A sample PyPy vs CPython wager script

5m.py

from __future__ import print_function

import cProfile
import random
import string

from datetime import datetime

def find_duplicates(seq):
    """
    Given a sequence of integers, returns a list of all the ones that
    occur more than once.
    
    >>> find_duplicates([1, 1, 2, 3, 3])
    [1, 3]

    """
    
    uniques = set()
    duplicates = [None] * 13000

    i = 0
    for item in seq:
        if item in uniques:
            duplicates[i] = item
            i += 1
        uniques.add(item)

    return duplicates[:i]


if __name__ == '__main__':
    print("Loading 5M msisdns... ")
    msisdns = [int(line) for line in open('5m.csv')]

    print("Finding duplicates...")

    t1 = datetime.now()
    duplicates = find_duplicates(msisdns)
    t2 = datetime.now()

    print("{0} duplicates found.".format(len(duplicates)))
    print("{0} seconds elapsed.".format(t2 - t1))