Re: BUG #18831: Particular queries using gin-indexes are not interruptible, resulting is resource usage concerns.

Vinod Sridharan <vsridh90@gmail.com> writes:
> I managed to get a test that fails without my patch, and added it to
> the intarray tests. Without the patch, the query returns 0 records.
> With my patch, the results are correct (12 records).

Cool, thanks.

> Please find attached the updated patch with a fix & a test in the
> intarray contrib. Thanks for the pointer to the contrib extensions.

Pushed with light editorialization.  I noticed one other bug in the
same area: shimTriConsistentFn wasn't paying attention to the recheck
flag returned by its first call to the underlying consistentFn.
I think this is probably just a latent issue, because if any of the
later calls return recheck = true we'll end with the correct
result anyway.  And at least for the contrib modules that depend
on this, the recheck flag doesn't depend on the key-presence inputs.
But it's theoretically possible we'd get the wrong answer.

            regards, tom lane