[SQL] Re: Auto incrementing an integer

Take a look at the SERIAL datatype as well as sequences (CREATE SEQUENCE, NEXTVAL, CURRVAL). good luck! On Mon, 14 May 2001, Sylte wrote: > How do I construct a datatype that autoincrement in postgreSQL? > > Thanks > Sylte > > > > > ---(end of broadcast)-

[SQL] fatal ERROR running simple join query...

I have a problem with postgres, I'm running a query that joins two tables, one with 129000+ records, and the other with 1172 records, it's a very simple join. But postgres can't handle it!!! (Both tables have an index on the field "cuentacb"). I get the following error after 3 minutes... migr

[SQL] Auto incrementing an integer

How do I construct a datatype that autoincrement in postgreSQL? Thanks Sylte ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])

Re: [SQL] can't get rid of unnesesary SORT step in explain plan for hash join

On Sun, 13 May 2001, Tom Lane wrote: > Alexey Nalbat <[EMAIL PROTECTED]> writes: > > So, my question is: how can I get rid of this unnesesary "Sort" step > > in the execution plan for hash join? > > You can't, because it's not unnecessary. Hash join doesn't promise > to produce its outputs in an

Re: [SQL] Using ORDER BY with AGGREGATE/GROUP BY in a SELECT statement

Tom, Thanks for the reply, and for correcting my broken SQL. :^) A couple days after I sent my own follow-up message, I realized how silly my question was. I'll let future questions sit a bit longer next time. To make up for my silly question, here is a function + aggregate I created while w

[SQL] working with stored procedures

Hi , I am trying to create a stored procedure that returns a user defined type. The following piece of code shows the way I am trying to do it. regy1=> create function regyuser_rec(int) returns varchar as ' declareregy1'> v_userid alias for $1;regy1'> output_rec varchar(400);regy1'> user_rec

[SQL] error message...

Hi I have Postgres 7.0.3/6 on a Suse Professional 7.1 (kernel 2.4.0) machine. The database is used very often and I see in the logfile the error message: Sorry, too many clients already Can I set the number of 'active' clients? And where can I set this? And How? Thanx in advance! Sven Frank

[SQL] Case

Hello, what's wrong when you receive this message from an select statement : CASE types "date" and "text" not matched thank you for any help -- Best regards, Cwhisperer mailto:[EMAIL PROTECTED] ---(end of broadcast)---

Re: [SQL] execute client application from PL/pgSql

hi, phpPGAdmin is a web based php driven postgresql admin tool. not sure of the exact url, try google :) it has a pg_dump option in it. jeff On Sat, 12 May 2001, datactrl wrote: > Date: Sat, 12 May 2001 10:23:39 +1000 > From: datactrl <[EMAIL PROTECTED]> > To: [EMAIL PROTECTED] > Subject: Re

Re: [SQL] Re: Informix->PostgreSQL database convertion

is there an informix utility called dbschema which does this? dbschema -t all -d databasename > schema.sql Sylte writes: > How is the SQL command UNLOAD (informix) used to extract both data and > database architecture? > > Is it able to create a file of SQL statements like when using pos

[SQL] Counting rows in a table

MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii I have a query as follows: SELECT date(c.timestart) as ondate, c.userid, u.username, count(c.callinfoid) as numCalls, sum(b.broquant) as numBroc, sum(b.broquant)/count(c.callinfoid) as BrochuresaCall

Re: [SQL] can't get rid of unnesesary SORT step in explain plan for hash join

Alexey Nalbat <[EMAIL PROTECTED]> writes: > While executing this query postgres at first creates hash on table > "resellers", then get from index "products_mcr" for rows with > "m_id=123" already ordered (!!!) pairs "c_id,r_id", for each that > pair it checks join condition using hash. If postger

[SQL] unwritable /tmp causes initdb to crash

this is a goofy situtation. we had a new linux 2.4.4 kernel with a /tmp not writable by postgres. granted, this is categorically hosed, but initdb still dumped core under these circumstances. changing the /tmp to writable immediately fixed the problem. -j = John Scott Senior Partner Au