capture text, including tags from string, and then reorder tags with text

I have the following text:

abcabcabcabc<2007-01-12><name1><2007-01-12>abcabcabcabc<name2><2007-01-11>abcabcabcabc<name3><2007-02-12>abcabcabcabc<name4>abcabcabcabc<2007-03-12><name5><date>abcabcabcabc<name6>

I need to use regular expressions in order to clean the above text:

The basic extraction rule is:

<2007-01-12>abcabcabcabc<name2>

I have no problem extracting this pattern. My issue is that within th text I have malformed sequences: If the text doesn't start with a date, and end with a name my extraction fails. For example, the text above may have several mal formed sequences, such as:

abcabcabcabc<2007-01-12><name1>

Should be:

<2007-01-12>abcabcabcabc<name1>

Is it possible to have a regular expression that would clean the above, pri开发者_如何学Pythonor to extracting my consistent pattern. In short, i need to find all mal formed patterns, and then take the date tag and put it in front of it, as provided in the example above.

Thanks.

---

Do you need something like this perhaps?

public class Extract {
    public static void main(String[] args) {
        String text =
            "abcabcabcabc<2007-01-12><name1>" +
            "<2007-01-12>abcabcabcxxx<name2>" +
            "<2007-01-11>abcabcabcyyy<name3>" +
            "<2007-02-12>abcabcabczzz<name4>" +
            "abcabcabc123<2007-03-12><name5>" +
            "<date>abcabcabc456<name6>";
        System.out.println(
            text.replaceAll(
                "(text)<(text)>(text)<(text)>"
                    .replace("text", "[^<]*"),
                "$1$3 - $2 - $4\n"
            )
        );
    }
}

This prints:

abcabcabcabc - 2007-01-12 - name1
abcabcabcxxx - 2007-01-12 - name2
abcabcabcyyy - 2007-01-11 - name3
abcabcabczzz - 2007-02-12 - name4
abcabcabc123 - 2007-03-12 - name5
abcabcabc456 - date - name6

Essentially, there are 3 parts:

• The naked text is captured by \1 and \3 -- one of these should be an empty string
• The date is \2
• The name is \4

You can of course use a Matcher and extract individual group too.

References

• regular-expressions.info/Grouping