各种hash函数

public class HashAlgorithms
{
/**
* 加法hash
* @param key 字符串
* @param prime 一个质数
* @return hash结果
*/
public static int additiveHash(String key, int prime)
{
int hash, i;
for (hash = key.length(), i = ; i < key.length(); i++)
hash += key.charAt(i);
return (hash % prime);
} 

/**
* 旋转hash
* @param key 输入字符串
* @param prime 质数
* @return hash值
*/
public static int rotatingHash(String key, int prime)
{
int hash, i;
for (hash=key.length(), i=; i<key.length(); ++i)
hash = (hash<<)^(hash>>)^key.charAt(i);
return (hash % prime);
// return (hash ^ (hash>>10) ^ (hash>>20));
} 

// 替代：
// 使用：hash = (hash ^ (hash>>10) ^ (hash>>20)) & mask;
// 替代：hash %= prime; 

/**
* MASK值，随便找一个值，最好是质数
*/
static int M_MASK = 0x8765fed1;
/**
* 一次一个hash
* @param key 输入字符串
* @return 输出hash值
*/
public static int oneByOneHash(String key)
{
int hash, i;
for (hash=, i=; i<key.length(); ++i)
{
hash += key.charAt(i);
hash += (hash << );
hash ^= (hash >> );
}
hash += (hash << );
hash ^= (hash >> );
hash += (hash << );
// return (hash & M_MASK);
return hash;
} 

/**
* Bernstein's hash
* @param key 输入字节数组
* @param level 初始hash常量
* @return 结果hash
*/
public static int bernstein(String key)
{
int hash = ;
int i;
for (i=; i<key.length(); ++i) hash = *hash + key.charAt(i);
return hash;
} 

//
//// Pearson's Hash
// char pearson(char[]key, ub4 len, char tab[256])
// {
// char hash;
// ub4 i;
// for (hash=len, i=0; i<len; ++i)
// hash=tab[hash^key[i]];
// return (hash);
// } 

//// CRC Hashing，计算crc,具体代码见其他
// ub4 crc(char *key, ub4 len, ub4 mask, ub4 tab[256])
// {
// ub4 hash, i;
// for (hash=len, i=0; i<len; ++i)
// hash = (hash >> 8) ^ tab[(hash & 0xff) ^ key[i]];
// return (hash & mask);
// } 

/**
* Universal Hashing
*/
public static int universal(char[]key, int mask, int[] tab)
{
int hash = key.length, i, len = key.length;
for (i=; i<(len<<); i+=)
{
char k = key[i>>];
if ((k&0x01) == ) hash ^= tab[i+];
if ((k&0x02) == ) hash ^= tab[i+];
if ((k&0x04) == ) hash ^= tab[i+];
if ((k&0x08) == ) hash ^= tab[i+];
if ((k&0x10) == ) hash ^= tab[i+];
if ((k&0x20) == ) hash ^= tab[i+];
if ((k&0x40) == ) hash ^= tab[i+];
if ((k&0x80) == ) hash ^= tab[i+];
}
return (hash & mask);
} 

/**
* Zobrist Hashing
*/
public static int zobrist( char[] key,int mask, int[][] tab)
{
int hash, i;
for (hash=key.length, i=; i<key.length; ++i)
hash ^= tab[i][key[i]];
return (hash & mask);
} 

// LOOKUP3
// 见Bob Jenkins(3).c文件 

// 32位FNV算法
static int M_SHIFT = ;
/**
* 32位的FNV算法
* @param data 数组
* @return int值
*/
public static int FNVHash(byte[] data)
{
int hash = (int)2166136261L;
for(byte b : data)
hash = (hash * ) ^ b;
if (M_SHIFT == )
return hash;
return (hash ^ (hash >> M_SHIFT)) & M_MASK;
}
/**
* 改进的32位FNV算法1
* @param data 数组
* @return int值
*/
public static int FNVHash1(byte[] data)
{
final int p = ;
int hash = (int)2166136261L;
for(byte b:data)
hash = (hash ^ b) * p;
hash += hash << ;
hash ^= hash >> ;
hash += hash << ;
hash ^= hash >> ;
hash += hash << ;
return hash;
}
/**
* 改进的32位FNV算法1
* @param data 字符串
* @return int值
*/
public static int FNVHash1(String data)
{
final int p = ;
int hash = (int)2166136261L;
for(int i=;i<data.length();i++)
hash = (hash ^ data.charAt(i)) * p;
hash += hash << ;
hash ^= hash >> ;
hash += hash << ;
hash ^= hash >> ;
hash += hash << ;
return hash;
} 

/**
* Thomas Wang的算法，整数hash
*/
public static int intHash(int key)
{
key += ~(key << );
key ^= (key >>> );
key += (key << );
key ^= (key >>> );
key += ~(key << );
key ^= (key >>> );
return key;
}
/**
* RS算法hash
* @param str 字符串
*/
public static int RSHash(String str)
{
int b = ;
int a = ;
int hash = ; 

for(int i = ; i < str.length(); i++)
{
hash = hash * a + str.charAt(i);
a = a * b;
} 

return (hash & 0x7FFFFFFF);
}
/* End Of RS Hash Function */ 

/**
* JS算法
*/
public static int JSHash(String str)
{
int hash = ; 

for(int i = ; i < str.length(); i++)
{
hash ^= ((hash << ) + str.charAt(i) + (hash >> ));
} 

return (hash & 0x7FFFFFFF);
}
/* End Of JS Hash Function */ 

/**
* PJW算法
*/
public static int PJWHash(String str)
{
int BitsInUnsignedInt = ;
int ThreeQuarters = (BitsInUnsignedInt * ) / ;
int OneEighth = BitsInUnsignedInt / ;
int HighBits = 0xFFFFFFFF << (BitsInUnsignedInt - OneEighth);
int hash = ;
int test = ; 

for(int i = ; i < str.length();i++)
{
hash = (hash << OneEighth) + str.charAt(i); 

if((test = hash & HighBits) != )
{
hash = (( hash ^ (test >> ThreeQuarters)) & (~HighBits));
}
} 

return (hash & 0x7FFFFFFF);
}
/* End Of P. J. Weinberger Hash Function */ 

/**
* ELF算法
*/
public static int ELFHash(String str)
{
int hash = ;
int x = ; 

for(int i = ; i < str.length(); i++)
{
hash = (hash << ) + str.charAt(i);
if((x = (int)(hash & 0xF0000000L)) != )
{
hash ^= (x >> );
hash &= ~x;
}
} 

return (hash & 0x7FFFFFFF);
}
/* End Of ELF Hash Function */ 

/**
* BKDR算法
*/
public static int BKDRHash(String str)
{
int seed = ; // 31 131 1313 13131 131313 etc..
int hash = ; 

for(int i = ; i < str.length(); i++)
{
hash = (hash * seed) + str.charAt(i);
} 

return (hash & 0x7FFFFFFF);
}
/* End Of BKDR Hash Function */ 

/**
* SDBM算法
*/
public static int SDBMHash(String str)
{
int hash = ; 

for(int i = ; i < str.length(); i++)
{
hash = str.charAt(i) + (hash << ) + (hash << ) - hash;
} 

return (hash & 0x7FFFFFFF);
}
/* End Of SDBM Hash Function */ 

/**
* DJB算法
*/
public static int DJBHash(String str)
{
int hash = ; 

for(int i = ; i < str.length(); i++)
{
hash = ((hash << ) + hash) + str.charAt(i);
} 

return (hash & 0x7FFFFFFF);
}
/* End Of DJB Hash Function */ 

/**
* DEK算法
*/
public static int DEKHash(String str)
{
int hash = str.length(); 

for(int i = ; i < str.length(); i++)
{
hash = ((hash << ) ^ (hash >> )) ^ str.charAt(i);
} 

return (hash & 0x7FFFFFFF);
} 

各种hash函数的实现

其中比较有名的是elfhash函数。它对字串数组的算hash方式为

将一个字符串的数组中的每个元素依次按前四位与上一个元素的低四位相与，组成一个长整形，如果长整的高四位大于零，那么就将它折回再与长整的低四位相异或，这样最后得到的长整对HASH表长取余，得到在HASH中的位置。

/**
* ELF算法
*/
public static int ELFHash(String str)
{
    int hash = ;
    int x = ; 

    for(int i = ; i < str.length(); i++)
    {
        hash = (hash << ) + str.charAt(i);
        if((x = (int)(hash & 0xF0000000L)) != )
        {
            hash ^= (x >> );
            hash &= ~x;
        }
    }
｝ 

下面这个是所有最靠近2的幂次方的质数表，这些质数可以用来做hash表的扩增

glib质数表

static const gint prime_mod [] =
{
, /* For 1 << 0 */
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
, /* For 1 << 16 */
,
,
,
,
,
,
,
,
,
,
,
,
,
,
 /* For 1 << 31 */
};