import os
import streamlit as st
import pickle
import time
from langchain import OpenAI
from langchain.chains import RetrievalQAWithSourcesChain
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.document_loaders import UnstructuredURLLoader
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS

from dotenv import load_dotenv

load_dotenv()

st.title("Finance News Research Tool")
st.sidebar.title("News Articles")

# Function to load and process URLs and split the data 
def load_and_process_urls(urls):
    loader = UnstructuredURLLoader(urls=urls)
    st.text("Loading Data ...")
    data = loader.load()


    text_splitter = RecursiveCharacterTextSplitter(
        separators=['\n\n', '\n', '.', ','],
        chunk_size=1000
    )

    docs = text_splitter.split_documents(data)

    # convert to vectors using embeddings
    embeddings = OpenAIEmbeddings()
    vectorstore_openai = FAISS.from_documents(docs, embeddings)

    st.text("Embedding Vectors...")

    # Save FAISS index to pickle file
    with open(file_path, "wb") as f:
        pickle.dump(vectorstore_openai, f)

# Function to handle query processing
def process_query(query, file_path):
    if os.path.exists(file_path):
        with open(file_path, "rb") as f:
            vectorstore = pickle.load(f)

            chain = RetrievalQAWithSourcesChain.from_llm(llm=llm, retriever=vectorstore.as_retriever())
            result = chain({"question": query}, return_only_outputs=True)
            
            st.header("Answer")
            st.write(result["answer"])

            sources = result.get("sources", "")
            if sources:
                st.subheader("Sources:")
                for source in sources.split("\n"):  
                    st.write(source)

# User inputs for URLs
urls = [st.sidebar.text_input(f"URL {i+1}") for i in range(5)]

# Button to process URLs
process_url_clicked = st.sidebar.button("Process URLs")

# Placeholder for displaying messages
main_placeholder = st.empty()

# File path for storing FAISS index
file_path = "faiss_store_openai.pkl"

# Load and process data when button is clicked
if process_url_clicked:
    load_and_process_urls(urls)

# User input for query
query = main_placeholder.text_input("Question: ")

# Process the query if provided
if query:
    process_query(query, file_path)