wysiwyg is a set of editor commands that can be used to build a full wysiwyg editor.
Check the demo to see these commands in action.
To use as a component:
$ component install adamsanderson/wysiwyg
Assuming you have an element on the page being edited with contenteditable:
var el = document.getElementById('content');
el.setAttribute('contenteditable', true);
You can use the functions exposed by wysiwyg to modify the state of the edited text:
var wysiwyg = require('wysiwyg');
wysiwyg.commands.bold(); // Make the text at the cursor bold
wysiwyg.commands.h1(); // Convert the block element to a header
wysiwyg.commands.ul(); // Now convert it to an unordered list
wysiwyg.commands.indent(); // Indent the list
You can also query the state of the edited content:
wysiwyg.state.bold(); // Returns true if the cursor is in bold text
wysiwyg.state.h1(); // Returns true if the block is a header
Commands
Commands modify the content being edited.
bold, italic, strike, underline, sub, and sup will format text. Meanwhile, clear will remove all formatting from the current selection.
h1, h2, h3, h4, h5, h6, p, div, and blockquote with change the current block. So if you have the following markup:
<p>This is just <b>great</b></p>
Then calling wysiwyg.commands.h1() will result in:
<h1>This is just <b>great</b></h1>
ol and ul will turn a line into a list with that line being the first list item. If the line is already a list item, it will revert the line.
indent and outdent will indent or outdent lists. When used on other elements, it will typically indent using blockquote elements.
hr, br, img, and html will insert their respective elements. img should be called with the url of the image. html can be used to insert arbitrary html at the cursor.
States
States let you determine how the text being edited is styled. These are useful for styling toolbar buttons.
bold, italic, strike, underline, sub, and sup can all be called to determine the formatting the text at the cursor. Each one will return true or false depending on if the style applies to the text.
h1, h2, h3, h4, h5, h6, p, div, and blockquote will return true if the current block of text is of that type. These are mutually exclusive.
Much of this depends on the browser's implementation of the contenteditable commands, different browsers may mark up content differently. For more details see Mozilla's documentation on Rich Text Editing
For the best results, wrap the content you want to edit in a div. If you edit paragraphs directly, some browsers will not allow you to insert lists.
If possible, style the element being edited with white-space: pre-wrap. This allows browsers to preserve white space while editing, and avoids some odd behavior around inserting all over the place.
To work with the user's text caret, for instance to enable and disable commands, you may want to also use caret.
MIT