Update app.py

app.py

@@ -145,56 +145,25 @@ def send_feedback_via_email(name, email, feedback):
         st.error(f"Error sending email: {e}")
 
 def clean_text(text):
-    # Replace escaped newlines with actual newlines
-    text = text.replace('\\n', '\n')
-
-    # Convert <a> tags to Markdown links
-    def convert_links(match):
-        url = match.group(1)
-        link_text = match.group(2)
-        return f"[{link_text}]({url})"
-
-    # Handle <a> tags to preserve clickable URLs in Markdown format
-    text = re.sub(r'<a [^>]*href="([^"]+)"[^>]*>(.*?)</a>', convert_links, text)
+    # Parse HTML with BeautifulSoup
+    soup = BeautifulSoup(text, 'html.parser')
     
-    # Preserve and correctly format headings (e.g., **Heading**)
-    def preserve_headings(match):
-        content = match.group(2)
-        return f"**{content.strip()}**"
-
-    text = re.sub(r'(<h[1-6][^>]*>)(.*?)(</h[1-6]>)', preserve_headings, text)
-
-    # Remove all other HTML tags including <span>, <p>, <b>, <i>, etc.
-    text = re.sub(r'<[^>]+>', '', text)
+    # Convert <a> tags to Markdown links
+    for a in soup.find_all('a'):
+        a.replace_with(f"[{a.get_text()}]({a['href']})")
     
-    # Clean up any redundant asterisks or special characters
-    text = re.sub(r'\*|\−|\∗', '', text)
+    # Remove unwanted tags but preserve text
+    for tag in ['span', 'i', 'b']:
+        for element in soup.find_all(tag):
+            element.unwrap()  # Remove the tag but keep the content
     
-    # Fix merged words by adding a space between camel case words
-    text = re.sub(r'([a-z])([A-Z])', r'\1 \2', text)
-
-    # Split the text into paragraphs
-    paragraphs = text.split('\n\n')
-
-    cleaned_paragraphs = []
-    for paragraph in paragraphs:
-        lines = paragraph.split('\n')
-        cleaned_lines = []
-        for line in lines:
-            # Remove extra spaces
-            cleaned_line = re.sub(r'\s+', ' ', line).strip()
-            cleaned_lines.append(cleaned_line)
-        
-        # Join the lines within each paragraph
-        cleaned_paragraph = '\n'.join(cleaned_lines)
-        cleaned_paragraphs.append(cleaned_paragraph)
+    # Get the cleaned text
+    cleaned_text = soup.get_text()
     
-    # Join the paragraphs back together
-    cleaned_text = '\n\n'.join(para for para in cleaned_paragraphs if para)
+    # Fix text formatting (e.g., add newlines for paragraphs)
+    cleaned_text = re.sub(r'\n\s*\n', '\n\n', cleaned_text)
     
     return cleaned_text
-
-
 def side():        
     with st.sidebar.form(key='feedback_form'):
     
@@ -581,57 +550,57 @@ suggestions = [
 ]
 
 
-
-prompt = st.chat_input("")
-if prompt :
-
-    st.session_state["chat_started"] = True
-    # Add user message to chat history
-    st.session_state.chat_history.append({"role": "user", "content": prompt})
+with st.container():
+    prompt = st.chat_input("")
+    if prompt :
     
-    # Display user message
-    with st.chat_message("user"):
-        st.markdown(prompt)
-
-    # Generate AI response
-    with st.chat_message("assistant"):
-        full_response = ""
+        st.session_state["chat_started"] = True
+        # Add user message to chat history
+        st.session_state.chat_history.append({"role": "user", "content": prompt})
         
-        try:
-            # Generate response using the agent executor
-            output = agent_executor.invoke({
-                    "input": f"{prompt} Always Be specific with numbers, dates, people, and $dollar amounts( Do Mention them) provide Latest information . Always Be specific and provide bold point examples and also Embed source links directly next to each point.  All outputs should be in structured format.Remove itallic and span tags  ",
-
-                "chat_history": st.session_state.chat_history
-            })
-            full_response = output["output"]
-            full_response= replace_terms(full_response)
-            
-            cleaned_text = clean_text(full_response)
-
+        # Display user message
+        with st.chat_message("user"):
+            st.markdown(prompt)
+    
+        # Generate AI response
+        with st.chat_message("assistant"):
+            full_response = ""
             
-            #cleaned_text = re.sub(r'</span>', '', cleaned_text) 
-            #cleaned_text = re.sub(r'<span[^>]*>', '', cleaned_text)
-            # Display the response
-            if "**Trust Tip**" not in cleaned_text and "**Suggestion**" not in cleaned_text:
-                trust_tip = random.choice(trust_tips)
-                suggestion = random.choice(suggestions)
-                cleaned_text += f"\n\n**Trust Tip**: {trust_tip}  \n\n **Suggestion**: {suggestion}"
+            try:
+                # Generate response using the agent executor
+                output = agent_executor.invoke({
+                        "input": f"{prompt} Always Be specific with numbers, dates, people, and $dollar amounts( Do Mention them) provide Latest information . Always Be specific and provide bold point examples and also Embed source links directly next to each point.  All outputs should be in structured format.Remove itallic and span tags  ",
+    
+                    "chat_history": st.session_state.chat_history
+                })
+                full_response = output["output"]
+                full_response= replace_terms(full_response)
                 
-            st.markdown(cleaned_text,unsafe_allow_html=True)
-
-
-
-
-
-
-        except Exception as e:
-            logger.error(f"Error generating response: {e}")
-            full_response = "I apologize, but an error occurred while generating the response. Please try again."
-            st.write(full_response)
-
-    # Add AI response to chat history
-    st.session_state.chat_history.append({"role": "assistant", "content": full_response})
-    copy_to_clipboard(full_response)
+                cleaned_text = clean_text(full_response)
+    
+                
+                #cleaned_text = re.sub(r'</span>', '', cleaned_text) 
+                #cleaned_text = re.sub(r'<span[^>]*>', '', cleaned_text)
+                # Display the response
+                if "**Trust Tip**" not in cleaned_text and "**Suggestion**" not in cleaned_text:
+                    trust_tip = random.choice(trust_tips)
+                    suggestion = random.choice(suggestions)
+                    cleaned_text += f"\n\n**Trust Tip**: {trust_tip}  \n\n **Suggestion**: {suggestion}"
+                    
+                st.markdown(cleaned_text,unsafe_allow_html=True)
+    
+    
+    
+    
+    
+    
+            except Exception as e:
+                logger.error(f"Error generating response: {e}")
+                full_response = "I apologize, but an error occurred while generating the response. Please try again."
+                st.write(full_response)
+    
+        # Add AI response to chat history
+        st.session_state.chat_history.append({"role": "assistant", "content": full_response})
+        copy_to_clipboard(full_response)