# Import necessary libraries
import os  # Interacting with the operating system (reading/writing files)
from dotenv import load_dotenv  # Loading environment variables from a .env file
import json  # Parsing and handling JSON data


# LangChain OpenAI imports
from langchain_openai import AzureChatOpenAI  # OpenAI embeddings and models
from langchain.prompts import ChatPromptTemplate  # Template for chat prompts

# Other utilities
import numpy as np  # Numpy for numerical operations
import streamlit as st
from datetime import datetime


#====================================SETUP=====================================#
# Fetch secrets from Hugging Face Spaces
api_key = os.environ['AT_4O_AZURE_OPENAI_KEY']
endpoint = os.environ['AT_4O_AZURE_OPENAI_ENDPOINT']
api_version = os.environ['AT_4O_AZURE_OPENAI_APIVERSION']
model_name = os.environ['AT_4O_CHATGPT_MODEL']

# Initialize the Azure Chat OpenAI model
# llm = AzureChatOpenAI(
#     azure_endpoint=endpoint,
#     api_key=api_key,
#     api_version=api_version,
#     azure_deployment='gpt-4o',
#     temperature=0
# )
# This initializes the Azure Chat OpenAI model with the provided endpoint, API key, deployment name, and a temperature setting of 0 (to control response variability).


class MyChatBot:
    def __init__(self):
        """
        Initialize the MyChatBot class, setting up the LLM client.
        """

        # Initialize the Azure OpenAI client using the provided credentials
        self.client = AzureChatOpenAI(
            azure_endpoint=endpoint,  # Endpoint URL for Azure OpenAI
            api_key=api_key,  # API key for authentication
            api_version=api_version,  # API version being used
            model_name=model_name,  # Specify the model to use (e.g., GPT-4 optimized version)
            temperature=0  # Controls randomness in responses; 0 ensures deterministic results
        )
        
        # Define the system prompt to set the behavior of the chatbot
        system_prompt = """You are a Travel agent AI assistant specializing in Travel industry helping users to answer their questions. Your goal is to provide accurate, answer and recommendations to user question.
                          Guidelines for Interaction:
                          Maintain a polite, professional, and reassuring tone.
                          If any detail is unclear or missing, proactively ask for clarification.
        """

        # Build the prompt template for the agent
        self.prompt = ChatPromptTemplate.from_messages([
            ("system", system_prompt),  # System instructions
            ("human", "{input}"),  # Placeholder for human input
            ("placeholder", "{agent_scratchpad}")  # Placeholder for intermediate reasoning steps
        ])

    def get_llm_response(self, user_input):
      """Sends a prompt to the LLM and returns the response.

      Args:
        user_input: The user's prompt.
        llm: The LLM object.

      Returns:
        The LLM's response.
      """
      final_prompt = self.prompt.format(input=user_input, agent_scratchpad=[])
      response = self.client(final_prompt)  # Assuming your LLM has a __call__ method
      return response



def simple_chat_bot_streamlit():
  """
  A Streamlit-based UI for Simple chatbot.
  """
  st.title("Travel Chatbot")
  st.write("Ask me anything and type 'exit' to end the conversation.")
  chatbot = MyChatBot()
  # while True:
  st.write("How can I help you?")
  user_query = st.chat_input("Type your question here (or 'exit' to end)!")
  if user_query:
    if user_query.lower() == "exit":
      goodbye_msg = "Goodbye! Feel free to return if you have more questions."
      st.rerun()
      return
    try:
      response = chatbot.get_llm_response(user_query)
      st.write(response.content)
    except Exception as e:
      print(f"Error: {e}")  

if __name__ == '__main__':
  simple_chat_bot_streamlit()