Обсуждение: [SQL] Index on date_trunc

X-From-Line: owner-pgsql-sql@hub.org  Fri Apr 23 06:48:33 1999
Return-Path: <owner-pgsql-sql@hub.org>
Delivered-To: labouiss@cybercable.fr
Received: from mail.cybercable.fr
    by localhost with POP3 (fetchmail-5.0.1)
    for moi@localhost (single-drop); Fri, 23 Apr 1999 06:48:33 +0200 (CEST)
Received: (qmail 4517 invoked from network); 22 Apr 1999 22:31:12 +0200
Received: from hub.org (209.47.145.100)
  by bouvreuil.cybercable.fr with SMTP; 22 Apr 1999 20:31:12 -0000
Received: from localhost (majordom@localhost)
    by hub.org (8.9.3/8.9.1) with SMTP id QAA12080;
    Thu, 22 Apr 1999 16:30:50 -0400 (EDT)
    (envelope-from owner-pgsql-sql@hub.org)
Received: by hub.org (TLB v0.10a (1.23 tibbs 1997/01/09 00:29:32)); Thu, 22 Apr 1999 16:29:27 +0000 (EDT)
Received: (from majordom@localhost)
    by hub.org (8.9.3/8.9.1) id QAA11449
    for pgsql-sql-outgoing; Thu, 22 Apr 1999 16:29:25 -0400 (EDT)
    (envelope-from owner-pgsql-sql@postgreSQL.org)
Received: from gabuzo.meunet (IDENT:root@d224.paris-204.cybercable.fr [212.198.204.224])
    by hub.org (8.9.3/8.9.1) with ESMTP id QAA11262
    for <pgsql-sql@postgreSQL.org>; Thu, 22 Apr 1999 16:27:09 -0400 (EDT)
    (envelope-from christophe.labouisse@dial.oleane.com)
Received: (from moi@localhost)
    by gabuzo.meunet (8.9.1a/8.9.1/Christophe Labouisse - 10/01/99) id WAA06056;
    Thu, 22 Apr 1999 22:21:50 +0200
Mime-Version: 1.0 (generated by tm-edit 7.108)
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <m3r9pcbe3m.fsf@gabuzo.meunet>
X-Mailer: Gnus v5.5/XEmacs 20.4 - "Emerald"
Sender: owner-pgsql-sql@postgreSQL.org
Precedence: bulk
X-Sorted: ML-PSQL-SQL
Lines: 18
Xref: gabuzo.meunet ML-PSQL-SQL:676

I'm trying to index on a part of a datetime field. I tried the
following statement :

create index ns_dt1_idx on netstats (date_trunc('day',NS_DATE) datetime_ops);

It doesn't work and I get :

ERROR:  parser: parse error at or near "'"

Any idea ?

Christophe Labouisse : Cinéma, typographie, Unix
christophe.labouisse@dial.oleane.com http://www.multimania.com/gabuzo/
Le cinéma en Lumière : http://www.lumiere.org/

At 08:19 +0300 on 30/04/1999, Christophe Labouisse wrote:


> create index ns_dt1_idx on netstats (date_trunc('day',NS_DATE) datetime_ops);
>
> It doesn't work and I get :
>
> ERROR:  parser: parse error at or near "'"
>
> Any idea ?

Seems as if the syntax requires that all the arguments for the function
should be attributes. That is, columns from the table rather than literals
of whatever kind.

Solution: create a function that hides the 'day' in it:

testing=> \d test

Table    = test
+-----------------------------+----------------------------------+-------+
|              Field          |              Type                | Length|
+-----------------------------+----------------------------------+-------+
| zman                        | datetime                         |     8 |
+-----------------------------+----------------------------------+-------+
testing=> create index zman_index on test
testing-> (date_trunc( 'day', zman ) datetime_ops );
ERROR:  parser: parse error at or near "'"
testing=> create function day_trunc( datetime ) returns datetime
testing-> as 'SELECT date_trunc( ''day'', $1 )'
testing-> language 'sql';
CREATE
testing=> create index zman_index on test
testing-> (day_trunc( zman ) datetime_ops );
CREATE

Just remember to use the same function for the queries you make on the
tables. Otherwise PostgreSQL won't use this index, like
  SELECT *  FROM test  WHERE day_trunc( zman ) = '1999-08-01';

Herouth

--
Herouth Maoz, Internet developer.
Open University of Israel - Telem project
http://telem.openu.ac.il/~herutma

Herouth Maoz <herouth@oumail.openu.ac.il> writes:
> At 08:19 +0300 on 30/04/1999, Christophe Labouisse wrote:
>> create index ns_dt1_idx on netstats (date_trunc('day',NS_DATE) datetime_ops);

> Seems as if the syntax requires that all the arguments for the function
> should be attributes.

Yes, I believe that's the case.  It's not only the parser that's
limited, either: the physical representation of indices doesn't
currently have room for anything more than a function OID to define
what the index sort function is.  So there's noplace to put a constant
value.

> Solution: create a function that hides the 'day' in it:

Good thought, but it doesn't actually work, as you'll find as soon
as there are any entries in the table:

insert into test values ('1999-08-01 10:15');
ERROR:  SQL-language function not supported in this context.

(6.4.x gives a misleading error message, but it's the same restriction.)

It'd be nice to allow SQL functions to be used for indexes, but I'm not
sure what it'd take to make it happen.  At the very least there'd have
to be some drastic restrictions on what the function could do (imagine
the carnage if the function tries to modify the table the index is
being built for...)

Bottom line is there's no easy way to do this right now :-(.  What I'd
suggest is keeping a separate column that is the day part of the date
and indexing that.  You could use a rule to update that column
automatically whenever the main timestamp column is set.  Alternatively,
try to restructure your queries so that you don't actually need an index
on the day part of the date...
        regards, tom lane

At 16:52 +0300 on 03/05/1999, Tom Lane wrote:


> Good thought, but it doesn't actually work, as you'll find as soon
> as there are any entries in the table:
>
> insert into test values ('1999-08-01 10:15');
> ERROR:  SQL-language function not supported in this context.

Argh. Chastisement taken. But it really ought to have said that when I
tried creating the index, oughtn't it? :)

Herouth

--
Herouth Maoz, Internet developer.
Open University of Israel - Telem project
http://telem.openu.ac.il/~herutma