DOM == document object model
document 对象是唯一同时属于 BOM 和 DOM 的
rows 是一种DOM集合，不是数组，所以没有sort() 函数。
NodeList NamedNodeMap :
document.all == document.getElementByTagName("*"); 取得document中所有元素
ie6 在使用getElementByName() 和 getElementById() 时都会同时匹配 id 和 name 属性，要注意
ie 在使用 setAttribute() 有可能出错
所以尽可能使用 img.src 形式取代 img.getAttribute("src")
Dom对象是host object，host 对象在ie8以及之前版本是通过COM实现的，而不是JScript
所以document.createElement 等Dom 方法是COM 对象的实例，因此typeof document.createElement 返回 object
每一个node节点都有自己的ownerDocument属性，如果将不同ownerDocument属性append到结构中会产生错误，需要使用importNode()方法。
注意直接定义在node上的属性，在标准浏览器和ie9,10中用getAttribute取不到,得到的结果是null
比如div.aa = 10,用div.getAttribute('aa')得到的是null,需要用div.aa或者div['aa']这样来取
Nodelist是类数组对象，但不是Array的实例( 转化为数组的话[].slice.call(eleList) )，当然用不了array的方法啦，同时它是动态更新的，实时变化的
function convertToArray(nodes) {
var array = null;
try {
array = Array.prototype.slice.call(nodes, 0); //non-IE and IE9+，throw an error because a NodeList is implemented as a COM object and thus cannot be used where a JScript object is necessary
}
catch (ex) {
array = new Array();
for (var i=0, len=nodes.length; i < len; i++) {
array.push(nodes[i]);
}
}
return array;
}
Dom把页面转化为层次节点结构的地图
Html或者xml的每一个部分都是一种包含不同数据的节点
通过这种树状结构来表示document，可以让开发者获得节点中的内容和修改它的结构
DOM Level 1 became a W3C recommendation in October 1998. It consisted of two modules（Dom core and Dom HTML interface):
the DOM Core, which provided a way to map the structure of an XML-based document to allow for easy access to and manipulation of any part of a document, and the DOM HTML, which extended the DOM Core by adding HTML-specific objects and methods.
Dom 2 added support for mouse and user-interface events (long supported by DHTML), ranges, traversals (methods to iterate over a DOM document), and support for Cascading Style Sheets (CSS) through object interfaces. The original DOM Core introduced in Level 1 was also extended to include support for XML namespaces.
- ➤ DOM Views — Describes interfaces to keep track of the various views of a document (the document before and after CSS styling, for example)
- ➤ DOM Events — Describes interfaces for events and event handling
- ➤ DOM Style — Describes interfaces to deal with CSS-based styling of elements
- ➤ DOM Traversal and Range — Describes interfaces to traverse and manipulate a document tree
DOM Level 3 further extends the DOM with the introduction of methods to load and save documents in a uniform way (contained in a new module called DOM Load and Save) and methods to validate a document (DOM Validation). In Level 3, the DOM Core is extended to support all of XML 1.0, including XML Infoset, XPath, and XML Base.