app.py

import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt
import base64
from fpdf import FPDF
from scripts.data_loader import load_running_data
from scripts.visualization import (
    plot_fastest_pace_per_shoe,
    plot_monthly_trends,
    plot_shoes_usage,
    plot_elevation_gain,
    plot_monthly_distance
)
from io import BytesIO
import tempfile

df = None

st.title("🏃🏻‍♀️ Running Data Dashboard")

# Load data from MongoDB
try:
    df = load_running_data()

    if df is None or df.empty:
        st.warning("No running data found for 2024.")
    else:
        # Remove "Unknown" 
        df = df[df['shoes'] != 'Unknown']

        # Sidebar selection 
        shoes_list = ["All"] + list(df['shoes'].unique())
        selected_shoe = st.sidebar.selectbox("Select Shoes", shoes_list, index=0)

        # Filter 
        if selected_shoe != "All":
            df = df[df['shoes'] == selected_shoe]
            st.write(f"✅ Filter applied: {selected_shoe}")
        else:
            st.write("✅ Showing all shoes")

        # Plot common graphs
        st.subheader("📊 Monthly Trends")
        monthly_trends_fig = plot_monthly_trends(df)
        st.pyplot(monthly_trends_fig)

        st.subheader("⛰️ Monthly Elevation Gain")
        elevation_gain_fig = plot_elevation_gain(df)
        st.pyplot(elevation_gain_fig)

        st.subheader("📍 Monthly Distance")
        monthly_distance_fig = plot_monthly_distance(df)
        st.pyplot(monthly_distance_fig)

        st.subheader("🚀 Fastest Pace per Shoe")
        fastest_pace_fig = plot_fastest_pace_per_shoe(df)
        st.pyplot(fastest_pace_fig)

        # Only show shoes usage for ALL
        shoes_usage_fig = None  
        if selected_shoe == "All":
            st.subheader("👟 Shoes Usage")
            shoes_usage_fig = plot_shoes_usage(df)
            st.pyplot(shoes_usage_fig)

except Exception as e:
    st.error(f"An error occurred: {e}")

# Generate CSV for download
if df is not None and not df.empty:
    csv = df.to_csv(index=False)
    b64_csv = base64.b64encode(csv.encode()).decode()
    href_csv = f'data:text/csv;base64,{b64_csv}'

    # Generate PDF for download
    pdf = FPDF()

    def save_plot_to_bytes(fig):
        """Save a Matplotlib figure to a BytesIO object"""
        img_stream = BytesIO()
        fig.savefig(img_stream, format='png')
        img_stream.seek(0)
        return img_stream

    def save_image_to_tempfile(image_stream):
        """Save image from BytesIO stream to a temporary file."""
        temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.png')
        temp_file.write(image_stream.read())
        temp_file.close()
        return temp_file.name

    # Convert figures to images 
    monthly_trends_img = save_plot_to_bytes(monthly_trends_fig)
    elevation_gain_img = save_plot_to_bytes(elevation_gain_fig)
    monthly_distance_img = save_plot_to_bytes(monthly_distance_fig)
    fastest_pace_img = save_plot_to_bytes(fastest_pace_fig)

    monthly_trends_temp = save_image_to_tempfile(monthly_trends_img)
    elevation_gain_temp = save_image_to_tempfile(elevation_gain_img)
    monthly_distance_temp = save_image_to_tempfile(monthly_distance_img)
    fastest_pace_temp = save_image_to_tempfile(fastest_pace_img)

    # Only save shoes usage if it exists
    shoes_usage_temp = None
    if shoes_usage_fig:
        shoes_usage_img = save_plot_to_bytes(shoes_usage_fig)
        shoes_usage_temp = save_image_to_tempfile(shoes_usage_img)

    # New page function
    def add_image_to_new_page(pdf, image_path):
        """Add image to a new page in the PDF."""
        pdf.add_page()
        pdf.image(image_path, x=10, y=30, w=180)

    # Add images to the PDF
    add_image_to_new_page(pdf, monthly_trends_temp)
    add_image_to_new_page(pdf, elevation_gain_temp)
    add_image_to_new_page(pdf, monthly_distance_temp)
    add_image_to_new_page(pdf, fastest_pace_temp)

    if shoes_usage_temp:
        add_image_to_new_page(pdf, shoes_usage_temp)

    # Create PDF
    pdf_output = pdf.output(dest='S').encode('latin1')
    pdf_output_io = BytesIO(pdf_output)
    pdf_output_io.seek(0)

    # Convert PDF to base64
    b64_pdf = base64.b64encode(pdf_output_io.read()).decode()
    href_pdf = f"data:application/pdf;base64,{b64_pdf}"

    # CSS for download buttons
    st.markdown(f"""
        <style>
            .download-button-container {{
                display: flex;
                justify-content: center;
                gap: 20px;
                margin-top: 50px;
            }}
            .download-button {{
                background-color: #FFD09B;
                color: black;
                border: none;
                padding: 15px 25px;
                font-size: 16px;
                border-radius: 5px;
                cursor: pointer;
                text-decoration: none;
                display: inline-block;
            }}
            .download-button:hover {{
                background-color: #FFB0B0;
            }}
        </style>
        <div class="download-button-container">
            <a href="{href_csv}" download="running_data.csv" class="download-button">Download Running Data as CSV</a>
            <a href="{href_pdf}" download="running_data.pdf" class="download-button">Download Charts as PDF</a>
        </div>
    """, unsafe_allow_html=True)