app.py

# import os
# import time
# from fastapi import FastAPI,Request
# from fastapi.responses import HTMLResponse
# from fastapi.staticfiles import StaticFiles
# from llama_index.core import StorageContext, load_index_from_storage, VectorStoreIndex, SimpleDirectoryReader, ChatPromptTemplate, Settings
# from llama_index.embeddings.huggingface import HuggingFaceEmbedding
# from pydantic import BaseModel
# from fastapi.responses import JSONResponse
# import uuid  # for generating unique IDs
# import datetime
# from fastapi.middleware.cors import CORSMiddleware
# from fastapi.templating import Jinja2Templates
# from huggingface_hub import InferenceClient
# import json
# import re
# from gradio_client import Client
# from simple_salesforce import Salesforce, SalesforceLogin
# from llama_index.llms.huggingface import HuggingFaceLLM
# # from llama_index.llms.huggingface import HuggingFaceInferenceAPI


# # Define Pydantic model for incoming request body
# class MessageRequest(BaseModel):
#     message: str
# repo_id = "meta-llama/Meta-Llama-3-8B-Instruct"
# llm_client = InferenceClient(
#     model=repo_id,
#     token=os.getenv("HF_TOKEN"),
# )


# os.environ["HF_TOKEN"] = os.getenv("HF_TOKEN")
# username = os.getenv("username")
# password = os.getenv("password")
# security_token = os.getenv("security_token")
# domain =  os.getenv("domain")# Using sandbox environment
# session_id, sf_instance = SalesforceLogin(username=username, password=password, security_token=security_token, domain=domain)

#     # Create Salesforce object
# sf = Salesforce(instance=sf_instance, session_id=session_id)

# app = FastAPI()


# @app.middleware("http")
# async def add_security_headers(request: Request, call_next):
#     response = await call_next(request)
#     response.headers["Content-Security-Policy"] = "frame-ancestors *; frame-src *; object-src *;"
#     response.headers["X-Frame-Options"] = "ALLOWALL"
#     return response


# # Allow CORS requests from any domain
# app.add_middleware(
#     CORSMiddleware,
#     allow_origins=["*"],
#     allow_credentials=True,
#     allow_methods=["*"],
#     allow_headers=["*"],
# )




# @app.get("/favicon.ico")
# async def favicon():
#     return HTMLResponse("")  # or serve a real favicon if you have one


# app.mount("/static", StaticFiles(directory="static"), name="static")

# templates = Jinja2Templates(directory="static")
# # Configure Llama index settings
# Settings.llm = HuggingFaceLLM(
#     model_name="meta-llama/Meta-Llama-3-8B-Instruct",
#     tokenizer_name="meta-llama/Meta-Llama-3-8B-Instruct",
#     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"
# )

# PERSIST_DIR = "db"
# PDF_DIRECTORY = 'data'

# # Ensure directories exist
# os.makedirs(PDF_DIRECTORY, exist_ok=True)
# os.makedirs(PERSIST_DIR, exist_ok=True)
# chat_history = []
# current_chat_history = []
# def data_ingestion_from_directory():
#     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 initialize():
#     start_time = time.time()
#     data_ingestion_from_directory()  # Process PDF ingestion at startup
#     print(f"Data ingestion time: {time.time() - start_time} seconds")
# def split_name(full_name):
#     # Split the name by spaces
#     words = full_name.strip().split()
    
#     # Logic for determining first name and last name
#     if len(words) == 1:
#         first_name = ''
#         last_name = words[0]
#     elif len(words) == 2:
#         first_name = words[0]
#         last_name = words[1]
#     else:
#         first_name = words[0]
#         last_name = ' '.join(words[1:])
    
#     return first_name, last_name

# initialize()  # Run initialization tasks


# def handle_query(query):
#     chat_text_qa_msgs = [
#         (
#             "user",
#             """
#             You are the Clara Redfernstech chatbot. Your goal is to provide accurate, professional, and helpful answers to user queries based on the company's data. Always ensure your responses are clear and concise. Give response within 10-15 words only       
#             {context_str}
#             Question:
#             {query_str}
#             """
#         )
#     ]
#     text_qa_template = ChatPromptTemplate.from_messages(chat_text_qa_msgs)
    
#     storage_context = StorageContext.from_defaults(persist_dir=PERSIST_DIR)
#     index = load_index_from_storage(storage_context)
#     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."
#     current_chat_history.append((query, response))
#     return response
# @app.get("/ch/{id}", response_class=HTMLResponse)
# async def load_chat(request: Request, id: str):
#     return templates.TemplateResponse("index.html", {"request": request, "user_id": id})
# # Route to save chat history
# @app.post("/hist/")
# async def save_chat_history(history: dict):
#     # Check if 'userId' is present in the incoming dictionary
#     user_id = history.get('userId')
#     print(user_id)

#     # Ensure user_id is defined before proceeding
#     if user_id is None:
#         return {"error": "userId is required"}, 400

#     # Construct the chat history string
#     hist = ''.join([f"'{entry['sender']}: {entry['message']}'\n" for entry in history['history']])
#     hist = "You are a Redfernstech summarize model. Your aim is to use this conversation to identify user interests solely based on that conversation: " + hist
#     print(hist)

#     # Get the summarized result from the client model
#     result = hist

#     try:
#         sf.Lead.update(user_id, {'Description': result})
#     except Exception as e:
#         return {"error": f"Failed to update lead: {str(e)}"}, 500
    
#     return {"summary": result, "message": "Chat history saved"}
# @app.post("/webhook")
# async def receive_form_data(request: Request):
#     form_data = await request.json()
#     # Log in to Salesforce
#     first_name, last_name = split_name(form_data['name'])
#     data = {
#     'FirstName': first_name,
#     'LastName': last_name,
#     'Description': 'hii',  # Static description
#     'Company': form_data['company'],  # Assuming company is available in form_data
#     'Phone': form_data['phone'].strip(),  # Phone from form data
#     'Email': form_data['email'],  # Email from form data
#     }
#     a=sf.Lead.create(data)
#     # Generate a unique ID (for tracking user)
#     unique_id = a['id']
    
#     # Here you can do something with form_data like saving it to a database
#     print("Received form data:", form_data)
    
#     # Send back the unique id to the frontend
#     return JSONResponse({"id": unique_id})

# @app.post("/chat/")
# async def chat(request: MessageRequest):
#     message = request.message  # Access the message from the request body
#     response = handle_query(message)  # Process the message
#     message_data = {
#         "sender": "User",
#         "message": message,
#         "response": response,
#         "timestamp": datetime.datetime.now().isoformat()
#     }
#     chat_history.append(message_data)
#     return {"response": response}
# @app.get("/")
# def read_root():
#     return {"message": "Welcome to the API"}

import os
import datetime
import json
import logging
from fastapi import FastAPI, Request, HTTPException
from fastapi.responses import HTMLResponse, JSONResponse
from fastapi.staticfiles import StaticFiles
from pydantic import BaseModel
from fastapi.middleware.cors import CORSMiddleware
from fastapi.templating import Jinja2Templates
from simple_salesforce import Salesforce, SalesforceLogin
from langchain_groq import ChatGroq
from langchain_core.prompts import ChatPromptTemplate
from llama_index.core import StorageContext, VectorStoreIndex, SimpleDirectoryReader, Settings
from llama_index.core import load_index_from_storage  # Added missing import
from llama_index.embeddings.huggingface import HuggingFaceEmbedding

# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# Define Pydantic model for incoming request body
class MessageRequest(BaseModel):
    message: str

# Initialize FastAPI app
app = FastAPI()

# Allow CORS requests (restrict in production)
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

# Mount static files
app.mount("/static", StaticFiles(directory="static"), name="static")
templates = Jinja2Templates(directory="static")

# Validate environment variables
required_env_vars = ["CHATGROQ_API_KEY", "username", "password", "security_token", "domain", "HF_TOKEN"]
for var in required_env_vars:
    if not os.getenv(var):
        logger.error(f"Environment variable {var} is not set")
        raise ValueError(f"Environment variable {var} is not set")

# Initialize Groq model
GROQ_API_KEY = os.getenv("CHATGROQ_API_KEY")
GROQ_MODEL = "llama3-8b-8192"
try:
    llm = ChatGroq(
        model_name=GROQ_MODEL,
        api_key=GROQ_API_KEY,
        temperature=0.1,
        max_tokens=50
    )
except Exception as e:
    logger.error(f"Failed to initialize Groq model: {e}")
    raise HTTPException(status_code=500, detail="Failed to initialize Groq model")

# Configure LlamaIndex settings
Settings.embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")

# Salesforce credentials
username = os.getenv("username")
password = os.getenv("password")
security_token = os.getenv("security_token")
domain = os.getenv("domain")  # e.g., 'test' for sandbox

# Initialize Salesforce connection (allow failure)
sf = None
try:
    session_id, sf_instance = SalesforceLogin(
        username=username, password=password, security_token=security_token, domain=domain
    )
    sf = Salesforce(instance=sf_instance, session_id=session_id)
    logger.info("Salesforce connection established")
except Exception as e:
    logger.error(f"Failed to connect to Salesforce: {e}. Continuing without Salesforce integration.")

# Chat history
chat_history = []
current_chat_history = []
MAX_HISTORY_SIZE = 100  # Limit chat history size

# Directories for data ingestion
PDF_DIRECTORY = "data"
PERSIST_DIR = "db"

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

def data_ingestion_from_directory():
    try:
        documents = SimpleDirectoryReader(PDF_DIRECTORY).load_data()
        if not documents:
            logger.warning("No documents found in PDF_DIRECTORY")
            return
        storage_context = StorageContext.from_defaults()
        index = VectorStoreIndex.from_documents(documents, storage_context=storage_context)
        index.storage_context.persist(persist_dir=PERSIST_DIR)
        logger.info("Data ingestion completed successfully")
    except Exception as e:
        logger.error(f"Error during data ingestion: {e}")
        raise HTTPException(status_code=500, detail=f"Data ingestion failed: {str(e)}")

def initialize():
    try:
        data_ingestion_from_directory()  # Process PDF ingestion at startup
    except Exception as e:
        logger.error(f"Initialization failed: {e}")
        raise HTTPException(status_code=500, detail="Initialization failed")

initialize()  # Run initialization tasks

def handle_query(query):
    # Prepare context from chat history
    chat_context = ""
    for past_query, response in reversed(current_chat_history[-10:]):  # Limit to last 10 exchanges
        if past_query.strip():
            chat_context += f"User asked: '{past_query}'\nBot answered: '{response}'\n"

    # Load vector index and retrieve relevant documents
    try:
        storage_context = StorageContext.from_defaults(persist_dir=PERSIST_DIR)
        index = load_index_from_storage(storage_context)
        query_engine = index.as_query_engine(similarity_top_k=2)
        retrieved = query_engine.query(query)
        doc_context = retrieved.response if hasattr(retrieved, 'response') else "No relevant documents found."
        logger.info(f"Retrieved documents for query '{query}': {doc_context[:100]}...")
    except Exception as e:
        logger.error(f"Error retrieving documents: {e}")
        doc_context = "Failed to retrieve documents."

    # Construct the prompt with Redferns Tech focus
    prompt_template = ChatPromptTemplate.from_messages([
        ("system", """
        You are Clara Redfernstech, a chatbot for Redferns Tech, a leader in data science, machine learning, and AI solutions.
        Provide accurate, professional answers in 10-15 words based on the provided document context and chat history.
        Focus on Redferns Tech's expertise in data science and AI.

        Document Context:
        {doc_context}

        Chat History:
        {chat_context}

        Question:
        {query}
        """),
    ])
    prompt = prompt_template.format(doc_context=doc_context, chat_context=chat_context, query=query)

    # Query Groq model
    try:
        response = llm.invoke(prompt)
        response_text = response.content.strip()
    except Exception as e:
        logger.error(f"Error querying Groq API: {e}")
        response_text = "Sorry, I couldn't find an answer."

    # Update chat history
    if len(current_chat_history) >= MAX_HISTORY_SIZE:
        current_chat_history.pop(0)
    current_chat_history.append((query, response_text))
    return response_text

@app.get("/ch/{id}", response_class=HTMLResponse)
async def load_chat(request: Request, id: str):
    return templates.TemplateResponse("index.html", {"request": request, "user_id": id})

@app.post("/hist/")
async def save_chat_history(history: dict):
    if not sf:
        logger.error("Salesforce integration is disabled")
        return {"error": "Salesforce integration is unavailable"}, 503

    user_id = history.get('userId')
    if not user_id:
        logger.error("userId is missing in history request")
        return {"error": "userId is required"}, 400

    hist = ''.join([f"'{entry['sender']}: {entry['message']}'\n" for entry in history['history']])
    hist = "You are a Redfernstech summarize model. Identify user interests from this conversation: " + hist

    try:
        sf.Lead.update(user_id, {'Description': hist})
        logger.info(f"Chat history updated for user {user_id}")
    except Exception as e:
        logger.error(f"Failed to update lead: {e}")
        return {"error": f"Failed to update lead: {str(e)}"}, 500
    
    return {"summary": hist, "message": "Chat history saved"}

@app.post("/webhook")
async def receive_form_data(request: Request):
    if not sf:
        logger.error("Salesforce integration is disabled")
        return {"error": "Salesforce integration is unavailable"}, 503

    try:
        form_data = await request.json()
    except json.JSONDecodeError:
        logger.error("Invalid JSON in webhook request")
        return {"error": "Invalid JSON"}, 400

    first_name, last_name = split_name(form_data.get('name', ''))
    data = {
        'FirstName': first_name,
        'LastName': last_name,
        'Description': 'hii',
        'Company': form_data.get('company', ''),
        'Phone': form_data.get('phone', '').strip(),
        'Email': form_data.get('email', ''),
    }

    try:
        result = sf.Lead.create(data)
        unique_id = result['id']
        logger.info(f"Lead created with ID {unique_id}")
        return JSONResponse({"id": unique_id})
    except Exception as e:
        logger.error(f"Failed to create lead: {e}")
        return {"error": f"Failed to create lead: {str(e)}"}, 500

@app.post("/chat/")
async def chat(request: MessageRequest):
    message = request.message
    response = handle_query(message)
    message_data = {
        "sender": "User",
        "message": message,
        "response": response,
        "timestamp": datetime.datetime.now().isoformat()
    }
    if len(chat_history) >= MAX_HISTORY_SIZE:
        chat_history.pop(0)
    chat_history.append(message_data)
    logger.info(f"Chat message processed: {message}")
    return {"response": response}

@app.get("/health")
async def health_check():
    try:
        storage_context = StorageContext.from_defaults(persist_dir=PERSIST_DIR)
        index = load_index_from_storage(storage_context)
        logger.info("Vector index loaded successfully")
        return {"status": "healthy", "pdf_ingestion": "successful"}
    except Exception as e:
        logger.error(f"Health check failed: {e}")
        return {"status": "unhealthy", "error": str(e)}

@app.get("/")
def read_root():
    return {"message": "Welcome to the Redferns Tech Chatbot API"}

def split_name(full_name):
    words = full_name.strip().split()
    if len(words) == 1:
        return '', words[0]
    elif len(words) == 2:
        return words[0], words[1]
    else:
        return words[0], ' '.join(words[1:])