从XML样式的String中删除无效字符

Question

I'm create a web service which receive string and convert it to XML. The XML creation is done through java dom4j. The string's format is like:

<form01><Textbox1 id="Textbox1" dataType="java.lang.String" perDataProId="">Site1</Textbox1><Textbox2 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1</Textbox2></form01>

To avoid XML invalid character I think I can use a StringReader to read string and remove &,but I wonder how to remove < and >?For example if the input string is

<form01><Textbox1 id="Textbox1" dataType="java.lang.String" perDataProId="">Site<1</Textbox1><Textbox2 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1</Textbox2></form01>

how to remove the '<' in "Site<1" while keep others like ? Any suggestion?Thx.

Answer 1

Normally the XML APIs will take care when setting the text in an XML DOM, or retrieving it later.

Some characters will be converted to an XML entity: < > " ' & .

There are also some Characters not allowed in some XML versions, like \ .

apache.commons.lang has a StringEscapeUtils.escapeXML if you need to do the conversion yourself as pure text.

Answer 2

You could use the following RegExp:

  public static void main(String[] args)
  {
    String str = "<form01><Textbox1 id=\"Textbox1\" dataType=\"java.lang.String\" perDataProId=\"\">Site<1</Textbox1><Textbox2 id=\"Textbox2\" dataType=\"java.lang.String\" perDataProId=\"\">Site1>a</Textbox2><Textbox3 id=\"Textbox2\" dataType=\"java.lang.String\" perDataProId=\"\">Site1&</Textbox3></form01>";
    System.out.println(str.replaceAll("(>[^<>]*)<([^<>]*<\\/)", "$1&lt;$2"));
    System.out.println(str.replaceAll("(>[^<>]*)>([^<>]*<\\/)", "$1&gt;$2"));
    System.out.println(str.replaceAll("(>[^<>]*)\\&([^<>]*<\\/)", "$1&amp;$2"));
  }

Results:

<form01><Textbox1 id="Textbox1" dataType="java.lang.String" perDataProId="">Site&lt;1</Textbox1><Textbox2 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1>a</Textbox2><Textbox3 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1&</Textbox3></form01>
<form01><Textbox1 id="Textbox1" dataType="java.lang.String" perDataProId="">Site<1</Textbox1><Textbox2 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1&gt;a</Textbox2><Textbox3 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1&</Textbox3></form01>
<form01><Textbox1 id="Textbox1" dataType="java.lang.String" perDataProId="">Site<1</Textbox1><Textbox2 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1>a</Textbox2><Textbox3 id="Textbox2" dataType="java.lang.String" perDataProId="">Site1&amp;</Textbox3></form01>

In any case, I would consider using Guava HtmlEscapers . The RegExp for this case are a fragile overhead.