Robert Haas writes:
>> 0003) Removes -fwrapv. I'm *NOT* suggesting we apply this right now, but
>> it seems like an important test for the new facilities. Without
>> 0002, tests would fail after this, after it all tests run
>> successfully.
>
> I suggest that if we think we don'
Hi Hackers,
The comment for get_collation_name() seems to have been copy-pasted from
get_constraint_name(), but missed one s/constraint/collation/.
Patch attached.
--
"I use RMS as a guide in the same way that a boat captain would use
a lighthouse. It's good to know where it is, but you gener
Peter Eisentraut writes:
> On 9/20/17 13:13, Dagfinn Ilmari Mannsåker wrote:
>> I have no opinion on the bulk of this patch set, but skimming it out of
>> curiosity I noticed that the plperl change seems to have lost the
>> dependency on plperl_helpers.h from the xsub
Hi Peter,
Peter Eisentraut writes:
> OK, I was not aware that people are using it that way. So updated patch
> set there, which separates coverage and coverage-html into two
> independent targets.
I have no opinion on the bulk of this patch set, but skimming it out of
curiosity I noticed that
Craig Ringer writes:
> On 20 September 2017 at 06:36, David Steele wrote:
>
>>
>> I just use:
>>
>> $SIG{__DIE__} = sub {Carp::confess @_};
That is the basic idea behind both Carp::Always and Devel::Confess, but
they also avoid breaking non-string exceptions (references and objects)
being throw
Tom Lane writes:
> Andrew Dunstan writes:
>> On 09/19/2017 01:31 PM, Andres Freund wrote:
>>> # Include module showing backtraces upon failures. As it's a
>>> non-standard module, don't fail if not installed.
>>> eval { use Carp::Always; }
>
>> Or maybe Devel::Confess ?
>
> Neither one seems to
Andrew Dunstan writes:
> On 09/19/2017 01:31 PM, Andres Freund wrote:
>> Hi,
>>
>> I've had a couple cases where tap tests died, and I couldn't easily see
>> where / why. For development of a new test I found it useful to show
>> backtraces in that case - just adding a
>> use Carp::Always;
>> at
Pavel Stehule writes:
> Hi
>
> I wrote a special pager for psql. Surely, this pager is not good for paging
> of man pages. So is not good to set it as global pager. We can introduce
> new env variable PSQL_PAGER for this purpose. It can work similar like
> PSQL_EDITOR variable.
>
> Notes, comment
Peter Eisentraut writes:
> Committed, adjusting for some of the suggestions.
Looks like one occurrence was still left in:
$ ag --no-group --ignore po --ignore release-\* --ignore wal.sgml
'\btransaction\s+log\b'
doc/src/sgml/func.sgml:18631:end of the transaction log at any instant,
while
Heikki Linnakangas writes:
> On 05/06/2017 01:56 PM, Gavin Flower wrote:
>> On 06/05/17 22:44, Vik Fearing wrote:
>>> On 05/05/2017 02:42 PM, Michael Paquier wrote:
+This option is obsolete but still accepted for backwards
+compatibility.
Isn't that incorrect Englis
Bruce Momjian writes:
> I have committed the first draft of the Postgres 10 release notes. They
> are current as of two days ago, and I will keep them current. Please
> give me any feedback you have.
I noticed a few niggles with the links in "my" entires, patch attached.
Cheers,
ilmari
--
Michael Paquier writes:
> On Fri, Apr 21, 2017 at 10:02 PM, Simon Riggs wrote:
>> On 21 April 2017 at 10:20, Heikki Linnakangas wrote:
>>> But looking more closely, I think I misunderstood RFC 5803. It *does* in
>>> fact specify a single string format to store the verifier in. And the format
>>
Tom Lane writes:
> Andres Freund writes:
>> On 2017-04-17 17:49:54 +0100, Dagfinn Ilmari Mannsåker wrote:
>>> I threw Devel::NYTProf at it and picked some more low-hanging fruit.
>
>> I'm a bit doubtful about improving the performance of genbki at the cost
>&
Tom Lane writes:
> There's certainly lots more that could be done in the genbki code,
> but I think all we can justify at this stage of the development
> cycle is to get the low-hanging fruit for testing speedups.
I threw Devel::NYTProf at it and picked some more low-hanging fruit.
Attached are
Tom Lane writes:
> I wonder if we shouldn't just do
>
> RangeTblEntry *rte PG_USED_FOR_ASSERTS_ONLY;
> ListCell *lc;
>
> /* Should only be applied to base relations that are subqueries */
> Assert(rel->relid > 0);
> -#ifdef USE_ASSERT_CHECKING
> rte = planner_rt_
Kevin Grittner writes:
> Unfortunately, I was unable to get the follow-on patch to allow
> setting by relation into a shape I liked. Let's see what we can do
> for the next release.
Okay, I'll try and crete a more comprehensive version of it for the next
commitfest.
> The first patch was appli
Peter Eisentraut writes:
> On 3/28/17 23:42, Michael Paquier wrote:
>> src/bin/pg_dump and src/test/modules/test_pgdump generate too much
>> output. If we could get tests to only print the final result, like how
>> many tests done and how many have passed, things would be much
>> friendlier.
>
>
Andrew Dunstan writes:
> On 03/28/2017 05:23 AM, Dagfinn Ilmari Mannsåker wrote:
>> +@opts = grep { !/\$\(/ && /^--/ }
>> +map { (my $x = $_) =~
>> s/\Q$(top_builddir)\E/\"$topdir\"/; $x;}
>> +s
Andrew Dunstan writes:
> I would try something like this:
>
> @opts = grep { $_ !~ /\$\(/ && $_ =~ /^--/ }
> map { s/\Q$(top_builddir)\E/\"$topdir\"/; }
> split(/\s+/, $1);
That map is not going to work: it'll modify the values returned by
split(), but s/// (without the /r modi
Tom Lane writes:
> I wrote:
>> Peter Eisentraut writes:
>>> Clean up Perl code according to perlcritic
>
>> This seems to have broken the regression tests (specifically, dblink)
>> on at least some of the Windows buildfarm critters.
>
> I'm hardly a Perl expert, but it looks to me like the culpr
Peter Eisentraut writes:
> On 3/1/17 11:21, Dagfinn Ilmari Mannsåker wrote:
>> diff --git a/src/pl/plperl/plc_perlboot.pl b/src/pl/plperl/plc_perlboot.pl
>> index 292c9101c9..b4212f5ab2 100644
>> --- a/src/pl/plperl/plc_perlboot.pl
>> +++ b/src/pl/plperl/plc_perl
Jim Nasby writes:
> On 2/25/17 10:27 AM, Peter Eisentraut wrote:
>> So I'm also wondering here which style people prefer so
>> I can implement it there.
>
> I think the more OO style is definitely better. I expect it would
> simplify the code as well.
I'm not a Python person, but I'd argue that
Tom Lane writes:
> For the basic build process, we've largely solved that through the
> use of "make -s". But we don't really have a comparable "be quiet"
> option for test runs, especially not the TAP tests. Maybe we need
> to think a bit more globally about what it is we're trying ton
> accom
Magnus Hagander writes:
> AFAIK travis-ci would require us to use github as our hoster for all those
> things, and embrace that workflow, they don't support anything else.
>
> There might be others that do, just not travis.
It merely requires the repository to exist on GitHub, and postgresql.git
Jeff Janes writes:
> On Thu, Mar 9, 2017 at 3:20 PM, Robert Haas wrote:
>
>> Committed. Hopefully this doesn't contain any Perl bits that are
>> sufficiently new as to cause problems for our older BF members ... I
>> guess we'll see.
>
> Bad luck there. I'm getting this error on CentOS6.8, pe
Simon Riggs writes:
> On 7 March 2017 at 20:36, Alvaro Herrera wrote:
>
>> FWIW, +1 on improving matters here.
>
> +1 also.
>
> I don't see what's wrong with relying on buildfarm though; testing is
> exactly what its there for.
>
> If we had a two-stage process, where committers can issue "trial
ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
> Hi Peter,
>
> Peter Eisentraut writes:
>
>> I posted this about 18 months ago but then ran out of steam. [ ] Here
>> is an updated patch. The testing instructions below still apply.
>> Especially welcome wo
David Christensen writes:
>> Hi David,
>>
>> Here's a review of your patch.
>
> Hi Ilmari, thanks for your time and review. I’m fine with the revised
> version.
Okay, I've marked the patch as Ready For Committer.
Thanks,
Ilmari
--
"The surreality of the universe tends towards a maximum" -
Hi David,
Here's a review of your patch.
David Christensen writes:
> Throws a build error if we encounter a different number of fields in a
> DATA() line than we expect for the catalog in question.
The patch is a good idea, and as-is implements the suggested feature.
Tested by removing an attr
Hi Peter,
Peter Eisentraut writes:
> I posted this about 18 months ago but then ran out of steam. [ ] Here
> is an updated patch. The testing instructions below still apply.
> Especially welcome would be ideas on how to address some of the places
> I have marked with ## no critic.
Attached is
Michael Paquier writes:
> On Mon, Feb 27, 2017 at 11:28 AM, Craig Ringer wrote:
>
>> Still needs applying to pg9.6 and pg10.
>
> I did not understand at first what you meant, but after looking at the
> commit message of the patch things are clear:
> Newer Perl or IPC::Run versions default to app
Kevin Grittner writes:
> It occurred to me that it would make sense to allow these settings
> to be attached to a database or table (though *not* a role). I'll
> look at that when I get back if you don't get to it first.
Attached is a draft patch (no docs or tests) on top of the previous one,
a
ilm...@ilmari.org (Dagfinn Ilmari Manns�ker) writes:
> Here's an updated patch wich adds it as a separate stanza.
I've added this to the current commit fest:
https://commitfest.postgresql.org/13/1043/
--
"I use RMS as a guide in the same way that a boat captain would use
a lighthouse. It's go
Bruce Momjian writes:
> On Sun, Jan 15, 2017 at 03:01:47PM -0600, Jim Nasby wrote:
>> Not sure how many people still use [1], as referenced by our git wiki[2],
>> but it appears git worktrees are a viable replacement for that technique. In
>> short, if you're already in your checkout:
>>
>> git
Tom Lane writes:
> ilm...@ilmari.org (Dagfinn Ilmari =?utf-8?Q?Manns=C3=A5ker?=) writes:
>> While playing with prepared statements in psql, I noticed that EXECUTE
>> tab-completes the list of active prepared statements, but DEALLOCATE
>> does not.
>> Attached is a patch to fix this.
>
> Good idea
Hi hackers,
While playing with prepared statements in psql, I noticed that EXECUTE
tab-completes the list of active prepared statements, but DEALLOCATE
does not.
Attached is a patch to fix this.
Cheers,
Ilmari
--
"I use RMS as a guide in the same way that a boat captain would use
a lighthous
Kevin Grittner writes:
> (1) The new GUCs are named max_pred_locks_per_*, but the
> implementation passes them unmodified from a function specifying at
> what count the lock should be promoted. We either need to change
> the names or (probably better, only because I can't think of a good
> name
ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
> One thing I don't like about this patch is that if a user has increased
> max_pred_locks_per_transaction, they need to set
> max_pred_locks_per_relation to half of that to retain the current
> behaviour, or they'll su
Kevin Grittner writes:
> On Wed, Dec 14, 2016 at 8:16 AM, Dagfinn Ilmari Mannsåker
> wrote:
>
>> Attached is a patch
>
> Please add this to the open CommitFest to ensure that it is reviewed.
Done.
https://commitfest.postgresql.org/12/910/
--
"The surreality of th
Hi hackers,
I have a workload using SSI that causes a lot of tuple predicate to be
promoted to page locks. This causes a lot of spurious serialisability
errors, since the promotion happens at once three tuples on a page are
locked, and the affected tables have 30-90 tuples per page.
PredicateLoc
Noah Misch writes:
> On Mon, Nov 21, 2016 at 05:37:50PM +0900, Kyotaro HORIGUCHI wrote:
>> At Mon, 21 Nov 2016 16:45:10 +0900, Michael Paquier
>> wrote in
>>
>> > No objections to this version. It's a good idea to document that.
>>
>> +1
>
> Committed after re-filling paragraphs.
I just not
Hi hackers,
Attached is a patch which adds diff= directives to .gitattributes for C,
Perl and (X|SG)ML files. This makes word diffs and the function
indicator in the diff chunk header and more useful.
>From 57d7d4ec5b94783bf68b2959128e33c28547a6b8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Dagfinn
Robert Haas writes:
> On Tue, Nov 8, 2016 at 8:53 AM, Dagfinn Ilmari Mannsåker
> wrote:
>> Thank you very much. I just looked at the patch again and realised the
>> completion of "TO" after RENAME VALUE can be merged with the one
>> for RENAME ATTRIBUTE .
Artur Zakirov writes:
> Hello,
Hi Artur,
> 2016-09-12 16:16 GMT+03:00 Dagfinn Ilmari Mannsåker :
>>
>> I've added it to the 2016-11 commit fest:
>> https://commitfest.postgresql.org/11/795/
>>
>> - ilmari
>
> I've tested your patch.
[...]
&
Jim Nasby writes:
> BTW, I've sometimes wished for a mode where queries would silently have
> result ordering intentionally futzed, to eliminate any possibility of
> dependence on tuple ordering (as well as having sequences start at some
> random value).
FWIW, SQLite has this, in the form of 'PR
ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
> Hi hackers,
>
> Here's a patch to add psql tab completion for the recently-added ALTER
> TYPE … RENAME VALUE feature (thanks to Tom for fixing it up and
> committing it).
I've added it to the 2
Hi hackers,
Here's a patch to add psql tab completion for the recently-added ALTER
TYPE … RENAME VALUE feature (thanks to Tom for fixing it up and
committing it).
It's modelled on the ALTER TYPE … RENAME ATTRIBUTE completion, but
tweaked to return string literals instead of identifiers.
- ilmari
Emre Hasegeli writes:
>> Bottom line here is that I'd rather commit ALTER TYPE RENAME VALUE with
>> no EXISTS features and then see it accrete those features together with
>> other types of RENAME, when and if there's a will to make that happen.
>
> This sounds like a good conclusion to me.
Work
Emre Hasegeli writes:
>> Here is v4, which changes the command from ALTER VALUE to RENAME VALUE,
>> for consistency with RENAME ATTRIBUTE.
>
> It looks like we always use "ALTER ... RENAME ... old_name TO
> new_name" syntax, so it is better that way. I have noticed that all
> the other RENAMEs g
ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
> ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
>
>> I was bored and thought "how hard could it be?", and a few hours'
>> hacking later, I have something that seems to work. It doesn't do IF
>
Magnus Hagander writes:
[pread/pwrite]
> Yeah, Windows does not have those API calls, but it shouldn't be rocket
> science to write a wrapper for it. The standard windows APIs can do the
> same thing -- but they'll need access to the HANDLE for the file and not
> the posix file descriptor.
>
> I
Peter Eisentraut writes:
> On 7/6/16 4:52 AM, Dagfinn Ilmari Mannsåker wrote:
>> This keeps the indentation consistent when editing the documentation
>> using Emacs.
>
> Unfortunately, sgml-basic-offset is not a "safe" variable, so if we put
> it into .dir-lo
This keeps the indentation consistent when editing the documentation
using Emacs.
>From c345671ae4704df500dd17719c5e9973001663c1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Dagfinn=20Ilmari=20Manns=C3=A5ker?=
Date: Sat, 26 Mar 2016 21:58:32 +
Subject: [PATCH] Set sgml-basic-offset to 1 in .dir-l
Tom Lane writes:
> Robert Haas writes:
>> Right. If there were a DEFAULT on the new column that would of course
>> be different, and you can also do thinks like CHECK (a != b) here.
>> However, if the CHECK constraint does not reference any column other
>> than the newly-added one, and if the n
Marko Tiikkaja writes:
> On 2016-03-27 19:30, Dagfinn Ilmari Mannsåker wrote:
>> ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
>>
>>> I was bored and thought "how hard could it be?", and a few hours'
>>> hacking later, I have something that
ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
> I was bored and thought "how hard could it be?", and a few hours'
> hacking later, I have something that seems to work. It doesn't do IF
> NOT EXISTS yet, and the error messaging could do with some improvement,
Andrew Dunstan writes:
> On 03/25/2016 04:13 AM, Matthias Kurz wrote:
>>
>> Hopefully at the commitfest at least the transaction limitation
>> will/could be tackled - that would help us a lot already.
>
> I don't believe anyone knows how to do that safely. Enums pose special
> problems here exact
ilm...@ilmari.org (Dagfinn Ilmari Mannsåker) writes:
>
> I was bored and thought "how hard could it be?", and a few hours'
> hacking later, I have something that seems to work. It doesn't do IF
> NOT EXISTS yet, and the error messaging could do with some impr
Matthias Kurz writes:
[altering and dropping enum values]
>>> Andrew Dunstan writes:
>>> > On 03/09/2016 11:07 AM, Tom Lane wrote:
>>> >> I have a vague recollection that we discussed this at the time the enum
>>> >> stuff went in, and there are concurrency issues? Don't recall details
>>> >>
Tom Lane writes:
> ilm...@ilmari.org (Dagfinn Ilmari =?utf-8?Q?Manns=C3=A5ker?=) writes:
>> 1) Perl's integers are at least pointer-sized and either signed or
>>unsigned, so can potentially hold up to 2⁶⁴-1. Floating point numbers
>>can also be larger than double (up to 128bit), allowing
Hi hackers,
Commit 23a27b039d94ba359286694831eafe03cd970eef changed the type of
numbers-of-tuples-processed counters to uint64 and adjusted various PLs
to cope with this. I noticed the PL/Perl changes did not take full
advantage of what Perl is capable of handling, so here's a patch that
improves
The comment in hv_store_string() says that negative key length to
hv_store() for UTF-8 is not documented, and mentions that 5.6 doesn't
track UTF-8-ness of keys. However, the negative length convention has
been documented since 5.16¹, and 5.6 is no longer supported. The
attached patch updates th
62 matches
Mail list logo