Michael Paquier <[email protected]> writes:
> On Sun, Dec 28, 2025 at 04:12:16AM +0000, Japin Li wrote:
>> 2. I noticed that btree_gin, btree_gist, and pgcrypto also have relatively
>> long REGRESS lists, should we also update them?
> If we think about that in terms of frequency where new test files are
> added, it seems to me that there would be more meaning to do that for
> pgcrypto, test_decoding and pgss. This is more true for the two last
> entries of this list.
> About the rest? Not so much. Still, I was surprised to see that we
> have added stuff to btree_gist recently for WITHOUT OVERLAPS, so
> perhaps there is a point there.
I think a reasonable policy for this going forward is to mop up
the stragglers when and if another test case is added to each one.
regards, tom lane
