I was studying the jQuery source when I found this (v1.5 line 2295):
namespace = new RegExp("(^|\\.)" +
jQuery.map( namespaces.slice(0).sort(), fcleanup ).join("\\.(?:.*\\.)?") + "(\\.|$)");
My question is, why use slice(0) here?
sort() modifies the array it's called on - and it isn't very nice to go around mutating stuff that other code might rely on.
slice() always returns a new array - the array returned by slice(0) is identical to the input, which basically means it's a cheap way to duplicate an array.
arr.slice(0) makes a copy of the original array by taking a slice from the element at index 0 to the last element.
It was also used to convert array-like objects into arrays. For example, a DOM NodeList (returned by several DOM methods like getElementsByTagName) is not an array, but it is an array-like object with a length field and is indexable in JavaScript. To convert it to an array, one often used:
var anchorArray = [].slice.call(document.getElementsByTagName('a'), 0)
Now, ES2015's spread syntax is used to convert iterable objects, including NodeList and HTMLCollection, to arrays:
const anchorArray = [...document.getElementsByTagName('a')]
slice(0) creates a new array identical to the original array. Many times you want to preserve your original array and create a new one.
If you use slice(1), it will create a different array starting from index position 1.
Similar things hold for strings as well.