Convert DOM trees into compact JSON objects, and vice versa, as fast as possible.

• Description
• Installation
• Demos
• Usage
• FilterLists
• API
• Performance
• Tests
• Contributing
• License

The purpose of domJSON is to create very accurate representations of the DOM as JSON, and to do it very quickly. While there are probably dozens of viable use cases for this project, I've made two quick demos to showcase the library's versatility. The first simply makes a copy of a given branch of the DOM tree, which could be useful for end user bug logging, state tracking, etc. The second demo does a batch update of a large number of DOM Nodes, but much more performantly than the "traditional" jQuery select > .each() > update pattern.

Broadly speaking, the goals of this project are:

• Provide as accurate a copy of a given node's DOM properties as possible, but allow option filtering to remove useless information
• Be able to rebuild JSON nodes into DOM nodes as performantly as possible
• Speed, speed, speed
• No frivolous data: produce JSON objects that are as compact as possible, removing all information not relevant to the developer
• Keep the library lightweight, with no dependencies

DomJSON works in the following browsers (mobile and desktop versions supported):

• Chrome 39+
• Firefox 24+
• Safari 7+
• IE 9+

Installing domJSON is easy. You can pull it from Bower...

bower install domjson

...or grab it from NPM and manually include it as a script tag...

npm install domjson --save

or just download this repo manually and include the file as a dependency.

<script src="./lib/domJSON.js"></script>

Coming soon...

Using domJSON is super simple: use the .toJSON() method to create a JSON representation of the DOM tree:

var someDOMElement = document.getElementById('sampleId');
var jsonOutput = domJSON.toJSON(myDiv);

And then rebuild the DOM Node from that JSON using .toDOM():

var DOMDocumentFragment = domJSON.toDOM(jsonOutput);
someDOMElement.parentNode.replaceChild(someDOMElement, DOMDocumentFragment);

When creating the JSON object, there are many precise options available, ensuring that developers can produce very specific and compact outputs. For example, the following will produce a JSON copy of someDOMElement's DOM tree that is only two levels deep, contains no "offset*," "client*," or "scroll*" type DOM properties, only keeps the "id" attribute on each DOM Node, and outputs a string (rather than a JSON-friendly object):

var jsonOutput = domJSON.toJSON(myDiv, {
	attributes: ['id'],
	domProperties: {
		exclude: true,
		values: ['clientHeight', 'clientLeft', 'clientTop', 'offsetWidth', 'offsetHeight', 'offsetLeft', 'offsetTop', 'offsetWidth', 'scrollHeight', 'scrollLeft', 'scrollTop', 'scrollWidth']
	},
	deep: 2,
	stringify: true
});

A FilterList is a custom type for certain options passed to domJSON's .toJSON() method. It allows for very granular control over which fields are included in the final JSON output, allowing developers to eliminate useless information from the result, and produce extremely compact JSON objects. It operates based on boolean logic: an inclusive FilterList allows the developer to explicitly specify which fields they would like to see in the output, while an exclusive FilterList allows them to specify which fields they would like to omit (e.g., "Copy every available field except X, Y, and Z"). The FilterList accepts an object as an argument, or a shorthand array (which is the recommended style, since it's much less verbose).

Take this example: suppose we have a single div that we would like to convert into a JSON object using domJSON. It looks like this in its native HTML:

<div id="myDiv" class="testClass" style="margin-top: 10px;" data-foo="bar" data-quux="baz">
	This is some sample text.
</div>

Let's try and make a JSON object out of this field, but only include the class and style attributes, and only the offsetTop and offsetLeft DOM properties. Both of the following inputs will have the same output:

var myDiv = document.getElementById('myDiv');

//Using object notation for our filterLists
var jsonOutput = domJSON.toJSON(myDiv, {
	attributes: {
		values: ['class', 'style']
	},
	domProperties: {
		values: ['offsetLeft', 'offsetTop']
	},
	metadata: false
});

//Same thing, using the array notation
var jsonOutput = domJSON.toJSON(myDiv, {
	attributes: [false, 'class', 'style'],
	domProperties: [false, 'offsetLeft', 'offsetTop'],
	metadata: false
});

The result:

{
	"attributes": {
		"class": "testClass",
		"style": "margin-top: 10px;"
	},
	"childNodes": [{
		"nodeType": 3,
		"nodeValue": "This is some sample text"
	}],
	"nodeType": 1,
	"nodeValue": "This is some sample text",
	"offsetLeft": 123,
	"offsetTop": 456,
	"tagName": "DIV"
}

The array notation is much shorter, and very easy to parse: the first value is a boolean that determines whether the list is exclusive or not, and the remaining elements are just the filter values themselves. Here is an example for making an exclusive list of attributes on the same div as above:

var myDiv = document.getElementById('myDiv');

//Using object notation for our filterLists, but this time specify values to EXCLUDE
var jsonOutput = domJSON.toJSON(myDiv, {
	attributes: {
		exclude: true,
		values: ['class', 'style']
	},
	domProperties: {
		values: ['offsetLeft', 'offsetTop']
	},
	metadata: false
});

//Same thing, using the array notation
var jsonOutput = domJSON.toJSON(myDiv, {
	attributes: [true, 'class', 'style'],
	domProperties: [false, 'offsetLeft', 'offsetTop'],
	metadata: false
});

The result:

{
	"attributes": {
		"id": "myDiv",
		"data-foo": "bar",
		"data-quux": "baz"
	},
	"childNodes": [{
		"nodeType": 3,
		"nodeValue": "This is some sample text"
	}],
	"nodeType": 1,
	"nodeValue": "This is some sample text",
	"offsetLeft": 123,
	"offsetTop": 456,
	"tagName": "DIV"
}

domJSON : object

domJSON is a global variable to store two methods: .toJSON() to convert a DOM Node into a JSON object, and .toDOM() to turn that JSON object back into a DOM Node

FilterList : Object | Array

An object specifying a list of fields and how to filter it, or an array with the first value being an optional boolean to convey the same information

domJSON is a global variable to store two methods: .toJSON() to convert a DOM Node into a JSON object, and .toDOM() to turn that JSON object back into a DOM Node

Kind: global namespace

• domJSON : object
  • .toJSON(node, [opts]) ⇒ Object | string
  • .toDOM(obj, [opts]) ⇒ DocumentFragment

Take a DOM node and convert it to simple object literal (or JSON string) with no circular references and no functions or events

Kind: static method of domJSON
Returns: Object | string - A JSON-friendly object, or JSON string, of the DOM node -> JSON conversion output
Todo

• {boolean|FilterList} [opts.parse=false] a FilterList of properties that are DOM nodes, but will still be copied PLANNED

Take the JSON-friendly object created by the .toJSON() method and rebuild it back into a DOM Node

Kind: static method of domJSON
Returns: DocumentFragment - A DocumentFragment (nodeType 11) containing the result of unpacking the input obj

An object specifying a list of fields and how to filter it, or an array with the first value being an optional boolean to convey the same information

Kind: global typedef
Properties

A major goal of this library is performance. That being said, there is one way to significantly slow it down: setting opts.computedStyle to true. This forces the browser to run window.getComputedStyle() on every node in the DOM Tree, which is really, really slow, since it requires a redraw each time it does it! Obviously, there are situations where this you need the computed style and this performance hit is unavoidable, but otherwise, keep opts.computedStyle set to false. Besides that, I'm working on writing some benchmark tests to give developers an idea of how each option affects the speed of domJSON, but this will take some time!

Generally speaking, avoid using FilterList type options for best performance. The optimal settings in terms of speed, without sacrificing any information, are as follows:

var jsonOutput = domJSON.toJSON(someNode, {
	absolutePaths: false,
	attributes: false,
});

A downside of the above setup is that you may well end up with a very bloated output containing a lot of frivolous fields. As of now, this is the cost of getting a quick turnaround - in the future, domJSON will be optimized to work more quickly for more precise setups.

You can give the test suite for domJSON a quick run through in the browser of your choice here. You can also view results from local Chrome tests, or the entire browser compatibility suite. Please note that Rawgit caches the tests after you run them the first time, so if something seems off, clear your cache!

Feel free to pull and contribute! If you do, please make a separate branch on your Pull Request, rather than pushing your changes to the Master. It would also be greatly appreciated if you ran the appropriate tests before submitting the request (there are three sets, listed below).

For unit testing the Chrome browser, which is the most basic target for functionality, type the following in the CLI:

gulp unit-chrome

To record the code coverage after your changes, use:

gulp coverage

And, if you have them all installed and are feeling so kind, you can also do the entire browser compatibility suite (Chrome, Canary, Firefox ESR, Firefox Developer Edition, IE 11, IE 10):

gulp unit-browsers

If you make changes that you feel need to be documented in the readme, please update the relevant files in the /docs directory, then run:

gulp docs

##License

The MIT License (MIT)

Copyright (c) 2014 Alex Zaslavsky

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.