backend.py

import pandas as pd
import requests
import random
from datetime import datetime, timedelta

# -----------------------------
# Fetch BTC Data
# -----------------------------
def fetch_btc_data():
    try:
        url = "https://api.coingecko.com/api/v3/coins/bitcoin/market_chart?vs_currency=usd&days=50"
        res = requests.get(url, timeout=10)
        res.raise_for_status()
        data = res.json()
        prices = data['prices']
        df = pd.DataFrame(prices, columns=['time', 'close'])
        df['time'] = pd.to_datetime(df['time'], unit='ms')
        return df
    except Exception as e:
        print(f"Error fetching BTC data: {e}")
        return pd.DataFrame()

# -----------------------------
# Simulated Prediction
# -----------------------------
def simulate_lstm_prediction(df, hours_ahead=0):
    if df.empty or len(df) < 5:
        return None, None, None
    
    current_price = df.iloc[-1]['close']
    momentum = (df.iloc[-1]['close'] - df.iloc[-5]['close']) / df.iloc[-5]['close']
    
    time_factor = hours_ahead * 0.015
    variance = (random.random() - 0.5) * (0.04 + time_factor)

    pred_price = current_price * (1 + (momentum * 0.35) + variance)
    trend = "HIGH ↑" if pred_price > current_price else "LOW ↓"
    confidence = random.uniform(65, 92)

    return pred_price, trend, confidence

# -----------------------------
# Create Plot
# -----------------------------
def create_plot(df):
    import plotly.graph_objects as go
    
    if df.empty:
        fig = go.Figure()
        fig.update_layout(template="plotly_dark", height=400,
                          annotations=[dict(text="No data available", x=0.5, y=0.5, showarrow=False, font=dict(size=20))])
        return fig
    
    fig = go.Figure()
    fig.add_trace(go.Scatter(x=df['time'], y=df['close'], mode='lines', name='BTC Price',
                             line=dict(color='#3b82f6', width=4), fill='tozeroy', fillcolor='rgba(59, 130, 246, 0.1)'))
    
    fig.update_layout(template="plotly_dark", height=400, margin=dict(l=0,r=0,t=0,b=0),
                      xaxis=dict(showgrid=False, title="Date"),
                      yaxis=dict(showgrid=True, gridcolor='rgba(255,255,255,0.05)', title="Price (USD)"))
    return fig

# -----------------------------
# Main Backend Function
# -----------------------------
def run_dashboard():
    df = fetch_btc_data()
    current_time = datetime.utcnow() + timedelta(hours=5)   # PKT Time

    if df.empty:
        error_html = "<div style='text-align:center;padding:30px;color:#ef4444;'>⚠️ Unable to fetch data</div>"
        return None, error_html, error_html, error_html, error_html, error_html

    current_price = df.iloc[-1]['close']
    current_box = f"""
    <div class='current-price-card' style='text-align: center; padding: 20px;'>
        <div style='margin-bottom: 8px;'><span style='font-size: 0.75rem; color: #fbbf24;'>🕐 {current_time.strftime('%H:%M')}</span></div>
        <div style='font-size: 2rem; font-weight: 900; color: #fbbf24;'>${current_price:,.2f}</div>
        <div style='font-size: 0.75rem; color: #fcd34d;'>Current Bitcoin Price</div>
    </div>
    """

    # 1 Hour
    p1, t1, c1 = simulate_lstm_prediction(df, 1)
    time1 = current_time + timedelta(hours=1)
    pred1 = format_prediction_box(time1.strftime("%H:%M"), p1, t1, c1)

    # 2 Hours
    p2, t2, c2 = simulate_lstm_prediction(df, 2)
    time2 = current_time + timedelta(hours=2)
    pred2 = format_prediction_box(time2.strftime("%H:%M"), p2, t2, c2)

    # 1 Day
    p3, t3, c3 = simulate_lstm_prediction(df, 24)
    time3 = current_time + timedelta(hours=24)
    pred3 = format_prediction_box(time3.strftime("%b %d, %H:%M"), p3, t3, c3)

    plot = create_plot(df)

    analysis = f"""
    <div class='analysis-box' style='text-align: center; padding: 16px;'>
        <p style='color: #e0e7ff; font-size: 0.85rem; font-style: italic;'>💡 {random.choice([
            "Momentum persistence detected in recent sequences.",
            "Strong correlation with historical breakout patterns.",
            "Attention mechanisms highlighting key support levels."
        ])}</p>
    </div>
    """

    return plot, current_box, pred1, pred2, pred3, analysis


# Helper function for prediction box
def format_prediction_box(time_str, price, trend, confidence):
    if price is None:
        return "<div class='price-card' style='text-align:center;padding:20px;'><p style='color:#ef4444;'>Prediction unavailable</p></div>"
    
    if "HIGH" in trend:
        color = "#34d399"; icon = "📈"; bg = "rgba(52, 211, 153, 0.1)"
    else:
        color = "#f87171"; icon = "📉"; bg = "rgba(248, 113, 113, 0.1)"
    
    return f"""
    <div class='price-card' style='text-align: center; padding: 20px;'>
        <div style='margin-bottom:5px;'><span style='font-size:0.75rem;color:#94a3b8;'>🕐 {time_str}</span></div>
        <div style='font-size:1.5rem;font-weight:800;color:#e0e7ff;margin:8px 0;'>${price:,.2f}</div>
        <div style='background:{bg};padding:6px 12px;border-radius:16px;display:inline-block;'>
            <span style='color:{color};font-weight:700;'>{icon} {trend}</span>
        </div>
        <div style='margin-top:5px;font-size:0.7rem;color:#cbd5e1;'>Confidence: {confidence:.0f}%</div>
    </div>
    """