If I have this code:

            Predicate documentFilter = ShipFields.ID == DocumentTransactionFields.ShipID &
                                       DocumentTransactionFields.VoyageID == DBNull.Value &
                                       DocumentTransactionFields.AllocatedFlag == true;

            ScalarQueryExpression followUpMessagesQuery =
                new ScalarQueryExpression(
                    DocumentFields.ID.SetAggregateFunction(AggregateFunction.Count), 
                    documentFilter & DocumentFields.FollowUpFlag == true, relations);

      ScalarQueryExpression hasPriorityMessagesQuery =
        new ScalarQueryExpression(
          DocumentFields.ID.SetAggregateFunction(AggregateFunction.Count),
          documentFilter & DocumentFields.PriorityFlag == true, relations);

I am performing almost the same query twice. Looking back in the forums, I see that Derived Tables were mentioned some time ago but I can't find anything current. Am I right in thinking this would be a candidate for a Derived Table so I can retrieve two pieces of information from one query?

Secondly, the second expression currently returns an aggregate Count but I actually only need to know if any matching rows are present. Seems a waste to count all the documents (there might be a lot) when ideally I would like it to short-circuit on finding one. This might be more of a SQL question but what is the best way of doing this?

Cheers Simon

Can't you use an EXISTS clause for the COUNT ?

Derived tables were cut as they were a pain to define (as in: the API to get them defined was pretty verbose)

The second query indeed should be an exist predicate. So do a getscalar where you fetch the id of the row with an exists filter.

Derived tables were cut as they were a pain to define

I really hope they will return/appear in 3.0.

arschr wrote:

Derived tables were cut as they were a pain to define

I really hope they will return/appear in 3.0.

I know you're waiting on them We'll do our best