What is this code doing? (size_t)-1

Question

Can someone explain what happens when size_t, or any other type identifier, is wrapped in parentheses. I know it is the old typecast syntax but in this context I don't follow what is happening.

I've seen it for defining the max size of a type as:

size_t max_size = (size_t)-1

It is a C-style cast. It casts `-1` to `size_t`, which presumably yields a very large number. — juanchopanza, Oct 07 '13 at 20:37
@juanchopanza: To be precise, the max number that `size_t` can represent :) — , Oct 07 '13 at 20:38
The cast is unnecessary. `size_t max_size = -1;` does the same thing. — Kerrek SB, Oct 07 '13 at 20:38
@juanchopanza: The C++ standard specifies that `size_t` is unsigned (because it inherits specifications about it from C) and that unsigned integer types use modulo arithmetic. — Eric Postpischil, Oct 07 '13 at 20:41
@KerrekSB Interesting. I thought there was some leeway for one's complement implementations, presumably for archaeological reasons. I must have been misinformed. — juanchopanza, Oct 07 '13 at 20:43
@jtetm Anyway, the C++ way to get the maximum value of `size_t` would be `std::numeric_limits::max()`. — juanchopanza, Oct 07 '13 at 20:47
@juanchopanza: You can have one's complement, but it means more work when you convert signed to unsigned... — Kerrek SB, Oct 07 '13 at 20:48

score 17 · Accepted Answer · edited May 24 '19 at 17:33

17

This code (unnecessarily) casts -1 to size_t. The most probable intent was getting the largest possible value of size_t on this system.

Although this code doesn't have Undefined Behavior, this code is ugly - in C++ you should use std::numeric_limits<size_t>::max() and in C use SIZE_MAX macro for exactly a purpose of getting the largest size_t value.

edited May 24 '19 at 17:33

Drew Dormann

59,987
13
123
180

answered Oct 07 '13 at 20:38

sasha.sochka

14,395
10
44
68

3

This code has well-defined behaviour. You might not like it, but it's sanctioned by the standard. (Both C and C++ standards, in fact.) And in C++ (the OP tag), you should use `std::numeric_limits::max()` – rici Oct 08 '13 at 01:10

score 1 · Answer 2 · edited May 23 '17 at 12:09

1

(size_t)-1 is in fact the equivalent of size_t(-1)

See also the following question c cast syntax styles

edited May 23 '17 at 12:09

Community

1
1

answered Oct 07 '13 at 20:39

Rod

52,748
3
38
55

score 0 · Answer 3 · answered Feb 06 '18 at 22:26

0

Some library methods intentionally return (size_t)(-1) to indicate an error condition. For example, the iconv method from the GNU libiconv library. I assume there is some good reason why these functions don't return ssize_t (signed) return values, which would allow you to check for -1 directly.

answered Feb 06 '18 at 22:26

MikeOnline

994
11
18

You wouldn't want to dedicate 16 bit only for the negative error codes. – Sep 09 '19 at 14:39

Osman GOKTURK · Answer 4 · 2023-03-28T14:24:27.253

0

for one of the code snippet I have encountered, size_t-1 aimed simply to have -1. They wanted to terminate for negative values.

The code snippet is bellow.

while (len != (size_t)-1) 
{
    mod = num % 16;
    if (mod < 10)
        str[len] = mod + '0';
    else if (mod >= 10)
        str[len] = (mod - 10) + 'a';
    num = num / 16;
    len--;
}

edited Mar 28 '23 at 14:24

answered Mar 22 '23 at 15:57

Osman GOKTURK

1
2

1

As it’s currently written, your answer is unclear. Please [edit] to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers [in the help center](/help/how-to-answer). – Community Mar 28 '23 at 00:53

What is this code doing? (size_t)-1

4 Answers4

Linked