Excerpts from Dimitri Fontaine's message of vie oct 15 16:15:23 -0300 2010:
> I updated the pg_execute_from_file branch with documentation for the
> function, and added documentation (catalog, functions, custom classes)
> to the main feature branch too.
Hmm. To be honest I don't like the directi
Excerpts from Alvaro Herrera's message of sáb oct 16 00:30:41 -0300 2010:
> Hmm, did you try "make install" in contrib? It fails for me in intagg:
>
> make[1]: Entering directory
> `/home/alvherre/Code/CVS/pgsql/build/HEAD/contrib/intagg'
> /bin/mkdir -p '/pgsql/install/HEAD/share/contrib'
> to
Excerpts from Dimitri Fontaine's message of vie oct 15 16:15:23 -0300 2010:
> Dimitri Fontaine writes:
> > So, you will find two new branches for those purposes at the repository,
> > named cfparser and pg_execute_from_file:
> >
> > http://git.postgresql.org/gitweb?p=postgresql-extension.git;a=s
On 15/10/2010 10:38 PM, Tom Lane wrote:
Yeah, actually what you need is DISCARD ALL when reassigning a
connection to another client. Anything less than that assumes the
clients are cooperating closely, ie they *want* the same prepared
statements etc.
For what it's worth, this is quite common
On Fri, Oct 15, 2010 at 8:45 PM, Robert Haas wrote:
> On Fri, Oct 15, 2010 at 8:10 PM, Tom Lane wrote:
>>> Perhaps we should think of pg_amop not so much
>>> as a way to tell the AM what to do, but just a way to tell it what
>>> operator is logically involved without relying on the name or OID.
>
--On 15. Oktober 2010 13:36:38 -0400 Tom Lane wrote:
Bernd Helmle writes:
Here is an updated version of the patch. It fixes the following issues
Andrew discovered during his review cycle:
I looked through this a bit. It's not ready to commit unfortunately.
Thanks for looking at this. I
On Fri, Oct 15, 2010 at 8:10 PM, Tom Lane wrote:
>> Perhaps we should think of pg_amop not so much
>> as a way to tell the AM what to do, but just a way to tell it what
>> operator is logically involved without relying on the name or OID.
>
> I already think of it that way ...
OK.
>>> Maybe we s
Robert Haas writes:
> On Fri, Oct 15, 2010 at 7:10 PM, Tom Lane wrote:
>> Well, we cannot avoid changing pg_amop, or at least changing its
>> interpretation, because the current scheme simply can't represent
>> indexable operators that are used for anything except simple boolean
>> WHERE tests.
Hitoshi Harada writes:
> 2010/10/5 Marko Tiikkaja :
>> This patch conflicted with Tom's WITH .. INSERT change. I tweaked the
>> patch just a bit and it now passes all regression tests so I can review
>> it. New version attached for documentation purposes.
> Thank you, I didn't notice that commi
On Fri, Oct 15, 2010 at 7:10 PM, Tom Lane wrote:
> Robert Haas writes:
>> I still feel vaguely uneasy about the fact that the proposed patch
>> can't handle ASC/DESC or NULLS FIRST/LAST, and that unease grew a bit
>> more last night when I read Peter's patch to add collation support.
>
> Good poi
Robert Haas writes:
> I still feel vaguely uneasy about the fact that the proposed patch
> can't handle ASC/DESC or NULLS FIRST/LAST, and that unease grew a bit
> more last night when I read Peter's patch to add collation support.
Good point.
> We could possibly cram ASC/DESC and NULLS FIRST/LAS
On Fri, Oct 15, 2010 at 5:42 PM, Tom Lane wrote:
> Robert Haas writes:
>> On Fri, Oct 15, 2010 at 10:54 AM, Tom Lane wrote:
>>> I think we should go into this with the idea that it might only do 10%
>>> of what sprintf can do initially, but there will be pressure to cover a
>>> lot of the other
On Fri, Oct 15, 2010 at 5:35 PM, Robert Haas wrote:
> On Fri, Oct 15, 2010 at 12:02 PM, Tom Lane wrote:
>> BTW, have we discussed the idea of embedding the role in the strategy
>> number? For example, require regular operators to have strategy
>> numbers 1-, while ordering operators have num
Robert Haas writes:
> On Fri, Oct 15, 2010 at 10:54 AM, Tom Lane wrote:
>> I think we should go into this with the idea that it might only do 10%
>> of what sprintf can do initially, but there will be pressure to cover a
>> lot of the other 90% eventually.
> Hmm. I have a feeling that's going t
On Fri, Oct 15, 2010 at 12:02 PM, Tom Lane wrote:
> BTW, have we discussed the idea of embedding the role in the strategy
> number? For example, require regular operators to have strategy
> numbers 1-, while ordering operators have numbers 1-1,
> leaving room for a couple more roles b
Andrew Dunstan writes:
> I have now implemented a policy of refusing results from snapshots older
> than 10 days. It's not quite what I said above, but I think it's sound
> policy.
+1.
Might be time to advise buildfarm owners to shut down the 7.4 and 8.0
branches, and remind them to start up 9
Stephen Frost writes:
> IOW, yeah, you're right, the role doesn't really matter. One thing that
> occurs to me when I last ran into a problem with this- it was painful to
> debug because the "permission denied" error didn't indicate which schema
> the table it was trying to access was in. I wond
On Fri, Oct 15, 2010 at 10:54 AM, Tom Lane wrote:
> Robert Haas writes:
>> On Fri, Oct 15, 2010 at 12:55 AM, Itagaki Takahiro
>> wrote:
>>> I agree that full-spec sprintf is too complex, but precision and
>>> zero-full for numeric types are commonly used. I think someone
>>> will ask us "Why don
On 09/27/2010 08:13 AM, Andrew Dunstan wrote:
In about a week I propose to have the buildfarm server start to reject
results based on snapshots where the newest file is older than the
time of the git migration a week ago. That means buildfarm owners need
to upgrade to using git. Many have a
* Tom Lane (t...@sss.pgh.pa.us) wrote:
> Stephen Frost writes:
> >> The whole problem with search_path and role is very frustrating. We've
> >> taken to just hacking things to be dynamic SQL whenever it's
> >> role-specific, but that's a really poor solution. I wonder if it would
> >> be possibl
Tom Lane writes:
> * Lots of bogus trailing whitespace. "git diff --check" can help you
> with that.
This is a repetitive common remark that I think sharing tips to avoid
that problem is a good idea. Here's an emacs setup to have trailing
whitespace hurt the eyes, and more-than-80 columns lines
Stephen Frost writes:
>> The whole problem with search_path and role is very frustrating. We've
>> taken to just hacking things to be dynamic SQL whenever it's
>> role-specific, but that's a really poor solution. I wonder if it would
>> be possible to have the function and prepare'd plan caches
Dimitri,
* Dimitri Fontaine (dimi...@2ndquadrant.fr) wrote:
[... much useful info ...]
Thanks for all of that, I certainly appreciate it.
> By default pgbouncer maintains a different pool per database and role,
> so you should be partly covered here.
No, not really.. We have a single unprivile
Stephen Frost writes:
> That assumption is certainly something I feel we should consider a valid
> and important use-case. I'd think a lot of the time your typical website
> is going to be using a dedicated pooler for connections and a dedicated
> database where having those queries cache'd would
I have addressed these issues in my final patch. One other change I did
was change the host name comparison to case insensitive (Apache and TCP
Wrappers do the same).
I'm planning to work on a wildcard mechanism soon. After that the issue
of which way the host resolution should work will have an
Hitoshi Harada writes:
> 2010/10/15 Tom Lane :
>> I think it'd be hard to describe without confusing people, because
>> while DISTINCT is a noise word there, it's definitely not a noise
>> word after SELECT.
> I thought adding DISTINCT next to ALL is enough like
> select_statement UNION [ ALL |
[Moving to -hackers]
On Fri, Oct 15, 2010 at 3:43 AM, Simon Riggs wrote:
> On Mon, 2010-10-11 at 09:41 -0400, Josh Kupershmidt wrote:
>> On Thu, Oct 7, 2010 at 7:43 PM, Josh Kupershmidt wrote:
>>
>> > I noticed that granting a user column-level update privileges doesn't
>> > allow that user to i
On Thu, Oct 14, 2010 at 7:35 PM, Tatsuo Ishii wrote:
>> What new public interfaces do you think are needed for 9.1 in this
>> regard?
>
> At this point I'm thinking of modifying existing pg_is_in_recovery(),
> thus 0 new public interface.
pg_is_in_recovery() returns a bool, are you proposing to c
Dimitri Fontaine writes:
> So, you will find two new branches for those purposes at the repository,
> named cfparser and pg_execute_from_file:
>
> http://git.postgresql.org/gitweb?p=postgresql-extension.git;a=summary
I updated the pg_execute_from_file branch with documentation for the
function,
On Fri, 2010-10-15 at 18:51 +0200, Stefan Kaltenbrunner wrote:
> >
> > When exactly don't the keepalives work?
>
> well tcp level keepalives are not terribly portable(or can only be
> partially controlledd from the app) and on some platforms have lower
> limits that are in the minutes which is t
Bernd Helmle writes:
> Here is an updated version of the patch. It fixes the following issues
> Andrew discovered during his review cycle:
I looked through this a bit. It's not ready to commit unfortunately.
The main gripe I've got is that you've paid very little attention to
updating comments
On 10/15/2010 05:43 PM, Simon Riggs wrote:
On Fri, 2010-10-15 at 21:41 +0900, Fujii Masao wrote:
As the result of the discussion, I think that we need the following two
parameters for the case where the standby goes down.
* replication_timeout
This is the maximum time to wait for the ACK
On Wed, 2010-10-13 at 22:29 +0200, Alastair Turner wrote:
> I am proposing altering psql to raise certain errors and exit before
> prompting for a password. These errors would have to be on items which
> didn't leak any information, my current list is:
> - Does the input file (-f) exist and is it
Hi.
Ahh, as for Japanese, it becomes the same.
C:\MinGW\home\HIROSHI>vcbuild /?
Microsoft(R) Visual C++ Project Builder - コマンド ライン バージョン 9.00.21022
Copyright (C) Microsoft Corporation. All rights reserved.
http://winpg.jp/~saito/pg_work/VisualStudio/vcbuild_help_SJIS.txt
Regards,
Hiroshi Saito
Tom Lane wrote:
> I'm all for doing this client-side.
Well, that makes a big difference. Unless someone can make a
convincing argument for why we should modify the server side to
support this, I think we should just focus on this one client-side
patch.
I'd be happy to give it a closer look
Teodor Sigaev writes:
> As I remember, there were several suggested designs:
> 1) 5-th boolean column (amopfamily, amoplefttype, amoprighttype,
> amopstrategy,
> amoporder) to point kind of operator (search or order)
>+ saves one record for operator in pg_amop
>- operator could not be us
forward. Tom and I laid out a technical design back in January and I
still think it's a good one, even though I know you think it's silly.
We may just have to agree to disagree on this point.
As I remember, there were several suggested designs:
1) 5-th boolean column (amopfamily, amoplefttype,
On Fri, 2010-10-15 at 21:41 +0900, Fujii Masao wrote:
> As the result of the discussion, I think that we need the following two
> parameters for the case where the standby goes down.
> * replication_timeout
> This is the maximum time to wait for the ACK from the standby. If this
> timeout exp
On Thu, Oct 14, 2010 at 19:37, Greg Smith wrote:
> We got an interesting documentation document left by Christian Freund
> recently, in regards to
> http://www.postgresql.org/docs/9.0/interactive/install-windows-full.html ;
> it says:
>
> Regarding 16.1.3 - "perl mkvcbuild.pl"
>
> In case you use
Robert Haas writes:
> On Fri, Oct 15, 2010 at 12:55 AM, Itagaki Takahiro
> wrote:
>> I agree that full-spec sprintf is too complex, but precision and
>> zero-full for numeric types are commonly used. I think someone
>> will ask us "Why don't have numeric formats though we have %s?".
> I think so
* Tom Lane (t...@sss.pgh.pa.us) wrote:
> Stephen Frost writes:
> > errr, I don't believe RESET ALL touches cache'd plans and whatnot (which
> > is actually a problem I've run into in the past, because changing the
> > search_path also doesn't invalidate plans, and neither does set role, so
> > you
On 14/10/10 15:53, Alastair Turner wrote:
It isn't a TODO item, or related to any previous thread I could find.
It's certainly something I can see a use for. When I'm having a bad
typing day I get annoyed that I find I've made a mistake after I've
typed the password. To me this is a featu
On 15/10/10 15:08, Tomáš Pospíšil wrote:
Hi Robert,
I would like to contribute for community. Under licence used by PostgreSQL.
So is (or was) there anybody with the same interest as me?
Yes, I've been looking at improving the XML support overall, or at least
working to finish the implemen
Stephen Frost writes:
> * RadosÅaw Smogura (rsmog...@softperience.eu) wrote:
>> But benefits of pooling statements are much more greater then RESET ALL,
>> because you can take advance of precompiling prepared statements,
>> increasing performance; it is comparable to using connection pool instea
David Boreham writes:
> Just wondering what was the motivation to GPL this code ?
It was written at Red Hat and they have (or at least had at the time)
a company policy of using GPL for any code written in-house.
regards, tom lane
--
Sent via pgsql-hackers mailing list
Robert Haas writes:
> On Fri, Oct 15, 2010 at 4:06 AM, Dean Rasheed
> wrote:
>> If these were the only problems with this patch, given that they
>> required only trivial changes to initialise variables to NIL/false,
>> perhaps it was premature to mark it as "returned with feedback".
> Sure. Is
=?UTF-8?Q?Rados=C5=82aw_Smogura?= writes:
> On Fri, 15 Oct 2010 10:37:05 +0200, Magnus Hagander
> wrote:
>>> But... connection pooler will not send RESET ALL in some situations,
>> To me, that sounds like a bug in the connection pooler. It is only
>> safe under quite limited circumstances.
> It
* Radosław Smogura (rsmog...@softperience.eu) wrote:
> > To me, that sounds like a bug in the connection pooler. It is only
> > safe under quite limited circumstances.
>
> It's hard to say this is bug. The GF connection pooler is "general pooler"
> for all drivers, so it can't know anything about
Hi Robert,
I would like to contribute for community. Under licence used by PostgreSQL.
So is (or was) there anybody with the same interest as me?
> Původní zpráva
> Od: Robert Haas
> Předmět: Re: [HACKERS] XML schema validation
> Datum: 15.10.2010 05:02:56
> --
On 10/15/2010 04:33 AM, Dean Rasheed wrote:
I started looking at this last night, but ran out of time. I'll
continue this evening / over the weekend. Here are my comments so far:
Patch applies cleanly to current git master with no offsets.
Compiles cleanly with no warnings.
Regression tests pa
Robert Haas writes:
> On Fri, Oct 15, 2010 at 2:36 AM, Greg Stark wrote:
>> I don't particularly see a problem with having GPL'd contrib modules.
> I think that's a bad idea for all kinds of reasons.
Yeah. From my viewpoint as a downstream packager, it creates a mess.
We've spent a great amou
On 10/15/2010 02:36 AM, Greg Stark wrote:
On Thu, Oct 14, 2010 at 2:53 PM, Tom Lane wrote:
Bruce Momjian writes:
Should we consider moving pg_filedump into our /contrib?
Can't: it's GPL.
I don't particularly see a problem with having GPL'd contrib modules.
It would mean any users hoping
Howdy folks,
Was wondering if there are any docs on which contrib modules don't work with
pg_upgrade? I seem to remember discussion on this during the 9.0 cycle, but
couldn't find it in the mail archive, and don't see anything in the wiki. What
brings this up is I'm currently working on an 8.3
On 10/15/2010 7:24 AM, Robert Haas wrote:
I think that's a bad idea for all kinds of reasons. For one thing, it
seems that someone could easily end up copying some of that code into
some other place. It would be *nice* to have this available as part
of our regular distribution but I don't want
On Fri, Oct 15, 2010 at 12:55 AM, Itagaki Takahiro
wrote:
> On Fri, Oct 15, 2010 at 12:59 PM, Pavel Stehule
> wrote:
>> then maybe %ls or %is - like "literal string" or "ident string".
>
> Yeah, good idea!
-1 from me. What does this do except make it more long-winded?
>> I don't think so merg
On Fri, Oct 15, 2010 at 2:36 AM, Greg Stark wrote:
> On Thu, Oct 14, 2010 at 2:53 PM, Tom Lane wrote:
>> Bruce Momjian writes:
>>> Should we consider moving pg_filedump into our /contrib?
>>
>> Can't: it's GPL.
>>
>
> I don't particularly see a problem with having GPL'd contrib modules.
> It wou
The new patch applies cleanly to head, there are no compile errors and
all the make check tests pass (linux).
Thanks
Andrew
On Thu, Oct 14, 2010 at 4:35 PM, Bernd Helmle wrote:
>
>
> --On 14. Oktober 2010 16:28:51 -0300 Alvaro Herrera
> wrote:
>
>>> Looking in that function, there is a similar
On Fri, Oct 15, 2010 at 4:06 AM, Dean Rasheed wrote:
> If these were the only problems with this patch, given that they
> required only trivial changes to initialise variables to NIL/false,
> perhaps it was premature to mark it as "returned with feedback".
Sure. Is someone available to do a deta
KaiGai,
* KaiGai Kohei (kai...@kaigai.gr.jp) wrote:
> However, it requires the plugin modules need to know everything;
> such as what is visible/invisible. It seems to me too closely-
> coupled interface.
I agree with Robert on this one. We're not trying to design a wholly
independent security m
Hi,
As the result of the discussion, I think that we need the following two
parameters for the case where the standby goes down.
* replication_timeout
This is the maximum time to wait for the ACK from the standby. If this
timeout expires, the master closes the replication connection and
dis
On Fri, 15 Oct 2010, Marios Vodas wrote:
Recently I worked a lot with gist and read about knngist. I have spotted
weaknesses in docs. So I would be happy to help if you tell me how.
We have some documentation already, Teodor will post soon. Marios,
I'd happy to help in writing docs. Do you ha
On Fri, Oct 15, 2010 at 8:20 PM, Pavel Stehule wrote:
> And the reply is - we have function to_char. I don't see any reason
> why we have to have two independent formatting systems.
The formats for literal and identifier can be replaced to quote_nullable() and
quote_ident(), too. Features to writ
2010/10/15 Itagaki Takahiro :
> On Fri, Oct 15, 2010 at 12:59 PM, Pavel Stehule
> wrote:
>> then maybe %ls or %is - like "literal string" or "ident string".
>
> Yeah, good idea!
>
>> I don't think so merging sprintf and format can be good. Sprintf is
>> too complex - so long years users don't kno
On Thu, 2010-10-14 at 18:05 -0400, Tom Lane wrote:
> Brendan Jurd writes:
> > Agreed that there are no doc bugs. The reason I suggested a backpatch
> > is that I'm concerned that a lot of people are going to be approaching
> > the whole Standby topic for the first time with 9.0, so it would be
>
On Fri, 15 Oct 2010 10:37:05 +0200, Magnus Hagander
wrote:
>> But... connection pooler will not send RESET ALL in some situations,
>> because JDBC driver can have no notify about switching client. In EE
>> platforms (e. g. Glassfish), server sometimes creates it's own pool and
>> in
>> certain con
Alvaro Herrera writes:
> Maybe it would be worthwhile to split the parts that parse a file and
> execute from a file, and submit separately. It is obviously
> self-contained and serves a useful purpose on its own. It also forces
> you to think harder about renaming the parse function :-)
So, yo
On Fri, Oct 15, 2010 at 08:56, Radosław Smogura
wrote:
> On Tue, 12 Oct 2010 20:03:29 +0200, Magnus Hagander
> wrote:
>> On Tue, Oct 12, 2010 at 17:55, David Fetter wrote:
>>> On Tue, Oct 12, 2010 at 10:37:00AM -0500, Kevin Grittner wrote:
David Fetter wrote:
> Is there something inco
On Thu, Oct 14, 2010 at 2:53 PM, Tom Lane wrote:
> Bruce Momjian writes:
>> Should we consider moving pg_filedump into our /contrib?
>
> Can't: it's GPL.
>
I don't particularly see a problem with having GPL'd contrib modules.
It would mean any users hoping to redistribute the package couldn't
in
On 15 October 2010 08:32, Bernd Helmle wrote:
>
>
> --On 14. Oktober 2010 20:47:27 +0100 Dean Rasheed
> wrote:
>
>> OK, here it is.
>>
>> I've not cured this compiler warning (in fact I'm not sure what it's
>> complaining about, because the variable *is* used):
>>
>> tablecmds.c: In function 'ATE
--On 14. Oktober 2010 20:47:27 +0100 Dean Rasheed
wrote:
OK, here it is.
I've not cured this compiler warning (in fact I'm not sure what it's
complaining about, because the variable *is* used):
tablecmds.c: In function 'ATExecSetNotNull':
tablecmds.c:4747: warning: unused variable 'is_new
Recently I worked a lot with gist and read about knngist. I have spotted
weaknesses in docs. So I would be happy to help if you tell me how.
On Fri, Oct 15, 2010 at 5:57 AM, Robert Haas wrote:
> On Wed, Oct 13, 2010 at 7:07 PM, Marios Vodas wrote:
> > I would like to ask in which future version
71 matches
Mail list logo
