finding unique nodes

I would like to find just the first element of certain nodes. My XML
document looks something like this:

<doc>

  <para name = "Normal" left-indent = "18"> text</para>

  <para name = "Normal" left-indent = "18"> more text</para>

  <para name = "any possible name" left-indent = "5"> more text</para>

  <para name = "any possible name" left-indent = "5"> some more text</para>

  <border top-border = "1">

    <para name = "another name" left-indent = "3"> text</para>

  </border>

  <table>
   <row>
    <cell><para name = "still another name" left-indent="2"> cell text</cell>
   </row>
  </table>

</doc>

I need my resulting document to look like this:

<doc>

  <table-of-styles>
   <para-style name = "Normal" left-indent = "18"/>
   <para-style name = "any possible name" left-indent = "3">
    <para-style = "another name" left-indent = "3"/>
    <para-style name = "still another name" left-indent="2"/> 
  </table-of-style>

  <!--rest of the document-->
   
</doc>

Notice that I have just one para-style elelement for each name--one for
"Normal", one for "another name", and one for "still another name", and
so on.

I need to then create a text string for each of the para-style elements.
I cannot duplicate the string.

Notice also that the para element can occur in many different places.

I know to start with something like this:

<xsl:template match = "doc">

 <xsl:for-each select = "descendant::para">
   <!--here is where I don't know what to do-->

 </xsl:for-each>

 <xsl:apply-templates/>
</xsl:template>

I think I have to use a generate-id function in my xsl:for-each loop?
But I've never done this before.

Thanks




-- 

************************
*Paul Tremblay         *
*phthenry@xxxxxxxxxxxxx*
************************

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