通过基于范围循环使用映射向量 c++ 中的 std::min_element 获取最小元素

Question

I'm trying to get min value using std::min_element in a function.

std::vector<std::map<int, double>> vectorMap;
std::map<int, double> data;


bool compare(std::pair<int, double> i, std::pair<int, double> j) {
  return i.second < j.second;
}

double findMin(std::vector<std::map<int, double>> const &vectorMap){
    for (auto const &vec : vectorMap){
        for (auto const &mymap : vec){
            std::pair<int, double> min = (*std::min_element(mymap.begin(), mymap.end(), compare));
        }
    }
    return static_cast<double>(min.second);
}

While compiling I'm getting these errors:

main.cpp:59:67: error: no member named 'begin' in 'std::pair<const int, double>'
            std::pair<int, double> min = (*std::min_element(mymap.begin(), mymap.end(), compare));
                                                            ~~~~~ ^

main.cpp:59:82: error: no member named 'end' in 'std::pair<const int, double>'
            std::pair<int, double> min = (*std::min_element(mymap.begin(), mymap.end(), compare));
                                                            ~~~~~ ^

main.cpp:62:32: error: use of undeclared identifier 'min'
    return static_cast<double>(min.second);

But when using the std::min_element in the main function it works fine.

std::vector<std::map<int, double>> vectorMap;
std::map<int, double> data;

...
...

std::pair<int, double> min = (*std::min_element(data.begin(), data.end(), compare));
    std::cout << "min value is " << min.second << std::endl; //works well

whole code

What am I doing wrong? Thank you.

Answer 1

You have confusing naming here:

double findMin(std::vector<std::map<int, double>> const &vectorMap){
    for (auto const &vec : vectorMap){
        for (auto const &mymap : vec){
            std::pair<int, double> min = (*std::min_element(mymap.begin(), mymap.end(), compare));
        }
    }
    return static_cast<double>(min.second);
}

In the for (auto const &vec: vectorMap){ line you iterate over elements of the vector, so each element of it is a map (it is obvious from its type - std::vector<std::map<int, double>> ). So each element of the vector is a map<int, double> . But you call each element as a vec and this name may mess things for you. Then you iterate over each map:

for (auto const &mymap : vec)

and here naming is also wrong. As you iterate over map then each element of it is a pair<int, double> which map consists of. That's why pair doesn't have begin/end methods. As as mentioned by Margarit Alexandru you don't need a second loop, as your vec is already a map that can be passed to std::min_element . Just use correct naming)

Answer 2

You are using min_element on each element of the map (of type std::pair<int, double>). You should have only one for instead of 2.

double findMin(std::vector<std::map<int, double>> const &vectorMap){
    for (auto const &vec : vectorMap){
        std::pair<int, double> min = (*std::min_element(vec.begin(), vec.end(), compare));
    }
    return static_cast<double>(min.second);
}