Re: [HACKERS] Removal of plaintext password type references

2017-05-20 Thread Heikki Linnakangas
On 05/20/2017 05:41 AM, Tom Lane wrote: Robert Haas writes: I guess it does seem likely that most users of the hook would need to do the same, but it seems confusing to pass the same function both x and f(x), so my vote is to not do that. Right, that was my thinking.

Re: [HACKERS] proposal psql \gdesc

2017-05-20 Thread Fabien COELHO
Hello Pavel, I am sending a variant with message instead empty result. Thanks. Patch looks good, applies, make check ok, code is neat. Personally I prefer empty result instead message. Hmm. I think that this version is less likely to raise questions from users, especially compared to

[HACKERS] type cache for concat functions

2017-05-20 Thread Pavel Stehule
Hi Now concat is 2x times slower than || operator. With cached FmgrInfo for output function it will be only 5%. Regards Pavel diff --git a/src/backend/utils/adt/varlena.c b/src/backend/utils/adt/varlena.c index be399f4..d3f04f8 100644 --- a/src/backend/utils/adt/varlena.c +++

Re: [HACKERS] Re: proposal - using names as primary names of plpgsql function parameters instead $ based names

2017-05-20 Thread Pavel Stehule
2017-05-19 5:48 GMT+02:00 Pavel Stehule : > > > 2017-05-19 3:14 GMT+02:00 Peter Eisentraut com>: > >> On 5/15/17 14:34, Pavel Stehule wrote: >> > Now, I when I working on plpgsql_check, I have to check function >> > parameters. I

[HACKERS] Allowing dash character in LTREE

2017-05-20 Thread Cyril Auburtin
It could be useful to allow the `-` char in allowed LTREE label characters (currently a-zA-Z0-9_ https://www.postgresql.org/docs/9.6/static/ltree.html) The reason is to allow to use more easily base64 ids, (like https://github.com/dylang/shortid, ...), uuid's too in labels `-` is also not used

Re: [HACKERS] statement_timeout is not working as expected with postgres_fdw

2017-05-20 Thread Amit Kapila
On Thu, May 18, 2017 at 7:56 PM, Robert Haas wrote: > On Wed, May 17, 2017 at 6:57 AM, Amit Kapila wrote: >> +1. Why not similar behavior for any other statements executed in >> this module by do_sql_command? > > The other cases are not quite the

Re: [HACKERS] Race conditions with WAL sender PID lookups

2017-05-20 Thread Michael Paquier
On Fri, May 19, 2017 at 3:01 PM, Masahiko Sawada wrote: > Also, as Horiguchi-san pointed out earlier, walreceiver seems need the > similar fix. Actually, now that I look at it, ready_to_display should as well be protected by the lock of the WAL receiver, so it is

[HACKERS] Tightening isspace() tests where behavior should match SQL parser

2017-05-20 Thread Tom Lane
The proximate cause of bug #14662, https://www.postgresql.org/message-id/flat/20170519162316.29945.5021%40wrigleys.postgresql.org appears to be that SplitIdentifierString's isspace() checks are identifying some bytes of a multibyte character as spaces. It's not quite clear to me whether there's

Re: [HACKERS] Allowing dash character in LTREE

2017-05-20 Thread David G. Johnston
On Sat, May 20, 2017 at 1:27 AM, Cyril Auburtin wrote: > It could be useful to allow the `-` char in allowed LTREE label characters > (currently > a-zA-Z0-9_ https://www.postgresql.org/docs/9.6/static/ltree.html) > > The reason is to allow to use more easily base64

Re: [HACKERS] Making replication commands case-insensitive

2017-05-20 Thread Andres Freund
Hi, On 2017-05-20 21:19:10 +0900, Michael Paquier wrote: > $subject has been raised in a recent thread here: > https://www.postgresql.org/message-id/CAB7nPqTmym5t-X6hvMF_P-KRc=ndXtbQCTiU=nhs_jvl7x1...@mail.gmail.com > > The idea is to make the replication protocol a bit more flexible, in a > way

Re: [HACKERS] Allowing dash character in LTREE

2017-05-20 Thread David G. Johnston
On Sat, May 20, 2017 at 11:26 AM, Cyril Auburtin wrote: > Ah sorry, first time, I thought it didn't pass > ​You should check our excellent online mailing list archives before re-sending. https://www.postgresql.org/list/ David J. ​

Re: [HACKERS] [PATCH v2] Progress command to monitor progression of long running SQL queries

2017-05-20 Thread Amit Kapila
On Wed, May 17, 2017 at 9:43 PM, Remi Colinet wrote: > > 2017-05-13 14:38 GMT+02:00 Amit Kapila : >> >> On Wed, May 10, 2017 at 10:10 PM, Remi Colinet >> wrote: >> > >> > Parallel queries can also be monitored. The same

[HACKERS] Making replication commands case-insensitive

2017-05-20 Thread Michael Paquier
Hi all, $subject has been raised in a recent thread here: https://www.postgresql.org/message-id/CAB7nPqTmym5t-X6hvMF_P-KRc=ndXtbQCTiU=nhs_jvl7x1...@mail.gmail.com The idea is to make the replication protocol a bit more flexible, in a way similar to what 5c837dd has done, but for repl_scanner.l.

Re: [HACKERS] Making replication commands case-insensitive

2017-05-20 Thread Magnus Hagander
On Sat, May 20, 2017 at 2:19 PM, Michael Paquier wrote: > Hi all, > > $subject has been raised in a recent thread here: > https://www.postgresql.org/message-id/CAB7nPqTmym5t- > X6hvMF_P-KRc=ndXtbQCTiU=nhs_jvl7x1...@mail.gmail.com > > The idea is to make the replication

Re: [HACKERS] OK, so culicidae is *still* broken

2017-05-20 Thread Noah Misch
On Sat, Apr 15, 2017 at 02:30:18PM -0700, Andres Freund wrote: > On 2017-04-15 17:24:54 -0400, Tom Lane wrote: > > Andres Freund writes: > > > On 2017-04-15 17:09:38 -0400, Tom Lane wrote: > > >> Why doesn't Windows' ability to map the segment into the new process > > >>

Re: [HACKERS] Allowing dash character in LTREE

2017-05-20 Thread Cyril Auburtin
Ah sorry, first time, I thought it didn't pass It's to support multiple ids (base64 ones are frequent) (or uuid's) in a path. I think it's more practical than an array, for searching and sorting But I understand that LTREE's main purpose is probably more for words 'breadcrumbs' etc.. 2017-05-20

Re: [HACKERS] proposal psql \gdesc

2017-05-20 Thread Pavel Stehule
Hi 2017-05-20 9:15 GMT+02:00 Fabien COELHO : > > Hello Pavel, > > I am sending a variant with message instead empty result. >> > > Thanks. Patch looks good, applies, make check ok, code is neat. > > Personally I prefer empty result instead message. >> > > Hmm. I think that

Re: [HACKERS] bumping HASH_VERSION to 3

2017-05-20 Thread Bruce Momjian
On Sun, May 21, 2017 at 09:11:29AM +0530, Amit Kapila wrote: > On Sun, May 21, 2017 at 5:26 AM, Bruce Momjian wrote: > > On Fri, May 19, 2017 at 04:54:27PM -0400, Robert Haas wrote: > >> On Tue, May 16, 2017 at 9:25 AM, Amit Kapila > >> wrote: > >> >

[HACKERS] [Q] When should I use reg* types instead of oid in the system catalog?

2017-05-20 Thread MauMau
Hello, Both pg_aggregate.aggfnoid and pg_trigger.tgfoid references pg_proc.oid, but the data types of them are regproc and oid respectively. Is there any criterion on when to which in the system catalog? Is the regproc choice just for readability of the catalog query output? Should

Re: pgindent (was Re: [HACKERS] [COMMITTERS] pgsql: Preventive maintenance in advance of pgindent run.)

2017-05-20 Thread Tom Lane
I wrote: > Also, I found two places where an overlength comment line is simply busted > altogether --- notice that a character is missing at the split point: I found the cause of that: you need to apply this patch: --- freebsd_indent/pr_comment.c~2017-05-17 14:59:31.548442801 -0400 +++

Re: [HACKERS] bumping HASH_VERSION to 3

2017-05-20 Thread Bruce Momjian
On Fri, May 19, 2017 at 04:54:27PM -0400, Robert Haas wrote: > On Tue, May 16, 2017 at 9:25 AM, Amit Kapila wrote: > > On Tue, May 16, 2017 at 5:14 PM, Robert Haas wrote: > >> On Tue, May 16, 2017 at 7:31 AM, Amit Kapila

Re: [HACKERS] Variable substitution in psql backtick expansion

2017-05-20 Thread Pavel Stehule
Hi 2017-04-02 21:57 GMT+02:00 Fabien COELHO : > > More to the point, we already have that. See c.h: >> >> #define CppAsString(identifier) #identifier >> #define CppAsString2(x) CppAsString(x) >> > > Thanks for the pointer. I grepped for them without success.

Re: [HACKERS] Regression in join selectivity estimations when using foreign keys

2017-05-20 Thread Tom Lane
David Rowley writes: > I've been analyzing a reported regression case between a 9.5 plan and > a 9.6 plan. I tracked this down to the foreign key join selectivity > code, specifically the use_smallest_selectivity code which is applied > to outer joins where the

Re: [HACKERS] bumping HASH_VERSION to 3

2017-05-20 Thread Amit Kapila
On Sun, May 21, 2017 at 5:26 AM, Bruce Momjian wrote: > On Fri, May 19, 2017 at 04:54:27PM -0400, Robert Haas wrote: >> On Tue, May 16, 2017 at 9:25 AM, Amit Kapila wrote: >> > On Tue, May 16, 2017 at 5:14 PM, Robert Haas wrote:

Re: [HACKERS] Race conditions with WAL sender PID lookups

2017-05-20 Thread Erik Rijkers
On 2017-05-20 14:40, Michael Paquier wrote: On Fri, May 19, 2017 at 3:01 PM, Masahiko Sawada wrote: Also, as Horiguchi-san pointed out earlier, walreceiver seems need the similar fix. Actually, now that I look at it, ready_to_display should as well be protected by the

Re: [HACKERS] proposal psql \gdesc

2017-05-20 Thread Fabien COELHO
Hello Pavel, - Maybe tests could also exercise unnamed columns, eg: SELECT 1, 2, 3 \gdesc \g done Can't see it. No big deal, but if you put it it did not get through, and there is a warning with git apply on the very last line of the patch which may be linked to that:

Re: pgindent (was Re: [HACKERS] [COMMITTERS] pgsql: Preventive maintenance in advance of pgindent run.)

2017-05-20 Thread Piotr Stefaniak
On 2017-05-21 03:00, Tom Lane wrote: > I wrote: >> Also, I found two places where an overlength comment line is simply busted >> altogether --- notice that a character is missing at the split point: > > I found the cause of that: you need to apply this patch: > > --- freebsd_indent/pr_comment.c~

Re: [HACKERS] [Q] When should I use reg* types instead of oid in the system catalog?

2017-05-20 Thread Tom Lane
"MauMau" writes: > Both pg_aggregate.aggfnoid and pg_trigger.tgfoid references > pg_proc.oid, but the data types of them are regproc and oid > respectively. Is there any criterion on when to which in the system > catalog? Is the regproc choice just for readability of the

Re: [HACKERS] proposal psql \gdesc

2017-05-20 Thread Pavel Stehule
2017-05-20 22:26 GMT+02:00 Fabien COELHO : > > Hello Pavel, > > - Maybe tests could also exercise unnamed columns, eg: >>> SELECT 1, 2, 3 \gdesc \g >>> >> >> done >> > > Can't see it. No big deal, but if you put it it did not get through, and > there is a warning with