Notes on Object Reference in JavaScript. It can be quite troublesome sometimes!

Notes on Object Reference in JavaScript. It can be quite troublesome sometimes!

Daily short news for you
  • One thing I find a bit confusing is that Postman requires an internet connection to function, everyone. This has probably been the case for several years now; back then, many people complained about it, but it’s still the same, still needing the internet to work.

    They must think that IT folks always have internet access, which is why they added the requirement for a network connection just to be safe. Indeed, I am rarely or never affected since I always have internet while working. However, I believe many others find this issue extremely frustrating. Proof of that is that after this incident, quite a few open-source projects have been created to replace Postman.

    Among them, I see the most notable one is hoppscotch. The interface looks quite similar to Postman, but it seems "nicer" 😁

    » Read more
  • Have you ever tried reading the Google Analytics (GA) API documentation? The first time I read it, I was completely lost, not understanding what they were presenting, what the next steps should be... in general, the way it was presented was so convoluted that I couldn’t read it all at once to get immediate results.

    However, if you interact with it more, you'll find that it follows a certain standard. This means that the next time you read it, you'll know what to read first and what to read next, and then move on to the steps to get results, which makes you a bit more proficient. While I was fumbling around with the GA API, I discovered this website that helps me quickly test queries for results, and then I just need to copy the query and put it into the API. Quick and easy! 😅

    GA4 Query Explorer

    » Read more
  • Aha! I'm leaving this here so I don't forget. act is a tool that helps us run GitHub Actions on... our own computers.

    There are two reasons to do this. The first is speed; you can immediately check if the GitHub action configuration in the workflows file is correct instead of having to commit and then wait to see. The second is that you can completely build a system of actions that works right on your machine without needing to use GitHub Actions anymore 😆

    » Read more

The Problem

Object Reference is a concept that represents a reference variable, which means that instead of storing the actual value, it stores the memory address & operates with data based on that address. This helps save memory for applications. However, it also comes with its fair share of troubles.

Primitive Data Types

In JavaScript, we have "primitive" data types including: undefined, null, string, number, boolean, and symbol.

Variables initialized with these values are allocated a certain memory location that is not related to each other. For example:

let name = "estacks";
let name2 = name; // estacks
name = "estacks.icu";
console.log(name); // estacks.icu
console.log(name2); //estacks

Whenever the name variable changes its value, the variables that were assigned with name will still retain their original values. This also applies to data types like number, boolean, etc.

Object & Array

Unlike Primitive Types, when there are more than one variable created to store an object, array, function, they are all pointing to the same allocated memory location.

const arr1 = ['e', 's', 't', 'a', 'c', 'k', 's'];
const arr2 = arr1;
arr1[0] = 'a';
console.log(arr2); // ['a, 's', 't', 'a', 'c', 'k', 's']

In the example above, when the first element in arr1 is changed, arr2 is also changed accordingly. Why is that?

When arr1 is declared, memory is allocated and an address is stored by it. Then arr2 is assigned with arr1. Since arr1 is an array, instead of creating a new copy of that array, arr2 simply points to the same address that stores arr1. By doing so, any changes made to arr1 will also be reflected in arr2, and vice versa, because they both point to the same location. This applies to object and function as well.

Troubles

Forgetting about the reference nature of a variable

This is probably the most common case. Declaring a variable based on another variable without realizing that the variable has a reference nature.

This is a common mistake for beginners, so when you learn about this characteristic, try to avoid declaring a variable based on another variable and make a copy instead.

const person1 = {
  name: 'Nguyễn Văn A',  
  age: 20,  
  address: {
    city: 'Hà Nội',  
    district: 'Cầu Giấy',  
  }
};
const person2 = { ...person1 };

In the above example, I just copied person1 to person2. Try changing the value of name or age in person1 and person2 will not be affected. But if you change city or district, person2 will still change accordingly. This is because address is declared with an object value, so address still has a reference nature.

To copy an object using spread syntax (...) as shown above, or many other copying methods like using Object.assign can only shallow copy the object. To copy deeply nested objects like that and avoid any reference, you can use one of the following three methods:

The first method is using JSON.parse & JSON.stringify. This is the simplest and fastest way.

const person1 = {
  name: 'Nguyễn Văn A',  
  age: 20,  
  address: {
    city: 'Hà Nội',  
    district: 'Cầu Giấy',  
  }
};
const person2 = JSON.parse(JSON.stringify(person1));

However, this is also the worst method as parsing a string into an object is not good for performance, especially if the string is large.

The second method is writing code to perform deep copy:

function deepCopy(obj) {
    if(typeof obj !== 'object' || obj === null) {
        return obj;
    }

    if(obj instanceof Date) {
        return new Date(obj.getTime());
    }

    if(obj instanceof Array) {
        return obj.reduce((arr, item, i) => {
            arr[i] = deepCopy(item);
            return arr;
        }, []);
    }

    if(obj instanceof Object) {
        return Object.keys(obj).reduce((newObj, key) => {
            newObj[key] = deepCopy(obj[key]);
            return newObj;
        }, {})
    }
}

Finally, you can use built-in libraries that have deepCopy functions such as lodash, ramda... or use the clone package available on npm.

Using a shared object

Imagine you have a config used as a default if no specific config is found, you export an object that contains those configs, and it would be catastrophic if you accidentally change the value somewhere while using it.

// config.js file
module.export = {
  appName: "estacks",  
  connection: {
    host: "0.0.0.0",  
    port: 80,  
}
// app.js file
const conf = require('./config.js');

let config = findConfig(); // null
if (!config) config = conf;
...  
// accidentally changed conf
config.connection.port = 443;

At this point, in other files that have imported config.js, connection.port will all be changed to 443.

The solution for this issue is to deep clone the config object before using it. This will prevent any accidental changes that could lead to silly bugs and take a week to debug :D.

Conclusion

Understanding Object Reference is simply understanding the nature of a reference variable. When working with reference variables, you need to be extra careful to avoid making mistakes like I did above!

Premium
Hello

5 profound lessons

Every product comes with stories. The success of others is an inspiration for many to follow. 5 lessons learned have changed me forever. How about you? Click now!

Every product comes with stories. The success of others is an inspiration for many to follow. 5 lessons learned have changed me forever. How about you? Click now!

View all

Subscribe to receive new article notifications

or
* The summary newsletter is sent every 1-2 weeks, cancel anytime.

Comments (0)

Leave a comment...