本文整理汇总了PHP中Math::partsToFloat方法的典型用法代码示例。如果您正苦于以下问题：PHP Math::partsToFloat方法的具体用法？PHP Math::partsToFloat怎么用？PHP Math::partsToFloat使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类Math的用法示例。
在下文中一共展示了Math::partsToFloat方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的PHP代码示例。
示例1: floatULP
/**
* Return the ulp for the given float argument. The ulp is the
* difference between the argument and the next larger float. Note
* that the sign of the float argument is ignored, that is,
* ulp(x) == ulp(-x). If the argument is a NaN, then NaN is returned.
* If the argument is an infinity, then +Inf is returned. If the
* argument is zero (either positive or negative), then
* {@link Float#MIN_VALUE} is returned.
*
* @param float $f
*
* @return float
*/
public static function floatULP($f)
{
if (is_nan($f)) {
return $f;
}
if (is_infinite($f)) {
return INF;
}
// This handles both +0.0 and -0.0.
if ($f == 0.0) {
return Float . MIN_VALUE;
}
$bits = Math::floatToParts($f);
$mantissa = $bits[2];
$exponent = $bits[1];
// Denormal number, so the answer is easy.
if ($bits[1] == 0) {
$bits[2] = 1;
//set fraction to smallest possible value
return Math::partsToFloat($bits);
}
// Conceptually we want to have '1' as the mantissa. Then we would
// shift the mantissa over to make a normal number. If this underflows
// the exponent, we will make a denormal result.
$bits[1] -= 23;
$bits[2] = 0;
if ($bits[1] == 0) {
$bits[2] = 1 << -($bits[1] - 1);
$bits[1] = 0;
}
return Math::partsToFloat($bits);
}