Update app.py

app.py

@@ -12,7 +12,6 @@ import json
 import uuid  # 🎲 For generating unique IDs
 from urllib.parse import quote  # 🔗 For encoding URLs
 from gradio_client import Client  # 🌐 For connecting to Gradio apps
-from bs4 import BeautifulSoup  # 🍲 For parsing HTML content
 
 # 🎉 Welcome to our fun-filled Cosmos DB and GitHub Integration app!
 st.set_page_config(layout="wide")
@@ -24,9 +23,11 @@ CONTAINER_NAME = os.environ.get("COSMOS_CONTAINER_NAME")
 Key = os.environ.get("Key")  # 🔑 Don't forget your key!
 
 # 🏠 Your local app URL (Change this to your app's URL)
-#LOCAL_APP_URL = "http://localhost:8501"
-LOCAL_APP_URL = "https://huggingface.co/spaces/awacke1/AzureCosmosDBUI"
+LOCAL_APP_URL = "http://localhost:8501"
 
+# 🤖 OpenAI configuration
+#openai.api_key = os.environ.get("OPENAI_API_KEY")
+#MODEL = "gpt-3.5-turbo"  # Replace with your desired model
 
 # 🐙 GitHub configuration
 def download_github_repo(url, local_path):
@@ -155,7 +156,7 @@ def archive_current_container(database_name, container_name, client):
         return f"An error occurred while archiving data: {str(e)} 😢"
 
 # 🔍 Search Glossary function
-def search_glossary(query, container):
+def search_glossary(query):
     # 🕵️‍♂️ Searching the glossary for: query
     all_results = ""
     st.markdown(f"- {query}")
@@ -183,33 +184,10 @@ def search_glossary(query, container):
     st.write('🔍 Run of Multi-Agent System Paper References is Complete')
     responseall = response2 + response1[0] + response1[1]
     st.markdown(responseall)
-    
-    # Save the response to a file
-    filename = create_and_save_file(responseall, file_type="md", prompt=query, is_image=False, should_save=True)
-    
-    # Convert markdown to JSON structure
-    json_content = markdown_to_json(responseall)
-    
-    # Create a record to insert into Cosmos DB
-    record = {
-        'id': generate_unique_id(),
-        'query': query,
-        'content': json_content,
-        'filename': filename,
-        'timestamp': datetime.now().isoformat()
-    }
-    
-    # Insert the record into Cosmos DB
-    success, message = insert_record(container, record)
-    if success:
-        st.success("Record inserted into Cosmos DB successfully!")
-    else:
-        st.error(f"Failed to insert record into Cosmos DB: {message}")
-    
     return responseall
 
 # 📝 Function to process text input
-def process_text(text_input, container):
+def process_text(text_input):
     if text_input:
         if 'messages' not in st.session_state:
             st.session_state.messages = []
@@ -220,45 +198,27 @@ def process_text(text_input, container):
             st.markdown(text_input)
             
         with st.chat_message("assistant"):
-        #    completion = openai.ChatCompletion.create(
-        #        model=MODEL,
-        #        messages=[
-        #            {"role": m["role"], "content": m["content"]}
-        #            for m in st.session_state.messages
-        #        ],
-        #        stream=False
-        #    )
-        #    return_text = completion.choices[0].message.content
-        #    st.write("Assistant: " + return_text)
-        #    filename = create_and_save_file(return_text, file_type="md", prompt=text_input, is_image=False, should_save=True)
-        #    st.session_state.messages.append({"role": "assistant", "content": return_text})
-            
-            # Convert markdown to JSON structure
-            #json_content = markdown_to_json(return_text)
-            json_content = markdown_to_json(container)
-            
-            # Create a record to insert into Cosmos DB
-            record = {
-                'id': generate_unique_id(),
-                'query': text_input,
-                'content': json_content,
-                'filename': filename,
-                'timestamp': datetime.now().isoformat()
-            }
-            
-            # Insert the record into Cosmos DB
-            success, message = insert_record(container, record)
-            if success:
-                st.success("Record inserted into Cosmos DB successfully!")
-            else:
-                st.error(f"Failed to insert record into Cosmos DB: {message}")
+            completion = openai.ChatCompletion.create(
+                model=MODEL,
+                messages=[
+                    {"role": m["role"], "content": m["content"]}
+                    for m in st.session_state.messages
+                ],
+                stream=False
+            )
+            return_text = completion.choices[0].message.content
+            st.write("Assistant: " + return_text)
+            filename = generate_filename(text_input, "md")
+
+            create_and_save_file(return_text, file_type="md", prompt=text_input, is_image=False, should_save=True)
+            st.session_state.messages.append({"role": "assistant", "content": return_text})
 
 # 📄 Function to generate a filename
 def generate_filename(text, file_type):
     # 📝 Generate a filename based on the text input
     safe_text = "".join(c if c.isalnum() or c in (' ', '.', '_') else '_' for c in text)
     safe_text = "_".join(safe_text.strip().split())
-    filename = f"{safe_text}"
+    filename = f"{safe_text}.{file_type}"
     return filename
 
 # 🏷️ Function to extract markdown title
@@ -270,26 +230,6 @@ def extract_markdown_title(content):
             return line.lstrip('#').strip()
     return None
 
-# 📄 Function to convert markdown to JSON structure
-def markdown_to_json(md_content):
-    html = markdown.markdown(md_content)
-    soup = BeautifulSoup(html, 'html.parser')
-    json_content = []
-    for element in soup.descendants:
-        if element.name == 'h1':
-            json_content.append({'type': 'heading', 'level': 1, 'text': element.get_text()})
-        elif element.name == 'h2':
-            json_content.append({'type': 'heading', 'level': 2, 'text': element.get_text()})
-        elif element.name == 'h3':
-            json_content.append({'type': 'heading', 'level': 3, 'text': element.get_text()})
-        elif element.name == 'p':
-            json_content.append({'type': 'paragraph', 'text': element.get_text()})
-        elif element.name == 'a':
-            json_content.append({'type': 'link', 'href': element.get('href'), 'text': element.get_text()})
-        elif element.name == 'img':
-            json_content.append({'type': 'image', 'src': element.get('src'), 'alt': element.get('alt')})
-    return json_content
-
 # 💾 Function to create and save a file
 def create_and_save_file(content, file_type="md", prompt=None, is_image=False, should_save=True):
     """
@@ -308,25 +248,14 @@ def create_and_save_file(content, file_type="md", prompt=None, is_image=False, s
         if title_from_content:
             filename = generate_filename(title_from_content, file_type)
 
-    # Add date time to filename
-    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-    filename = f"{filename}_{timestamp}.{file_type}"
-
-    # Ensure the directory exists
-    save_dir = "saved_files"
-    os.makedirs(save_dir, exist_ok=True)
-
-    # Full path
-    filepath = os.path.join(save_dir, filename)
-
     # Step 3: Save the file
-    with open(filepath, "w", encoding="utf-8") as f:
+    with open(filename, "w", encoding="utf-8") as f:
         if is_image:
             f.write(content)
         else:
-            f.write(content)
+            f.write(prompt + "\n\n" + content)
 
-    return filename  # Return just the filename
+    return filename
 
 # 🎈 Let's modify the main app to be more fun!
 def main():
@@ -353,16 +282,10 @@ def main():
     # ⚙️ q= Run ArXiv search from query parameters
     try:
         query_params = st.query_params
-        query = (query_params.get('q') or query_params.get('query') or [''])
+        query = (query_params.get('q') or query_params.get('query') or [''])[0]
         if query:
             # 🕵️‍♂️ We have a query! Let's process it!
-            # Ensure the client and container are initialized
-            if 'client' in st.session_state and st.session_state.client:
-                database = st.session_state.client.get_database_client(DATABASE_NAME)
-                container = database.get_container_client(CONTAINER_NAME)
-                process_text(query, container)
-            else:
-                st.error("Database client not initialized.")
+            process_text(query)
             st.stop()  # Stop further execution
     except Exception as e:
         st.markdown(' ')
@@ -610,36 +533,6 @@ def main():
                 else:
                     st.info("No documents to display. 🧐")
                     
-            # 🗂️ List saved files
-            st.sidebar.title("🗂️ Saved Files")
-
-            # Get list of saved files
-            def get_saved_files():
-                save_dir = "saved_files"
-                if not os.path.exists(save_dir):
-                    return []
-                files = [f for f in os.listdir(save_dir) if os.path.isfile(os.path.join(save_dir, f))]
-                # Sort files descending by date (assuming filenames include timestamp)
-                files.sort(reverse=True)
-                return files
-
-            saved_files = get_saved_files()
-            if saved_files:
-                selected_file = st.sidebar.radio("Select a saved file", saved_files)
-                st.session_state.selected_file = selected_file
-            else:
-                st.sidebar.info("No saved files available.")
-
-            # In the main area, display the selected file content
-            if 'selected_file' in st.session_state:
-                file_path = os.path.join("saved_files", st.session_state.selected_file)
-                if os.path.exists(file_path):
-                    with open(file_path, "r", encoding="utf-8") as f:
-                        file_content = f.read()
-                    st.markdown(file_content)
-                else:
-                    st.error("File not found.")
-            
             # 🐙 GitHub section
             st.subheader("🐙 GitHub Operations")
             github_token = os.environ.get("GITHUB")  # Read GitHub token from environment variable