import fitz # PyMuPDF
from ragstack_common.ocr import OcrService
def test_is_text_native_pdf():
"""Test PDF text detection with a generated PDF."""
# Create a simple PDF with text
pdf_doc = fitz.open()
page = pdf_doc.new_page()
page.insert_text((100, 100), "This is a test document with text content.")
page.insert_text((100, 150), "It has multiple lines to simulate a real document.")
# Convert to bytes
pdf_bytes = pdf_doc.tobytes()
pdf_doc.close()
# Test OCR service
ocr_service = OcrService(region="us-east-1", backend="textract")
is_text_native = ocr_service._is_text_native_pdf(pdf_bytes)
assert is_text_native
print(f"✓ Text-native PDF detection works: {is_text_native}")
def test_ocr_service_initialization():
"""Test OCR service initialization."""
# Test with textract backend
ocr1 = OcrService(region="us-east-1", backend="textract")
assert ocr1.backend == "textract"
assert ocr1.region == "us-east-1"
# Test with bedrock backend
ocr2 = OcrService(region="us-west-2", backend="bedrock", bedrock_model_id="test-model")
assert ocr2.backend == "bedrock"
assert ocr2.bedrock_model_id == "test-model"
print("✓ OCR service initialization works")
if __name__ == "__main__":
test_is_text_native_pdf()
test_ocr_service_initialization()
print("OCR module tests passed!")
