I have a parallel inheritance going on,
so I was wondering if there was a way
to re-name a derived column? This would
make my design clearer.
-
CREATE TABLE B
( NAME VARCHAR(10) );
CREATE TABLE C
( ... ) INHERITS(B);
CREATE TABLE X
(
A VARCHAR(10),
B VARCHAR(10),
CONS
Just for PgSQL's development group think about
I made a mistake typing a query that generates a strange result (Very strange).
The query: select text('12345678'::float8);
It returns a date in datetime format !!
If you use: select ('12345678'::float8)::text;everything runs well
Dear Thomas,
I began with ver.6.4 beta. What do I have to do ? Come back to 6.3 ?
However, I tested more and the problem appear when you use "check" clause of "create
table".
Could you reproduce the situation with your PostGres 6.3 ? See the procedures below.
create table TBL (FLD1 int4, FLD2
Hi,
I create a C function: create function FUNC (text) returns text as '/path/func.so'
language 'C';
As described in manual, I allocate space to text return: ret=(text *) palloc
(datasize+VARHDRSZ);
The function works fine with small tables, but some problem occurs during server pfree
with b
Just a general note. How close are PostgreSQL's
object extensions to those at the ODMG? How hard
would it be to write complant interface? Anyone
working on it?
See: http://www.odmg.org/
Clark
It depends of data type - unfortunately max was not defined for ALL
appropriate types. Through
testo=> select max(bypref) from inetspec;
max
---
24
where bypref is int2.
On Sat, 6 Feb 1999, dustin sallings wrote:
> Date: Sat, 6 Feb 1999 13:10:56 -0800 (PST)
> From: dustin sallings <[EMAIL
Do you plan the transaction log support in 6.5 release, and could you give us a brief description on ?Will it allow to recover until the last committed transaction in case of disk crash ?Thanks a lot for your answer and for all the things you do to make Postgres the best.RegardsJérôme Doucerain
Jeff Gerhart <[EMAIL PROTECTED]> writes:
| machine) without installing the entire postgres environment. I
| thought I'd seen a place where somebody had broken the client and
| backend into seperate RPMs, but I can't find where I saw that
| discussed.
Once, I've done this for SuSE Linux.
On Sun, 7 Feb 1999, Bob VonMoss wrote:
> Tom Lane wrote:
>
> > Bob VonMoss <[EMAIL PROTECTED]> writes:
> > > [ can connect via unix socket, but not via TCP ]
> >
> > > The administrator says this is how postmaster is invoked:
> > > /usr/local/pgsql/bin/postmaster -S -i -D /usr/local/pgsql/data -
Tom Lane wrote:
> Bob VonMoss <[EMAIL PROTECTED]> writes:
> > [ can connect via unix socket, but not via TCP ]
>
> > The administrator says this is how postmaster is invoked:
> > /usr/local/pgsql/bin/postmaster -S -i -D /usr/local/pgsql/data -p 5432
>
> OK, that eliminates the "forgot -i" gotcha.
10 matches
Mail list logo