Re: Add GiST support for mixed-width integer operators - Mailing list pgsql-hackers

From Paul Jungwirth
Subject Re: Add GiST support for mixed-width integer operators
Date
Msg-id 59dfd569-f22b-4bab-aee0-88dde7dc12eb@illuminatedcomputing.com
Whole thread Raw
In response to Re: Add GiST support for mixed-width integer operators  ("Andrey M. Borodin" <x4mmm@yandex-team.ru>)
List pgsql-hackers
On 7/6/24 05:04, Andrey M. Borodin wrote:>> On 5 Jul 2024, at 23:46, Paul Jungwirth 
<pj@illuminatedcomputing.com> wrote:
>>
>> this commit adds support for all combinations of int2/int4/int8 for all five btree operators (</<=/=/>=/>).
> 
> Looks like a nice feature to have.
> Would it make sense to do something similar to float8? Or, perhaps, some other types from btree_gist?

Here is another patch adding float4/8 and also date/timestamp/timestamptz, in the same combinations 
as btree.

No other types seem like they deserve this treatment. For example btree doesn't mix oids with ints.

> Also, are we sure such tests will be stable?

You're right, it was questionable. We hadn't analyzed the table, and after doing that the plan 
changes from a bitmap scan to an index-only scan. That makes more sense, and I doubt it will change 
now that it's based on statistics.

Yours,

-- 
Paul              ~{:-)
pj@illuminatedcomputing.com
Attachment

pgsql-hackers by date:

Previous
From: "David G. Johnston"
Date:
Subject: Re: ❓ JSON Path Dot Precedence
Next
From: Michael Christofides
Date:
Subject: Re: Detoasting optionally to make Explain-Analyze less misleading