Thanks for the hint and please excuse my not thouroughly enough
reading of the documentation. I did not suspect such Feature amongst
controll structures.
Zitat von Pavel Stehule :
https://www.postgresql.org/docs/current/static/plpgsql-control-structures.html#PLPGSQL-CALL-STACK
--
Öffentli
Zitat von Pavel Stehule :
No, there is not possible to read/write client side variables from server
side.
I did not mean that the variable be replaced by its value at execution
time but at Installation time.
--
Öffentlicher PGP-Schlüssel:
http://pgp.mit.edu/pks/lookup?op=get&search=0x8F7
Zitat von Rene Romero Benavides :
What about setting the search path at the user level?
ALTER ROLE act SET search_path = act,logger;
Best.
Thanks for the inspiration. Maybe it is best to create a dedicated
user for logging anyway...
--
Öffentlicher PGP-Schlüssel:
http://pgp.mit.edu/pks/
Zitat von Adrian Klaver :
In addition to my previous suggestions:
test=# SELECT public.dblink_connect('dbname=production ');
dblink_connect
OK
test=# select * from public.dblink('show search_path') as
t1(search_path text);
search_path
-
main
test=# select
Zitat von "David G. Johnston" :
?It does not. If the array is not naturally ordered you will want to
attach a "with ordinality" clause to it for performing future ordering.
Thanks for the hints.
Kind regards
--
Öffentlicher PGP-Schlüssel:
http://pgp.mit.edu/pks/lookup?op=get&search=0x8F70
Zitat von Adrian Klaver :
What do you have your log levels set to?
Thanks for pointing this out. I put client level to debug1. So, I am
just lucky not to have got flooded with Messages?
--
Öffentlicher PGP-Schlüssel:
http://pgp.mit.edu/pks/lookup?op=get&search=0x8F70EFD2D972CBEF
-
Zitat von Vik Fearing :
For many people, this is why sourceforge died:
https://en.wikipedia.org/wiki/SourceForge#Project_hijackings_and_bundled_malware
Wow! I missed that completely. This is grave.
This message was sent using IM
Zitat von Ryan Murphy :
I could see how I could revoke permissions from, say, all users that aren't
superusers to INSERT or UPDATE certain views. However, if possible it
would be nice to get an error message about the VIEW not being updatable,
rather than a user access error, which could be mis
Andreas, thanks for your reply.
I try to set up synchronous streaming replication as try-out. I use my
laptop with Debian 9 and PostgreSQL package 10+189.pgdg90+1. And of
this PostgreSQL installation I have two clusters main (master) and
main2 (hot standby). I tried with Rigg's book and the Post
Zitat von Magnus Hagander :
On Tue, Feb 6, 2018 at 1:37 PM, Thiemo Kellner
wrote:
pgcrypto and uuid-ossp are both in the core package for PostgreSQL 10.
So, I take the documentation would be outdated?
But why do I get following error?
ERROR: function gen_random_uuid() does not exist.
create
In my testing, gen_random_uuid() is quite a bit faster than
uuid_generate_v4().
Thx for sharing your experience.
--
Öffentlicher PGP-Schlüssel:
http://pgp.mit.edu/pks/lookup?op=get&search=0x8F70EFD2D972CBEF
This message was
Zitat von Tom Lane :
I believe the point of the recommendation is that if you only need type-4
UUIDs, you can get that from pgcrypto without needing the external
dependency of the OSSP UUID library. This is a bit less urgent than when
that text was written, because now we also support building
Zitat von Peter Eisentraut :
No, what was discussed was that the Debian packaging included them in
the same package as the server instead of having a separate -contrib
package. But the modules themselves are still extensions that you need
to install into the databases.
Oh, another misundersta
Thanks for answering.
Zitat von Pavel Stehule :
Why you don't create query like
EXECUTE 'SELECT xxx FROM TAB WHERE A = $1.x AND B = $1.y' USING NEW;
I shall try. This would be the direct way, but I doubt the placeholder
$1 can be a record.
I don't understand tou your case, but usually
14 matches
Mail list logo
