转自:https://www.cnblogs.com/evidd/p/8823092.html
详细介绍C++STL:unordered_map
不得不提一下,hash_map未加入在C++11标准中。
在VC中编译:
1 #include <hash_map> 2 using namespace stdext; 3 hash_map<int ,int> myhash;
在GCC中编译:
1 #include <ext/hash_map> 2 using namespace __gnu_cxx; 3 hash_map<int ,int> myhash;
既如此,还是用unordered_map吧!
C++ 11标准中加入了unordered系列的容器。unordered_map记录元素的hash值,根据hash值判断元素是否相同。map相当于java中的TreeMap,unordered_map相当于HashMap。无论从查找、插入上来说,unordered_map的效率都优于hash_map,更优于map;而空间复杂度方面,hash_map最低,unordered_map次之,map最大。
unordered_map与map的对比:
存储时是根据key的hash值判断元素是否相同,即unordered_map内部元素是无序的,而map中的元素是按照二叉搜索树存储(用红黑树实现),进行中序遍历会得到有序遍历。所以使用时map的key需要定义operator<。而unordered_map需要定义hash_value函数并且重载operator==。但是很多系统内置的数据类型都自带这些。
总结:结构体用map重载<运算符,结构体用unordered_map重载==运算符。
unordered_map与hash_map对比:
unordered_map原来属于boost分支和std::tr1中,而hash_map属于非标准容器。
unordered_map感觉速度和hash_map差不多,但是支持string做key,也可以使用复杂的对象作为key。
unordered_map编译时gxx需要添加编译选项:--std=c++11
unordered_map模板:
1 template < class Key, // unordered_map::key_type 2 class T, // unordered_map::mapped_type 3 class Hash = hash<Key>, // unordered_map::hasher 4 class Pred = equal_to<Key>, // unordered_map::key_equal 5 class Alloc = allocator< pair<const Key,T> > // unordered_map::allocator_type 6 > class unordered_map;
迭代器:
unordered_map的迭代器是一个指针,指向这个元素,通过迭代器来取得它的值。
1 unordered_map<Key,T>::iterator it; 2 (*it).first; // the key value (of type Key) 3 (*it).second; // the mapped value (of type T) 4 (*it); // the "element value" (of type pair<const Key,T>)
它的键值分别是迭代器的first和second属性。
1 it->first; // same as (*it).first (the key value) 2 it->second; // same as (*it).second (the mapped value)
成员函数:
=================迭代器=========================
begin 返回指向容器起始位置的迭代器(iterator)
end 返回指向容器末尾位置的迭代器
cbegin 返回指向容器起始位置的常迭代器(const_iterator)
cend 返回指向容器末尾位置的常迭代器
=================Capacity================
size 返回有效元素个数
max_size 返回 unordered_map 支持的最大元素个数
empty 判断是否为空
=================元素访问=================
operator[] 访问元素
at 访问元素
=================元素修改=================
insert 插入元素
erase 删除元素
swap 交换内容
clear 清空内容
emplace 构造及插入一个元素
emplace_hint 按提示构造及插入一个元素
================操作=========================
find 通过给定主键查找元素,没找到:返回unordered_map::end
count 返回匹配给定主键的元素的个数
equal_range 返回值匹配给定搜索值的元素组成的范围
================Buckets======================
bucket_count 返回槽(Bucket)数
max_bucket_count 返回最大槽数
bucket_size 返回槽大小
bucket 返回元素所在槽的序号
load_factor 返回载入因子,即一个元素槽(Bucket)的最大元素数
max_load_factor 返回或设置最大载入因子
rehash 设置槽数
reserve 请求改变容器容量
插入元素示例:
1 // unordered_map::insert 2 #include <iostream> 3 #include <string> 4 #include <unordered_map> 5 using namespace std; 6 7 void display(unordered_map<string,double> myrecipe,string str) 8 { 9 cout << str << endl; 10 for (auto& x: myrecipe) 11 cout << x.first << ": " << x.second << endl; 12 cout << endl; 13 } 14 15 int main () 16 { 17 unordered_map<string,double> 18 myrecipe, 19 mypantry = {{"milk",2.0},{"flour",1.5}}; 20 21 /****************插入*****************/ 22 pair<string,double> myshopping ("baking powder",0.3); 23 myrecipe.insert (myshopping); // 复制插入 24 myrecipe.insert (make_pair<string,double>("eggs",6.0)); // 移动插入 25 myrecipe.insert (mypantry.begin(), mypantry.end()); // 范围插入 26 myrecipe.insert ({{"sugar",0.8},{"salt",0.1}}); // 初始化数组插入(可以用二维一次插入多个元素,也可以用一维插入一个元素) 27 myrecipe["coffee"] = 10.0; //数组形式插入 28 29 display(myrecipe,"myrecipe contains:"); 30 31 /****************查找*****************/ 32 unordered_map<string,double>::const_iterator got = myrecipe.find ("coffee"); 33 34 if ( got == myrecipe.end() ) 35 cout << "not found"; 36 else 37 cout << "found "<<got->first << " is " << got->second<<"\n\n"; 38 /****************修改*****************/ 39 myrecipe.at("coffee") = 9.0; 40 myrecipe["milk"] = 3.0; 41 display(myrecipe,"After modify myrecipe contains:"); 42 43 44 /****************擦除*****************/ 45 myrecipe.erase(myrecipe.begin()); //通过位置 46 myrecipe.erase("milk"); //通过key 47 display(myrecipe,"After erase myrecipe contains:"); 48 49 /****************交换*****************/ 50 myrecipe.swap(mypantry); 51 display(myrecipe,"After swap with mypantry, myrecipe contains:"); 52 53 /****************清空*****************/ 54 myrecipe.clear(); 55 display(myrecipe,"After clear, myrecipe contains:"); 56 return 0; 57 } 58 /* 59 myrecipe contains: 60 salt: 0.1 61 milk: 2 62 flour: 1.5 63 coffee: 10 64 eggs: 6 65 sugar: 0.8 66 baking powder: 0.3 67 68 found coffee is 10 69 70 After modify myrecipe contains: 71 salt: 0.1 72 milk: 3 73 flour: 1.5 74 coffee: 9 75 eggs: 6 76 sugar: 0.8 77 baking powder: 0.3 78 79 After erase myrecipe contains: 80 flour: 1.5 81 coffee: 9 82 eggs: 6 83 sugar: 0.8 84 baking powder: 0.3 85 86 After swap with mypantry, myrecipe contains: 87 flour: 1.5 88 milk: 2 89 90 After clear, myrecipe contains: 91 */
遍历示例:
1 // unordered_map::bucket_count 2 #include <iostream> 3 #include <string> 4 #include <unordered_map> 5 using namespace std; 6 7 int main () 8 { 9 unordered_map<string,string> mymap = 10 { 11 {"house","maison"}, 12 {"apple","pomme"}, 13 {"tree","arbre"}, 14 {"book","livre"}, 15 {"door","porte"}, 16 {"grapefruit","pamplemousse"} 17 }; 18 /************begin和end迭代器***************/ 19 cout << "mymap contains:"; 20 for ( auto it = mymap.begin(); it != mymap.end(); ++it ) 21 cout << " " << it->first << ":" << it->second; 22 cout << endl; 23 /************bucket操作***************/ 24 unsigned n = mymap.bucket_count(); 25 26 cout << "mymap has " << n << " buckets.\n"; 27 28 for (unsigned i=0; i<n; ++i) 29 { 30 cout << "bucket #" << i << "'s size:"<<mymap.bucket_size(i)<<" contains: "; 31 for (auto it = mymap.begin(i); it!=mymap.end(i); ++it) 32 cout << "[" << it->first << ":" << it->second << "] "; 33 cout << "\n"; 34 } 35 36 cout <<"\nkey:'apple' is in bucket #" << mymap.bucket("apple") <<endl; 37 cout <<"\nkey:'computer' is in bucket #" << mymap.bucket("computer") <<endl; 38 39 return 0; 40 } 41 /* 42 mymap contains: door:porte grapefruit:pamplemousse tree:arbre apple:pomme book:livre house:maison 43 mymap has 7 buckets. 44 bucket #0's size:2 contains: [book:livre] [house:maison] 45 bucket #1's size:0 contains: 46 bucket #2's size:0 contains: 47 bucket #3's size:2 contains: [grapefruit:pamplemousse] [tree:arbre] 48 bucket #4's size:0 contains: 49 bucket #5's size:1 contains: [apple:pomme] 50 bucket #6's size:1 contains: [door:porte] 51 52 key:'apple' is in bucket #5 53 54 key:'computer' is in bucket #6 55 */