使Firefox對xml的處理兼容IE的節點處理方法。具體代碼列出如下。

具體代碼如下。

<! DOCTYPE Html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" >
< HTML >
< HEAD >
< TITLE >  使firefox對xml的處理兼容IE的selectSingleNode selectNodes方法  </ TITLE >
< META  NAME ="Author"  CONTENT ="emu" >
< META  NAME ="KeyWords"  CONTENT ="firefox IE selectSingleNode selectNodes" >
< META  NAME ="Descr< SCRIPT  LANGUAGE ="javascript" >
<!--
var  isIE  =   !! document.all;

function  parseXML(st){
     if (isIE){
         var  result  =   new  ActiveXObject( " microsoft.XMLDOM " );
        result.loadXML(st);
    } else {
         var  parser  =   new  DOMParser();
         var  result  =  parser.parseFromString(st,  " text/xml " );
    }
     return  result;
}

if ( ! isIE){
     var  ex;
    XMLDocument.         try {
             return   new  XMLSerializer().serializeToString( this );
        } catch (ex){
             var  d  =  document.createElement( " div " );
            d.appendChild( this .cloneNode( true ));
             return  d.innerHTML;
        }
    });
    Element.prototype.__proto__.__defineGetter__( " xml " ,  function (){
         try {
             return   new  XMLSerializer().serializeToString( this );
        } catch (ex){
             var  d  =  document.createElement( " div " );
            d.appendChild( this .cloneNode( true ));
             return  d.innerHTML;
        }
    });
    XMLDocument.prototype.__proto__.__defineGetter__( " text " ,  function (){
         return   this .firstChild.textContent
    });
    Element.prototype.__proto__.__defineGetter__( " text " ,  function (){
         return   this .textContent
    });

    XMLDocument.prototype.selectSingleNode = Element.prototype.selectSingleNode = function (XPath){
         var  x = this .selectNodes(xpath)
         if ( ! x    x.length < 1 ) return   null ;
         return  x[ 0 ];
    }
    XMLDocument.prototype.selectNodes = Element.prototype.selectNodes = function (xpath){
         var  xpe  =   new  XPathEvaluator();
         var  nsResolver  =  xpe.createNSResolver( this .ownerDocument  ==   null   ?
             this .documentElement :  this .ownerDocument.documentElement);
         var  result  =  xpe.evaluate(xpath,  this , nsResolver,  0 ,  null );
         var  found  =  [];
         var  res;
         while  (res  =  result.iterateNext())
            found.push(res);
         return  found;
    }
}
var  x  =  parseXML( " <people>  <person first-name=/ " eric/ "  middle-initial=/ " H/ "  last-name=/ " jung/ " >    <address street=/ " 321  south st/ "  city=/ " denver/ "  state=/ " co/ "  country=/ " usa/ " />    <address street=/ " 123  main st/ "  city=/ " arlington/ "  state=/ " ma/ "  country=/ " usa/ " />  </person>  <person first-name=/ " jed/ "  last-name=/ " brown/ " >    <address street=/ " 321  north st/ "  city=/ " atlanta/ "  state=/ " ga/ "  country=/ " usa/ " />    <address street=/ " 123  west st/ "  city=/ " seattle/ "  state=/ " wa/ "  country=/ " usa/ " />    <address street=/ " 321  south avenue/ "  city=/ " denver/ "  state=/ " co/ "  country=/ " usa/ " />  </person></people> " );

alert( " 搜索所有人的姓氏（last-name） " )
var  results  =  x.selectNodes( " //person/@last-name " );
for  ( var  i = 0 ; i < results.length;i ++ )
  alert( " Person # "   +  i  +   "  has the last name  "   +  results[i].nodeValue);
alert( " 搜索第二個人 " );
//  IE是以0為下標基數的，而不是1
if ( ! document.all)
    results  =  x.selectSingleNode( " /people/person[2] " );
else
    results  =  x.selectSingleNode( " /people/person[1] " );
alert(results.xml)

alert( " 獲得住址在donver街上的人 " );
results  =  x.selectNodes( " //person[address/@city='denver'] " );
for  ( var  i = 0 ; i < results.length;i ++ )alert(results[i].xml)

if ( ! document.all){
     //  獲得所有街名中帶south的地址
    results  =  x.selectNodes( " //address[contains(@street, 'south')] " );
    alert(results[ 0 ].xml);
} else {
    alert( " IE不支持 //address[contains(@street, 'south')] 這種查詢方式 " )
}
// -->
</ SCRIPT >
</ HEAD >
< BODY >
</ BODY >
</ HTML >

例子引自 http://kb.mozillazine.org/XPath 。從例子可以看到，IE對xpath的支持還是有限度的。