Py.Cafe

from dash import Dash, dcc, html, Input, Output
import dash_ag_grid as dag
import plotly.express as px
import pandas as pd
import numpy as np

df = pd.read_csv('Argentina-bilateral-instruments-1810-2023.csv')

# Filter out rows where the "Counterpart ENG" field is missing
df = df[df['Counterpart ENG'].notna()].copy()

# Standardize column names by stripping whitespace and replacing spaces with underscores
df.columns = df.columns.str.strip().str.replace(' ', '_')

# Use the provided "Sign_year" column as the Year of the treaty signing.
# Create a new column "Year" for consistency in further analysis.
df['Year'] = df['Sign_year']

# Create a "Decade" column by grouping "Year" into decades.
df['Decade'] = df['Year'].apply(lambda x: (x // 10 * 10) if pd.notnull(x) else np.nan)

# Use the provided "Region_ENG" column for region information.
df["Region"] = df["Region_ENG"].str.strip()

# Use a mapping dictionary to further standardize region names.
region_mapping = {
    # North America
    "United States": "North America",
    "USA": "North America",
    "Canada": "North America",
    "Mexico": "North America",
    
    # South America
    "Brazil": "South America",
    "Chile": "South America",
    "Peru": "South America",
    "Colombia": "South America",
    "Argentina": "South America",
    "Uruguay": "South America",
    "Paraguay": "South America",
    "Bolivia": "South America",
    "Ecuador": "South America",
    "Venezuela": "South America",
    
    # Europe
    "Germany": "Europe",
    "France": "Europe",
    "United Kingdom": "Europe",
    "Italy": "Europe",
    "Spain": "Europe",
    "Netherlands": "Europe",
    "Belgium": "Europe",
    "Sweden": "Europe",
    "Norway": "Europe",
    "Switzerland": "Europe",
    "Poland": "Europe",
    "Austria": "Europe",
    "Czech Republic": "Europe",
    "Slovakia": "Europe",
    "Austro-Hungarian Monarchy": "Europe",
    "Czechoslovakia": "Europe",
    "Yugoslavia": "Europe",
    "Serbia": "Europe",
    "Prussia": "Europe",
    
    # Eurasia / Near East
    "Russia": "Eurasia",
    "USSR": "Eurasia",
    "Turkey": "Eurasia",
    "Ottoman Empire": "Eurasia",
    
    # Asia
    "China": "Asia",
    "Japan": "Asia",
    "India": "Asia",
    "South Korea": "Asia",
    "Indonesia": "Asia",
    "Malaysia": "Asia",
    "Singapore": "Asia",
    
    # Oceania
    "Australia": "Oceania",
    "New Zealand": "Oceania",
    
    # Africa
    "Egypt": "Africa",
    "South Africa": "Africa",
    "Nigeria": "Africa",
    "Kenya": "Africa",
    "Morocco": "Africa",
    
    # Middle East
    "Saudi Arabia": "Middle East",
    "Israel": "Middle East",
    "Iran": "Middle East",
    "United Arab Emirates": "Middle East",
    "Qatar": "Middle East"
}
df["Region"] = df["Region"].map(lambda x: region_mapping.get(x, x))

# Aggregate treaty counts per year
treaties_per_year = df.groupby('Year').size().reset_index(name='Count')

# Aggregate treaty counts per decade
treaties_per_decade = df.groupby('Decade').size().reset_index(name='Count')

# Top 20 counterparts by number of treaties
top20 = df['Counterpart_ENG'].value_counts().nlargest(20).reset_index(name='Count')

def apply_argentina_style(fig):
    """
    Update the figure layout with Argentine color scheme.
    Sky blue (#74acdf) and dark blue (#0033a0) are used alongside white.
    """
    fig.update_layout(
        plot_bgcolor="#ffffff",    # white background for plot area
        paper_bgcolor="#ffffff",   # white background for the whole paper
        font_color="#0033a0",       # dark blue font color
        title_font_color="#0033a0", # dark blue title font color
        coloraxis=dict(colorbar=dict(bgcolor="#ffffff"))
    )
    return fig

# Create a bar chart for the top 20 counterparts
fig_bar = px.bar(
    top20,
    x='Counterpart_ENG',
    y='Count',
    title='Top 20 Counterparts by Number of Treaties',
    labels={
        'Counterpart_ENG': 'Counterpart',
        'Count': 'Number of Treaties'
    }
)

fig_bar = apply_argentina_style(fig_bar)

# Create a line chart for the treaty signings trend over the years
fig_line = px.line(
    treaties_per_year,
    x='Year',
    y='Count',
    title='Trend of Treaty Signings Over the Years',
    labels={
        'Year': 'Year',
        'Count': 'Number of Treaties'
    }
)
fig_line = apply_argentina_style(fig_line)

# Create a bar chart for treaties by decade
fig_decade = px.bar(
    treaties_per_decade,
    x='Decade',
    y='Count',
    title='Number of Treaties by Decade',
    labels={
        'Decade': 'Decade',
        'Count': 'Number of Treaties'
    }
)
fig_decade = apply_argentina_style(fig_decade)

# Enable suppress_callback_exceptions to allow callbacks for dynamically created components.
app = Dash(__name__, suppress_callback_exceptions=True)

app.layout = html.Div([
    html.H1(
        'Argentina Treaty Library Analysis',
        style={
            'textAlign': 'center',
            'color': '#0033a0'
        }
    ),

    # Year range slider filter
    html.Div([
        html.Label(
            'Select Year Range:',
            style={
                'fontWeight': 'bold',
                'color': '#0033a0'
            }
        ),
        dcc.RangeSlider(
            id='year-slider',
            min=int(df['Year'].min()),
            max=int(df['Year'].max()),
            value=[int(df["Year"].min()), int(df["Year"].max())],
            marks={
                str(year): str(year) for year in range(int(df["Year"].min()),
                                                       int(df["Year"].max()) + 1, 10)
            }
        )
    ], style={'margin': '20px'}),

    # Region Dropdown (single-select with an "All Regions" option)
    html.Div([
        html.Label(
            'Select Region:',
            style={
                'fontWeight': 'bold',
                'color': '#0033a0'
            }
        ),
        dcc.Dropdown(
            id='region-dropdown',
            options=[{'label': 'All Regions', 'value': 'all'}] +
                    [{'label': r, 'value': r} for r in sorted(df['Region'].unique())],
            value='all',
            multi= False,
            placeholder='Select region'
        )
    ], style={'margin': '20px'}),

    # Counterpart Dropdown (always in layout, initially hidden)
    html.Div([
        dcc.Dropdown(
            id='counterpart-dropdown',
            multi=True,
            placeholder='Select counterpart(s)',
            style={
                'display': 'none'
            }
        )
    ], id='counterpart-container', style={'margin': '20px'}),

    # Tabs for various analyses
    dcc.Tabs(
        id='tabs',
        value='tab-table',
        children=[
            dcc.Tab(
                label='Table',
                value='tab-table'
            ),
            dcc.Tab(
                label='Visualizations',
                value='tab-vis'
            ),
            dcc.Tab(
                label='Decade Analysis',
                value='tab-decade'
            ),
            dcc.Tab(
                label='Regional Analysis',
                value='tab-region'
            ),
            dcc.Tab(
                label='Deep Analytics',
                value='tab-deep'
            ),
            dcc.Tab(
                label='Tags Analysis',
                value='tab-tags'
            )
        ]
    ),

    html.Div(
        id='tabs-content'
    )
])

@app.callback(
    [Output('counterpart-dropdown', 'options'),
     Output('counterpart-dropdown', 'style')],
    Input('region-dropdown', 'value')
)
def update_counterpart_dropdown(selected_region):
    # If "All Regions" is selected or no region is selected, hide the counterpart dropdown.
    if not selected_region or selected_region == 'all':
        return [], {'display': 'none'}
    else:
        filtered = df[df['Region'] == selected_region]
        options = [{'label': i, 'value': i} for i in sorted(filtered['Counterpart_ENG'].unique())]
        return options, {'display': 'block'}

@app.callback(
    Output('tabs-content', 'children'),
    Input('tabs', 'value'),
    Input('year-slider', 'value'),
    Input('counterpart-dropdown', 'value'),
    Input('region-dropdown', 'value')
)
def render_tab_content(tab, year_range, selected_counterparts, selected_region):
    # Convert selected_region to string if it comes as a list
    if isinstance(selected_region, list):
        selected_region = selected_region[0] if selected_region else None
    
    # Create a copy of the data for filtering.
    filtered = df.copy()
    
    if year_range:
        filtered = filtered[(filtered['Year'] >= year_range[0]) & (filtered['Year'] <= year_range[1])]

    if selected_region and selected_region != 'all':
        filtered = filtered[filtered['Region'] == selected_region]
    
    if selected_counterparts:
        filtered = filtered[filtered['Counterpart_ENG'].isin(selected_counterparts)]
    
    if tab == 'tab-table':
        grid = dag.AgGrid(
            id="data-grid",
            rowData=filtered.to_dict('records'),
            columnDefs=[{'field': col, 'filter': True, 'sortable': True} for col in filtered.columns],
            dashGridOptions={
                'pagination': True,
                'paginationPageSize': 20
            }
        )
        return grid
    
    elif tab == 'tab-vis':
        treaties_year = filtered.groupby('Year').size().reset_index(name='Count')
        top20_filtered = filtered['Counterpart_ENG'].value_counts().nlargest(20).reset_index()
        top20_filtered.columns = ['Counterpart_ENG', 'Count']
        
        fig_line_filtered = px.line(
            treaties_year,
            x='Year',
            y='Count',
            title='Trend of Treaty Signings Over the Years (Filtered Data)',
            labels={
                'Year': 'Year',
                'Count': 'Number of Treaties'
            }
        )
        fig_line_filtered = apply_argentina_style(fig_line_filtered)
        
        fig_bar_filtered = px.bar(
            top20_filtered,
            x='Counterpart_ENG',
            y='Count',
            title='Top 20 Counterparts (Filtered Data)',
            labels={
                'Counterpart_ENG': 'Counterpart',
                'Count': 'Number of Treaties'
            }
        )
        fig_bar_filtered = apply_argentina_style(fig_bar_filtered)
        
        return html.Div([
            dcc.Graph(figure=fig_line_filtered),
            dcc.Graph(figure=fig_bar_filtered)
        ])
    
    elif tab == 'tab-decade':
        treaties_decade = filtered.groupby('Decade').size().reset_index(name='Count')
        fig_decade_filtered = px.bar(
            treaties_decade,
            x='Decade',
            y='Count',
            title='Number of Treaties by Decade (Filtered Data)',
            labels={
                'Decade': 'Decade',
                'Count': 'Number of Treaties'
            }
        )
        fig_decade_filtered = apply_argentina_style(fig_decade_filtered)
        return html.Div([dcc.Graph(figure=fig_decade_filtered)])
    
    elif tab == 'tab-region':
        treaties_by_region_year = filtered.groupby(['Year', 'Region']).size().reset_index(name='Count')
        fig_region_year = px.area(
            treaties_by_region_year,
            x='Year',
            y='Count',
            color='Region',
            title='Trend of Treaty Signings by Region Over Time',
            labels={
                'Year': 'Year',
                'Count': 'Number of Treaties',
                'Region': 'Region'
            }
        )
        fig_region_year = apply_argentina_style(fig_region_year)
        
        treaties_by_region = filtered.groupby('Region').size().reset_index(name='Count')
        fig_region_bar = px.bar(
            treaties_by_region,
            x='Region',
            y='Count',
            title='Total Number of Treaties by Region',
            labels={
                'Region': 'Region',
                'Count': 'Number of Treaties'
            }
        )
        fig_region_bar = apply_argentina_style(fig_region_bar)
        
        return html.Div([
            dcc.Graph(figure=fig_region_year),
            dcc.Graph(figure=fig_region_bar)
        ])
    
    elif tab == 'tab-deep':
        deep_df = filtered.copy()
        deep_df['Decade'] = deep_df['Decade'].fillna('Unknown')
        region_decade_counts = deep_df.groupby(['Region', 'Decade']).size().reset_index(name='Count')
        
        fig_treemap = px.treemap(
            region_decade_counts,
            path=['Region', 'Decade'],
            values='Count',
            title='Treemap of Treaties by Region and Decade'
        )
        fig_treemap = apply_argentina_style(fig_treemap)
        
        pivot_df = region_decade_counts.pivot(index='Region', columns='Decade', values='Count').fillna(0)
        fig_heatmap = px.imshow(
            pivot_df,
            title='Heatmap of Treaty Counts by Region and Decade',
            labels=
            {'x': 'Decade',
             'y': 'Region',
             'color': 'Count'},
            text_auto=True
        )
        fig_heatmap = apply_argentina_style(fig_heatmap)
        
        return html.Div([
            dcc.Graph(figure=fig_treemap),
            dcc.Graph(figure=fig_heatmap)
        ])
    
    elif tab == 'tab-tags':
        tags_series = filtered['Tags'].dropna().str.split('\n')
        all_tags = [tag.strip() for sublist in tags_series for tag in sublist if tag.strip() != '']
        tag_counts = pd.Series(all_tags).value_counts().reset_index()
        tag_counts.columns = ['Tag', 'Count']
        
        fig_tags = px.bar(
            tag_counts,
            x='Tag',
            y='Count',
            title='Frequency of Tags in Treaty Data',
            labels={'Tag': 'Tag', 'Count': 'Frequency'}
        )
        fig_tags = apply_argentina_style(fig_tags)
        
        return html.Div([dcc.Graph(figure=fig_tags)])

if __name__ == '__main__':
    app.run_server(debug=True)