Presidential Election 2022

from pynsee import get_geodata, get_geodata_list
import requests
import tempfile
import pandas as pd

import matplotlib.cm as cm
from matplotlib.colors import ListedColormap
import matplotlib.pyplot as plt
import descartes
import geopandas as gpd
import seaborn as sns

import logging
import sys
logging.basicConfig(stream=sys.stdout,
                    level=logging.INFO,
                    format="%(message)s")

# download data from data.gouv.fr
url = 'https://www.data.gouv.fr/fr/datasets/r/1ffb6125-1cea-4a03-80be-520c1c3a5411'

def download_data(url=url):

    results = requests.get(url)
    dirpath = tempfile.mkdtemp()
    dwnFile = dirpath + '\\fr_presidential_election_1stround_2022.txt'

    with open(dwnFile, 'wb') as f:
        f.write(results.content)
        f.close()

    return dwnFile

dwnFile = download_data()

def make_dataframes(dataFile=dwnFile):
    """
    function to make clean dataframes from data.gouv.fr file
        it returns one dataframe containing data about votes by candidate and commune,
        and a dataframe containing communes agregated data
    """

    # import data
    data = pd.read_csv(dataFile, dtype='str', sep=';', encoding='latin-1', skiprows=1, header=None)
    colNames = pd.read_csv(dataFile, dtype='str', sep=';', encoding='latin-1', nrows=1, header=None)
    colNamesCommunes = colNames.iloc[0,:18].to_list()
    colNamesBallots = colNames.iloc[0,19:len(colNames.columns)].to_list()

    # make dataframe only on communes agregated data
    dfCommunes = data.iloc[:,:18]
    dfCommunes.columns = colNamesCommunes

    # make dataframe only with detailed data, candidate by candidate
    icols = [x for x in range(4)] + [x for x in range(19, len(data.columns))]
    dfBallots = data.iloc[:, icols]
    dfBallots.columns = [x for x in range(len(dfBallots.columns))]
    colnamesCandidates = colNamesCommunes[:4] + colNamesBallots

    # reshape detailed data
    list_df = []
    for icandidate in range(1, 12):
        icols = [x for x in range(4)] + [x for x in range(7 * (icandidate - 1) + 4, 7 * icandidate + 4)]
        dfBallotsCandidate = dfBallots.iloc[:, icols]
        dfBallotsCandidate.columns = colnamesCandidates
        list_df += [dfBallotsCandidate]

    dfFinal = pd.concat(list_df).reset_index(drop=True)

    dfFinal['Voix'] = pd.to_numeric(dfFinal['Voix'])

    dfFinal = dfFinal.rename(columns={"Code du département":"dep", "Code de la commune":"com"})
    dfCommunes = dfCommunes.rename(columns={"Code du département":"dep", "Code de la commune":"com"})

    # add clean communes id
    for r in dfCommunes.index:
        dep = str(dfCommunes.loc[r, "dep"])
        comId = str(dfCommunes.loc[r, "com"])
        if dep.startswith('Z'):
            dep = '97'
        dfCommunes.loc[r,"insee_com"] = dep + comId

    dfFinal = dfFinal.merge(dfCommunes, on = ["dep", "com"], how="left")

    return dfFinal, dfCommunes

dfBallots, dfCommunes = make_dataframes()

# select only the candidate who has won the most votes for each commune
colList = dfBallots.columns[:4].to_list() + ["insee_com"]
FirstCandidate = dfBallots.loc[dfBallots.reset_index().groupby(colList)['Voix'].idxmax()].reset_index(drop=True)

# download geographical limits of communes from IGN, and merge it with
IGNCom = get_geodata('ADMINEXPRESS-COG-CARTO.LATEST:commune').translate().zoom()
com = IGNCom.merge(FirstCandidate, how='left', on ='insee_com')
com = com.dropna()
comMap = gpd.GeoDataFrame(com)

Previously saved data has been used:
/home/onyxia/.cache/pynsee/pynsee/a4f589755961ab3aa6bdff99605334d1
Set update=True to get the most up-to-date data

# make color palette
set1 = sns.color_palette("Set1")
colors = [sns.color_palette("Paired")[4]] + [set1[2], set1[1]] + set1[3:5] + [set1[0]] + set1[6:len(set1)]
cmap = ListedColormap(colors)

fig, ax = plt.subplots(1,1,figsize=[15,15])
comMap.plot(column='Nom', legend=True, ax=ax, cmap=cmap)
plt.suptitle('The results of the first round of the 2022 presidential election by commune', x = 0.3, y = 0.85,
            horizontalalignment= 'left',
            fontsize = 12, fontweight='bold')
plt.show()