Reputation: 517
I'trying to convert a float to an int value in php:
var_dump((int)(39.3 * 100.0)); //Returns 3929 but should be 3930!
var_dump((int)(39.2 * 100.0)); //Returns 3920
I can use ceil to make it work but can somebody explain this to me?
var_dump((int)ceil(39.3 * 100.0)); //Returns 3930
Upvotes: 21
Views: 10243
Reputation: 6877
This might be late but the proper way to do it is as follows:
(int) bcmul("39.3", "100.0"); // 3930
(int) bcmul("39.2", "100.0"); // 3920
and for dealing with digits / calculation for floats, or anything that involves money transaction, you should never use directly multiplication/division + casting.
please refer to this as well:
http://php.net/manual/en/book.bc.php
notice: the casting in my answer is just to convert String to Int, (you don't need to do it)
Upvotes: 18
Reputation: 97815
This is because numbers that have a finite representation in base 10 may or may not have an exact representation in the floating point representation PHP uses.
See
>php -r "echo var_dump(sprintf('%.40F', 39.3 * 100.0));" string(45) "3929.9999999999995452526491135358810424804688"
Since int
always rounds the number down, a small error in the representation makes the cast round it one number down that you would otherwise expect.
Consider using round
instead.
Upvotes: 25
Reputation: 85
// gives: int(3930)
var_dump(intval((39.3 * 100.0) . ''));
or, for use with a function:
function floatToInteger ($fValue)
{
return (intval(($fValue + 0) . ''));
}
// gives: int(3930)
var_dump(floatToInteger(39.3 * 100.0));
Upvotes: 4