bfs-tree-layout

Operations on nodes for balanced binary trees stored in bfs layout. These are useful if you are building data structures, like binary search trees, implicitly (ie not storing pointers to subtrees).

BFS layout, unlike inorder layout, is much faster for performing binary searches and bounds queries. The disadvantage is that it is much slower at inorder traversal.

Install

npm install bfs-tree-layout

Example

Assume that the tree is filled in level order, and laid out in memory via an bfs traversal. For example:

The tree:

          0
        /   \
       1     2
      / \   / \
     3   4 5   6
    / \  |
   7   8 9

With this picture in mind, here are some operations:

var layout = require("bfs-tree-layout")
 
console.log(layout.left(10, 3))     //Prints:  7
 
console.log(layout.parent(10, 4))   //Prints:  1
 
console.log(layout.height(10, 2))   //Prints:  1
 

API

var layout = require("bfs-tree-layout")

Conventions:

• n is always the size of the tree
• x is the index of a node in the tree

layout.root(n)

Returns the index of the root of a tree of size n.

layout.begin(n)

Returns the index of the first node of the tree

layout.end(n)

Returns the index of the last node in the tree

layout.height(n, x)

Returns the height of node x in a tree of size n

layout.prev(n, x)

Returns the predecessor of x in an in-order traversal

layout.next(n, x)

Returns the successor of x in an in-order traversal

layout.parent(n, x)

Returns the parent of x in a tree of size n

layout.left(n, x)

Returns the left child of x

layout.right(n, x)

Returns the right child of x

layout.leaf(n, x)

Returns true if the node x is a leaf

layout.lo(n, x)

Returns the left-most ancestor of x

layout.hi(n, x)

Returns the right-most ancestor of x

Credits

(c) 2013 Mikola Lysenko. MIT License