Update app.py

app.py

@@ -4,7 +4,7 @@ from dotenv import load_dotenv
 import os
 import re
 import json
-from helper.openai_api import chat_completion
+from helper.openai_api import chat_completion  # Ensure this import is correct
 
 # Load environment variables from .env file
 load_dotenv()
@@ -16,6 +16,10 @@ VERIFY_TOKEN = os.getenv('VERIFY_TOKEN')
 PAGE_ACCESS_TOKEN = os.getenv('PAGE_ACCESS_TOKEN')
 OPENAI_API_KEY = os.getenv('OPENAI_API_KEY')
 
+# Check if environment variables are set
+if not all([VERIFY_TOKEN, PAGE_ACCESS_TOKEN, OPENAI_API_KEY]):
+    raise EnvironmentError("Some environment variables are missing. Please check your .env file.")
+
 # Debugging: Print the loaded environment variables (remove in production)
 print(f"VERIFY_TOKEN: {VERIFY_TOKEN}")
 print(f"PAGE_ACCESS_TOKEN: {PAGE_ACCESS_TOKEN}")
@@ -71,41 +75,23 @@ def webhook():
                     # Get response with possible image URLs
                     response_data = chat_completion(message_text, sender_id)
                     print("ChatBot Response:\n", response_data)
+                    response_data = parse_response(response_data)
+                    print("#"*10)
+                    print("Parsed Response:\n", response_data)
+                    print("#"*10)
+                    response_text = response_data.get('msg', '')
+                    image_urls = response_data.get('image_urls', [])
+
+                    print(f"ChatBot Response Text: {response_text}")
+                    print(f"ChatBot Response Image URLs: {image_urls}")
+
+                    # Send text and/or images
+                    if response_text:
+                        send_message(sender_id, response_text)
+                    if image_urls:
+                        for url in image_urls:
+                            send_image(sender_id, url)
 
-                    # Ensure response_data is a dictionary
-                    if isinstance(response_data, str):
-                        try:
-                            response_data = json.loads(response_data)
-                        except json.JSONDecodeError:
-                            # Extract text and URLs from the string
-                            response_text, image_urls = extract_text_and_urls(response_data)
-                            print(f"Extracted Text: {response_text}")
-                            print(f"Extracted Image URLs: {image_urls}")
-
-                            # Send text and/or images
-                            if response_text:
-                                send_message(sender_id, response_text)
-                            if image_urls:
-                                for url in image_urls:
-                                    send_image(sender_id, url)
-                            response_data = {}
-
-                    if isinstance(response_data, dict):
-                        response_text = response_data.get('msg', '')
-                        image_urls = response_data.get('image_urls', [])
-
-                        print(f"ChatBot Response Text: {response_text}")
-                        print(f"ChatBot Response Image URLs: {image_urls}")
-
-                        # Send text and/or images
-                        if response_text:
-                            send_message(sender_id, response_text)
-                        if image_urls:
-                            for url in image_urls:
-                                send_image(sender_id, url)
-                    else:
-                        print("Error: ChatBot response is not a dictionary.")
-                        send_message(sender_id, "Sorry, I couldn't process your request.")
                 except Exception as e:
                     print(f"Exception occurred: {e}")
                     send_message(sender_id, "Sorry, something went wrong. Please try again later.")
@@ -117,14 +103,46 @@ def webhook():
 
         return jsonify({'status': 'ok'})
 
-def extract_text_and_urls(input_string):
+def parse_response(input_string):
     # Regex pattern to find URLs
     url_pattern = re.compile(r'https?://\S+|www\.\S+')
-    # Find all URLs using the pattern
     urls = re.findall(url_pattern, input_string)
-    # Remove URLs from the input string to get the remaining text
-    text = re.sub(url_pattern, '', input_string).strip()
-    return text, urls
+
+    # Clean URLs by removing unwanted characters
+    cleaned_urls = [re.sub(r'[(){}"\']', '', url) for url in urls]
+
+    # Extract JSON-like dictionaries from the input string
+    json_pattern = re.compile(r'\{.*?\}')
+    json_matches = json_pattern.findall(input_string)
+    extracted_text = input_string
+
+    for json_str in json_matches:
+        try:
+            json_data = json.loads(json_str)
+            if isinstance(json_data, dict):
+                extracted_text = json_data.get('msg', extracted_text)
+                image_urls = json_data.get('image_url', [])
+                if isinstance(image_urls, str):
+                    cleaned_urls.append(re.sub(r'[(){}"\']', '', image_urls))
+                elif isinstance(image_urls, list):
+                    cleaned_urls.extend([re.sub(r'[(){}"\']', '', url) for url in image_urls])
+        except json.JSONDecodeError:
+            continue
+
+    # Remove URLs from the extracted text
+    text = re.sub(url_pattern, '', extracted_text).strip()
+
+    # Further clean the text from any leftover JSON artifacts and unwanted parts
+    text = re.sub(r'[\)\{\}\[\]\"\(\']', '', text).strip()
+    text = re.sub(r'msg:', '', text).strip()
+    text = re.sub(r'image_url:', '', text).strip()
+    text = text.replace('\n', '').strip()
+    text = text.replace("  ", " ")
+    
+    return {
+        'msg': text,
+        'image_urls': list(set(cleaned_urls))
+    }
 
 def send_message(recipient_id, message_text):
     url = f'https://graph.facebook.com/v12.0/me/messages?access_token={PAGE_ACCESS_TOKEN}'
@@ -205,4 +223,4 @@ def mark_seen(recipient_id):
     print(f"Mark seen response: {response.status_code}, {response.text}")
 
 if __name__ == '__main__':
-    app.run(host="0.0.0.0", port=7860, debug=True)
+    app.run(host="0.0.0.0", port=7860, debug=True)