Re: CI and test improvements - Mailing list pgsql-hackers

From Justin Pryzby
Subject Re: CI and test improvements
Date
Msg-id ZDYD0aoGaGWMymz2@telsasoft.com
Whole thread Raw
In response to Re: CI and test improvements  (Peter Eisentraut <peter.eisentraut@enterprisedb.com>)
Responses Re: CI and test improvements
List pgsql-hackers
On Wed, Mar 15, 2023 at 04:57:34PM +0100, Peter Eisentraut wrote:
> On 15.03.23 15:56, Justin Pryzby wrote:
> > I'm surprised if there's any question about the merits of making
> > documentation easily available for review.  Several people have agreed;
> > one person mailed me privately specifically to ask how to show HTML docs
> > on cirrusci.
> > 
> > Anyway, all this stuff is best addressed either before or after the CF.
> > I'll kick the patch forward.  Thanks for looking.
> 
> I suppose this depends on what you want to use this for.  If your use is to
> prepare and lay out as much information as possible about a patch for a
> reviewer, some of your ideas make sense.
> 
> I'm using this primarily to quickly test local work in progress.  So I want
> a quick feedback cycle.  I don't need it to show me which HTML docs changed,
> for example.
> 
> So maybe there need to be different modes.

I'm opened to that - for example, mingw is currently opt-in.  Maybe this
should be a separate task - it was implemented like that based on an
earlier suggestion (and then changed back again based on another
suggestion).  The task could be triggered manually or by cfbot's
message.

But a primary goal for cirrus.yml was to allow developers to do the same
things as cfbot, and without everyone needing to reimplement it for
themselves.

You want quick feedback, like everyone else - but I doubt you disable
the documentation build when you don't need it, even though that would
shave off a whole minute.  And I doubt that you'd comment it out even
the documentation was built twice.

Anyway - I think this patch is probably waiting on Andres' patch to
"convert CompilerWarnings to meson".

> > 7e09035f588 WIP: ci/meson: allow showing only failed tests ..
> 
> I'm not sure I like this one.  I sometimes look up the logs of non-failed
> tests to compare them with failed tests, to get context to could lead to
> failures.  Maybe we can make this behavior adjustable. But I've not been
> bothered by the current behavior.

I suggest to try the patch; I doubt you'd prefer the existing behavior.

The patch is rebased now that meson is updated to avoid the windows
python warnings (thanks Andres).

-- 
Justin

Attachment

pgsql-hackers by date:

Previous
From: Justin Pryzby
Date:
Subject: Re: Add LZ4 compression in pg_dump
Next
From: Michael Paquier
Date:
Subject: Re: Add LZ4 compression in pg_dump