Benchmark: spread vs Object.assign for reduce callback with large dataset

Tests:

with spread operator
const range = (from, to) => { const output = [] for(var x = from; x < to; x++){ output.push(x) } return output } range(0, 1000).reduce((acc, num) => { return { ...acc, [num]: num } }, {})
with object assign
const range = (from, to) => { const output = [] for(var x = from; x < to; x++){ output.push(x) } return output } range(0, 1000).reduce((acc, num) => { return Object.assign(acc, {[num]: num}) }, {})

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
with spread operator
with object assign

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated 27 days ago):

LLMs can make mistakes. Check important info.

Let's dive into the benchmark.

**What is being tested?**

MeasureThat.net is testing two approaches to merge an object with another object using different methods:

1. **Spread Operator (`...`)**: The first approach uses the spread operator to merge an object with a new property. Specifically, it takes an array created by `range(0, 1000)` and reduces it to create an object where each number is mapped to itself.
2. **Object Assign (`Object.assign()`)**: The second approach uses `Object.assign()` to merge two objects. Again, the input is an array created by `range(0, 1000)`, but this time it's being merged into an empty object.

**Options comparison**

The two approaches have different pros and cons:

* **Spread Operator (`...`)**:
	+ Pros: concise, efficient, and easy to read.
	+ Cons: not supported in older browsers, may cause issues with certain browser versions or configurations.
* **Object Assign (`Object.assign()`)**: More widely supported across browsers, but can be less readable and more verbose.

**Other considerations**

When using either approach, consider the following:

* In both cases, the `reduce()` method is used to create a new object. However, if you need to merge multiple arrays or objects, other methods (like `Object.assign()` with an array of properties) might be more suitable.
* If performance is critical, measuring the execution time can help determine which approach is better for your specific use case.

**Library and special JS features**

No libraries are used in these benchmarks. However, JavaScript does support some interesting syntax:

* **Destructuring**: The spread operator (`...`) uses destructuring to extract properties from an object.
* **Arrow functions**: Both benchmark definitions use arrow functions (`() => { ... }`), which provide a concise way to define small functions.

**Alternative approaches**

If you need more control or flexibility, consider these alternative approaches:

1. **Manual object merge using `for...in` loop**: Iterate over the properties of one object and add them to another.
```javascript
const obj1 = { a: 1, b: 2 };
const obj2 = {};

for (const key in obj1) {
  obj2[key] = obj1[key];
}
```
2. **Using `Object.fromEntries()`**: A newer method for creating objects from arrays of key-value pairs.
```javascript
const arr = [['a', 1], ['b', 2]];
const obj = Object.fromEntries(arr);
```

These alternatives may offer more control, but they might also be slower or less concise than the original approaches.

Latest run results:

Run details: (Test run date: 2 years ago)

User agent: Mozilla/5.0 (iPhone; CPU iPhone OS 15_4 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.4 Mobile/15E148 Safari/604.1

Browser/OS: Mobile Safari 15 on iOS 15.4

View result in a separate tab

Test name	Executions per second
with spread operator	14.6 Ops/sec
with object assign	854.8 Ops/sec

Related benchmarks:

spread vs Object.assign for reduce callback with large dataset (version: 0)

Comparing performance of: with spread operator vs with object assign

Created: 6 years ago by: Guest

Jump to the latest result

with spread operator

with object assign

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated 27 days ago):