What is structuredClone()?
It is a method that creates a deep copy of a given value using the structured clone algorithm and without using a 3rd-party library like Lodash. This makes it much easier for developers to transfer data between different applications without having to worry about compatibility issues.
Before we go on structuredClone, let's learn about the differences between shallow copy and deep copy!
Shallow Copy vs Deep Copy
There are 2 types of clone methods, shallow copy and deep copy. Let's see what are the differences between them.
Shallow Copy
A shallow copy
creates a reference to the original object rather than creating a new copy of the object's data. You can use spread operator
or Object.assign
or clone()
from Lodash.
const original = {
hello: "world",
motto: "it is so awesome",
address: {
province: "Awesome Province",
},
};
const copied = { ...original };
console.log("original", original);
console.log("copied", copied);
original.motto = "This only change the ____original____ object";
As long as the properties are 1 level deep when the change is made to either original
or copied
, they keep their values.
original.motto = "This only change the ____original____ object";
copied.motto = "This only change the ____copied____ object";
As the screenshot shows, both mottos are changed, and they don't affect each other.
original.address.province = "Original Province, this changes both objects";
The province property is 2 levels deep and remembers a shallow copy creates an object with the same reference to the original object. This is why when the province is changed, both objects will have the same value.
Deep Copy
A deep copy
creates a completely new copy of the entire object, including any nested objects, and any changes to the copied object will not affect the original object. There are 2 ways, use JSON.parse(JSON.stringify(obj))
or cloneDeep()
from Lodash. Now, structuredClone()
can be used.
const obj = {
hello: "world",
a: {
b: "b",
c: "c",
},
};
const original = {
a: new Date(),
b: NaN,
c: new Function(),
d: undefined,
e: function () {},
f: Number,
g: false,
h: Infinity,
i: obj,
};
The JSON.parse(JSON.stringify(obj))
has unexpected consequences. The original
object will lose any JavaScript property that has no equivalent type in JSON, like Function or Infinity.
const cloned = JSON.parse(JSON.stringify(original));
Now, structuredClone
can be used to copy objects. Keep in mind, it doesn't support Function
and there are supported types by the structured clone algorithm such as Array, Boolean and more. You can read more on the algorithm at this link.
const obj = {
hello: "world",
a: {
b: "b",
c: "c",
},
};
// Functions are not supported, which is why they are removed from this.
const original = {
a: new Date(),
b: NaN,
d: undefined,
g: false,
h: Infinity,
i: obj,
j: [1, 2, 3, 4, 5],
};
const cloned = structuredClone(original);
original.i.a.b = "hey";
cloned.i.a.b = "hey, I have changed!";
The b
had "b" as the value before. With structuredClone()
, both original
and copied
can have their values without affecting each other, unlike the shallow copy.
Recap
A shallow copy of an object means that the copy has the same references to the properties of the original object. As a result, if one of the properties at two levels deep is modified, both the original and the copy will have the same new value. On the other hand, in a deep copy, the copy does not share references with the original object. This means that if a deeply nested property is changed, only that particular object will be affected and not the other.
It is advisable to use the structuredClone()
method for creating a deep copy of an object, as it makes sure that the data types are preserved during the copying process. In contrast, using JSON.parse(JSON.stringify(obj))
results in data type changes. If you prefer using Lodash, keep in mind that additional steps are required.
Resources
- MDN Docs - structuredClone()
- MDN Docs - structured clone algorithm
- MDN Docs - Shallow copy
- MDN Docs - Deep copy
- Lodash - clone() and cloneDeep()
Thank you!
Thank you for your time and for reading this!
Originally published at victoriacheng15.vercel.app