On Tue, May 08, 2018 at 04:07:41PM +0900, Amit Langote wrote: > I have to agree to go with this conservative approach for now. Although > we might be able to evaluate the array elements by applying the coercion > specified by ArrayCoerceExpr, let's save that as an improvement to be > pursued later.
Thanks for confirming. Yes, non-volatile functions would be actually safe, and we'd need to be careful about NULL handling as well, but that's definitely out of scope for v11. > FWIW, constraint exclusion wouldn't prune in this case either (that is, if > you try this example with PG 10 or using HEAD as of the parent of > 9fdb675fc5), but it doesn't crash like the new pruning code does. Yeah, I have noticed that. -- Michael
signature.asc
Description: PGP signature
