Test Generation Skill

When to Use

Use this skill when:

• •Creating unit tests for functions (indexer, search, organize)
• •Writing integration tests for full workflows
• •Testing PDF extraction (pdfplumber and OCR)
• •Testing database operations (SQLite/FTS5)
• •Validating document classification logic

Testing Framework

Currently no test framework configured, but recommended approach:

• •Framework: pytest (recommended for Python projects)
• •Test Location: tests/ directory
• •Naming: test_*.py files
• •Coverage: Aim for >80% coverage

Test Organization

tests/
├── test_indexer.py      # Tests for indexer.py
├── test_search.py       # Tests for search.py
├── test_organize.py     # Tests for organize.py
├── test_extraction.py   # Tests for PDF extraction
└── fixtures/            # Test fixtures (sample PDFs)

Key Functions to Test

From indexer.py:

• •create_db() - Database creation and schema
• •extract_text_from_pdf() - PDF text extraction
• •extract_text_from_pdf_ocr() - OCR fallback
• •classify_document() - Document classification logic
• •extract_metadata() - Metadata extraction
• •index_files() - Full indexing workflow

From search.py:

• •search() - Full-text search with filters
• •list_categories() - Category listing
• •list_contracts() - Contract listing

From organize.py:

• •organize_documents() - Document organization
• •generate_summary_report() - Report generation

Testing Patterns

Unit Test Example

import pytest
from contrato.indexer import classify_document

def test_classify_document_contrato():
    """Test classification of main contract."""
    filename = "TNE_JU_COM_0002-13 - SIEMENS - Contrato.pdf"
    result = classify_document(filename, "/path/to/file")
    
    assert result["category"] == "contrato"
    assert result["document_type"] == "contrato_principal"
    assert result["contract_number"] == "TNE_JU_COM_0002-13"

def test_classify_document_aditivo():
    """Test classification of addendums."""
    filename = "TNE_JU_COM_0002-13 - SIEMENS - 1ºAditivo.pdf"
    result = classify_document(filename, "/path/to/file")
    
    assert result["category"] == "aditivo"
    assert result["document_number"] == "1"

def test_classify_document_anexo():
    """Test classification of attachments."""
    filename = "TNE_JU_COM_0002-13 (Anexo I).pdf"
    result = classify_document(filename, "/path/to/file")
    
    assert result["category"] == "anexo"
    assert result["document_number"] == "I"

Integration Test Example

def test_full_indexing_workflow(tmp_path):
    """Test complete indexing workflow."""
    # Setup: Create test PDF in tmp_path
    # Execute: Run index_files()
    # Assert: Verify documents in database
    # Assert: Verify FTS index created

Mocking Strategies

For external dependencies:

• •PDFs: Use sample PDF files in fixtures/
• •OCR: Mock pytesseract when testing extraction
• •Database: Use in-memory SQLite for tests
• •File System: Use tmp_path from pytest

Test Coverage Goals

• •

  Critical Functions: 100% coverage

  • •classify_document() - Core classification logic
  • •extract_text_from_pdf() - Extraction methods
  • •Database operations
• •

  Overall Coverage: >80%

• •

  Edge Cases: Test error handling, empty files, corrupted PDFs

Test Data

Create test fixtures:

• •Sample PDFs (with text)
• •Sample scanned PDFs (for OCR testing)
• •Sample contract filenames
• •Database fixtures with known data

Best Practices

1. •Isolation: Each test should be independent
2. •Fixtures: Reuse test PDFs and data
3. •Clear Names: Test names should describe what they test
4. •Portuguese: Test descriptions can be in Portuguese
5. •Assertions: Use descriptive assertion messages

Common Test Scenarios

PDF Extraction Tests

• •Extract text from normal PDFs
• •Fallback to OCR for scanned PDFs
• •Handle corrupted PDFs gracefully
• •Extract metadata correctly

Classification Tests

• •Classify contracts correctly
• •Extract contract numbers
• •Identify addendums by number
• •Identify attachments by number

Database Tests

• •Verify schema creation
• •Test FTS5 index updates
• •Validate triggers work
• •Test query performance

Error Handling Tests

• •Missing PDF files
• •Corrupted PDFs
• •Database connection errors
• •OCR failures

Checklist

When writing tests:

• • Test file created in tests/ directory
• • Test functions named test_*
• • Edge cases covered
• • Error cases tested
• • Fixtures created for reusable data
• • Assertions are clear and descriptive
• • Integration tests for workflows
• • Mock external dependencies appropriately