Here we will introduce another tree traversal method: breadth-first search. In contrast to the depth-first search methods from the last challenge, breadth-first search explores all the nodes in a given level within a tree before continuing on to the next level. Typically, queues are utilized as helper data structures in the design of breadth-first search algorithms.
In this method, we start by adding the root node to a queue. Then we begin a loop where we dequeue the first item in the queue, add it to a new array, and then inspect both its child subtrees. If its children are not null, they are each enqueued. This process continues until the queue is empty.
# --instructions--
Let's create a breadth-first search method in our tree called `levelOrder`. This method should return an array containing the values of all the tree nodes, explored in a breadth-first manner. Be sure to return the values in the array, not the nodes themselves. A level should be traversed from left to right. Next, let's write a similar method called `reverseLevelOrder` which performs the same search but in the reverse direction (right to left) at each level.
# --hints--
The `BinarySearchTree` data structure should exist.
```js
assert(
(function () {
var test = false;
if (typeof BinarySearchTree !== 'undefined') {
test = new BinarySearchTree();
}
return typeof test == 'object';
})()
);
```
The binary search tree should have a method called `levelOrder`.
```js
assert(
(function () {
var test = false;
if (typeof BinarySearchTree !== 'undefined') {
test = new BinarySearchTree();
} else {
return false;
}
return typeof test.levelOrder == 'function';
})()
);
```
The binary search tree should have a method called `reverseLevelOrder`.