import streamlit as st
import requests
from bs4 import BeautifulSoup
from transformers import pipeline
import yfinance as yf

# Set up models
ner_model = "Cassie-0121/fin-bert-finetuned-ner"
sentiment_model = "yiyanghkust/finbert-tone"
text_gen_model = "gpt2"

ner = pipeline("ner", model=ner_model)
sentiment_analyzer = pipeline("sentiment-analysis", model=sentiment_model)
text_generator = pipeline("text-generation", model=text_gen_model)

# Function to fetch stock news using web scraping from Yahoo Finance
def get_stock_news(stock_symbol):
    url = f'https://finance.yahoo.com/quote/{stock_symbol}?p={stock_symbol}'
    headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'}
    
    # Send request and parse the page
    response = requests.get(url, headers=headers)
    soup = BeautifulSoup(response.content, 'html.parser')
    
    # Extracting news headlines (For Yahoo Finance, it's under the 'Recent News' section)
    headlines = []
    news_section = soup.find_all('li', {'class': 'js-stream-content'})
    for news_item in news_section:
        headline = news_item.find('h3')
        if headline:
            headlines.append(headline.get_text())
    
    return headlines

# Function to fetch stock data using Yahoo Finance (yfinance library)
def get_stock_data(stock_symbol):
    stock = yf.Ticker(stock_symbol)
    stock_info = stock.history(period="1d")  # Fetching the latest stock data (1-day)
    return stock_info

# App title
st.title("AI-Powered Financial Analysis App")

# Sidebar for stock symbol input
st.sidebar.header("Stock Data & Analysis")
stock_symbol = st.sidebar.text_input("Enter Stock Symbol (e.g., AAPL for Apple):", "AAPL")

# Fetch stock data and news
if stock_symbol:
    st.sidebar.write(f"Fetching data for {stock_symbol}...")

    # Fetch stock news
    stock_news = get_stock_news(stock_symbol)

    # Fetch stock data (latest price, volume, etc.)
    stock_data = get_stock_data(stock_symbol)
    
    # Display stock data
    st.subheader(f"Latest Stock Data for {stock_symbol}")
    st.write(stock_data)

    # Display stock news
    st.subheader(f"Latest News for {stock_symbol}")
    for news in stock_news:
        st.write(news)

    # Text for analysis
    input_text = ' '.join(stock_news)  # Combine news for analysis

    # Display extracted entities
    st.subheader("Extracted Key Financial Entities")
    if input_text:
        entities = ner(input_text)
        for entity in entities:
            st.write(f"Entity: {entity['word']}, Label: {entity.get('entity', 'N/A')}, Score: {entity.get('score', 0.0):.2f}")
    
    # Sentiment Analysis
    st.subheader("Sentiment Analysis")
    if input_text:
        sentiment = sentiment_analyzer(input_text)
        for result in sentiment:
            st.write(f"Sentiment: {result['label']}, Score: {result['score']:.2f}")
    
    # Investment Advice or Strategy
    st.subheader("Investment Advice or Strategy")
    if input_text:
        prompt = f"Provide a focused and complete investment strategy for {stock_symbol} based on the latest news and trends."
        advice = text_generator(prompt + input_text, max_length=100, num_return_sequences=1, do_sample=True, temperature=0.7)
        st.write(advice[0]['generated_text'])
else:
    st.write("Please enter a valid stock symbol to get analysis.")

# Optional: Footer for extra branding
st.sidebar.write("Powered by Hugging Face, Streamlit, and Web Scraping")