Rod Taylor <rbt@rbt.ca> writes:
>> It's the only way to prevent it from simplifying when you don't want it
>> to.
> I'm having a difficult time coming up with a circumstance where that is
> beneficial except when stats are out of whack.
Try trawling the archives --- I recall several cases in which people
were using sub-selects for this purpose.
In any case, I don't see the value of having the planner check to see if
a sub-select is just a trivial arithmetic expression. The cases where
people write that and expect it to be simplified are so few and far
between that I can't believe it'd be a good use of planner cycles.
regards, tom lane