On Tue, 12 Dec 2000 12:04:46 +0100, Roger Wernersson wrote:
>
> I can't BEGIN - SELECT FOR UPDATE - INSERT or UPDATE - COMMIT as someone
> might insert after my SELECT and before my INSERT.
>
correct me if i'm wrong, but a select for update locks the table for
an insert or an update until the
According to my copy of SQL For Smarties by Joe Celko (2nd ed, p. 411):
> The UNION removes all duplicate rows from the results and does
> not care from which table the duplicate rows came. We could
> use this feature to write a query to remove duplicates from a
> table:
>(TABLE tableA)
>
According to my copy of SQL For Smarties by Joe Celko (2nd ed, p. 411):
> The UNION removes all duplicate rows from the results and does
> not care from which table the duplicate rows came. We could
> use this feature to write a query to remove duplicates from a
> table:
>(TABLE tableA)
>
n.b. to clarify, i got PL/Perl working on RH Linux 6.2
am in the process of attempting to get it working on NT.
==BEGIN FORWARDED MESSAGE==
when i did what's described below, i had no problems at all with
PL/Perl (and, in fact, am using it for a couple of trig
when i did what's described below, i had no problems at all with
PL/Perl (and, in fact, am using it for a couple of triggers that
are lightly used in a production environment)
http://www.postgresql.org/cgi/cvsweb.cgi/pgsql/src/pl/plperl/README?rev=1.2&content-type=text/x-cvsweb-
markup
you can
On Wed, 20 Sep 2000 21:24:17 -0400, Neil Conway wrote:
> I believe he is using straight Pg - the perl interface to Postgres.
> AFAIK, that's independant of DBI::Pg, which is the Postgres
> driver for DBI.
>
ahhh! "i see" said the blind man :)
> I agree with you, however: I've found DBD::Pg q
actually they're saying two different things :)
first, to explain my example a bit better:
the difference between this:
> > begin;
> > insert into foo (A,B) values (B);
> > select currval('foo_A_seq');
> > commit;
and this:
> > insert into foo (A,B) values (B);
> > select currval('foo_A
in effect, this turns the filesystem into a "poor-mans" balanced tree.
the rdbms gives you a "rich-mans" balanced tree, but along with the
overhead of the rdbms.
cheers
--e--
On Thu, 21 Sep 2000 15:20:39 +0300, Alessio Bragadini wrote:
> Neil Conway wrote:
>
> > > a BLOB. Conversely, Un
i can't seem to locate the reference to the documentation
you mention in the perldoc for DBD::Pg, can you reference
it?
as a general note, you should be using the DBI interface
with DBD::Pg and not calling the methods in DBD::Pg
directly. they're "private" and likely to change.
in other words
On Wed, 20 Sep 2000 17:54:40 -0400, Joe Kislo wrote:
> > you should not use the OID value for application level work. for one thing,
> > it's not portable, and if you rebuild the database it'll change. you should
> > consider it a strictly internal value.
>
> Hmm, I think you missed what
to replace the one currently in $PGSRC/src/pl/plperl
it encompasses the information in that document while adding more structure
and more specific details about what is needed. it also addresses
a couple of issues that came up when i had personally installed it.
since there is no email address
for what it's worth, when i run these two tests, i
get the correct results
i'm using RedHat 6.2 also.
here are more details:
[ebridges@sleeepy]$ uname -a
Linux sleeepy 2.2.16 #2 SMP Mon Jul 31 14:51:33 EDT 2000 i686 unknown
[ebridges@sleeepy]$ psql -V
psql (PostgreSQL) 7.0.2
Portions Copyright
12 matches
Mail list logo
