Update app.py

app.py

@@ -109,9 +109,12 @@ class OCRProcessor:
             ocr_logger.info("PDF document detected, enforcing page limit")
             payload['filetype'] = 'PDF'
         
-        # Prepare file for OCR API - using file object directly as in the test script
+        # Prepare file for OCR API - using file data as bytes to avoid file handle issues
+        with open(file_path, 'rb') as f:
+            file_data = f.read()
+            
         files = {
-            'file': (os.path.basename(file_path), open(file_path, 'rb'))
+            'file': (os.path.basename(file_path), file_data, file_type)
         }
         
         headers = {
@@ -132,9 +135,18 @@ class OCRProcessor:
             ocr_logger.info(f"OCR API status code: {response.status_code}")
             
             # Log response text for debugging (first 200 chars)
-            ocr_logger.info(f"OCR API response preview: {response.text[:200]}...")
+            response_preview = response.text[:200] if hasattr(response, 'text') else "No text content"
+            ocr_logger.info(f"OCR API response preview: {response_preview}...")
             
-            response.raise_for_status()
+            try:
+                response.raise_for_status()
+            except Exception as e:
+                ocr_logger.error(f"HTTP Error: {str(e)}")
+                return {
+                    "success": False,
+                    "error": f"OCR API HTTP Error: {str(e)}",
+                    "text": ""
+                }
             
             try:
                 result = response.json()
@@ -177,8 +189,7 @@ class OCRProcessor:
                 "text": ""
             }
         finally:
-            # Close the file handle
-            files['file'][1].close()
+            # No need to close file handle as we're using bytes directly
     
     def _extract_text_from_result(self, result: Dict) -> str:
         """