scrape april 7 2015 chicago mayoral runoff election results

scrape.py

from __future__ import print_function

from collections import namedtuple
import csv
import sys
import time
from urllib import urlopen

from bs4 import BeautifulSoup

Precinct = namedtuple('Precinct', ['ward', 'precinct', 'candidate', 'votes'])

def votes_by_precinct(ward):
    candidates = [
        'Rahm Emanuel',
        'Jesus "Chuy" Garcia',
    ]

    # Column index that corresponds to the candidate's vote total in precinct.
    cols = [
        2,
        4,
    ]

    f = urlopen('http://www.chicagoelections.com/en/pctlevel3.asp?Ward={}&elec_code=9&race_number=10'.format(ward))
    soup = BeautifulSoup(f)
    f.close()

    rows = soup.body.table.find_all('tr')
    i = 3
    row = rows[i]
    while row.td.string != 'Total':
        precinct = int(row.td.string)
        cells = row.find_all('td')
        for j, col in enumerate(cols):
            votes = int(list(cells[col].stripped_strings)[0])
            yield Precinct(ward, precinct, candidates[j], votes)
        i += 1
        row = rows[i]

def main():
    results = []

    for ward in range(1, 50+1):
        print('Ward {} ...'.format(ward), end='', file=sys.stderr)

        for res in votes_by_precinct(ward):
            results.append(res)

        print('done.', file=sys.stderr)

        # Don't hammer them.
        time.sleep(2)

    w = csv.writer(sys.stdout)
    w.writerow(Precinct._fields)
    w.writerows(results)

if __name__ == '__main__':
    main()

xresults.csv

Run it like:

python scrape.py > results.csv