# routes.py

import logging

from fastapi import APIRouter, Body, HTTPException
from models import (
    PageSpeedURLRequest,
    ReportIngestRequest,
    NewChatRequest,
    ChatQueryRequest,
)
from config import GEMINI_API_KEY, PAGESPEED_API_KEY
from pagespeed_fetch import get_pagespeed_data
from vectorstore_manager import ingest_report
from gemini_report import generate_report_with_gemini

from chatbot import create_new_chat, get_chain_for_user, summarize_messages
from langchain_mongodb.chat_message_histories import MongoDBChatMessageHistory

# Configure logger to integrate with Uvicorn’s logging
logger = logging.getLogger("uvicorn.error")

router = APIRouter()


# ----- EXISTING ENDPOINTS -----

@router.post("/ingest-report")
async def ingest_user_report(request: ReportIngestRequest):
    """
    Ingests a user’s PageSpeed report text into the FAISS vectorstore.
    """
    logger.info(f"[ingest-report] Received ingest request for user_id={request.user_id}")
    try:
        vectorstore, retriever = ingest_report(request.user_id, request.report_text)
        logger.info(f"[ingest-report] Successfully ingested report for user_id={request.user_id}")
        return {"status": "success", "message": f"Report ingested for user {request.user_id}"}
    except Exception as e:
        logger.error(f"[ingest-report] Error ingesting report for user_id={request.user_id}: {e}")
        raise HTTPException(status_code=500, detail=str(e))


@router.post("/generate-report")
async def generate_report(data: dict = Body(...)):
    """
    Generates a report using Gemini given arbitrary data in the request body.
    """
    logger.info(f"[generate-report] Received data for Gemini report generation: keys={list(data.keys())}")
    try:
        report = generate_report_with_gemini(data, GEMINI_API_KEY)
        logger.info(f"[generate-report] Successfully generated report")
        return {"report": report}
    except Exception as e:
        logger.error(f"[generate-report] Error generating report: {e}")
        raise HTTPException(status_code=500, detail=str(e))


@router.post("/fetch-pagespeed-data")
async def fetch_pagespeed_data(request: PageSpeedURLRequest):
    """
    Fetches PageSpeed Insights data for the provided target URL.
    """
    logger.info(f"[fetch-pagespeed-data] Fetching PageSpeed for URL={request.target_url}")
    try:
        data = get_pagespeed_data(request.target_url, PAGESPEED_API_KEY)
        logger.info(f"[fetch-pagespeed-data] Successfully fetched PageSpeed data for URL={request.target_url}")
        return {"status": "success", "data": data}
    except Exception as e:
        logger.error(f"[fetch-pagespeed-data] Error fetching PageSpeed data for URL={request.target_url}: {e}")
        raise HTTPException(status_code=500, detail=str(e))


@router.get("/")
async def home():
    """
    Simple health check endpoint.
    """
    logger.info("[home] Health check endpoint called")
    return {"API is UP and running"}


# ----- NEW ENDPOINTS FOR CHATBOT -----

@router.post("/new-chat")
async def new_chat(request: NewChatRequest):
    """
    Creates a new chat session for the given user_id and returns a chat_id.
    Expected JSON body: {"user_id": "<string>"}
    """
    logger.info(f"[new-chat] Creating new chat session for user_id={request.user_id}")
    try:
        chat_id = create_new_chat(request.user_id)
        logger.info(f"[new-chat] Created chat_id={chat_id} for user_id={request.user_id}")
        return {"status": "success", "chat_id": chat_id}
    except Exception as e:
        logger.error(f"[new-chat] Error creating chat session for user_id={request.user_id}: {e}")
        raise HTTPException(status_code=500, detail=str(e))


@router.post("/chat")
async def chat_query(request: ChatQueryRequest):
    """
    Sends a question to an existing chat session and returns the AI's response.
    Expected JSON body:
      {
        "user_id": "<string>",
        "chat_id": "<string>",
        "query": "<string>"
      }
    """
    logger.info(f"[chat] Received query for user_id={request.user_id}, chat_id={request.chat_id}")
    try:
        # Reconstruct the ConversationalRetrievalChain for this user & chat_id
        chain = get_chain_for_user(request.user_id, request.chat_id)
        chat_history: MongoDBChatMessageHistory = chain.memory  # type: ignore

        # Optionally summarize if the history is too long
        if summarize_messages(chat_history):
            logger.info(f"[chat] Summarized chat history for chat_id={request.chat_id}")

        # Add the user’s message to history
        chat_history.add_user_message(request.query)
        logger.info(f"[chat] Added user message to history: \"{request.query}\"")

        # Run the chain synchronously (non-streaming)
        result = chain({
            "question": request.query,
            "chat_history": chat_history.messages,
        })

        answer = result.get("answer", "").strip()
        logger.info(f"[chat] LLM returned answer=\"{answer}\" for user_id={request.user_id}")

        # Persist the AI’s message
        if answer:
            chat_history.add_ai_message(answer)
            logger.info(f"[chat] Persisted AI message to history for chat_id={request.chat_id}")

        return {
            "status": "success",
            "answer": answer,
            "source_documents": result.get("source_documents", []),
        }

    except ValueError as ve:
        logger.warning(f"[chat] No vectorstore for user_id={request.user_id}: {ve}")
        raise HTTPException(status_code=404, detail=str(ve))
    except Exception as e:
        logger.error(f"[chat] Error processing chat request for user_id={request.user_id}, chat_id={request.chat_id}: {e}")
        raise HTTPException(status_code=500, detail=str(e))