On T, 2004-06-29 at 13:11, jacob koehler (RRes-Roth) wrote:
> seems evgen has got a sql99 compliant version of recursive SQL
> out and would agree to relicense it.
> are there any other concerns regarding evgens work? or do you want to
> encourage him to go this route? i have the impression he would happily
> join in on the postgres developement wrt recursive SQL, and some friendly
> words might encourage him to do the required work.
> > the funny thing that i almost get working implementation
> > of SQL99 recursive queries,"just for fun";), and i think
> > it would not be included too;)
Maybe he knows of some fundamental flaws with his implementation ?
Why else does he think that it will not be accepted ?
He could at least try to submit it again, once it is past the "almost"
part :)
BTW, I also "almost" got a working implementation of SQL99 recursive
queries (at least it parsed the syntax ;), but gave up due to lack of
time and also because the SQL99 recursive queries are a desceptively
complex beast which I was unable to fully understand from the specs, at
least the part beyond the simple parent-child tree queries.
And the simple parent-child tree queriest are now doable using
set-returning functions.
----------------
Hannu
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
