pgsql: Fix incorrect format placeholder.
Hackorum builds and tests every patch posted to the lists, not only commitfest submissions. This is Hackorum's own CI rather than the PostgreSQL project's, and it is still under testing - please report anything that looks wrong.
You can run a PostgreSQL built from this patch straight from Docker, with no checkout and no build:
docker run --rm -p 5432:5432 ghcr.io/hackorum-dev/postgres-patch:t225578psql -h localhost -U postgresBuilt from patchset v6 (message #6), July 29, 2026 at 12:53 AM.
Every patchset is also pushed to a branch of our PostgreSQL fork, so you can check out the same tree CI built. Without a PostgreSQL checkout:
git clone --branch t225578_6 https://github.com/hackorum-dev/postgres.gitIn a checkout you already have, add the fork once:
git remote add hackorum https://github.com/hackorum-dev/postgres.gitthen, for this patchset and every later one:
git fetch hackorum t225578_6 && git checkout t225578_6Patchset v6 (message #6) is on t225578_6
Fix incorrect format placeholder.
Per buildfarm warnings.
Branch
------
master
Details
-------
https://git.postgresql.org/pg/commitdiff/b0cf5444f9a8d915b2e9b44790025f17a7dc107f
Modified Files
--------------
src/bin/pg_basebackup/pg_receivewal.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
On Mon, Nov 08, 2021 at 07:32:41PM +0000, Tom Lane wrote:
Fix incorrect format placeholder.
Per buildfarm warnings.
Thanks, that's embarassing. The last time I bumped into this issue,
the report came from lapwing which would directly report a failure,
but it did not happen this time. Which animal has reported that?
--
Michael
Michael Paquier <michael@paquier.xyz> writes:
Thanks, that's embarassing. The last time I bumped into this issue,
the report came from lapwing which would directly report a failure,
but it did not happen this time. Which animal has reported that?
Umm ... florican and dangomushi spotted it. Maybe we need a 32-bit
machine with -Werror?
regards, tom lane
On Tue, Nov 09, 2021 at 12:19:36AM -0500, Tom Lane wrote:
Umm ... florican and dangomushi spotted it. Maybe we need a 32-bit
machine with -Werror?
Indeed, I can see it in the logs of dangomushi. I could add a -Werror
there, but there are also a bunch of warnings generated by plperl.
I'd rather keep --with-perl and clang, and there seems to be a fix
regarding that stuff in upstream:
https://www.nntp.perl.org/group/perl.perl5.changes/2021/07/msg57758.html
But it also looks like Archlinux ARM did not catch up that yet..
--
Michael
Michael Paquier <michael@paquier.xyz> writes:
On Tue, Nov 09, 2021 at 12:19:36AM -0500, Tom Lane wrote:
Umm ... florican and dangomushi spotted it. Maybe we need a 32-bit
machine with -Werror?
Indeed, I can see it in the logs of dangomushi. I could add a -Werror
there, but there are also a bunch of warnings generated by plperl.
Yeah, we had complaints before about -Wcompound-token-split-by-macro
warnings.
I'd rather keep --with-perl and clang, and there seems to be a fix
regarding that stuff in upstream:
https://www.nntp.perl.org/group/perl.perl5.changes/2021/07/msg57758.html
I figured the Perl guys would do something about it, but it will be a
very long time before that propagates everywhere. Maybe we should
teach configure to add -Wno-compound-token-split-by-macro? By my
count there are now five buildfarm members besides dangomushi that
are producing these warnings, and I think the number is going to
go up before it goes down.
(Meanwhile, I guess I could enable -Werror on florican.)
regards, tom lane
On Tue, Nov 09, 2021 at 01:06:26AM -0500, Tom Lane wrote:
I figured the Perl guys would do something about it, but it will be a
very long time before that propagates everywhere. Maybe we should
teach configure to add -Wno-compound-token-split-by-macro? By my
count there are now five buildfarm members besides dangomushi that
are producing these warnings, and I think the number is going to
go up before it goes down.(Meanwhile, I guess I could enable -Werror on florican.)
Yeah, you have a point here. -Wcompound-token-split-by-macro is new
as of clang-12 from last April, so I am a bit surprised that it is
that present in the buildfarm TBH.
The attached is sufficient to remove those warnings. Debian provides
clang-12, so it is easy to reproduce the problem.
What do you think?
--
Michael
Michael Paquier <michael@paquier.xyz> writes:
Yeah, you have a point here. -Wcompound-token-split-by-macro is new
as of clang-12 from last April, so I am a bit surprised that it is
that present in the buildfarm TBH.
The attached is sufficient to remove those warnings. Debian provides
clang-12, so it is easy to reproduce the problem.
OK by me, but probably should raise the issue in -hackers not just
-committers. Also, please back-patch once the release freeze lifts.
regards, tom lane