On Wed, Jul 13, 2011 at 2:53 AM, Peter Eisentraut wrote:
> On tis, 2011-07-12 at 08:51 -0400, Tom Lane wrote:
> > Peter Eisentraut writes:
> > > It has occurred to me a few times that it could be useful to clarify
> the
> > > approach here. If we could somehow have a separable cleanup step for
On Tue, Jul 12, 2011 at 05:59:01PM -0400, Alvaro Herrera wrote:
> Excerpts from Noah Misch's message of vie mar 11 12:51:14 -0300 2011:
> > On Fri, Feb 11, 2011 at 02:13:22AM -0500, Noah Misch wrote:
> > > Automated tests would go a long way toward building confidence that this
> > > patch
> > > d
Excerpts from Alvaro Herrera's message of mié jul 13 01:11:41 -0400 2011:
> Blind attempt at fixing isolation_tester on Win32
If this doesn't work, I'm afraid I'll have to ask some Windows person
for help :-)
--
Álvaro Herrera
The PostgreSQL Company - Command Prompt, Inc.
PostgreSQL Replication
On Jul 12, 2011, at 5:06 PM, Josh Berkus wrote:
>> I'm okay with this, though given the fact that ftp.ossp.org has been down
>> for *months*, I'm inclined to think that we ought to include it in the
>> contrib distribution for easy linking.
>
> What license is it under?
COPYRIGHT AND LICENSE
On Jul 12, 2011, at 5:07 PM, Tom Lane wrote:
> Curious considering that the machine is there (responds to ping), and
> the ossp.org webserver works fine. Has anyone bugged the owner about
> that?
I've sent him email and Twitter DMs, to no avail.
Best,
David
--
Sent via pgsql-hackers mailing
On Jul13, 2011, at 00:10 , Robert Haas wrote:
> On Jul 12, 2011, at 8:03 AM, Florian Pflug wrote:
>> The algorithm is quite straight forward, if one assumes a lock-free
>> implementation of a queue (More on that below)
>
> This is similar to the CAS-based LWLocks I played around with, though
> I
There's a thread over in pgsql-performance
http://archives.postgresql.org/pgsql-performance/2011-07/msg00046.php
in which the main conclusion was that we need to take a fresh look at the
heuristics the planner uses when dealing with small or empty relations.
The code in question is in estimate_rel_
On Tuesday, July 12, 2011 08:57:44 PM Dean Rasheed wrote:
> On 12 July 2011 19:26, Josh Berkus wrote:
> > On 7/12/11 9:46 AM, Andres Freund wrote:
> >> Hi,
> >>
> >> I guess $subject wasn't implemented because plain unique indexes aren't
> >> represented in pg_constraint and thus do not have a pl
"David E. Wheeler" writes:
> On Jul 12, 2011, at 1:40 PM, k...@rice.edu wrote:
>> That is why I think having the UUID generators be a contrib module
>> is the correct place for them to be, but the UUID type is better as
>> a core function.
> I'm okay with this, though given the fact that ftp.ossp
David,
> I'm okay with this, though given the fact that ftp.ossp.org has been down for
> *months*, I'm inclined to think that we ought to include it in the contrib
> distribution for easy linking.
What license is it under?
--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com
--
Sent
On Jul 12, 2011, at 1:40 PM, k...@rice.edu wrote:
> That is why I think having the UUID generators be a contrib module
> is the correct place for them to be, but the UUID type is better as
> a core function.
I'm okay with this, though given the fact that ftp.ossp.org has been down for
*months*,
On 07/12/2011 01:29 PM, Andrew Dunstan wrote:
On 07/12/2011 03:44 PM, Joshua D. Drake wrote:
What about extensions makes them less usable?
It is an extra step, that is less usable. Does it matter? Shrug, I
know I hate having to type apt-get just to use xyz, does it mean it is
a big deal? P
Peter Eisentraut writes:
> When you create a column with a plain "interval" column, the typmod is
> set to -1 and the information schema reports this as 6, because that's
> what the internal default value is (see _pg_datetime_precision
> function). But when you create a column such as "interval y
When you create a column with a plain "interval" column, the typmod is
set to -1 and the information schema reports this as 6, because that's
what the internal default value is (see _pg_datetime_precision
function). But when you create a column such as "interval year to
month"), the typmod is actu
On Jul 12, 2011, at 8:03 AM, Florian Pflug wrote:
> The algorithm is quite straight forward, if one assumes a lock-free
> implementation of a queue (More on that below)
This is similar to the CAS-based LWLocks I played around with, though I didn't
use a lock-free queue. I think I probably need
On Jul 12, 2011, at 12:19 PM, Alex Hunsaker wrote:
> All Arrays in 9.0 and lower are strings, regardless of if they are
> comprised of composite types. Its not so much a bug as a limitation.
> Alexey Klyukin fixed this for 9.1 :-)
Oh?
dump
--
Excerpts from Noah Misch's message of vie mar 11 12:51:14 -0300 2011:
> On Fri, Feb 11, 2011 at 02:13:22AM -0500, Noah Misch wrote:
> > Automated tests would go a long way toward building confidence that this
> > patch
> > does the right thing. Thanks to the SSI patch, we now have an in-tree test
(2011/07/12 22:56), Alvaro Herrera wrote:
> Speaking of which -- what's the difference between ADD and SET for SQL/MED
> options?
ADD can only add new option; it can't overwrite existing option's value.
To overwrite existing option's value, you need to use SET instead.
Regards,
--
Shigeru Hanad
On 07/12/2011 05:42 PM, Alvaro Herrera wrote:
Hi,
I think we're not running the isolation test suite. I noticed that the
stage seemed to take too little time (zero seconds in fact), and looking
into the stage logs only says
'make check' is not supported.
Install PostgreSQL, th
Hi,
I think we're not running the isolation test suite. I noticed that the
stage seemed to take too little time (zero seconds in fact), and looking
into the stage logs only says
'make check' is not supported.
Install PostgreSQL, then 'make installcheck' instead.
http://buildfarm.
On tis, 2011-07-12 at 09:56 -0400, Alvaro Herrera wrote:
> Speaking of which -- what's the difference between ADD and SET for
> SQL/MED options?
ADD add to the existing options, SET overwrites all options with what
you specify.
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.o
On tis, 2011-07-12 at 08:51 -0400, Tom Lane wrote:
> Peter Eisentraut writes:
> > It has occurred to me a few times that it could be useful to clarify the
> > approach here. If we could somehow have a separable cleanup step for
> > every test, and eliminate interdependencies between tests, we cou
Hi everybody:
Does anybody has an example to send an email in html format using pgmail.
Best Regards,
Fernando Acosta
Hi All,
As per discussion here
http://archives.postgresql.org/pgsql-hackers/2011-05/msg01119.php
PFA a patch which implements the idea with some variation.
At the start of the first pass, we remember the current LSN. Every page that
needs some work is HOT-pruned so that dead tuples are truncated
On Tue, Jul 12, 2011 at 04:29:33PM -0400, Andrew Dunstan wrote:
>
>
> On 07/12/2011 03:44 PM, Joshua D. Drake wrote:
> >>
> >>What about extensions makes them less usable?
> >
> >
> >It is an extra step, that is less usable. Does it matter? Shrug, I
> >know I hate having to type apt-get just to u
On 07/12/2011 03:44 PM, Joshua D. Drake wrote:
What about extensions makes them less usable?
It is an extra step, that is less usable. Does it matter? Shrug, I
know I hate having to type apt-get just to use xyz, does it mean it is
a big deal? Probably not.
By that argument we wouldn't
On 07/12/2011 09:15 AM, Andrew Dunstan wrote:
On 07/12/2011 12:03 PM, Joshua D. Drake wrote:
On 07/03/2011 11:54 AM, Peter Eisentraut wrote:
On sön, 2011-07-03 at 13:42 -0500, Michael Gould wrote:
I would like to request that full support for the UUID data type can
added.
I think that even t
On Tue, 2011-07-12 at 13:32 -0500, Robert Haas wrote:
> On Jul 12, 2011, at 12:02 PM, Jeff Davis wrote:
> > Yeah, I think you're right here. It's probably not much of a practical
> > concern.
> >
> > I was slightly bothered because it seemed a little unpredictable. But it
> > seems very minor, an
On Tue, Jul 12, 2011 at 12:45, David E. Wheeler wrote:
> Hackers,
> That is, if a record is passed to a PL/Perl function, it's correctly
> converted into a hash. If, however, an array of records are passed, the
> record are stringified, rather than turned into hashes. This seems
> inconsistent
On 07/12/2011 11:56 AM, Josh Berkus wrote:
Thom,
The functions to produce UUIDs are in contrib, but the UUID data type
itself is in core. You get the type uuid whether you install the
contrib module or not.
http://www.postgresql.org/docs/current/static/datatype-uuid.html
Oh!
I guess that s
On 12 July 2011 19:26, Josh Berkus wrote:
> On 7/12/11 9:46 AM, Andres Freund wrote:
>> Hi,
>>
>> I guess $subject wasn't implemented because plain unique indexes aren't
>> represented in pg_constraint and thus do not have a place to store
>> information
>> about being deferred?
>> Other than tha
Thom,
> The functions to produce UUIDs are in contrib, but the UUID data type
> itself is in core. You get the type uuid whether you install the
> contrib module or not.
>
> http://www.postgresql.org/docs/current/static/datatype-uuid.html
Oh!
I guess that shows you how much I use the type then
Hackers,
Given this script:
BEGIN;
CREATE TYPE foo AS ( this int, that int );
CREATE OR REPLACE FUNCTION dump(foo[]) returns text language plperlu AS $$
use Data::Dumper; Dumper shift;
$$;
CREATE OR REPLACE FUNCTION dump(foo) returns text language plperlu AS $$
On Jul 12, 2011, at 1:24 PM, Josh Berkus wrote:
> Magnus, JD,
>
>> UUID *is* in core. It's just the generation functions that aren't.
>
> No, it's not. It's in /contrib, which makes it an extension.
>
>> Uh UUID/GUID is used pervasively throughout enterprise apps,
>> especially Java apps.
On Jul 12, 2011, at 12:02 PM, Jeff Davis wrote:
> Yeah, I think you're right here. It's probably not much of a practical
> concern.
>
> I was slightly bothered because it seemed a little unpredictable. But it
> seems very minor, and if we wanted to fix it later I think we could.
Yes, I agree. I
On 12 July 2011 19:24, Josh Berkus wrote:
> Magnus, JD,
>
>> UUID *is* in core. It's just the generation functions that aren't.
>
> No, it's not. It's in /contrib, which makes it an extension.
The functions to produce UUIDs are in contrib, but the UUID data type
itself is in core. You get the t
On 7/12/11 9:46 AM, Andres Freund wrote:
> Hi,
>
> I guess $subject wasn't implemented because plain unique indexes aren't
> represented in pg_constraint and thus do not have a place to store
> information
> about being deferred?
> Other than that I do not see any special complications in imple
Magnus, JD,
> UUID *is* in core. It's just the generation functions that aren't.
No, it's not. It's in /contrib, which makes it an extension.
> Uh UUID/GUID is used pervasively throughout enterprise apps,
> especially Java apps.
Oh, I guess I encounter it a lot less than you. Time for a s
On 7/12/11 7:38 AM, Simon Riggs wrote:
> On Sun, Jul 10, 2011 at 7:13 PM, Josh Berkus wrote:
>
>> This patch[1] is for some reason marked "waiting on Author". But I
>> can't find that there's been any review of it searching the list.
>> What's going on with it? Has it been reviewed?
>
> Yes, I
Florian, Radoslaw,
Please both of you calm down. Florian is trying to improve our XML
type. Radoslaw is trying to help out by reviewing it. It's not a
benefit to anyone for you two to get into an argument about who said
what ... especially if the argument is based on (as far as I can see)
not u
Radoslaw,
> For me this discussion is over. I putted my objections and suggestions. Full
>> review is available in archives, and why to not escape is putted in review
>> of your 2nd patch, about scalar values.
Did you install and test the functionality of the patch? I can't tell
from your review
On Tue, 2011-07-12 at 07:55 -0500, Robert Haas wrote:
> I haven't been that worried about overflow of the fast path table. If
> you are locking more than 16 relations at once, you probably have at
> least 5 tables in the query, maybe more - it depends in how many
> indexes you have, of course. My
Hi,
I guess $subject wasn't implemented because plain unique indexes aren't
represented in pg_constraint and thus do not have a place to store information
about being deferred?
Other than that I do not see any special complications in implementing it?
Is there any reasons not to store unique in
On 07/12/2011 06:54 AM, Alvaro Herrera wrote:
Excerpts from Magnus Hagander's message of mar jul 12 09:34:56 -0400 2011:
Agreed. On one level I like the sponsor message, but on the other
having "Sponsored by RedHat" on every Tom Lane item will get tiring.
;-)
Create a macro ;)
Can we add
On 07/12/2011 12:03 PM, Joshua D. Drake wrote:
On 07/03/2011 11:54 AM, Peter Eisentraut wrote:
On sön, 2011-07-03 at 13:42 -0500, Michael Gould wrote:
I would like to request that full support for the UUID data type can
added.
I think that even though there is a contrib module, since this is
On 07/03/2011 11:54 AM, Peter Eisentraut wrote:
On sön, 2011-07-03 at 13:42 -0500, Michael Gould wrote:
I would like to request that full support for the UUID data type can added.
I think that even though there is a contrib module, since this is a standard
datatype that Postgres ought to be the
Peter Eisentraut wrote:
> pg_upgrade's pg_scandir_internal() makes use of the non-standard %m
> format:
>
> pg_log(PG_FATAL, "could not open directory \"%s\": %m\n", dirname);
>
> Is this an oversight, or is there an undocumented assumption that this
> code will only be used on platforms wher
On Mon, Jul 11, 2011 at 16:58, Peter Eisentraut wrote:
> These files are last updated 2001 or 2002 and I'm pretty sure they are
> outdated. It looks like no one is maintaining them, so we should remove
> them.
+1.
--
Magnus Hagander
Me: http://www.hagander.net/
Work: http://www.redpill-lin
Alvaro Herrera wrote:
> Excerpts from Bruce Momjian's message of lun jul 11 18:58:35 -0400 2011:
> > I have updated the TODO wiki to remove the 9.1-completed items:
> >
> > http://wiki.postgresql.org/wiki/Todo
> >
> > This will allow us to now mark 9.2-completed items.
>
> I have created Tod
On Sun, Jul 10, 2011 at 7:13 PM, Josh Berkus wrote:
> This patch[1] is for some reason marked "waiting on Author". But I
> can't find that there's been any review of it searching the list.
> What's going on with it? Has it been reviewed?
Yes, I reviewed it on list. Some minor changes were disc
On Sun, Jul 10, 2011 at 20:46, Brar Piening wrote:
> Sorry for the late response - I've been on a wedding this weekend.
>>
>> Something is strange here. Did you run perltidy with the exact
>> parameters documented in the README file?
>
> Yes - I usually even copy paste it from the README as "perlt
Excerpts from Bruce Momjian's message of lun jul 11 18:58:35 -0400 2011:
> I have updated the TODO wiki to remove the 9.1-completed items:
>
> http://wiki.postgresql.org/wiki/Todo
>
> This will allow us to now mark 9.2-completed items.
I have created TodoDone91 from the items marked TodoDone
On Tue, Jul 12, 2011 at 01:10, Andrew Dunstan wrote:
>
>
> On 07/11/2011 07:59 PM, Bruce Momjian wrote:
>>
>> Andrew Dunstan wrote:
>>>
>>> On 06/28/2011 05:31 PM, Peter Eisentraut wrote:
On tis, 2011-06-28 at 17:05 -0400, Andrew Dunstan wrote:
>>
>> Couldn't you just put a text
Alvaro Herrera wrote:
> Excerpts from Magnus Hagander's message of mar jul 12 09:34:56 -0400 2011:
>
> > > Agreed. ??On one level I like the sponsor message, but on the other
> > > having "Sponsored by RedHat" on every Tom Lane item will get tiring.
> > > ;-)
> > >
> > > Can we add text if the emp
Excerpts from Shigeru Hanada's message of mar jul 12 03:11:54 -0400 2011:
> (2011/07/12 0:44), Peter Eisentraut wrote:
> > On lör, 2011-07-09 at 23:49 -0400, Alvaro Herrera wrote:
> >> The new ALTER TABLE grammar seems a bit strange -- ADD, SET, DROP. Is
> >> this defined by the SQL/MED standard?
Excerpts from Magnus Hagander's message of mar jul 12 09:34:56 -0400 2011:
> > Agreed. On one level I like the sponsor message, but on the other
> > having "Sponsored by RedHat" on every Tom Lane item will get tiring.
> > ;-)
> >
> > Can we add text if the employer is _not_ the feature sponsor?
>
On Tue, Jul 12, 2011 at 02:34, Bruce Momjian wrote:
> Simon Riggs wrote:
>> On Sun, Jul 3, 2011 at 7:51 PM, Peter Eisentraut wrote:
>> > On tor, 2011-06-30 at 15:09 -0400, Alvaro Herrera wrote:
>> >> Robert Hass (whose name I misspelled in the commit message above) just
>> >> mentioned to me (in
On Jul7, 2011, at 03:35 , Robert Haas wrote:
> Some poking around suggests that the problem isn't that
> spinlocks are routinely contended - it seems that we nearly always get
> the spinlock right off the bat. I'm wondering if the problem may be
> not so much that we have continuous spinlock conte
On Jul 11, 2011, at 11:45 AM, Jeff Davis wrote:
> * ... It's also possible that
> * we're acquiring a second or third lock type on a relation we have
> * already locked using the fast-path, but for now we don't worry about
> * that case either.
> */
>
> How common is that case? There are only 16
Peter Eisentraut writes:
> It has occurred to me a few times that it could be useful to clarify the
> approach here. If we could somehow have a separable cleanup step for
> every test, and eliminate interdependencies between tests, we could more
> easily support a number of uses cases such as cre
On Jul 11, 2011, at 8:34 PM, Bruce Momjian wrote:
> Can we add text if the employer is _not_ the feature sponsor?
I don't see that as much better. Commit messages should not be ads, IMHO.
There are plenty of ways to give credit without polluting the commit log with
it.
...Robert
--
Sent via
On Jul12, 2011, at 12:57 , Radosław Smogura wrote:
> On Tue, 12 Jul 2011 11:45:59 +0200, Florian Pflug wrote:
>> On Jul12, 2011, at 11:00 , Radosław Smogura wrote:
>>> On Sun, 10 Jul 2011 17:06:22 -0500, Robert Haas wrote:
Unless I am missing something, Florian is clearly correct here.
>>> Fo
(2011/07/12 21:19), Robert Haas wrote:
> On Jul 12, 2011, at 12:31 AM, Shigeru Hanada wrote:
>> I'm afraid that I've misunderstood the discussion. Do you mean that
>> per-table options should be stored in reloptions, but per-column should
>> be separated from attoptions? (I think I've misread...
On Jul 12, 2011, at 12:31 AM, Shigeru Hanada wrote:
> (2011/07/11 10:21), Robert Haas wrote:
>> On Jul 9, 2011, at 10:49 PM, Alvaro Herrera
>> wrote:
>>> In short: in my opinion, attoptions and attfdwoptions need to be one
>>> thing and the same.
>>
>> I feel the opposite. In particular, what h
On Jul 12, 2011, at 4:46 AM, Peter Eisentraut wrote:
> On fre, 2011-07-08 at 22:27 -0400, Robert Haas wrote:
>> On Fri, Jul 8, 2011 at 1:45 AM, Ashutosh Bapat
>> wrote:
>>> I think, tab1 and tab2 are too common names, for anyone to pick up for the
>>> tables. Also, the test alter_table.sql is dro
On Tue, 12 Jul 2011 11:45:59 +0200, Florian Pflug wrote:
On Jul12, 2011, at 11:00 , Radosław Smogura wrote:
On Sun, 10 Jul 2011 17:06:22 -0500, Robert Haas wrote:
Unless I am missing something, Florian is clearly correct here.
For me not, because this should be fixed internally by making xml
On 12 Jul 2011, at 3:25, Chris Travers wrote:
>> Right now I can emulate a hierarchical schema structure via a naming scheme
>> - for example "schemabase_sub1_sub2_etc". I am simply looking for a formal
>> way to do the above AND also tell the system that I want all schemas under
>> "schemabase"
On ons, 2011-07-06 at 12:40 -0400, Robert Haas wrote:
> I think perhaps we should create a
> big static array, each row of which would contain:
>
> - ObjectType
> - system cache ID for OID lookups
> - system catalog table OID for scans
> - attribute number for the name attribute, where applicable
On fre, 2011-07-08 at 22:27 -0400, Robert Haas wrote:
> On Fri, Jul 8, 2011 at 1:45 AM, Ashutosh Bapat
> wrote:
> > I think, tab1 and tab2 are too common names, for anyone to pick up for the
> > tables. Also, the test alter_table.sql is dropping many other tables (even
> > those which have undergo
On Jul12, 2011, at 11:00 , Radosław Smogura wrote:
> On Sun, 10 Jul 2011 17:06:22 -0500, Robert Haas wrote:
>> Unless I am missing something, Florian is clearly correct here.
> For me not, because this should be fixed internally by making xml type sefe
Huh??. Making the xml type safe is *exactly*
On Sun, 10 Jul 2011 17:06:22 -0500, Robert Haas wrote:
On Jul 10, 2011, at 1:40 PM, Josh Berkus wrote:
Hackers,
B. 6. Current behaviour _is intended_ (there is "if" to check
node type) and _"natural"_. In this particular case user ask for
text content of some node, and this content is actu
New version of patch with a little more refactoring and comments.
--
With best regards,
Alexander Korotkov.
gist_fast_build-0.6.0.patch.gz
Description: GNU Zip compressed data
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http:/
> This version of the patch adds a field into pg_controldata that tries to
> store the source of the base backup while in recovery mode.
> I think your ultimate goal with this patch is to be able to take a
> backup of a running hot-standby slave and recover it as another
> instance. This patch see
This is an update of a patch for NEXT CommitFest 2011/09.
Please ignore this message.
1 Additional Feature - EUC-JP incrementer
2 Bug fixes - bytea incrementer, libpq compilation.
--
Kyotaro Horiguchi
NTT Open Source Software Center
diff --git a/src/backend/utils/adt/selfuncs.c b/src/backend/ut
(2011/07/12 0:44), Peter Eisentraut wrote:
On lör, 2011-07-09 at 23:49 -0400, Alvaro Herrera wrote:
The new ALTER TABLE grammar seems a bit strange -- ADD, SET, DROP. Is
this defined by the SQL/MED standard? It seems at odds with our
handling of attoptions
Well, I believe the SQL/MED options
On Fri, Jul 8, 2011 at 6:18 PM, Tom Lane wrote:
> For test purposes, you could turn off synchronize_seqscans to prevent
> that.
Thanks, it helps. I'm rerunning tests now.
--
With best regards,
Alexander Korotkov.
On Jul11, 2011, at 21:49 , David Johnston wrote:
> Right now I can emulate a hierarchical schema structure via a naming scheme
> - for example "schemabase_sub1_sub2_etc". I am simply looking for a formal
> way to do the above AND also tell the system that I want all schemas under
> "schemabase" t
77 matches
Mail list logo
