leetcode-187. 重复的DNA序列-C语言

/**
 * 算法思想：
 * 遍历所有以10为长度的字符串，边遍历边添加进哈希表，如果哈希表中存在，并且添加进结果的标志还未置位，
 * 则将其添加进结果（注意cnt用来标识是否已经添加进结果）；
 * 如果哈希表中不存在，将当前遍历的字符串添加到哈希表中；
 * 哈希值的计算方法要尽量将其均匀分布在哈希表中，否则容易超时。
 */
#define L 0x1ffff

int get(char c){
    if(c == 'A') return 0;
    if(c == 'C') return 1;
    if(c == 'G') return 2;
    return 3;
}

typedef struct hnode{
    int key;
    char *s;
    int cnt;
    struct hnode *next;
} HNode;

void hash_init(HNode ***htb, int len){
    int i;
    *htb = (HNode **)malloc(sizeof(HNode*) * len);
    for(i=0; i<len; i++){
        (*htb)[i] = NULL;
    }
}

void hash_exit(HNode **htb, int len){
    int i, cnt=0;
    HNode *node, *last;

    for(i=0; i<len; i++){
        node=(htb)[i];
        if(node) cnt++;
        while(node){
            last = node;
            node = node->next;
            free(last);
        }
    }    
    printf("len = %d, cnt = %d\n", len,cnt);
}

void hash_insert(HNode **hash_table, int key, char *s){
    int index = key % L;
    
    //printf("index = %d\n", index);
    HNode *node = (HNode *)malloc(sizeof(HNode));
    
    node->key = key;
    node->s = s;
    node->cnt = 0;
    node->next = hash_table[index];
    
    hash_table[index] = node;
}

HNode *hash_find(HNode **table, int key, char *s) {
    int index = key % L;
    HNode *p = table[index];
    
    while(p){
        if(!memcmp(p->s, s, sizeof(char)*10)){
            break;
        }
        p = p->next;
    }
    
    return p;
}


char** findRepeatedDnaSequences(char* s, int* returnSize) {
    int i, j, k;
    int len = strlen(s);
    char **ret;
    ret = (char **)malloc(sizeof(char *) * 1000);
    int index = 0;
    int flag = 0, key;
    HNode **htable, *node, *p;
    
    hash_init(&htable, L);
    
    *returnSize = 0;
    if(len <= 9) return NULL;
    
    int table[len-9][4];
    memset(table, 0, sizeof(table));
    
    for(i=0; i<len-9; i++){
        /* 计算哈希值，尽量均匀落在哈希表中 */
        table[i][0] = s[i] * s[i+9] + s[i+1] * s[i+8]+ s[i+2] * s[i+7] + s[i+3] * s[i+6] + s[i+4] * s[i+5];
        table[i][0] += s[i] * s[i+8] + s[i+2] * s[i+9]+ s[i+2] * s[i+6] + s[i+3] * s[i+5] + s[i+4] * s[i+7];
    }
    printf("len=%d\n", len);
    
    for(i=0; i<len-9; i++){
           key = table[i][0];
           
           p = hash_find(htable, key, &s[i]);
           
           if(!p) {
               hash_insert(htable, key, &s[i]);
           }else{
               if(!p->cnt) {
                   p->cnt = 1;
                   ret[index] = (char *)malloc(sizeof(char) * 11);
                   memcpy(ret[index], p->s, 10);
                   ret[index++][10]= '\0';                       
               }
           }   

    }
    
    
    hash_exit(htable, L);
    
    *returnSize = index;
    return ret;
}