How can I remove the decimal part from JavaScript number?

You can also use bitwise operators to truncate the decimal.

e.g.

var x = 9 / 2;
console.log(x); // 4.5

x = ~~x;
console.log(x); // 4

x = -3.7
console.log(~~x) // -3
console.log(x | 0) // -3
console.log(x << 0) // -3

Bitwise operations are considerably more efficient than the Math functions. The double not bitwise operator also seems to slightly outperform the x | 0 and x << 0 bitwise operations by a negligible amount.

// 952 milliseconds
for (var i = 0; i < 1000000; i++) {
    (i * 0.5) | 0;
}

// 1150 milliseconds
for (var i = 0; i < 1000000; i++) {
    (i * 0.5) << 0;
}

// 1284 milliseconds
for (var i = 0; i < 1000000; i++) {
    Math.trunc(i * 0.5);
}

// 939 milliseconds
for (var i = 0; i < 1000000; i++) {
    ~~(i * 0.5);
}

Also worth noting is that the bitwise not operator takes precedence over arithmetic operations, so you may need to surround calculations with parentheses to have the intended result:

x = -3.7

console.log(~~x * 2) // -6
console.log(x * 2 | 0) // -7
console.log(x * 2 << 0) // -7

console.log(~~(x * 2)) // -7
console.log(x * 2 | 0) // -7
console.log(x * 2 << 0) // -7

More info about the double bitwise not operator can be found at Double bitwise NOT (~~)

You could use...

• Math.trunc() (truncate fractional part, also see below)
• Math.floor() (round down)
• Math.ceil() (round up)
• Math.round() (round to nearest integer)

...dependent on how you wanted to remove the decimal.

Math.trunc() isn't supported on all platforms yet (namely IE), but you could easily use a polyfill in the meantime.

Another method of truncating the fractional portion with excellent platform support is by using a bitwise operator (.e.g |0). The side-effect of using a bitwise operator on a number is it will treat its operand as a signed 32bit integer, therefore removing the fractional component. Keep in mind this will also mangle numbers larger than 32 bits.

---

You may also be talking about the inaccuracy of decimal rounding with floating point arithmetic.

Required Reading - What Every Computer Scientist Should Know About Floating-Point Arithmetic.

For Example:

var x = 9.656;
x.toFixed(0);           // returns 10
x.toFixed(2);           // returns 9.66
x.toFixed(4);           // returns 9.6560
x.toFixed(6);           // returns 9.656000 

or

parseInt("10");         // returns 10
parseInt("10.33");      // returns 10
parseInt("10 20 30");   // returns 10
parseInt("10 years");   // returns 10
parseInt("years 10");   // returns NaN