"Pavel Stehule" <[EMAIL PROTECTED]> writes:
> I need new function
> SPI_cursor_fetch_with_direction(Portal portal, int direction, long
count)
> where is possible set direction of fetch statement.
Huh? SPI_cursor_fetch already lets you specify forward or backward.
re
uwcssa <[EMAIL PROTECTED]> writes:
> The problem is that each join method has some value that seems to
> change value each time I cost the SAME path.
There's a bug in your code; but you've not provided nearly enough detail
to identify exactly what it is.
regards, tom lane
"Jochem van Dieten" <[EMAIL PROTECTED]> wrote:
> On 12/28/06, ITAGAKI Takahiro wrote:
> >
> > | [TODO item] Allow data to be pulled directly from indexes
> > | Another idea is to maintain a bitmap of heap pages where all rows are
> > | visible to all backends, and allow index lookups to reference
I wish to estimate the total cost for a given path of a query. This
path is a tree consists of nestloop, hashjoin, mergejoin as internal
nodes with seqscan or idxscan at the leaves.
My approach is to cost the path bottom-up, and invoke the
"cost_nestloop", "cost_merge_join" etc.
The problem is t
Your patch has been added to the PostgreSQL unapplied patches list at:
http://momjian.postgresql.org/cgi-bin/pgpatches
It will be applied as soon as one of the PostgreSQL committers reviews
and approves it.
---
Eul
Andrew Dunstan wrote:
> Should this be cleared up (maybe a nice first project for lurking new
> contributors)?\
>
Maybe.
> If we don't want long forms for some reason, then a comment in the code
> saying why would make sense.
>
I don't see a strong reason for not to do it. But if you look clo
Peter Eisentraut wrote:
> On the name of the option, it's not actually a "data" directory, so I'd just
> call it --xlogdir, parallel to --datadir.
>
Seems reasonable. Patch modified is attached.
--
Euler Taveira de Oliveira
http://www.timbira.com/
*** ./doc/src/sgml/ref/initdb.sgml.orig 2
On 1/2/07, Tom Lane <[EMAIL PROTECTED]> wrote:
Lukas Kahwe Smith <[EMAIL PROTECTED]> writes:
> Err, I think you misunderstood what I said. My implementation uses
> SAVEPOINTs already. The point is having some API where you do not have
> to care of you are already in a transaction or not.
It's no
On Jan 2, 2007, at 2:01 PM, Lukas Kahwe Smith wrote:
Tom Lane wrote:
Lukas Kahwe Smith <[EMAIL PROTECTED]> writes:
Err, I think you misunderstood what I said. My implementation
uses SAVEPOINTs already. The point is having some API where you
do not have to care of you are already in a transac
You just proved the case for why the units shouldn't be case sensitive:
On Dec 30, 2006, at 6:36 PM, Andrew Hammond wrote:
I agree. But perhaps the solution instead of failing is to throw a
warning to the effect of "Not to be pedantic, but you said mb and
millibits as a unit doesn't make sense i
On Dec 29, 2006, at 12:30 PM, Chris Browne wrote:
How you get the work to spread consistently across 6 hours is a
challenge; personally, my preference would generally be to try to get
the work done ASAP, so the goal seems a tad off to me...
Agreed. If we're going to monkey with automatically se
On Jan 1, 2007, at 2:24 AM, Gurjeet Singh wrote:
The comment, "This should be improved someday" sure sounds like a
TODO to me.
I don't know if it should make it to the TODO doc, as that lists
high-level/abstract feature-request-like items.
Given that the TODO list is the official compilati
Hello all,
a friend of mine ran into a problem installing PostgreSQL 8.0.9 on a
Windows XP Pro machine. Before anyone is asking: it has to be a 8.0.x
version and we even tried to install 8.2 and it did not work.
Ok, the problem is: after installing all the files the installer wants
to init the d
On Dec 21, 2006, at 9:56 AM, Simon Riggs wrote:
On Thu, 2006-12-21 at 09:36 -0500, Matthew O'Connor wrote:
Richard Huxton wrote:
Simon Riggs wrote:
- improve RI check perf by caching small, static tables in each
backend
- apply index filter conditions on index scan to avoid heap lookup
For
On 1/3/07, David Fetter <[EMAIL PROTECTED]> wrote:
As I recall, it was decided long ago, and the conclusions were:
* Only BSD-compatibly licensed code goes in PostgreSQL's code base,
* PostgreSQL will only support the SQL:2003 standard WITH (RECURSIVE)
syntax in the main line code.
Yes, see la
On Wed, Jan 03, 2007 at 04:19:19PM -0700, Jonah H. Harris wrote:
> On 1/3/07, Hubert FONGARNAND <[EMAIL PROTECTED]> wrote:
> >Why not looking at http://gppl.moonbone.ru/ evgen potemkin. has
> >ever made a patch for WITH and CONNECT BY?
>
> Nope, no good. This is what I started with last time and
On 1/3/07, Mark Cave-Ayland <[EMAIL PROTECTED]> wrote:
IIRC there were two issues - firstly the license for the patch was GPL
as opposed to BSD used for PostgreSQL
Yes, however Evgen was kind enough to grant me a BSD license for it
should I get it committed into PostgreSQL. However, with the a
On 1/3/07, Hubert FONGARNAND <[EMAIL PROTECTED]> wrote:
Why not looking at http://gppl.moonbone.ru/ evgen potemkin. has ever made a
patch for WITH and CONNECT BY?
Nope, no good. This is what I started with last time and the
refactoring attempt at WITH is just too massive. As for CONNECT BY,
P
"Pavel Stehule" <[EMAIL PROTECTED]> writes:
> I need new function
> SPI_cursor_fetch_with_direction(Portal portal, int direction, long count)
> where is possible set direction of fetch statement.
Huh? SPI_cursor_fetch already lets you specify forward or backward.
regar
i> >
> Is "int" a reasonable domain for directions? I'd think there would be
> at most values.
Er, "at most two."
enum is better, true
tree fields: FORWARD, BACKWARD, ABSOLUTE
Cheers,
D
>
> Cheers,
> D
> > where is possible set direction of fetch statement.
>
> >
> > any comments?
> >
>
In response to "Simon Riggs" <[EMAIL PROTECTED]>:
> On Tue, 2007-01-02 at 18:20 -0500, Tom Lane wrote:
> > Bill Moran <[EMAIL PROTECTED]> writes:
> > > In response to Alvaro Herrera <[EMAIL PROTECTED]>:
> > >> Please change things to save the stat() syscall when the feature is not
> > >> in use.
>
On Wed, 2007-01-03 at 19:44 +, alfranio correia junior wrote:
> I have some questions on execution plans.
>
> When are the following plans used ?
>
> nodeFunctionscan.c
When there is a Set Returning Function emitting tuples.
> nodeTidscan.c
When there is a query that accesses data usin
On Wed, Jan 03, 2007 at 12:36:27PM -0800, David Fetter wrote:
> On Wed, Jan 03, 2007 at 06:30:48PM +0100, Pavel Stehule wrote:
> > Hello,
> >
> > I am working on support scrollable cursors in plpgpsm. Scrollable cursors
> > are in ToDo for plpgsql too. I need new function
> > SPI_cursor_fetch_w
On Wed, Jan 03, 2007 at 06:30:48PM +0100, Pavel Stehule wrote:
> Hello,
>
> I am working on support scrollable cursors in plpgpsm. Scrollable cursors
> are in ToDo for plpgsql too. I need new function
> SPI_cursor_fetch_with_direction(Portal portal, int direction, long count)
Is "int" a reaso
Hi,
I have some questions on execution plans.
When are the following plans used ?
nodeFunctionscan.c nodeTidscan.c
In particular, is the plan nodeBitmapHeapscan.c always used along with
the nodeBitmapIndexscan.c ?
Please, let me know if there is a previous thread on this subject or any
d
> I am working on support scrollable cursors in plpgpsm. Scrollable
cursors
> are in ToDo for plpgsql too. I need new function
> SPI_cursor_fetch_with_direction(Portal portal, int direction, long
count)
> where is possible set direction of fetch statement.
Sounds good.
Please ensure the def
"Tom Lane" <[EMAIL PROTECTED]> writes:
> Gregory Stark <[EMAIL PROTECTED]> writes:
>> Merge Join (cost=10149.78..10448.70 rows=13161 width=36)
>>Merge Cond: (a.a = "inner"."?column2?")
>>-> Index Scan using aa on a (cost=0.00..62.45 rows=1230 width=32)
>>-> Sort (cost=100
Merge Joins require us to potentially Mark and Restore positions in the
tuples arriving from executor sub-nodes.
This currently means that if the tuples arrive from a Sort node, as they
often do in an MJ, the sort node will be instructed to prepare a random
access version of the sort result. That
Gregory Stark <[EMAIL PROTECTED]> writes:
> Merge Join (cost=10149.78..10448.70 rows=13161 width=36)
>Merge Cond: (a.a = "inner"."?column2?")
>-> Index Scan using aa on a (cost=0.00..62.45 rows=1230 width=32)
>-> Sort (cost=10149.78..10155.13 rows=2140 width=4)
>
On Wed, 2007-01-03 at 18:30 +0100, Pavel Stehule wrote:
> I am working on support scrollable cursors in plpgpsm. Scrollable cursors
> are in ToDo for plpgsql too. I need new function
> SPI_cursor_fetch_with_direction(Portal portal, int direction, long count)
> where is possible set direction o
Hope you had a nice holiday.
"Tom Lane" <[EMAIL PROTECTED]> writes:
> Gregory Stark <[EMAIL PROTECTED]> writes:
>> So the only reason we needed the cross-data-type operators was to get better
>> estimates? I thought without them you couldn't get an index-based plan at
>> all.
>
> Oh, hm, there
Hello,
I am working on support scrollable cursors in plpgpsm. Scrollable cursors
are in ToDo for plpgsql too. I need new function
SPI_cursor_fetch_with_direction(Portal portal, int direction, long count)
where is possible set direction of fetch statement.
any comments?
Regards
Pavel Stehu
Tom Lane wrote:
It should happen automatically at commit of the CREATE DATABASE ... and
you'd not be able to see the pg_database row before that anyway. So I'm
not clear on what you're worried about.
Okay, thanks. I'll have to investigate on why exactly I still get the
error, then. That's unc
Markus Schiltknecht <[EMAIL PROTECTED]> writes:
> Most probably I better go another way, than fiddling with postgres
> startup internals. Is there a way to be sure the flatfile has been
> written to disk after a CREATE DATABASE? I would like to ensure I can
> connect to a newly created database.
Hello Tom,
Tom Lane wrote:
In what PG version?
Postgres-R 8.3devel ;-)
Because the postmaster doesn't have direct database access. If it did,
any corruption of shared memory would risk crashing the postmaster
along with the backends.
Understood, thanks.
Most probably I better go another
Markus Schiltknecht <[EMAIL PROTECTED]> writes:
> I've just run into a race condition with creating a database and
> connecting to it immediately afterwards. I'm getting a "database %s not
> found" error just after the first flatfiles check in InitPostgres().
In what PG version?
> What that Fin
"Greg Sabino Mullane" <[EMAIL PROTECTED]> writes:
> So as a general rule, the system tables should be considered a special
> case as far as transactional activity? To be more precise, you are saying
> that a system table must be locked in access exclusive mode before any
> change is made to guarant
Hi,
I've just run into a race condition with creating a database and
connecting to it immediately afterwards. I'm getting a "database %s not
found" error just after the first flatfiles check in InitPostgres().
What that FindMyDatabase() there does, besides checking if the database
exists, is
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Tom Lane wrote:
> No, it's not a violation of ACID. In this case what you are doing is
> altering a table's schema without a sufficiently strong lock on the
> table, and that's a no-no, whether you would like it to be or not.
So as a general rule,
On Tue, 2007-01-02 at 18:20 -0500, Tom Lane wrote:
> Bill Moran <[EMAIL PROTECTED]> writes:
> > In response to Alvaro Herrera <[EMAIL PROTECTED]>:
> >> Please change things to save the stat() syscall when the feature is not
> >> in use.
>
> > Do you have a suggestion on how to do that and still ha
* Florian Weimer ([EMAIL PROTECTED]) wrote:
> * Stephen Frost:
> > Ah, this does sound rather ugly and not something we'd want. The
> > particular library doesn't make a whole heck of alot of difference to me
> > provided it has the general functionality necessary and a compatible
> > license (whe
* Stephen Frost:
> Ah, this does sound rather ugly and not something we'd want. The
> particular library doesn't make a whole heck of alot of difference to me
> provided it has the general functionality necessary and a compatible
> license (where 'compatible' in this case really means 'Debian fee
I notice that quite a few pg_ctl options have no long form equivalents,
namely these: NopPwW
Also, none of the long forms seems to be documented at all.
Should this be cleared up (maybe a nice first project for lurking new
contributors)?\
If we don't want long forms for some reason, then a
> > Yes, and I can't think of a single reason why we'd let people
specify
> > anything in millibytes, or kilobits.
>
> How about a configuration option related to connection throughput,
which is
> typically measured in bits?
We'd use "kbit". I don't see us using "kb" in that case (or was it kB
> > I believe there's something similar for OS X as well. The question
is:
> > would it be better to do that, or to just delay calling fsync until
the
> > OS has had a chance to write things out.
>
> A delay is not going to help unless you can suppress additional writes
> to the file, which I don
On Wed, 2007-01-03 at 09:45 +0100, Hubert FONGARNAND wrote:
> Why not looking at http://gppl.moonbone.ru/ evgen potemkin. has ever
> made a patch for WITH and CONNECT BY?
>
> I'm ready to test these features... (RECURSIVE) when they'll land in
> CVS...
Hi Hubert,
IIRC there were two issues - f
Le mardi 02 janvier 2007 à 18:08 -0700, Jonah H. Harris a écrit :
> On 12/30/06, Mark Cave-Ayland <[EMAIL PROTECTED]> wrote:
> > In short, if people don't mind waiting for my free cycles to come along
> > then I will continue to chip away at it; otherwise if it's considered an
> > essential for 8.
Am Mittwoch, 27. Dezember 2006 02:56 schrieb Euler Taveira de Oliveira:
> This simple patch lets someone specifies the xlog directory at initdb
> time. It uses symlinks to do it, and create and/or set permissions at
> the directory as appropriate.
On the name of the option, it's not actually a "da
48 matches
Mail list logo
