Dom4j是Java访问XML的利器之一，另一个是JDom。记得当年因为粗掌握点JDomAPI但项目要求使用Dom4j还闹一阵情绪，现在看来真是没必要，只花一些时间成本就进去一个新世界绝对是值得做的一件事。更何况JDom因无人更新而停顿了。
Dom4j有两个包，一个是dom4j-1.6.1.jar，它提供基本的XML API支持，如访问节点，属性等。
还有一个是jaxen-1.1-beta-9.jar，它提供XPath支持。
关于XPath的语法，请见转载：XPath基本语法
言归正传，下面请看例程。
1.访问节点群
XML样本：
<applications> <application name='chat'> <mtLanguage source='ar_ar' target='en_us' /> <mtLanguage source='zh_cn' target='en_us' /> <mtLanguage source='zh_tw' target='en_us' /> <mtLanguage source='en_us' target='ar_ar' /> <mtLanguage source='en_us' target='zh_cn' /> <mtLanguage source='en_us' target='zh_tw' /> <mtLanguage source='en_us' target='fr_fr' /> <mtLanguage source='en_us' target='de_de' /> <mtLanguage source='en_us' target='it_it' /> <mtLanguage source='en_us' target='ja_jp' /> <mtLanguage source='en_us' target='ko_kr' /> <mtLanguage source='en_us' target='pt_br' /> <mtLanguage source='en_us' target='ru_ru' /> <mtLanguage source='en_us' target='es_es' /> <mtLanguage source='fr_fr' target='en_us' /> <mtLanguage source='de_de' target='en_us' /> <mtLanguage source='it_it' target='en_us' /> <mtLanguage source='ja_jp' target='en_us' /> <mtLanguage source='ko_kr' target='en_us' /> <mtLanguage source='pt_br' target='en_us' /> <mtLanguage source='ru_ru' target='en_us' /> <mtLanguage source='es_es' target='en_us' /> </application> <application name='doc'> <mtLanguage source='ar_ar' target='en_us' /> <mtLanguage source='zh_cn' target='en_us' /> <mtLanguage source='zh_tw' target='en_us' /> <mtLanguage source='en_us' target='ar_ar' /> <mtLanguage source='en_us' target='zh_cn' /> <mtLanguage source='en_us' target='zh_tw' /> <mtLanguage source='en_us' target='fr_fr' /> <mtLanguage source='en_us' target='de_de' /> <mtLanguage source='en_us' target='hi_in' /> <mtLanguage source='en_us' target='it_it' /> <mtLanguage source='en_us' target='ja_jp' /> <mtLanguage source='en_us' target='ko_kr' /> <mtLanguage source='en_us' target='pt_br' /> <mtLanguage source='en_us' target='ru_ru' /> <mtLanguage source='en_us' target='es_es' /> <mtLanguage source='en_us' target='ur_pk' /> <mtLanguage source='fr_fr' target='en_us' /> <mtLanguage source='de_de' target='en_us' /> <mtLanguage source='hi_in' target='en_us' /> <mtLanguage source='it_it' target='en_us' /> <mtLanguage source='ja_jp' target='en_us' /> <mtLanguage source='ko_kr' target='en_us' /> <mtLanguage source='pt_br' target='en_us' /> <mtLanguage source='ru_ru' target='en_us' /> <mtLanguage source='es_es' target='en_us' /> <mtLanguage source='ur_pk' target='en_us' /> </application> </applications>
现在，如果我想要访问属性为chat的application节点下的所有mtLanguage子节点，XPath应该这样写：
//applications/application[@name='chat']/mtLanguage
而具体操作的Java语句是：
Document doc= DocumentHelper.parseText(responseXML);// 这个responseXML就是上面的XML样例 List<?> elms=doc.selectNodes("//applications/application[@name='chat']/mtLanguage"); System.out.println("There are "+elms.size()+" language pairs available in text translation"); for(Object obj:elms){ Element elm=(Element)obj; System.out.println("From "+elm.attributeValue("source")+" to "+elm.attributeValue("target")); }
执行上面语句输出如下：
There are 22 language pairs available in text translation
From ar_ar to en_us
From zh_cn to en_us
From zh_tw to en_us
From en_us to ar_ar
From en_us to zh_cn
From en_us to zh_tw
From en_us to fr_fr
From en_us to de_de
From en_us to it_it
From en_us to ja_jp
From en_us to ko_kr
From en_us to pt_br
From en_us to ru_ru
From en_us to es_es
From fr_fr to en_us
From de_de to en_us
From it_it to en_us
From ja_jp to en_us
From ko_kr to en_us
From pt_br to en_us
From ru_ru to en_us
From es_es to en_us
2.访问特定节点
XML样本：
<rep sts="OK" a="trep" tl="zh-CN"> <docs> <d dt="ndoc" did="d20160223213120480009045125076363146" lang="en-US" ctime="2016-02-23T21:31:20" mtime="2016-02-23T21:31:20" orig="1" mime="text/x-mt-xml" wc="2"> <p pid="1" wc="2"> <s sid="1"> <t tid="1" tt="orig" wc="2">Good evening</t> </s> </p> </d> <d dt="ndoc" did="d20160223213120480009045125076363146" lang="zh-CN" ctime="2016-02-23T21:31:20" mtime="2016-02-23T21:31:20" orig="0" mime="text/x-mt-xml" sc="100.00" wc="1"> <p pid="1" wc="1"> <s sid="1"> <t tid="1" tt="mt" src="mt" sc="100.00" wc="1">晚上好</t> </s> </p> </d> </docs> </rep>
如果我想得到上文中“晚上好”这段文字，XPath应该这样写
//rep/docs/d[last()]/p/s/t
对应的Java代码是：
Document doc= DocumentHelper.parseText(responseXML); Element elm = (Element) doc.selectSingleNode("//rep/docs/d[last()]/p/s/t"); targetTxt=elm.getText();
3.取属性
XML样本：
<rep sts="OK" a="trep" tl="zh-CN"> <docs>1</docs> </rep>
要取根节点rep的sts属性，XPath可以这样写：
//rep/@sts
而对应的Java语句是：
System.out.println("XML="+responseXML);
Document doc= DocumentHelper.parseText(responseXML);
Attribute attr = (Attribute) doc.selectSingleNode("//rep/@sts");
return attr.getText();
，如需转载请自行联系原作者
