import streamlit as st
import yfinance as yf
import plotly.graph_objects as go
from prophet import Prophet
import pandas as pd
import matplotlib.pyplot as plt
from statsmodels.tsa.seasonal import seasonal_decompose

from statsmodels.graphics.gofplots import qqplot

# Set Streamlit options
st.set_page_config(layout="wide")  # Use wide page layout
st.set_option('deprecation.showPyplotGlobalUse', False)


# Header section with enhanced aesthetics
def display_header():
    st.sidebar.title("Navigation")
    st.sidebar.info("""
        This application is designed to provide users with financial insights and predictive analysis 
        on various stocks using real-time data from Yahoo Finance. Created by Gokul Palanisamy.
    """)
    st.sidebar.title("About Us")
    st.sidebar.info("""
        Developed by Gokul Palanisamy, this tool helps users make informed investment decisions by 
        analyzing historical data and predicting future stock trends.
    """)
    st.sidebar.title("Contact Us")
    st.sidebar.info("""
        Email: [gokulp@bu.edu](mailto:gokulp@bu.edu)
        Phone: +1 (857) 832-0441
        More Information: [Gokul Palanisamy](https://www.linkedin.com/in/gokulp/)
    """)


display_header()

# Input widgets for stock symbols and settings
st.markdown("## StocX AI")
col1, col2, col3 = st.columns(3)
with col1:
    stock1 = st.text_input('Enter Stock Ticker Symbol 1', value='', key='stock1_input')
with col2:
    stock2 = st.text_input('Enter Stock Ticker Symbol 2', value='', key='stock2_input')
with col3:
    daysago = st.text_input('Select Time Frame in Days (write "max" for maximum time)', value='1y', key='daysago_input')
forecast_out = st.slider('Predicted Days Ahead', 1, 180, 30)


# Fetch and display stock data
def fetch_data(ticker, period):
    return yf.Ticker(ticker).history(period=period)


data1, data2 = None, None
if stock1:
    data1 = fetch_data(stock1, daysago)
    st.write(f'### {stock1} Stock Data')
    st.write(data1)

if stock2:
    data2 = fetch_data(stock2, daysago)
    st.write(f'### {stock2} Stock Data')
    st.write(data2)


# Function to analyze and compare stocks
def compare_stocks(stock1, data1, stock2, data2):
    if data1 is not None and data2 is not None:
        # Calculate percentage change for each stock
        pct_change1 = (data1['Close'].iloc[-1] - data1['Close'].iloc[0]) / data1['Close'].iloc[0] * 100
        pct_change2 = (data2['Close'].iloc[-1] - data2['Close'].iloc[0]) / data2['Close'].iloc[0] * 100

        better_stock = stock1 if pct_change1 > pct_change2 else stock2
        reason = f"{better_stock} had a higher percentage change ({max(pct_change1, pct_change2):.2f}%) over the period."

        st.write("## Stock Performance Comparison")
        st.write(f"**{stock1}** percentage change: {pct_change1:.2f}%")
        st.write(f"**{stock2}** percentage change: {pct_change2:.2f}%")
        st.write(f"**Better Performing Stock:** {better_stock}")
        st.write(f"**Reason:** {reason}")


if stock1 and stock2 and data1 is not None and data2 is not None:
    compare_stocks(stock1, data1, stock2, data2)


# Comprehensive analysis functions
def perform_analysis(stock, data):
    if data is not None and not data.empty:
        st.write(f"### {stock} Detailed Analysis")

        # Time Series Decomposition
        st.write(f"#### Time Series Decomposition")
        decomposition = seasonal_decompose(data['Close'], period=30, model='additive')
        fig, (ax1, ax2, ax3, ax4) = plt.subplots(4, 1, figsize=(10, 8))
        decomposition.observed.plot(ax=ax1, title='Observed')
        decomposition.trend.plot(ax=ax2, title='Trend')
        decomposition.seasonal.plot(ax=ax3, title='Seasonal')
        decomposition.resid.plot(ax=ax4, title='Residual')
        plt.tight_layout()
        st.pyplot()

        # Prophet Prediction
        st.write(f"#### Prophet Forecast")
        df_prophet = pd.DataFrame(data={'ds': data.index, 'y': data['Close']})
        df_prophet['ds'] = pd.to_datetime(df_prophet['ds']).dt.tz_localize(None)  # Make timezone naive
        model = Prophet()
        model.fit(df_prophet)
        future = model.make_future_dataframe(periods=forecast_out)
        forecast = model.predict(future)
        fig = plot_prophet_forecast(model, forecast)
        st.plotly_chart(fig)


# Helper function for Plotly forecast visualization
def plot_prophet_forecast(model, forecast):
    fig = go.Figure()
    fig.add_trace(go.Scatter(x=model.history['ds'], y=model.history['y'], mode='lines', name='Actual'))
    fig.add_trace(go.Scatter(x=forecast['ds'], y=forecast['yhat'], mode='lines+markers', name='Forecast'))
    fig.add_trace(go.Scatter(x=forecast['ds'], y=forecast['yhat_upper'], fill=None, mode='lines',
                             line=dict(color='gray', dash='dash'), name='Upper Confidence Interval'))
    fig.add_trace(go.Scatter(x=forecast['ds'], y=forecast['yhat_lower'], fill='tonexty', mode='lines',
                             line=dict(color='gray', dash='dash'), name='Lower Confidence Interval'))
    fig.update_layout(title='Prophet Forecast and Confidence Intervals', xaxis_title='Date', yaxis_title='Values',
                      hovermode='x')
    return fig


# Call analysis functions if data is available
if stock1 and data1 is not None:
    perform_analysis(stock1, data1)
if stock2 and data2 is not None:
    perform_analysis(stock2, data2)