JavaScript floating-point value arithmetic rounding error

Problem

In JavaScript, where integers and floating-point numbers belong to the Number data type (one of the simple data types), we often find that the result of printing 1.0 such a float is not 1 1.0 , because the memory space that holds the floating-point numbers is twice times the value of the saved integer. So ECMAScript will lose no chance to convert the floating-point number to an integer.
This situation is a bit uncomfortable for obsessive-compulsive patients, but it's not a big mistake at all, and the next thing is scary. For example, when we calculate the 0.1 addition 0.2 , it's output is not 0.3 , but 0.3000000000000004 . What the fuck?! Have you ever felt the challenge of the world view for the first time in a child's shoes?

Cause

So quickly to save his soul and body, found the book is impressively written: Ecmascrip is based on IEEE754 numerical floating point calculation, this numerical method will save the value as a binary and then calculate, because the floating-point number in binary expression is infinite, so in the arithmetic calculation will produce rounding error , because of the rounding error, the accuracy of floating-point calculation is far less than that of integers, and finally the value of a particular floating-point number is never tested .

Solution Solutions

The so-called remedy, know the cause of the problem so that you can find a solution to the problem. Since the binary of floating-point numbers is an infinite number of errors, this error does not exist in integer arithmetic, is it possible for you to get a glimpse of the truth? Yes, that is to convert the floating-point number to an integer in an arithmetic project, and then convert the resulting result to a floating-point number. Sir, here is the new code ~

1 //addition2 functionFloatadd (arg1,arg2) {3        varr1,r2,m;4        Try{r1=arg1.tostring (). Split (".") [1].length}Catch(e) {r1=0;//parameter 1 is an integer}; Parameter 1 number of digits after the decimal point5        Try{r2=arg2.tostring (). Split (".") [1].length}Catch(e) {r2=0;//parameter 2 is an integer}; Parameter 2 number of digits after the decimal point6M=math.pow (10,math.max (R1,R2));//take the larger number of digits7        return(arg1*m+arg2*m)/m; Convert arg1 and arg2 to integers before converting back to floating point numbers8}

The above reprint attached the original website 1190000007207259

The toFixed () method should also handle part of the rounding error problem in a dummy manner.

Syntax: number.tofixed (x) x: Specifies the number of decimal places, which are values between 0 and 20, including 0 and 20, and some implementations can support a larger range of values. If this argument is omitted, 0 will be used instead.

END

JavaScript floating-point value arithmetic rounding error