#add missing packages to kernel
import sys
!{sys.executable} -m pip install geopandas gpdvega

import pandas as pd
import datadotworld as dw
import os
import altair as alt
import datetime
import geopandas as gpd
import gpdvega

#def
cwd= os.getcwd()
cwd

#Mother Jones 
df1_details=pd.read_csv('https://docs.google.com/spreadsheets/d/e/2PACX-1vQBEbQoWMn_P81DuwmlQC0_jr2sJDzkkC0mvF6WLcM53ZYXi8RMfUlunvP1B5W0jRrJvH-wc-WGjDB1/pub?gid=0&single=true&output=csv')
df1_details.info()
print('-----------------------')
print('\n')
print('\n')
df1_details.head(2)

# The Associated Press/USA TODAY/Northeastern University Mass
loc=os.path.join(cwd,'datasets', 'associatedpress-mass-killings-public')
df2_incidents=pd.read_csv(loc+'/'+'mass_killing_incidents_public.csv')
df2_offenders=pd.read_csv(loc+'/'+'mass_killing_offenders_public.csv')
df2_victims=pd.read_csv(loc+'/'+'mass_killing_victims_public.csv')
df2_weapons=pd.read_csv(loc+'/'+'mass_killing_weapons_public.csv')

print('mass_killing_incidents_public.csv')
print(df2_incidents.head(2))
print('-----------------------')
print('\n')
print('\n')
print('mass_killing_offenders_public.csv')
print(df2_offenders.head(2))
print('-----------------------')
print('\n')
print('\n')
print('mass_killing_victims_public.csv')
print(df2_victims.head(2))
print('-----------------------')
print('\n')
print('\n')
print('mass_killing_weapons_public.csv')
print(df2_weapons.head(2))
print('-----------------------')

loc=os.path.join(cwd, 'dataframes')
df2_offenders.to_csv(loc + '/df2_offenders.csv')
df2_victims.to_csv(loc + '/df2_victims.csv')
df2_weapons.to_csv(loc + '/df2_weapons.csv')

#creating source data by joining all sources by state
import geopandas as gpd

column_names=['state_name', 'drawseq', 'state_fips', 'sub_region', 'state','geometry']        
gdf = gpd.read_file(os.path.join(os.getcwd(),'datasets','states_21basic','states.shp'))
gdf.columns=column_names

df_gun_violence=(df2_incidents
      .loc[:, ['state', 'tot_victims', 'longitude','latitude', 'incident_id']]
      .groupby(df2_incidents.state)
      .agg(
          tot_victims=pd.NamedAgg(column="tot_victims", aggfunc="sum"),
          num_incidents=pd.NamedAgg(column="incident_id", aggfunc="count"),
          latitude=pd.NamedAgg(column="latitude", aggfunc="mean"),
          longitude=pd.NamedAgg(column="longitude", aggfunc="mean"),
      )
      .reset_index()  
     )

latest_state_population
df3_states=pd.merge(left=gdf,right=df_gun_violence, how='right', on='state') 
df3_states=pd.merge(left=latest_state_population,right=source, how='right', on='state')

loc=os.path.join(cwd, 'dataframes')
df2_incidents.to_csv((loc + '/df3_states'))

def my_theme():
    markColor = '#30a2da';
    axisColor = '#cbcbcb';
    guideLabelColor = '#999';
    guideTitleColor = '#333';
    backgroundColor = '#f0f0f0';
    blackTitle = '#333';
    
    return {
        'config': {
            'arc': {'fill': markColor},
            'area': {'fill': markColor},
            'axis': {
                'domainColor': axisColor,
                'grid': True,
                'gridColor': axisColor,
                'gridOpacity':0.3,
                'gridWidth': 1,
                'labelColor': guideLabelColor,
                'labelFontSize': 10,
                'titleColor': guideTitleColor,
                'tickColor': axisColor,
                'tickSize': 10,
                'titleFontSize': 14,
                'titlePadding': 10,
                'labelPadding': 4,
              },
            'axisBand': {
                'grid': False,
            },
            'background': backgroundColor,

            'group': {
                'fill': backgroundColor,
              },

            'legend': {
                'labelColor': blackTitle,
                'labelFontSize': 11,
                'padding': 1,
                'symbolSize': 30,
                'symbolType': 'square',
                'titleColor': blackTitle,
                'titleFontSize': 14,
                'titlePadding': 10,
              },

            'line': {
                'stroke': markColor,
                'strokeWidth': 2,
              },

            'path': {'stroke': markColor, 'strokeWidth': 0.5},
            'rect': {'fill': markColor},

              'range': {
                'category': [
                  '#30a2da',
                  '#fc4f30',
                  '#e5ae38',
                  '#6d904f',
                  '#8b8b8b',
                  '#b96db8',
                  '#ff9e27',
                  '#56cc60',
                  '#52d2ca',
                  '#52689e',
                  '#545454',
                  '#9fe4f8',
                ],

                'diverging': ['#cc0020', '#e77866', '#f6e7e1', '#d6e8ed', '#91bfd9', '#1d78b5'],
                'heatmap': ['#d6e8ed', '#cee0e5', '#91bfd9', '#549cc6', '#1d78b5'],
              },

              'point': {
                'filled': True,
                'shape': 'circle',
              },

              'shape': {'stroke': markColor},

              'bar': {
                'binSpacing': 2,
                'fill': markColor,
                'stroke': None
              },

              'title': {
                'anchor': 'start',
                'fontSize': 24,
                'fontWeight': 600,
                'offset': 20,
              },
#             
            'mark': {
                'color': 'black',
                'fill': None
            }
        }
    }
# register the custom theme under a chosen name
alt.themes.register('my_theme', my_theme)
# enable the newly registered theme
alt.themes.enable('my_theme')

title= 'Number shootings per year'
subtitle="2019 holds the record of shooting incidents"

source=df2_incidents

base = alt.Chart(source).properties(width=800)

chart=base.mark_bar().encode(
    x=alt.X('year(date):O', axis=alt.Axis(title='year')),
    y=alt.Y('count(incident_id):Q',axis=alt.Axis(title='count shootings'))
    ).properties(
    title={
      "text": [title], 
      "subtitle": [subtitle]
    }
)
    
bar_inc_per_year=chart
bar_inc_per_year

var='situation_type'
title= 'Most shootings happen in residences due to interpersonal conflict. '
subtitle='However, indiscriminate shootings are much more deadly'
brush = alt.selection_multi(
    fields=[var], nearest=True, on='click'
)
source=df2_incidents

base = alt.Chart(source).properties(width=800).properties(
    title={
      "text": [title], 
      "subtitle": [subtitle]
    })

chart=base.mark_bar().encode(
    x=alt.X('count(incident_id):Q', axis=alt.Axis(title='num. of shootings')),
    y=alt.Y(var,sort=alt.EncodingSortField(field='incident_id', order='descending', op='count'),axis=alt.Axis(title=var.replace('_', " "))),
    color=alt.condition(brush, 'sum(tot_victims)', alt.value('lightgray')),
    order=alt.Order('count(incident_id)', sort='descending'),
    tooltip=['count(incident_id)']
).add_selection(brush)

    
chart

chart & (chart_1 | chart_2)

df2_incidents.head(3)

title= 'Total Victims per year'
subtitle="2019 holds the record of shooting incidents"

source=df2_incidents

base = alt.Chart(source).properties(width=800)

chart=base.mark_bar().encode(
    x=alt.X('year(date):O', axis=alt.Axis(title=None)),
    y=alt.Y('sum(tot_victims):Q',axis=alt.Axis(title='victims'))
    ).properties(
    title={
      "text": [title], 
      "subtitle": [subtitle]
    }
)

    
bar_vic_per_year=chart
bar_vic_per_year

#comb_inc_and_vict
from altair import datum

title= 'Total Victims vs Number of Incidents per year '
subtitle=["Although significant decrease in incidents tends to follow decrease in victims, the relationship between both variables is not directly proportional as shown in 2017", "and 2019."]

source=df2_incidents

base = alt.Chart(source).properties(width=600).encode(alt.X('year(date):O', axis=alt.Axis(title=None)))

bar=base.mark_bar(opacity=0.8).encode(y=alt.Y('sum(tot_victims):Q',axis=alt.Axis(title='victims', titleColor='#30a2da')))

line=base.mark_line(color='#fc4f30').encode(alt.Y('count(incident_id):Q',axis=alt.Axis(title='count shootings', titleColor='#fc4f30')))


chart_2=alt.layer(bar,line).resolve_scale(y='independent').properties(
    title={
      "text": title, 
      "subtitle": subtitle
    })

chart_2

## 2017 increase in victims is caused by Las Vegas shooting with >900 victims
df2_incidents[(df2_incidents.date.dt.year==2017)&(df2_incidents.state=='NV')].loc[:,'narrative']

#trying to make intearctive
from altair import datum

title= 'Total Victims vs Number of Incidents per year '
subtitle=["Although significant decrease in incidents tends to follow decrease in victims, the relationship between both variables is not directly proportional as shown in 2017", "and 2019."]

brush = alt.selection_single(on='mouseover', nearest=True)
# brush = alt.selection_multi(
#     fields=['year(date)'], nearest=True, on='click'
# )

source=df2_incidents

base = alt.Chart(source).properties(width=600).encode(alt.X('year(date):O', axis=alt.Axis(title=None)))

bar=base.mark_bar(opacity=0.8).encode(
    y=alt.Y('sum(tot_victims):Q',axis=alt.Axis(title='victims', titleColor='#30a2da')),
    tooltip=['year(date):O','sum(tot_victims)','count(incident_id)' ]
)

line=base.mark_line(color='#fc4f30').encode(alt.Y('count(incident_id):Q',axis=alt.Axis(title='count shootings', titleColor='#fc4f30')))


chart_2=alt.layer(bar,line).resolve_scale(y='independent').properties(
    title={
      "text": title, 
      "subtitle": subtitle
    }).add_selection(brush)

chart_2

world= gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
world.head(3)

world= gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))

brush= alt.selection_single(encodings=['y'],on='mouseover', empty='none')
color=alt.Color('pop_est', scale=alt.Scale(type='pow', exponent=0.4))

alt.hconcat(
    alt.Chart().mark_bar().encode(
        x=alt.X('pop_est', scale=alt.Scale(nice=False)),
        y=alt.Y('name', sort=alt.EncodingSortField(field='pop_est', op='sum', order='descending')),
        tooltip=['name', 'pop_est','gdp_md_est' ],
        color=alt.condition(brush, alt.value('lightgray'), color)
    ).add_selection(
        brush
    ).properties(
        width=200,
        height=450
),
    alt.Chart().mark_geoshape().project().encode(
        color=alt.condition(
        brush,
            alt.value('lightgray'),
            color,
        ),
        tooltip=['name', 'pop_est','gdp_md_est' ],
    ).properties(
        width=300,
        height=450,
        title='North America population'
    ),
    data=world[world.continent=='North America']
)

chart_2 & chart_3

#creating source data by joining all sources by state
import geopandas as gpd

column_names=['state_name', 'drawseq', 'state_fips', 'sub_region', 'state','geometry']        
gdf = gpd.read_file(os.path.join(os.getcwd(),'datasets','states_21basic','states.shp'))
gdf.columns=column_names

df_gun_violence=(df2_incidents
      .loc[:, ['state', 'tot_victims', 'longitude','latitude', 'incident_id']]
      .groupby(df2_incidents.state)
      .agg(
          tot_victims=pd.NamedAgg(column="tot_victims", aggfunc="sum"),
          num_incidents=pd.NamedAgg(column="incident_id", aggfunc="count"),
          latitude=pd.NamedAgg(column="latitude", aggfunc="mean"),
          longitude=pd.NamedAgg(column="longitude", aggfunc="mean"),
      )
      .reset_index()  
     )

latest_state_population
source=pd.merge(left=gdf,right=df_gun_violence, how='right', on='state') 
source=pd.merge(left=latest_state_population,right=source, how='right', on='state')

print('printing example:')
source[source.state=='CA']

#sum victims last decade
last_year=df2_incidents.date.dt.year.max()
state_victims_last_decade=(df2_incidents
      [df2_incidents.date.dt.year>=last_year-10]
      .groupby(df2_incidents.state)
      .agg(
          tot_victims=pd.NamedAgg(column="tot_victims", aggfunc="sum")
       )
      .sort_values(by='tot_victims', ascending=False)
)
state_victims_last_decade.head(3)

state='CA'
state_population=source[source.state==state]['latest_population']
state_victims=source[source.state==state]['tot_victims']

data=pd.DataFrame([dict(id=i) for i in range(1, num_people+1)])
person = (
    "M1.7 -1.7h-0.8c0.3 -0.2 0.6 -0.5 0.6 -0.9c0 -0.6 "
    "-0.4 -1 -1 -1c-0.6 0 -1 0.4 -1 1c0 0.4 0.2 0.7 0.6 "
    "0.9h-0.8c-0.4 0 -0.7 0.3 -0.7 0.6v1.9c0 0.3 0.3 0.6 "
    "0.6 0.6h0.2c0 0 0 0.1 0 0.1v1.9c0 0.3 0.2 0.6 0.3 "
    "0.6h1.3c0.2 0 0.3 -0.3 0.3 -0.6v-1.8c0 0 0 -0.1 0 "
    "-0.1h0.2c0.3 0 0.6 -0.3 0.6 -0.6v-2c0.2 -0.3 -0.1 "
    "-0.6 -0.4 -0.6z"
)

population= alt.Chart(data).transform_calculate(
        row="ceil(datum.id/10)"
    ).transform_calculate(
        col="datum.id - datum.row*10"
    ).mark_point(
        filled=False,
        size=40
    ).encode(
        x=alt.X("col:O", axis=None),
        y=alt.Y("row:O", axis=None),
        shape=alt.ShapeValue(person)
    ).properties(
        width=300,
        height=300
    ).configure_view(
        strokeWidth=0
)

population 

from collections import ChainMap
data=source
dict_sub_region=(
    data.loc[:,['state','sub_region']]
    .set_index('state')
    .set_axis(['sub_region'], axis = 1)#renaming columns
    .to_dict()
)
dict_sub_region
dict_sub_region = dict(ChainMap(*dict_sub_region.values()))#removes top level of dictionary
print(dict_sub_region)

chart_4=(chart_2 & chart_5 &chart_3)
chart_4.save('altair_charts/page_2.html', embed_options={'actions': True})
chart_4

title= ''
subtitle=''

source=df2_incidents


columns=sorted(source.date.dt.year.unique())


a = alt.Chart(source).mark_bar().encode(
    x=alt.X('month(date):O', axis=alt.Axis(title=None)),
    y=alt.Y('count(incident_id):Q',axis=alt.Axis(title='Num. of Shootings')),
    color='situation_type', 
    tooltip=['year(date)', 'count(incident_id)', 'situation_type'],
    order=alt.Order('situation_type', sort='descending')
).properties(width=800)

explore=a
explore

df1_details.info()
df1_details.head(3)
df1_details.columns.values

cols=['case', 'geo_location', 'date', 'summary', 'fatalities', 'injured',
       'total_victims', 'location', 'age',
       'prior_signs_mental_health_issues', 'mental_health_details',
       'weapons_obtained_legally', 'where_obtained', 'weapon_type',
       'weapon_details', 'race', 'gender', 'sources',
       'mental_health_sources', 'sources_additional_age', 'latitude',
       'longitude', 'type', 'year', 'month']

data=(df1_details
    .assign(year=df2_incidents.date.dt.year, month=df2_incidents.date.dt.month)
    .set_axis(cols, axis = 1)
    .pipe(catchstate, var_name="df2")
)
data.head(2)

# from vega_datasets import data
# source = data.population.url
source=(data
#         .dropna(subset=['gender'])
       )

color_scale = alt.Scale(domain=['M', 'F'],
                        range=['#1f77b4', '#e377c2'])
base = alt.Chart(source).transform_calculate(
    gender=alt.expr.if_(alt.datum.gender == "M", 'M', 'F')
).properties(
    width=250
)

left = base.transform_filter(
    alt.datum.gender == 'F'
).encode(
    y=alt.Y('binned_age:O', axis=None),
    x=alt.X('count(case):Q',
            title='age (binned)',
            sort=alt.SortOrder('descending')),
    color=alt.Color('gender:N', scale=color_scale, legend=None),
    tooltip=['count(case)']
).mark_bar().transform_bin(
    'binned_age', field='age'
).properties(title='Female')

middle = base.encode(
    y=alt.Y('binned_age:O', axis=None),
    text=alt.Text('binned_age:Q'),
).mark_text().transform_bin(
    'binned_age', field='age'
).properties(width=20)

right = base.transform_filter(
    alt.datum.gender == 'M'
).encode(
    y=alt.Y('binned_age:O', axis=None),
    x=alt.X('count(case):Q', title='age (binned)'),
    color=alt.Color('gender:N', scale=color_scale, legend=None),
    tooltip=['count(case)']
).mark_bar().transform_bin(
    'binned_age', field='age'
).properties(title='Male')

chart_2_1=alt.concat(left, middle, right, spacing=5)
chart_2_1

var='race'
title= "Race"
subtitle=''
dict_race={'white':'White', "-":'Unknown', 'unclear':'Unknown', 'black':'Black', 'White ':'White'}


# brush = alt.selection_single(nearest=True, on='mouseover')

source=(data
        .assign(race=data.race.replace(dict_race))
)

base = alt.Chart(source).properties(width=300).properties(
    title={
      "text": [title], 
      "subtitle": [subtitle]
    })

chart=base.mark_bar().encode(
    x=alt.X('count(case):Q', axis=alt.Axis(title='num. of shootings')),
    y=alt.Y(var,sort=alt.EncodingSortField(field='incident_id', order='descending', op='count'),axis=alt.Axis(title=var.replace('_', " "))),
    order=alt.Order('count(case)', sort='descending'),
    tooltip=['count(case)']
)

    
chart_2_2=chart
chart_2_2

var='prior_signs_mental_health_issues'
title= ['Prior Signs of', 'Mental Health']
subtitle=''
dict_values={'yes':'Yes', "-":'Unknown', 'unclear':'Unknown', 'Unclear':'Unknown','Unclear ':'Unknown','TBD':'Unknown', 'no':'No'}

# brush = alt.selection_single(nearest=True, on='mouseover')

source=(data
        .assign(prior_signs_mental_health_issues=data.prior_signs_mental_health_issues.replace(dict_values))
        .pipe(catchstate, var_name="df2")
        [(df2.prior_signs_mental_health_issues!='Unknown')]
        .groupby(by=['prior_signs_mental_health_issues'])
        .agg(
          num_incidents=pd.NamedAgg(column="case", aggfunc="count")
        ).reset_index()
)
source

base = alt.Chart(source).properties(width=200).properties(
    title={
      "text": title, 
      "subtitle": subtitle
    })

chart=base.mark_bar().encode(
    x=alt.X('num_incidents:Q', axis=alt.Axis(title='num. of shootings')),
    y=alt.Y(var,sort=alt.EncodingSortField(field='num_incidents', order='descending', op='sum'),axis=alt.Axis(title='')),
    order=alt.Order('num_incidents', sort='descending'),
    tooltip=['num_incidents',var]
)

# chart=base.mark_arc().encode(
#     theta=alt.Theta(field='num_incidents', type='quantitative'),
#     color=alt.Color(field=var, type="nominal", title=var.replace('_', " ")),
# )
    
chart_2_3=chart
chart_2_3

page_3=chart_2_1 & (chart_2_2 | chart_2_3)
page_3.save('altair_charts/page_3.html', embed_options={'actions': True})
page_3

## NPL to Desc:  Wordcloud of all data
# libraries for NLP
import nltk
nltk.download('omw-1.4')
from nltk import FreqDist
nltk.download('stopwords')
nltk.download('averaged_perceptron_tagger')
nltk.download('wordnet')
import string
from nltk import pos_tag
from nltk.corpus import stopwords
from nltk.tokenize import WhitespaceTokenizer
from nltk.stem import WordNetLemmatizer


# return the wordnet object value corresponding to the POS tag
from nltk.corpus import wordnet

def get_wordnet_pos(pos_tag):
    if pos_tag.startswith('J'):
        return wordnet.ADJ
    elif pos_tag.startswith('V'):
        return wordnet.VERB
    elif pos_tag.startswith('N'):
        return wordnet.NOUN
    elif pos_tag.startswith('R'):
        return wordnet.ADV
    else:
        return wordnet.NOUN
    
# clean text data

def clean_text(text):
    # lower text
    text = text.lower()
    # tokenize text and remove puncutation
    text = [word.strip(string.punctuation) for word in text.split(" ")]
    # remove words that contain numbers
    text = [word for word in text if not any(c.isdigit() for c in word)]
    # remove stop words
    stop = stopwords.words('english')
    text = [x for x in text if x not in stop]
    # remove empty tokens
    text = [t for t in text if len(t) > 0]
    # pos tag text
    pos_tags = pos_tag(text)
    # lemmatize text
    text = [WordNetLemmatizer().lemmatize(t[0], get_wordnet_pos(t[1])) for t in pos_tags]
    # remove words with only one letter
    text = [t for t in text if len(t) > 1]
    # join all
    text = " ".join(text)
    return(text)

#define original data
dirty = dirty

##clean data
content_clean=dirty.desc.apply(lambda x: clean_text(x))

##graphic wordcloud
from wordcloud import WordCloud
import matplotlib.pyplot as plt
 
def show_wordcloud(data, title = None):
    wordcloud = WordCloud(
        background_color = 'white',
        max_words = 200,
        max_font_size = 40, 
        scale = 3,
        random_state = 42
    ).generate(str(data))
 
    fig = plt.figure(1, figsize = (20, 20))
    plt.axis('off')
    if title: 
        fig.suptitle(title, fontsize = 20)
        fig.subplots_adjust(top = 2.3)
 
    plt.imshow(wordcloud)
    plt.show()

show_wordcloud(content_clean)

var='weapons_obtained_legally'
title= ''
subtitle=''
dict_values={'yes':'Yes', "-":'Unknown', 'unclear':'Unknown', 'Unclear':'Unknown','Unclear ':'Unknown','TBD':'Unknown', 'no':'No'}


source=(data
        .assign(weapons_obtained_legally=data[var].replace(dict_values))
        .pipe(catchstate, var_name="df2")
        [(df2[var]=='Yes')|(df2[var]=='No')]
        .groupby(by=[var])
        .agg(
          num_incidents=pd.NamedAgg(column="case", aggfunc="count")
        ).reset_index()
)

# print(source)

base = alt.Chart(source).properties(width=200).properties(
    title={
      "text": [title], 
      "subtitle": [subtitle]
    })

chart=base.mark_arc().encode(
    theta=alt.Theta(field='num_incidents', type='quantitative'),
    color=alt.Color(field=var, type="nominal", title=var.replace('_', " ")),
)
    
chart

var='weapon_type'
title= ''
subtitle=''
dict_race={'white':'White', "-":'Unknown', 'unclear':'Unknown', 'black':'Black', 'White ':'White'}


# brush = alt.selection_single(nearest=True, on='mouseover')

source=(data
        .assign(race=data.race.replace(dict_race))
)

base = alt.Chart(source).properties(width=800).properties(
    title={
      "text": [title], 
      "subtitle": [subtitle]
    })

chart=base.mark_bar().encode(
    x=alt.X('count(case):Q', axis=alt.Axis(title='num. of shootings')),
    y=alt.Y(var,sort=alt.EncodingSortField(field='incident_id', order='descending', op='count'),axis=alt.Axis(title=var.replace('_', " "))),
    order=alt.Order('count(case)', sort='descending'),
    tooltip=['count(case)']
)

    
chart_6=chart
chart_6

data.head(2)

var='weapons_obtained_legally'
title= ''
subtitle=''
dict_values={' Yes':'Yes','Yes ':'Yes','yes':'Yes', "-":'Unknown', 'unclear':'Unknown', 'Unclear':'Unknown','Unclear ':'Unknown','TBD':'Unknown', 'no':'No'}


# brush = alt.selection_single(nearest=True, on='mouseover')

source=(data
        .assign(weapons_obtained_legally=data[var].replace(dict_values))
        .pipe(catchstate, var_name="df2")
        [(df2[var]=='Yes')|(df2[var]=='No')]
        .groupby(by=[var])
        .agg(
          num_incidents=pd.NamedAgg(column="case", aggfunc="count")
        ).reset_index()
)
source
base = alt.Chart(source).properties(width=200).properties(
    title={
      "text": [title], 
      "subtitle": [subtitle]
    })

# # chart=base.mark_bar().encode(
# #     x=alt.X('count(case):Q', axis=alt.Axis(title='num. of shootings')),
# #     y=alt.Y(var,sort=alt.EncodingSortField(field='incident_id', order='descending', op='count'),axis=alt.Axis(title=var.replace('_', " "))),
# #     order=alt.Order('count(case)', sort='descending'),
# #     tooltip=['count(case)','weapons_obtained_legally']
# # )

chart=base.mark_arc().encode(
    theta=alt.Theta(field='count(case)', type='quantitative'),
    color=alt.Color(field=var, type="nominal", title=var.replace('_', " ")),
)
    
chart