Difference between String() and new String() in Javascript
Strings returned from String
calls in a non-constructor context (i.e., without using the new
keyword) are primitive strings.
Strings created with new String()
(constructor mode) is an object and can store property in them.
Demonstrating the difference:
var strPrimitive = String('word');
strPrimitive.prop = "bar";
console.log(strPrimitive.prop); // undefined
var strObject = new String('word');
strObject.prop = "bar";
console.log(strObject.prop); // bar
Using the String()
constructor without new
gives you the string (primitive) value of the passed parameter. It's like boxing the parameter in a native object if necessary (like a Number or Boolean), and then calling .toString()
on it. (Of course if you pass a plain object reference it just calls .toString()
on that.)
Calling new String(something)
makes a String instance object.
The results look the same via console.log()
because it'll just extract the primitive string from the String instance you pass to it.
So: just plain String()
returns a string primitive. new String(xyz)
returns an object constructed by the String constructor.
It's rarely necessary to explicitly construct a String instance.
String() returns a string primitive and new String() returns a Object String. This has some real consequences for your code.
- Using String() returns 'true' with other primitives both with == and === operator.
- Using String() gives you a primitive so it cannot use the "instanceOf" method to check its type. You can check only value type with "typeof" operator
- Using new String() with "instanceOf" method with String or Object prototypes - both assert to true.
- Using new String() will return 'true' with string primitives only by calling valueOf() method. String() has also this method and returns true when compared to string of the same value.
- Using new String() allows you to add some other properties and methods to the Object to allow more complex behaviour.
From my coding experience you should avoid using new String() if you have no need for adding special methods to your String Object.
var x = String('word');
console.log(typeof x); // "string"
var y = new String('word');
console.log(typeof y); // "object"
// compare two objects !!!
console.log(new String('') === new String('')) // false!!!
// compare with string primitive
console.log('' == String('')) // true
console.log('' === String('')) // true
//compare with string Object
console.log('' == new String('')) // true
//!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
console.log('' === new String('')) // false !!!!
//!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
// instance of behavior
console.log(x instanceof String); // false
console.log(x instanceof Object); // false
// please note that only new String() is a instanceOf Object and String
console.log(y instanceof String); // true
console.log(y instanceof Object); // true
//valueOf behavior
console.log('word' == x.valueOf()); // true
console.log('word' === x.valueOf()); // true
console.log('word' == y.valueOf()); // true
console.log('word' === y.valueOf()); // true
//create smart string
var superString = new String('Voice')
superString.powerful = 'POWERFUL'
String.prototype.shout = function () {
return `${this.powerful} ${this.toUpperCase()}`
};
console.log(superString.shout()) //"POWERFUL VOICE"
Here is an example in addition to the good answers already provided:
var x = String('word');
console.log(typeof x); // "string"
var y = new String('word');
console.log(typeof y); // "object"
The exact answer to your question is here in the documentation.
String literals (denoted by double or single quotes) and strings returned from
String
calls in a non-constructor context (i.e., without using the new keyword) are primitive strings.