Recently, while studying some code in javascript, I came across the following code:
return !!navigator.userAgent.match(/iPad/i);
I know that:
!true == false
!false == true
And consequently:
!!true == true
!!false == false
So I was in doubt: what is the logic of doing this type of operation?
This is a quick way to convert the result of an expression to a boolean value True / False (or in technical terms, providing typecast for boolean ), simplifying thus conditional evaluation processes.
There is a post on the original Stack Overflow that perfectly describes this behavior . The translation of the relevant part follows:
The first sign
!evaluates the expression, converting the logical value and denying it. The second sign!reverses the value of the first.In the documentation, regarding the logical operator : / p>
Returns
falseif its single-operator can be converted totrue; otherwise, it returnstrue.The following values are evaluated as "false":
falseNaNundefinednull""(empty string)0
Since the purpose of return !!navigator.userAgent.match(/iPad/i); is to find out whether the current device is an iPad or not, it is interesting / elegant that the function return is Boolean ( true or false ).
navigator.userAgent.match(/iPad/i); function (without the double negation), the return will be null or ['iPad'] (or maybe something else) instead of true or false .
Using !! is a way to force the return of this match to be converted to a Boolean value, while maintaining the correct logical value for the target in question. In other words, ['iPad'] turns true and null false . =)
In stages it becomes clearer:
navigator.userAgent.match(/iPad/i); // ex: retorno ['iPad'] !navigator.userAgent.match(/iPad/i); // primeira negação, !['iPad'] == false !!navigator.userAgent.match(/iPad/i); // segunda conversão, !false == true The double negation serves to make the return of the function more typesafe. It is a good technique to ensure that certain functions only return Boolean values, thus abstracting the user from the returns function that may not make sense.
For example
function isIpad() {
return navigator.userAgent.match(/iPad/i);
}
if ( isIpad() ) { ... }
This code will work because if iPad is an array is returned (and this is true ) if it is not going to return null (which is evaluated as false ). But for the developer to see in the output of isIpad() null or ["iPad"] it escapes the semantics of "is" so the double negation forces the function isIpad () to return exclusively or false or true