On Thu, Nov 29, 2018 at 11:00:11AM +0900, Michael Paquier wrote:
> I am attaching the patch I would like to commit to close this thread,
> which has as single change TAP_TESTS commented out in contrib/bloom/ to
> not explode the buildfarm. Any objections to that?
And I gave this stuff another sho
On Fri, Nov 30, 2018 at 02:18:04PM +0300, Arthur Zakirov wrote:
> As far as I understand bloom TAP tests are stable on some platforms, such as
> linux. Can be TAP test disabled only for some specific platforms?
That can be done with PORTNAME. However I am doubting that those tests
are even reliab
On 29.11.2018 05:00, Michael Paquier wrote:
The main patch had to be reverted a couple of days ago via 1d7dd18, and
I have been digging and fixing the set of issues the buildfarm has been
complaining about since. There were two problems:
Thank you for working on the patch.
I run world-check.
On Fri, Nov 23, 2018 at 09:43:45AM +0900, Michael Paquier wrote:
> That's for the main regression test suite within src/test/regress, TAP
> is also a regression test suite, the page of the link reflects the
> general set of test suites available.
The main patch had to be reverted a couple of days
On Mon, Nov 26, 2018 at 10:50:35AM +0300, Nikolay Shaplov wrote:
> I see there PGXS mentioned several times as "a build infrastructure
> for extensions" and as an environment variable it is mentioned only in
> code sample
As a variable PGXS stands for the location of the makefile which holds
all t
В письме от 26 ноября 2018 08:53:20 Вы написали:
> > I've carefully read this documentation. And did not get clear explanation
> > of what is the true purpose of PGXS environment variable. Only
> >
> > "The last three lines should always be the same. Earlier in the file, you
> > assign variables
On Sun, Nov 25, 2018 at 05:49:42PM +0300, Nikolay Shaplov wrote:
> I've carefully read this documentation. And did not get clear explanation of
> what is the true purpose of PGXS environment variable. Only
>
> "The last three lines should always be the same. Earlier in the file, you
> assign v
В письме от 23 ноября 2018 09:43:45 пользователь Michael Paquier написал:
> > So, I continued exploring your patch. First I carefully read changes in
> > pgxs.mk. The only part of it I did not understand is
> >
> > .PHONY: submake
> >
> > -submake:
> > ifndef PGXS
> >
> > +submake:
> > +ifdef
On Fri, Nov 23, 2018 at 05:29:00PM +0300, Arthur Zakirov wrote:
> The patch is very useful. Using TAP_TESTS is more convenient and clearer
> than adding wal-check target. Every time I was adding TAP tests for a
> extension I had to remember that I should add wal-check.
wal-check is a custom option
Hello,
On 21.11.2018 03:39, Michael Paquier wrote:
I have added a reference to regress-tap in one of the new paragraphs.
Linking the existing stuff to point to "regress" is a separate issue
though, and while pointing to the TAP section is adapted as its
guidelines are rather general, I am not su
On Thu, Nov 22, 2018 at 10:01:26PM +0300, Nikolay Shaplov wrote:
> В письме от 21 ноября 2018 09:39:53 пользователь Michael Paquier написал:
>> Those are mentioned here as part of the additional test suites:
>> https://www.postgresql.org/docs/11/regress-run.html#id-1.6.20.5.5
>
> Oh thanks (I feel
В письме от 21 ноября 2018 09:39:53 пользователь Michael Paquier написал:
> > For me name "output_iso" means nothing. iso is something about CD/DVD or
> > about standards. I would not guess that iso stands for isolation if I did
> > not know it already. isolation_output is more sensible: I have he
On Tue, Nov 20, 2018 at 07:30:39PM +0300, Nikolay Shaplov wrote:
> Is it ok, if I join the reviewing? I like test, especially TAP one, you know
> ;-)
>
> Since you are much more experienced in postgres then me, I'd try to
> understand how does the patch work, try to use if for writing more TAP
> t
В письме от 10 ноября 2018 09:14:19 пользователь Michael Paquier написал:
> > Nice cleanup. Also, I like the ability to enable/control more types of
> > tests easily from the Makefile. What are the next steps for this
> > patch?
>
> Thanks. It seems to me that a complete review is still in order
On Fri, Nov 09, 2018 at 03:14:00PM +, Adam Berlin wrote:
> Nice cleanup. Also, I like the ability to enable/control more types of
> tests easily from the Makefile. What are the next steps for this
> patch?
Thanks. It seems to me that a complete review is still in order,
particularly regarding
Nice cleanup. Also, I like the ability to enable/control more types of tests
easily from the Makefile. What are the next steps for this patch?
On Wed, Sep 05, 2018 at 06:48:49PM -0700, Michael Paquier wrote:
> On a recent thread of pgsql-committers has been discussed the fact that
> we lacked a bit of infrastructure to allow extensions to control
> isolation and TAP tests:
> https://www.postgresql.org/message-id/20180905174527.ga2...@paqu
17 matches
Mail list logo
