2B / app /ui /streamlit_app.py
37-AN
Update for Hugging Face Space deployment
2a735cc
import streamlit as st
import os
import sys
import tempfile
from datetime import datetime
import pandas as pd
from typing import List, Dict, Any
import time
import logging
# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
# Add project root to path for imports
sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
# Use relative imports when running as part of the app package
try:
from app.core.agent import AssistantAgent
from app.core.ingestion import DocumentProcessor
from app.core.telegram_bot import TelegramBot
from app.core.chat_history import ChatHistoryManager
from app.utils.helpers import get_document_path, format_sources, save_conversation, copy_uploaded_file
from app.config import (
LLM_MODEL, EMBEDDING_MODEL, TELEGRAM_ENABLED,
TELEGRAM_BOT_TOKEN, TELEGRAM_ALLOWED_USERS,
HF_DATASET_NAME
)
except ImportError:
# Fallback to direct imports if app is not recognized as a package
sys.path.append(os.path.abspath('.'))
from app.core.agent import AssistantAgent
from app.core.ingestion import DocumentProcessor
from app.core.telegram_bot import TelegramBot
from app.core.chat_history import ChatHistoryManager
from app.utils.helpers import get_document_path, format_sources, save_conversation, copy_uploaded_file
from app.config import (
LLM_MODEL, EMBEDDING_MODEL, TELEGRAM_ENABLED,
TELEGRAM_BOT_TOKEN, TELEGRAM_ALLOWED_USERS,
HF_DATASET_NAME
)
# Set page config
st.set_page_config(
page_title="Personal AI Second Brain",
page_icon="🧠",
layout="wide"
)
# Function to initialize the agent safely
@st.cache_resource
def get_agent():
logger.info("Initializing AssistantAgent (should only happen once)")
try:
return AssistantAgent()
except Exception as e:
logger.error(f"Error initializing agent: {e}")
st.error(f"Could not initialize AI assistant: {str(e)}")
# Return a dummy agent as fallback
class DummyAgent:
def query(self, question):
return {
"answer": "I'm having trouble starting up. Please try refreshing the page.",
"sources": []
}
def add_conversation_to_memory(self, *args, **kwargs):
pass
return DummyAgent()
# Function to initialize document processor safely
@st.cache_resource
def get_document_processor(_agent):
"""Initialize document processor with unhashable agent parameter.
The leading underscore in _agent tells Streamlit not to hash this parameter.
"""
logger.info("Initializing DocumentProcessor (should only happen once)")
try:
return DocumentProcessor(_agent.memory_manager)
except Exception as e:
logger.error(f"Error initializing document processor: {e}")
st.error(f"Could not initialize document processor: {str(e)}")
# Return a dummy processor as fallback
class DummyProcessor:
def ingest_file(self, *args, **kwargs):
return ["dummy-id"]
def ingest_text(self, *args, **kwargs):
return ["dummy-id"]
return DummyProcessor()
# Function to initialize chat history manager
@st.cache_resource
def get_chat_history_manager():
logger.info("Initializing ChatHistoryManager")
try:
return ChatHistoryManager(dataset_name=HF_DATASET_NAME)
except Exception as e:
logger.error(f"Error initializing chat history manager: {e}")
st.error(f"Could not initialize chat history: {str(e)}")
# Return a dummy manager as fallback
class DummyHistoryManager:
def load_history(self, *args, **kwargs):
return []
def save_conversation(self, *args, **kwargs):
return True
def sync_to_hub(self, *args, **kwargs):
return False
return DummyHistoryManager()
# Function to initialize Telegram bot
@st.cache_resource
def get_telegram_bot(_agent):
"""Initialize Telegram bot with unhashable agent parameter."""
if not TELEGRAM_ENABLED or not TELEGRAM_BOT_TOKEN:
logger.info("Telegram bot disabled or token missing")
return None
logger.info("Initializing Telegram bot")
try:
bot = TelegramBot(
agent=_agent,
token=TELEGRAM_BOT_TOKEN,
allowed_user_ids=TELEGRAM_ALLOWED_USERS
)
return bot
except Exception as e:
logger.error(f"Error initializing Telegram bot: {e}")
return None
# Initialize session state variables
if "messages" not in st.session_state:
st.session_state.messages = []
if "telegram_status" not in st.session_state:
st.session_state.telegram_status = "Not started"
if "history_filter" not in st.session_state:
st.session_state.history_filter = ""
if "current_tab" not in st.session_state:
st.session_state.current_tab = "Chat"
# Initialize agent and other components with caching
agent = get_agent()
document_processor = get_document_processor(agent)
chat_history_manager = get_chat_history_manager()
telegram_bot = get_telegram_bot(agent)
# Load initial messages from history
if not st.session_state.messages:
try:
recent_history = chat_history_manager.load_history()
# Take the last 10 conversations and convert to messages format
for conv in recent_history[-10:]:
if "user_query" in conv and "assistant_response" in conv:
st.session_state.messages.append({"role": "user", "content": conv["user_query"]})
st.session_state.messages.append({"role": "assistant", "content": conv["assistant_response"]})
except Exception as e:
logger.error(f"Error loading initial history: {e}")
# Main UI
st.title("🧠 Personal AI Second Brain")
# Create tabs for different functionality
tabs = st.tabs(["Chat", "Documents", "History", "Settings"])
# Chat tab
with tabs[0]:
if st.session_state.current_tab != "Chat":
st.session_state.current_tab = "Chat"
# Display chat messages from history
for message in st.session_state.messages:
with st.chat_message(message["role"]):
st.markdown(message["content"])
# Accept user input
if prompt := st.chat_input("Ask me anything..."):
# Add user message to chat history
st.session_state.messages.append({"role": "user", "content": prompt})
# Display user message in chat
with st.chat_message("user"):
st.markdown(prompt)
# Generate and display assistant response
with st.chat_message("assistant"):
message_placeholder = st.empty()
message_placeholder.markdown("Thinking...")
try:
response = agent.query(prompt)
answer = response["answer"]
sources = response["sources"]
# Update the placeholder with the response
message_placeholder.markdown(answer)
# Add assistant response to chat history
st.session_state.messages.append({"role": "assistant", "content": answer})
# Save conversation to history manager
chat_history_manager.save_conversation({
"user_query": prompt,
"assistant_response": answer,
"sources": [s["source"] for s in sources] if sources else [],
"timestamp": datetime.now().isoformat()
})
# Display sources if available
if sources:
with st.expander("Sources"):
st.markdown(format_sources(sources))
# Add to agent's memory
agent.add_conversation_to_memory(prompt, answer)
except Exception as e:
logger.error(f"Error generating response: {e}")
error_message = f"I'm sorry, I encountered an error: {str(e)}"
message_placeholder.markdown(error_message)
st.session_state.messages.append({"role": "assistant", "content": error_message})
# Documents tab (existing functionality)
with tabs[1]:
if st.session_state.current_tab != "Documents":
st.session_state.current_tab = "Documents"
st.header("Upload & Manage Documents")
col1, col2 = st.columns(2)
with col1:
st.subheader("Upload a File")
# Show supported file types info
with st.expander("Supported File Types"):
st.markdown("""
- **PDF** (.pdf) - Best for formatted documents
- **Text** (.txt) - Simple text files
- **CSV** (.csv) - Structured data
- **Word** (.doc, .docx) - Microsoft Word documents
- **Markdown** (.md) - Formatted text
- **HTML** (.html, .htm) - Web pages
Other file types may work but are not fully supported.
""")
uploaded_file = st.file_uploader("Choose a file", type=["pdf", "txt", "csv", "doc", "docx", "md", "html", "htm", "xml", "json"])
if uploaded_file is not None:
# Display file info
file_details = {
"Filename": uploaded_file.name,
"File size": f"{uploaded_file.size / 1024:.1f} KB",
"File type": uploaded_file.type
}
st.json(file_details)
# Handle the uploaded file
if st.button("Process Document"):
with st.spinner("Processing document..."):
status_placeholder = st.empty()
status_placeholder.info("Starting document processing...")
try:
# Create a temporary file with proper error handling
status_placeholder.info("Creating temporary file...")
temp_dir = tempfile.gettempdir()
temp_path = os.path.join(temp_dir, uploaded_file.name)
logger.info(f"Saving uploaded file to temporary path: {temp_path}")
# Write the file data to the temporary file
with open(temp_path, "wb") as temp_file:
temp_file.write(uploaded_file.getvalue())
# Get a path to store the document permanently
status_placeholder.info("Preparing document storage location...")
doc_path = get_document_path(uploaded_file.name)
# Copy the file to the documents directory
logger.info(f"Copying file to documents directory: {doc_path}")
copy_success = copy_uploaded_file(temp_path, doc_path)
if not copy_success:
logger.warning("Using temporary file path instead of documents directory")
doc_path = temp_path
status_placeholder.warning("Using temporary storage (document won't be permanently saved)")
# Ingest the document with retry logic for 403 errors
status_placeholder.info("Analyzing and indexing document content...")
progress_bar = st.progress(0)
max_retries = 3
for attempt in range(max_retries):
try:
progress_bar.progress((attempt * 30) / 100) # Show progress as we attempt
ids = document_processor.ingest_file(temp_path, {"original_name": uploaded_file.name})
progress_bar.progress(100)
break
except Exception as e:
error_str = str(e).lower()
if ("403" in error_str or "forbidden" in error_str or "permission" in error_str) and attempt < max_retries - 1:
status_placeholder.warning(f"Permission error ({attempt+1}/{max_retries}), retrying...")
logger.warning(f"Permission error ({attempt+1}/{max_retries}), retrying...")
time.sleep(1.5) # Add delay between retries
elif attempt < max_retries - 1:
# General retry for any error
status_placeholder.warning(f"Error ({attempt+1}/{max_retries}), retrying...")
logger.warning(f"Error during ingestion ({attempt+1}/{max_retries}): {e}")
time.sleep(1.5)
else:
raise # Re-raise on last attempt
# Clean up the temporary file if different from doc_path
if temp_path != doc_path and os.path.exists(temp_path):
try:
os.unlink(temp_path)
logger.info(f"Temporary file removed: {temp_path}")
except Exception as e:
logger.warning(f"Could not remove temporary file: {e}")
# Check if ingestion was successful based on IDs
if ids and not all(str(id).startswith("error-") for id in ids):
status_placeholder.success(f"✅ Document processed successfully!")
st.balloons() # Celebrate success
else:
status_placeholder.warning("⚠️ Document processed with warnings. Some content may not be fully indexed.")
except Exception as e:
progress_bar = st.progress(100) if 'progress_bar' in locals() else st.progress(0)
logger.error(f"Error processing document: {str(e)}")
status_placeholder.error(f"❌ Error processing document: {str(e)}")
if "403" in str(e) or "forbidden" in str(e).lower():
st.warning("This appears to be a permissions issue. Try using a different file format or using the text input option instead.")
elif "unsupported" in str(e).lower() or "not supported" in str(e).lower() or "no specific loader" in str(e).lower():
st.warning("This file format may not be supported. Try converting to PDF or TXT first.")
with col2:
st.subheader("Add Text Directly")
# Text input for adding content directly
text_content = st.text_area("Enter text to add to your knowledge base:", height=200)
text_title = st.text_input("Give this text a title:")
if st.button("Process Text") and text_content and text_title:
with st.spinner("Processing text..."):
status_placeholder = st.empty()
status_placeholder.info("Processing your text...")
try:
# Process the text content
metadata = {"title": text_title, "source": "direct_input"}
ids = document_processor.ingest_text(text_content, metadata)
if ids:
status_placeholder.success("✅ Text processed successfully!")
else:
status_placeholder.warning("⚠️ Text processed with warnings.")
except Exception as e:
logger.error(f"Error processing text: {str(e)}")
status_placeholder.error(f"❌ Error processing text: {str(e)}")
# History tab (new)
with tabs[2]:
if st.session_state.current_tab != "History":
st.session_state.current_tab = "History"
st.header("Chat History")
# Search and filtering options
col1, col2, col3 = st.columns([2, 1, 1])
with col1:
search_query = st.text_input("Search conversations:", st.session_state.history_filter)
if search_query != st.session_state.history_filter:
st.session_state.history_filter = search_query
with col2:
st.text("Date Range (optional)")
start_date = st.date_input("Start date", None)
with col3:
st.text("\u00A0") # Non-breaking space for alignment
end_date = st.date_input("End date", None)
# Load and filter history
try:
history = chat_history_manager.load_history()
# Apply search filter if provided
if search_query:
history = chat_history_manager.search_conversations(search_query)
# Apply date filtering if provided
if start_date or end_date:
# Convert datetime.date to datetime.datetime for filtering
start_datetime = datetime.combine(start_date, datetime.min.time()) if start_date else None
end_datetime = datetime.combine(end_date, datetime.max.time()) if end_date else None
history = chat_history_manager.get_conversations_by_date(start_datetime, end_datetime)
# Display history
if not history:
st.info("No conversation history found matching your criteria.")
else:
# Sort by timestamp (newest first)
history.sort(key=lambda x: x.get("timestamp", ""), reverse=True)
# Create a DataFrame for display
df = pd.DataFrame(history)
if not df.empty:
# Select and rename columns for display
if all(col in df.columns for col in ["timestamp", "user_query", "assistant_response"]):
display_df = df[["timestamp", "user_query", "assistant_response"]]
display_df = display_df.rename(columns={
"timestamp": "Date",
"user_query": "Your Question",
"assistant_response": "AI Response"
})
# Format timestamp
if "Date" in display_df.columns:
display_df["Date"] = pd.to_datetime(display_df["Date"]).dt.strftime('%Y-%m-%d %H:%M')
# Truncate long text
for col in ["Your Question", "AI Response"]:
if col in display_df.columns:
display_df[col] = display_df[col].apply(lambda x: x[:100] + "..." if isinstance(x, str) and len(x) > 100 else x)
# Display as table
st.dataframe(display_df, use_container_width=True)
# Add option to view full conversation
if not df.empty:
selected_idx = st.selectbox("Select conversation to view details:",
range(len(df)),
format_func=lambda i: f"{df.iloc[i].get('timestamp', 'Unknown')} - {df.iloc[i].get('user_query', '')[:30]}...")
if selected_idx is not None:
selected_conv = df.iloc[selected_idx]
st.subheader("Conversation Details")
st.markdown("**Your Question:**")
st.markdown(selected_conv.get("user_query", ""))
st.markdown("**AI Response:**")
st.markdown(selected_conv.get("assistant_response", ""))
# Display sources if available
if "sources" in selected_conv and selected_conv["sources"]:
st.markdown("**Sources:**")
for src in selected_conv["sources"]:
st.markdown(f"- {src}")
# Option to use this conversation in chat
if st.button("Continue this conversation"):
# Add to current chat session
st.session_state.messages.append({"role": "user", "content": selected_conv.get("user_query", "")})
st.session_state.messages.append({"role": "assistant", "content": selected_conv.get("assistant_response", "")})
# Switch to chat tab
st.session_state.current_tab = "Chat"
st.experimental_rerun()
else:
st.error("Unexpected history format. Some columns are missing.")
else:
st.info("No conversation history found.")
except Exception as e:
logger.error(f"Error displaying history: {e}")
st.error(f"Error loading conversation history: {str(e)}")
# Sync to Hugging Face Hub button
if HF_DATASET_NAME:
if st.button("Sync History to Hugging Face Hub"):
with st.spinner("Syncing history..."):
success = chat_history_manager.sync_to_hub()
if success:
st.success("History successfully synced to Hugging Face Hub!")
else:
st.error("Failed to sync history. Check logs for details.")
# Settings tab (new)
with tabs[3]:
if st.session_state.current_tab != "Settings":
st.session_state.current_tab = "Settings"
st.header("Settings")
# System information
st.subheader("System Information")
system_info = {
"LLM Model": LLM_MODEL,
"Embedding Model": EMBEDDING_MODEL,
"HF Dataset": HF_DATASET_NAME or "Not configured",
"Telegram Enabled": "Yes" if TELEGRAM_ENABLED else "No"
}
for key, value in system_info.items():
st.markdown(f"**{key}:** {value}")
# Telegram settings
st.subheader("Telegram Integration")
telegram_status = "Not configured"
if telegram_bot:
telegram_status = st.session_state.telegram_status
st.markdown(f"**Status:** {telegram_status}")
col1, col2 = st.columns(2)
with col1:
if telegram_bot and st.session_state.telegram_status != "Running":
if st.button("Start Telegram Bot"):
try:
success = telegram_bot.start()
if success:
st.session_state.telegram_status = "Running"
st.success("Telegram bot started!")
else:
st.error("Failed to start Telegram bot. Check logs for details.")
except Exception as e:
logger.error(f"Error starting Telegram bot: {e}")
st.error(f"Error: {str(e)}")
with col2:
if telegram_bot and st.session_state.telegram_status == "Running":
if st.button("Stop Telegram Bot"):
try:
telegram_bot.stop()
st.session_state.telegram_status = "Stopped"
st.info("Telegram bot stopped.")
except Exception as e:
logger.error(f"Error stopping Telegram bot: {e}")
st.error(f"Error: {str(e)}")
if telegram_bot:
with st.expander("Telegram Bot Settings"):
st.markdown("""
To configure the Telegram bot, set these environment variables:
- `TELEGRAM_ENABLED`: Set to `true` to enable the bot
- `TELEGRAM_BOT_TOKEN`: Your Telegram bot token
- `TELEGRAM_ALLOWED_USERS`: Comma-separated list of Telegram user IDs (optional)
""")
if telegram_bot.allowed_user_ids:
st.markdown("**Allowed User IDs:**")
for user_id in telegram_bot.allowed_user_ids:
st.markdown(f"- {user_id}")
else:
st.markdown("The bot will respond to all users (no user restrictions configured).")
# Show Telegram bot instructions
st.markdown("### Telegram Bot Commands")
st.markdown("""
- **/start**: Start a conversation with the bot
- **/help**: Shows available commands
- **/search**: Use `/search your query` to search your knowledge base
- **Direct messages**: Send any message to chat with your second brain
#### How to Set Up Your Telegram Bot
1. Talk to [@BotFather](https://t.me/botfather) on Telegram
2. Use the `/newbot` command to create a new bot
3. Get your bot token and add it to your `.env` file
4. Set `TELEGRAM_ENABLED=true` in your `.env` file
5. To find your Telegram user ID, talk to [@userinfobot](https://t.me/userinfobot)
""")
else:
st.info("Telegram integration is not enabled. Configure your .env file to enable it.")
# Settings for Hugging Face Dataset persistence
st.subheader("Hugging Face Dataset Settings")
if HF_DATASET_NAME:
st.markdown(f"**Dataset Name:** {HF_DATASET_NAME}")
st.markdown(f"**Local History File:** {chat_history_manager.local_file}")
# HF Dataset instructions
with st.expander("Setup Instructions"):
st.markdown("""
### Setting up Hugging Face Dataset Persistence
1. Create a private dataset repository on Hugging Face Hub
2. Set your API token in the `.env` file as `HF_API_KEY`
3. Set your dataset name as `HF_DATASET_NAME` (format: username/repo-name)
Your chat history will be automatically synced to the Hub.
""")
else:
st.info("Hugging Face Dataset persistence is not configured. Set HF_DATASET_NAME in your .env file.")
# Run Telegram bot on startup if enabled
if telegram_bot and TELEGRAM_ENABLED and st.session_state.telegram_status == "Not started":
try:
success = telegram_bot.start()
if success:
st.session_state.telegram_status = "Running"
logger.info("Telegram bot started automatically")
except Exception as e:
logger.error(f"Error auto-starting Telegram bot: {e}")
st.session_state.telegram_status = "Error"
if __name__ == "__main__":
# This is used when running the file directly
pass