From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Steven Lembark <lembark(at)wrkhors(dot)com> |
Cc: | pgsql-general(at)lists(dot)postgresql(dot)org |
Subject: | Re: Selecting strict, immutable text for a composite type. |
Date: | 2018-05-10 21:38:48 |
Message-ID: | 13325.1525988328@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Steven Lembark <lembark(at)wrkhors(dot)com> writes:
> On Thu, 10 May 2018 14:41:26 -0400
> Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Steven Lembark <lembark(at)wrkhors(dot)com> writes:
>>> The problem is with gists telling me that they cannot index
>>> the type. This works for enums, just not the composite type.
>> Oh, well, they can't. There's no GiST opclass covering arbitrary
>> composite types. This doesn't seem very surprising to me given
>> the lack of operators that such an opclass might accelerate.
> But I thought that they could include functions of composite
> types that were indexable (e.g., text)?
Yeah, but that's not what you did.
I think you could make that work with
exclude using gist (
lat_lng_text(location) with =,
effective with &&
)
but it's not going to apply the function without you telling it to.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | David G. Johnston | 2018-05-10 21:41:53 | Re: Selecting strict, immutable text for a composite type. |
Previous Message | Peter J. Holzer | 2018-05-10 21:17:41 | Re: Domain based on TIMEZONE WITH TIME ZONE |