Peter Eisentraut <[EMAIL PROTECTED]> writes:
> Bruce Momjian writes:
>> If a file is needed by three non-backend directories, /port seems to be
>> the proper place for it.
> src/port is intended for replacement implementations of standard library
> functions.
I concur, src/port is *not* the right
I'm not sure this is a fair assessment of statement level triggers.
Yes, in MSSQL you can access the rows involved in the statement, but
in Oracle you cannot (emphasis added):
"Accessing Column Values in Row Triggers
Within a trigger body of a *row trigger*, the PL/SQL code and SQL
statements
> Seriously, I think there's a good case for banning a few characters in
> at least some names - like []<>'"~#*|\ , say
Why? They're allowed in all other identifiers. And what if someone
already has a database full of usernames with those chars? They wouldn't
be able to load their dump properly
forwarded to the compiler guys at SCO.
LER
--On Thursday, August 07, 2003 19:10:04 +0200 [EMAIL PROTECTED] wrote:
Hi Tom, Hi Larry
After updating from cvs I'm going a little further. But still have
problems:
UX:acomp: ERROR: "fe-protocol3.c", line 1402: internal compiler error:
can't deal with
"Tom Lane" <[EMAIL PROTECTED]> writes:
> Josh Berkus <[EMAIL PROTECTED]> writes:
> > From my perspective, we could really use a "delimiter" between the
> > "fields" of log output which is unlikely to appear within those fields
> > instead of parsing by character count, rather than making dbname a
>
To fix it, remove -K inline from src/port/unixware's CFLAGS.
Dave Prosser of SCO has a fix, but this is the most expedient fix for now.
--On Thursday, August 07, 2003 18:58:16 -0500 Larry Rosenman
<[EMAIL PROTECTED]> wrote:
I've filed a bug with the folks I know that work on the compilers.
w
Robert Creager <[EMAIL PROTECTED]> writes:
> Anything to look at before I kick it?
pg_locks and pg_stat_activity, if you can select from them in a non-stuck
backend.
> It's not built with debug, but I
> can still get a backtrace.
Might be useful.
regards, tom lane
-
Tom Lane wrote:
Christopher Kings-Lynne <[EMAIL PROTECTED]> writes:
So if you agree that there is a quoting problem,and you don't mind
breaking backwards compatibility for it, I'll do a complete patch...
I don't see any backwards-compatibility issue, because usernames
containing double quo
I'm curious to know what that will do to performance.
andrew
Larry Rosenman wrote:
To fix it, remove -K inline from src/port/unixware's CFLAGS.
Dave Prosser of SCO has a fix, but this is the most expedient fix for
now.
---(end of broadcast)---
On Fri, 8 Aug 2003, Tom Lane wrote:
> "Mendola Gaetano" <[EMAIL PROTECTED]> writes:
> > Again the error:
>
> > kalman=# select bar();
> > ERROR: consistency check on SPI tuple count failed
> > CONTEXT: PL/pgSQL function "bar" line 5 at for over select rows
> > kalman=# select bar();
> > ERROR:
On Sat, 9 Aug 2003, Kurt Roeckx wrote:
> Date: Sat, 9 Aug 2003 17:45:55 +0200
> From: Kurt Roeckx <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> Cc: Tom Lane <[EMAIL PROTECTED]>,
> pgsql-hackers list <[EMAIL PROTECTED]>,
> Larry Rosenman <[EMAIL PROTECTED]>
> Subject: Re: [HACKERS] 7.4bet
From: "Tom Lane" <[EMAIL PROTECTED]>
> To forestall this scenario, I'm thinking of introducing backoff into the
> sleep intervals --- that is, after first failure to get the spinlock,
> sleep 10 msec; after the second, sleep 20 msec, then 40, etc, with a
> maximum sleep time of maybe a second. Th
Thanks, Bruce
On Tue, 5 Aug 2003, Bruce Momjian wrote:
>
> Done.
>
> ---
>
> Oleg Bartunov wrote:
> > On Tue, 5 Aug 2003, Bruce Momjian wrote:
> >
> > > Oleg Bartunov wrote:
> > > > Bruce, you forgot new contrib/tsearch2 modu
> More to the point, this is highly incomplete... you did not teach the
> adjacent getid routine about this, and there is code in (at least)
> pg_dump.c that knows the quoting conventions used here.
Hang on - those routines can parse the acls just fine? How? How do they
handle usernames with equ
Does anyone mind if I remove src/interfaces/cli? It's clearly outdated
and useless.
--
Peter Eisentraut [EMAIL PROTECTED]
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
--On Saturday, August 09, 2003 11:47:43 -0500 Larry Rosenman
<[EMAIL PROTECTED]> wrote:
I tried(!) to load my 7.3.4 data into 7.4CVS.
the Bricolage folks have managed to make a circular definition (at least
not loadable).
why does each setval() call invoke the pager?
the dump I used is at:
h
Ok, you're right it works with -g until...
UX:cc: WARNING: debugging and optimization mutually exclusive; -O disabled
Undefined first referenced
symbol in file
PQparameterStatus pg_backup_db.o
UX:ld: ERROR: Symbol referencing erro
Bruce Momjian wrote:
It might be a bit risky getting pg_dump to use it though?
I don't think we every want pg_dump to use it --- better accurate than
pretty in there.
Agreed.
There seems to be some tough assumptions that have to
be made in that function that are better used for visual-only c
When running make check, I get:
The program
/usr/local/pgsql/bin/postgres
needed by initdb does not belong to PostgreSQL version 7.4beta1,
or there may be a configuration problem.
What's the point of running a regression tests if it's going to
use the old binary?
Kurt
---
On Sat, Aug 09, 2003 at 12:24:36PM -0400, Tom Lane wrote:
> Kurt Roeckx <[EMAIL PROTECTED]> writes:
> > What's the point of running a regression tests if it's going to
> > use the old binary?
>
> It doesn't, ordinarily. Feel free to trace through the scripts and
> find out what went wrong.
I'm n
Andreas Pflug <[EMAIL PROTECTED]> writes:
> - Implement a way to enable triggers to check which columns are affected
> by the triggering statement.
This can already be done by comparing old and new values, no?
I don't have a lot of sympathy for the idea that checking what the
original UPDATE tou
On Fri, Jul 25, 2003 at 05:47:50PM -0400, Bruce Momjian wrote:
> I am seeing the following parallel regression test failures. Any idea
> on the cause?
I think I saw about the same thing once, but I run the test again
and it didn't show up anymore at all. I'm not sure what it
exactly was, but it
Tom Lane wrote:
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> > Seems we have a problem with pooled connections and WITH HOLD cursors.
> > We have code to reset transaction state and variables via RESET ALL, but
> > how do we remove WITH HOLD cursors when we pass a connection to a new
> > client?
>
Hi Tom, Hi Larry
After updating from cvs I'm going a little further. But still have
problems:
UX:acomp: ERROR: "fe-protocol3.c", line 1402: internal compiler error: can't deal
with op BMOVE
gmake[3]: *** [fe-protocol3.o] Error 1
gmake[2]: *** [all] Error 2
gmake[1]: *** [all] Error 2
gmake: ***
Bruce Momjian <[EMAIL PROTECTED]> writes:
> What would be interesting is a CREATE OR REPLACE functionality for
> prepared cursors, where you could ask for it to be prepared, but if it
> already existed, it would do nothing, or something like that.
I don't think you could call that CREATE OR REPLAC
> > Hrm, after an hour of searching and reading, I think one of the
> > better papers on the subject can be found here:
> > http://www.cs.ust.hk/faculty/dimitris/PAPERS/TKDE-NNmodels.pdf
>
> Interesting paper, but I don't see the connection to index order
> correlation?
Nothing that I found was n
26 matches
Mail list logo
