Re: Incremental sort for access method with ordered scan support (amcanorderbyop) - Mailing list pgsql-hackers

From Richard Guo
Subject Re: Incremental sort for access method with ordered scan support (amcanorderbyop)
Date
Msg-id CAMbWs49m7GCDw_gsSrE4etZR1FwvOUfor0P8J2y8sXW0EcK86Q@mail.gmail.com
Whole thread Raw
In response to Re: Incremental sort for access method with ordered scan support (amcanorderbyop)  (David Rowley <dgrowleyml@gmail.com>)
Responses Re: Incremental sort for access method with ordered scan support (amcanorderbyop)
List pgsql-hackers

On Tue, Jul 4, 2023 at 7:15 PM David Rowley <dgrowleyml@gmail.com> wrote:
On Tue, 4 Jul 2023 at 20:12, Richard Guo <guofenglinux@gmail.com> wrote:
> The v4 patch looks good to me (maybe some cosmetic tweaks are still
> needed for the comments).  I think it's now 'Ready for Committer'.

I agree. I went and hit the comments with a large hammer and while
there also adjusted the regression tests. I didn't think having "t" as
a table name was a good idea as it seems like a name with a high risk
of conflicting with a concurrently running test. Also, there didn't
seem to be much need to insert data into that table as the tests
didn't query any of it.

The only other small tweak I made was to not call list_copy_head()
when the list does not need to be shortened. It's likely not that
important, but if the majority of cases are not partial matches, then
we'd otherwise be needlessly making copies of the list.

I pushed the adjusted patch.

The adjustments improve the patch a lot.  Thanks for adjusting and
pushing the patch.

Thanks
Richard

pgsql-hackers by date:

Previous
From: Amul Sul
Date:
Subject: Dumping policy on a table belonging to an extension is expected?
Next
From: Masahiko Sawada
Date:
Subject: Re: doc: improve the restriction description of using indexes on REPLICA IDENTITY FULL table.