Subscripting a Multidimensional Array

As with any array, we can use a subscript to access the elements of a multidimensional array. To do so, we use a separate subscript for each dimension.

If an expression provides as many subscripts as there are dimensions, we get an element with the specified type. If we supply fewer subscripts than there are dimensions, then the result is the inner-array element at the specified index:

// assigns the first element of arr to the last element in the last row of ia
ia[2][3] = arr[0][0][0];
int (&row)[4] = ia[1]; // binds row to the second four-element array in ia

In the first example we supply indices for all the dimensions for both arrays. On the left-hand side, ia[2] returns the last row in ia. It does not fetch an element from that array but returns the array itself. We subscript that array, fetching element [3], which is the last element in that array.

Similarly, the right-hand operand has three dimensions. We first fetch the array at index 0 from the outermost array. The result of that operation is a (multidimensional) array of size 20. We take the first element from that 20-element array, yielding an array of size 30. We then fetch the first element from that array.

In the second example, we define row as a reference to an array of four ints. We bind that reference to the second row in ia.

As another example, it is common to use a pair of nested for loops to process the elements in a multidimensional array:

constexpr size_t rowCnt = 3, colCnt = 4;
int ia[rowCnt][colCnt];   // 12 uninitialized elements
// for each row
for (size_t i = 0; i != rowCnt; ++i) {
    // for each column within the row
    for (size_t j = 0; j != colCnt; ++j) {
        // assign the element's positional index as its value
        ia[i][j] = i * colCnt + j;
    }
}

The outer for loops through each of the array elements in ia. The inner for loops through the elements of those interior arrays. In this case, we set the value of each element as its index in the overall array.