CareerCup

public static int calculate_hash(String input) {
    int h = 0;
    int len = input.length();
    for (int i = 0; i < len; i++) {
        h = 31 * h + input.charAt(i);
    }
    return h;
}

The polynomial function will work. I guess other variation could be using unique prime numbers to each of the alphabets. Hash would be multiplication of each character in the string.

Actually, what u guys are talk about are the same algo , they are both polynominal hash function , the code use Horner's rule , such as: h = k1 + 31*k2 + 31^2*k3 is computed by ((k3)*31 + k2)*31+k1

its xn(h^n)
sorry...

index hash(const char *key, int tablesize)
{

index hash(const char *key, int tablesize)
{ int hashval =0;
while(*key != '\0') {
hashval += (hashval<<5)+ *key++;
}
return hashval%tabesize ;}

int HashFunc1 (const char* s, int size)
{
  unsigned int h = (unsigned int)strlen(s);
  int step = (int)((h>>5)+1);
  
  for (int i=h; i>=step; i-=step)
    h = h ^ ((h<<5)+(h>>2)+(unsigned char)(s[i-1]));
  
  return h%(size-1);
}

above answer might not be good because it will cause collisions. consider using strings
1. apple
2. aplpe

will create the same result

what they were probably looking for was the polynomial hashing algorithm
its like
x0+x1(h^1)+x2(h^2)+x3(h^3)+ .... + xn(h^3)

x_n is a character in a string

hash function outputs dont need to be positive numbers between 1 to M, there's a compressor function that is used after the hashing algorithm to make a hashtable

i used a book to answer this question (just for fyi! :P)