app.py

from dotenv import load_dotenv
import gradio as gr
import os
from llama_index.core import StorageContext, load_index_from_storage, VectorStoreIndex, SimpleDirectoryReader, ChatPromptTemplate, Settings
from llama_index.llms.huggingface import HuggingFaceInferenceAPI
from llama_index.embeddings.huggingface import HuggingFaceEmbedding
import firebase_admin
from firebase_admin import db, credentials
import datetime
import uuid
import random

def select_random_name():
    names = ['Clara', 'Lily']
    return random.choice(names)

# Load environment variables
load_dotenv()

# Authenticate to Firebase
cred = credentials.Certificate("redfernstech-fd8fe-firebase-adminsdk-g9vcn-0537b4efd6.json")
firebase_admin.initialize_app(cred, {"databaseURL": "https://redfernstech-fd8fe-default-rtdb.firebaseio.com/"})

# Configure Llama index settings
Settings.llm = HuggingFaceInferenceAPI(
    model_name="facebook/rag-token-nq",
    tokenizer_name="facebook/rag-token-nq",
    context_window=3000,
    token=os.getenv("HF_TOKEN"),
    max_new_tokens=512,
    generate_kwargs={"temperature": 0.1},
)
Settings.embed_model = HuggingFaceEmbedding(
    model_name="BAAI/bge-small-en-v1.5"
)

# Define directories for storage and data
PERSIST_DIR = "db"
PDF_DIRECTORY = 'data'  # Directory containing PDFs

# Ensure directories exist
os.makedirs(PDF_DIRECTORY, exist_ok=True)
os.makedirs(PERSIST_DIR, exist_ok=True)

# Variable to store current chat conversation
current_chat_history = []

def data_ingestion_from_directory():
    # Use SimpleDirectoryReader on the directory containing PDF files
    documents = SimpleDirectoryReader(PDF_DIRECTORY).load_data()
    storage_context = StorageContext.from_defaults()
    index = VectorStoreIndex.from_documents(documents)
    index.storage_context.persist(persist_dir=PERSIST_DIR)

def handle_query(query):
    chat_text_qa_msgs = [
        (
            "user",
            """
            You're Clara, working in customer care at RedfernsTech. Continue the conversation flow, giving responses within 10-15 words only. Convert all questions into company-related inquiries. Use the entire conversation context to craft responses, ensuring each answer relates to previous questions and answers. If you don't know the answer, say, 'You can directly contact us at +91 7972628566 or email us at [email protected]'
            {context_str}
            Question:
            {query_str}
            """
        )
    ]
    text_qa_template = ChatPromptTemplate.from_messages(chat_text_qa_msgs)

    # Load index from storage
    storage_context = StorageContext.from_defaults(persist_dir=PERSIST_DIR)
    index = load_index_from_storage(storage_context)

    # Use chat history to enhance response
    context_str = ""
    for past_query, response in reversed(current_chat_history):
        if past_query.strip():
            context_str += f"User asked: '{past_query}'\nBot answered: '{response}'\n"

    query_engine = index.as_query_engine(text_qa_template=text_qa_template, context_str=context_str)
    answer = query_engine.query(query)

    if hasattr(answer, 'response'):
        response = answer.response
    elif isinstance(answer, dict) and 'response' in answer:
        response = answer['response']
    else:
        response = "Sorry, I couldn't find an answer."

    # Update current chat history
    current_chat_history.append((query, response))

    return response

def predict(message, history):
    logo_html = '''
    <div class="circle-logo">
      <img src="https://rb.gy/8r06eg" alt="FernAi">
    </div>
    '''
    response = handle_query(message)
    response_with_logo = f'<div class="response-with-logo">{logo_html}<div class="response-text">{response}</div></div>'
    return response_with_logo

def save_chat_message(session_id, message_data):
    ref = db.reference(f'/chat_history/{session_id}')
    ref.push().set(message_data)

def chat_interface(message, history):
    try:
        # Generate a unique session ID for this chat session
        session_id = str(uuid.uuid4())

        # Process the user message and generate a response
        response = handle_query(message)

        # Capture the message data
        message_data = {
            "sender": "user",
            "message": message,
            "response": response,
            "timestamp": datetime.datetime.now().isoformat()
        }

        # Save the chat message to Firebase
        save_chat_message(session_id, message_data)

        return response
    except Exception as e:
        return str(e)

# Custom CSS for styling
css = '''
  .circle-logo {
    display: inline-block;
    width: 40px;
    height: 40px;
    border-radius: 50%;
    overflow: hidden;
    margin-right: 10px;
    vertical-align: middle;
  }
  .circle-logo img {
    width: 100%;
    height: 100%;
    object-fit: cover;
  }
  .response-with-logo {
    display: flex;
    align-items: center;
    margin-bottom: 10px;
  }
  footer {
    display: none !important;
    background-color: #F8D7DA;
  }
  label.svelte-1b6s6s {display: none}
'''

gr.ChatInterface(
    chat_interface,
    css=css,
    description="Clara",
    clear_btn=None, undo_btn=None, retry_btn=None
).launch()