A simple and fast zero-dependencies-library to transform text-selections into serializable markings.
marklib can be installed with npm
or bower
.
npm install --save-dev marklib
bower install marklib --save
// obtain a selection from document
var selection = document.getSelection();
// create a new rendering based on the current document
var renderer = new Marklib.Rendering(document, options, context)
renderer.setId('myRenderId') // if an ID is not provided, a autogenerated one will be used
// renders the given selection and returns a result (`RenderResult`).
var result = renderer.renderWithRange(selection.getRangeAt(0));
Important: After a Rendering has been used to render a selection/serialized result,
it can't be used to render something again. You need to create a new Instance of Rendering
.
You can pass options to each rendering instance, the following shows the default options
var renderer = new Marklib.Rendering(document, {
hoverClass: 'marklib--hover',
treeClass: 'marklib--tree',
// Supports arrays and/or strings
className: ['marking']
});
Marklib triggers events that can be listened to with instance.on('event-name')
. Events are build with
wolfy87-eventemitter
(https://github.com/Olical/EventEmitter). The following Events are available:
Before you can actually receive events, you need to register the event handler with registerEvents
(use import { registerEvents } from 'marklib/src/main/RenderingEvents';
on your application bootstrap code.)
Event-Name | Description | Arguments |
---|---|---|
click |
triggered when clicked on a marking. | (originalEvent, instanceHierarchy) |
hover-enter |
triggered when a pointer-device starts hovering over a marking | (originalEvent, instanceHierarchy) |
hover-leave |
triggered when a pointer-device leaves a marking | (originalEvent, instanceHierarchy) |
Additionally, marklib will add hover classes to the current hovered marking.
-
HTMLDocument
document -> the document instance used
-
Object
[options], optional -> an object containing setting for marklib (see Options)
-
HTMLElement
[context], optional -> the context used to serialize / deserialize the rendering, if not given the document instance.
A Serialized results consist of 2 strings (start end end) in the following form
'body>section;0;1'`
-▲------------▲-▲
- ▲ The first part defines a css-selector (queryable with document.querySelector).
- ▲ The second part defines the text-node inside the given selector
- ▲ The third part defines the string-offset inside this text-node
// This is the result we get from `RenderResult#serialize()`
var result = {
startContainerPath: 'body>section;0',
endContainerPath: 'body>section;1',
startOffset: 2,
endOffset: 5
}
var rendering = new Marklib.Rendering(document);
rendering.renderWithResult(result);
- Annotations
- Collaboration tools
- Inline-Commenting (I actually started a project that will do something like this: https://github.com/BowlingX/commentp)
npm run develop
or npm run tdd
(to start karma in watch mode)
The MIT License (MIT)
Copyright (c) 2015 David Heidrich
Any contribution is welcome, just issue a pull-request or bug/feature if you found something :)