Tom,
> Damn! When is the "holy grail" of PostgreSQL going to be
> stable enough to use? Beta3 still has a "not advisable
> for
> production" warning, and I'm being tied up in knots by
> the
> number of things I need in 7.1.
Ooops! That may have sounded a little harsh. I am a bit
desperate, b
Folks,
ALTER TABLE won't work until 7.1. CUrrently, I have a table
that needs one small change, but it's refrenced as a foriegn
key by 7 other tables. Any suggestions on how I can make
the table change without having to drop and re-create 8
tables?
-Josh Berkus
Tom,
> Not only varchar --- any other parameters, period. And
> not only that,
> but the result is taken as NULL no matter what you try to
> return.
Not quite. I tried the following:
Parameters: $1=integer, $2=NULL, $3=varchar
And I had the function test for nulls. It read the first
paramet
"Josh Berkus" <[EMAIL PROTECTED]> writes:
> 4. If I pass a NULL to any of the parameters of a PL/PGSQL
> function, any (other) VARCHAR parameters are set to NULL as
> well.
Not only varchar --- any other parameters, period. And not only that,
but the result is taken as NULL no matter what you
Folks,
Oh, yes, one more:
4. If I pass a NULL to any of the parameters of a PL/PGSQL
function, any (other) VARCHAR parameters are set to NULL as
well.
Thanks!
-Josh
Folks,
1. While I am able to use the %TYPE declaration within
PL/PGSQL functions, I am unable to use this declaration in
the parameters for the function -- I get 'Parse Error at or
near "."'
2. When I have a PL/PGSQL function return a custom message
using a VARCHAR return value, I get backslashe
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
Can not create pgdump_oid table. Explanation from backend: 'ERROR:
cannot create pgdump_oid
Is there a way to repair this?
I would guess that this means there is a file named "pgdump_oid" hanging
around in the database directory, no d
> >Can not create pgdump_oid table. Explanation from backend: 'ERROR:
> >cannot create pgdump_oid
> >
> >Is there a way to repair this?
> >
>
> If there was an earlier crash while running pg_dump, the table may already
> exist. If so, try deleting it. Also, can you confirm that the username
> un
> > create table testdate (field1 date);
> > insert into testdate values ('2000-09-30');
> > insert into testdate values ('2000-10-20');
> > insert into testdate values ('2000-11-25');
> > select * from testdate where field1 between '2000-10-01' and
> > '2000-11-30' ;
>
> >field1
> > -
At 22:13 15/01/01 +0200, Johann Spies wrote:
>
>Can not create pgdump_oid table. Explanation from backend: 'ERROR:
>cannot create pgdump_oid
>
>Is there a way to repair this?
>
If there was an earlier crash while running pg_dump, the table may already
exist. If so, try deleting it. Also, can you
Mauricio Hipp Werner writes:
> I need help, which is the symbol used in postgreSql to carry out the outer
> join.
>
> in oracle the is used (+)
> in sybase the is used * and
> in postgreSql?
No symbol, just words.
http://www.postgresql.org/devel-corner/docs/postgres/sql-select.htm
--
Peter E
Version 6.5.3 on Debian Potato
My daily cron job reported :
Can not create pgdump_oid table. Explanation from backend: 'ERROR:
cannot create pgdump_oid
Is there a way to repair this?
Johann
--
J.H. Spies - Tel. 082 782 0336
"For I know that my redeemer liveth, and that he shall
Hi,
About a week back I'd posted a message asking for help with a script
that as far as I could tell was well formed. The problem was with 2
INSERT statements directly after the CREATE TABLE statement of the
table that I was trying to insert data into. I'd used C single/multi
line comments of t
Hi together,
I am still trying to implement my "userlog". At the moment I can insert
a message like "User xy added". What I want to have is, that I can
insert something like "User xy added by z". Can anyone tell me, how to
parse a parameter to the trigger (I think, only the application knows
what
Hello,I am trying to use a perl script to break an input string
into keywords andpull out all the records that match any of them:my
(@chips)= split(' ', $f_name);foreach my $chip (@chips)
{ next if ($stoppers=~/ $chip / || length($chip) <
3); if ($query) {$query .= " OR name ~* '.*$chi
> select user_name, date_part( 'epoch' , timestamp 'acct_timestamp') from
> tbacct limit 2;
>
> it said ERROR: Bad timestamp external representation 'acct_timestamp'
> how should i represent date_part( 'epoch' , timestamp 'acct_timestamp')
> to work?
select user_name, date_part ('epoch', acct
Hello!
again i was faced to a problem i was trying to do create view in wich i
had the next fields:
CREATE TABLE "tbacct" (
"user_name" character varying(32),
"nas_identifier" character varying(15),
"nas_port" int4,
"acct_session_id" character varying(15),
- Original Message -
From: "Renato De Giovanni" <[EMAIL PROTECTED]>
> Hi,
>
> Is there any SQL workaround to get the right results from the select
> statement bellow? Or am I doing something wrong??
>
> select * from testdate where field1 between '2000-10-01' and
> '2000-11-30' ;
>
>
Alexaki Sofia <[EMAIL PROTECTED]> writes:
> Initially I loaded all data in one transaction. Subsequently, I increased
> the number of buffers and disabled fsync() (-o -F) and I loaded the
> data again but the performance was almost unchanged. Does it make sense??
> How can I improve performance?
Renato De Giovanni <[EMAIL PROTECTED]> writes:
> create table testdate (field1 date);
> insert into testdate values ('2000-09-30');
> insert into testdate values ('2000-10-20');
> insert into testdate values ('2000-11-25');
> select * from testdate where field1 between '2000-10-01' and
> '2000-11-
On Mon, Jan 15, 2001 at 12:19:56 -0200, Renato De Giovanni wrote:
> select * from testdate where field1 between '2000-10-01' and
> '2000-11-30' ;
>
>field1
>
> 2000-09-30 < why is it here??
> 2000-10-20
> 2000-11-25
It works fine for me (7.0.3, Debian GNU/Linux "unstabl
Hi,
Is there any SQL workaround to get the right results from the select
statement bellow? Or am I doing something wrong??
create table testdate (field1 date);
insert into testdate values ('2000-09-30');
insert into testdate values ('2000-10-20');
insert into testdate values ('2000-11-25');
sel
Hello,
A)
I am going to load a huge amount of data in the DBMS using JDBC
and I want to reduce as much as possible the required loading time.
Initially I loaded all data in one transaction. Subsequently, I increased
the number of buffers and disabled fsync() (-o -F) and I loaded the
data again
Hi,
I recently posted a problem with connecting to a pg db server from access.
Now I found that the pg/odbc driver under windows lets choose between three
protocols, 6.x, 6.y, 6.z. I am using pg 7.x as the db server. Could this be
the problem? I managed to get the tables linked and shown, but
24 matches
Mail list logo
