max/min return wrong value if signed (negative) zero

[kkmuffme]

Description

The following code:

https://3v4l.org/JXU4T#v8.4.13

<?php
var_dump( min( round( -0.01 / 2, 0 ), 0 ) );
var_dump( max( round( -0.01 / 2, 0 ), 0 ) );

Resulted in this output:

int(0)
float(-0)

But I expected this output instead:

float(-0)
int(0)

For 99.9% of applications, this output is actually what people would expect however: (instead of -0)

float(0)
int(0)

so I guess it would make sense to also change that.

This does not have any impact on calculations or comparisons, this is purely visual but can lead to unexpected behavior if the returned number is used as a string (e.g. substr, strtok,...) since there suddenly is an additional character in front of the numbers.
This is especially unexpected, when the float is the result of float arithmetics, which can minimally negative numbers in some cases

PHP Version

8.4

Operating System

No response

[ndossche]

Versions prior to 8.4 output:

float(-0)
float(-0)

[TimWolla]

Changing the order of the parameters

<?php
var_dump( min( 0, round( -0.01 / 2, 0 ) ) );
var_dump( max( 0, round( -0.01 / 2, 0 ) ) );

returns:

float(-0)
int(0)

So this seems to be related to how 0.0 and -0.0 compare equal and thus the behavior is returning either the first or the last of all equal values.

[TimWolla]

Simpler reproducer:

<?php
var_dump(min(0.0, -0.0));
var_dump(min(-0.0, 0.0));
var_dump(max(0.0, -0.0));
var_dump(max(-0.0, 0.0));

[TimWolla]

JavaScript:

> Math.min(0.0, -0.0);
-0
> Math.min(-0.0, 0.0);
-0
> Math.max(-0.0, 0.0);
0
> Math.max(0.0, -0.0);
0

[TimWolla]

Full test case:

<?php
foreach ([-0.0, 0.0] as $a) {
	foreach ([-0.0, 0.0] as $b) {
		if ("$a $b" === "0 0" || "$a $b" === "-0 -0") {
			continue;
		}

		printf("min(%.17g, %.17g)   = %.17g\n", $a, $b, min($a, $b));
		printf("min([%.17g, %.17g]) = %.17g\n", $a, $b, min([$a, $b]));
		printf("max(%.17g, %.17g)   = %.17g\n", $a, $b, max($a, $b));
		printf("max([%.17g, %.17g]) = %.17g\n", $a, $b, max([$a, $b]));
		echo "\n";
	}
}

echo "======\n\n";

foreach ([-0.0, 0.0] as $a) {
	foreach ([-0.0, 0.0] as $b) {
		foreach ([-0.0, 0.0] as $c) {
			if ("$a $b $c" === "0 0 0" || "$a $b $c" === "-0 -0 -0") {
				continue;
			}

			printf("min(%.17g, %.17g, %.17g)   = %.17g\n", $a, $b, $c, min($a, $b, $c));
			printf("min([%.17g, %.17g, %.17g]) = %.17g\n", $a, $b, $c, min([$a, $b, $c]));
			printf("max(%.17g, %.17g, %.17g)   = %.17g\n", $a, $b, $c, max($a, $b, $c));
			printf("max([%.17g, %.17g, %.17g]) = %.17g\n", $a, $b, $c, max([$a, $b, $c]));
			echo "\n";
		}
	}
}

[eyupcanakman]

Every path keeps the first of two equal values except frameless min. The variadic loop replaces only on min_dval > Z_DVAL(args[i]), zend_hash_minmax only on compar(res, zv) > 0, and frameless max uses lhs >= rhs ? lhs : rhs. Frameless min uses lhs < rhs ? lhs : rhs, which takes the false branch on a tie and returns the second argument.

So only min changes with the call form:

min(0.0, -0.0)                    // -0  frameless 2-arg
min(0.0, -0.0, 0.0)               // +0  variadic loop
min([0.0, -0.0])                  // +0  zend_hash_minmax
call_user_func('min', 0.0, -0.0)  // +0  dynamic call, no frameless path

max agrees with itself across the forms, though keep-first stays order-dependent. max(0.0, -0.0) is +0 and max(-0.0, 0.0) is -0.

Changing frameless min to <= would line the paths up, and every form would still flip with argument order. Order-independence needs a sign check on equal doubles, giving min = -0 and max = +0 on a signed-zero tie.

That is the tie rule IEEE 754-2019 minimum/maximum use, and what Math.min/max implement. Would you want that, or is path agreement enough?