On 02/23/2017 04:41 PM, Tom Lane wrote:
> BTW, while I'm looking at this ... isn't gin_enum_cmp broken on its face?
> It's using DirectFunctionCall2 to call enum_cmp, and that's wrong because
> DirectFunctionCall2 does not pass through a flinfo but enum_cmp needs to
> have one. I've not tested, but I'm certain that this would dump core if
> asked to compare odd-numbered enum OIDs.
>
>
Yes, that's what I'm trying to fix.
cheers
andrew
--
Andrew Dunstan https://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services