cpluspluscheck vs vpath
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:t40699psql -h localhost -U postgresBuilt from patchset v2 (message #2), July 28, 2026 at 04:12 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 t40699_2 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 t40699_2 && git checkout t40699_2Patchset v2 (message #2) is on t40699_2
Hi,
right now cpluspluscheck doesn't work with vpath builds. That's pretty
annoying, because it does require cloning the git tree into a separate
directory + doing configure there just to run cpluspluscheck.
Attached is a small patch allowing cpluspluscheck to run from different
directories. I needs the src and build directories for that,
unsurprisingly.
As that makes it more complicated to invoke, I added a makefile target
(in the top level) for it.
Seems we could round the edges a good bit further than what's done in
the attached (argument checking, for example). But I think this would
already be an improvement?
Greetings,
Andres Freund
Hi,
On 2019-05-30 15:02:44 -0700, Andres Freund wrote:
right now cpluspluscheck doesn't work with vpath builds. That's pretty
annoying, because it does require cloning the git tree into a separate
directory + doing configure there just to run cpluspluscheck.Attached is a small patch allowing cpluspluscheck to run from different
directories. I needs the src and build directories for that,
unsurprisingly.As that makes it more complicated to invoke, I added a makefile target
(in the top level) for it.Seems we could round the edges a good bit further than what's done in
the attached (argument checking, for example, but also using the C++
compiler from configure). But I think this would already be an
improvement?
Ugh, sent the previous email too early.
Greetings,
Andres Freund
On 2019-May-30, Andres Freund wrote:
On 2019-05-30 15:02:44 -0700, Andres Freund wrote:
Seems we could round the edges a good bit further than what's done in
the attached (argument checking, for example, but also using the C++
compiler from configure). But I think this would already be an
improvement?
+1 I've stumbled upon this too.
--
�lvaro Herrera https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
Andres Freund <andres@anarazel.de> writes:
Attached is a small patch allowing cpluspluscheck to run from different
directories. I needs the src and build directories for that,
unsurprisingly.
No objection to changing this, but you could reduce the surprise
factor for existing workflows with a couple of defaults for the
arguments --- allow srcdir to default to "." and builddir to default
to the same as srcdir.
regards, tom lane
Hi,
On 2019-05-31 09:56:45 -0400, Tom Lane wrote:
Andres Freund <andres@anarazel.de> writes:
Attached is a small patch allowing cpluspluscheck to run from different
directories. I needs the src and build directories for that,
unsurprisingly.No objection to changing this, but you could reduce the surprise
factor for existing workflows with a couple of defaults for the
arguments --- allow srcdir to default to "." and builddir to default
to the same as srcdir.
Pushed, with that modification.
Would be kinda nice to do the check in parallel...
- Andres