Tomas Vondra <[email protected]> writes: > This should fix it I guess, and it's how we deal with unused return > values elsewhere. I've considered using USE_ASSERT_CHECKING here, but it > seems rather ugly with that. I'll wait for Emre's opinion ...
I think what you want is to mark the variable with
PG_USED_FOR_ASSERTS_ONLY.
regards, tom lane
