Update app.py

app.py

@@ -33,15 +33,22 @@ def get_ocr(lang, use_gpu=False):
     return ocr_cache.get(lang)
 
 # Function to extract images from PDF
-def pdf_to_images(file_contents):
+def pdf_to_images(uploaded_file):
     try:
-        doc = fitz.open(io.BytesIO(file_contents))
-        images = []
-        for page in doc:
+        doc = fitz.open(stream=uploaded_file.read(),filetype="pdf")
+        image_parts = []
+
+        for page_number in range(len(pdf_document)):
+            page = pdf_document.load_page(page_number)
             pix = page.get_pixmap()
-            img = Image.open(io.BytesIO(pix.tobytes("png")))
-            images.append(img)
-        return images
+            image_data = pix.tobytes("png")
+            
+            image_parts.append({
+                "mime_type": "image/png",
+                "data": image_data
+            })
+    
+        return image_parts
     except Exception as e:
         logger.error(f"Error processing PDF: {str(e)}")
         raise HTTPException(status_code=500, detail="Error processing PDF file")
@@ -56,7 +63,7 @@ async def create_upload_file(
 
         # Determine if the uploaded file is a PDF or image
         if file.content_type == "application/pdf":
-            images = pdf_to_images(contents)
+            images = pdf_to_images(file)
         else:
             # If it's an image file
             images = [Image.open(io.BytesIO(contents))]