Update app.py

app.py

@@ -1,14 +1,44 @@
 import os
 import gradio as gr
 import pandas as pd
-from openai import OpenAI
+import json
 from dotenv import load_dotenv
+import httpx
 
 # Load environment variables
 load_dotenv()
 
-# Set up OpenAI client
-client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
+# Get OpenAI API key
+api_key = os.getenv("OPENAI_API_KEY")
+
+# Helper function for OpenAI API calls
+def openai_chat_completion(messages, temperature=0, response_format=None, max_tokens=None):
+    """Make a direct API call to OpenAI without using the client library"""
+    url = "https://api.openai.com/v1/chat/completions"
+    
+    headers = {
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {api_key}"
+    }
+    
+    payload = {
+        "model": "gpt-3.5-turbo",
+        "messages": messages,
+        "temperature": temperature
+    }
+    
+    if response_format:
+        payload["response_format"] = response_format
+        
+    if max_tokens:
+        payload["max_tokens"] = max_tokens
+    
+    response = httpx.post(url, json=payload, headers=headers)
+    
+    if response.status_code != 200:
+        raise Exception(f"OpenAI API error: {response.status_code} - {response.text}")
+        
+    return response.json()
 
 # Simple database using pandas DataFrames
 class SimpleDatabase:
@@ -83,26 +113,24 @@ class QueryRouter:
     def _classify_query(self, query):
         """Classify the query to determine which agent should handle it"""
         # Use OpenAI to classify the query
-        response = client.chat.completions.create(
-            model="gpt-3.5-turbo",
-            messages=[
-                {"role": "system", "content": """
-                You are a query classifier for a shop assistant system. 
-                Classify customer queries into one of these categories:
-                - max_revenue_product: Questions about which product generated the most revenue (today or on a specific date)
-                - inventory_check: Questions about product availability or stock levels
-                - product_info: Questions about product details, pricing, etc.
-                - general_knowledge: Questions that require general knowledge not related to specific shop data
-                
-                Return ONLY the category as a single word without any explanation.
-                """},
-                {"role": "user", "content": query}
-            ],
-            temperature=0
-        )
+        messages = [
+            {"role": "system", "content": """
+            You are a query classifier for a shop assistant system. 
+            Classify customer queries into one of these categories:
+            - max_revenue_product: Questions about which product generated the most revenue (today or on a specific date)
+            - inventory_check: Questions about product availability or stock levels
+            - product_info: Questions about product details, pricing, etc.
+            - general_knowledge: Questions that require general knowledge not related to specific shop data
+            
+            Return ONLY the category as a single word without any explanation.
+            """},
+            {"role": "user", "content": query}
+        ]
+        
+        response = openai_chat_completion(messages, temperature=0)
         
         # Extract the query type from the response
-        query_type = response.choices[0].message.content.strip().lower()
+        query_type = response["choices"][0]["message"]["content"].strip().lower()
         return query_type
         
     def _extract_parameters(self, query, query_type):
@@ -122,15 +150,19 @@ class QueryRouter:
         Example: {{"product_name": "laptop"}} or {{"date": "2025-04-29"}}
         """
         
-        response = client.chat.completions.create(
-            model="gpt-3.5-turbo",
-            messages=[
-                {"role": "system", "content": "You extract parameters from customer queries for a shop assistant."},
-                {"role": "user", "content": prompt_content}
-            ],
-            temperature=0,
-            response_format={"type": "json_object"}
-        )
+        messages = [
+            {"role": "system", "content": "You extract parameters from customer queries for a shop assistant."},
+            {"role": "user", "content": prompt_content}
+        ]
+        
+        response = openai_chat_completion(messages, temperature=0, response_format={"type": "json_object"})
+        
+        # Parse the JSON response
+        try:
+            parameters = json.loads(response["choices"][0]["message"]["content"])
+            return parameters
+        except json.JSONDecodeError:
+            return {}
         
         # Parse the JSON response
         import json
@@ -142,20 +174,17 @@ class QueryRouter:
             
     def _handle_general_knowledge(self, query):
         """Handle general knowledge queries using OpenAI"""
-        response = client.chat.completions.create(
-            model="gpt-3.5-turbo",
-            messages=[
-                {"role": "system", "content": """
-                You are a helpful assistant for a shop. Answer the customer's question 
-                using your general knowledge. Keep answers brief and focused.
-                """},
-                {"role": "user", "content": query}
-            ],
-            temperature=0.7,
-            max_tokens=150
-        )
-        
-        return response.choices[0].message.content
+        messages = [
+            {"role": "system", "content": """
+            You are a helpful assistant for a shop. Answer the customer's question 
+            using your general knowledge. Keep answers brief and focused.
+            """},
+            {"role": "user", "content": query}
+        ]
+        
+        response = openai_chat_completion(messages, temperature=0.7, max_tokens=150)
+        
+        return response["choices"][0]["message"]["content"]
         
     def _format_response(self, query_type, data):
         """Format the response based on query type and data"""