Stefan,
> My LEAST function would not work because
> my_timestamp_field has a datatype of "timestamp
> without time zone", and the TO_TIMESTAMP(...) creates
> a "timestamp *with* time zone". I could not find
> anything in the documentation about this behavior. I
> am running v7.3.2.
Just do a:
Hi Chris,
You need to start a transaction explicitly:
Session 1:
buns=# begin;
BEGIN
buns=# create table test_table (col1 int);
CREATE TABLE
Session 2:
buns=# \d test_table
Did not find any relation named "test_table".
Session 1:
buns=# commit;
COMMIT
Session 2:
buns=# \d test_table
Tabl
Thanks for all your help, Scott. A friend of mine whipped out this
script which runs a lot faster than trying to use the original query I
posted.
-M@
begin;
create temporary table cart_temp as select distinct a.cart_id,
a.cart_cookie from v_carts a, v_cart_contents b where a.cart_id =
b.cart
It's good to hear this kind of discussion going on!
I solved my problem (for now) by creating a bunch of
overloaded LEAST and GREATEST functions, one for each
datatype. They only take two parameters, but that's
fine for what we're doing.
However, I ran into another, unrelated problem. I
created
I know this is basic, but couldn\'t find and in a hurry to know the answer. When
interfacing with PostgreSQL through PSQL, it appears that DML statements are
auto-commited, that is, a change I make in one session is seen from another without
the original session issueing a COMMIT. Is this a re
Yep, we get asked to do that quite often. Use statement_timeout before
the LOCK command. If the timeout happens, the LOCK, and hence
transaction will abort.
---
The Hermit Hacker wrote:
>
> Simple (I think) question ... i
Simple (I think) question ... is there a way of having an application
attempt to acquire a LOCK on a table *without* it blocking? Right now, if
you try to LOCK a table that another process has LOCKed, it will hang
indefinitely waiting for the other LOCK to drop ... is there a way of
setting a 'ti
On Wednesday 02 July 2003 18:49, Andreas Schmitz wrote:
> Hello *,
>
> I have a litlle problem writing a plpgsql trigger function. I am in need to
> get some interpretaion of the TG_OP within an IF clause to decide with
> action will be taken. I tried it that way:
>
> IF (TG_OP=DELETE AND
Hello *,
I have a litlle problem writing a plpgsql trigger function. I am in need to
get some interpretaion of the TG_OP within an IF clause to decide with action
will be taken. I tried it that way:
IF (TG_OP=DELETE AND check_count > 1) THEN
UPDATE tipp_team
Ang,
> Any chance of this making it into 7.4's contrib? Maybe with enough
> documentation to make it a tutorial for PostgreSQL's user functions?
Er, no. Feature freeze was Tuesday. And you haven't submitted a patch yet.
> Wow, less reasons to use VIEWs when CREATE FUNCTION can be just as fast,
Ang Chin Han wrote:
Tom Lane wrote:
create function greatest(anyelement, anyelement) returns anyelement as
'select case when $1>$2 then $1 else $2 end' language sql;
Any chance of this making it into 7.4's contrib? Maybe with enough
documentation to make it a tutorial for PostgreSQL's user functi
On Wed, 2003-07-02 at 11:30, Ralf Werny wrote:
> Hi,
> many clients like webmin and openoffice makes an
> INSERT NULL if i give no value for a field because it is a sequence.
> Is there a better way to solve this problem as using a trigger ?
A trigger (maybe a rule) is the way to go about th
Hi,
many clients like webmin and openoffice makes an
INSERT NULL if i give no value for a field because it is a sequence.
Is there a better way to solve this problem as using a trigger ?
CREATE TABLE test (
id integer PRIMARY KEY
);
CREATE SEQUENCE test_seq_id MINVALUE
What locales are the two databases initdb'd to?
On Wed, 2 Jul 2003 [EMAIL PROTECTED] wrote:
> Hello,
>
> I used to run a program on a box with postgres 7.1
> There was data like:
> 'abc 234'
> 'abc 1234'
>
> (not the extra blank before 234)
>
> so after a select * from x order by field i got
Hello,
I used to run a program on a box with postgres 7.1
There was data like:
'abc 234'
'abc 1234'
(not the extra blank before 234)
so after a select * from x order by field i got
'abc 234'
'abc 1234'
but since this program runs on a 7.3 postgres version
i get.
'abc 1234'
'abc 234'
it see
Both link work nicely, thanks.
I don't know what could be the problem with the others.
But it's not important by now
Thank you.
-- Csaba
- Original Message -
From: "Tomasz Myrta" <[EMAIL PROTECTED]>
To: "Együd Csaba" <[EMAIL PROTECTED]>
Sent: Wednesday, July 02, 2003 12:06 PM
Subject
Dear All,
thank you for your help, it was really efficient.
I'll get by with it now. Thanks.
-- Csaba
- Original Message -
From: "Bruno Wolff III" <[EMAIL PROTECTED]>
To: "Együd Csaba" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Friday, June 27, 2003 5:10 PM
Subject: Re: [SQL] Gett
Tom Lane wrote:
create function greatest(anyelement, anyelement) returns anyelement as
'select case when $1>$2 then $1 else $2 end' language sql;
Any chance of this making it into 7.4's contrib? Maybe with enough
documentation to make it a tutorial for PostgreSQL's user functions?
You do have to
18 matches
Mail list logo
