algorithm/hello-algo
1
0
Fork 0
mirror of https://github.com/krahets/hello-algo.git synced 2025-11-01 20:12:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

Refactor the hash_map_open_addressing implementation with lazy reallocation. (#776)

Browse Source
This commit is contained in:
Yudong Jin
2023-09-21 04:43:15 -05:00
committed by GitHub
parent 45e20e57a1
commit a46b482951
4 changed files with 190 additions and 145 deletions
Download Patch File Download Diff File Expand all files Collapse all files

177
codes/cpp/chapter_hashing/hash_map_open_addressing.cpp
View File

@ -6,26 +6,28 @@
#include "./array_hash_map.cpp"
/* 开放寻址哈希表 */
class HashMapOpenAddressing {
private:
int size; // 键值对数量
int capacity; // 哈希表容量
double loadThres; // 触发扩容的负载因子阈值
int extendRatio; // 扩容倍数
vector<Pair *> buckets; // 桶数组
Pair *removed; // 删除标记
int size; // 键值对数量
int capacity = 4; // 哈希表容量
const double loadThres = 2.0 / 3; // 触发扩容的负载因子阈值
const int extendRatio = 2; // 扩容倍数
vector<Pair *> buckets; // 桶数组
Pair *TOMBSTONE = new Pair(-1, "-1"); // 删除标记
public:
/* 构造方法 */
HashMapOpenAddressing() {
// 构造方法
size = 0;
capacity = 4;
loadThres = 2.0 / 3.0;
extendRatio = 2;
buckets = vector<Pair *>(capacity, nullptr);
removed = new Pair(-1, "-1");
HashMapOpenAddressing() : size(0), buckets(capacity, nullptr) {
}
/* 析构方法 */
~HashMapOpenAddressing() {
for (Pair *pair : buckets) {
if (pair != nullptr && pair != TOMBSTONE) {
delete pair;
}
}
delete TOMBSTONE;
}
/* 哈希函数 */
@ -35,67 +37,75 @@ class HashMapOpenAddressing {
/* 负载因子 */
double loadFactor() {
return static_cast<double>(size) / capacity;
return (double)size / capacity;
}
/* 搜索 key 对应的桶索引 */
int findBucket(int key) {
int index = hashFunc(key);
int firstTombstone = -1;
// 线性探测，当遇到空桶时跳出
while (buckets[index] != nullptr) {
// 若遇到 key ，返回对应桶索引
if (buckets[index]->key == key) {
// 若之前遇到了删除标记，则将键值对移动至该索引
if (firstTombstone != -1) {
buckets[firstTombstone] = buckets[index];
buckets[index] = TOMBSTONE;
return firstTombstone; // 返回移动后的桶索引
}
return index; // 返回桶索引
}
// 记录遇到的首个删除标记
if (firstTombstone == -1 && buckets[index] == TOMBSTONE) {
firstTombstone = index;
}
// 计算桶索引，越过尾部返回头部
index = (index + 1) % capacity;
}
// 若 key 不存在，则返回添加点的索引
return firstTombstone == -1 ? index : firstTombstone;
}
/* 查询操作 */
string get(int key) {
int index = hashFunc(key);
// 线性探测，从 index 开始向后遍历
for (int i = 0; i < capacity; i++) {
// 计算桶索引，越过尾部返回头部
int j = (index + i) % capacity;
// 若遇到空桶，说明无此 key ，则返回 nullptr
if (buckets[j] == nullptr)
return nullptr;
// 若遇到指定 key ，则返回对应 val
if (buckets[j]->key == key && buckets[j] != removed)
return buckets[j]->val;
// 搜索 key 对应的桶索引
int index = findBucket(key);
// 若找到键值对，则返回对应 val
if (buckets[index] != nullptr && buckets[index] != TOMBSTONE) {
return buckets[index]->val;
}
return nullptr;
// 若键值对不存在，则返回空字符串
return "";
}
/* 添加操作 */
void put(int key, string val) {
// 当负载因子超过阈值时，执行扩容
if (loadFactor() > loadThres)
if (loadFactor() > loadThres) {
extend();
int index = hashFunc(key);
// 线性探测，从 index 开始向后遍历
for (int i = 0; i < capacity; i++) {
// 计算桶索引，越过尾部返回头部
int j = (index + i) % capacity;
// 若遇到空桶、或带有删除标记的桶，则将键值对放入该桶
if (buckets[j] == nullptr || buckets[j] == removed) {
buckets[j] = new Pair(key, val);
size += 1;
return;
}
// 若遇到指定 key ，则更新对应 val
if (buckets[j]->key == key) {
buckets[j]->val = val;
return;
}
}
// 搜索 key 对应的桶索引
int index = findBucket(key);
// 若找到键值对，则覆盖 val 并返回
if (buckets[index] != nullptr && buckets[index] != TOMBSTONE) {
buckets[index]->val = val;
return;
}
// 若键值对不存在，则添加该键值对
buckets[index] = new Pair(key, val);
size++;
}
/* 删除操作 */
void remove(int key) {
int index = hashFunc(key);
// 线性探测，从 index 开始向后遍历
for (int i = 0; i < capacity; i++) {
// 计算桶索引，越过尾部返回头部
int j = (index + i) % capacity;
// 若遇到空桶，说明无此 key ，则直接返回
if (buckets[j] == nullptr)
return;
// 若遇到指定 key ，则标记删除并返回
if (buckets[j]->key == key) {
delete buckets[j]; // 释放内存
buckets[j] = removed;
size -= 1;
return;
}
// 搜索 key 对应的桶索引
int index = findBucket(key);
// 若找到键值对，则用删除标记覆盖它
if (buckets[index] != nullptr && buckets[index] != TOMBSTONE) {
delete buckets[index];
buckets[index] = TOMBSTONE;
size--;
}
}
@ -109,19 +119,22 @@ class HashMapOpenAddressing {
size = 0;
// 将键值对从原哈希表搬运至新哈希表
for (Pair *pair : bucketsTmp) {
if (pair != nullptr && pair != removed) {
if (pair != nullptr && pair != TOMBSTONE) {
put(pair->key, pair->val);
delete pair;
}
}
}
/* 打印哈希表 */
void print() {
for (auto &pair : buckets) {
if (pair != nullptr) {
cout << pair->key << " -> " << pair->val << endl;
} else {
for (Pair *pair : buckets) {
if (pair == nullptr) {
cout << "nullptr" << endl;
} else if (pair == TOMBSTONE) {
cout << "TOMBSTONE" << endl;
} else {
cout << pair->key << " -> " << pair->val << endl;
}
}
}
@ -129,29 +142,29 @@ class HashMapOpenAddressing {
/* Driver Code */
int main() {
/* 初始化哈希表 */
HashMapOpenAddressing map = HashMapOpenAddressing();
// 初始化哈希表
HashMapOpenAddressing hashmap;
/* 添加操作 */
// 在哈希表中添加键值对 (key, value)
map.put(12836, "小哈");
map.put(15937, "小啰");
map.put(16750, "小算");
map.put(13276, "小法");
map.put(10583, "小鸭");
// 添加操作
// 在哈希表中添加键值对 (key, val)
hashmap.put(12836, "小哈");
hashmap.put(15937, "小啰");
hashmap.put(16750, "小算");
hashmap.put(13276, "小法");
hashmap.put(10583, "小鸭");
cout << "\n添加完成后，哈希表为\nKey -> Value" << endl;
map.print();
hashmap.print();
/* 查询操作 */
// 向哈希表输入键 key ，得到值 value
string name = map.get(13276);
// 查询操作
// 向哈希表输入键 key ，得到值 val
string name = hashmap.get(13276);
cout << "\n输入学号 13276 ，查询到姓名 " << name << endl;
/* 删除操作 */
// 在哈希表中删除键值对 (key, value)
map.remove(16750);
// 删除操作
// 在哈希表中删除键值对 (key, val)
hashmap.remove(16750);
cout << "\n删除 16750 后，哈希表为\nKey -> Value" << endl;
map.print();
hashmap.print();
return 0;
}