I have a table t1 in a Postgres database which is periodically updated
with data from a proprietary database. To date, the easiest way to update
the data is to
begin transaction;
delete from t1;
insert into t1 (each row from the proprietary database);
commit;
Table
* Tom Lane in "Re: [GENERAL] Can this be done?" dated 2000/11/15 00:42
* wrote:
> Ashley Clark <[EMAIL PROTECTED]> writes:
> > Can I defer constraint checking on unique indexes?
>
> No, not at present. Consider restructuring your UPDATE into
> sequential steps so that the intermediate states ca
Ashley Clark <[EMAIL PROTECTED]> writes:
> Can I defer constraint checking on unique indexes?
No, not at present. Consider restructuring your UPDATE into sequential
steps so that the intermediate states can't cause index failures. Or
just don't make the index unique...
I am implementing some set-based trees in a table and I've been writing
some functions to manipulate (delete branch, insert node, move branch)
them and I've run into a problem with the move_branch function.
When I try a select move_tree(13, 25, 0); to move the branch labeled g
to be a child of th
Philip Hallstrom <[EMAIL PROTECTED]> writes:
> I downloaded 7.0.3 from ftp.postgresql.org yesterday (monday 14th). It
> didn't say anything about a pre-release... is there some way to tell if I
> have the final version?
Look at src/backend/utils/mmgr/portalmem.c --- if it says v 1.36.2.1
then yo
Ever since I upgraded to 6.5.2 I've had problems upgrading to the 7.x
series. I'm running Red Hat 6.1 (2.2.17 kernel) on this box.
Now I'm trying to upgrade to 7.0.3 via the rpm. When I tried both the
freshen (-F) and upgrade (-U) switches, I found a failed dependency:
/lib/gcc! and a bunch o
Putting the WHERE clause before the GROUP BY clause worked. I'm too
stupid or too tired to think clearly. :)
Thanks all
Shawn
On Tue, 14 Nov 2000, Michael Fork wrote:
> Try these queries...
>
> SELECT injdate, outcome, count(outcome) FROM safety GROUP BY
> injdate,outcome;
>
> or
>
> SELECT o
Try these queries...
SELECT injdate, outcome, count(outcome) FROM safety GROUP BY
injdate,outcome;
or
SELECT outcome, count(outcome) FROM safety WHERE injdate='2000-11-14'
GROUP BY outcome;
Michael Fork - CCNA - MCP - A+
Network Support - Toledo Internet Access - Toledo Ohio
On Tue, 14 Nov
> Adding a WHERE clause to the query doesn't work for example:
>
> SELECT outcome,count(outcome) from safety
> GROUP BY outcome
> WHERE InjDate='2000-11'14';
A WHERE clause must come before a GROUP BY clause.
Also note that if you are using a table with the row defined as datetime
instead of dat
when will be released?
valter mazzola, Italy.
_
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
Share information about yourself, create your own public profile at
http://profiles.msn.com.
I have a table defined as follows:
CREATE TABLE safety (
sid SERIAL,
side INTEGER,
InjDate DATE,
InjType INTEGER,
Outcome INTEGER,
PRIMARY KEY (sid)
);
What I'm trying to generate is a list of all outcomes and how many of each
that occured for a pa
I downloaded 7.0.3 from ftp.postgresql.org yesterday (monday 14th). It
didn't say anything about a pre-release... is there some way to tell if I
have the final version?
On Sun, 12 Nov 2000, The Hermit Hacker wrote:
>
> After a couple of pre-release tarballs, the PostgreSQL Developers are
> pro
On Tue, 14 Nov 2000, Beth K. Gatewood wrote:
> >
>
> Stephan-
>
> Thank you so much for taking the effort to answer this these questions. You
> help is truly appreciated
>
> I just have a few points for clarification.
>
> >
> > MATCH PARTIAL is a specific match type which describes whic
13 matches
Mail list logo
