how to create a database

On Tue, 4 Dec 2001, [iso-8859-1] ranjitha ullasa wrote:

hi ,
i am encoutering the following error while trying to create a database ,
kindly provide the solution to this as soon as possible,
regards
ranjitha

bash-2.05$ createdb sat_demo1
/Program Files/Mercury Interactive/LoadRunner/bin/sed.exe: Unexpected End-of-file
psql: connectDBStart() -- connect() failed: No such file or directory
Is the postmaster running locally
and accepting connections on Unix socket '/tmp/.s.PGSQL.5432'?
createdb: database creation failed

For the first message, Are you running under cygwin or something and is
the sed above actually the correct sed? It may just be having problems
with spaces or something too, but it doesn't hurt to check.

The second does seem to imply that the server isn't actually running, have
you started the postmaster?

Le Tuesday 04 December 2001 01:09, ranjitha ullasa a �crit :

i am encoutering the following error while trying to create a database ,
kindly provide the solution to this as soon as possible,

bash-2.05$ createdb sat_demo1
/Program Files/Mercury Interactive/LoadRunner/bin/sed.exe: Unexpected

hum, so you are not using a _normal_ linux ?
This is a pseudo_linux running under Windows ?
Not sure Everything can work on this kind of _pseudo_linux_

End-of-file psql: connectDBStart() -- connect() failed: No such file or
directory Is the postmaster running locally
and accepting connections on Unix socket '/tmp/.s.PGSQL.5432'?
createdb: database creation failed

This means that the postmaster is not running.
You should launch the postmaster before
( with '/etc/init.d/posgresql start' for example ) see below :

----------------

[root@waisse william]# createdb a
psql: connectDBStart() -- connect() failed: No such file or directory
Is the postmaster running at 'localhost'
and accepting connections on Unix socket '5432'?
createdb: database creation failed
[root@waisse william]# /etc/init.d/po
portmap postfix postgresql
[root@waisse william]# /etc/init.d/pos
postfix postgresql
[root@waisse william]# /etc/init.d/postgresql start
Checking postgresql installation: looks good!
Starting postgresql service: postmaster [14253]
[root@waisse william]# createdb a
CREATE DATABASE
[root@waisse william]#

-----------

This works all the time on a _normal_ linux.

--
cordialement, William WAISSE fpr = 9CCD 7DA2 7050 8805 F471 03D1 DF76 B78C
690B 4E07
--
Computers are like air conditionners. They work better when you close windows.
--
Visitez donc mon site perso (version 0.4 du 29/05/2001) :
http://www.neofutur.net
--
Vous habitez dans l'Essonne et vous aimez les Pingouins ???
visitez donc la GAULE(=LUG91) : http://www.gaule.org
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCS/E/MU d- s: a- C++++ UL+++ P+ L++++ E--- W++ N o-- K- w---
O- M- V- PS+ PE-- Y++ PGP++ t+ 5 X++ R++ tv- b+++ DI- D G e++ h* r++ y+++
------END GEEK CODE BLOCK------

"=?iso-8859-1?Q?Paul_Wehr?=" <postgresql@industrialsoftworks.com> writes:

I need to find the minimum of dates in a number of tables, but "min(date)"
is, of course, an aggregate function. For example:

select key, min(a.date, b.date, c.date) as first_date
from table_a a, table_b b, table_c c
where a.key=b.key and a.key=c.key

Does that really express the computation you want, ie produce a result
only for key values that occur in all three tables?

I was going to suggest

select key, min(date) as first_date from
(select key, date from table_a
union all
select key, date from table_b
union all
select key, date from table_c) subsel
group by key;

but it occurs to me that this produces different results, ie, it will
include key values that only occur in one or two of the tables ...

regards, tom lane

"=?iso-8859-1?Q?Paul_Wehr?=" <postgresql@industrialsoftworks.com> writes:

Since you didn't volunteer "just use not_aggregate_max()", I'm assuming
there's no built-in thing, so the follow-up question(s) would be:
1) can a function be defined with a variable number of arguments

No, although sometimes you can fake it by defining a family of functions
with the same name and different numbers of arguments.

2) can a function be created that calls itself (i.e. recursive)

Sure, although I don't see how that helps in this case.

3) can you do it for me? :)

Nope, got too many things to do already...

regards, tom lane

On Mon, Dec 10, 2001 at 09:07:28PM -0800, Chris Albertson wrote:

All you need is a "max" function with two arguments. To find the
max of four numbers you can do this

max(max(max(a,b),c),d)

It is not even all that ugly.

This function even exists, though it's called int4larger. I know the
function exists but I can never remember the name and have to scan through
the function list each time to find it.

There's also cashlarger, date_larger, float4larger, float8larger,
int2larger, int8larger, interval_larger, numeric_larger, text_larger,
time_larger, timestamp_larger and timetz_larger. (Consistant naming huh?).
The opposites are *smaller.

HTH,
--
Martijn van Oosterhout <kleptog@svana.org>
http://svana.org/kleptog/

All you need is a "max" function with two arguments. To find the
max of four numbers you can do this

max(max(max(a,b),c),d)

It is not even all that ugly.

Tom Lane wrote:

"=?iso-8859-1?Q?Paul_Wehr?=" <postgresql@industrialsoftworks.com> writes:

Since you didn't volunteer "just use not_aggregate_max()", I'm assuming
there's no built-in thing, so the follow-up question(s) would be:
1) can a function be defined with a variable number of arguments

No, although sometimes you can fake it by defining a family of functions
with the same name and different numbers of arguments.

2) can a function be created that calls itself (i.e. recursive)

Sure, although I don't see how that helps in this case.

3) can you do it for me? :)

Nope, got too many things to do already...

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org

--

--
Chris Albertson
Redondo Beach, California
home: 310-376-1029 chrisalbertson90278@yahoo.com
cell: 310-990-7550
office: 310-336-5189 Christopher.J.Albertson@aero.org

I need to find the minimum of dates in a number of tables, but "min(date)"
is, of course, an aggregate function. For example:

select key, min(a.date, b.date, c.date) as first_date
from table_a a, table_b b, table_c c
where a.key=b.key and a.key=c.key

right now I'm using:

case
when a.date < b.date and a.date < c.date then a.date
when b.date < c.date then b.date
else c.date
end

But there must be a better way. I've even looked into createing a
recursive function using an date[] as an argument (since I don't know how
to specify a function with a variable number of arguments)

Sorry if this is a FAQ but muc.lists.postgres doesn't return anything
useful, and the searchable archive seems to be down.

Thanks,

-paul

I knew I shouldn't have cut corners. More specifically, I have tables with
date ranges, for example:

--table company--
ssn company employ_from employ_to
123456789 whiznet 1999-01-01 2000-06-30

--table hmo--
ssn hmo enroll_from enroll_to
123456789 goodhealth 1999-01-01 1999-07-31
123456789 careplan 1999-08-01 2000-06-30

--table ira--
ssn broker member_from member_to
123456789 bigbroker 1999-01-01 1999-12-31
123456789 tinybroker 2000-01-01 2000-06-30

I want to get:

ssn company hmo broker from to
123456789 whiznet goodhealth bigbroker 1999-01-01 1999-07-31
123456789 whiznet careplan bigbroker 1999-08-01 1999-12-31
123456789 whiznet careplan tinybroker 2000-01-01 2000-05-30

The idea is to smash all the tables together (by ssn), then keep the
records where the dates from each table share at least one common day, then
pick the max start date, and min end date, but the max and min are the
stickler.

Since you didn't volunteer "just use not_aggregate_max()", I'm assuming
there's no built-in thing, so the follow-up question(s) would be:
1) can a function be defined with a variable number of arguments
2) can a function be created that calls itself (i.e. recursive)
3) can you do it for me? :)

Thanks,

-paul

hmmm... nice name I seem to have set for myself...

I was reading the Momjian section on Large Objects (BLOBS) and
was wondering what I might do if I am on one machine and my
Postgresql database is on another machine and I want to store
BLOBS in my database.

Terrence Brannon <metaperl@mac.com> writes:

I was reading the Momjian section on Large Objects (BLOBS) and was wondering
what I might do if I am on one machine and my Postgresql database is on
another machine and I want to store BLOBS in my database.

You can do all BLOB operations over the network.

-Doug
--
Let us cross over the river, and rest under the shade of the trees.
--T. J. Jackson, 1863