Re: forced sequential scan when condition has current_user - Mailing list pgsql-performance

From Erik Jones
Subject Re: forced sequential scan when condition has current_user
Date
Msg-id E56D1346-6E42-40D5-9CF1-239F5104F3A5@engineyard.com
Whole thread Raw
In response to Re: forced sequential scan when condition has current_user  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: forced sequential scan when condition has current_user
Re: forced sequential scan when condition has current_user
List pgsql-performance
On Jan 4, 2010, at 1:59 PM, Robert Haas wrote:

> The thing is, PostgreSQL doesn't know at planning time what the value of
> current_user() will be, so the plan can't depend on that; the planner
> just takes its best shot.

current_user() is a stable function and the manual is explicit that the result of stable function can be used in an
indexscan: 

"A STABLE function cannot modify the database and is guaranteed to return the same results given the same arguments for
allrows within a single statement. This category allows the optimizer to optimize multiple calls of the function to a
singlecall. In particular, it is safe to use an expression containing such a function in an index scan condition.
(Sincean index scan will evaluate the comparison value only once, not once at each row, it is not valid to use a
VOLATILEfunction in an index scan condition.)" 

postgres=# select provolatile from pg_proc where proname = 'current_user';
 provolatile
-------------
 s

So, I think the OP's question is still valid.

Erik Jones, Database Administrator
Engine Yard
Support, Scalability, Reliability
866.518.9273 x 260
Location: US/Pacific
IRC: mage2k






pgsql-performance by date:

Previous
From: Scott Marlowe
Date:
Subject: Re: DB is slow until DB is reloaded
Next
From: Madison Kelly
Date:
Subject: Re: DB is slow until DB is reloaded