app.py

import streamlit as st
import pandas as pd
import plotly.express as px
#import bar_chart_race as bcr
from raceplotly.plots import barplot



# Configuration and Constants
COUNTRY_MAPPING = {
    "Italy, San Marino and the Holy See": "Italy",
    "France and Monaco": "France",
    "Belgium and Luxembourg": "Belgium",
    "China (mainland)": "China",
    "United States of America": "United States",
    "United Kingdom of Great Britain and Northern Ireland": "United Kingdom",
    "Spain and Andorra": "Spain"
}

DEFAULT_COUNTRIES = ["Italy", "France", "Germany"]
YEAR_RANGE = (2000, 2020)

# Data Loading and Processing Functions
@st.cache_data
def load_data(sheet_name, year_range = (2000, 2020), sector=None):
    df = pd.read_excel("dati/fossilco2emission.xlsx", sheet_name=sheet_name)
    df_mapped = df.copy()
    df_mapped['Country'] = df_mapped['Country'].replace(COUNTRY_MAPPING)
    
    year_cols = list(range(year_range[0], year_range[1] + 1))
    if sector:
        selected_cols = ['Country', 'Sector'] + year_cols
    else:
        selected_cols = ['Country'] + year_cols
    return df_mapped[selected_cols].copy()

def process_data_for_line_plot(df, selected_countries, year_range):
    mask = df['Country'].isin(selected_countries)
    filtered_df = df[mask]
    
    df_melted = filtered_df.melt(
        id_vars=['Country'],
        value_vars=range(year_range[0], year_range[1] + 1),
        var_name='Year',
        value_name='Emissions'
    )
    df_melted['Year'] = pd.to_numeric(df_melted['Year'])
    return df_melted

# Visualization Functions
def create_line_plot(data):
    fig = px.line(
        data,
        x='Year',
        y='Emissions',
        color='Country',
        title='CO2 Emissions Over Time',
        labels={'Emissions': 'CO2 Emissions per Capita (Mton)'},
        hover_data={'Year': True, 'Emissions': ':.2f'}
    )
    fig.update_layout(height=600, hovermode='x unified')
    return fig

def create_animated_choropleth(data, start_year, end_year):
    df_map = data.melt(
        id_vars=['Country'],
        value_vars=range(start_year, end_year + 1),
        var_name='Year',
        value_name='Emissions'
    )
    
    fig_map = px.choropleth(
        df_map,
        locations='Country',
        locationmode='country names',
        color='Emissions',
        animation_frame='Year',
        title='CO2 Emissions per Capita Over Time',
        color_continuous_scale='Reds',
        range_color=[0, df_map['Emissions'].quantile(0.95)],
        labels={'Emissions': 'CO2 Emissions per Capita (Mton)'}
    )
    
    fig_map.update_layout(
        height=600,
        margin=dict(l=0, r=0, t=30, b=0),
        updatemenus=[{
            'type': 'buttons',
            'showactive': False,
            'buttons': [
                dict(label='Play',
                     method='animate',
                     args=[None, {'frame': {'duration': 500, 'redraw': True},
                                'fromcurrent': True}]),
                dict(label='Pause',
                     method='animate',
                     args=[[None], {'frame': {'duration': 0, 'redraw': False},
                                  'mode': 'immediate',
                                  'transition': {'duration': 0}}])
            ]
        }]
    )
    return fig_map


def create_race_plot(df, year_range):
    # Prepare data for race plot
    # Convert year columns to rows for raceplotly format
    df_race = df.melt(
        id_vars=['Country'],
        value_vars=range(year_range[0], year_range[1] + 1),
        var_name='Year',
        value_name='Emissions'
    )
    
    # Create the race plot
    race_plot = barplot(
        df_race,
        item_column='Country',
        value_column='Emissions',
        time_column='Year',
        top_entries=10,
    )
    
    # Plot with custom settings
    fig = race_plot.plot(
        title='Top 10 Countries by CO2 Emissions',
        orientation='horizontal',
        item_label='Country',
        value_label='CO2 Emissions (Mton)',
        time_label='Year: ',
        frame_duration=800
    )

    # fig.update_layout(
    # height=700,  # Make plot taller
    # font=dict(size=12),  # Increase base font size
    # title_font_size=20,  # Larger title
    # xaxis_title_font_size=16,  # Larger axis titles
    # yaxis_title_font_size=16,
    # yaxis_tickfont_size=14,  # Larger tick labels
    # xaxis_tickfont_size=14
    # )
    
    return fig

def create_covid_impact_plot(df1, df2):
    # Create tabs for different COVID analyses
    tab1, tab2 = st.tabs(["Global Impact", "Sectoral Impact"])
    print(df2.head())
    
    with tab1:
        # Global emissions around COVID
        years_covid = range(2017, 2023)
        global_emissions = df1[df1['Country'] == 'GLOBAL TOTAL']
        emissions_covid = global_emissions[list(years_covid)].values[0]
        
        fig_global = px.line(
            x=years_covid, 
            y=emissions_covid,
            title='Global CO2 Emissions Around COVID-19',
            labels={'x': 'Year', 'y': 'CO2 Emissions (Mt CO2)'}
        )
        fig_global.add_vline(x=2020, line_dash="dash", line_color="red",
                            annotation_text="COVID-19")
        st.plotly_chart(fig_global, use_container_width=True)
        
        # Calculate and display percentage changes
        col1, col2 = st.columns(2)
        with col1:
            change_2020 = ((emissions_covid[3] - emissions_covid[2])/emissions_covid[2]*100)
            st.metric("2020 Emissions Change", f"{change_2020:.1f}%")
        with col2:
            change_2021 = ((emissions_covid[4] - emissions_covid[3])/emissions_covid[3]*100)
            st.metric("2021 Recovery", f"{change_2021:.1f}%")
    
    with tab2:
        # Sectoral analysis
        sectors = ['Power Industry', 'Industrial Combustion', 'Transport', 'Processes']
        sector_data = {}
        
        for sector in sectors:
            sector_emissions = df2[(df2['Sector'] == sector) & 
                                 (df2['Country'] == 'GLOBAL TOTAL')]
            sector_data[sector] = sector_emissions[list(years_covid)].values[0]
        
        # Create DataFrame for plotly
        df_sectors = pd.DataFrame(sector_data, index=years_covid).reset_index()
        df_sectors_melted = df_sectors.melt('index', var_name='Sector', 
                                          value_name='Emissions')
        
        fig_sectors = px.line(
            df_sectors_melted,
            x='index',
            y='Emissions',
            color='Sector',
            title='CO2 Emissions by Sector Around COVID-19'
        )
        fig_sectors.add_vline(x=2020, line_dash="dash", line_color="red",
                             annotation_text="COVID-19")
        st.plotly_chart(fig_sectors, use_container_width=True)
        
        # Display sector-specific impacts
        st.subheader("Sector Impact (2019-2020)")
        cols = st.columns(len(sectors))
        for i, sector in enumerate(sectors):
            change = ((sector_data[sector][3] - sector_data[sector][2])/
                     sector_data[sector][2]*100)
            cols[i].metric(sector, f"{change:.1f}%")

def create_agreements_timeline(df1):
    # Get the global total data
    global_data = df1[df1['Country'] == 'GLOBAL TOTAL'].iloc[0]
    
    # Get only the year columns (1970 to 2023)
    year_columns = [col for col in df1.columns if str(col).isdigit()]
    years = [int(col) for col in year_columns]
    values = [global_data[year] for year in years]
    
    # Create the plot using plotly
    fig = px.line(
        x=years, 
        y=values,
        title='Global CO2 Emissions and Key Climate Agreements',
        labels={'x': 'Year', 'y': 'CO2 Emissions (Mt CO2)'}
    )
    
    # Add vertical lines for key agreements
    agreements = {
        1997: 'Kyoto Protocol Adopted',
        2005: 'Kyoto Protocol Enforced',
        2015: 'Paris Agreement'
    }
    
    colors = {'1997': 'red', '2005': 'green', '2015': 'orange'}
    
    for year, agreement in agreements.items():
        fig.add_vline(
            x=year,
            line_dash="dash",
            line_color=colors[str(year)],
            annotation_text=agreement,
            annotation_position="top"
        )
    
    # Customize layout
    fig.update_layout(
        hovermode='x unified',
        showlegend=False,
        height=600
    )
    
    return fig



# Main App Function
def main():
    st.set_page_config(page_title="CO2 Emissions Dashboard", layout="wide")
    st.title("Global CO2 Emissions Dashboard")
    
    # Load Data
    df1 = load_data("fossil_CO2_per_capita_by_countr")
    df2 = load_data("fossil_CO2_totals_by_country")
    df3 = load_data("fossil_CO2_by_sector_country_su", year_range=(2017, 2023), sector="Sector")
    df_covid = load_data("fossil_CO2_totals_by_country", year_range=(2017, 2023))
    df_poliicy = load_data("fossil_CO2_totals_by_country", year_range=(1970,2023))
    
    df2 = df2[df2['Country'] != 'International Shipping']
    df_only_countries = df2.copy()[:210]
    
    # Sidebar Controls
    #st.sidebar.header("Controls")

    #st.sidebar.markdown("---")
    st.sidebar.image("dati/SIAM-logo.jpg", width=150)

    visualization_type = st.sidebar.radio(
        "Choose Visualization",
        ["Time Series Plot", "Animated World Map", "Bar Chart Race", "COVID-19 Impact", "Climate Agreements Timeline"]
    )
    
    # Year range selector (common to both visualizations)
    year_range = st.sidebar.slider(
        "Select Year Range",
        min_value=YEAR_RANGE[0],
        max_value=YEAR_RANGE[1],
        value=YEAR_RANGE
    )
    
    # Conditional controls and display
    if visualization_type == "Time Series Plot":
        st.subheader("CO2 Emissions Time Series")
        
        # Show country selector only for time series
        countries = df1['Country'].unique().tolist()
        selected_countries = st.sidebar.multiselect(
            "Select countries to compare",
            options=countries,
            default=DEFAULT_COUNTRIES
        )
        
        # Process and display time series plot
        df_processed = process_data_for_line_plot(df1, selected_countries, year_range)
        fig = create_line_plot(df_processed)
        st.plotly_chart(fig, use_container_width=True)
        
    elif visualization_type == "Animated World Map":
        st.subheader("Global Emissions Map (Animated)")
        fig_map = create_animated_choropleth(df1, year_range[0], year_range[1])
        st.plotly_chart(fig_map, use_container_width=True)

    elif visualization_type == "Climate Agreements Timeline":
        st.subheader("Global Emissions and Climate Agreements")
        
        # Add some context about the agreements
        with st.expander("About the Climate Agreements"):
            st.markdown("""
            - **Kyoto Protocol (1997)**: First legally binding agreement to reduce greenhouse gases
            - **Kyoto Protocol Enforcement (2005)**: The protocol came into force
            - **Paris Agreement (2015)**: Global agreement to limit temperature rise to well below 2°C
            """)
        
        # Create and display the plot
        fig = create_agreements_timeline(df_poliicy)
        st.plotly_chart(fig, use_container_width=True)
        
        # Add some analysis
        st.markdown("### Key Observations")
        col1, col2, col3 = st.columns(3)
        
        # Calculate some metrics
        kyoto_change = ((float(df_poliicy[df_poliicy['Country'] == 'GLOBAL TOTAL'][2005]) - 
                        float(df_poliicy[df_poliicy['Country'] == 'GLOBAL TOTAL'][1997])) / 
                       float(df_poliicy[df_poliicy['Country'] == 'GLOBAL TOTAL'][1997]) * 100)
        
        paris_change = ((float(df_poliicy[df_poliicy['Country'] == 'GLOBAL TOTAL'][2023]) - 
                        float(df_poliicy[df_poliicy['Country'] == 'GLOBAL TOTAL'][2015])) / 
                       float(df_poliicy[df_poliicy['Country'] == 'GLOBAL TOTAL'][2015]) * 100)
        
        with col1:
            st.metric("Emissions Change 1997-2005", f"{kyoto_change:.1f}%")
        with col2:
            st.metric("Emissions Change 2015-2023", f"{paris_change:.1f}%")
        with col3:
            latest_emissions = float(df_poliicy[df_poliicy['Country'] == 'GLOBAL TOTAL'][2023])
            st.metric("Current Emissions (Mt CO2)", f"{latest_emissions:.1f}")

    elif visualization_type == "COVID-19 Impact":
        st.subheader("COVID-19 Impact Analysis")
        create_covid_impact_plot(df_covid, df3)

    else:
        st.subheader("Top 10 CO2 Emitters Race")
        fig_race = create_race_plot(df_only_countries, year_range)
        st.plotly_chart(fig_race, use_container_width=True)
    
    st.sidebar.markdown("---")
    st.sidebar.markdown(""" GRUPPO 5 (EMANUELA, FULVIO, MARCO, TINSAE) """)

if __name__ == "__main__":
    main()