Re: [GENERAL] index scan and functions

2010-07-23 Thread arno
Le mardi 20 juillet 2010, à 10:11:21 +0200, Harald a écrit : 
 In article 20100719162547.ga17...@localhost,
 arno a...@renevier.net writes:
 
  Thanks, that's exactly what I was looking for.
 
 No, I'd say you're looking for the ip4r package which provides
 an indexable IP address range type.

Thanks, I'll look into it.


signature.asc
Description: Digital signature


Re: [GENERAL] index scan and functions

2010-07-20 Thread Harald Fuchs
In article 20100719162547.ga17...@localhost,
arno a...@renevier.net writes:

 Thanks, that's exactly what I was looking for.

No, I'd say you're looking for the ip4r package which provides
an indexable IP address range type.


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general


Re: [GENERAL] index scan and functions

2010-07-19 Thread Sam Mason
On Mon, Jul 19, 2010 at 05:55:48PM +0200, arno wrote:
 But when using a custom function to compute my where parameter

 inet_to_bigint is a function that transform an inet address its integer 
 representation.

 Is there a way, either to put function return value in a variable, or to tell
 postgres to still use a sequential scan ?

I'd guess your function is labeled as VOLATILE.  This is saying that
the function has side effects, but from the function's name I'd guess
that it doesn't and the only purpose of calling the function is to get
its return value.  I'd expect that labeling it as STABLE would cause PG
to do what you're expecting.  More details here:

  http://www.postgresql.org/docs/current/static/xfunc-volatility.html

-- 
  Sam  http://samason.me.uk/

-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general


Re: [GENERAL] index scan and functions

2010-07-19 Thread arno
Le lundi 19 juillet 2010, à 17:09:02 +0100, Sam a écrit : 
 On Mon, Jul 19, 2010 at 05:55:48PM +0200, arno wrote:
  But when using a custom function to compute my where parameter
 
  inet_to_bigint is a function that transform an inet address its integer 
  representation.
 
  Is there a way, either to put function return value in a variable, or to 
  tell
  postgres to still use a sequential scan ?
 
 I'd guess your function is labeled as VOLATILE.  This is saying that
 the function has side effects, but from the function's name I'd guess
 that it doesn't and the only purpose of calling the function is to get
 its return value.  I'd expect that labeling it as STABLE would cause PG
 to do what you're expecting.  More details here:
 
   http://www.postgresql.org/docs/current/static/xfunc-volatility.html

Thanks, that's exactly what I was looking for.


signature.asc
Description: Digital signature