pointer to multidimensional array
A multidimensional array is a common data structure that is actually an array of arrays. Although this meaning is well understood, it is a bit troublesome when defining pointers that manipulate array data. Suppose you have the following two-dimensional array.
int matrix[10][10];
Which of the following is the correct way to define a pointer to one of the rows?
int *row[10];
int (*row)[10];
The answer is the second. Of course, it is a way to memorize it here, but you can also look at the code in another way:
int* row[10];
int(*row)[10];
The first case is an array of pointers, and the rest is an array of pointers.
How C++11 handles it
After C++11, with auto descriptors and begin/end functions, it is easier to define pointers to multidimensional arrays. For example, you can write code like this:
int matrix[10][10];
int number =1;
for(auto row = begin(matrix); row != end(matrix); ++row)
{
for(auto data = begin(*row); data != end(*row); ++data){
*data = number++;
}
}
Because row is an array pointer, and begin and end require a reference type, the parameter used when calling the begin and end functions to obtain the data pointer is *row, not row.
Other than that, it's all natural.
Author's point of view
Today's description is not so much a new content as it is a comprehensive exercise of the previous content.
