A projection query like this
nda2.Trace.proj('trace', presyn='segment_id') & key
gives
*scan_idx *segmentation trace *presyn
+----------+ +------------+ +--------+ +--------+
5 1 =BLOB= 421208
(1 tuples)
so that primary keys are not grouped at the beginning. Would make sense for all table results to consistently give primary key attributes at the beginning.
A projection query like this
gives
so that primary keys are not grouped at the beginning. Would make sense for all table results to consistently give primary key attributes at the beginning.