Squire – 简洁的 HTML5 富文本编辑器Squire-master.zip

Squire ====== Squire is an HTML5 rich text editor, which provides powerful cross-browser normalisation, whilst being supremely lightweight and flexible. It is built for the present and the future, and as such does not support truly ancient browsers. It should work fine back to around Opera 12, Firefox 3.5, Safari 5, Chrome 9 and IE9. An example UI integration can be tried at http://neilj.github.io/Squire/. Unlike other HTML5 rich text editors, Squire was written as a component for writing documents (emails, essays, etc.), not doing wysiwyg websites. If you are looking for support for inserting form controls or flash components or the like, you'll need to look elsewhere. However for many purposes, Squire may be just what you need, providing the power without the bloat. The key features are: ### Lightweight ### * Only 11.5KB of JS after minification and gzip (35KB before gzip). * Does not include its own XHR wrapper, widget library or lightbox overlays. * No dependencies. * No UI for a toolbar is supplied, allowing you to integrate seamlessly with the rest of your application and lose the bloat of having two UI toolkits loaded. Instead, you get a component you can drop in in place of a `<textarea>` and manipulate programatically. ### Powerful ### Squire provides an engine that handles the heavy work for you, making it easy to add extra features. With the `changeFormat` method you can easily add or remove any inline formatting you wish. And the `modifyBlocks` method can be used to make complicated block-level changes in a relatively easy manner. If you need more commands than in the simple API, I suggest you check out the source code (it's not very long), and see how a lot of the other API methods are implemented in terms of these two methods. The general philosophy of Squire is to allow the browser to do as much as it can (which unfortunately is not very much), but take control anywhere it deviates from what is required, or there are significant cross-browser differences. As such, the `document.execCommand` method is not used at all; instead all formatting is done via custom functions, and certain keys, such as 'enter' and 'backspace' are handled by the editor. Installation and usage ---------------------- 1. Copy the contents of the `build/` directory onto your server. 2. Edit the `<style>` block in document.html to add the default styles you would like the editor to use (or link to an external stylesheet). 3. In your application, instead of a `<textarea>`, use an `<iframe src="path/to/document.html">`. 4. In your JS, attach an event listener to the `load` event of the iframe. When this fires you can grab a reference to the editor object through `iframe.contentWindow.editor`. 5. Use the API below with the `editor` object to set and get data and integrate with your application or framework. Advanced usage -------------- If you load the library into a top-level document (rather than an iframe), it will not turn the page into an editable document, but will instead add a function named `Squire` to the global scope. Call `new Squire( document )`, with the `document` from an iframe to instantiate multiple rich text areas on the same page efficiently. ### Setting the default block style By default, the editor will use a `<div>` for blank lines, as most users have been conditioned by Microsoft Word to expect <kbd>Enter</kbd> to act like pressing <kbd>return</kbd> on a typewriter. If you would like to use `<p>` tags (or anything else) for the default block type instead, then after calling `var editor = new Squire( document )` (or getting your reference to the ready-made `editor` instance if using the simple setup), set `editor.defaultBlockTag = 'P';`. You can also set an object of attributes to apply to each default block node by setting the *defaultBlockProperties* property, e.g. `editor.defaultBlockProperties = { style: 'font-size: 16px;' }`. License ------- Squire is released under the MIT license. See LICENSE for full license. API --- ### addEventListener Attach an event listener to the editor. The handler can be either a function or an object with a `handleEvent` method. This function or method will be called whenever the event fires, with an event object as the sole argument. The following events may be observed: * **focus**: The editor gained focus. * **blur**: The editor lost focus * **keydown**: Standard DOM keydown event. * **keypress**: Standard DOM keypress event. * **keyup**: Standard DOM keyup event. * **input**: The user inserted, deleted or changed the style of some text; in other words, the result for `editor.getHTML()` will have changed. * **pathChange**: The path (see getPath documentation) to the cursor has changed. The new path is available as the `path` property on the event object. * **select**: The user selected some text. * **undoStateChange**: The availability of undo and/or redo has changed. The event object has two boolean properties, `canUndo` and `canRedo` to let you know the new state. * **willPaste**: The user is pasting content into the document. The content that will be inserted is available as the `fragment` property on the event object. You can modify this fragment in your event handler to change what will be pasted. You can also call the `preventDefault` on the event object to cancel the paste operation. The method takes two arguments: * **type**: The event to listen for. e.g. 'focus'. * **handler**: The callback function to invoke Returns self (the Squire instance). ### removeEventListener Remove an event listener attached via the addEventListener method. The method takes two arguments: * **type**: The event type the handler was registered for. * **handler**: The handler to remove. Returns self (the Squire instance). ### focus Focuses the editor. The method takes no arguments. Returns self (the Squire instance). ### blur Removes focus from the editor. The method takes no arguments. Returns self (the Squire instance). ### getDocument Returns the `document` object of the editable area. May be useful to do transformations outside the realm of the API. ### getHTML Returns the HTML value of the editor in its current state. This value is equivalent to the contents of the `<body>` tag and does not include any surrounding boilerplate. ### setHTML Sets the HTML value for the editor. The value supplied should not contain `<body>` tags or anything outside of that. The method takes one argument: * **html**: The html to set. Returns self (the Squire instance). ### getSelectedText Returns the text currently selected in the editor. ### insertImage Inserts an image at the current cursor location. The method takes one argument: * **src**: The source path for the image. Returns a reference to the newly inserted image element. ### getPath Returns the path through the DOM tree from the `<body>` element to the current current cursor position. This is a string consisting of the tag, id and class names in CSS format. For example `BODY>BLOCKQUOTE>DIV#id>STRONG>SPAN.font>EM`. If a selection has been made, so different parts of the selection may have different paths, the value will be `(selection)`. The path is useful for efficiently determining the current formatting for bold, italic, underline etc, and thus determining button state. If a selection has been made, you can has the `hasFormat` method instead to get the current state for the properties you care about. ### getSelection Returns a W3C Range object representing the current selection/cursor position. ### setSelection Changes the current selection/cursor position. The method takes one argument: * **range**: The W3C Range object representing the desired selection. Returns self (the Squire instance). ### undo Undoes the most recent change. Returns self (the Squire instance). ### redo If the user has just undone a change, this will reapply that change. Returns self (the Squire instance). ### hasFormat Querie