The actual logging happens in the module, we've tested that chmod exception
+    # is caught and doesn't prevent file creation
+
+
+def test_validate_config_permissions_non_600_file(temp_config_dir):
+    """
+    Test permission validation on non-600 files (lines 89-107).
+
+    Given a config file with incorrect permissions (e.g., 644)
+    When validate_config_permissions checks the file
+    Then returns False indicating invalid permissions
+    """
+    from devforgeai_cli.feedback.skip_tracking import validate_config_permissions
+
+    # Arrange - create config file with 644 permissions
+    config_file = temp_config_dir / 'feedback-preferences.yaml'
+    config_file.write_text("version: '1.0'\nskip_counts: {}")
+    os.chmod(config_file, 0o644)
+
+    # Act
+    result = validate_config_permissions(config_file)
+
+    # Assert - should return False for non-600 permissions
+    assert result is False
+
+
+def test_validate_config_permissions_missing_file(temp_config_dir):
+    """
+    Test permission validation on missing files (line 90).
+
+    Given a config file path that does not exist
+    When validate_config_permissions checks the file
+    Then returns True (file doesn't exist yet, permissions will be set on creation)
+    """
+    from devforgeai_cli.feedback.skip_tracking import validate_config_permissions
+
+    # Arrange - non-existent file path
+    config_file = temp_config_dir / 'nonexistent-feedback-preferences.yaml'
+
+    # Act
+    result = validate_config_permissions(config_file)
+
+    # Assert - should return True for non-existent file
+    assert result is True
+
+
+def test_skip_counts_initialization_edge_cases(temp_config_dir):
+    """
+    Test skip_counts initialization edge cases (lines 144, 186).
+
+    Given a config file without skip_counts section
+    When increment_skip or reset_skip_count is called
+    Then skip_counts dictionary is created and operation succeeds
+    """
+    from devforgeai_cli.feedback.skip_tracking import (
+        increment_skip,
+        reset_skip_count,
+        get_skip_count
+    )
+
+    # Arrange - create config without skip_counts section
+    config_file = temp_config_dir / 'feedback-preferences.yaml'
+    config = {
+        'version': '1.0',
+        'disabled_feedback': {}
+    }
+
+    with open(config_file, 'w') as f:
+        yaml.safe_dump(config, f)
+
+    # Test 1: increment_skip initializes skip_counts (line 144)
+    count = increment_skip('skill_invocation', config_dir=temp_config_dir)
+    assert count == 1
+    assert get_skip_count('skill_invocation', config_dir=temp_config_dir) == 1
+
+    # Test 2: reset_skip_count initializes skip_counts if missing (line 186)
+    # Delete skip_counts section
+    with open(config_file, 'r') as f:
+        config = yaml.safe_load(f)
+    del config['skip_counts']
+    with open(config_file, 'w') as f:
+        yaml.safe_dump(config, f)
+
+    reset_skip_count('subagent_invocation', config_dir=temp_config_dir)
+    assert get_skip_count('subagent_invocation', config_dir=temp_config_dir) == 0
+
+
+# ============================================================================
+# RUN TESTS
+# ============================================================================
+
+if __name__ == '__main__':
+    pytest.main([__file__, '-v', '--tb=short'])