Update extract_insights.py

extract_insights.py

@@ -200,52 +200,110 @@ def adjust_prompt_tokens_v1(prompt: str) -> str:
     return trimmed_text
 
 
-async def process_synchronous_job(user_id,email,message_id,raw_text):
+async def process_synchronous_job(dataset: Dict[str, Any]):
     """
     Background task to process the batch job
     """
 
-    try:
-        prompt = receipt_radar_prompt(raw_text)
-
-        completion = client.chat.completions.create(
-            model="gpt-4o-mini",
-            messages=[
-                {
-                    "role": "user",
-                    "content": prompt
-                }
-            ],
-            response_format={"type": "json_object"},
-            temperature=0.1
-        )
-        print("Printing GPT response")
-        print(completion.choices[0].message)
-
-        # inserting data into supabase
-        insert_data = json.loads(completion.choices[0].message.content)
-        insert_data['email'] = email
-        insert_data['user_id'] = user_id
-        insert_data['message_id'] = message_id
-        insert_data['total_cost'] = parse_number(insert_data['total_cost'])
-        print("Printing user_id")
-        print(user_id)
-
-        insert_response = (
-            supabase.table("receipt_radar_structured_data_duplicate")
-            .insert(insert_data)
-            .execute()
-        )
-
-        update_status_response = (
-            supabase.table("receipt_ocr_data")
-            .update({"status": "processing completed"})
-            .eq("message_id", message_id)
-            .eq("user_id", user_id)
-            .eq("email", email)
-            .execute()
-        )
-
-        print(update_status_response)
-    except Exception as e:
-        print(f"Error occurred during processing: {e}")
+    for ds in dataset.get('data'):
+        user_id = ds.get('user_id')
+            
+        message_id = ds.get('message_id')
+            
+        raw_text = ds.get('receipt_text')
+        email = ds.get('email')
+        try:
+            prompt = adjust_prompt_tokens_v1(receipt_radar_prompt(raw_text))
+    
+            completion = client.chat.completions.create(
+                model="gpt-4o-mini",
+                messages=[
+                    {
+                        "role": "user",
+                        "content": prompt
+                    }
+                ],
+                response_format={"type": "json_object"},
+                temperature=0.1
+            )
+            print("Printing GPT response")
+            print(completion.choices[0].message)
+    
+            # inserting data into supabase
+            insert_data = json.loads(completion.choices[0].message.content)
+            insert_data['email'] = email
+            insert_data['user_id'] = user_id
+            insert_data['message_id'] = message_id
+            insert_data['total_cost'] = parse_number(insert_data['total_cost'])
+            print("Printing user_id")
+            print(user_id)
+    
+            insert_response = (
+                supabase.table("receipt_radar_structured_data_duplicate")
+                .insert(insert_data)
+                .execute()
+            )
+    
+            update_status_response = (
+                supabase.table("receipt_ocr_data")
+                .update({"status": "processing completed"})
+                .eq("message_id", message_id)
+                .eq("user_id", user_id)
+                .eq("email", email)
+                .execute()
+            )
+    
+            print(update_status_response)
+        except Exception as e:
+            print(f"Error occurred during processing: {e}")
+        
+
+
+
+
+    
+
+    # try:
+    #     prompt = receipt_radar_prompt(raw_text)
+
+    #     completion = client.chat.completions.create(
+    #         model="gpt-4o-mini",
+    #         messages=[
+    #             {
+    #                 "role": "user",
+    #                 "content": prompt
+    #             }
+    #         ],
+    #         response_format={"type": "json_object"},
+    #         temperature=0.1
+    #     )
+    #     print("Printing GPT response")
+    #     print(completion.choices[0].message)
+
+    #     # inserting data into supabase
+    #     insert_data = json.loads(completion.choices[0].message.content)
+    #     insert_data['email'] = email
+    #     insert_data['user_id'] = user_id
+    #     insert_data['message_id'] = message_id
+    #     insert_data['total_cost'] = parse_number(insert_data['total_cost'])
+    #     print("Printing user_id")
+    #     print(user_id)
+
+    #     insert_response = (
+    #         supabase.table("receipt_radar_structured_data_duplicate")
+    #         .insert(insert_data)
+    #         .execute()
+    #     )
+
+    #     update_status_response = (
+    #         supabase.table("receipt_ocr_data")
+    #         .update({"status": "processing completed"})
+    #         .eq("message_id", message_id)
+    #         .eq("user_id", user_id)
+    #         .eq("email", email)
+    #         .execute()
+    #     )
+
+    #     print(update_status_response)
+    # except Exception as e:
+    #     print(f"Error occurred during processing: {e}")