所以我使用的库有一个枚举(假设它名为LibEnum
).我需要有一个 LibEnum
的 std::unordered_set
,但是我得到编译错误,没有专门的 std::hash
用于它.我可以轻松地编写它并只返回值的数量(第一个元素是 0,第二个元素是 1 等),但是我应该把这个专业化放在哪里,它应该是什么样子的?我无法修改库源.
So the library I use has an enum (say it's named LibEnum
). I need to have an std::unordered_set
of LibEnum
, but I get compilation error that there is no specialized std::hash
for it. I could easily write it and just return the number of value (first element is 0, second 1 etc), but where exactly I should put this specialization and how should it look like? I can't modify the library sources.
enum LibEnum { A, B, C, D};
std::unordered_set <LibEnum> mySet;
//need std::hash for LibEnum
//how should it look like?
你可以专门针对你的类型std::hash
:
You can just specialise std::hash
for your type:
namespace std {
template <>
struct hash<FullyQualified::LibEnum> {
size_t operator ()(FullyQualified::LibEnum value) const {
return static_cast<size_t>(value);
}
};
}
或者,您可以在任何您喜欢的地方定义 hash
类型,并在实例化 std::unordered_map
时将其作为附加模板参数提供::>
Alternatively, you can define a hash
type where ever you like and just provide it as the additional template argument when instantiating std::unordered_map<FooEnum>
:
// Anywhere in your code prior to usage:
struct myhash {
std::size_t operator ()(LibEnum value) const {
return static_cast<std::size_t>(value);
}
};
// Usage:
std::unordered_map<LibEnum, T, myhash> some_hash;
这两种方法都不需要您修改库.
Neither methods require you to modify the library.
这篇关于如何为来自其他库的类型专门化 std::hash的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持跟版网!