Array Atom feed

Prototype extends all native Javascript arrays with quite a few powerful methods.

This is done in two ways:

  • It mixes in the Enumerable module, which brings a ton of methods in already.
  • It adds quite a few extra methods, which are documented in this section.

With Prototype, arrays become much, much more than the trivial objects we were used to manipulate, limiting ourselves to using their length property and their [] indexing operator. They become very powerful objects, that greatly simplify the code for 99% of the common use cases involving them.

Why you should stop using forin to iterate (or never take it up)

Many JavaScript authors have been misled into using the forin JavaScript construct to loop over array elements. This kind of code just won’t work with Prototype.

You see, the ECMA 262 standard, which defines ECMAScript 3rd edition, supposedly implemented by all major browsers including MSIE, defines numerous methods on Array (§15.4.4), including such nice methods as concat, join, pop and push, to name but a few among the ten methods specified.

This same standard explicitely defines that the forin construct (§12.6.4) exists to enumerate the properties of the object appearing on the right side of the in keyword. Only properties specifically marked as non-enumerable are ignored by such a loop. By default, the prototype and the length properties are so marked, which prevents you from enumerating over array methods when using forin. This comfort led developers to use forin as a shortcut for indexing loops, when it is not its actual purpose.

However, Prototype has no way to mark the methods it adds to Array.prototype as non-enumerable. Therefore, using forin on arrays when using Prototype will enumerate all extended methods as well, such as those coming from the Enumerable module, and those Prototype puts in the Array namespace (described in this section, and listed further below).

What is a developer to do?

You can revert to vanilla loops:

for (var index = 0; index  myArray.length; ++index) {
  var item = myArray[index];
  // Your code working on item here...

Or you can use iterators, such as each :

myArray.each(function(item) {
  // Your code working on item here...

This side-effect enforcement of the true purpose of forin is actually not so much of a burden: as you’ll see, most of what you used to loop over arrays for can be concisely done using the new methods provided by Array or the mixed-in Enumerable module. So manual loops should be fairly rare.

A note on performance

Should you have a very large array, using iterators with lexical closures (anonymous functions that you pass to the iterators, that get invoked at every loop iteration), such as each, or relying on repetitive array construction (such as uniq), may yield unsatisfactory performance. In such cases, you’re better off writing manual indexing loops, but take care then to cache the length property and use the prefix ++ operator:

// Custom loop with cached length property: maximum full-loop performance on very large arrays!
for (var index = 0, len = myArray.length; index  len; ++index) {
  var item = myArray[index];
  // Your code working on item here...



clear() -> Array

Clears the array (makes it empty).


clone() -> newArray

Returns a duplicate of the array, leaving the original array intact.


compact() -> newArray

Returns a new version of the array, without any null/undefined values.


each(iterator) -> Array

Iterates over the array in ascending numerical index order.


first() -> value

Returns the first item in the array, or undefined if the array is empty.


flatten() -> newArray

Returns a “flat” (one-dimensional) version of the array. Nested arrays are recursively injected “inline.” This can prove very useful when handling the results of a recursive collection algorithm, for instance.


Array.from(iterable) -> actualArray

Clones an existing array or creates a new one from an array-like collection.

This is an alias for the $A() method. Refer to its page for complete description and examples.


indexOf(value) -> position

Returns the position of the first occurrence of the argument within the array. If the argument doesn’t exist in the array, returns -1.


inspect() -> String

Returns the debug-oriented string representation of an array.


last() -> value

Returns the last item in the array, or undefined if the array is empty.


reduce() -> Array | singleValue

Reduces arrays: one-element arrays are turned into their unique element, while multiple-element arrays are returned untouched.


reverse([inline = true]) -> Array

Returns the reversed version of the array. By default, directly reverses the original. If inline is set to false, uses a clone of the original array.


size() -> Number

Returns the size of the array.


toArray() -> newArray

This is just a local optimization of the mixed-in toArray from Enumerable.


toJSON() -> String

Returns a JSON string.


uniq() -> newArray

Produces a duplicate-free version of an array. If no duplicates are found, the original array is returned.


without(value...) -> newArray

Produces a new version of the array that does not contain any of the specified values.