"Nathan Barnett" <[EMAIL PROTECTED]> writes:
> I am trying to create an index on the date_trunc('day', columna) in a table.
> When I try to create this index, I get a parse error near the single quote.
> Is there a workaround to create this index? I am using v7.0.3 on FreeBSD.
Functional indexes
Michael Meskes <[EMAIL PROTECTED]> writes:
> Anyone willing to spend some time digging the hostname problem? That is why
> ecpg programs need to specify a hostname and not an IP number to connect,
> while psql for instance works with both?
I do not see how that can be ecpg's fault --- it just doe
"Robert B. Easter" <[EMAIL PROTECTED]> writes:
UNION JOIN is deprecated
>>
>> Oh? By whom?
> I read it in the SQL spec. ANSI/ISO 9075-2 1999 (final).
> It is intended that the following features will be removed at a
> later date from a revised version of this part of I
On Sunday 07 January 2001 13:13, Tom Lane wrote:
> "Robert B. Easter" <[EMAIL PROTECTED]> writes:
> > UNION JOIN is deprecated
>
> Oh? By whom?
>
> The reason 7.1 doesn't have it is I didn't have time for it, not that
> we don't plan to do it ever.
I read it in the SQL spec. ANSI/ISO 9075-2 199
Radoslaw Stachowiak <[EMAIL PROTECTED]> writes:
> Could You write what is suggested path to change encoding from SQLASCII to
> another (ISO88592) when I'd upgrade to 7.1 from 7.03 ?
> I know have SQLASCII, but I'll need to change it to sth more suitable.
Should be sufficient to make sure you hav
"Robert B. Easter" <[EMAIL PROTECTED]> writes:
> UNION JOIN is deprecated
Oh? By whom?
The reason 7.1 doesn't have it is I didn't have time for it, not that
we don't plan to do it ever.
regards, tom lane
Denis Perchine <[EMAIL PROTECTED]> writes:
> Cost is something really wierd Why?
That's how the enable stuff works for plan types that can't be ignored
completely: it just adds a big constant to the estimated cost. If there
is no other alternative plan, you get the unwanted plan type anyway.
Denis Perchine <[EMAIL PROTECTED]> writes:
>> You could probably get a plan without the sort step if you said
>> ... order by variant_id, rcptdate;
> No way, it just get all tuples for the qual, sort them, and the limiting.
> That's horrible...
> slygreetings=> explain select * from users where
"Robert B. Easter" <[EMAIL PROTECTED]> writes:
> Just for the heck of it, I tried to execute all this sql on 7.0.3 and got
> this:
> psql:join2.sql:23: pqReadData() -- backend closed the channel unexpectedly.
> I knew it wouldn't run it, but didn't think it would crash.
7.0 had the beginnings of
Hi
my Postgres is running on Linux. The Clients
are partly linux, partly Win boxes.
Now i've the problem with the typical german letter (ä,ö, and so on)
On the linux machine the are shown different to windows.
So if somebody enters a record with special chars on Win
the user on Linux have othe
*** Tom Lane <[EMAIL PROTECTED]> [Tuesday, 02.January.2001, 18:38 -0500]:
> To enlarge a little more: the most common way to get burnt by this is
> to have different LC_xxx environment variables when starting the
> postmaster from a boot script as you do when running initdb or starting
> the postm
"Ian deSouza" wrote:
>Anyone know how to list what sequences already exist?
In psql: \ds
--
Oliver Elphick[EMAIL PROTECTED]
Isle of Wight http://www.lfix.co.uk/oliver
PGP: 1024R/32B8FAA1: 97 EA 1D 47 72 3F 28 47 6B 7E 39 CC 56 E4
On Sunday 07 January 2001 00:53, Robert B. Easter wrote:
> Union join:
> T1 UNION JOIN T2
>
> is not implemented. Nice! :)
[snip]
> SELECT * FROM a UNION JOIN b;
>
> ERROR: UNION JOIN is not implemented yet
> psql:/home/reaster/sql/join/join.sql:37: ERROR: UNION JOIN is not
> implemented yet
Are the progresql c++ wrappers ready for prime time?
I am having some problems with them while trying to
compile and link on my RH6.2 machine:
(1) Some of the headers have the following comments:
* NOTES
* Currently under construction.
(2) Compiling gives the following warnings
Hi
How can I change the default transaction isolation level to be
SERIALIZABLE, so that I don't have to use
BEGIN;
SET TRANSACTION ISOLATION LEVEL SERIALIZABLE;
in each transaction ?
Karol Wieloch
HI,
I couldn't find an answer in doc, so here's the question. Is it possible
to to have result similar to explain command in postgres log. I've turned
on plan in pg_options but results are not to readable for me (anyone has a
doc about it?). I need it to find queries which I can optimize by using
On Thu, Jan 04, 2001 at 10:39:14AM -0700, T F wrote:
> Does anyone know how to get pgaccess and libpq/libpq++ to crypt the
> passwords? I'd like to use crypt in my pg_hba.conf file for all hosts,
> but when I do the only way I can connect is via psql and a relatively
> new version of the perl mod
Hi,
another interesting thing...
This is current 7.1.
slygreetings=> explain select * from users where variant_id=5 AND active='f'
order by rcptdate,variant_id,active limit 60;
NOTICE: QUERY PLAN:
Limit (cost=13005.10..13005.10 rows=60 width=145)
-> Sort (cost=13005.10..13005.10 rows=34
Id> Figure it out:
metamousetrap=>> create table my_table (first serial, second varchar);
Id> NOTICE: CREATE TABLE will create implicit sequence 'my_table_first_seq' for
SERIAL column 'my_table.first'
Id> NOTICE: CREATE TABLE/UNIQUE will create implicit index 'my_table_first_key' for
table 'm
DROP SEQUENCE EMPLOYEE_ID_seq;
And it is in the docs..
Id> The PostgreSQL doc referencing this situation is at
Id> http://www.postgresql.org/devel-corner/docs/user/datatype.htm#AEN1181
Id> but it doesn't explain what the syntax is nor does it give an example.
Id> So if Employee has an attribute
20 matches
Mail list logo
