Travis Hume writes:
> So at this point all selects are screwed unless I issue a ROLLBACK;
> Why? Is there a way for SELECTs (or all SQL) statement that fail to NOT
> have this behaviour?
You probably set autocommit to off. Turn it on.
--
Peter Eisentraut [EMAIL PROTECTED]
culley harrelson <[EMAIL PROTECTED]> writes:
> http://marc.theaimsgroup.com/?l=postgresql-general&m=105551123915183&w=2
> I am wondering if things are going to change in 7.4?
Did you actually read the cited links?
regards, tom lane
---(end of broad
"Nigel J. Andrews" <[EMAIL PROTECTED]> writes:
> If in a trigger one does a select * from the table the trigger is on
> is it safe to assume that tg_trigtuple (and so tg_newtuple) will use
> the same TupleDesc as that returned from the select *, i.e. the
> tg_relation->rd_att TupleDesc?
Given rece
On Thu, 19 Jun 2003, Peter Eisentraut wrote:
> Travis Hume writes:
>
> > So at this point all selects are screwed unless I issue a ROLLBACK;
> > Why? Is there a way for SELECTs (or all SQL) statement that fail to NOT
> > have this behaviour?
>
> You probably set autocommit to off. Turn it on.
On Wed, 18 Jun 2003, Tom Lane wrote:
> "scott.marlowe" <[EMAIL PROTECTED]> writes:
> > On Wed, 18 Jun 2003, Tom Lane wrote:
> >> I do now seem to recall an agreement that a GUC switch to disable
> >> date-interpretation guessing would be okay, though.
>
> > I'm pretty sure it was the other way ar
Antonios Christofides <[EMAIL PROTECTED]> writes:
> Is this filenames-instead-of-BLOBs for easier backup common practice?
> Any other ideas or comments?
This is a major point of contention. Some people think keeping all data in the
database is a better approach, others think data that isn't inhe
"Peter Haworth" <[EMAIL PROTECTED]> writes:
> On Wed, 18 Jun 2003 15:09:44 -0600 (MDT), scott.marlowe wrote:
>> Why not accept a date of 04/44/2003 and just wrap it into May?
> Unbelievably, to_date() already does that in 7.3.3:
It's only "unbelievable" if Oracle doesn't do the same thing. to_da
On Thu, 19 Jun 2003, Tom Lane wrote:
> "Nigel J. Andrews" <[EMAIL PROTECTED]> writes:
> > If in a trigger one does a select * from the table the trigger is on
> > is it safe to assume that tg_trigtuple (and so tg_newtuple) will use
> > the same TupleDesc as that returned from the select *, i.e. th
On Thu, 19 Jun 2003, Andrew Snow wrote:
>
> This is an old postgres "gotcha" that I've lived with since a long, long
> time ago. It forced me to always use the datestyle -MM-DD so there
> is never any confusion in my programming.
create table p (d date);
CREATE TABLE
insert into p values
On Thu, 2003-06-19 at 08:39, Jonathan Bartlett wrote:
> The problem is that you can't do the check on the _field_ since it has
> already been converted to a date.
You're right. How about a "before insert" trigger?
> On 19 Jun 2003, Ron Johnson wrote:
>
> > On Thu, 2003-06-19 at 05:35, Bruno Wol
On Thu, 19 Jun 2003, Nigel J. Andrews wrote:
> On Thu, 19 Jun 2003, Peter Eisentraut wrote:
>
> > Travis Hume writes:
> >
> > > So at this point all selects are screwed unless I issue a ROLLBACK;
> > > Why? Is there a way for SELECTs (or all SQL) statement that fail to NOT
> > > have this behav
On 19/06/2003 14:04 Carlos Oliva wrote:
Hi Paul,
This worked very well. Perhaps, you can help me with another question:
If I write an aaa bbb ccc of 200.200.200.0, do all clientes with IP
addresses between 0.0.0.0 and 200.200.200.0 can connect?
No. The client ips would have to be in the range 200.
On Thu, 19 Jun 2003, Bruno Wolff III wrote:
> On Thu, Jun 19, 2003 at 02:43:12 -0500,
> Ron Johnson <[EMAIL PROTECTED]> wrote:
> >
> > OTOH, Andrew Snow's method (alway use ANSI standard -MM-DD)
> > is guaranteed to work. Have your app convert to that format before
> > inserting, and then P
Did you actually read the cited links?
I read the 7.3 docs then fired off an email and forgot about 7.4 docs.
duh! sorry.
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
Frank Miles <[EMAIL PROTECTED]> writes:
> If the application always passes the date to Postgres with the three-letter
> month name where appropriate, and use the 4-digit year, it should be
> comparatively bulletproof.
That pretty much assumes that you've already validated the input and
converted i
On Thu, 19 Jun 2003, Frank Miles wrote:
> On Thu, 19 Jun 2003, Bruno Wolff III wrote:
>
> > On Thu, Jun 19, 2003 at 02:43:12 -0500,
> > Ron Johnson <[EMAIL PROTECTED]> wrote:
> > >
> > > OTOH, Andrew Snow's method (alway use ANSI standard -MM-DD)
> > > is guaranteed to work. Have your app
> > The problem is that you can't do the check on the _field_ since it has
> > already been converted to a date.
>
> You're right. How about a "before insert" trigger?
This suffers from the same problem, I believe. If I weren't a lazy
bastard, I check it :)
Jon
---(end
On your second question:
Keeping old data helps with data analysis, i.e., data mining. I would do the fired date as transactions. To see if an employee is still and employee, look for the latest transation, hired, rehired, contracted with as a temp/consultant, fired, laid off, etc.
Antonios Chris
I think he was thiniing about making the **IN** and out of the date functions to be ANSI only.
Bruno Wolff III wrote:
On Thu, Jun 19, 2003 at 02:43:12 -0500,
Ron Johnson <[EMAIL PROTECTED]> wrote:
OTOH, Andrew Snow's method (alway use ANSI standard -MM-DD)
is guaranteed to work. Have your
WHOA!
Peter Haworth wrote:
On Wed, 18 Jun 2003 15:09:44 -0600 (MDT), scott.marlowe wrote:
Why not accept a date of 04/44/2003 and just wrap it into May?
Unbelievably, to_date() already does that in 7.3.3:
pmh=> select to_date('2003-04-44','-mm-dd');
to_date
2003-05-14
(
On Thu, 2003-06-19 at 09:54, Jonathan Bartlett wrote:
> > > The problem is that you can't do the check on the _field_ since it has
> > > already been converted to a date.
> >
> > You're right. How about a "before insert" trigger?
>
> This suffers from the same problem, I believe. If I weren't a
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> But when you are dealing with hand-entered input, you *do not know* what
> the user meant by input such as '01/03/2003'. You may think you know,
> but you're just fooling yourself.
I agree with this, but the actual test case someone posted was:
Shouldn't dates be validated using the *LOCALE setting and not try to
guess?
Tom Lane wrote:
>
> Frank Miles <[EMAIL PROTECTED]> writes:
> > If the application always passes the date to Postgres with the three-letter
> > month name where appropriate, and use the 4-digit year, it should be
> > com
Thanks again for your help. I think that the picture is getting clearer
in my mind.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Paul Thomas
Sent: Thursday, June 19, 2003 10:37 AM
To: pgsql-general @ postgresql . org
Subject: Re: FW: [GENERAL] Allowing
> Anyone care to run some tests to see how lax Oracle's to_date is?
I just did some checking on 9i to make sure things haven't changed since
Oracle 7, it appears they have not.
Oracle's to_date function is not lax at all. In fact, one of the
things that has always bugged me about Oracle's to_dat
I could not select from pg_catalog. Is it called something else in
Postgres 7.1.3?
Postgres 7.1.3 is included in RED HAT ADVANCED SERVER 2.1 which I am using.
Jim Apsey
[EMAIL PROTECTED] wrote:
This is an a view I have found somewhat useful in helping partially
automate definition of functions d
Jean-Luc Lachance <[EMAIL PROTECTED]> writes:
> Shouldn't dates be validated using the *LOCALE setting and not try to
> guess?
It would make sense to offer a "strict" mode in which the date order
has to be what DateStyle suggests. I'm astonished that no one seems
to get the point that there are a
[EMAIL PROTECTED] writes:
>> Anyone care to run some tests to see how lax Oracle's to_date is?
> Oracle's to_date function is not lax at all.
Then we need to fix ours. Karel?
regards, tom lane
---(end of broadcast)---
TIP
On Mon, 16 Jun 2003, Bruce Momjian wrote:
> Curt Sampson wrote:
> >
> > How do I, in SQL, find out the page size used in the current database
> > cluster?
>
> Sure, but I am not sure how to determine this from SQL:
>
> $ pg_controldata
> ...
> Database block size: 8192
Right. So,
Hi,
I'm trying to write a very simple SQL function, but I'm having problems
with it. My function is as follows:
CREATE OR REPLACE FUNCTION fnCPUpdateVolunteerType(int, text) RETURNS
INTEGER AS '
UPDATE vblVolunteerType
SET vchDisplayName = ''$2''
WHERE intVolunteerTypeID= $1;
Hi all!
I'm having a very strange deadlock problem in PostgreSQL 7.3.2 under RedHat Linux.
All connections to this database come from Tomcat's 4.1.24 connection pooling
mechanism, and for some reason frequently I have lots of INSERTs in "waiting" state.
But what they're writing for?? To me s
Shalom,
I need to know does Postgre has email receiving and sending capability?
(Like xp_mail in M$ SQL Server). If Postgre has this capability, can
someone give me pointer where could I found the documentation/tutorial
about using this mail processing capability? Thank you for your help.
Living
Thanks for nailing this down for me.
If you're interested in how I stumbled across pg_options ..
In going back, I can see that I was looking at 7.0 documentation (
http://www.postgresql.org/docs/view.php?version=7.0&idoc=0&file=postgres
.htm) as downloaded to some user site.
I would observe t
Hi,
this sounds typical for M$ - overload the
backend with rarely used functionality :)
I think this can be much more elegant
be done outside the DB with a host language.
For example with any script language
where is should not cover more then a
few lines.
You can even trigger this from DB if
this
I need to change a column of one table from int2 to int4. What is the best way to do
it and retain the data in the 7.2 version?
Thanks,
Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail!
http://login.mail.lycos.com
I strongly protest the pulling of plpython.
Make it untrusted if necessary, but don't
pull it. The capabilities of plpython are
much too valuable to trash.
At this point I am available as a tester,
but not as a developer for any changes to the
interface. Drag me into the frey. I may also
be fr
As I understand it, plpython currently relies on the rexec call /
interface to function, and that functionality is going away, so SOMEONE
has to step up to the plate and change the way plpython is called in order
to keep it.
I do believe someone on hackers proferred an interest in doing this
elein <[EMAIL PROTECTED]> writes:
> I strongly protest the pulling of plpython.
I seem to have missed the part of this where you volunteer to fix it.
Complaining will not change anything; someone stepping up and doing the
work will change things.
regards, tom lane
--
Hi,
I have a database with a few tables in it (this is just a "learning"
database). Originally I set these tables up with sequences on them
which increment the primary key of the table when a new record is
inserted. I would like to drop these sequences.
1. First I wanted to copy the database
Erik Price <[EMAIL PROTECTED]> writes:
> So, is the solution to drop the column first, then drop the sequence,
> then re-create the column by using ALTER TABLE ?
Perhaps what you are really looking to do is just to reset the sequence
counters with setval() ?
regards, tom
Tom Lane wrote:
Erik Price <[EMAIL PROTECTED]> writes:
So, is the solution to drop the column first, then drop the sequence,
then re-create the column by using ALTER TABLE ?
Perhaps what you are really looking to do is just to reset the sequence
counters with setval() ?
Believe it or not, I r
On Thu, 19 Jun 2003, Erik Price wrote:
> Hi,
>
> I have a database with a few tables in it (this is just a "learning"
> database). Originally I set these tables up with sequences on them
> which increment the primary key of the table when a new record is
> inserted. I would like to drop thes
I just didn't want anyone to say "no one complained".
That has been a reason some features were changed
without a lot of notice in the past. It is important to
understand who uses what.
And I have offered to do as much as I can at this
time. It is not as if I were whining about something
and b
Thanks, Scott.
I believe that was Kevin Jacobs in a message to pgsql-general
and I've sent him a message about being available for testing.
-elein
On Thursday 19 June 2003 14:27, scott.marlowe wrote:
> As I understand it, plpython currently relies on the rexec call /
> interface to function, a
elein wrote:
> Thanks, Scott.
>
> I believe that was Kevin Jacobs in a message to pgsql-general
> and I've sent him a message about being available for testing.
It sure is a shame the Python folks just decided to punt instead of
rolling up their sleeves and fixing the problem. Now a cascade of
d
On Tue, Jun 17, 2003 at 06:10:06PM -0300, Felipe Schnack wrote:
> Hi all!
> I'm having a very strange deadlock problem in PostgreSQL 7.3.2 under
> RedHat Linux. All connections to this database come from Tomcat's 4.1.24
> connection pooling mechanism, and for some reason frequently I have
Regarding backup history:
I have an application designed for novices. Apparently it's easy to hit the
"Delete" button, and then say yes to the "Are you sure you want to delete
this?" question even when they don't want to. Therefore I simply mark a
record as deleted. For example,
UPDATE table SE
adding comments to a tableHi All,
here are my three tables. I wold like to list them in the following way: I
need all the columns from t_stockchanges, and one field for the pgoductgroup
the t_stockchanges.productid belongs to.
But one product can belong to many group, so i would need one row for e
First off, I've nerver heard of using Floating point for money. There
**IS** rounding and truncation that goes on, in even double precision.
The bean counters get upset if you lose .001 dollars, especially if
it occurs for several million transactions. The usual choice is DECIMAL.
And I'd u
Test
"Jeff Eckermann" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> --- Ivar <[EMAIL PROTECTED]> wrote:
> > Hi,
> >
> > Is there any way to make post to newsgroups faster ?
> > At moment moderators must check it, while they chek
> > it message is outdated
> > already.
> >
> > Are
To all postgresql users,
I've just installed the new version of postgresql 7.3.3 (using the
source and compile it myself). I find out there is a problem with the
big5 characters. I have a perl script and it is used to extract data
from some xbase file to the postgresql. It works under the versi
Yee works ok
"Ivar" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
>
> Test
>
> "Jeff Eckermann" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]
> > --- Ivar <[EMAIL PROTECTED]> wrote:
> > > Hi,
> > >
> > > Is there any way to make post to newsgroups faster ?
> > > At m
On Thu, 19 Jun 2003, Dennis Gearon wrote:
> First off, I've nerver heard of using Floating point for money. There
> **IS** rounding and truncation that goes on, in even double precision.
> The bean counters get upset if you lose .001 dollars, especially if
> it occurs for several million tr
We came across an error while using the hash_search function to find a
database entry in the pg_statdbhash hash table. Just wondering if there
are any recorded errors using this function when doing the initdb. It
seems to work fine while using sql statements but internally something is
going wrong.
What i am trying to do is to maintain a linked list of all the relations
in a database. When i create a db then i want it to insert into the linked
list the relation ids. As it stands now, i can create a db fine and i see
the relation id's in the linked list. BUT, as soon as i psql into the db
then
On Thu, Jun 19, 2003 at 17:07:43 -0400,
Nailah Ogeer <[EMAIL PROTECTED]> wrote:
Please don't respond to other messages to start a new thread.
> What i am trying to do is to maintain a linked list of all the relations
> in a database. When i create a db then i want it to insert into the linked
>
56 matches
Mail list logo
