Re: Materialized views WIP patch - Mailing list pgsql-hackers

From Thom Brown
Subject Re: Materialized views WIP patch
Date
Msg-id CAA-aLv5oq9SXJqcERdPQnR8epM=rOMUVJrJeousdfrxOVzw==w@mail.gmail.com
Whole thread Raw
In response to Re: Materialized views WIP patch  ("Kevin Grittner" <kgrittn@mail.com>)
List pgsql-hackers
On 25 January 2013 06:00, Kevin Grittner <kgrittn@mail.com> wrote:
> Noah Misch wrote:
>
>> The patch conflicts with git master; I tested against master@{2013-01-20}.
>
> New patch rebased, fixes issues raised by Thom Brown, and addresses
> some of your points.

Thanks for the new version.  All previous issues I raised have been resolved.

I have an inconsistency to note between VIEWs and MATERIALIZED VIEWs:

CREATE VIEW v_test8 (meow, "?column?") AS SELECT 1 bark, 2;
CREATE MATERIALIZED VIEW mv_test8 (meow, "?column?") AS SELECT 1 bark, 2;

pg_dump output:

CREATE VIEW v_test8 AS   SELECT 1 AS meow, 2;

CREATE MATERIALIZED VIEW mv_test8 (   meow,   "?column?"
) AS   SELECT 1 AS meow, 2 WITH NO DATA;

The materialized view adds in column name parameters, whereas the
standard view doesn't.  But it seems to add column parameters
regardless:

CREATE VIEW v_test9 AS SELECT 1;
CREATE MATERIALIZED VIEW v_test9 AS SELECT 1;

CREATE VIEW v_test9 AS   SELECT 1;

CREATE MATERIALIZED VIEW mv_test9 (   "?column?"
) AS   SELECT 1 WITH NO DATA;

VIEWs never seem to use column parameters, MATERIALIZED VIEWs always
appear to use them.

--
Thom



pgsql-hackers by date:

Previous
From: Amit Kapila
Date:
Subject: Re: Proposal for Allow postgresql.conf values to be changed via SQL [review]
Next
From: Kohei KaiGai
Date:
Subject: Re: [sepgsql 1/3] add name qualified creation label