Diving Deeper in JavaScripts Objects
A Closer Look at JavaScript Object Descriptors
JavaScript objects pack more things than their terse and concise syntax would naturally exhibit. Creating and using objects in JavaScript is so easy, effortless, and so flexible that a lot of developers never realize that there is more to it.
We are going to uncover some of the hidden layers and understand the intricacies of JavaScript objects. After reading this article, you should be able to answer the following questions —
- How to make a property undeletable?
- What are accessor properties and what are their features?
- How to make a property immutable or hidden?
- Why some properties do not appear in
for-in
loops orObject.keys
and some do? - How to “protect” objects from modification?
- How to make sense of code such as
obj.id = 5;
console.log(obj.id)
// => '101' ( 5 in binary )
If you want a refresher of the basics of JS objects, I have written another article.
Tip: Use Bit to organize and reuse your JS components. It will help you build faster, using components like Lego. Example: Ramda, Semantic-UI. Try it out.
Types of Properties
Data Properties
You must have made countless objects like these —
const obj = {
name: 'Arfat',
id: 5
}obj.name
// => 'Arfat'
In object obj
, name
and id
properties are called Data Properties. These are the normal kind of properties that constitute most of the JavaScript code. What is, then, the other type of property?
Accessor Properties
They can also be understood as getters and setters of other languages such as C# or Python. An accessor property is a combination of two functions: the get
and the set
function.
Instead of using the traditional key: value
syntax, we use the following syntax —
const accessorObj = {
get name() {
return 'Arfat';
}
};accessorObj.name;
// => 'Arfat'const dataObj = {
name: 'Arfat',
};dataObj.name;
// => 'Arfat'
Look at accessorObj
and compare it with dataObj
. Right now, they exhibit the same behavior. We use the get
keyword and follow it with a function declaration. Reading accessor properties do not need to use parentheses to invoke the function. That is, accessorObj.name();
is wrong.
When we access(read) accessorObj.name
, the name
function is executed and its return value becomes the final value of the name
key.
The get function is called a getter since it is involved in getting the value of a key. If you update accessorObj.name = 'New Person’;
, the update won’t happen. This is because we don’t have a corresponding setter function for the name
key. Setter functions help in setting values of getter properties.
const accessorObj = {
_name: 'Arfat',
get name() {
return this._name;
},
set name(value) {
this._name = value;
}
};
The setter function receives the assigned value as a parameter. Now, you can store the value in property or a global variable. In this case, we make a conventional “private” property _name
and store the name value in it.
In the getter function, we can modify or override the property before returning its value. The following example should demonstrate this. It also answers one of the above questions.
Why would anybody use accessor properties if normal data properties exist? There are often cases where you need to log the property access or maintain a history of all the values that the property has had. Accessor properties give us the full power of a function with the ease of use of object properties. You can read more about accessor usage here.
So how does JavaScript know which is an accessor property and which is a data property? Let’s find out.
Object Property Descriptors
At first glance, it might look like there is a one-to-one mapping between keys and values of an object. However, that’s not entirely true.
Property Attributes
Every key of an object contains a set of property attributes that define the characteristics of the value associated with the key. They can also be thought of as meta-data describing the key-value pair. In short, attributes are used to define and explain the state of object properties.
The set of property attributes is called the property descriptor.
In total, there are six property attributes. They are —
- [[Value]]
- [[Get]]
- [[Set]]
- [[Writable]]
- [[Enumerable]]
- [[Configurable]]
Why have we wrapped the attribute names in [[]]? Double square brackets mark internal properties used by the ECMA specifications. These are properties that JS programmer cannot touch directly from within the code. To manipulate internal properties, we’d need to use methods provided to us by the language.
Let’s see an example —
In the above image, the object has 2 keys: x
and y
. You can see the corresponding list of attributes associated with each property.
How can you get the same information in JavaScript? We can use the function Object.getOwnPropertyDescriptor
to get that information. It takes an object and a property name and returns an object containing the required attributes. Here’s a code sample —
const object = {
x: 5,
y: 6
};Object.getOwnPropertyDescriptor(object, 'x');/*
{
value: 5,
writable: true,
enumerable: true,
configurable: true
}
*/
Let’s see the attributes in more detail to know how they are helpful and then we’ll revisit the image.
[[Value]]
It stores the value retrieved by a get access of the property. Which means that when we do object.x
in the above example, we actually retrieve its [[Value]] attribute. Any dot-access or square-bracket access of a Data property will work in this way.
[[Get]]
It stores the reference to the function that we declare while making a getter property. It is called with an empty arguments list to retrieve the property value each time a get access of the property is performed.
[[Set]]
It stores the reference to the function that we declare while making a setter property. It is called with an arguments list containing the assigned value as its sole argument each time a set access of the property is performed.
const obj = {
set x(val) {
console.log(val)
// => 23
}
}obj.x = 23;
In the above example, the right-hand-side of the assignment is passed as the val
argument to a setter function. You can see this code for a demonstration.
[[Writable]]
This is a boolean value. It tells whether we can overwrite the value or not. If false, attempts to change the property’s value will not succeed.
[[Enumerable]]
This is also a boolean value. This attribute dictates whether the property is going to appear in for-in
loops or not. If true, the for-in
loop will be able to iterate on this property.
[[Configurable]]
This is a boolean too.
When it is false —
- Attempts to delete the property will fail.
- Also, converting a Data Property to be an Accessor Property or vice-versa will fail too. That is, inter-conversion between two property types will not be possible.
- It will also prevent further changing the attributes values. That is, current values of
enumerable
,configurable
,get
orset
will become fixed.
The effect of this property is also dependent on the property type. Apart from the above effects, it also does the following.
- When the property is a Data Property, you can only set
writable
fromtrue
tofalse
. - Before
writable
becomes false, you can also change its[[Value]]
attribute. However, oncewritable
is false, andconfigurable
is false too, the property becomes unwritable, undeletable and unchangeable.
All six properties do not exist for each property type.
- For Data Properties, only
value
,writable
,enumerable
andconfigurable
exists. - For Accessor Properties, instead of
value
andwritable
, we haveget
andset
.
Working with Descriptors
Having learned about these descriptors, how can we set or update them on our own objects? There are a couple of functions in JavaScript that can be used to work with these descriptors. Let’s see them —
Object.getOwnPropertyDescriptor
We have seen this function above. It takes an existing object, and a property name. It returns either undefined
or an object containing the descriptors.
Object.defineProperty
It’s a static method on Object
that can define or modify a new property on a given object. It takes three arguments — the object, the property name, and descriptors. It returns the modified object. Let’s see an example —
It seems like a long example but it’s simple. Let’s go step by step —
- On #3, we use
defineProperty
passing itobj
, property nameid
and a descriptor containing only the [[Value]] field asvalue
having an actual value of42
. - Remember that if you fail to specify any more property attributes such as
enumerable
orconfigurable
, the default value is set tofalse
. In this case,id
haswritable
,enumerable
, andconfigurable
all set tofalse
. - On #7, we print
obj
. Sinceid
property is not enumerable, it is not printed. However, it does exist as #10 shows. - On #13, we specify the full descriptor set. We set
writable
tofalse
. - On #20 and #25, we print
name
property. However, we modify the value in between on line #23. Because of non-writability, it has no effect. Hence, we see the original value twice. - On #37, we try to delete
id
whoseconfigurable
is set tofalse
. Line #39 proves that it is not deleted. - On #42, we use the batch version of
defineProperty
calledObject.defineProperties
.
Object.defineProperty
sets one property at a time. There is a different variant of this function that can set multiple properties with their descriptors at the same time. It is aptly named Object.defineProperties
.
Protecting Objects
There are often cases where we don’t want anyone to tamper with our objects. Given the flexibility of JavaScript, it’s really easy to mistakenly re-assign properties of objects that we are not meant to touch. There are three major ways of protecting objects in JavaScript. Let’s look at them —
Object.preventExtensions
The Object.preventExtensions
method prevents new properties from ever being added to an object (i.e. prevents future extensions to the object). It takes an object and makes it non-extensible.
Note that the properties can be deleted though.
const obj = {
id: 42
};Object.preventExtensions(obj);obj.name = 'Arfat';console.log(obj);
// => { id: 42 }
You can check whether an object is non-extensible or not by using Object.isExtensible
. If it returns true
, you can add more properties to the object.
Object.seal
The seal
method seals an object. It means —
- It prevents new properties from being added just like
Object.preventExtensions
. - It marks all existing properties as non-configurable.
- Values of present properties can still be changed as long as they are writable.
- In short, it prevents adding and/or removing properties.
const obj = {
id: 42
};Object.seal(obj);delete obj.id
// (does not work)obj.name = 'Arfat';
// (does not work)console.log(obj);
// => { id: 42 }Object.isExtensible(obj);
// => falseObject.isSealed(obj);
//=> true
You can use Object.isSealed
to test whether an object has been sealed or not.
Object.freeze
freeze
is the maximum protection any object can have in JavaScript. It —
- seals the object using
Object.seal
- It also prevents modifying any existing properties at all.
- It also prevents the descriptors from being changed as the object is already sealed.
const obj = {
id: 42
};Object.freeze(obj);delete obj.id
// (does not work)obj.name = 'Arfat';
// (does not work)console.log(obj);
// // => { id: 42 }Object.isExtensible(obj);
// // => falseObject.isSealed(obj);
// //=> trueObject.isFrozen(obj);
// => true
You can check whether an object is frozen or not using the Object.isFrozen
function.
An important point to note is that these methods deal only with the direct properties of the objects. They do not modify nested objects.
Here’s a table summarizing the same.
The operations are in terms of properties. [CREATE] a new property, [READ] an existing property, [UPDATE] an existing property and [DELETE] an existing property.
Conclusion
Given how pervasive objects are on JavaScripts, it is important to understand the true power of objects. I hope the article was able to convey that effectively to you. I also hope that now you are able to answer the questions listed at the beginning of the article. Thanks for reading.