I agree that it looks like "overenthusiastic propagation
of implied equalities." (Nicely put.) I haven't tested
whether this fails in any other version (we've moved all
our instances to 8.1beta2 to pound it the best we can),
but I see that Klint Gore has reported the test case fails
in 7.4.7 and 8.0.0.
=20
Any idea on when a fix might be available, or what
conditions must exist in a query for this failure to occur?
If this is too big to fix in 8.1, we would want to try to
identify vulnerable queries proactively, to work around
it somehow.
=20
-Kevin
=20
=20
I think this is a case of overenthusiastic propagation of implied
equalities. Do you know offhand if it fails in 8.0.* or earlier
branches?

regards, tom lane


---------------------------(end of broadcast)---------------------------
TIP 1: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo (AT) postgresql (DOT) org so that your
message can get through to the mailing list cleanly