您好,我的哈希表有一个问题,它的实现如下:
#define HT_SIZE 10
typedef struct _list_t_ {
char key[20];
char string[20];
char prevValue[20];
struct _list_t_ *next;
} list_t;
typedef struct _hash_table_t_ {
int size; /* the size of the table */
list_t ***table; /* first */
sem_t lock;
} hash_table_t;
我有一个包含 3 个指针的链表,因为我想要一个包含多个分区(分片)的哈希表,这是我的哈希表的初始化:
hash_table_t *create_hash_table(int NUM_SERVER_THREADS, int num_shards){
hash_table_t *new_table;
int j,i;
if (HT_SIZE<1) return NULL; /* invalid size for table */
/* Attempt to allocate memory for the hashtable structure */
new_table = (hash_table_t*)malloc(sizeof(hash_table_t)*HT_SIZE);
/* Attempt to allocate memory for the table itself */
new_table->table = (list_t ***)calloc(1,sizeof(list_t **));
/* Initialize the elements of the table */
for(j=0; j<num_shards; j++){
new_table->table[j] = (list_t **)calloc(1,sizeof(list_t *));
for(i=0; i<HT_SIZE; i++){
new_table->table[j][i] = (list_t *)calloc(1,sizeof(list_t ));
}
}
/* Set the table's size */
new_table->size = HT_SIZE;
sem_init(&new_table->lock, 0, 1);
return new_table;
}
这是我在哈希表中搜索的搜索函数
list_t *lookup_string(hash_table_t *hashtable, char *key, int shardId){
list_t *list ;
int hashval = hash(key);
/* Go to the correct list based on the hash value and see if key is
* in the list. If it is, return return a pointer to the list element.
* If it isn't, the item isn't in the table, so return NULL.
*/
sem_wait(&hashtable->lock);
for(list = hashtable->table[shardId][hashval]; list != NULL; list =list->next) {
if (strcmp(key, list->key) == 0){
sem_post(&hashtable->lock);
return list;
}
}
sem_post(&hashtable->lock);
return NULL;
}
还有我的插入函数:
char *add_string(hash_table_t *hashtable, char *str,char *key, int shardId){
list_t *new_list;
list_t *current_list;
unsigned int hashval = hash(key);
/*printf("|%d|%d|%s|\n",hashval,shardId,key);*/
/* Lock for concurrency */
sem_wait(&hashtable->lock);
/* Attempt to allocate memory for list */
new_list = (list_t*)malloc(sizeof(list_t));
/* Does item already exist? */
sem_post(&hashtable->lock);
current_list = lookup_string(hashtable, key,shardId);
sem_wait(&hashtable->lock);
/* item already exists, don't insert it again. */
if (current_list != NULL){
strcpy(new_list->prevValue,current_list->string);
strcpy(new_list->string,str);
strcpy(new_list->key,key);
new_list->next = hashtable->table[shardId][hashval];
hashtable->table[shardId][hashval] = new_list;
sem_post(&hashtable->lock);
return new_list->prevValue;
}
/* Insert into list */
strcpy(new_list->string,str);
strcpy(new_list->key,key);
new_list->next = hashtable->table[shardId][hashval];
hashtable->table[shardId][hashval] = new_list;
/* Unlock */
sem_post(&hashtable->lock);
return new_list->prevValue;
}
我的主类通过从哈希表的元素执行插入/读取/删除来运行一些测试,问题是当我有超过 4 个分区/分片时,测试会在第一个读取元素处停止,说它返回了错误搜索函数上的值为 NULL,当它小于 4 时,它运行得很好并通过了所有测试。
如果你想看一下,你可以在这里看到我的 main.c:
http://hostcode.sourceforge.net/view/1105
我的完整哈希表代码:
http://hostcode.sourceforge.net/view/1103
以及执行哈希表代码的其他函数:
.c 文件 http://hostcode.sourceforge.net/view/1104
.h 文件 http://hostcode.sourceforge.net/view/1106
感谢您抽出宝贵的时间,我感谢您能给我的任何帮助,这是我正在努力解决的一个大学重要项目,我在这里被困了 2 天。
最佳答案
您好,已经解决了这个问题,我在初始化中进行了错误的分配:
new_table->table = (list_t ***)calloc(1,sizeof(list_t **));
应该是这样的:
new_table->table = (list_t ***)calloc(num_shards,sizeof(list_t **));
关于c - C 中的哈希表插入/搜索,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/19745026/