[RESOLVED] casting question

jrough

I did this
<?php
$var1= (int) ((0.1 +0.7) *10);
echo $var1;
var_dump ($var1);
?>
and I expected the answer to be 8 yet I got 7.
Why is that?

thanks,

dagon

Converting to integer

Warning
Floating point precision

It is typical that simple decimal fractions like 0.1 or 0.7 cannot be converted into their internal binary counterparts without a small loss of precision. This can lead to confusing results: for example, floor((0.1+0.7)*10) will usually return 7 instead of the expected 8, since the internal representation will be something like 7.9.

This is due to the fact that it is impossible to express some fractions in decimal notation with a finite number of digits. For instance, 1/3 in decimal form becomes 0.3.

So never trust floating number results to the last digit, and never compare floating point numbers for equality. If higher precision is necessary, the arbitrary precision math functions and gmp functions are available.

jrough

0.7 + 0.1 = 0.8
0.8 * 10 = 8

Isn't 8 already an integer?
But if you
echo (int) 8.0 you get 8.
So I still don't get it.
thanks,

dagon

0.7 is a float
0.1 is a float

float math suffers from the issues described in the warning

jrough

thanks, I learned something

bradgrafelman

jrough wrote:
0.7 + 0.1 = 0.8

This is true on paper, but not in a computer's memory - see dagon's post.

jrough wrote:
0.8 10 = 8

This statement is incorrect. The result of this equation will be a float, since 0.8 is a float. Then again, your script above never carries out this operation to begin with - it actually does something like 0.799999999 10, which is, of course, 7.99999999.

EDIT: By the way, this whole discussion reminds me about movies like Office Space. :p

dagon

jrough;10950799 wrote:
thanks, I learned something

math was so much easier in primary school, 1 apple + 1 apple made sense.

halojoy

There are at least 2 fixes I can think of.

My second example is the use of one very small correction konstant $k.
Here I use 1E-10 ( 0.000 000 000 1)
We add this very small value to make sure the value is not a fraction too small.

The third example is to use [man]round[/man]
This will round to nearest integer (int)

The code below will output

<?php

echo ((0.1 +0.7));
echo '<br />';
echo (int)((0.1 +0.7) * 10);
echo '<br />';
echo '<br />';

$k = 1E-10;
echo ((0.1 +0.7));
echo '<br />';
echo (int)((0.1 +0.7) * 10 + $k);
echo '<br />';
echo '<br />';

echo round((0.1 +0.7) * 10);

?>

Weedpacket

Another approach is to scale everything up at the beginning and work as much as possible with integers, scaling back down at the end (e.g., instead of working with $4.10, call it 410¢ instead). Not only will this help avoid rounding issues, but integer arithmetic is pretty much guaranteed to be faster than the floating-point equivalent.

jrough

You have a good 2010 also, thanks for the example