I have a Python list of the prime factors of a number. How do I (pythonically) find all the factors?

Question

I'm working on a Project Euler problem which requires factorization of an integer. I can come up with a list of all of the primes that are the factor of a given number. The Fundamental Theorem of Arithmetic implies that I can use this list to derive every factor of the number.

My current plan is to take each number in the list of base primes and raise its power until it is no longer an integer factor to find the maximum exponents for each prime. Then, I will multiply every possible combination of prime-exponent pairs.

So for example, for 180:

Given: prime factors of 180: [2, 3, 5]
Find maximum exponent of each  factor: 
    180 / 2^1 = 90
    180 / 2^2 = 45
    180 / 2^3 = 22.5 - not an integer, so 2 is the maximum exponent of 2.

    180 / 3^1 = 60
    180 / 3^2 = 20
    180 / 3^3 = 6.6 - not an integer, so 2 is the maximum exponent of 3.

    180 / 5^1 = 36
    180 / 5^2 = 7.2 - not an integer, so 1 is the maximum exponent of 5.

Next, do every combination of these up to the maximum exponent to get the factors:

    2^0 * 3^0 * 5^0 = 1
    2^1 * 3^0 * 5^0 = 2
    2^2 * 3^0 * 5^0 = 4
    2^0 * 3^1 * 5^0 = 3
    2^1 * 3^1 * 5^0 = 6
    2^2 * 3^1 * 5^0 = 12
    2^0 * 3^2 * 5^0 = 9
    2^1 * 3^2 * 5^0 = 18
    2^2 * 3^2 * 5^0 = 36
    2^0 * 3^0 * 5^1 = 5
    2^1 * 3^0 * 5^1 = 10
    2^2 * 3^0 * 5^1 = 20
    2^0 * 3^1 * 5^1 = 15
    2^1 * 3^1 * 5^1 = 30
    2^2 * 3^1 * 5^1 = 60
    2^0 * 3^2 * 5^1 = 45
    2^1 * 3^2 * 5^1 = 90
    2^2 * 3^2 * 5^1 = 180

So the list of factors = [1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 30, 36, 45, 60, 90, 180]

Here is the code I have so far. Two problems: First, I don't think it is very Pythonic at all. I'd like to fix that. Second, I really don't have a Pythonic way to do the combinatoric second step. Out of shame, I've spared you from the ridiculous set of loops.

n is the number we want to factor. listOfAllPrimes is a precalculated list of the primes up to 10 million.

def getListOfFactors(n, listOfAllPrimes):
    maxFactor = int(math.sqrt(n)) + 1
    eligiblePrimes = filter(lambda x: x <= maxFactor, listOfAllPrimes)
    listOfBasePrimes = filter(lambda x: n % x ==0, eligiblePrimes)

    listOfExponents = [] #(do I have to do this?)
    for x in listOfBasePrimes:
        y = 1
        while (x**(y+1)) % n == 0:
            y += 1
        listOfExponents.append(y)

Answer 1

Instead of a list of exponents, consider simply repeating each prime factor by the number of times it is a factor. After that, working on the resulting primefactors list-with-repetitions, itertools.combinations does just what you need -- you'll just require the combinations of length 2 to len(primefactors) - 1 items included (the combinations of just one are the prime factors, that of all of them will be the original number -- if you want those too, just use range(1, len(primefactors) + 1) instead of the range(2, len(primefactors)) which my main suggestion would use).

There will be repetitions in the results (e.g., 6 will appear twice as a factor of 12, since the latter's primefactors will be [2, 2, 3]) and they can of course be weeded out in the usual ways (i.e. sorted(set(results)) for example).

To compute primefactors given listOfAllPrimes, consider for example:

def getprimefactors(n):
    primefactors = []
    primeind = 0
    p = listOfAllPrimes[primeind]
    while p <= n:
        if n % p == 0:
            primefactors.append(p)
            n //= p
        else:
            primeind += 1
            p = listOfAllPrimes[primeind]
    return primefactors

Answer 2

You could use itertools.combinations() to get all possible combinations of factors once you've gotten your list of repeated-primes, such as [2, 2, 3, 3, 5] for 180. Then, simply multiplying the components from each combination will get you a factor.

Answer 3

Here's a simple and efficient solution to the original problem:

def getDivisors(n):
    divisors = []
    d = 1
    while d*d < n:
        if n % d == 0:
            divisors.append(d)
            divisors.append(n / d);
        d += 1
    if d*d == n:
        divisors.append(d)
    return divisors

The output list is unsorted. You can make it more "pythonic" if you want, whatever that means.

Answer 4

Why do you begin your solution from the set of prime factors? when you factorize a number you can as easily get all its prime factors (repeated) and from them the exponents for each factor. With this in mind, you can write this:

import itertools
prime_factors = get_prime_factors(180) 
# 2, 2, 3, 3, 5 
factorization = [(f, len(list(fs))) for (f, fs) in itertools.groupby(prime_factors)]
# [(2, 2), (3, 2), (5, 1)]
values = [[(factor**e) for e in range(exp+1)] for (factor, exp) in factorization]
# [[1, 2, 4], [1, 3, 9], [1, 5]]
print sorted(product(xs) for xs in itertools.product(*values))
# [1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 30, 36, 45, 60, 90, 180]

get_prime_factors and product are not implemented here, but you get the idea (both are pretty simple to write)

IMHO, being mathematic problems, the Euler problems can be nicely solved using functional instead of imperative style (though I acknowledge that some solutions may not come out as pythonic as desired).

Answer 5

With a few cooler Python features:

def factors( num ):
    for p in primes:
        if num==1: break # stop when num is 1
        while True: # these factors may be repeated 
            new, rest = divmod(num, p) # does div and mod at once
            if rest==0: # its divisible
                yield p # current prime is factor
                num = new # continue with the div'd number
            else:
                break # no factor so break from the while


print list(factors(2*2*3*3*5*7*11*11*13)) # [2, 2, 3, 3, 5, 7, 11, 11, 13] ofc

Answer 6

An all in one solution; i.e. no need for an existing list of the prime factors.

#!/usr/bin/python3 -O

from primegen import erat3 as generate_primes # see Note[1]
import operator as op, functools as ft, itertools as it

def all_factors(number):
    prime_powers= []

    for prime in generate_primes(): # for prime in listOfAllPrimes
        if prime > number: break

        this_prime_powers= [1]
        new_number, modulo= divmod(number, prime)

        while not modulo:
            number= new_number
            this_prime_powers.append(this_prime_powers[-1] * prime)
            new_number, modulo= divmod(number, prime)

        if len(this_prime_powers) > 1:
            prime_powers.append(this_prime_powers)

    # at this point:
    # if number was 360, prime_powers is [[1, 2, 4, 8], [1, 3, 9], [1, 5]]
    # if number was 210, prime_powers is [[1, 2], [1, 3], [1, 5], [1, 7]]

    return sorted(
        ft.reduce(op.mul, combination, 1)
        for combination in it.product(*prime_powers))

if __name__ == "__main__":
    def num_result(number):
        return number, all_factors(number)
    print(num_result(360))
    print(num_result(210))
    print(num_result(7))

Note[1]: As a prime number generator, you can choose one from How to implement an efficient infinite generator of prime numbers in Python? or use your own (e.g. your listOfAllPrimes).

This produces a full factor list, i.e. including 1 and the number argument itself. If you want to omit these, you can use all_factors(number)[1:-1].

$ allfactors.py 
(360, [1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 15, 18, 20, 24, 30, 36, 40, 45, 60, 72, 90, 120, 180, 360])
(210, [1, 2, 3, 5, 6, 7, 10, 14, 15, 21, 30, 35, 42, 70, 105, 210])
(7, [1, 7])