@zerodep/struct-tree NPM

@zerodep/struct-tree

license

A factory function that returns an optionally-typed Tree data structure instance

Full documentation is available at the zerodep.app page.

Signature

declare const structTreeFactory: <T = unknown>(json?: TreeJson<T> | undefined) => Tree<T>;

interface Tree<T> {
  getRootNode: () => TreeNode<T> | null;
  setRootNode: (data: T) => TreeNode<T>;
  addChild: (node: TreeNode<T>, data: T) => TreeNode<T>;
  addChildren: (node: TreeNode<T>, data: T[]) => TreeNode<T>[];
  dfs: (comparator: (data: T) => boolean) => TreeNode<T> | null;
  bfs: (comparator: (data: T) => boolean) => TreeNode<T> | null;
  dfsFilter: (comparator: (data: T) => boolean) => TreeNode<T>[];
  bfsFilter: (comparator: (data: T) => boolean) => TreeNode<T>[];
  deleteNode: (node: TreeNode<T>) => void;
  toJSON: () => TreeJson<T> | null;
}

interface TreeNode<T> {
  data: T;
  parent: TreeNode<T> | null;
  children: TreeNode<T>[] | null;
}

type TreeJson<T> =
  | {
      data?: T;
      children?: TreeJson<T>[];
    }
  | {
      children?: TreeJson<T>[];
      [key: string]: any;
    };

Factory Parameters

The structTreeFactory accepts an optional json object with the following shape:

children - the array of children values of the node (may have their own children properties)
data - optional data of the node, will be used as the data source if provided
key:string: any - optional key:value tuples of data, will be used to fill the data of the node

Tree Methods

The structTreeFactory returns an object with the following methods:

getRootNode() - returns the root node of the tree
setRootNode() - sets and returns the root node of the tree
addChild(node, data) - add a child node to the specified node
addChildren(node, data) - add multiple child nodes to the specified node
dfs(comparator) - depth-first search using a provided comparator that returns the first match
bfs(comparator) - breadth-first search using a provided comparator that returns the first match
dfsFilter(comparator) - depth-first filtering of a tree returning all nodes that match the comparator
bfsFilter(comparator) - breadth-first filtering of a tree returning all nodes that match the comparator
deleteNode(node) - remove a node and all of its children from three
toJSON() - converts the tree to an object suitable for serialization

Tree Node Properties

The getRootNode(), dfs(), bfs(), dfsFilter() and bfsFilter() methods return node(s) that have the following properties:

data - the value of the node, may be anything (string, number, object, etc...)
parent - the reference/pointer to the parent node in the tree
children - the reference/pointer to the children nodes of the current node, if any

Examples

// ESM
import { structTreeFactory } from '@zerodep/struct-tree';

// CJS
const { structTreeFactory } = require('@zerodep/struct-tree');

Building a Tree

const tree = structTreeFactory<string>();

// populate the root node
const rootNode = tree.setRootNode('root'); // { "data": "root", "parent": null, "children": null }

// add children
const aNode = tree.addChild(rootNode, 'aaa'); // { "data": "aaa", "parent": [Object], "children": null }
const bNode = tree.addChild(rootNode, 'bbb'); // { "data": "bbb", "parent": [Object], "children": null }

// add grandchildren
tree.addChild(aNode, 'aaa111');
tree.addChild(aNode, 'aaa222');
tree.addChild(bNode, 'bbb333');
tree.addChild(bNode, 'bbb444');
tree.addChild(bNode, 'bbb555');

// serialization
const json = tree.toJSON();
// {
//   "data": "root",
//   "children": [
//     {
//       "data": "aaa",
//       "children": [
//         { "data": "aaa111" },
//         { "data": "aaa222" },
//       ]
//     },
//     {
//       "data": "bbb",
//       "children": [
//         { "data": "bbb333" },
//         { "data": "bbb444" },
//         { "data": "bbb555" }
//       ]
//     }
//   ]
// }

Populating and Searching a Tree

const json: TreeJson<string> = {
  data: 'root',
  children: [
    {
      data: 'aaa',
      children: [{ data: 'aaa111' }, { data: 'aaa222' }],
    },
    {
      data: 'bbb',
      children: [{ data: 'bbb111' }, { data: 'bbb222' }, { data: 'bbb333' }],
    },
  ],
};
const tree = structTreeFactory<string>(json);

// depth-first search
const comparator1 = (data: string) => data.endsWith('222');
const match1 = tree.dfs(comparator1); // { "data": "aaa222", "parent": [Object], "children": null }

// breadth-first search
const comparator2 = (data: string) => data.endsWith('333');
const match2 = tree.bfs(comparator2); // { "data": "bbb333", "parent": [Object], "children": null }

// depth-first filter
const comparator3 = (data: string) => data.startsWith('aaa');
const matches3 = tree.bfs(comparator3);
// [
//   { "data": "aaa", "parent": [Object], "children": [[Object], [Object]] }
//   { "data": "aaa111", "parent": [Object], "children": null }
//   { "data": "aaa222", "parent": [Object], "children": null }
// ]

// breadth-first filter
const comparator4 = (data: string) => data.startsWith('bbb');
const matches4 = tree.bfs(comparator4);
// [
//   { "data": "bbb", "parent": [Object], "children": [[Object], [Object], [Object]] }
//   { "data": "bbb111", "parent": [Object], "children": null }
//   { "data": "bbb222", "parent": [Object], "children": null }
//   { "data": "bbb333", "parent": [Object], "children": null }
// ]

ZeroDep Advantages

Zero npm dependencies - completely eliminates all risk of supply-chain attacks, decreases node_modules folder size
ESM & CJS - supports both ECMAScript modules and common JavaScript exports
Tree Shakable - built to be fully tree shakable ensuring your packages are the smallest possible size
Fully Typed - typescript definitions are provided/built-in to every package for a superior developer experience
Semantically Named - package and method names are easy to grok, remember, use, and read
Documented - actually useful documentation with examples at zerodep.app
Intelligently Packaged - multiple npm packages of different sizes available allowing a menu or a-la-carte composition of capabilities
100% Tested - all methods and packages are fully unit tested
Predictably Versioned - semantically versioned for peace-of-mind upgrading, valuable changelogs for understand changes
MIT Licensed - permissively licensed for maximum usability