On 08/01/2019 23:26, Corey Huinker wrote:
> On Fri, Jan 4, 2019 at 7:49 AM Peter Eisentraut
> <peter.eisentraut@2ndquadrant.com> wrote:
>>
>> On 25/12/2018 00:56, Corey Huinker wrote:
>>> The regression diff (attached) seems to imply that the triggers simply
>>> are not firing, though.
>>
>> The reason for this was explained by Dean. If you take out the check
>> that he mentioned, then your trigger fires but crashes. In your changed
>> unique_key_recheck(), "slot" is not initialized before use (or ever).
>
> Thanks. I'll be revisiting this shortly. Dean's information made me
> think the potential for a gain is smaller than initially imagined.
I think those are independent considerations. The "recheckIndexes"
logic just tracks what indexes have potential conflicts to check later.
Whether that checking later happens in a row or statement trigger should
not matter. What you need to do is adapt the "recheckIndexes" logic
from row triggers to statement triggers, e.g., expand
ExecASInsertTriggers() to look more like ExecARInsertTriggers().
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services