The Archiving functionality of guix lint should be opt-in and Documented more prominently

  • Open
  • quality assurance status badge
Details
3 participants
  • MSavoritias
  • Richard Sent
  • Simon Tournier
Owner
unassigned
Submitted by
MSavoritias
Severity
normal
M
M
MSavoritias wrote on 21 Jun 2024 18:45
(address . bug-guix@gnu.org)
20240621194531.40c24620@fannys.me
Hey,

There was recently a discussion around SWH and it came up that `guix lint` actually by default when you run it without arguments, runs all the linters.
One of them being the archive linter that contacts SWH archive to let it know where to download the source code from (if its a public repo).

I would like to propose to make that linter off by default. Because:

The tool is name `guix lint` and it is not obvious (unless somebody runs --list-linters after --help) that it also does code archiving.
To that end it breaks the expectations of the person using the tool to have their code silently uploaded to SWH. (if its a public repo again)

What we should do instead:

Instead we should document more prominently in the manual that `guix lint` also does archiving and encourage people to actually archive the software they write to SWH.
(assuming they are the authors that is. A disclaimer to get permission from the author of the software should be also added if they are not.)
And for the usecase of Guix, they flag can just be turned on by default since as a project we are interested in code archival.

MSavoritias
S
S
Simon Tournier wrote on 21 Jun 2024 20:46
(name . MSavoritias)(address . email@msavoritias.me)(address . 71700@debbugs.gnu.org)
877ceikuat.fsf@gmail.com
Hi,

On Fri, 21 Jun 2024 at 19:45, MSavoritias <email@msavoritias.me> wrote:

Toggle quote (2 lines)
> I would like to propose to make that linter off by default.

Somehow I disagree with this. And I propose the generic approach that
allows to exclude any checkers from the package definition using the
field properties.


Cheers,
simon
R
R
Richard Sent wrote on 22 Jun 2024 15:21
(name . MSavoritias)(address . email@msavoritias.me)
87iky1azb6.fsf@freakingpenguin.com
I think channel level configuration of some form for code archival is a
good idea so individual channels can choose to disable it. I also agree
that we should make the fact that guix lint does archival more
prominent.

I disagree with a statement that permission is required, but I'll avoid
rehashing the discussion ongoing in guix-devel. [1]

I think there is a good reason to support disabling archival at the
channel level. Simon, do you think your patch can/will manage that?

Toggle quote (6 lines)
> Somehow I disagree with this. And I propose the generic approach that
> allows to exclude any checkers from the package definition using the
> field properties.
>
> See <https://issues.guix.gnu.org/71697#1>.


--
Take it easy,
Richard Sent
Making my computer weirder one commit at a time.
M
M
Msavoritias wrote on 22 Jun 2024 16:24
(name . Richard Sent)(address . richard@freakingpenguin.com)
20240622172419.477c5b32@fannys.me
On Sat, 22 Jun 2024 09:21:01 -0400
Richard Sent <richard@freakingpenguin.com> wrote:

Toggle quote (11 lines)
> I think channel level configuration of some form for code archival is a
> good idea so individual channels can choose to disable it. I also agree
> that we should make the fact that guix lint does archival more
> prominent.
>
> I disagree with a statement that permission is required, but I'll avoid
> rehashing the discussion ongoing in guix-devel. [1]
>
> I think there is a good reason to support disabling archival at the
> channel level. Simon, do you think your patch can/will manage that?

That is still missing the usage of people wanting to run `guix lint` without having a channel.
A channel level mechanism would be nice indeed but we still need a way to account for the archiving functionality for people who dont have channels or dont run channels.
The proposal of making it explicitely enabled would work as a solution for that use case.
That way the channel configuration would be to enable it instead of disabling it. opt-in/opt-out and all that.

It also avoids the mistake of not realizing it exists or is enabled and accidentally somebodys code ends up in SWH without them meaning too.
Not everybody reads the manual after all and we shouldnt do stuff we havent been explicitly required to do.

In short I would say a channel level mechanism would help to "automate" the opt-in of running `--archival` everywhere with `guix lint`.

MSavoritias

Toggle quote (8 lines)
> > Somehow I disagree with this. And I propose the generic approach that
> > allows to exclude any checkers from the package definition using the
> > field properties.
> >
> > See <https://issues.guix.gnu.org/71697#1>.
>
> [1]: https://lists.gnu.org/archive/html/guix-devel/2024-06/msg00192.html
>
S
S
Simon Tournier wrote on 22 Jun 2024 17:30
(address . 71700@debbugs.gnu.org)
87bk3tuh97.fsf@gmail.com
Hi Richard,

On Sat, 22 Jun 2024 at 09:21, Richard Sent <richard@freakingpenguin.com> wrote:

Toggle quote (3 lines)
> I think there is a good reason to support disabling archival at the
> channel level. Simon, do you think your patch can/will manage that?

Yeah it could be helpful. However, my patch does not address at this
level.

I agree it could be an other complementary direction. But the design at
channel needs to be thought a bit, IMHO.

Cheers,
simon
?
Your comment

Commenting via the web interface is currently disabled.

To comment on this conversation send an email to 71700@debbugs.gnu.org

To respond to this issue using the mumi CLI, first switch to it
mumi current 71700
Then, you may apply the latest patchset in this issue (with sign off)
mumi am -- -s
Or, compose a reply to this issue
mumi compose
Or, send patches to this issue
mumi send-email *.patch