Perfect use case for recursion, which could handle even deeper structure:
function flatten(ary) {
var ret = [];
for(var i = 0; i < ary.length; i++) {
if(Array.isArray(ary[i])) {
ret = ret.concat(flatten(ary[i]));
} else {
ret.push(ary[i]);
}
}
return ret;
}
flatten([[[[[0]], [1]], [[[2], [3]]], [[4], [5]]]]) // [0, 1, 2, 3, 4, 5]
Alternatively, as an Array method:
Array.prototype.flatten = function() {
var ret = [];
for(var i = 0; i < this.length; i++) {
if(Array.isArray(this[i])) {
ret = ret.concat(this[i].flatten());
} else {
ret.push(this[i]);
}
}
return ret;
};
[[[[[0]], [1]], [[[2], [3]]], [[4], [5]]]].flatten() // [0, 1, 2, 3, 4, 5]
EDIT #1: Well, think it a little bit functional way (except for the named recursion which should be using Y-combinator for pure functional :D).
function flatten(ary) {
return ary.reduce(function(a, b) {
if (Array.isArray(b)) {
return a.concat(flatten(b))
}
return a.concat(b)
}, [])
}
Let's adopt some ES6 syntax which makes it even shorter, in one line.
const flatten = (ary) => ary.reduce((a, b) => a.concat(Array.isArray(b) ? flatten(b) : b), [])
But remember, this one cannot be applied as an array method, because arrow functions don't have theirs own this
.
EDIT #2: With the latest Array.prototype.flat
proposal this is super easy. The array method accepts an optional parameter depth
, which specifies how deep a nested array structure should be flattened (default to 1
).
[[[[[0]], [1]], [[[2], [3]]], [[4], [5]]]].flat() // [[[[0]], [1]], [[[2], [3]]], [[4], [5]]]
[[[[[0]], [1]], [[[2], [3]]], [[4], [5]]]].flat(2) // [[[0]], [1], [[2], [3]], [4], [5]]
[[[[[0]], [1]], [[[2], [3]]], [[4], [5]]]].flat(3) // [[0], 1, [2], [3], 4, 5]
[[[[[0]], [1]], [[[2], [3]]], [[4], [5]]]].flat(4) // [0, 1, 2, 3, 4, 5]
So to flatten an array of arbitrary depth, just call flat
method with Infinity
.
[[[[[0]], [1]], [[[2], [3]]], [[4], [5]]]].flat(Infinity) // [0, 1, 2, 3, 4, 5]