> Personally I would prefer to have a trigger on a metadata table
> where I could
> trigger vacuuming a particular table each n number of
> transactions(Oh it would
> be great if that vacuum runs in background not blocking meta data
> table.. just
> a wishlist...). Can anybody tell me which table
On 3 Sep 2002 at 8:55, Mario Weilguni wrote:
> I know everyone is busy with the 7.3beta, but maybe this is something to think of
>before releasing the beta. Currently VACUUM will vacuum every table, but sometimes
> it's desireable to leave tables untouched because the're mostly static or protoco
Am Dienstag, 3. September 2002 00:26 schrieb Peter Eisentraut:
> Mario Weilguni writes:
> > Timestamps in 7.2.x are this format: 2002-07-04 15:19:11.363562+02
> > 7.3 expects a timestamp per default in this format: 2002-09-02
> > 08:51:27,223455+02
>
> If you're not running the reallly latest 7.3
I know everyone is busy with the 7.3beta, but maybe this is something to think of
before releasing the beta. Currently VACUUM will vacuum every table, but sometimes
it's desireable to leave tables untouched because the're mostly static or protocol
tables. In my case this would be the pg_largeobj
Christopher Kings-Lynne dijo:
> findoidjoins doens't seem to compile:
>
> gmake[1]: Entering directory `/home/chriskl/pgsql-head/contrib/findoidjoins'
> gcc -pipe -O -g -Wall -Wmissing-prototypes -Wmissing-declarations -I../../sr
> c/interfaces/libpgeasy -I../../src/interfaces/libpq -I. -I../..
findoidjoins doens't seem to compile:
gmake[1]: Entering directory `/home/chriskl/pgsql-head/contrib/findoidjoins'
gcc -pipe -O -g -Wall -Wmissing-prototypes -Wmissing-declarations -I../../sr
c/interfaces/libpgeasy -I../../src/interfaces/libpq -I. -I../../src/include
-c -o findoidjoins.o findo
Le Mardi 3 Septembre 2002 04:28, Gerhard Häring a écrit :
> PostgreSQL will have a
> native win32 port,
Just out of interest, what is the advancement of the Windows port.
Best regards, Jean-Michel
---(end of broadcast)---
TIP 3: if posting/reading
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Yes, I haven't gotten to the release checklist yet. Let's delay a day.
Or at least late in the day tomorrow. I have some loose ends to clean
up yet as well, but I'm beat and am going to bed.
But I assume we are now officially in feature freeze, right
Tom Lane wrote:
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> > I don't know of any other items holding up the packaging.
>
> Gotta brand the thing as 7.3beta1 not 7.3devel, no?
Yes, I haven't gotten to the release checklist yet. Let's delay a day.
I have a 17k line log file down to 3.5k lines
Bruce Momjian <[EMAIL PROTECTED]> writes:
> I don't know of any other items holding up the packaging.
Gotta brand the thing as 7.3beta1 not 7.3devel, no?
regards, tom lane
---(end of broadcast)---
TIP 5: Have you checked ou
I am still working on the 7.3 HISTORY file. I have extracted the items,
but I have to worksmith them and write an introduction.
It is midnight here now. I don't think I can finish before ~3am and at
that point, I am not sure I will know what I am writing.
Basically, one day of feature freeze w
Marc, do you want to start trimming /contrib? I know the MySQL and
Oracle tools each have their own websites.
--
Bruce Momjian| http://candle.pha.pa.us
[EMAIL PROTECTED] | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ C
If we got rid of the other NOT NULL != CHECK (a IS NOT NULL) instance, may
as well get rid of the one on this page:
http://developer.postgresql.org/docs/postgres/sql-createdomain.html
Chris
---(end of broadcast)---
TIP 1: subscribe and unsubscrib
> Actually, Linux Journal (and their editors) are fans of PostgreSQL.
>
> This year, MySQL may actually have clued in to transactions and a few
> other big database features. I don't know that they actually *have*
> these features polished up, but LJ is giving them credit for trying...
It still d
Thomas Lockhart wrote:
> Actually, Linux Journal (and their editors) are fans of PostgreSQL.
>
> This year, MySQL may actually have clued in to transactions and a few
> other big database features. I don't know that they actually *have*
> these features polished up, but LJ is giving them credit f
Bruce Momjian wrote:
> No, that's OK. I have things pointing to my machine like the open
> patches list and stuff. It turns out that my ISP is all I need to
> change, and .US can do it whenever they get to it. I thought I was more
> vulnerable than I thought.
Actually, now that I think of it,
Actually, Linux Journal (and their editors) are fans of PostgreSQL.
This year, MySQL may actually have clued in to transactions and a few
other big database features. I don't know that they actually *have*
these features polished up, but LJ is giving them credit for trying...
-
Vince Vielhaber wrote:
> On Mon, 2 Sep 2002, Bruce Momjian wrote:
>
> > Vince Vielhaber wrote:
> > > On Fri, 30 Aug 2002, Bruce Momjian wrote:
> > >
> > > > I have moved my main computer to my new house. The DNS is still being
> > > > updated, but the new IP address is:
> > > >
> > > > 2
On Mon, 2 Sep 2002, Bruce Momjian wrote:
> Vince Vielhaber wrote:
> > On Fri, 30 Aug 2002, Bruce Momjian wrote:
> >
> > > I have moved my main computer to my new house. The DNS is still being
> > > updated, but the new IP address is:
> > >
> > > 207.106.42.251
> > >
> > > Vince, would you upda
* Christopher Kings-Lynne <[EMAIL PROTECTED]> [2002-09-03 10:17 +0800]:
> > On Tue, 3 Sep 2002, Christopher Kings-Lynne wrote:
> >
> > > Database: MySQL Honorable Mention: PostgresSQL
> >
> > Nothing wrong with that. From your list it seemed that in the categories
> > where there were competing op
Christopher Kings-Lynne wrote:
> > On Tue, 3 Sep 2002, Christopher Kings-Lynne wrote:
> >
> > > Database: MySQL Honorable Mention: PostgresSQL
> >
> > Nothing wrong with that. From your list it seemed that in the categories
> > where there were competing open source and open source/commercial back
On Tue, 3 Sep 2002, Christopher Kings-Lynne wrote:
> database market now includes SapDB for crying out loud - how can MySQL (and
> even postgres really) compete with that? And what about Firebird? I think
And berkeley db. *Easily* the most widely used open source database and
the most profitab
> On Tue, 3 Sep 2002, Christopher Kings-Lynne wrote:
>
> > Database: MySQL Honorable Mention: PostgresSQL
>
> Nothing wrong with that. From your list it seemed that in the categories
> where there were competing open source and open source/commercial backed
> software then the latter seemed to win
Gavin Sherry wrote:
> Okay, my bad. From my reading of the email exchange, I thought people
> wanted this on -- always. The best solution for this, in my opinion, is to
> have a magic value "off" which the error code lookup translates to some
> number > PANIC.
What do people think? I thought we
On Tue, 3 Sep 2002, Christopher Kings-Lynne wrote:
> Database: MySQL Honorable Mention: PostgresSQL
Nothing wrong with that. From your list it seemed that in the categories
where there were competing open source and open source/commercial backed
software then the latter seemed to win over.
Thi
Bruce suggested that we need a porting guide to help people look for
application and client-library code that will be broken by the changes
in PG 7.3. Here is a first cut at documenting the issues.
Comments welcome --- in particular, what have I missed?
regards, tom lane
On Mon, 2 Sep 2002, Bruce Momjian wrote:
> > >> Oh, didn't you put in that patch to provide a GUC level control?
> >
> > > Yes, but what level do you set it at to turn it off?
> >
> > FATAL? PANIC?
>
> He doesn't support those levels:
>
> test=> set log_min_error_statement = fata
Looks like we got an honourable mention *sigh*:
Server Appliance: SnapGear for Lite/Lite+ SOHO Firewall/VPN Client
Honorable Mention: Sun Microsystems for Cobalt Qube
Security Tool: GPG
Web Server: IBM for xSeries
Honorable Mention: Sun Microsystems for Cobalt RaQ XRT
Enterprise Application Serve
Christopher Kings-Lynne wrote:
> > o -ALTER TABLE ADD PRIMARY KEY (Tom)
> > o -ALTER TABLE ADD UNIQUE (Tom)
> >
> > AFAIR, I didn't do either of those; I think Chris K-L gets the credit.
>
> Actually, I did ADD UNIQUE originally after lots of coding and then you went
> and made it work by
> o -ALTER TABLE ADD PRIMARY KEY (Tom)
> o -ALTER TABLE ADD UNIQUE (Tom)
>
> AFAIR, I didn't do either of those; I think Chris K-L gets the credit.
Actually, I did ADD UNIQUE originally after lots of coding and then you went
and made it work by changing a couple of lines in the gramma
Vince Vielhaber wrote:
> On Fri, 30 Aug 2002, Bruce Momjian wrote:
>
> > I have moved my main computer to my new house. The DNS is still being
> > updated, but the new IP address is:
> >
> > 207.106.42.251
> >
> > Vince, would you update the developer's page and change
> > 'candle.pha.pa.us'
On Sat, 31 Aug 2002, Marc G. Fournier wrote:
>
> This is all in Vince's area ...
Correction. You're not looking, it's in the users lounge. We've
covered the button thing already.
>
>
> On 23 Aug 2002, Greg Copeland wrote:
>
> > I must be blind. I don't see links to gborg anywhere on the deve
On Sat, 31 Aug 2002, Marc G. Fournier wrote:
>
> This is all in Vince's area ...
You're right. You don't see 'em.
>
>
> On 23 Aug 2002, Greg Copeland wrote:
>
> > I must be blind. I don't see links to gborg anywhere on the developer
> > or main site web pages. Perhaps more obvious a sister s
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> Now that the "multibyte"-based character set recoding is a fixed part of
> the feature set, is there any need to keep the "Cyrillic" recode support?
It's probably not worth maintaining anymore ...
regards, tom lane
-
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> Perhaps one should consider removing the autocommit option. It's no use
> if it's there but everything breaks when you turn it on.
As far as I'm concerned, it's in there for one reason only (as far as
7.3 goes): so that we can run the NIST SQL compl
Alvaro Herrera <[EMAIL PROTECTED]> writes:
> Tom Lane dijo:
>> Hm. Now that I look at it, the forms REINDEX TABLE and REINDEX INDEX
>> don't seem to have any permissions checks at all :-(. One would think
>> that they need to require ownership rights on the target table.
> They are done in tco
Just a quick one before I package up the wrong thing ... where should I be
pulling docs from? :)
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL P
Tom Lane writes:
> There are some things we can tweak to make the clients less broken than
> they are now --- for instance, all of libpq's startup-time SET commands
> could be switched to "BEGIN; SET ...; COMMIT;" which will work the same
> with or without autocommit --- but I don't think we can
On Fri, 30 Aug 2002, Bruce Momjian wrote:
> I have moved my main computer to my new house. The DNS is still being
> updated, but the new IP address is:
>
> 207.106.42.251
>
> Vince, would you update the developer's page and change
> 'candle.pha.pa.us' to this fixed IP address. It may take
Gettext cannot handle compile-time string concatenation with macros. This
is a made-up example:
printf(gettext("now at file position " INT64_FORMAT), (int64) offset);
At the time when the message catalogs are extracted, INT64_FORMAT is
unknown.
The solution in the Gettext manual is to rewr
Mario Weilguni writes:
> Timestamps in 7.2.x are this format: 2002-07-04 15:19:11.363562+02
> 7.3 expects a timestamp per default in this format: 2002-09-02 08:51:27,223455+02
If you're not running the reallly latest 7.3 tip, update and try again.
Something related to this was fixed recently.
-
Now that the "multibyte"-based character set recoding is a fixed part of
the feature set, is there any need to keep the "Cyrillic" recode support?
--
Peter Eisentraut [EMAIL PROTECTED]
---(end of broadcast)---
TIP 5: Have you checked our extens
On Mon, 2 Sep 2002, Tom Lane wrote:
> So my take is that anything that needs initdb doesn't get in after
> beta1, unless it's a "must fix" bug. What have we got in the queue
> that would require system catalog changes?
Agreed, but, have we ever done such when it *wasn't* required? I know in
th
Tom Lane dijo:
> Hm. Now that I look at it, the forms REINDEX TABLE and REINDEX INDEX
> don't seem to have any permissions checks at all :-(. One would think
> that they need to require ownership rights on the target table.
They are done in tcop/utility.c AFAICS.
--
Alvaro Herrera ()
"Cuand
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom, when you loosened the restriction on reindexing toast tables, did
> you continue to restrict indexing of system tables to the superuser? Is
> that required?
Hm. Now that I look at it, the forms REINDEX TABLE and REINDEX INDEX
don't seem to have a
Tom, when you loosened the restriction on reindexing toast tables, did
you continue to restrict indexing of system tables to the superuser? Is
that required?
--
Bruce Momjian| http://candle.pha.pa.us
[EMAIL PROTECTED] | (610) 359-1001
+ If your lif
Tom Lane writes:
> What I'm currently thinking we should do is default largefile support to
> off in HPUX < 11.0; is there a convenient way to accomplish that in
> autoconf?
Something like this maybe (before AC_SYS_LARGEFILE):
case $host_os in hpuxZYX*)
if test "${enable_largefile+set}" != set;
Tom Lane wrote:
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> > Tom Lane wrote:
> >> I don't think it *can* be fixed in a reasonable fashion until we have
> >> notification to the client side about what the backend's transaction
> >> state is; which is one of the protocol-change items for 7.4.
>
> > * Have SERIAL generate non-colliding sequence names when we have
> > auto-destruction
> >
> > They should be pretty well non-colliding now. What's the gripe exactly?
>
> The issue was that when there were name collisions, we threw an error
> instead of trying other sequence names. We h
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> I don't think it *can* be fixed in a reasonable fashion until we have
>> notification to the client side about what the backend's transaction
>> state is; which is one of the protocol-change items for 7.4.
> Why can't we just turn on
OK, we just need to make that clear in the release notes. I had
forgotten about the change.
---
Tom Lane wrote:
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> > Tom Lane wrote:
> >> Yes. The release notes mention:
> >> SER
On Mon, 2002-09-02 at 09:35, Tatsuo Ishii wrote:
> You need not to specify --enable-syslog in 7.3 BTW.
OK, thanks.
> This happens because the path to shared objs are defined at the
> compile time. I think you don't get the failure once you install
> PostgreSQL. However it's not convenience since
Tom Lane wrote:
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> > Tom Lane wrote:
> >> This isn't really done; the backend side is probably okay, but we have
> >> a ton of client-side code that will malfunction if you try to run it in
> >> autocommit-off state.
>
> > My feeling is that we have to f
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> Yes. The release notes mention:
>> SERIAL no longer implies UNIQUE; specify explicitly if index is wanted
> What was the logic for this change?
See the thread back around 17 Aug:
http://archives.postgresql.org/pgsql-hackers/2002-08/
Patch applied to test for Ant >= 1.5. Autconf run.
---
Bruce Momjian wrote:
> Tatsuo Ishii wrote:
> > > Yeah, it seems to require Ant 1.5 now.
> >
> > Thanks I got Ant 1.5 and now PostgreSQL builds fine.
> >
> > > This s
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> This isn't really done; the backend side is probably okay, but we have
>> a ton of client-side code that will malfunction if you try to run it in
>> autocommit-off state.
> My feeling is that we have to fix this during beta.
No we do
On Mon, 2 Sep 2002, Christopher Kings-Lynne wrote:
> Ummm...I totally don't understand the format of that array at all!
stop words in suffix tree :) I've told it's default list of stop
words and it's certainly far from perfect. Please find attached file
with stop words we used (I just found on m
Serguei A. Mokhov wrote:
>
>
> On Mon, 2 Sep 2002, Tom Lane wrote:
>
> > Date: Mon, 02 Sep 2002 10:33:49 -0400
> > From: Tom Lane <[EMAIL PROTECTED]>
> > To: Christopher Kings-Lynne <[EMAIL PROTECTED]>
> > Cc: Bruce Momjian <[EMAIL PROTECTED]>,
> > PostgreSQL-development <[EMAIL PROTECTED]
Tom Lane wrote:
> Chris <[EMAIL PROTECTED]> writes:
> > I'm playing around with the CVS version and noticed a change from 7.2 in
> > regards to serial datatypes - they no longer automatically have an
> > index. Is this a deliberate thing?
>
> Yes. The release notes mention:
>
> SERIAL no long
Tom Lane wrote:
> "Christopher Kings-Lynne" <[EMAIL PROTECTED]> writes:
> > You can probably nail some TODOs:
>
> > * Allow autocommit so always in a transaction block
>
> This isn't really done; the backend side is probably okay, but we have
> a ton of client-side code that will malfunction if
Tom Lane dijo:
> "Christopher Kings-Lynne" <[EMAIL PROTECTED]> writes:
> > You can probably nail some TODOs:
>
> o -Cluster all tables at once using pg_index.indisclustered set during
> previous CLUSTER
>
> This is not done, unless we are going to accept Alvaro's last-minute
>
Chris <[EMAIL PROTECTED]> writes:
> I'm playing around with the CVS version and noticed a change from 7.2 in
> regards to serial datatypes - they no longer automatically have an
> index. Is this a deliberate thing?
Yes. The release notes mention:
SERIAL no longer implies UNIQUE; specify expli
On Mon, 2 Sep 2002, Tom Lane wrote:
> Date: Mon, 02 Sep 2002 10:33:49 -0400
> From: Tom Lane <[EMAIL PROTECTED]>
> To: Christopher Kings-Lynne <[EMAIL PROTECTED]>
> Cc: Bruce Momjian <[EMAIL PROTECTED]>,
> PostgreSQL-development <[EMAIL PROTECTED]>
> Subject: Re: [HACKERS] I am done
>
> "C
"Christopher Kings-Lynne" <[EMAIL PROTECTED]> writes:
> You can probably nail some TODOs:
> * Allow autocommit so always in a transaction block
This isn't really done; the backend side is probably okay, but we have
a ton of client-side code that will malfunction if you try to run it in
autocommi
Hi,
I'm playing around with the CVS version and noticed a change from 7.2 in
regards to serial datatypes - they no longer automatically have an
index. Is this a deliberate thing? I did a search in the archives but
didn't come across mention of the change. A pointer to discussion on
this would
Tatsuo Ishii wrote:
> > Yeah, it seems to require Ant 1.5 now.
>
> Thanks I got Ant 1.5 and now PostgreSQL builds fine.
>
> > This should probably go in the
> > configure script, ie a test for Ant 1.5 if java enabled?
>
> Sounds nice idea.
Yes, I am going to add a test for >= Ant 1.5 to config
I have already received a patch that modifies the regression tests and
it seems things are working. However, it would be good for you to take
a look. Fortunately, that can be done anytime during beta. Thanks.
---
Oleg Bar
> I'm still getting conversion test failures on RH 7.2, 7.3, and Null beta.
>
> My confiugre arguments are:
>
> ./configure --prefix=/opt/postgresql --with-java --with-python --with-openssl
>--enable-syslog --enable-debug --enable-cassert
> --enable-depend
You need not to specify --enable-sys
On Mon, 2002-09-02 at 02:36, Christopher Kings-Lynne wrote:
> Whatver happened to Rod's work on the BETWEEN command? I remember he got to
> the stage of realising a lot of execetor changes had to be made...?
I've not had time to implement the optimizer portion.
---(end
> Yeah, it seems to require Ant 1.5 now.
Thanks I got Ant 1.5 and now PostgreSQL builds fine.
> This should probably go in the
> configure script, ie a test for Ant 1.5 if java enabled?
Sounds nice idea.
--
Tatsuo Ishii
---(end of broadcast)---
T
Ummm...I totally don't understand the format of that array at all!
static ESWNODE engstoptree[] = {
{'m',L,9,126},
{'d',L,4,71},
{'b',L,2,40},
{'a',F,0,14},
{'c',0,0,62},
{'f',L,2,79},
{'e',0,0,75},
{'h',0,1,90},
{'i',F,0,108
Christopher,
current implementation is ugly, we still didn't move functionality
from OpenFTS to tsearch. Look at makedict subdirectory to create your
custom dictionary. Default list is in engstoptree[] defined
in dic/porter_english.dct
On Mon, 2 Sep 2002, Christopher Kings-Lynne wrote:
> How do
We'll be back to work next week and look into code.
Oleg
On Sun, 1 Sep 2002, Bruce Momjian wrote:
>
> Can someone address the intagg issue here, or is the code OK?
>
> ---
>
> Tom Lane wrote:
> > Joe Conway and I hav
Yeah, it seems to require Ant 1.5 now. This should probably go in the
configure script, ie a test for Ant 1.5 if java enabled?
Tom.
On Mon, 2002-09-02 at 16:30, Tatsuo Ishii wrote:
> I see:
>
> /usr/jakarta-ant-1.4.1/bin/ant -buildfile ./build.xml all \
> -Dmajor=7 -Dminor=3 -Dfullversion=7.3
I see:
/usr/jakarta-ant-1.4.1/bin/ant -buildfile ./build.xml all \
-Dmajor=7 -Dminor=3 -Dfullversion=7.3devel -Ddef_pgport=5432 -Denable_debug=yes
Buildfile: ./build.xml
BUILD FAILED
/usr/local/src/pgsql/current/pgsql/src/interfaces/jdbc/./build.xml:51: Class
org.apache.tools.ant.taskdefs.co
75 matches
Mail list logo
