Hi,
Was there ever any discussion.thought about being able to follow a non-strict
by name
column mapping between inherited tables and father tables?
If for instance someone wants to build an hierarchy on a schema of tables being
defined
in an established production system, designed with no inh
On 10/2/2014 4:51 AM, Achilleas Mantzios wrote:
After all, generally speaking invoices have invoice_no's while general
docs have doc_no's , right?
So I think, the above scenario could be indeed be found a lot of times
in systems designed with no OO in mind.
database relations are not 'obje
Achilleas Mantzios writes:
> Was there ever any discussion.thought about being able to follow a non-strict
> by name
> column mapping between inherited tables and father tables?
No. You could use a view with UNION ALL perhaps.
(The subtext here is that past discussion of inheritance has genera
Testing that now. Initial results are not looking too performant.
I have one single table which had 234575 updates done to it. I am rolling
back 13093 of them. It's been running 20 min now, using 100% of a single
core, and almost 0 disk. No idea how long it'll run at this point.
This is on an
Ended up running for 28 min, but it did work as expected.
On Thu, Oct 2, 2014 at 10:27 AM, Adam Brusselback wrote:
> Testing that now. Initial results are not looking too performant.
> I have one single table which had 234575 updates done to it. I am rolling
> back 13093 of them. It's been ru
Hi,
I'm writing an extention (FDW), and I need, in my C code, the name of
my database ("contrib_regression_test" for example), and I've two
questions:
1) What is the easiest way to get that directly in C?
2) Is there a way to get this information in the SQL extension
installation/update scripts
Cedric Berger writes:
> I'm writing an extention (FDW), and I need, in my C code, the name of
> my database ("contrib_regression_test" for example), and I've two
> questions:
> 1) What is the easiest way to get that directly in C?
The usual locution is "get_database_name(MyDatabaseId)".
> 2) Is
Hi
Ho do I create a PL/pgSQL function that returns multiple out parameters and
a refcursor. Something like this:
create function myfunction(p_cursor out refcursor, p_code out varchar,
p_message out varchar) returns record as $$
begin
open p_cursor is select * from table;
p_code := 'OK'
Hi Tom,
First of all, thank you for your fast response!
I did not try to set up keepalives in client yet. This is a good point. We are
using JDBC to connect to PG from JBoss application server.
I have seen the JDBC documentation and keepalive can be enabled when creating
the connection to the
Does anyone have a good solution for benching queries under various conditions,
and collecting the EXPLAIN data ?
I looked at pgbench, but it doesn't seem to be what I want.
My situation is this-
- For a given query, there are 3-5 different ways that I can run it.
- Each form of the query ha
I’m looking for a way to increase select statement speed in Postgres 9.0.
Table has required index present. Desired result can obtained using index
(kuupaev,kellaaeg) immediately.
However Postgres scans all rows:
explain analyze SELECT
max( kuupaev||kellaaeg ) as res
from firma2.A
Steps to reproduce:
Run commands
create temp table test (kuupaev date, kellaaeg char(5) ) on commit drop;
create index test on test ((kuupaev||kellaaeg));
in
"PostgreSQL 9.3.4, compiled by Visual C++ build 1600, 32-bit"
Observed result:
ERROR: functions in index expres
Hi,
I'm writing an extention (FDW), and I need, in my C code, the name of
my database ("contrib_regression_test" for example), and I've two
questions:
1) What is the easiest way to get that directly in C?
2) Is there a way to get this information in the SQL extension
installation/update scripts
we've always installed on linux so need help with a new mac air running
latest osx
in the instructions it shows several methods:
1) enterprisedb (but this does not look open source ?)
2) fink
3) macports
4) source
etc
what do most use ? thks
ps: is there a mac build for 9.4 beta 3 yet ? we
On 10/2/2014 8:11 AM, Cedric Berger wrote:
I'm writing an extention (FDW), and I need, in my C code, the name of
my database ("contrib_regression_test" for example), and I've two
questions:
1) What is the easiest way to get that directly in C?
2) Is there a way to get this information in the SQ
On 14/10/03 7:50, john.tiger wrote:
> we've always installed on linux so need help with a new mac air running
> latest osx
>
> in the instructions it shows several methods:
> 1) enterprisedb (but this does not look open source ?)
>
> 2) fink
> 3) macports
> 4) source
> etc
>
> what do most use ?
The GUI installer for Mac OS X downloaded from postgresql.org works fine.
~ john
On Oct 2, 2014, at 3:50 PM, john.tiger wrote:
> we've always installed on linux so need help with a new mac air running
> latest osx
>
> in the instructions it shows several methods:
> 1) enterprisedb (b
On Oct 2, 2014, at 7:30 PM, john gale wrote:
> The GUI installer for Mac OS X downloaded from postgresql.org works fine.
Unless you NEED to use the source/etc version, use the GUI installer.
Unless you are already on a system where installing from Fink/Macports/Source
is commonplace... you'
On 10/02/2014 03:50 PM, john.tiger wrote:
we've always installed on linux so need help with a new mac air running
latest osx
in the instructions it shows several methods:
1) enterprisedb (but this does not look open source ?)
It is just the community version of Postgres behind a graphical
ins
On 10/02/2014 01:49 PM, Andrus wrote:
Steps to reproduce:
Run commands
create temp table test (kuupaev date, kellaaeg char(5) ) on commit
drop;
create index test on test ((kuupaev||kellaaeg));
in
"PostgreSQL 9.3.4, compiled by Visual C++ build 1600, 32-bit"
Observed result:
E
So kellaaeg is a time? Your best bet here would be to create an index that is
an actual timestamp comprised of both kuupaev and kellaaeg. You could do this
with to_timestamp by concatinating both fields together, or it may be easier to
replace the space in kellaaeg with a colon and cast it to t
On 10/2/14, 9:00 AM, Tom Lane wrote:
Achilleas Mantzios writes:
Was there ever any discussion.thought about being able to follow a non-strict
by name
column mapping between inherited tables and father tables?
No. You could use a view with UNION ALL perhaps.
FWIW, I've had some less than ste
Andrus Moor wrote
> Table has both indexes present but postgres will not use them.
> Indexes can changed and query can re-written if this helps.
> Table structure cannot changed. char columns cannot replaced with varchar
> columns. kuupaev must be date and kellaaeg must be char(5) type.
Don't di
On 10/2/2014 4:37 PM, Adrian Klaver wrote:
On 10/02/2014 03:50 PM, john.tiger wrote:
we've always installed on linux so need help with a new mac air running
latest osx
in the instructions it shows several methods:
1) enterprisedb (but this does not look open source ?)
It is just the community
Jim Nasby writes:
> So kellaaeg is a time? Your best bet here would be to create an index
> that is an actual timestamp comprised of both kuupaev and kellaaeg.
The real problem with this query, or at least with the index design,
is that the index design isn't accounting for the need to constrain
On Oct 2, 2014, at 8:04 PM, John R Pierce wrote:
> On 10/2/2014 4:37 PM, Adrian Klaver wrote:
>> On 10/02/2014 03:50 PM, john.tiger wrote:
>>> we've always installed on linux so need help with a new mac air running
>>> latest osx
>>>
>>> in the instructions it shows several methods:
>>> 1) ente
On Fri, Oct 3, 2014 at 1:15 AM, Néstor Boscán wrote:
> Ho do I create a PL/pgSQL function that returns multiple out parameters and
> a refcursor.
Using a plain "RETURN;" is just but fine, your output parameters are
being set internally in the function:
=# create function myfunction(id_base in int
Hi!
So kellaaeg is a time? Your best bet here would be to create an index that
is an actual timestamp comprised of both >kuupaev and kellaaeg. You could
do this with to_timestamp by concatinating both fields together, or it may
be easier to replace the space in kellaaeg with a colon and cast i
28 matches
Mail list logo
