记录几个经典的字符串hash算法

1 #include <stdio.h>
2 #include <string.h>
3 #include <stdlib.h>
4 #include <stdint.h>
5 #include <stdbool.h>
6
7 #define MAXTABLELEN 102400000
8
9 typedef struct  _HASHTABLE
10 {
11     long nHashA;
12     long nHashB;
13     bool bExists;
14 }HASHTABLE, *PHASHTABLE;
15
16 const unsigned long nTableLength = MAXTABLELEN;
17 unsigned long m_tablelength;    // 哈希索引表长度
18 HASHTABLE *m_HashIndexTable;
19 unsigned long cryptTable[0x500];
20
21 int collc = 0;
22 int errstr = 0;
23
24 void InitCryptTable()
25 {
26     unsigned long seed = 0x00100001, index1 = 0, index2 = 0, i;
27
28     for( index1 = 0; index1 < 0x100; index1++ )
29     {
30         for( index2 = index1, i = 0; i < 5; i++, index2 += 0x100 )
31         {
32             unsigned long temp1, temp2;
33             seed = (seed * 125 + 3) % 0x2AAAAB;
34             temp1 = (seed & 0xFFFF) << 0x10;
35             seed = (seed * 125 + 3) % 0x2AAAAB;
36             temp2 = (seed & 0xFFFF);
37             cryptTable[index2] = ( temp1 | temp2 );
38         }
39     }
40 }
41
42 /************************************************************************/
43 /*函数名：HashString
44  *功  能：求取哈希值
45  *返回值：返回hash值
46  ************************************************************************/
47 unsigned long HashString(char *lpszString, unsigned long dwHashType)
48 {
49     unsigned char *key = (unsigned char *)lpszString;
50     unsigned long seed1 = 0x7FED7FED, seed2 = 0xEEEEEEEE;
51     int ch;
52
53     while(*key != 0)
54     {
55         ch = toupper(*key++);
56
57         seed1 = cryptTable[(dwHashType << 8) + ch] ^ (seed1 + seed2);
58         seed2 = ch + seed1 + seed2 + (seed2 << 5) + 3;
59     }
60     return seed1;
61 }
62 /************************************************************************/
63 /*函数名：Hashed
64  *功  能：检测一个字符串是否被hash过
65  *返回值：如果存在，返回位置；否则，返回-1
66  ************************************************************************/
67 unsigned long Hashed(char * lpszString)
68
69 {
70     const unsigned long HASH_OFFSET = 0, HASH_A = 1, HASH_B = 2;
71     //不同的字符串三次hash还会碰撞的率无限接近于不可能
72     unsigned long nHash = HashString(lpszString, HASH_OFFSET);
73     unsigned long nHashA = HashString(lpszString, HASH_A);
74     unsigned long nHashB = HashString(lpszString, HASH_B);
75     unsigned long nHashStart = nHash % m_tablelength;
76     unsigned long nHashPos = nHashStart;
77
78     while (m_HashIndexTable[nHashPos].bExists)
79     {
80         if (m_HashIndexTable[nHashPos].nHashA == nHashA && m_HashIndexTable[nHashPos].nHashB == nHashB)
81             return nHashPos;
82         else
83             nHashPos = (nHashPos + 1) % m_tablelength;
84
85         if (nHashPos == nHashStart)
86             break;
87     }
88     errstr++;
89
90     return -1; //没有找到
91 }
92
93 /************************************************************************/
94 /*函数名：Hash
95  *功  能：hash一个字符串
96  *返回值：成功，返回true；失败，返回false
97  ************************************************************************/
98 bool Hash(char * lpszString)
99 {
100     const unsigned long HASH_OFFSET = 0, HASH_A = 1, HASH_B = 2;
101     unsigned long nHash = HashString(lpszString, HASH_OFFSET);
102     unsigned long nHashA = HashString(lpszString, HASH_A);
103     unsigned long nHashB = HashString(lpszString, HASH_B);
104     unsigned long nHashStart = nHash % m_tablelength,
105     nHashPos = nHashStart;
106
107     while (m_HashIndexTable[nHashPos].bExists)
108     {
109         nHashPos = (nHashPos + 1) % m_tablelength;
110         if (nHashPos == nHashStart) //一个轮回
111         {
112             collc ++;
113             //hash表中没有空余的位置了,无法完成hash
114             return false;
115         }
116     }
117     m_HashIndexTable[nHashPos].bExists = true;
118     m_HashIndexTable[nHashPos].nHashA = nHashA;
119     m_HashIndexTable[nHashPos].nHashB = nHashB;
120
121     return true;
122 }
123
124 int InitHashTable()
125 {
126     int i;
127
128     InitCryptTable();
129     m_tablelength = nTableLength;
130
131     m_HashIndexTable = (HASHTABLE *)malloc(nTableLength * sizeof(HASHTABLE));
132     if (NULL == m_HashIndexTable) {
133         printf("Init HashTable failure!!\n");
134         return -1;
135     }
136
137     for (i = 0; i < nTableLength; i++ )
138     {
139         m_HashIndexTable[i].nHashA = 0;
140         m_HashIndexTable[i].nHashB = 0;
141         m_HashIndexTable[i].bExists = false;
142     }
143
144     return 0;
145 }
146
147 void do_test()
148 {
149     int count = 0;
150     FILE *fp;
151     char url[2048] = {0};
152
153     fp = fopen("urllist", "rb+");
154     if (NULL == fp) {
155         return;
156     }
157
158     if (InitHashTable()) {
159         return;
160     }
161
162     while (!feof(fp)) {
163         fgets(url, 2048, fp);
164         Hash(url);
165         count++;
166     }
167
168     printf("count: %d\n", count);
169
170     fclose(fp);
171 }
172
173 /*test main*/
174 int main()
175 {
176     do_test();
177
178     printf("conflict: %d\n", collc);
179     printf("not find: %d\n", errstr);
180
181     return 0;
182 }

#include <stdio.h>
#include <string.h>
#include <stdlib.h>

#define MOD 10

int ELFhash(char*key)
{
unsigned long h=0;
while(*key)
{
h = (h << 4) + *key++;
unsigned long g = h & 0xF0000000L;
if(g)
h ^= g >> 24;
h &= ~g;
}
return h % MOD;
}

int main(int argc, char **argv)
{
if (argc < 2) {
printf("using %s <string>\n", argv[0]);
return -1;
}

int num = 0;
num = ELFhash(argv[1]);

printf("num is %d\n", num);
}