modulus operator is not working if i give number more than 24 digits

If number is more than 24 digits, modulus operator is not giving correct output
here attached sample code
[code]<script type="text/javascript">
var a=10000000000000000000000.0;
var b=10.0;
var c=a % b;
alert("c"+c);
</script>[code]

please tel me solution

The largest number that Javascript can handle reliably without loss of precision is 9e15 or 9000000000000000. Any number greater than that is liable to return incorrect values for parseInt(), % modulus etc.

See also:- http://jsfromhell.com/classes/bignumber

All advice is supplied packaged by intellectual weight, and not by volume. Contents may settle slightly in transit.

If number is more than 24 digits, modulus operator is not giving correct output
here attached sample code
[code]<script type="text/javascript">
var a=10000000000000000000000.0;
var b=10.0;
var c=a % b;
alert("c"+c);
</script>[code]

please tel me solution
your problem is that you use decimals. It work for me if I remove decimal part.

best regards

your problem is that you use decimals. It work for me if I remove decimal part.


Not so! The real problem is that which I mentioned above. Integer numbers greater than 9e15 may or may not render accurately, depending of course on whether they are amenable to binary or not. Just as you cannot write 1/3 as a binary floating point number (resolves to 0.3333333333333333) but 1/4 is correctly evaluated to 0.25.


<script type="text/javascript">
var a=100000000000000000000000;
var b=10;
var c=a % b;
alert("c"+c); // 2

var a=90000000000000000000000; // 9e22
var b=10;
var c=a % b;
alert("c"+c); // 6

var a=9000000000000000000000; // 9e21
var b=10;
var c=a % b;
alert("c"+c); // 0

var a = 9971992547409847;
document.write(a); // 9971992547409848

</script>

As a further point of clarification:

JavaScript doesn't *REALLY* use integer arithmetic when ANY value involved exceeds 2147483647 (2^31-1). (And the spec doesn't require it to ever use integers, but I would strongly suspect that all modern implementations do so when they can.) That number is the largest positive integer that can be held in a standard 32-bit signed integer. (The smallest negative number is one greater, thanks to the foibles of 2's-complement binary representation.)

So...instead, JS must use a double precision floating point number. And the IEEE/ANSI format for such numbers (used by all modern CPUs) gives only 53 bits for the "mantissa". And 2^53 is 9007199254740992, whence the number that Philip is citing as the maximum possible integer before you begin losing precision. (It's probably actually 9007199254740991, one less than 2^53, again because of 2's-complement notation, but it's been too long since I investigated the format for me to remember that for sure.)