> That last line has me puzzled Mike. Would someone expand
    please, for
    > the uninitiated?

    In the input we have a set of tables each containing zero
    or more columns, each column has a name. We want to find
    the names of the columns that appear in each table. Group
    the columns by name; for each group (of columns with the
    same name), the column appears in every table if the number
    of tables containing a column in that group equals the
    total number of tables.

    Any clearer?

 <tables>
   <table name="table1">
     <column name="col1"/>
     <column name="col2"/>
     <column name="col3"/>
     <column name="col4"/>
   </table>
   <table name="table2">
     <column name="col1"/>
     <column name="col2"/>
     <column name="col5"/>
     <column name="col6"/>
   </table>

Yes, the basics are clear. Your last line(s) were
  you want something like

   count(current-group()/parent::table) = $number-of-tables

  relying on the fact that "/" eliminates duplicates.


My question was, how does the '/' eliminate duplicates?
That's the bit I don't understand.

regards DaveP