c++ trying to understand clockwise rules for deciphering complicated syntax

Question

I have the following code:

int ia[3][4] = {    //
{0, 1, 2, 3},   //
{4, 5, 6, 7},   //
{8, 9, 10, 11}  //
};

int (*p4)[4] = ia;
cout << \"(*(p4 + 0))         


        

Answer 1

Postfix expressions have higher priority than unary expressions.

So this expression

*(p4 + 0)[3]

is equivalent to the expression

*( (p4 + 0)[3] )

As it is seen it is not the same as

( *(p4 + 0) )[3]

This expression *( (p4 + 0)[3] ) can be written just like *p4[3] that is equivalent to *ia[3] and due to the priority is considered like

*( ia[3] )

or like

ia[3][0]

The valid range for indices of the first dimension for the array ia is [0, 2] because the array is declared like

int ia[3][4] = { /*...*/ };
      ^^^

Thus expression ia[3] tries to access memory beyond the array. As result the code snippet has undefined behavior.

Shortly speaking you should understand that these expressions

(*(p4 + 0))[3]
 *(p4 + 0)[3]

that in turn are equivalent to the following pair of expressions

ia[0][3]
io[3][0]

are not equivalent due to the priorities of the operations.

Answer 2

I don't understand the last one how it arrives at 1.

Undefined behavior got you there.

Due to operator precedence (the array indexing operator binds tighter than the pointer dereference operator),

*(p4 + 0)[3] is the same as:
*((p4 + 0)[3]), which is the same as:
*(p4[3]), which is the same as:
p4[3][0].

For your array. the valid indices for the first dimension are: 0, 1, and 2. Accessing the array using the index value of 3 accesses memory beyond valid range, leading to undefined behavior.