Btukfyl

In XQuery 3.1 (under eXistDB 4.4) I have two functions which return two sets of nodes that describe family relations of a person. The results of the two functions can overlap.

A first function person:person-relationship-as-object("#Guilhem_Faure_SML-AU") returns:

<person relation="Peire_Faure_SML-AU">

  <span class="en">Sibling(s)</span>

  <span class="fr">Frère(s)/sœur(s)</span>

</person>



<person relation="Arnald_Faure_SML-AU">

  <span class="en">Sibling(s)</span>

  <span class="fr">Frère(s)/sœur(s)</span>

</person>



<person relation="Esteve_Faure_SML-AU">

  <span class="en">Sibling(s)</span>

  <span class="fr">Frère(s)/sœur(s)</span>

</person>



<person relation="Martin_de_Verazilh_SML-AU">

  <span class="en">Aunt(s)/Uncle(s)</span>

  <span class="fr">Tante(s)/Oncle(s)</span>

</person>



<person relation="Raimund_de_Verazilh_SML-AU">

  <span class="en">Aunt(s)/Uncle(s)</span>

  <span class="fr">Tante(s)/Oncle(s)</span>

</person>

A second function person:person-relationship-as-subject("#Guilhem_Faure_SML-AU") returns

<person relation="Arnald_Faure_SML-AU">

  <span class="en">Sibling(s)</span>

  <span class="fr">Frère(s)/sœur(s)</span>

</person>



<person relation="Peire_Faure_SML-AU">

  <span class="en">Sibling(s)</span>

  <span class="fr">Frère(s)/sœur(s)</span>

</person>



<person relation="Esteve_Faure_SML-AU">

  <span class="en">Sibling(s)</span>

  <span class="fr">Frère(s)/sœur(s)</span>

</person>



<person relation="Bernarda_Guilhem_Faure_SML-AU">

  <span class="en">Spouse(s)</span>

  <span class="fr">Époux/épouse(s)</span>

</person>

Now I want to merge the two sets of nodes into one, and return each distinct node only once.

I attempted a rough SQL-like structure, but failed:

let $x := person:person-relationship-as-object("#Guilhem_Faure_SML-AU") 

          | person:person-relationship-as-subject("#Guilhem_Faure_SML-AU")



for $y in $x



order by $y/@relation ascending



group by $y/@relation, $y/span[@class="en"], $y/span[@class="fr"]



return $y

I don't think I really understand why the group by doesn't work here. Moreover, if group by doesn't do the job, how else can I approach what (I think?) should be a simple exercise in grouping/distinct results?

With XQuery and group by, the variable $y in your return clause is bound to the whole group you have created so to eliminate duplicates you need to use return $y[1]. See https://www.w3.org/TR/xquery-31/#id-group-by which says

In the post-grouping tuple generated for a given group, each
non-grouping variable is bound to a sequence containing the
concatenated values of that variable in all the pre-grouping tuples
that were assigned to that group.

and then

This behavior may be surprising to SQL programmers, since SQL reduces
the equivalent of a non-grouping variable to one representative value.