gistfile1.py

import collections
import itertools
import hashlib
import json
import requests
import mixpanel  # https://pypi.python.org/pypi/mixpanel-py/4.0.2

def get_mixpanel_datapoints():
    url = 'https://data.mixpanel.com/api/2.0/export/'
    params = {
        'from_date': '2015-02-20',
        'to_date': '2015-04-19',
        # ...
    }
    params_http = itertools.starmap("{0}={1}".format, params.items())
    params_str = ''.join(sorted(fmt_params)) + '<secret>'
    params['sig'] = hashlib.md5(sig).hexdigest()
    return map(json.loads, requests.get(URL, params=PARAMS).content.split('\n'))

def main():
    mp = mixpanel.Mixpanel('<token>')
    users = get_users_from_database()
    datapoints = get_mixpanel_datapoints()
    existing_ids = collections.defaultdict(set)
    for datapoint in datapoints:
        existing_ids[datapoint['user']].add(datapoint['distinct_id'])
    for user in users:
        # link profile to previous distinct ids
        # https://mixpanel.com/help/reference/python#alias
        for original_id in existing_ids[user['email']]:
            mp.alias(user['email'], original_id)
    for user in users:
        # link profile to user email
        mp.people_set(user['email'], {'$email': user['email']})


if __name__ == '__main__':
    main()