# ===================================================
# "the-very-latest-latest-POST-it"-----app.py
# ===================================================

import asyncio
import os
import re
import time
import json

import chainlit as cl
from dotenv import load_dotenv

from pydantic import BaseModel, PrivateAttr

from langchain import hub
from langchain_openai import OpenAI
from tiktoken import encoding_for_model
from langchain.chains import LLMChain #APIChain
from langchain_core.prompts import PromptTemplate

from langchain_community.tools.requests.tool import RequestsPostTool
from langchain_community.utilities.requests import TextRequestsWrapper

from langchain.memory.buffer import ConversationBufferMemory
from langchain.memory import ConversationTokenBufferMemory
from langchain.memory import ConversationSummaryMemory

from api_docs import api_docs_str

load_dotenv()
OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY")
auth_token = os.getenv("DAYSOFF_API_TOKEN")



class EnhancedRequestsPostTool(RequestsPostTool, BaseModel):
    api_docs: str = api_docs_str

    # --PrivateAttr@dynanmc init attrbts
    _url_chain: LLMChain = PrivateAttr()
    _response_chain: LLMChain = PrivateAttr()

    def __init__(self, requests_wrapper, llm, api_docs_str, api_url_prompt, api_response_prompt):
        super().__init__(requests_wrapper=requests_wrapper, allow_dangerous_requests=True)
        object.__setattr__(self, 'api_docs', api_docs_str) # self.api_docs = api_docs_str
        object.__setattr__(self, 'url_chain', LLMChain(llm=llm, prompt=api_url_prompt))  # --dynanmc init1
        object.__setattr__(self, 'response_chain', LLMChain(llm=llm, prompt=api_response_prompt)) # --dynanmc init2

    async def ainvoke(self, input_data, callbacks=None):
        # -- create API URL
        url_response = await self.url_chain.ainvoke({
            "api_docs": self.api_docs,
            "question": input_data.get("question")
        })

        api_response = await super().ainvoke(input_data, callbacks) # --make POST request

        # --form(at) response/:response_chain
        formatted_response = await self.response_chain.ainvoke({
            "api_docs": self.api_docs,
            #"question": input_data.get("question"),
            #"api_url": url_response.get("text"),
            "api_response": api_response
        })

        return formatted_response

daysoff_assistant_template = """
#You are a customer support assistant (’kundeservice AI assistent’) for Daysoff.
#By default, you respond in Norwegian language, using a warm, direct, and professional tone.
Your expertise is exclusively in retrieving booking information for a given booking ID assistance related to
to this.
You do not provide information outside of this scope. If a question is not about this topic, respond with
"Jeg driver faktisk kun med henvendelser omkring bestillingsinformasjon. Gjelder det andre henvendelser
må du nok kontakte kundeservice på kundeservice@daysoff.no😊"
Chat History: {chat_history}
Question: {question}
Answer:
"""
daysoff_assistant_prompt = PromptTemplate(
    input_variables=['chat_history', 'question'],
    template=daysoff_assistant_template
)

api_url_template = """
Given the following API Documentation for Daysoff's official
booking information API: {api_docs}
Your task is to construct the most efficient API URL to answer
the user's question, ensuring the
call is optimized to include only the necessary information.
Question: {question}
API URL:
"""
api_url_prompt = PromptTemplate(input_variables=['api_docs', 'question'],
                                template=api_url_template)

api_response_template = """
With the API Documentation for Daysoff's official API: {api_docs} in mind,
and the specific user question: {question},
and given this API URL: {api_url} for querying,
and response from Daysoff's API: {api_response},
never refer the user to the API URL as your answer!
You should always provide a clear and concise summary (in Norwegian) of the booking information retrieved.
This way you directly address the user's question in a manner that reflects the professionalism and warmth
of a human customer service agent.
Summary:
"""
api_response_prompt = PromptTemplate(
    input_variables=['api_docs', 'question', 'api_url', 'api_response'],
    template=api_response_template
)


@cl.on_chat_start
def setup_multiple_chains():
    llm = OpenAI(
        model='gpt-3.5-turbo-instruct',
        temperature=0.7,
        openai_api_key=OPENAI_API_KEY,
        max_tokens=2048,
        top_p=0.9,
        frequency_penalty=0.1,
        presence_penalty=0.1
    )

    conversation_memory = ConversationBufferMemory(memory_key="chat_history",
                                                   max_len=30,
                                                   return_messages=True,
    )

    llm_chain = LLMChain(
        llm=llm,
        prompt=daysoff_assistant_prompt,
        memory=conversation_memory,
    )

    requests_wrapper = TextRequestsWrapper(
        headers={
            "Authorization": f"Bearer <auth_token>",
            "Content-Type": "application/json"
        }
    )

    post_tool = EnhancedRequestsPostTool(
        requests_wrapper=requests_wrapper,
        llm=llm,
        api_docs_str=api_docs_str,
        api_url_prompt=api_url_prompt,
        api_response_prompt=api_response_prompt
    )

    cl.user_session.set("llm_chain", llm_chain)
    cl.user_session.set("post_tool", post_tool)


@cl.on_message
async def handle_message(message: cl.Message):
    user_message = message.content
    llm_chain = cl.user_session.get("llm_chain")
    post_tool = cl.user_session.get("post_tool")

    booking_pattern = r'\b[A-Z]{6}\d{6}\b'
    endpoint_url = "https://aivisions.no/data/daysoff/api/v1/booking/"

    match = re.search(booking_pattern, user_message)
    if match:
        
        bestillingskode = match.group()
        post_data = {
            "url": endpoint_url,
            "body": { 
                "booking_id": bestillingskode
            }
        }

        response = await post_tool.ainvoke(
            json.dumps(post_data),
            [cl.AsyncLangchainCallbackHandler()]  # config={"callbacks": 
        )

        # --degug!
        print(f"Raw response: {response}")

        if response:
            try:
                booking_data = json.loads(response.get("output", "{}"))
                table = "| Field | Information |\n|---|---|\n"  

                table = f"""
                | Field       | Value              |
                |-------------|--------------------|
                | Booking ID  | {booking_data.get('booking_id', 'N/A')} |
                | Name        | {booking_data.get('full_name', 'N/A')} |
                | Amount      | {booking_data.get('amount', 'N/A')} kr |
                | Check-in    | {booking_data.get('checkin', 'N/A')} |
                | Check-out   | {booking_data.get('checkout', 'N/A')} |
                | Address     | {booking_data.get('address', 'N/A')} |
                | User ID     | {booking_data.get('user_id', 'N/A')} |
                | Info        | {booking_data.get('infotext', 'N/A')} |
                | Included    | {booking_data.get('included', 'N/A')} |
                """
                await cl.Message(content=table).send()
                
            except Exception as e:
                error_msg = f"Error: Could not parse the booking information. Details: {str(e)}"
            await cl.Message(error_msg).send()

    else:
        response = await llm_chain.ainvoke(
            user_message,                                # {"chat_history": "", "question": user_message}
        )
        await cl.Message(content=response).send()

    response_key = "output" if "output" in response else "text"

    await cl.Message(response.get(response_key, "")).send()
    return message.content