Re: Rewriting the test of pg_upgrade as a TAP test - take three - remastered set - Mailing list pgsql-hackers

From Thomas Munro
Subject Re: Rewriting the test of pg_upgrade as a TAP test - take three - remastered set
Date
Msg-id CA+hUKGJDZt0MRhVSaaO5DX8KLjWf9RPD4YcsnfTSRcRjzCoY1g@mail.gmail.com
Whole thread Raw
In response to Re: Rewriting the test of pg_upgrade as a TAP test - take three - remastered set  (Andres Freund <andres@anarazel.de>)
Responses Re: Rewriting the test of pg_upgrade as a TAP test - take three - remastered set  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: Rewriting the test of pg_upgrade as a TAP test - take three - remastered set  (Andres Freund <andres@anarazel.de>)
List pgsql-hackers
On Sun, Feb 13, 2022 at 5:50 PM Andres Freund <andres@anarazel.de> wrote:
> On 2022-01-18 11:20:16 +0900, Michael Paquier wrote:
> > +REGRESS_OUTPUTDIR=$(abs_top_builddir)/src/bin/pg_upgrade
> > +export REGRESS_OUTPUTDIR
>
> I don't really understand why 027_stream_regress.pl is using this (and thus
> not why it's used here). The tap tests create files all the time, why is this
> different?
>
> It's not like make / msvc put the data in different places:
> src/test/recovery/Makefile:REGRESS_OUTPUTDIR=$(abs_top_builddir)/src/test/recovery/tmp_check
> src/tools/msvc/vcregress.pl: $ENV{REGRESS_OUTPUTDIR} = "$topdir/src/test/recovery/tmp_check";

As I wrote in
https://www.postgresql.org/message-id/CA%2BhUKGK-%2Bmg6RWiDu0JudF6jWeL5%2BgPmi8EKUm1eAzmdbwiE_A%40mail.gmail.com,

>> > +# required for 027_stream_regress.pl
>> > +REGRESS_OUTPUTDIR=$(abs_top_builddir)/src/test/recovery
>> > +export REGRESS_OUTPUTDIR
>>
>> Why do we need this?
>
> The Make macro "prove_check" (src/Makefile.global.in) always changes
> to the source directory to run TAP tests. Without an explicit
> directive to control where regression test output goes, it got
> splattered all over the source tree in VPATH builds. I didn't see an
> existing way to adjust that (did I miss something?). Hence desire to
> pass down a path in the build tree. Better ideas welcome.

I thought it was a goal that VPATH builds shouldn't pollute the source
tree, but the Make macro prove_check is explicitly doing so by
default.  Perhaps *that* should be fixed?



pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: Rewriting the test of pg_upgrade as a TAP test - take three - remastered set
Next
From: Tom Lane
Date:
Subject: Re: Rewriting the test of pg_upgrade as a TAP test - take three - remastered set