app.py

import gradio as gr
from huggingface_hub import InferenceClient
import yfinance as yf
import plotly.express as px
import pandas as pd
from datetime import datetime
import feedparser

# ---------- LLM Setup ----------
client = InferenceClient("microsoft/phi-4")

def random_respond(message, history):
    messages = [{
        "role": "system",
        "content": "You are an expert financial advisor and teacher from LSE and Oxford. You help young people understand finance by breaking down complex terms clearly."
    }]
    
    if history:
        for user_msg, bot_msg in history:
            messages.append({"role": "user", "content": user_msg})
            messages.append({"role": "assistant", "content": bot_msg})
    
    messages.append({"role": "user", "content": message})
    response = client.chat_completion(messages, max_tokens=1000)
    return response['choices'][0]['message']['content'].strip()

# ---------- Investment Simulator Setup ----------

class Investment_Simulator:

    def __init__(self, portfolio, history, portfolio_history):
        self.portfolio = {"cash": 500.0, "stocks": {}}
        self.history = []
        self.portfolio_history = []
    
    def get_price(self, ticker):
        try:
            data = yf.Ticker(ticker)
            todays_data = data.history(period='1d', interval='1m')
            latest_price = todays_data['Close'].dropna().iloc[-1]
            return float(latest_price)
        except:
            return None
    
    def compute_portfolio_value(self):
        total_stock_value = 0.0
        for ticker, qty in self.portfolio["stocks"].items():
            price = get_price(ticker)
            if price:
                total_stock_value += qty * price
        return self.portfolio["cash"] + total_stock_value
    
    def update_portfolio_history(self):
        now = datetime.now()
        total_value = self.compute_portfolio_value()
        self.portfolio_history.append((now, total_value))
    
    def plot_portfolio_value(self):
        if not self.portfolio_history:
            return None
        df = pd.DataFrame(self.portfolio_history, columns=["Time", "Value"])
        fig = px.line(df, x="Time", y="Value", title="Portfolio Value Over Time", labels={"Time": "Time", "Value": "Value (£)"})
        fig.update_layout(xaxis_rangeslider_visible=True)
        return fig

    def process_input(self, user_input):
        user_input = user_input.lower()
    
        try:
            if "buy" in user_input:
                amount = int(user_input.split("£")[1].split()[0])
                ticker = user_input.split("of")[1].strip().upper()
                price = self.get_price(ticker)
                if not price:
                    return f"Couldn't fetch price for {ticker}."
                qty = round(amount / price, 4)
                if amount > self.portfolio["cash"]:
                    return f"Insufficient funds (£{self.portfolio['cash']:.2f})."
                self.portfolio["cash"] -= amount
                self.portfolio["stocks"][ticker] = self.portfolio["stocks"].get(ticker, 0) + qty
                self.history.append(f"Bought £{amount} of {ticker} ({qty} shares at £{price:.2f})")
                self.update_portfolio_history()
                return f"Bought {qty} shares of {ticker} at £{price:.2f}"
    
            elif "sell" in user_input:
                amount = int(user_input.split("£")[1].split()[0])
                ticker = user_input.split("of")[1].strip().upper()
                price = self.get_price(ticker)
                if not price:
                    return f"Couldn't fetch price for {ticker}."
                qty_to_sell = round(amount / price, 4)
                current_qty = self.portfolio["stocks"].get(ticker, 0)
                if qty_to_sell > current_qty:
                    return f"You don't own enough of {ticker}."
                self.portfolio["stocks"][ticker] -= qty_to_sell
                self.portfolio["cash"] += amount
                self.history.append(f"Sold £{amount} of {ticker} ({qty_to_sell} shares at £{price:.2f})")
                self.update_portfolio_history()
                return f"Sold {qty_to_sell} shares of {ticker} at £{price:.2f}"
    
            elif "portfolio" in user_input:
                cash = self.portfolio['cash']
                total_stock_value = 0.0
                summary = [f"Cash: £{cash:.2f}"]
                for ticker, qty in self.portfolio["stocks"].items():
                    live_price = self.get_price(ticker)
                    if live_price:
                        value = qty * live_price
                        total_stock_value += value
                        summary.append(f"{ticker}: {qty:.4f} shares (£{value:.2f})")
                    else:
                        summary.append(f"{ticker}: {qty:.4f} shares (price unavailable)")
                total_value = cash + total_stock_value
                summary.append(f"\nTotal Portfolio Value: £{total_value:.2f}")
                return "\n".join(summary)
    
            elif "history" in user_input:
                return "\n".join(self.history[-5:]) or "No trades yet."
    
            elif "reset" in user_input:
                self.portfolio["cash"] = 500.0
                self.portfolio["stocks"].clear()
                self.history.clear()
                self.portfolio_history.clear()
                self.update_portfolio_history()
                return "Portfolio reset."
    
            else:
                return "Try commands like:\n• Buy £100 of AAPL\n• Sell £50 of TSLA\n• Show portfolio"
    
        except Exception as e:
            return f"Error: {e}"

    def run_all(self, text):
        response = self.process_input(text)
        fig = self.plot_portfolio_value()
        return response, fig
    
    self.update_portfolio_history()


class News:
    def __init__(self):
        pass
    
    def get_news(self, tickers):
        ticker_list = [t.strip().upper() for t in tickers.split(",")]
    
        market_url = "https://finance.yahoo.com/rss/topstories"
        market_feed = feedparser.parse(market_url)
    
        result = "## General Market News\n"
        if market_feed.entries:
            for entry in market_feed.entries[:5]:
                result += f"**{entry.title}**\n{entry.link}\n{entry.published}\n\n"
        else:
            result += "_No general market news found_\n\n"
    
        for ticker in ticker_list:
            rss_url = f"https://feeds.finance.yahoo.com/rss/2.0/headline?s={ticker}&region=US&lang=en-US"
            feed = feedparser.parse(rss_url)
    
            result += f"## News for {ticker}\n"
            if feed.entries:
                for entry in feed.entries[:5]:
                    result += f"**{entry.title}**\n{entry.link}\n{entry.published}\n\n"
            else:
                result += "_No news found_\n\n"
            
        return result
    

# ---------- UI Layout ----------
custom_css = """
#ZenoLogo {
    display: block;
    margin-left: auto;
    margin-right: auto;
    width: 200px;
}
#landing-content {
    text-align: center;
    margin-top: 100px;
}
#landing_page {
    background-color: #AEC3B0;
    padding: 50px;
    min-height: 100vh;
    overflow: hidden;
}
#sidebar-toggle {
    font-size: 5em;
    background: none;
    border: none;
    margin: 10px;
    cursor: pointer;
}
#sidebar {
    color: Black;
    background-color: #124559;
    border-right: 1px solid #ddd;
    padding: 10px;
}
"""

news = News()

with gr.Blocks(css=custom_css) as demo:
    active_section = gr.State("chat")
    sidebar_open = gr.State(True)

    landing_page = gr.Column(visible=True, elem_id="landing_page")
    with landing_page:
        gr.HTML("""
            <div id="landing-content">
                <img id="ZenoLogo" src="https://i.imgur.com/iCdIzOR.png" alt="ZENO" />
                <h2>REDUCING YOUR MONEY PROBLEMS TO ZERO</h2>
            </div>
        """)
        landing_input = gr.Textbox(
            placeholder="Type your first finance question...",
            show_label=False,
            lines=1
        )

    app = gr.Row(visible=False)
    with app:
        with gr.Column(scale=1, min_width=200):
            sidebar_toggle = gr.Button("≡", elem_id="sidebar-toggle")
            with gr.Column(visible=True, elem_id="sidebar") as sidebar:
                gr.Markdown("📊 Zeno Tools")
                btn_chat = gr.Button("General Finance")
                btn_news = gr.Button("News/Updates")
                btn_mock = gr.Button("Mock Investment")
                btn_about = gr.Button("About Us")

        with gr.Column(scale=5):
            Zeno_Chat = gr.Column(visible=True)
            with Zeno_Chat:
                chatbot = gr.Chatbot()
                textbox = gr.Textbox(
                    show_label=False,
                    placeholder="Ask me anything about finance!",
                    lines=1
                )

            Zeno_Investments = gr.Column(visible=False, elem_id="Investments")
            with Zeno_Investments:
                gr.Markdown("## Live Investment Simulator (Practice £500) + Portfolio Tracker")
                
                simulator = Investment_Simulator()
                
                invest_input = gr.Textbox(label="Enter a command", placeholder="Try: Buy £100 of AAPL")
                invest_output = gr.Textbox(label="Bot Response")
                invest_chart = gr.Plot(label="Portfolio Value Over Time")
                
                invest_input.submit(
                    simulator.run_all,
                    inputs=invest_input,
                    outputs=[invest_output, invest_chart]
                )
            
            Zeno_News = gr.Column(visible=False, elem_id="Stock and Market News")
            with Zeno_News:
                gr.Markdown("## 📰 Stock and Market News")
                
                ticker_input = gr.Textbox(
                label="Enter stock tickers (comma-separated, e.g., AAPL, TSLA, MSFT)"
                )
                news_output = gr.Markdown()

                fetch_button = gr.Button("Get News")
                fetch_button.click(
                fn=news.get_news,
                inputs=ticker_input,
                outputs=news_output
                )
                
                
    def handle_first_input(msg):
        return gr.update(visible=False), gr.update(visible=True), msg, []

    landing_input.submit(
        handle_first_input,
        inputs=landing_input,
        outputs=[landing_page, app, textbox, chatbot]
    )

    def respond_to_user(message, chat_history):
        bot_response = random_respond(message, chat_history)
        chat_history.append((message, bot_response))
        return "", chat_history

    textbox.submit(
        respond_to_user,
        inputs=[textbox, chatbot],
        outputs=[textbox, chatbot]
    )

    def switch_view(section):
        return (
            gr.update(visible=section == "chat"),
            gr.update(visible=section == "investments"),
            section
        )

    btn_chat.click(switch_view, inputs=[], outputs=[Zeno_Chat, Zeno_Investments, active_section], _js="() => 'chat'")
    btn_mock.click(switch_view, inputs=[], outputs=[Zeno_Chat, Zeno_Investments, active_section], _js="() => 'investments'")

    def toggle_sidebar(is_open):
        new_state = not is_open
        return gr.update(visible=new_state), new_state

    sidebar_toggle.click(
        toggle_sidebar,
        inputs=[sidebar_open],
        outputs=[sidebar, sidebar_open]
    )

demo.launch()