Re: Nested query performance issue - Mailing list pgsql-performance

From Tom Lane
Subject Re: Nested query performance issue
Date
Msg-id 21534.1239370946@sss.pgh.pa.us
Whole thread Raw
In response to Re: Nested query performance issue  (Glenn Maynard <glennfmaynard@gmail.com>)
List pgsql-performance
Glenn Maynard <glennfmaynard@gmail.com> writes:
> http://www.postgresql.org/docs/8.3/static/xfunc-sql.html says this is
> deprecated (though no deprecation warning is being generated):

>> Currently, functions returning sets can also be called in the select list of a query. For each row that the query
generatesby itself, the function returning set is invoked, and an output row is generated for each element of the
function'sresult set. Note, however, that this capability is deprecated and might be removed in future releases. 

The way to parse that is "we don't like this and we will get rid of it
if we can ever figure out a good substitute".  Right now there is no
100% substitute, so it stays.  (In fact, 8.4 will extend the feature so
it works in cases that don't work today, like for PL functions.)

There are, however, good reasons not to like it, such as the rather
questionable behavior if there's more than one SRF in the same select
list.  Don't complain if you run into that wart.

            regards, tom lane

pgsql-performance by date:

Previous
From: Ognjen Blagojevic
Date:
Subject: Postgres 8.x on Windows Server in production
Next
From: Tom Lane
Date:
Subject: Re: Shouldn't the planner have a higher cost for reverse index scans?