Testing Guide for Duplicate Logic Detector

This guide explains how to test the duplicate logic detection functionality both locally and as part of your development workflow.

Quick Start

1. Install Dependencies

# Install main dependencies
pip install -r requirements.txt

# Install test dependencies
pip install -r test_requirements.txt

# Or use the test runner to install everything
python run_tests.py install

2. Run All Tests

# Run complete test suite
python run_tests.py all

# Or run specific test types
python run_tests.py unit      # Unit tests only
python run_tests.py sample    # Sample analysis

Test Types

Unit Tests

Located in tests/test_duplicate_detector.py, these tests verify individual components:

• CodeFunction class: Tests function data structure
• DuplicateMatch class: Tests match result structure
• DuplicateLogicDetector: Tests core detection logic
• Integration tests: Tests complete detection workflow

# Run unit tests
python run_tests.py unit

# With verbose output
python run_tests.py unit --verbose

# With coverage report
python run_tests.py unit --coverage

Sample Analysis

Test with pre-built sample files that contain known duplicates:

# Analyze sample files
python run_tests.py sample

Sample files:

• test_samples/original_code.py - Original functions
• test_samples/duplicate_code.py - Near-identical duplicates
• test_samples/similar_but_different.py - Similar but different logic

Integration Tests

Test the detector on real repository scenarios:

# Test on current repository
python run_tests.py integration

# Test specific files
python run_tests.py integration --changed-files file1.py file2.py

# Test different repository
python run_tests.py integration --repository-path /path/to/repo

Manual Testing

Direct Script Usage

You can also run the detection script directly:

cd scripts
python duplicate_logic_detector.py \
    --repository-path ../test_samples \
    --changed-files duplicate_code.py \
    --output-format json

Testing with Your Own Code

1. Create test files with intentional duplicates:

# file1.py
def validate_email(email):
    import re
    pattern = r'^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$'
    return re.match(pattern, email) is not None

# file2.py  
def check_email_format(email_addr):
    import re
    regex = r'^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$'
    return re.match(regex, email_addr) is not None

2. Run detection:

python run_tests.py integration --changed-files file2.py

Test Configuration

Custom Configuration

Create a test-specific config file:

# test-config.yml
thresholds:
  high_similarity: 0.70    # Lower threshold for testing
  moderate_similarity: 0.50

similarity_weights:
  semantic: 0.50           # Emphasize semantic similarity
  signature: 0.30
  ast_structure: 0.15
  function_calls: 0.05

min_function_lines: 3      # Test smaller functions

Use with:

python scripts/duplicate_logic_detector.py \
    --config-file test-config.yml \
    --changed-files test_file.py

Expected Results

High-Confidence Duplicates

Functions that should be detected as high-confidence duplicates:

• validate_email() vs check_email_format() - Same regex, similar logic
• calculate_discount() vs compute_discount_amount() - Identical calculation
• process_user_data() vs validate_and_process_users() - Same workflow

Medium-Confidence Matches

Functions with similar patterns but different purposes:

• validate_email() vs validate_phone_number() - Similar validation pattern
• calculate_discount() vs calculate_tax() - Similar math operations

Expected Output

{
  "summary": {
    "total_matches": 8,
    "high_confidence": 5,
    "medium_confidence": 2,
    "low_confidence": 1
  },
  "matches": [
    {
      "original_function": {
        "name": "validate_email",
        "file": "test_samples/original_code.py",
        "line_start": 8
      },
      "duplicate_function": {
        "name": "check_email_format", 
        "file": "test_samples/duplicate_code.py",
        "line_start": 8
      },
      "similarity_score": 0.92,
      "confidence_level": "very_high",
      "suggestions": [
        "Extract common email validation logic into a shared utility function",
        "Consider using the existing validate_email function instead"
      ]
    }
  ]
}

Troubleshooting

Common Issues

1. ImportError: No module named 'duplicate_logic_detector'

   # Make sure the script exists
   ls scripts/duplicate_logic_detector.py
   
   # Install dependencies
   python run_tests.py install

2. NLTK Data Missing

   python -c "
   import nltk
   nltk.download('punkt')
   nltk.download('stopwords')
   "

3. No Matches Found

   • Lower similarity thresholds in config
   • Check file patterns (include/exclude)
   • Verify functions meet minimum complexity/length requirements

4. Too Many False Positives

   • Increase similarity thresholds
   • Add exclusion patterns for utility functions
   • Adjust similarity weights

Debug Mode

Enable detailed logging:

export DEBUG=1
python run_tests.py sample

Test Coverage

Generate detailed coverage reports:

python run_tests.py unit --coverage
# Open htmlcov/index.html in browser

CI/CD Integration

GitHub Actions Testing

Add to .github/workflows/test-duplicate-detection.yml:

name: Test Duplicate Detection

on: [push, pull_request]

jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.11'
          
      - name: Install dependencies
        run: python run_tests.py install
        
      - name: Run tests
        run: python run_tests.py all --verbose

Local Pre-commit Hook

Add to .git/hooks/pre-commit:

#!/bin/bash
python run_tests.py unit
if [ $? -ne 0 ]; then
    echo "Tests failed. Commit aborted."
    exit 1
fi

Performance Testing

Large Codebase Testing

Test with larger codebases:

# Clone a Python project
git clone https://github.com/python/cpython.git test_large_repo
cd test_large_repo

# Test duplicate detection
python ../run_tests.py integration \
    --repository-path . \
    --changed-files Lib/email/utils.py

Benchmark Testing

Time the detection process:

time python run_tests.py sample

Expected performance:

• Small files (< 100 functions): < 5 seconds
• Medium files (100-500 functions): < 30 seconds
• Large files (500+ functions): < 2 minutes

Contributing Tests

When adding new features, include tests:

1. Add unit tests in tests/test_duplicate_detector.py
2. Add sample cases in test_samples/
3. Update documentation in this file
4. Run full test suite before submitting

# Verify all tests pass
python run_tests.py all --coverage