Question

I’m attempting to build a regular expression that will match against the contents of an XML element containing some un-encoded data. Eg:

<myElement><![CDATA[<p>The <a href="http://blah"> draft </p>]]></myElement>

Usually in this circumstance I’d use

[^<]*

to match everything up to the less than sign but this isn’t working in this case. I’ve also tried this unsuccessfully:

[^(</myElement>)]*

I’m using Groovy, i.e. Java.

Answer 1

Please don’t do this, but you’re probably looking for:

<myElement>(.*?)</myElement>

This won’t work if <myElement> (or the closing tag) can appear in the CDATA. It won’t work if the XML is malformed. It also won’t work with nested <myElement>s. And the list goes on…

The proper solution is to use a real XML parser.

Your [^(</myElement>)]* regex was saying: match any number of characters that are not in the set (, <, /, m, etc., which is clearly not what you intended. You cannot place a group within a character class in order for it to be treated atomically — the characters will always be treated as a set (with ( and ) being literal characters, too).