Tom Lane, 07.11.2007 06:14:
Thomas Kellerer <[EMAIL PROTECTED]> writes:
If everyone simply top-posted, there would be no need for me to scroll down,
just to find a two line answer below a forty line quote - which I personally
find more irritating than top-posting.
I think you're ignoring my b
Il Wednesday 07 November 2007 01:29:44 Alvaro Herrera ha scritto:
> Reg Me Please wrote:
> > Il Tuesday 06 November 2007 22:13:15 hai scritto:
> > That's the "branch and bound". Editing 29M+ lines file takes some time.
> > But this is the way I'm going to go right now.
>
> Huh, why not just use pgl
Alvaro Herrera wrote:
Please try "thread apply all bt full" on gdb.
The first lines where the symbols are loaded are of course identical.
The output of the command is in my opinion not very helpful:
(gdb) thread apply all bt full
Thread 1 (Thread 47248855881456 (LWP 7129)):
#0 0x2af
Il Wednesday 07 November 2007 06:35:55 Tom Lane ha scritto:
> nick <[EMAIL PROTECTED]> writes:
> > Just wondering, if it is possible to do the following
> > create function foo() RETURNS Record(int, varchar, int)
>
> OUT parameters (in reasonably recent PG releases).
>
> regression=# create functio
Reece Hart <[EMAIL PROTECTED]> writes:
> However, it's not clear that you've considered a clause like 'ORDER BY
> (foo IS NULL), foo', which I believe is not implementation dependent.
Yeah, that should work reasonably portably ... where "portable" means
"equally lousy performance in every implemen
nick <[EMAIL PROTECTED]> writes:
> Just wondering, if it is possible to do the following
> create function foo() RETURNS Record(int, varchar, int)
OUT parameters (in reasonably recent PG releases).
regression=# create function foo (f1 out int, f2 out varchar, f3 out int) as
$$ select 42, 'foo'::v
On Tue, 2007-11-06 at 11:38 +0400, rihad wrote:
> Is there any way to "hardcode" the NULL handling in an index (as per
> http://www.postgresql.org/docs/8.3/static/indexes-ordering.html) so
> that
> SELECT * FROM t ORDER BY foo automatically implies NULLS FIRST (and,
> similarly so that SELECT
Thomas Kellerer <[EMAIL PROTECTED]> writes:
> If everyone simply top-posted, there would be no need for me to scroll down,
> just to find a two line answer below a forty line quote - which I personally
> find more irritating than top-posting.
I think you're ignoring my basic point, which was tha
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I would be interested as well. Our changes are perhaps once or twice a
week in development but its nice to make sure that its being tracked.
Mario
Tim Goodaire wrote:
> Robert,
>
> I'd like to have a copy of your script if you're willing to shar
Hello,
Could someone tell me if the ‘--pwfile’ command can be applied to createdb.exe
and psql.exe?
I found a description of ‘--pwfile’ at
http://www.it-manuals.info/postgressql/app-initdb.html, which states:
--pwfile=filename
makes initdb read the database superuser's password from a file. th
Just wondering, if it is possible to do the following
create function foo() RETURNS Record(int, varchar, int)
then inside of the function return a set int, varchar, int.
Then be able to call the function
select * From foo()
instead of having to write
select * from foo() as (int, varchar, int
On Nov 7, 2007 2:40 PM, Rainer Bauer <[EMAIL PROTECTED]> wrote:
> That's nice to hear. But I respect licences as they are and the ODBCng driver
> is licenced under the GPL.
That doesn't mean that you're not allowed to use it with commercial
applications; it just means that you need to be happy to
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Wed, 07 Nov 2007 02:40:08 +0100
Rainer Bauer <[EMAIL PROTECTED]> wrote:
>
> That's nice to hear. But I respect licences as they are and the
> ODBCng driver is licenced under the GPL.
>
> So is this an _official_ statement from CommandPrompt that
"Joshua D. Drake" wrote:
>> psqlodbc is licenced under LGPL, but ODBCng uses the GPL.
>>
>> That means that commercial software cannot use the ODBCng driver. Are
>> there any plans to release it under LGPL licence?
>
>The only time this would be an issue is if you tried to embed the
>driver. Othe
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Tue, 06 Nov 2007 22:03:52 +0100
Rainer Bauer <[EMAIL PROTECTED]> wrote:
> Alvaro Herrera wrote:
>
> >FYI there's another Postgres ODBC driver that is said to have better
> >performance.
> >
> >https://projects.commandprompt.com/public/odbcng
> >
>
"SHARMILA JOTHIRAJAH" <[EMAIL PROTECTED]> writes:
> I understand that. But why is that when oracle is given a hint to do full
> table scan instead of using index to get the count, it is still faster than
> postgres when both has the same explain plan? Oracle takes 34 sec and
> postgres takes 1 m10
Reg Me Please wrote:
> Il Tuesday 06 November 2007 22:13:15 hai scritto:
> > Split it in half, with the appropriate sql on each end so the data
> > still works, and see which half causes a problem. Keep splitting the
> > one that causes a problem in half until you have a small one with the
> > pr
Christian Schröder wrote:
> Since "ps -Lw" only showed me my own processes, I did "ps -AT -o
> pid,ppid,lwp,nlwp,command" instead. This is the relevant line:
>
> PID PPID LWP NLWP COMMAND
> ...
> 7129 3934 71291 postgres: dpyrek uspi 88.70.241.205(10612) SELECT
> ...
>
> I'm not sure if
On Nov 6, 2007, at 3:43 PM, Tim Goodaire wrote:
Robert,
I'd like to have a copy of your script if you're willing to share.
We're
currently working on schema control stuff, and your script may be
useful.
On Wed, Oct 24, 2007 at 11:23:39PM -0400, Robert Treat wrote:
On Wednesday 24 October
Hi,
Le Tuesday 06 November 2007 22:40:50 Tom Lane, vous avez écrit :
> Reg Me Please <[EMAIL PROTECTED]> writes:
> > I'm trying to understand what and where.
> > The point is that I have this 29M+ lines script telling me there's a
> > problem somewhere.
You could use pgloader, which reads CSV inp
Tom Lane wrote on 06.11.2007 21:21:
The real point here is that when writing to a mailing list, you should
make an effort to conserve other peoples' time. Hundreds of people are
going to read what you wrote, possibly thousands when you count in
people searching the list archives in the future.
Il Tuesday 06 November 2007 22:37:12 Scott Marlowe ha scritto:
> On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > Il Tuesday 06 November 2007 22:13:15 hai scritto:
> > > On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > > > Il Tuesday 06 November 2007 19:43:38 Scott Marlowe ha scritt
Robert,
I'd like to have a copy of your script if you're willing to share. We're
currently working on schema control stuff, and your script may be useful.
On Wed, Oct 24, 2007 at 11:23:39PM -0400, Robert Treat wrote:
> On Wednesday 24 October 2007 15:11, Roberts, Jon wrote:
> > Yeah. I think ha
Reg Me Please <[EMAIL PROTECTED]> writes:
> I'm trying to understand what and where.
> The point is that I have this 29M+ lines script telling me there's a problem
> somewhere.
It told you exactly where it detected the problem.
Actually ... are you looking at the right error message? If this is
On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> Il Tuesday 06 November 2007 22:13:15 hai scritto:
> > On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > > Il Tuesday 06 November 2007 19:43:38 Scott Marlowe ha scritto:
> > > > On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > > >
Il Tuesday 06 November 2007 22:13:15 hai scritto:
> On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > Il Tuesday 06 November 2007 19:43:38 Scott Marlowe ha scritto:
> > > On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > > > That seems not to be the case.
> > > > The last line has a \
On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> Il Tuesday 06 November 2007 19:43:38 Scott Marlowe ha scritto:
> > On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > > That seems not to be the case.
> > > The last line has a \. by its own and the last but one is
> > > well formed.
> >
Alvaro Herrera wrote:
>FYI there's another Postgres ODBC driver that is said to have better
>performance.
>
>https://projects.commandprompt.com/public/odbcng
>
>(Yes, my company maintains it)
psqlodbc is licenced under LGPL, but ODBCng uses the GPL.
That means that commercial software cannot use
Il Tuesday 06 November 2007 19:43:38 Scott Marlowe ha scritto:
> On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> > That seems not to be the case.
> > The last line has a \. by its own and the last but one is
> > well formed.
>
> (Please don't top post...)
>
> Got a self contained test case y
Marc <[EMAIL PROTECTED]> writes:
> My original reason for reaching out to the list was over confusion as to
> when an EXCLUSIVE lock would be taken table level since the documentation
> says this should never happen except to some system catalogs. Is there
> something missing from the documentatio
Ok. I'll keep looking at pg_locks.
My original reason for reaching out to the list was over confusion as to
when an EXCLUSIVE lock would be taken table level since the documentation
says this should never happen except to some system catalogs. Is there
something missing from the documentation?
? wrote:
> Hi all,
>
> Recently I found my C/S program becomes slow. So I analysed the tcp
> traffic between my program and the Postgresql database. I found there
> are occasionally tcp re-transmission and that's why my porgram slow
> down.
>
> But when I changed to npsql, the tcp traf
Marc <[EMAIL PROTECTED]> writes:
> Version of postgres is 8.2.4.
> ...
> second report that I always see that UPDATE statement and in the list of
> locks I see ExclusiveLock granted on the users table for one of the running
> pids but not the others.
Well, if it's not a foreign key issue then I th
Version of postgres is 8.2.4.
Maybe it will help to give more b/g on how I'm identifying the problem?
The way this materializes as a real issue surrounds transactions left idle.
There is a bug in our app that we haven't tracked down yet where on occasion
we end up with connections marked " in tra
"Scott Marlowe" <[EMAIL PROTECTED]> writes:
> But the other reason is that top posters are often the biggest
> offenders of not editing their replies. there's nothing like seeing a
> one line reply to a 500 line conversation.
> When inline posting, one can delete the parts of the converstation
> t
Marc <[EMAIL PROTECTED]> writes:
> This is the query that I'm running to view locks:
> SELECT pg_class.relname AS table,
>pg_database.datname AS database,
>transaction, pid, mode, granted
> FROM pg_locks, pg_class, pg_database
> WHERE pg_locks.relation = pg_class.oid
> AND pg_loc
On Nov 6, 2007 1:16 PM, Collin Kidder <[EMAIL PROTECTED]> wrote:
> This is offtopic but there is nothing wrong with top posting. Is there a
> mail list policy on it or are you just picky about it?
OK, I was being a bit flippant with my last response. The reasoning
on technical mailing lists for p
This is the query that I'm running to view locks:
SELECT pg_class.relname AS table,
pg_database.datname AS database,
transaction, pid, mode, granted
FROM pg_locks, pg_class, pg_database
WHERE pg_locks.relation = pg_class.oid
AND pg_locks.database = pg_database.oid
ORDER BY pg_clas
Marc wrote:
Hi,
I'm seeing an "EXCLUSIVE" lock being taken on a table even though the
documentation says that "This lock mode is not automatically acquired on
user tables by any PostgreSQL command."
Hmm - are you sure?
My SQL is
UPDATE users SET online = $1 where username = $2
username is t
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
On Nov 6, 2007 1:16 PM, Collin Kidder <[EMAIL PROTECTED]> wrote:
> This is offtopic but there is nothing wrong with top postin
This is offtopic but there is nothing wrong with top posting. Is there a
mail list policy on it or are you just picky about it?
Scott Marlowe wrote:
On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
That seems not to be the case.
The last line has a \. by its own and the last but one is
On 11/6/07, Reg Me Please <[EMAIL PROTECTED]> wrote:
> That seems not to be the case.
> The last line has a \. by its own and the last but one is
> well formed.
(Please don't top post...)
Got a self contained test case you can post?
---(end of broadcast)--
Hi,
I'm seeing an "EXCLUSIVE" lock being taken on a table even though the
documentation says that "This lock mode is not automatically acquired on
user tables by any PostgreSQL command."
My SQL is
UPDATE users SET online = $1 where username = $2
username is the PK on the users table.
Other lock
That seems not to be the case.
The last line has a \. by its own and the last but one is
well formed.
Il Tuesday 06 November 2007 19:14:00 Tom Lane ha scritto:
> Reg Me Please <[EMAIL PROTECTED]> writes:
> > At a certain point I get an error telling about a
> > "invalid input syntax for type numer
Reg Me Please <[EMAIL PROTECTED]> writes:
> At a certain point I get an error telling about a
> "invalid input syntax for type numeric"
> The incriminated line number is the last one (the one containing the \.).
> Is there a way to know which line is really malformed?
Why do you think the report
Hi all.
I'm generating an SQL script to load some million rows into a table.
I'm trying to use the COPY command in order to speed the load up.
At a certain point I get an error telling about a
"invalid input syntax for type numeric"
The incriminated line number is the last one (the one containin
Tom Lane wrote:
* The only place internal_flush would call errmsg is here:
ereport(COMMERROR,
(errcode_for_socket_access(),
errmsg("could not send data to client: %m")));
So why is it unable to send data to the client?
The user
In response to André Volpato <[EMAIL PROTECTED]>:
> Richard Huxton escreveu:
> > Reg Me Please wrote:
> >> While I would not spend resources in fine tuning the count(*), I would
> >> spend some to underastand why and how the other ones do it better.
> >>
> >> Just to be better.
> >
> > The problem
Richard Huxton escreveu:
Reg Me Please wrote:
While I would not spend resources in fine tuning the count(*), I would
spend some to underastand why and how the other ones do it better.
Just to be better.
The problem is well understood, and there is extensive discussion in
the mailing lists ar
[EMAIL PROTECTED] wrote:
I have a table with a lot of fields, some of which are type TEXT. I
seem to have a lot of occasions where I'd like to see almost all of
the fields in a query but a SELECT * will pull in the text fields,
which then wrap across the screen, even using \x. Is there a variatio
phazon <[EMAIL PROTECTED]> writes:
> I've "**" before the name in order to display them at first, but the 2
> chars "**" are ignored in the order by.
What's your lc_collate setting? In locales other than "C" this is not
at all unexpected.
regards, tom lane
--
Hi,
I've to order by q query by a column.
SELECT a.id, a.nom
FROM table a
ORDER BY a.nom;
And i've something strange. It's order like that:
5 |
8 | aaab
15 | ** aaac
6 | aaad
I've "**" before the name in order to display them at first, but the 2
chars "**" are ignored in the order b
"Eric Johnson" <[EMAIL PROTECTED]> writes:
> I can find all the check constraints, primary keys and foreign keys in
> the information schema, but I cannot find the non-unique indexes. Is
> there anyplace to get that information (name, columns, etc)?
Not from the information_schema, because "index"
On 11/6/07, John Smith <[EMAIL PROTECTED]> wrote:
> http://www.nabble.com/need-good-editor-for-postgreSQL-tf3660882.html#a10229560
>
> how do u get gvim (portable?) to do syntax color highlighting for
> sql/plpgsql on files in buffer?? you're right- notepad is not enough
> (and pgedit is not free).
Hi,
Oracle, eventhough is a timestamp based database, stores only one
version for each row in the table and the rest of the versions of the
same-row(which might have got created due to updates) in a separate
place called undo log. In postgres, all the versions are stored in the
table. So the ta
Il Tuesday 06 November 2007 16:40:27 John Smith ha scritto:
> http://www.nabble.com/need-good-editor-for-postgreSQL-tf3660882.html#a10229
>560
>
> how do u get gvim (portable?) to do syntax color highlighting for
> sql/plpgsql on files in buffer?? you're right- notepad is not enough
> (and pgedit i
On 11/6/07, Christian Schröder <[EMAIL PROTECTED]> wrote:
> Hi list,
> once again I do not understand how the query planner works and why it
> apparently does not find the best result.
> I have a table with about 125 million rows. There is a char(5) column
> with a (non-unique) index. When I try to
I've never used Npsql and I'm not a C# developer (so take this
response with more than a few grains of salt) but from what I
remember reading somewhere, Npsql implements postgres protocols
natively, with the protocol handling done directly in C# and getting
the raw data without going throug
=?ISO-8859-1?Q?Christian_Schr=F6der?= <[EMAIL PROTECTED]> writes:
> db2:/home/pgsql/data # gdb --pid=7129
> GNU gdb 6.5
> Copyright (C) 2006 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you are
> welcome to change it and/or distribute copies
It may depend on the index itself against the locales.
1. define the insex with the text_pattern_ops operato class
2. run vacuum analyze on the table
3. re-run the explain
See chapter 11 (especiallu 11.8) for v8.2
Il Tuesday 06 November 2007 16:25:09 Christian Schröder ha scritto:
> Hi list,
>
http://www.nabble.com/need-good-editor-for-postgreSQL-tf3660882.html#a10229560
how do u get gvim (portable?) to do syntax color highlighting for
sql/plpgsql on files in buffer?? you're right- notepad is not enough
(and pgedit is not free).
cheers, jzs
---(end of broadcast
Hi list,
once again I do not understand how the query planner works and why it
apparently does not find the best result.
I have a table with about 125 million rows. There is a char(5) column
with a (non-unique) index. When I try to find the distinct values in
this column using the following sql
Reg Me Please wrote:
While I would not spend resources in fine tuning the count(*), I would
spend some to underastand why and how the other ones do it better.
Just to be better.
The problem is well understood, and there is extensive discussion in the
mailing lists archives. The basic problem
I can find all the check constraints, primary keys and foreign keys in
the information schema, but I cannot find the non-unique indexes. Is
there anyplace to get that information (name, columns, etc)?
Sorry if this is an FAQ. I couldn't find it with Google.
--
Eric Johnson
"Where your pleasure
On 11/3/07, Abandoned <[EMAIL PROTECTED]> wrote:
>
> I tryed pg_dump but it is very slowly. Are there any faster way to
> copy database?
Assuming it's all happening on the same db server, yes:
psql template1
create database newdb template olddb
---(end of broadcast)--
Tom Lane wrote:
What we can be reasonably certain of is that that backend wasn't
reaching any CHECK_FOR_INTERRUPTS() macros. Whether it was hung up
waiting for something, or caught in a tight loop somewhere, is
impossible to say without more data than we have. AFAIR the OP didn't
even mention w
On Sat, Nov 03, 2007 at 01:42:09PM -, Abandoned wrote:
> I tryed pg_dump but it is very slowly. Are there any faster way to
> copy database?
Have you tried CREATE DATABASE .. TEMPLATE ? (See amual for syntax)
Have a nice day,
--
Martijn van Oosterhout <[EMAIL PROTECTED]> http://svana.org
On 11/2/07, Lew <[EMAIL PROTECTED]> wrote:
> Scott Marlowe wrote:
> > PostgreSQL is ... still not in the
> > same realm for really really big transactional sites, but man is it
> > geting close fast.
>
> I beg to differ. All anecdotal evidence, and also Sun benchmarks with
> Postgres, show that it
On 11/5/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> I have a table with a lot of fields, some of which are type TEXT. I
> seem to have a lot of occasions where I'd like to see almost all of
> the fields in a query but a SELECT * will pull in the text fields,
> which then wrap across the scre
rihad <[EMAIL PROTECTED]> writes:
> Is there any way to "hardcode" the NULL handling in an index (as per
> http://www.postgresql.org/docs/8.3/static/indexes-ordering.html) so that
> SELECT * FROM t ORDER BY foo automatically implies NULLS FIRST (and,
> similarly so that SELECT * FROM t ORDER BY
I understand that. But why is that when oracle is given a hint to do full table
scan instead of using index to get the count, it is still faster than postgres
when both has the same explain plan? Oracle takes 34 sec and postgres takes 1
m10 sec . Is there anything that can be done in postgresql
While I would not spend resources in fine tuning the count(*), I would
spend some to underastand why and how the other ones do it better.
Just to be better.
Il Tuesday 06 November 2007 15:29:34 Bill Moran ha scritto:
> In response to Reg Me Please <[EMAIL PROTECTED]>:
> > I have no doubt you're r
In response to Reg Me Please <[EMAIL PROTECTED]>:
> I have no doubt you're right, Pavel.
> But why not?
> It could be a simple enhacement.
It's not simple. Do some searches on the mailing lists and you will
find discussion of why it's difficult to do.
>
> Il Tuesday 06 November 2007 15:11:02 P
I'm interested in some guidance on how to best deal with the following
issue.
I'm building postgres-8.2.5 on Solaris 8 SPARC, using a gcc built (not
by me) for our environment. We have an old home-grown software
distribution/configuration management system that arranges
shared-objects and bi
> There were other JDBC clients but they never
> lock any tables. They all used transactions with the serializable isolation
> level.
If you have concurrent access and some isolation you have locking.
If two inserts/updates try to change the same data (even via
triggers/rules) you get a failed c
Hi,
I have a problem starting the server on Mac OSX 10.3.9. I compiled
v8.2.5 from source. The compilation, regression testing, and
installation all went smoothly with no errors. I installed PostgreSQL
in the default location and created a new user account named postgres.
I ran initdb which e
Hi all,
Recently I found my C/S program becomes slow. So I analysed the tcp
traffic between my program and the Postgresql database. I found there
are occasionally tcp re-transmission and that's why my porgram slow
down.
But when I changed to npsql, the tcp traffic was cut off to only 1/4
and t
I have a table with a lot of fields, some of which are type TEXT. I
seem to have a lot of occasions where I'd like to see almost all of
the fields in a query but a SELECT * will pull in the text fields,
which then wrap across the screen, even using \x. Is there a variation
of SELECT that is somethi
Kevin Hunter wrote:
"what do I need to able to do to be able to honestly say that 'I am a
Postgres DBA' on my resume"
Greg Smith wrote:
Organizing things as above, this turns into a somewhat fuzzy question
about how much of the grid one has to cover before achieving that goal.
Consider this;
On Nov 2, 5:30 pm, Sascha Bohnenkamp <[EMAIL PROTECTED]> wrote:
> if you dump/undump using a pipe there is no temp file ...
>
> Abandoned schrieb:
>
>
>
> > On Nov 2, 3:49 pm, Sascha Bohnenkamp <[EMAIL PROTECTED]> wrote:
> >> maybe pg_dump | pg_undump could work?
>
> >> I mean pg_dump with the appr
On Nov 2, 3:49 pm, Sascha Bohnenkamp <[EMAIL PROTECTED]> wrote:
> maybe pg_dump | pg_undump could work?
>
> I mean pg_dump with the appropriate parameters and undump directly to
> the other database?
>
> This may one of the fastest ways to do it I think.
>
> Abandoned schrieb:
>
> > Hi..
> > I want
Hi..
I want to copy my database..
I have a database which is name db01 and i want to copy it as name
db01copy ?
How can i do this ?
This database is very big 200GB so i want to the fastest way. Also if
you suggest pg_dump how can i remove the dump data after copying ?
Note: I have a root account wi
maybe pg_dump | pg_undump could work?
I mean pg_dump with the appropriate parameters and undump directly to
the other database?
This may one of the fastest ways to do it I think.
Abandoned schrieb:
> Hi..
> I want to copy my database..
> I have a database which is name db01 and i want to copy i
if you dump/undump using a pipe there is no temp file ...
Abandoned schrieb:
> On Nov 2, 3:49 pm, Sascha Bohnenkamp <[EMAIL PROTECTED]> wrote:
>> maybe pg_dump | pg_undump could work?
>>
>> I mean pg_dump with the appropriate parameters and undump directly to
>> the other database?
>>
>> This may
I am facing a difficult problem with setting up a rule.
The rule is essentially designed to allow updates to a view made up of
data from two tables.
Here is a simplified example to reproduce the problem:
--
-- clear the table and view
--
DROP VIEW IF EXISTS full_view;
DROP TABLE IF EXISTS chil
Kynn Jones wrote:
I have large database that needs to be built from
scratch roughly once every month. I use a Perl script to do this.
The tables are very large, so I avoid as much as possible using
in-memory data structures, and instead I rely heavily on temporary
flat files.
I have solved th
Scott Marlowe wrote:
PostgreSQL is ... still not in the
same realm for really really big transactional sites, but man is it
geting close fast.
I beg to differ. All anecdotal evidence, and also Sun benchmarks with
Postgres, show that it is not only "in the same realm" but can actually come
In response to SHARMILA JOTHIRAJAH <[EMAIL PROTECTED]>:
> Hi
> We are in the process of testing for migration of our database from Oracle to
> Postgresql.
> I hava a simple query
>
> Select count(*) from foo
This is asked a lot. The quick answer is that PostgreSQL method of MVCC
makes it impos
Thanks Tom! I already solved the problem by putting "127.0.0.1
localhost" in /etc/hosts file..
On Nov 2, 12:04 am, [EMAIL PROTECTED] (Tom Lane) wrote:
> novice <[EMAIL PROTECTED]> writes:
> > I just installed postgres-8.2 using the command line
> > apt-get install postgres-8.2 and it seems the i
Hi,
As i am Working for Decimal Formats. is there any method to do it in
the same way.
On 11/2/07, Moiz Kothari <[EMAIL PROTECTED]> wrote:
>
> Hi,
>
> You can use what bruce has suggested and get the value you want.
>
> eg.
>
> select trim(substring(cash_words('1520'), 1, position('dolla
I have no doubt you're right, Pavel.
But why not?
It could be a simple enhacement.
Il Tuesday 06 November 2007 15:11:02 Pavel Stehule ha scritto:
> Hello
>
> PostgreSQL doesn't use index for COUN(*)
>
> http://www.varlena.com/GeneralBits/18.php
> http://sql-info.de/en/postgresql/postgres-gotchas.h
On 06/11/2007, Toni Casueps <[EMAIL PROTECTED]> wrote:
>
> I'm using the PostgreSQL crosstab functions to do something like this:
>
> From this table ...
>
> ax 10
> bx 13
> ay 11
> by 17
>
>
> ... I get:
>
> xy
> a10 11
> b13
Hello
PostgreSQL doesn't use index for COUN(*)
http://www.varlena.com/GeneralBits/18.php
http://sql-info.de/en/postgresql/postgres-gotchas.html#1_7
Regards
Pavel Stehule
On 06/11/2007, SHARMILA JOTHIRAJAH <[EMAIL PROTECTED]> wrote:
>
> Hi
> We are in the process of testing for migration of our
Hi
We are in the process of testing for migration of our database from Oracle to
Postgresql.
I hava a simple query
Select count(*) from foo
This table has 29384048 rows and is indexed on foo_id
The tables are vacuumed and the explain plan for postgresql is
QUERY PLAN
Albe Laurenz wrote:
Richard Huxton wrote:
What'd be the syntax to create a primary key on an already
build table?
ALTER TABLE my_table ADD PRIMARY CONSTRAINT (existing_column);
I'm sure it's just a typo, but that is wrong.
It is not ADD PRIMARY CONSTRAINT, but ADD PRIMARY KEY.
It is indeed
Richard Huxton wrote:
>> What'd be the syntax to create a primary key on an already
>> build table?
>
> ALTER TABLE my_table ADD PRIMARY CONSTRAINT (existing_column);
I'm sure it's just a typo, but that is wrong.
It is not ADD PRIMARY CONSTRAINT, but ADD PRIMARY KEY.
Yours,
Laurenz Albe
--
On 11/5/07, andy <[EMAIL PROTECTED]> wrote:
> Hey, I was just thinking about this... instead of disabling the FK's,
> what about adding a temp table where you could COPY into, then fire off
> a bunch of update's to setup the id fields, etc, etc, then do an Insert
> into realtable select * from temp
I was having the same problem.
I defined a "crosstab definition table" where I declare the things I want
out of crosstab and then I dynamically build views that will add the
needed "AS ct42( )".
Maybe someone else can suggest a better solution ...
Il Tuesday 06 November 2007 13:46:06 Toni C
I'm using the PostgreSQL crosstab functions to do something like this:
From this table ...
ax 10
bx 13
ay 11
by 17
... I get:
xy
a10 11
b13 17
To get the second table I have to specify the field list when I call t
Man ... well :-), thanks for the answer. At least I'll stop running against
the wall.
-Original Message-
From: Alvaro Herrera [mailto:[EMAIL PROTECTED]
Sent: 06 ??? 2007 00:11
To: Hristo Filipov
Cc: 'Shane Ambler'; pgsql-general@postgresql.org
Subject: Re: [GENERAL] How to transfer fr
1 - 100 of 110 matches
Mail list logo
