Greg Stark wrote:
"Tom Dunstan" <[EMAIL PROTECTED]> writes:
I didn't really want to go down that path in this thread
since it would turn what should be a fairly non-intrusive
patch to add a new type into a big thing, and I really just
wanted to get enums in. :) I tend to think of it the other
way around from how you put it: if a general solution to
that problem can be found which does fall afoul of the
security issues that were the reason for multi-argument
output functions to be killed off in the first place, then
great, and enums can directly benefit.
True. Perhaps it's reasonable to use a 8-byte representation in the name of
getting the user-visible feature in. Knowing that the fundamental problem will
eventually be solved and the implementation can eventually be improved
transparently to use 1 to 4 byte storage.
8 bytes is dead. We are going with 4 bytes, which will in fact be an oid
which will uniquely identify a <typeoid,value> combination.
cheers
andrew
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
