Soylent Customer Ranking Script

rank_users.py

# The MIT License (MIT)

# Copyright (c) 2015 John Coogan

# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:

# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.

# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.

import argparse
import csv
import time
from collections import defaultdict


def dollars_to_seconds(dollars):
    decayed_days = dollars ** 0.75
    seconds_per_day = 24 * 60 * 60
    active_seconds = int(seconds_per_day * decayed_days)
    return active_seconds


def get_active_users(all_transactions):
    trans_by_time = sorted(all_transactions, key=lambda x: x[1])
    result_dict = {}
    for email, timestamp, amount in trans_by_time:
        timestamp = int(timestamp)
        amount = float(amount)
        current_value = result_dict.get(email)
        additional_seconds = dollars_to_seconds(amount)
        if current_value and current_value > timestamp:
            # user was active at the time of this purchase.
            new_value = current_value + additional_seconds
        else:
            # user is inactive and needs to either activate or reactivate.
            new_value = timestamp + additional_seconds

        result_dict[email] = new_value

    sorted_results = sorted(list(result_dict.items()), key=lambda x: x[1], 
                            reverse=True)
    current_timestamp = int(time.time())
    active_users = [(email, score) for email, score in sorted_results
                        if score > current_timestamp]
    print("Total users: {}".format(len(sorted_results)))
    print("Total actives: {}".format(len(active_users)))

    return sorted_results


def get_score(timestamp, amount, gravity=1.8):
    timestamp = int(timestamp)
    amount = float(amount)
    seconds_ago = time.time() - timestamp
    item_hour_age = seconds_ago / 3600
    score = (amount - 1) / ((item_hour_age + 2) ** gravity)
    return score

def get_agg_score(trans):
    return sum(get_score(timestamp, amount) 
                for email, timestamp, amount in trans)

def get_user_scores(all_transactions):
    user_transctions = defaultdict(list)
    for trans in all_transactions:
        user_transctions[trans[0]].append(trans)
    user_scores = [(e, get_agg_score(t)) for e, t in user_transctions.items()]
    sorted_results = sorted(user_scores, key=lambda x: x[1], reverse=True)
    return sorted_results


if __name__ == '__main__':
    parser = argparse.ArgumentParser(description='Calculate active users.')
    parser.add_argument('--infile', type=argparse.FileType('r'), 
                        help='filename of csv with user data', required=True)
    parser.add_argument('--outfile', type=argparse.FileType('w'),
                        help='filename of output file', required=True)
    parser.add_argument('--method', type=str, help='pick either hn or soylent',
                        default='soylent')
    args = parser.parse_args()
    in_csv = csv.reader(args.infile)
    all_transactions = [x for x in in_csv]

    if args.method == 'hn':
        sorted_results = get_user_scores(all_transactions)
    else:
        sorted_results = get_active_users(all_transactions)
    out_csv = csv.writer(args.outfile)
    out_csv.writerows(sorted_results)