accessing specific elements

Hi,

I have the following problem.  Imagine the concept of  a flight
'Brussels - London - Brussels' encoded in XML as follows:

<input_stream>
 <chunk>
   <token>
...
...
   </token>
  </chunk>
 <chunk >
    <token>
      <normalized.form>Brussels</normalized.form>
      <domlex attribute="location" subattribute="val" value="Brussels"/>

   </token>
   <token>
      <normalized.form>London</normalized.form>
      <domlex attribute="location" subattribute="val" value="London"/>
  </token>
  <token>
      <normalized.form>and</normalized.form>
  </token>
    <token>
      <normalized.form>Brussels</normalized.form>
      <domlex attribute="location" subattribute="val" value="Brussels"/>

  </token>
</chunk>
...
...
</input_stream>


So, <chunk> contains 4 <token> elements out of which
only 3 have a domlex[@attribute='location'], namely the 1st, 2nd and
4th.

Now *irrespective of their sequential order*, I would like to

-     count the tokens which have a domlex[@attribute='location']
        (this is no problem, I just use
count(//domlex[@attribute='location'])"/>))

-     access only those tokens with an [@attribute='location'],  so that
I can express
        the condition: "If the first element with  @attribute='location'
and the third have the same
        value, then write out only one as the origin (and the middle one
as the destination).

I thought of storing the value of the first in a variable and comparing
it with subsequent ones... but I can't get it work...

I don't know how to access only the tokens which do have
domlex[@attribute='location'].

Can anybody help me?

Thanks in advance

Sabine




 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list