Округление чисел в PHP
В PHP для округления чисел применяются следующие функции:
Округление к ближайшему целому
Функция round($num, $precision) возвращает округлённое значение $num с указанной точностью $precision (количество цифр после запятой).
echo round(111.455); // 111 // Один знак после запятой echo round(111.455, 1); // 111.5 // Два знака после запятой echo round(111.455, 2); // 111.46Также функция round() может округлять целую часть числа, для этого нужно указать отрицательный $precision .
// Один знак перед запятой echo round(111.455, -1); // 110 // Два знака перед запятой echo round(111.455, -2); // 100Третьим аргументом функции можно повлиять способ округления.
| PHP_ROUND_HALF_UP | Округляет от нуля, когда следующий знак находится посередине. То есть округляет 1.5 в 2 и -1.5 в -2. |
| PHP_ROUND_HALF_DOWN | Округляет к нулю, когда следующий знак находится посередине. То есть округляет 1.5 в 1 и -1.5 в -1. |
| PHP_ROUND_HALF_EVEN | Округляет до ближайшего чётного значения, когда следующий знак находится посередине. То есть округляет 1.5 и 2.5 в 2. |
| PHP_ROUND_HALF_ODD | Округляет до ближайшего нечётного значения, когда следующий знак находится посередине. То есть округляет 1.5 в 1 и 2.5 в 3. |
Округление к большему
Функция ceil($num) – всегда округляет число в большую сторону.
echo ceil(1.1); // 2 echo ceil(1.5); // 2 echo ceil(1.6); // 2ceil
Returns the next highest integer value by rounding up num if necessary.
Parameters
Return Values
num rounded up to the next highest integer. The return value of ceil() is still of type float as the value range of float is usually bigger than that of int .
Changelog
Examples
Example #1 ceil() example
See Also
User Contributed Notes 21 notes
I needed this and couldn’t find it so I thought someone else wouldn’t have to look through a bunch of Google results-
// duplicates m$ excel’s ceiling function
if( ! function_exists ( ‘ceiling’ ) )
function ceiling ( $number , $significance = 1 )
return ( is_numeric ( $number ) && is_numeric ( $significance ) ) ? ( ceil ( $number / $significance )* $significance ) : false ;
>
>
echo ceiling ( 0 , 1000 ); // 0
echo ceiling ( 1 , 1 ); // 1000
echo ceiling ( 1001 , 1000 ); // 2000
echo ceiling ( 1.27 , 0.05 ); // 1.30
Caution!
$value = 77.4 ;
echo ceil ( $value * 100 ) / 100 ; // 77.41 — WRONG!
echo ceil ( round ( $value * 100 )) / 100 ; // 77.4 — OK!
I couldn’t find any functions to do what ceiling does while still leaving I specified number of decimal places, so I wrote a couple functions myself. round_up is like ceil but allows you to specify a number of decimal places. round_out does the same, but rounds away from zero.
// round_up:
// rounds up a float to a specified number of decimal places
// (basically acts like ceil() but allows for decimal places)
function round_up ( $value , $places = 0 ) if ( $places < 0 ) < $places = 0 ; >
$mult = pow ( 10 , $places );
return ceil ( $value * $mult ) / $mult ;
>
// round_out:
// rounds a float away from zero to a specified number of decimal places
function round_out ( $value , $places = 0 ) if ( $places < 0 ) < $places = 0 ; >
$mult = pow ( 10 , $places );
return ( $value >= 0 ? ceil ( $value * $mult ): floor ( $value * $mult )) / $mult ;
>
echo round_up ( 56.77001 , 2 ); // displays 56.78
echo round_up (- 0.453001 , 4 ); // displays -0.453
echo round_out ( 56.77001 , 2 ); // displays 56.78
echo round_out (- 0.453001 , 4 ); // displays -0.4531
?>
Actual behaviour:
echo ceil(-0.1); //result «-0» but i expect «0»
Workaround:
echo ceil(-0.1)+0; //result «0»
function roundaway ( $num ) return(( $num > 0 ) ? ceil ( $num ) : floor ( $num ));
>
?>
Slightly pointless, but there you have it, in one line only..
Please see http://www.php.net/manual/en/language.types.float.php for information regarding floating point precision issues.
$test_1 = ceil($test);
$test_2 = ceil($test * 10);
var_dump($test_1); // float 4
var_dump($test_2); // float 31
Caution!
$value = 77.4 ;
echo ceil ( $value * 100 ) / 100 ; // 77.41 — WRONG!
echo ceil ((string)( $value * 100 )) / 100 ; // 77.4 — OK!
function fCeil($val,$pressision=2) $p = pow(10,$pressision);
$val = $val*$p;
$val = ceil($val);
return $val /$p;
>
Note that ‘ceil’ can show unexpected behaviour when using float values due to inaccuracies in the float system and PHP tendency to freely limiting the number of digits it displays to the user.
It might like it is rounding up values that end in 0 (zero) but actually PHP is not showing that the value is not exactly the value it shows.
echo( ‘Displaying 13915.0000000000018190 : ‘ . 13915.0000000000018190 );
echo( ‘
‘ );
echo( ‘Result ceil(13915.0000000000018190) :’ . ceil ( 13915.0000000000018190 ));
echo( ‘
‘ );
?>
For example 13915 cannot be represented as exactly 13915 but becomes
13916.0000000000018190 and ‘ceil’ will therefore round it up to 13916.
$value = 12650 * 1.1 ;
echo( ‘Expected behaviour : ceil(12650 * 1.1) = ceil(13915) = 13915’ );
echo( ‘
‘ );
echo( ‘Using only ceil :’ . ceil ( $value ));
echo( ‘
‘ );
echo( ‘Showing decimals : ‘ . number_format ( $value — floor ( $value ), 16 ));
echo( ‘
‘ );
echo( ‘Using ceil + round : ‘ . ceil ( round ( $value , 4 )));
?>
So if ‘ceil’ looks like it is not working correctly and rounding up figures that end in 0 (zero) then this might be the cause of this behaviour.
Adding a simple ’round’ before the applying the ‘ceil’ solves this problem.
$k = 0.14 * 100;
echo ceil($k); // results 15
solution is in converting float number to string
Example 1.
echo ceil («»); // results 14
Example 2.
$totalSum1 = 102.1568;
$k = $totalSum1 / 100;
echo ceil («»); // results 102.16
Example 3.
$totalSum2 = 102.15;
$k = $totalSum1 / 100;
echo ceil («»); // results 102.15
useful for ‘ceil’ with precision capability
Here is a navbar using the ceil function.
function navbar ( $num_rows , $page , $link ) <
$nbrlink = 10 ; /* Number of link to display per page */
$page = (int) $page ; /* Page now displayed */
$num_rows = (int) $num_rows ;
if( $num_rows > 0 ) <
$total_page = ceil ( $num_rows / $nbrlink );
return $ret ;
>
>
/* Let say that $num_rows content the numbre of rows of your sql query */
$navbar = navbar ( $num_rows , $page , «listmovie.php?id= $id » );
IceKarma said: «If you want, say, 2.6 to round to 3, and -2.6 to round to -3, you want round(), which rounds away from zero.»
That’s not always true. round() doesn’t work that way, like zomis2k said it just rounds up _or_ down to the nearest non-decimal number. However this should work.
function roundaway ( $num ) if ( $num > 0 )
return ceil ( $num );
elseif ( $num < 0 )
return floor ( $num );
elseif ( $num == 0 )
return 0 ;
>
Ceil for decimal numbers with precision:
function ceil_dec($number,$precision,$separator)
$numberpart=explode($separator,$number);
$numberpart[1]=substr_replace($numberpart[1],$separator,$precision,0);
if($numberpart[0]>=0)
else
$ceil_number= array($numberpart[0],$numberpart[1]);
return implode($separator,$ceil_number);
>
echo ceil_dec(1.125,2,».»); //1.13
echo ceil_dec(-1.3436,3,».»); //-1.343
echo ceil_dec(102938.1,4,».»); //102938.1
Here is a more accurate way to round on superior decimal.
function round_sup($nb, $precision) $p = pow(10, $precision);
return ceil(round($nb * $p, 1)) / $p;
>
$k = 10 * 4.98; // k = 49.8
echo ceil($k * 10) / 10; // 49.9 !!
echo round_sup($k, 1); // 49.8
Remember that floating point precision means behavior can be «correct» — though not what you expect:
php > echo 100 * 1 * 0.07;
7
php > echo ceil(100 * 1 * 0.07);
8
Some people asking on rounding -1.5 to -2 and 1.5 to 2, the way is this:
The code below rounds a value up to a nearest multiple, away from zero. The multiple does not have to be a integer. So you could round, say, to the nearest 25.4, allowing you to round measurements in mm to the nearest inch longer.
// $x is the variable
// $c is the base multiple to round to, away from zero
$result = ( ( $y = $x / $c ) == ( $y = (int) $y ) ) ? $x : ( $x >= 0 ?++ $y :— $y )* $c ;
?>
I originally developed this as an example of write-only code: to make the point that being cleverly terse might save clock ticks but wastes more in programmer time generating un-maintainable code.
The inline code above nests one conditional statement inside another. The value of y changes twice within the same line (three times, if you count the pre-increment). The value of each assignment is used to determine branching within the conditional statement.
How it works can more easily be seen from the expansion below:
function myCeilingLong ( $x , $c )
// $x is variable
// $c is ceiling multiple
$a = $x / $c ;
$b = (int) $a ;
if ( $a == $b )
return $x ; // x is already a multiple of c;
else
if ( $x >= 0 )
return ( $b + 1 )* $c ; // return ((int)(x/c)+1 ) * c
else
return ( $b — 1 )* $c ; // return ((int)(x/c)-1 ) * c
>
>
?>
function myCeilingShort ( $x , $c )
return ( ( $y = $x / $c ) == ( $y = (int) $y ) ) ? $x : ( $x >= 0 ?++ $y :— $y )* $c ;
>
?>
Comparing the versions for speed: the in-line version is about three times faster than myCeilingLong() — but this is almost entirely down to function call overhead.
Putting the in-line code inside the function: the difference in execution speed between myCeilingLong() and myCeilingShort() is around 1.5%.
ceil() is still around 25% faster than the in-line statement so if you are a speed hound your efforts might be better devoted to compiling your own library .