Semantics of GROUP_CONCAT, part 2

Hello, sorry for asking again about a relevant question concerning GROUP_CONCAT but I think I found a problematic situation while I was running my queries. The previous question was this: http://answers.semanticweb.com/questions/20337/semantics-of-group_concat

I have a case in my queries where I want to combine GROUP_CONCAT with UNION and the VALUES clause.

AndyS suggested to use GROUP_CONCAT and UNION as following:

SELECT *
WHERE { 
   { SELECT (GROUP_CONCAT(DISTINCT ?variable1; separator = ',') as ?var1) 
     { <http://dbpedia.org/resource/Peter> some_predicate_1 ?variable1 } }
   } UNION {
   { SELECT (GROUP_CONCAT(DISTINCT ?variable2; separator = ',') as ?var2)
     {?variable2 some_predicate_2 <http://dbpedia.org/resource/John> }
   }
}

In this example, suppose we have for the ?variable1 and ?variable2 VALUES clauses bound to them with some values. Then something very strange happens.

I get for ?var1 and ?var2 all the values that both ?variable1 and ?variable2 contain. So both ?var1 and ?var2 are identical containing all the bindings for both variables!

I tested the query with FILTER (IN ()) and it work correctly. Also the results when I am not using GROUP_CONCAT and use UNION with VALUES are again correct.

Is this a Jena bug with GROUP_CONCAT and VALUES?

Thank you again for your help! Chris

EDIT ADDED EXAMPLE:

SELECT * 
WHERE {  
{ SELECT (GROUP_CONCAT(DISTINCT ?variable1; separator = ',') as ?var1) 
    { VALUES ?variable1 { <http://dbpedia.org/resource/Peter> } ?variable1 some_predicate_1 ?object } 
} UNION 
{ SELECT (GROUP_CONCAT(DISTINCT ?variable2; separator = ',') as ?var2) 
    { VALUES ?variable2 { <http://dbpedia.org/resource/John> } ?subject some_predicate_2 ?variable2 } 
}

The result from Jena is: (?var1 = http://dbpedia.org/resource/Peter, http://dbpedia.org/resource/John" ) -> [Root]

and the same is the result for ?var2.