Problem of the Day
A new programming or logic puzzle every Mon-Fri

Sieve Of Eratosthenes

Welcome back to the last Monday in March!

Today's problem will be to reimplement a classic algorithm for finding small primes. The Sieve of Eratosthenes has been around for centuries. Follow the link to the Wikipedia page to learn the idea behind the algorithm then implement it in a language of your choice.

Permalink: http://problemotd.com/problem/sieve-of-eratosthenes/

Comments:

  • David - 9 years ago

    Python with a touch of feature creep.

    from math import sqrt
    import time
    
    debug = 0
    
    start_time = time.time()
    
    
    def list_primes(largest_int):
        if(largest_int < 2):
            return []
        if(largest_int == 2):
            return [2]
        if(largest_int < 5):
            return [2, 3]
        if(largest_int > 50000):
            print "Sorry, %d is just too big to run through this algorithm" % largest_int
            return []
        if(largest_int > 2500):
            factors = list_primes(int(sqrt(largest_int))+1)
        else:
            factors = range(2, int(sqrt(largest_int))+1)
        nums = range(2, largest_int)
        if(debug):
            print largest_int
        for i in factors:
        # for i in range(2, 7):
            nums = filter(lambda x: x==i or x%i, nums)
        return nums
    
    
    
    print -1, list_primes(-1)
    print 0, list_primes(0)
    print 1, list_primes(1)
    print 2, list_primes(2)
    print 3, list_primes(3)
    print 4, list_primes(4)
    print 5, list_primes(5)
    print 6, list_primes(6)
    print 50, list_primes(50)
    print 50000, list_primes(50000)
    
    print int(sqrt(5))
    
    
    
    
    
    print "ExecutionTime %f seconds" % (time.time() - start_time)
    

    reply permalink

  • Jack le - 9 years ago

    This was actually an assignment in my system level programming class. The program is actually to solve project euler 10.

    #include <stdio.h>

    int main(){ // This program uses the sieve of Eratosthenes to solve the problem. int numbers[2000000] = {}; //Create an array of 2000000 integers of value 0 long int sum = 0; // Initialize sum to 0 // Create i and j for the nested for loops, they have to be long int because j*j will exceed the limit of int at some point long int i, j; for(j = 2; j <= sizeof(numbers)/4; j++){ // j goes from 2 to n if(!numbers[j-1]){ // If the number is not marked sum += j; // That is a prime number and add it to sum for(i = j*j; i <= sizeof(numbers)/4; i+=j) // Mark all the numbers from j2 to the end by j interval numbers[i-1] = 1; } } printf("sum: %ld\n",sum); // Display the sum return 0; }

    reply permalink

  • Werner D - 9 years ago

    C# solution:

            public static List<int> Sieve(int largestInt)
            {
                List<int> primes = new List<int>();
                List<int> matrix = new List<int>();
                for (int i = 2; i <= largestInt; i++)
                    matrix.Add(i);
    
                while (matrix.Count > 0)
                {
                    int p = matrix[0];
                    primes.Add(p);
    
                    for (int x = p; x <= matrix[matrix.Count - 1]; x++)
                    {
                        if (x % p == 0)
                            matrix.Remove(x);
    
                        if (matrix.Count == 0)
                            break;
                    }
                }
    
                return primes;
            }
    

    reply permalink

  • Anonymous - 9 years ago

    I've run this with up to 10 million which takes about a minute. Haven't tried 100 million.

    #python 2.7
    n = int(raw_input("Enter a number to find primes> "))
    
    sieve_array = []
    for i in range(n+1):
        sieve_array.append(i)
    
    sieve_array[1] = 0
    prime_array = []
    
    c = 2
    while c < n:
        m = 2
    
        while c*m <= n:
            sieve_array[c*m] = 0
            m = m + 1
        c = c + 1
    
    for item in sieve_array:
        if item != 0:
            prime_array.append(item)
    
    print prime_array
    print "There are %d primes between 1 and %d" % (len(prime_array), n)
    

    reply permalink

Content curated by @MaxBurstein