Fix flake8 errors in image_service.py

Author: sidmohan0

datafog/processing/image_processing/donut_processor.py

@@ -73,13 +73,17 @@ async def extract_text_from_image(self, image: Image.Image) -> str:
 
         # If we're in a test environment and PYTEST_DONUT is not enabled, return a mock response
         if IN_TEST_ENV and not DONUT_TESTING_ENABLED:
-            logging.info("Running in test environment without PYTEST_DONUT=yes, returning mock OCR result")
+            logging.info(
+                "Running in test environment without PYTEST_DONUT=yes, returning mock OCR result"
+            )
             mock_result = {"text": "Mock OCR text for testing"}
             return json.dumps(mock_result)
-        
+
         # If PYTEST_DONUT is enabled, log that we're running real OCR in test mode
         if IN_TEST_ENV and DONUT_TESTING_ENABLED:
-            logging.info("PYTEST_DONUT=yes is set, running actual OCR in test environment")
+            logging.info(
+                "PYTEST_DONUT=yes is set, running actual OCR in test environment"
+            )
 
         # Only import torch and transformers when actually needed and not in test environment
         try:

tests/test_ocr_integration.py

@@ -5,16 +5,16 @@
 Otherwise, they will use a mock implementation.
 """
 
-import os
+import io
 import json
-import pytest
+import os
 from unittest.mock import patch
+
+import pytest
 from PIL import Image
-import io
 
-from datafog.services.image_service import ImageService
 from datafog.processing.image_processing.donut_processor import DonutProcessor
-
+from datafog.services.image_service import ImageService
 
 # Mark all tests in this file as integration tests
 pytestmark = pytest.mark.integration
@@ -24,7 +24,7 @@
 def sample_image():
     """Create a simple test image."""
     # Create a small white image with some black text
-    img = Image.new('RGB', (200, 100), color='white')
+    img = Image.new("RGB", (200, 100), color="white")
     return img
 
 
@@ -42,21 +42,24 @@ def image_service_donut():
 
 def test_ocr_with_tesseract(image_service_tesseract, sample_image):
     """Test OCR extraction using Tesseract.
-    
+
     This test should always run regardless of the PYTEST_DONUT flag.
     """
     # Save the image to a bytes buffer
     img_buffer = io.BytesIO()
-    sample_image.save(img_buffer, format='PNG')
+    sample_image.save(img_buffer, format="PNG")
     img_buffer.seek(0)
-    
+
     # Create a temporary file-like object that PIL can open
-    with patch('PIL.Image.open', return_value=sample_image):
-        with patch('os.path.isfile', return_value=True):
+    with patch("PIL.Image.open", return_value=sample_image):
+        with patch("os.path.isfile", return_value=True):
             # Run the OCR extraction
             import asyncio
-            result = asyncio.run(image_service_tesseract.ocr_extract(['dummy_path.png']))
-            
+
+            result = asyncio.run(
+                image_service_tesseract.ocr_extract(["dummy_path.png"])
+            )
+
             # Verify that we got some result (even if empty for a blank image)
             assert result is not None
             assert isinstance(result, list)
@@ -65,54 +68,62 @@ def test_ocr_with_tesseract(image_service_tesseract, sample_image):
 
 def test_ocr_with_donut(sample_image):
     """Test OCR extraction using Donut.
-    
+
     This test will use a mock implementation if PYTEST_DONUT is not set to 'yes'.
     It will use the actual implementation if PYTEST_DONUT=yes.
     """
     # Save the image to a bytes buffer
     img_buffer = io.BytesIO()
-    sample_image.save(img_buffer, format='PNG')
+    sample_image.save(img_buffer, format="PNG")
     img_buffer.seek(0)
-    
+
     # Force the test environment flag to be recognized
-    with patch('datafog.processing.image_processing.donut_processor.IN_TEST_ENV', True):
-        with patch('datafog.processing.image_processing.donut_processor.DONUT_TESTING_ENABLED', False):
+    with patch("datafog.processing.image_processing.donut_processor.IN_TEST_ENV", True):
+        with patch(
+            "datafog.processing.image_processing.donut_processor.DONUT_TESTING_ENABLED",
+            False,
+        ):
             # Create a new image service with Donut enabled
             image_service = ImageService(use_donut=True, use_tesseract=False)
-            
+
             # Create a temporary file-like object that PIL can open
-            with patch('PIL.Image.open', return_value=sample_image):
-                with patch('os.path.isfile', return_value=True):
+            with patch("PIL.Image.open", return_value=sample_image):
+                with patch("os.path.isfile", return_value=True):
                     # Run the OCR extraction
                     import asyncio
-                    result = asyncio.run(image_service.ocr_extract(['dummy_path.png']))
-                    
+
+                    result = asyncio.run(image_service.ocr_extract(["dummy_path.png"]))
+
                     # Verify that we got some result
                     assert result is not None
                     assert isinstance(result, list)
                     assert len(result) == 1
-                    
+
                     # We should get the mock result since PYTEST_DONUT is not set
                     assert "Mock OCR text for testing" in result[0]
 
 
 def test_donut_processor_directly(sample_image):
     """Test the DonutProcessor directly.
-    
+
     This test will use a mock implementation if PYTEST_DONUT is not set to 'yes'.
     It will use the actual implementation if PYTEST_DONUT=yes.
     """
     # Force the test environment flag to be recognized
-    with patch('datafog.processing.image_processing.donut_processor.IN_TEST_ENV', True):
-        with patch('datafog.processing.image_processing.donut_processor.DONUT_TESTING_ENABLED', False):
+    with patch("datafog.processing.image_processing.donut_processor.IN_TEST_ENV", True):
+        with patch(
+            "datafog.processing.image_processing.donut_processor.DONUT_TESTING_ENABLED",
+            False,
+        ):
             processor = DonutProcessor()
-            
+
             # Run the OCR extraction
             import asyncio
+
             result = asyncio.run(processor.extract_text_from_image(sample_image))
-            
+
             # Verify that we got some result
             assert result is not None
-            
+
             # If PYTEST_DONUT is not set, we should get the mock result
             assert "Mock OCR text for testing" in result