From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Paul A Jungwirth <pj(at)illuminatedcomputing(dot)com> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: GiST support for UUIDs |
Date: | 2015-06-25 15:06:31 |
Message-ID: | 2724.1435244791@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Paul A Jungwirth <pj(at)illuminatedcomputing(dot)com> writes:
> I'm interested in adding GiST support for the UUID column type from
> the uuid-ossp extension. This has been requested and attempted before:
> I've used Postgres for a long time, but I've only dabbled a bit in the
> source code. So I'm curious where this change would go? The btree_gist
> extension? The uuid-ossp extension? Somewhere else?
btree_gist, I'd think, assuming you are only thinking of btree-equivalent
semantics and not anything more outre. uuid-ossp is only concerned with
UUID generation algorithms, not with storage or indexing. btree_gist
already has a bunch of infrastructure for this type of GiST opclass,
so it should be pretty straightforward to add it there (at least up to
making it work; the overhead of an extension version bump will probably
exceed the useful payload :-( ).
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2015-06-25 15:12:36 | Re: Oh, this is embarrassing: init file logic is still broken |
Previous Message | Tom Lane | 2015-06-25 15:00:29 | Schedule for 9.5alpha1 |