In jQuery's attr and prop, it is mentioned that improper use of property in versions before IE9 will cause memory leaks, and the difference between Attribute and Property is also very troublesome. Add it in HTML5 The data-* method is used to customize attributes. The so-called data-* is actually the data- prefix plus the customized attribute name. Data can be stored using such a structure. Using data-* can solve the current situation of confusing and unmanaged custom attributes.
data-* has two setting methods, you can write directly on the HTML element tag
<div id="test" data-age="24"> Click Here </div>
The data -age is a custom attribute. Of course, we can also operate it through JavaScript. Elements in HTML5 will have a dataset attribute, which is a collection of key-value pairs of the DOMStringMap type
var test = document.getElementById('test'); test.dataset.my = 'Byron';
In this way, a data-my custom attribute is added to the div. There are two things to pay attention to when using JavaScript to operate the dataset
1. We need to when adding or reading attributes Remove the prefix data-*. Like the above example, we do not use the form test.dataset.data-my = 'Byron';.
2. If the attribute name also contains a hyphen (-), it needs to be converted to camel case naming. But if a selector is used in CSS, we need to use the hyphen format
as just now The code appends content
<style type="text/css"> [data-birth-date] { background-color: #0f0; width:100px; margin:20px; } </style>
test.dataset.birthDate = '19890615';
In this way we set data-birth- through JavaScript Date custom attribute, add some styles to the div in the CSS style sheet, see the effect
is also read through the dataset object, using "." To get the attributes, you also need to remove the data- prefix, and the hyphens need to be converted into camel case naming
var test = document.getElementById('test'); test.dataset.my = 'Byron'; test.dataset.birthDate = '19890615'; test.onclick = function () { alert(this.dataset.my + ' ' + this.dataset.age+' '+this.dataset.birthDate); }
Some students may ask if there is any difference between this and getAttribute/setAttribute except naming. Let’s take a look
var test = document.getElementById('test'); test.dataset.birthDate = '19890615'; test.setAttribute('age', 25); test.setAttribute('data-sex', 'male'); console.log(test.getAttribute('data-age')); //24 console.log(test.getAttribute('data-birth-date')); //19890516 console.log(test.dataset.age); //24 console.log(test.dataset.sex); //male
In this way we can see that both have set the attributes to attributes (nonsense, otherwise they can call them custom attributes), that is It is said that getAttribute/setAttribute can operate all dataset contents. The dataset contents are only a subset of attributes. The special thing is the naming, but there are only attributes with data- prefix in the dataset (there is no age=25 one).
So why do we still use data-*? One of the biggest advantages is that we can manage all custom attributes in a unified manner in the dataset object. It is very convenient to traverse everything without going to zero. It’s scattered, so it’s still good to use.
The bad news is that the browser compatibility of data-* is very unoptimistic
Among them, IE11 is simply blinding the eyes of friends. It seems that if you want to fully use this attribute The road is long and the journey is long
This article is reproduced from: http://www.cnblogs.com/dolphinX/p/3348458.html