There is an IFRAME on the page:
Copy Code code as follows:
<iframe src= ' a.html ' ></iframe>
<button> Test iframe Leak </button>
The contents of a.html are as follows:
Copy Code code as follows:
<! DOCTYPE html>
<meta charset= "UTF-8" >
<title>insert title here</title>
<script type= "Text/javascript" src= "Jquery/jquery.min.js" ></script>
<style>
. hack{
/* 1. All browsers are valid/*
Background-color:green;
/* 2.ie8~ie10,opera valid, but opera compatibility can not be considered * *
Background-color:blue\0;
/* 3.ie9~ie10 valid, combined with 2, in 2 to write first for IE8, in this section for ie9| IE10 *
background-color:red\9\0;
/* 4.ie7 valid, with 23 combination can distinguish ie7/ie8/(ie9| IE10) * * *
+background-color:yellow;
}
/* for ie10+ * *
@media screen and (-ms-high-contrast:active), (-ms-high-contrast:none) {
. hack{
Background-color:pink;
}
}
</style>
<body>
The suffix "\9" ie6/ie7/ie8/ie9/ie10 is in effect.
The suffix "ie8/ie9/ie10" is in force and is the IE8/9/10 hack
The suffix "\9\0" only takes effect on Ie9/ie10, is IE9/10 hack
Prefix "*" is valid for IE7
Prefix "+" is valid for IE7
Selector prefix @media screen and (-ms-high-contrast:active), (-ms-high-contrast:none)
Valid for IE10
Copy Code code as follows:
<input type= ' text ' value= ' Chinese ' ></input>
<div class= ' hack ' style= ' width:100px;
height:100px; " ></div>
<div id= "1" style= "padding:5px;position:relative;background-color:green;margin:10px;border:20px solid Red;width : 100%;height:200px; " >
</div>
<div id= "2" ></div>
<span>SPAN</span>
</body>
B.html contents are as follows:
Copy Code code as follows:
<! DOCTYPE html>
<meta charset= "UTF-8" >
<title>insert title here</title>
<script type= "Text/javascript" src= "Jquery/jquery.min.js" ></script>
<body>
<span>SPAN</span>
</body>
There is a transmission on the Internet, the following writing can reduce memory leakage:
Copy Code code as follows:
var framedom = $ (' iframe:eq (0) ') [0];
var framewin = Framedom.contentwindow;
try{
FrameWin.document.write (");
FrameWin.document.clear ();
}catch (e) {};
FRAMEDOM.SRC = ' b.html ';
So what's the effect?
To set a URL directly
Copy Code code as follows:
var flag = true;
var framedom = $ (' iframe:eq (0) ') [0];
$ (' button '). On (' click ', function () {
if (flag) {
var framedom = $ (' iframe:eq (0) ') [0];
var framewin = Framedom.contentwindow;
/*
try{
FrameWin.document.write (");
FrameWin.document.clear ();
}catch (e) {};
*/
FRAMEDOM.SRC = ' b.html ';
Flag = false;
}else{
var framedom = $ (' iframe:eq (0) ') [0];
var framewin = Framedom.contentwindow;
/*
try{
FrameWin.document.write (");
FrameWin.document.clear ();
}catch (e) {};
*/
FRAMEDOM.SRC = ' a.html ';
Flag = true;
}
$ (' #console '). Append (flag?) ' Switch to a.html ': ' Switch to b.html ');
});
Use the Sieve test: Each time you switch, #leaks增加28左右.
Writing two: According to the network transmission
Copy Code code as follows:
<script>
var flag = true;
var framedom = $ (' iframe:eq (0) ') [0];
$ (' button '). On (' click ', function () {
if (flag) {
var framedom = $ (' iframe:eq (0) ') [0];
var framewin = Framedom.contentwindow;
try{
FrameWin.document.write (");
FrameWin.document.clear ();
}catch (e) {};
FRAMEDOM.SRC = ' b.html ';
Flag = false;
}else{
var framedom = $ (' iframe:eq (0) ') [0];
var framewin = Framedom.contentwindow;
try{
FrameWin.document.write (");
FrameWin.document.clear ();
}catch (e) {};
FRAMEDOM.SRC = ' a.html ';
Flag = true;
}
$ (' #console '). Append (flag?) ' Switch to a.html ': ' Switch to b.html ');
});
</script>
Use the Sieve test: Each time you switch, #leaks增加28左右. There's no difference between the wording
Writing three:
Copy Code code as follows:
var flag = true;
var framedom = $ (' iframe:eq (0) ') [0];
$ (' button '). On (' click ', function () {
if (flag) {
/*
try{
FrameDom.contentWindow.document.write (");
FrameDom.contentWindow.document.clear ();
FrameDom.contentWindow.close ();
}catch (e) {};
*/
$ (' iframe:eq (0) '). Remove ();
$ (' body '). Append ("<iframe src= ' b.html ' ></iframe>");
Flag = false;
}else{
/*
try{
FrameDom.contentWindow.document.write (");
FrameDom.contentWindow.document.clear ();
FrameDom.contentWindow.close ();
}catch (e) {};
*/
$ (' iframe:eq (0) '). Remove ();
$ (' body '). Append ("<iframe src= ' a.html ' ></iframe>");
Flag = true;
}
});
Using the Sieve test: #leaks平均为 3, the difference between the first two is huge
Writing four: Note that the writing of the three notes a piece of code, remove the annotation what?
Copy Code code as follows:
var flag = true;
var framedom = $ (' iframe:eq (0) ') [0];
$ (' button '). On (' click ', function () {
if (flag) {
try{
FrameDom.contentWindow.document.write (");
FrameDom.contentWindow.document.clear ();
FrameDom.contentWindow.close ();
}catch (e) {};
$ (' iframe:eq (0) '). Remove ();
$ (' body '). Append ("<iframe src= ' b.html ' ></iframe>");
Flag = false;
}else{
try{
FrameDom.contentWindow.document.write (");
FrameDom.contentWindow.document.clear ();
FrameDom.contentWindow.close ();
}catch (e) {};
$ (' iframe:eq (0) '). Remove ();
$ (' body '). Append ("<iframe src= ' a.html ' ></iframe>");
Flag = true;
}
});
There is no significant difference between this writing and writing 3, each switch #leaks still increase by about 3
Therefore, it can be concluded that the best solution to reset the IFRAME address memory leak method is to kill it, and then add one!
It's not necessarily the net.
Note: The native test environment is WIN7 x64 IE9
