Re: SPI access to PostgreSQL query plan

From: "Florian G(dot) Pflug" <fgp(at)phlo(dot)org>
To: Cristiano Duarte <cunha17(at)gmail(dot)com>
Cc: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: SPI access to PostgreSQL query plan
Date: 2007-09-18 00:46:14
Message-ID: 46EF1FD6.8010305@phlo.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Cristiano Duarte wrote:
> 2007/9/17, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>:
>> "Cristiano Duarte" <cunha17(at)gmail(dot)com> writes:
>>> Is there a way to have access to PostgreSQL query plan and/or predicates
>>> inside a function using spi (or any other way)?
>> No.
>>
>> Hi Tom,
>
> "No" means: there is no way since the query plan is stored in a
> secret/safe/protected/non-visible/fort-knox like place :) or it means, there
> is no friendly way to do it with spi or casual c language programming?

"No" as in: You function is not told by the executor which filters are applied
to it's results, and since it might be called multiple times within a query you
cannot figure that out yourself, even if you somehow got hold of the currently
executed plan.

So unless you start to hack the executor in serious ways, you'll either have to
pass the filter condition manually to your function, or live with it producing
unnecessary output rows.

Thats only holds true for functions in languages other than pl/sql (Which is
*not* the same as pl/pgsql) - SQL functions can be inlined by the executor, and
then are subject to the usual optimizations. (So they essentially behave like
views).

greetings, Florian Pflug

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Sibte Abbas 2007-09-18 02:45:23 Re: Raw device I/O for large objects
Previous Message Georgi Chulkov 2007-09-18 00:17:04 Raw device I/O for large objects