normalize.py

import csv
from collections import defaultdict

__author__ = 'davedash'


def main():

    # Store the most recent/complete data here.
    companies = defaultdict(lambda: defaultdict(str))

    with open('results-download.csv', 'rb') as csvfile:
        for row in csv.reader(csvfile):
            if row[0] == 'survey_id':
                continue

            company = row[2]
            mom = row[28]
            dad = row[31]
            adopt = row[34]

            current = companies[company]
            if mom:
                current['mom'] = mom
            if dad:
                current['dad'] = dad
            if adopt:
                current['adopt'] = adopt

    for company, data in companies.iteritems():
        if not data:
            continue
        print ', '.join([company, data['mom'], data['dad'], data['adopt']])


if __name__ == '__main__':
    main()