Re: PoC plpgsql - possibility to force custom or generic plan

2017-09-05 15:01 GMT+02:00 Daniel Gustafsson <daniel(at)yesql(dot)se>:

> > On 08 Apr 2017, at 09:42, Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> wrote:
> >
> > 2017-04-08 2:30 GMT+02:00 Peter Eisentraut <
> peter(dot)eisentraut(at)2ndquadrant(dot)com <mailto:peter(dot)eisentraut(at)2ndquadrant(dot)com
> >>:
> > On 4/6/17 14:32, Pavel Stehule wrote:
> > > I like to see any proposals about syntax or implementation.
> > >
> > > Using PRAGMA is one variant - introduced by PLpgSQL origin - Ada
> > > language. The PRAGMA syntax can be used for PRAGMA autonomous with well
> > > known syntax. It scales well - it supports function, block or command
> > > level.
> >
> > I had pragmas implemented in the original autonomous transactions patch
> > (https://www.postgresql.org/message-id/659a2fce-b6ee-06de-
> 05c0-c8ed6a01979e(at)2ndquadrant(dot)com <https://www.postgresql.org/
> message-id/659a2fce-b6ee-06de-05c0-c8ed6a01979e(at)2ndquadrant(dot)com>).
> > However, the difference there is that the behavior is lexical, specific
> > to plpgsql, whereas here you are really just selecting run time
> > behavior. So a GUC, and also something that could apply to other
> > places, should be considered.
> >
> > I'll look there - we coordinate work on that.
>
> This patch was moved to the now started commitfest, and is marked as “Needs
> review”. Based on this thread I will however change it to "waiting for
> author”,
> since there seems to be some open questions. Has there been any new work
> done
> on this towards a new design/patch?
>

I didn't any work on this patch last months. I hope so this week I reread
this thread and I'll check what I do.

There are few but important questions:

1. we want this feature? I hope so we want - because I don't believe to
user invisible great solution - and this is simple solution that helps with
readability of some complex PL procedures.

2. what syntax we should to use (if we accept this feature)? There was not
another proposal if I remember well - The PRAGMA syntax is strong because
we can very well specify to range where the plans caching will be
explicitly controlled. It is well readable and static.

Regards

Pavel

>
> cheers ./daniel