Re: Partitions not Working as Expected - Mailing list pgsql-performance

From Andrew Dunstan
Subject Re: Partitions not Working as Expected
Date
Msg-id 51CC9746.5060804@dunslane.net
Whole thread Raw
In response to Re: Partitions not Working as Expected  (Shaun Thomas <sthomas@optionshouse.com>)
Responses Re: Partitions not Working as Expected  (Shaun Thomas <sthomas@optionshouse.com>)
List pgsql-performance
On 06/27/2013 03:14 PM, Shaun Thomas wrote:
> On 06/27/2013 01:42 PM, Tom Lane wrote:
>
>> That will break things: CURRENT_DATE will then be equivalent to just
>> writing today's date as a literal.
>
> Interesting. I tested it by creating a view and a table with a
> default, and it always seems to get translated to:
>
> ('now'::text)::date
>
> But I'll take your explanation at face value, since that doesn't imply
> what the output would be. What's interesting is that EnterpriseDB has
> their own pg_catalog.current_date function that gets called by the
> CURRENT_DATE keyword. So unlike in vanilla PG, I could mark just the
> current_date function as immutable without affecting a lot of other
> internals.
>
> On EDB, this actually works:
>
> UPDATE pg_proc
>    SET provolatile = 'i'
>  WHERE proname = 'current_date';


But that's a lie, surely. If it breaks you have nobody to blame but
yourself. There's a reason EDB haven't marked their function immutable -
it's not.

cheers

andrew



pgsql-performance by date:

Previous
From: Shaun Thomas
Date:
Subject: Re: Partitions not Working as Expected
Next
From: Shaun Thomas
Date:
Subject: Re: Partitions not Working as Expected