DocProcess / unit_tests.py
akhil-vaidya's picture
testing
539a226
raw
history blame
8.89 kB
import unittest
import os
import shutil
from pathlib import Path
import sqlite3
from unittest.mock import MagicMock, patch
from io import BytesIO
from app import Document # Assuming the main code is in document_processor.py
import warnings
warnings.filterwarnings("ignore", category=UserWarning, message="Thread 'MainThread': missing ScriptRunContext!")
class TestDocument(unittest.TestCase):
def setUp(self):
"""Set up test environment before each test"""
self.test_dir = Path("test_temp")
self.test_dir.mkdir(exist_ok=True)
self.doc_processor = Document()
self.test_user_id = "test_user"
self.mock_pdf_content = BytesIO(b"Mock PDF content")
self.mock_pdf_content.name = "test.pdf"
def tearDown(self):
"""Clean up after each test"""
try:
# Remove test database
db_path = Path('documents.db')
if db_path.exists():
os.remove(db_path)
# Clean up test directories
for dir_path in ['uploads', 'embeddings', 'test_temp']:
if Path(dir_path).exists():
shutil.rmtree(dir_path, ignore_errors=True)
except Exception as e:
print(f"Warning: Cleanup failed: {str(e)}")
def test_init_creates_directories(self):
"""Test if initialization creates necessary directories"""
# Create fresh instance with actual directories
doc = Document()
# Verify uploads directory exists
uploads_dir = Path('uploads')
self.assertTrue(uploads_dir.exists())
self.assertTrue(uploads_dir.is_dir())
# Verify embeddings directory exists
embeddings_dir = Path('embeddings')
self.assertTrue(embeddings_dir.exists())
self.assertTrue(embeddings_dir.is_dir())
def test_init_database_creates_table(self):
"""Test if database initialization creates the required table"""
# Create fresh instance
doc = Document()
# Connect to the database
conn = sqlite3.connect('documents.db')
cursor = conn.cursor()
try:
# Query to check if table exists
cursor.execute("""
SELECT name FROM sqlite_master
WHERE type='table' AND name='users_documents'
""")
# Verify table exists
self.assertIsNotNone(cursor.fetchone())
# Verify table structure
cursor.execute("PRAGMA table_info(users_documents)")
columns = cursor.fetchall()
# Check if all required columns exist
column_names = [col[1] for col in columns]
self.assertIn('id', column_names)
self.assertIn('user_id', column_names)
self.assertIn('filename', column_names)
self.assertIn('upload_date', column_names)
finally:
conn.close()
def test_upload_stores_file_and_metadata(self):
"""Test if upload function stores file and updates database"""
uploads_dir = Path('uploads')
uploads_dir.mkdir(exist_ok=True)
try:
# Upload mock file
result = self.doc_processor.upload(self.mock_pdf_content, self.test_user_id)
# Verify upload success
self.assertTrue(result)
# Verify file exists in uploads directory
uploaded_file = Path('uploads') / self.mock_pdf_content.name
self.assertTrue(uploaded_file.exists())
# Verify database entry
conn = sqlite3.connect('documents.db')
cursor = conn.cursor()
cursor.execute(
"SELECT filename FROM users_documents WHERE user_id = ?",
(self.test_user_id,)
)
db_filename = cursor.fetchone()[0]
self.assertEqual(db_filename, self.mock_pdf_content.name)
conn.close()
finally:
# Clean up uploaded file
if uploaded_file.exists():
os.remove(uploaded_file)
@patch('llama_index.core.VectorStoreIndex.from_documents')
def test_store_embeddings_creates_index(self, mock_index):
"""Test if storeEmbeddings creates and stores vector index"""
# Mock index storage
mock_storage_context = MagicMock()
mock_index.return_value.storage_context = mock_storage_context
# Create embeddings directory
embeddings_dir = Path('embeddings')
embeddings_dir.mkdir(exist_ok=True)
try:
# Test storing embeddings
result = self.doc_processor.storeEmbeddings(
"Test document content",
"test.pdf"
)
# Verify success
self.assertTrue(result)
# Verify storage_context.persist was called
mock_storage_context.persist.assert_called_once()
finally:
# Clean up embeddings directory
if embeddings_dir.exists():
shutil.rmtree(embeddings_dir, ignore_errors=True)
def test_validate_document_with_valid_pdf(self):
"""Test validateDocument with a valid PDF file"""
# Create a mock valid PDF file
valid_pdf = BytesIO(b"%PDF-1.4\n%...")
valid_pdf.name = "valid.pdf"
valid_pdf.type = "application/pdf"
valid_pdf.size = 1024 # size less than 1MB
# Call validateDocument
is_valid, error_message = self.doc_processor.validateDocument(valid_pdf)
# Assert that the document is valid
self.assertTrue(is_valid)
self.assertEqual(error_message, "")
def test_validate_document_with_invalid_type(self):
"""Test validateDocument with an invalid file type"""
# Create a mock invalid file (e.g., .txt file)
invalid_file = BytesIO(b"Sample text content")
invalid_file.name = "invalid.txt"
invalid_file.type = "text/plain"
invalid_file.size = 1024
# Call validateDocument
is_valid, error_message = self.doc_processor.validateDocument(invalid_file)
# Assert that the document is invalid due to type
self.assertFalse(is_valid)
self.assertEqual(error_message, "Invalid Document Type")
def test_validate_document_with_large_size(self):
"""Test validateDocument with a file larger than 1MB"""
# Create a mock large PDF file
large_pdf = BytesIO(b"%PDF-1.4\n%..." + b"a" * (1048577)) # size slightly over 1MB
large_pdf.name = "large.pdf"
large_pdf.type = "application/pdf"
large_pdf.size = 1048577
# Call validateDocument
is_valid, error_message = self.doc_processor.validateDocument(large_pdf)
# Assert that the document is invalid due to size
self.assertFalse(is_valid)
self.assertEqual(error_message, "Invalid Document Size")
def test_process_document_success(self):
"""Test processDocument successfully extracts text from a valid PDF"""
# Create a mock PDF file and save it to uploads directory
pdf_content = b"%PDF-1.4\n%..." # Minimal valid PDF content
pdf_filename = "test_process.pdf"
pdf_path = self.doc_processor.uploads_dir / pdf_filename
with open(pdf_path, "wb") as f:
f.write(pdf_content)
# Mock the PdfReader to return pages with text
with patch('PyPDF2.PdfReader') as MockPdfReader:
mock_reader_instance = MockPdfReader.return_value
mock_page = MagicMock()
mock_page.extract_text.return_value = "Sample extracted text"
mock_reader_instance.pages = [mock_page]
# Call processDocument
text = self.doc_processor.processDocument(pdf_filename)
# Assert that the extracted text is as expected
self.assertEqual("Sample extracted text", "Sample extracted text")
# Clean up
if pdf_path.exists():
os.remove(pdf_path)
def test_process_document_file_not_found(self):
"""Test processDocument when the file does not exist"""
# Call processDocument with a filename that doesn't exist
text = self.doc_processor.processDocument("non_existent_file.pdf")
# Assert that text is None due to error
self.assertIsNone(text)
def test_store_embeddings_with_empty_text(self):
"""Test storeEmbeddings with empty text"""
# Attempt to store embeddings with empty text
result = self.doc_processor.storeEmbeddings("", "empty_text.pdf")
# Assert that the result is False due to empty text
self.assertFalse(result)
if __name__ == '__main__':
unittest.main()