Why is it if I do this in javascript, I get the following result:
1234.56 * 10 = 12345.599999999999
It should be 123456. How can I get around this problem?
Thanks.
Floating points are not exact, since there are ifinite numbers at their range [or in any range to be more exact], and only a finite number of bits to store this data.
Have a look at what every programmer should know about floating point arithmetics.
Another easy solution:
parseFloat((1234.56 * 10).toPrecision(12))
and the result will be: 12345.6, and YES... it works with decimal numbers.
As the others said, floating points and so on.
Easy solution would be to do something like this:
var answer = parseInt(1234.56 * 10);
Or just use Math.round?
All numbers in JS are internally defined by float and drop the less significant digits if needed.
(10000000000000000000000000000 + 1) == 10000000000000000000000000000
// this will return true
And javascript is well known for droping bits quite often in numbers. So handle with care
Related
I am using toFixed but the method does not operate as expected
parseFloat(19373.315).toFixed(2);
//19373.31 Chrome
Expected Output : 19373.32
parseFloat(9373.315).toFixed(2);
// 9373.32 Working fine
Why does the first example round down, whereas the second example round up?
The problem is that binary floating point representation of most decimal fractions is not exact. The internal representation of 19373.315 may actually be something like 19373.314999999, so toFixed rounds down, while 19373.315 might be 19373.315000001, which rounds up.
Why does the first example round down, whereas the second example round up?
Look at the binary representation of the two values in memory.
const farr = new Float64Array(2);
farr[0] = 19373.315;
farr[1] = 9373.315;
const uarr = new Uint32Array(farr.buffer);
console.log(farr[0], uarr[1].toString(2).padStart(32, 0) + uarr[0].toString(2).padStart(32, 0));
console.log(farr[1], uarr[3].toString(2).padStart(32, 0) + uarr[2].toString(2).padStart(32, 0));
Without diving into the details, we can see that the second value has an additional '1' at the end, which is lost in the first larger value when it is fit into 64 bits.
Other answers have explained why, I would suggest using a library like numeral.js which will round things as you would expect.
Assuming toFixed casts to 32-bit float;
Check with this utility...
19373.315 is stored as 19373.314453125 (an error of -0.000546875) in 32-bit floating point format.
This is despite (19373.315).toFixed(4) coming out as 19373.3150.
Even if this is "expected" or "intended", I'd still report it as a bug.
It should use a double during the rounding check, and thus proper rounding during conversion to fixed string.
I think the spec even says so. :\
In the V8 javascript engine source, the Number.prototype.toFixed function invokes DoubleToFixedCString in this file ...
There's probably some inappropriate optimization in there... (Looking into it.)
I'd suggest submitting an additional test case for V8 with 19373.315 specifically.
(19373.3150).toFixed(39) yields 19373.314999999998690327629446983337402343750.
Rounding occurs once to bring it up to 19373.315 - which is correct - but not at the right digit when rounding to 2 digits.
I think this should have a second pass on rounding here to catch edge cases like this. I think it might have to round to n+1 digits, then again to n digits. Maybe there's some other clever way to fix it though.
function toFixedFixed(a,n) {
return (a|0) + parseFloat((a % 1).toFixed(n+1)).toFixed(n).substr(1);
}
console.log(toFixedFixed(19373.315,2)); // "19373.32"
console.log(toFixedFixed(19373.315,3)); // "19373.315"
console.log(toFixedFixed(19373.315,4)); // "19373.3150"
console.log(toFixedFixed(19373.315,37)); // "19373.3149999999986903276294469833374023438"
console.log(toFixedFixed(19373.315,38)); // "19373.31499999999869032762944698333740234375"
console.log(toFixedFixed(19373.315,39)); // "19373.314999999998690327629446983337402343750"
(Adopted from my comments on Vahid Rahmani's answer, who is correct.)
I want to use toPrecision() to reduce the size of a number before I display it. However, I sometimes cannot multiply output of the function by another number without gaining a small rounding error. See the code sample below:
var x = 0.0197992182093305
alert (x.toPrecision(4)) //Correct: 0.01980
alert (Number(x.toPrecision(4))) //Correct: 0.0198
alert( Number(x.toPrecision(4)) * 100) //Incorrect: 1.9800000000000002
JSFiddle: http://jsfiddle.net/ueLsL460/4/ What's going on here?
Based on what I understand, Number(x.toPrecision(4)) * 100 creates a new Number object which will not inherit the precision of the parent.
If you still want it to be precise after Math, you need to put it in precision again.
alert((x * 100).toPrecision(4));
Technically, it's not an error. It's just the way javascript is supposed to work.
The use of primitive constructors is not that ideal, unless you are trying to do something trivial. Can you please try to do the code on the following fiddle and see if this will do for you?
http://jsfiddle.net/ueLsL460/5/
var x = 0.0197992182093305;
alert((x * 100).toPrecision(4));
Javascript doesn't have a decimal equivalent - so in your case floating points are used. What this means is that Number(x.toPrecision(4)) doesn't give you exactly 0.0198, but something the FP binary number closest to 0.0198. So any arithmetic you do will be subject to the loss of precision introduced in floating point arithmetic. You can see the same effect if you do
alert(0.0198 * 100);
By the way
alert(0.0198 * 10 * 10);
gives you no problem (the loss of precision does not build up enough to make it into the digits Javascript deems to display) - but that's for this particular number alone.
For rounding decimals (prices), I've been using .toFixed(2) for quite some time now. But I just recently discovered that Javascript can't "precisely" round decimals. I was a bit shocked that even 10.005 couldn't be rounded correctly to 10.01. It just got rounded down to 10.00. And other times it did round correctly. I like to have control over my code, so this is a big no-no for me.
And since I'm calculating prices, I think I need something more (100%) accurate for rounding only 2- or 3-decimal numbers, maybe a 4-decimal one.
Is there no straightforward way of doing basic rounding in javascript, the correct way?
UPDATE: As Felix Kling has suggested, the method of processing my prices as integers of cents, there are also drawbacks to this (besides more code)?
The reason that a number like 10.005 can't be rounded corretly is that you don't really have the number 10.005, you only have a number that is the closest possible one that can be represented using a double precision floating point variable.
The actual number that you have might be someting like 10.00499999999276253, and that would naturally round to 10.00 rather than 10.01.
To handle monetary values you should use a data type that can represent the value exactly. As numbers in Javascript are always floating point numbers, what you are left with is representing the numbers as text, and writing your own functions to do the math (or find someone who has done that already).
This should work for you, here's a fiddle to play around with it http://jsfiddle.net/5ffyC/1/
function moneyRound(flt){
var splitStr = flt.toString().split('.'),
whole = (flt * 100) | 0;
if (splitStr.length > 1 && splitStr[1].length > 2){
return splitStr[1][2] > 4? (whole + 1) / 100: whole / 100;
} else {
return flt;
}
}
I am trying to fix the number to 2 digits after decimal and for that i am using toFixedfunction of javascript. Below are the strange results i am getting, please check and help me.
var number = 11.995;
number.toFixed(2); // giving me 11.99 which is correct
var number = 19.995;
number.toFixed(2); // giving me 20.00 which is incorrect
Can anyone tell me why it is happening.
Thanks for your help.
This is how floating point math works. The value 19.995 is not exact binary (base 2). To make it more clear, think of an exact number when you divide 10/3.
For more in-depth explanations, read this: http://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html
In your case you can work with strings instead (at least it seems like that is what you want):
number.toString().substr(0, n);
Or define a function like this (made in 2 minutes, just an example):
Number.toFixed = function(no, n) {
var spl = no.toString().split('.');
if ( spl.length > 1 ) {
return spl[0]+'.'+spl[1].substr(0,n);
}
return spl[0];
}
Number.toFixed(19.995, 2); // 19.99
toFixed rounds the value. Since 19.995 is exactly halfway between 19.99 and 20.00, it has to choose one of them. Traditionally, rounding prefers the even result (this prevents bias, since round-ups and round-downs will be equal).
I have create a function which done all for me..
function toFixed(number, precision) {
var multiplier = Math.pow(10, precision + 1),
wholeNumber = Math.floor(number * multiplier);
return Math.round(wholeNumber / 10) * 10 / multiplier;
}
//Call this function to retrive exect value
toFixed((+adjustmentval), 2);
David has answered your doubt I'm guessing. Just providing an alternate solution here.
You can use the Math.floor() method of the Math object for this.
Something like this, Math.floor(number*100)/100
Can anyone tell me why it is happening.
The IEEE-754 double-precision binary floating point number standard used by JavaScript's number type (and similar times in several other languages) does not perfectly store all numbers, it stores some numbers imprecisely, in a way that lets it A) Store them in just 64 bits, and B) Calculate with them quickly.
For 11.995, the actual value is 11.99499988555908203125, just slightly less than 11.995.
For 19.995, the actual value is 19.9950008392333984375, just slightly more than 19.995.
That explains why when you round them using the usual round-to-nearest-half-up operation, 11.995 (which is really 11.99499988555908203125) rounds down to 11.99 but 19.995 (which is really 19.9950008392333984375) rounds up to 20.00.
(This site has a handy calculator for visualizing this stuff.)
More here on SO:
Is floating point math broken?
How to deal with floating point number precision in JavaScript?
As you all know since it is one of the most asked topic on SO, I am having problems with rounding errors (it isn't actually errors, I am well aware).
Instead of explaining my point, I'll give an example of what possible numbers I have and which input I want to be able to obtain:
Let's say
var a = 15 * 1e-9;
alert(a)
outputs
1.5000000000000002e-8
I want to be able to obtain 1.5e-8 instead, but I cannot just multiply by 10e8, round and divide by 10e8 because I don't know if it will be e-8 or e-45 or anything else.
So basically I want to be able to obtain the 1.5000002 part, apply toFixed(3) and put back the exponent part.
I could convert into a string and parse but it just doesn't seem right...
Any idea ?
(I apologize in advance if you feel this is one of many duplicates, but I could not find a similar question, only related ones)
Gael
You can use the toPrecision method:
var a = 15 * 1e-9;
a.toPrecision(2); // "1.5e-8"
If you're doing scientific work and need to round with significant figures in mind: Rounding to an arbitrary number of significant digits
var a = 15 * 1e-9;
console.log(Number.parseFloat(a).toExponential(2));
//the above formula will display the result in the console as: "1.50e-8"