Re: Materialized views WIP patch

On 2013-02-21 14:11:10 -0800, Kevin Grittner wrote:
> Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > Kevin Grittner <kgrittn(at)ymail(dot)com> writes:
>
> >> What did you think of the idea of something like DISCARD
> >> MATERIALIZED VIEW DATA as a new statment?  Or maybe RESET
> >> MATERIALIZED VIEW?
> >
> > I could live with either DISCARD or RESET.
>
> I figured this was worth a pass through the keyword list to look
> for all imperative verbs suitable for this, which could support the
> needed syntax without adding a new keyword.  Here are the
> possibilities I came up with, along with a note about why they are
> keywords already.
>
> DISABLE MATERIALIZED VIEW mv;  -- ALTER clause for constraints
> DISCARD MATERIALIZED VIEW DATA mv;  -- session state
> RELEASE MATERIALIZED VIEW DATA mv;  -- savepoint
> RESET MATERIALIZED VIEW DATA mv;  -- run-time parameter
>
> I think any of these could work.  I'm personally most inclined
> toward DISABLE MATERIALIZED VIEW.  It seems to convey the semantics
> better, especially if you leave out DATA as an additonal word.

> Given that a materialized view will retain its query, tablespace,
> indexes, statistics targets, etc. with this operation, and will
> just not be available for scanning, some of the above seem
> downright misleading without DATA thrown in.

I vote for RESET or DISCARD. DISABLE sounds more like you disable
automatic refreshes or somesuch.

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services