import geopandas as gpd
import pandas as pd
import matplotlib.pyplot as plt

#! pip install mapclassify

def clean_df(df):
    df.columns = df.columns.str.lower()
    # add country column
    df['first_letter_code'] = df['geo_code'].str[0]
    diz_country = {'E':'England','S':'Scotland','W':'Wales','N':'Northen Ireland'}
    df['Country'] = df['first_letter_code'].map(lambda x: diz_country[x])
    df = df[['geo_code','Country','car or van availability : total\\ car or van availability - unit : households']]
    # rename column 
    df = df.rename(columns={'car or van availability : total\\ car or van availability - unit : households':'car_or_van_availability_households'})
    return df

df = pd.read_csv('data/CARVAN_OASA_England_Northern_Ireland_Scotland_Wales_Descriptions.csv')
df = clean_df(df)
df.sample(3)