import gradio as gr
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import io

def trading_dashboard(digit, bias, digit_bias_list, bias_file, digit_sequence):
    # Bias interpretation + gauge zone
    if bias < 30:
        bias_msg = f"Bias {bias}% → Weak (Red Zone)"
        gauge_color = "red"
    elif 30 <= bias < 60:
        bias_msg = f"Bias {bias}% → Moderate (Yellow Zone)"
        gauge_color = "yellow"
    else:
        bias_msg = f"Bias {bias}% → Strong (Green Zone)"
        gauge_color = "green"

    # --- Bias Gauge Plot ---
    fig_gauge, axg = plt.subplots(figsize=(4,1))
    axg.barh(["Bias"], [bias], color=gauge_color)
    axg.set_xlim(0,100)
    axg.set_title("Bias Strength Gauge")
    axg.set_xlabel("%")
    for spine in axg.spines.values():
        spine.set_visible(False)

    # --- Digit Bias Analyzer (0–9 bar chart) ---
    if bias_file is not None:
        try:
            df = pd.read_csv(bias_file) if bias_file.name.endswith(".csv") else pd.read_excel(bias_file)
            digit_bias_values = df.iloc[0].values.tolist()[:10]
        except Exception:
            digit_bias_values = np.random.randint(0,100,10)
    elif digit_bias_list is not None and len(digit_bias_list) == 10:
        digit_bias_values = digit_bias_list
    else:
        digit_bias_values = np.random.randint(0,100,10)

    fig_digits, axd = plt.subplots()
    axd.bar(range(10), digit_bias_values, color="skyblue")
    axd.set_xticks(range(10))
    axd.set_xticklabels([str(i) for i in range(10)])
    axd.set_title("Digit Bias Analyzer (0–9)")
    axd.set_ylabel("Bias %")

    # --- Digit Psychology Module ---
    streaks = []
    bias_summary = {}
    if digit_sequence:
        digits = [int(d) for d in str(digit_sequence) if d.isdigit()]
        if digits:
            # Streak awareness
            current_streak = 1
            for i in range(1, len(digits)):
                if digits[i] == digits[i-1]:
                    current_streak += 1
                else:
                    streaks.append((digits[i-1], current_streak))
                    current_streak = 1
            streaks.append((digits[-1], current_streak))

            # Bias summary (frequency %)
            freq = pd.Series(digits).value_counts(normalize=True) * 100
            bias_summary = freq.to_dict()

    psychology_text = "📊 Digit Psychology Analysis\n\n"
    if streaks:
        psychology_text += "Streaks:\n" + "\n".join([f"Digit {d} → {s} times" for d,s in streaks]) + "\n\n"
    if bias_summary:
        psychology_text += "Bias Summary (%):\n" + "\n".join([f"Digit {d}: {round(p,1)}%" for d,p in bias_summary.items()]) + "\n\n"
    psychology_text += "Checklist:\n- Bias > 30%\n- Streak awareness checked\n- Indicators aligned (MACD/RSI)\n- Candle confirmation done"

    # Example market data
    data = pd.DataFrame(np.random.randn(50, 3), columns=['Price', 'Volume', 'Signal'])
    data['MACD'] = data['Price'].ewm(span=12).mean() - data['Price'].ewm(span=26).mean()
    data['RSI'] = 100 - (100 / (1 + (data['Price'].diff().clip(lower=0).rolling(14).mean() /
                                    data['Price'].diff().clip(upper=0).abs().rolling(14).mean())))

    # Plot MACD
    fig1, ax1 = plt.subplots()
    ax1.plot(data['Price'], label="Price")
    ax1.plot(data['MACD'], label="MACD", color="orange")
    ax1.set_title("Price vs MACD")
    ax1.legend()

    # Plot RSI
    fig2, ax2 = plt.subplots()
    ax2.plot(data['RSI'], label="RSI", color="green")
    ax2.axhline(70, linestyle="--", color="red")
    ax2.axhline(30, linestyle="--", color="blue")
    ax2.set_title("RSI Indicator")
    ax2.legend()

    # Checklist summary
    checklist = [
        "Bias > 30%",
        "Even/Odd bias matches entry",
        "Over/Under bias matches entry",
        "Indicators aligned (MACD/RSI)",
        "Final candle confirmation"
    ]

    # Master Trading Board Poster text
    poster = """
📌 Master Trading Board Poster

Bias Rules:
- Trade only if Bias > 30%
- Strong bias preferred (>60%)

Even/Odd Rules:
- Trade EVEN if bias favors even digits
- Trade ODD if bias favors odd digits

Over/Under Rules:
- Trade OVER if bias favors digits 5–9
- Trade UNDER if bias favors digits 0–4

Indicator Rules:
- MACD confirms trend
- RSI not overbought/oversold
- Candle pattern matches strategy
"""

    # --- Export results to CSV ---
    export_df = pd.DataFrame({
        "Digit Bias Values": digit_bias_values,
        "Checklist": checklist
    })
    buffer = io.StringIO()
    export_df.to_csv(buffer, index=False)
    buffer.seek(0)

    return bias_msg, fig_gauge, fig_digits, fig1, fig2, checklist, poster, psychology_text, buffer

with gr.Blocks() as demo:
    with gr.Row():
        with gr.Column():
            digit = gr.Number(label="Digit (0–9)")
            bias = gr.Slider(0, 100, step=5, label="Bias %")
            digit_bias_list = gr.Dataframe(headers=[str(i) for i in range(10)], row_count=1, col_count=10,
                                           label="Digit Bias Input (0–9)", type="numpy")
            bias_file = gr.File(label="Upload Bias Data (CSV/Excel)", file_types=[".csv", ".xlsx"])
            digit_sequence = gr.Textbox(label="Digit Sequence (e.g. 1234555777)")
            bias_out = gr.Textbox(label="Bias Interpretation")
            gauge_plot = gr.Plot(label="Bias Gauge")
            digit_bias_plot = gr.Plot(label="Digit Bias Analyzer")
            macd_plot = gr.Plot(label="MACD Chart")
            rsi_plot = gr.Plot(label="RSI Chart")
            psychology_out = gr.Textbox(label="Digit Psychology")
            download_out = gr.File(label="Download Results (CSV)")
        with gr.Column():
            checklist_out = gr.Label(label="Checklist")
            poster_out = gr.Textbox(label="Master Trading Board Poster")

    demo.load(trading_dashboard,
              inputs=[digit, bias, digit_bias_list, bias_file, digit_sequence],
              outputs=[bias_out, gauge_plot, digit_bias_plot, macd_plot, rsi_plot, checklist_out, poster_out, psychology_out, download_out])

demo.launch()