I hope someone on this list can help me with this problem.  I am trying to use XSLT to transform a XML document into another XML document.  The problem here is that the source XML document is highly denormalized.  I say denormalized to mean that the XML document has much redundant data.  I would like to transform the denormalized XML document, via XSLT, into a normalized XML document.  For example, a denormalized XML document may be like:

<GROUP>group A</GROUP>
    <SECTION>section 1</SECTION>
        <STUFF>0001</STUFF>
        <MORE_STUFF>000000001</MORE_STUFF>
<GROUP>group A</GROUP>
    <SECTION>section 1</SECTION>
        <STUFF>0002</STUFF>
        <MORE_STUFF>000000002</MORE_STUFF>
    
The normalized form of the above XML document would look like:

<GROUP>group A</GROUP>
    <SECTION>section 1</SECTION>
        <STUFF>0001</STUFF>
        <STUFF>0002</STUFF>
        <MORE_STUFF>000000001</MORE_STUFF>
        <MORE_STUFF>000000002</MORE_STUFF>

The above XML document represents the same data as the first XML document in a more compact manner (we are assuming that the corresponding DTD defines element "STUFF" and "MORE_STUFF" to be repeating elements).

Is there a way to do the above with XSLT and not comparing any of the actual element values?
Any help is appreciated.