Programming Tutorials

Using parseInt() and parseFloat() in JavaScript to convert data types to Numbers

By: Nicholas C. Zakas Printer Friendly Format    

JavaScript provides two methods for converting non-number primitives into numbers: parseInt() and parseFloat() . As you may have guessed, the former converts a value into an integer whereas the latter converts a value into a floating-point number. These methods only work properly when called on strings; all other types return NaN .

Both parseInt() and parseFloat() look at a string carefully before deciding what its numeric value should be. The parseInt() method starts with the character in position 0 and determines if this is a valid number; if it isn’t, the method returns NaN and doesn’t continue. If, however, the number is valid, the method goes on to the character in position 1 and does the same test. This process continues until a character isn’t a valid number, at which point parseInt() takes the string (up to that point) and converts it into a number. For example, if you want to convert the string “1234blue” to an integer, parseInt() would return a value of 1234 because it stops processing one it reaches the character b .

Any number literal contained in a string is also converted correctly, so the string “0xA” is properly converted into the number 10. However, the string “22.5” will be converted to 22 , because the decimal point is an invalid character for an integer. Some examples:

var iNum1 = parseInt(“1234blue”); //returns 1234

var iNum2 = parseInt(“0xA”); //returns 10

var iNum3 = parseInt(“22.5”); //returns 22

var iNum4 = parseInt(“blue”); //returns NaN

The parseInt() method also has a radix mode, allowing you to convert strings in binary, octal, hexadecimal, or any other base into an integer. The radix is specified as a second argument to parseInt() , so a call to parse a hexadecimal value looks like this:

var iNum1 = parseInt(“AF”, 16); //returns 175

Of course, this can also be done for binary, octal, and even decimal (which is the default mode):

var iNum1 = parseInt(“10”, 2); //returns 2

var iNum2 = parseInt(“10”, 8); //returns 8

var iNum2 = parseInt(“10”, 10); //returns 10

If decimal numbers contain a leading zero, it’s always best to specify the radix as 10 so that you won’t accidentally end up with an octal value. For example:

var iNum1 = parseInt(“010”); //returns 8

var iNum2 = parseInt(“010”, 8); //returns 8

var iNum3 = parseInt(“010”, 10); //returns 10

In this code, both lines are parsing the string “010” into a number. The first line thinks that the string is an octal value and parses it the same way as the second line (which specifies the radix as 8). The last line specifies a radix of 10, so iNum3 ends up equal to 10.

The parseFloat() method works in a similar way to parseInt() , looking at each character starting in position 0. It also continues until the first invalid character and then converts the string it has seen up to that point. For this method, however, the decimal point is a valid character the first time it appears. If two decimal points are present, the second is considered invalid and the parseFloat() method converts the string up until that position. This means that the string “22.34.5” will be parsed into 22.34 .

Another difference when using parseFloat() is that the string must represent a floating-point number in decimal form, not octal or hexadecimal. This method ignores leading zeros, so the octal number 0908 will be parsed into 908 , and the hexadecimal number 0xA will return NaN because x isn’t a valid character for a floating-point number. There is also no radix mode for parseFloat() .

Some examples of using parseFloat() :

var fNum1 = parseFloat(“1234blue”); //returns 1234.0

var fNum2 = parseFloat(“0xA”); //returns NaN

var fNum3 = parseFloat(“22.5”); //returns 22.5

var fNum4 = parseFloat(“22.34.5”); //returns 22.34

var fNum5 = parseFloat(“0908”); //returns 908

var fNum6 = parseFloat(“blue”); //returns NaN