数组<int，2> dim在这段代码中意味着什么？

Question

I came across this piece of code while reading The c++ Programming Language 4th edition

template<class T>
class Matrix {
    array<int,2> dim; // two dimensions

    T∗ elem; // pointer to dim[0]*dim[1] elements of type T

public:
    Matrix(int d1, int d2) :dim{d1,d2}, elem{new T[d1∗d2]} {} // error handling omitted

    int size() const { return dim[0]∗dim[1]; }

    Matrix(const Matrix&); // copy constructor

    Matrix& operator=(const Matrix&); // copy assignment

    Matrix(Matrix&&); // move constructor

    Matrix& operator=(Matrix&&); // move assignment

    ˜Matrix() { delete[] elem; }
    // ...
};

There are two data members in the class of which one is a pointer of type T . I am not able to understand what array< int, 2 > dim means.

Answer 1

The member variable dim stores the size of the first and second dimension of your 2D matrix, Matrix< T > . Those two sizes are stored as an array< int, 2 > (I assume std::array< int, 2 > : an array of two values of type int ).

Without this member variable dim , Matrix< T > has no idea how many elements are contained in its heap-allocated array elem ( note that elem is a pointer to the first element contained in the array of contiguous elements). So Matrix< T > has no way of safely iterating these elements, since it would not know when to stop. (In fact the only useful operation Matrix< T > can perform, is deallocating the heap-allocated array, as is the case in your destructor.) Therefore, the size of the heap-allocated array (ie dim[0] * dim[1] ) is explicitly stored as well.

Answer 2

This is making use of the std::array from the standard library. You can find a detailed reference here: https://en.cppreference.com/w/cpp/container/array

array<int,N> x;

declares an array of integers of length x; x is 2 in your case.

This is later used to store the shape of your matrix.

Answer 3

它是一个类型为array<int, 2>的成员变量dim的声明（可能是std::array 。）