Thread: Re: [PATCH] Fix OAuth feature detection on OpenBSD+Meson

Hi,

On Tue, 24 Jun 2025 at 02:37, Jacob Champion
<jacob.champion@enterprisedb.com> wrote:
> As a potential follow-up, is there any interest in switching the
> Cirrus Meson setup to explicitly enable the features that we expect to
> test? That would match what we do for Autoconf, I think; it would have
> helped me catch my mistake earlier.

I think this is a good idea. Another point is that CI images and their
packages are updated automatically, so it would be easier to catch if
something breaks when the VM is updated.

-- 
Regards,
Nazir Bilal Yavuz
Microsoft

On Tue, Jun 24, 2025 at 1:27 AM Nazir Bilal Yavuz <byavuz81@gmail.com> wrote:
> I think this is a good idea. Another point is that CI images and their
> packages are updated automatically, so it would be easier to catch if
> something breaks when the VM is updated.

Yes, that's a great point too. Okay, sounds like there is some
interest, and I'll add it to my list of patchsets to try (but if
anyone wants to beat me to it, please go ahead!).

--Jacob

On Wed, Jun 25, 2025 at 3:46 AM Nazir Bilal Yavuz <byavuz81@gmail.com> wrote:
> I wanted to experiment with it.

That was fast, thank you!

> First, I got the current list of
> features from upstream, then disabled the auto features, then
> explicitly enabled these features. I did this only for FreeBSD to show
> my idea roughly.
>
> There are two patches; 0001 disables auto features for all of the
> tasks and 0002 explicitly enables these features for FreeBSD.

Just from a stylistic perspective, I think I'd prefer for
`--auto-features=disabled` to be folded into the beginning of the
MESON_FEATURES variable. Or is there another reason to control them
fully independently?

Also: how do we ensure that none of the previously enabled features
were missed in this list, for all the OSes we've modified? (I'm a
Meson novice.)

> What do you think about this approach? If you are okay with this, I
> can apply it to all CI tasks (and possibly create another thread for
> it).

Seems good to me. I think a new top-level thread would be good too, to
get more eyes on the changes.

Thanks!
--Jacob

Hi,

On Mon, 30 Jun 2025 at 17:45, Jacob Champion
<jacob.champion@enterprisedb.com> wrote:
>
> On Wed, Jun 25, 2025 at 3:46 AM Nazir Bilal Yavuz <byavuz81@gmail.com> wrote:
> > I wanted to experiment with it.
>
> That was fast, thank you!
>
> > First, I got the current list of
> > features from upstream, then disabled the auto features, then
> > explicitly enabled these features. I did this only for FreeBSD to show
> > my idea roughly.
> >
> > There are two patches; 0001 disables auto features for all of the
> > tasks and 0002 explicitly enables these features for FreeBSD.
>
> Just from a stylistic perspective, I think I'd prefer for
> `--auto-features=disabled` to be folded into the beginning of the
> MESON_FEATURES variable. Or is there another reason to control them
> fully independently?

No, I had thought people would prefer to control auto-features from
one central place but apparently I was wrong :) I will update the
patch according to this.

> Also: how do we ensure that none of the previously enabled features
> were missed in this list, for all the OSes we've modified? (I'm a
> Meson novice.)

I am not sure, I do not know if there is a way to ensure that. I just
manually checked features from the end of the 'meson setup' command's
output. I think this should be enough but of course an automated way
would be better if there is any.

> > What do you think about this approach? If you are okay with this, I
> > can apply it to all CI tasks (and possibly create another thread for
> > it).
>
> Seems good to me. I think a new top-level thread would be good too, to
> get more eyes on the changes.

Thanks! I will create the thread soon.

--
Regards,
Nazir Bilal Yavuz
Microsoft