原文:http://www.cnlei.org/blog/article.asp?id=496
<div id="d1"></div>
<script >
function createButton(){
var obj = document.createElement("button");
obj.innerHTML="点我!";
obj.οnclick=function(){
//处理click事件
}
obj.οnmοuseοver=function(){
//处理mouseover事件
}
return obj;//这里由于需要返回创建的对象,所以不能把obj直接设为null. return 后obj是局部变量,不能在外部断开其与HTMLElement的引用.ie中将出现泄漏问题
}
var 按钮 = document.getElementsById("d1").appendChild( createButton());
按钮.做某些事();
按钮.做某些事();
........
某些东西.某些事(按钮);
......
</script>
<script >
function createButton(){
var obj = document.createElement("button");
obj.innerHTML="点我!";
obj.οnclick=function(){
//处理click事件
}
obj.οnmοuseοver=function(){
//处理mouseover事件
}
return obj;//这里由于需要返回创建的对象,所以不能把obj直接设为null. return 后obj是局部变量,不能在外部断开其与HTMLElement的引用.ie中将出现泄漏问题
}
var 按钮 = document.getElementsById("d1").appendChild( createButton());
按钮.做某些事();
按钮.做某些事();
........
某些东西.某些事(按钮);
......
</script>
这种写法在IE中100%内存泄漏
使用try finally很容易解决些问题
function createButton(){
var obj = document.createElement("button");
obj.innerHTML="点我!";
obj.οnclick=function(){
//处理click事件
}
obj.οnmοuseοver=function(){
//处理mouseover事件
}
try{
return obj;
}finally{
obj = null;//这句话在return 之后才执行,有效地解决了需在return后将obj置null的问题
}
}
var obj = document.createElement("button");
obj.innerHTML="点我!";
obj.οnclick=function(){
//处理click事件
}
obj.οnmοuseοver=function(){
//处理mouseover事件
}
try{
return obj;
}finally{
obj = null;//这句话在return 之后才执行,有效地解决了需在return后将obj置null的问题
}
}
一个函数或方法中,其实有很多地方都需要这种选返回值,最后执行某些事的
=====================================================================
附一:JavaScript Error (try/catch/finally)
Introduction
Like other programming languages, JavaScript provides the possibility to make use of the try/catch/finally block. Usually when an error is encountered then the script stops and doesn’t continue with the rest of the page. The try/catch/finally block can be used to continue the processing with the rest of the page. You just have to put the code in your try block and when an error in encountered there, then it will call the catch block. The finally block is called always regardless of an error occurred or not. The following example makes the usage clear.
Example:
<html>
<head>
</head>
<body>
<script type="text/javascript">
try
{
document.write(unknownVariable)
}
catch(e)
{
document.write(e.message + "<br/>")
}
finally
{
document.write("This is the finally message.")
}
</script>
</body>
</html>
<head>
</head>
<body>
<script type="text/javascript">
try
{
document.write(unknownVariable)
}
catch(e)
{
document.write(e.message + "<br/>")
}
finally
{
document.write("This is the finally message.")
}
</script>
</body>
</html>