Update app.py

app.py

@@ -22,6 +22,8 @@ from datetime import datetime
 import pandas as pd
 import re
 import random
+from bs4 import BeautifulSoup
+
 # Initialize logging and load environment variables
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
@@ -242,7 +244,18 @@ def search_knowledge_base(query):
     except Exception as e:
         logger.error(f"Error searching knowledge base: {e}")
         return ["Error occurred during knowledge base search"]
-
+def sanitize_and_normalize(text):
+    # Use BeautifulSoup to parse and clean the HTML
+    soup = BeautifulSoup(text, 'html.parser')
+    
+    # Remove unwanted tags or attributes if necessary
+    for tag in soup.find_all(['i', 'em']):  # Example: removing <i> and <em> tags
+        tag.unwrap()
+    
+    # Normalize the text by ensuring proper spacing
+    normalized_text = ' '.join(soup.get_text().split())
+    return normalized_text
+    
 def google_search(query):
     try:
         search_client = serpapi.Client(api_key=serper_api_key)
@@ -291,7 +304,11 @@ prompt_message = f"""
                                            Strictly Do not mention"Trust" , "Stability" , "Development" , "Relationship" , "Competence" , "benefit" , "vision" words in your response . 
                                            Do not generate imaginary/Creative content . It should not look like creative story . 
                                            When asking for trust builders it has tibe returned short/concised  facts as dot points.
-                                           text with dollar figures needs to be concised and readable with spacing .
+                                           text with dollar figures needs to be concised and readable with spacing.
+                                           
+                                           
+                                           
+                                           
 
                                                 
                               
@@ -518,16 +535,18 @@ if prompt :
         try:
             # Generate response using the agent executor
             output = agent_executor.invoke({
-                    "input": f" {prompt} . Always INCLUDE & Be specific with numbers, dates, people, and  [$ dollar amounts]   (include them).* Search and provide sources for all the results. Respond directly to the query without any introductory phrases or meta-commentary. Your response should be natural and read as if it's addressing the query immediately, without any preamble. Use creative techniques like metaphors, analogies, and juxtaposition to enhance the content. Interconnect ideas smoothly to create meaning for the reader.Always write in nicely flowing text and create an active language headline. Avoid using colons in the sub-headings and incorporate a creative technique in every headline and sub-header. Interweave the examples contextually to create meaning for donors and beneficiaries. When asked about to write annual report ONLY THEN  write report in 2-3 pargagraphs(with no sub-headings) and end then  : do mention  these 3 in sub headings with example related to the report: -example proof points , -Heuristics used , -creative techniques used. DO NOT USE COLONS in the sub-headers Otherwise ignore it . DONOT WRITE  CONCLUSION in the end related to given prompt. remove all tags present with your response so response format is great",
+                    "input": f" {prompt} . Always INCLUDE & Be specific with numbers, dates, people, and  [$ dollar amounts]   (include them).* Search and provide sources for all the results. Respond directly to the query without any introductory phrases or meta-commentary. Your response should be natural and read as if it's addressing the query immediately, without any preamble. Use creative techniques like metaphors, analogies, and juxtaposition to enhance the content. Interconnect ideas smoothly to create meaning for the reader.Always write in nicely flowing text and create an active language headline. Avoid using colons in the sub-headings and incorporate a creative technique in every headline and sub-header. Interweave the examples contextually to create meaning for donors and beneficiaries. When asked about to write annual report ONLY THEN  write report in 2-3 pargagraphs(with no sub-headings) and end then  : do mention  these 3 in sub headings with example related to the report: -example proof points , -Heuristics used , -creative techniques used (DO-NOT MENTION THIS WHEN ASKED TO WRITE EMAIL or news letter). When asked about email  news letter or news letter it should be in proper email format . DO NOT USE COLONS in the sub-headers Otherwise ignore it . DONOT WRITE  CONCLUSION in the end related to given prompt. remove all tags present with your response so response format is great",
                 "chat_history": st.session_state.chat_history
             })
             full_response = output["output"]
 
+            full_response = sanitize_and_normalize(full_response)
 
             # Display the response
-            trust_tip = random.choice(trust_tips)
-            suggestion = random.choice(suggestions)
-            full_response += f"\n\n**Trust Tip**: {trust_tip}  \n \n **Suggestion**: {suggestion}"
+            if "**Trust Tip**" not in full_response and "**Suggestion**" not in full_response:
+                trust_tip = random.choice(trust_tips)
+                suggestion = random.choice(suggestions)
+                full_response += f"\n\n**Trust Tip**: {trust_tip}  \n\n **Suggestion**: {suggestion}"
             
             st.write(full_response, unsafe_allow_html=True)