You are not logged in.

#1 2022-01-21 09:31:08

xerus
Member
Registered: 2021-05-11
Posts: 34

How about automating orphaning?

I heard orphan requests have to be approved manually, which is bewildering to me.
Sure there are cases where this does not apply, but how about orphaning a package immediately when it has had an OOD flag for a while (maybe 2 weeks or a month) and an orphan request. This seems to me one of the most common and uncontroversial cases.
I should not have to wait another multiple weeks to update a PKGBUILD that has already been outdated for weeks.

Alternatively, when there is an appropriately old OOD flag, how about an option to immediately adopt the package?
Mostly I file orphan requests when I actually want to adopt the package, so currently I have to endure a months-long process:
- I add an OOD flag while updating the PKGBUILD for myself
- When I stumble upon it again weeks later, I file an orphan request
- When I notice that having been approved, I adopt and update the package

I think it is in our all best interest to simplify that process.

Offline

#2 2022-01-21 10:10:50

ayekat
Member
Registered: 2011-01-17
Posts: 1,589

Re: How about automating orphaning?

People appear to flag packages as out-of-date and file orphan requests for all sorts of nonsense reasons. I think it's perfectly reasonable for there to be at least one human in the orphaning process, I don't think that's really "bewildering".

how about orphaning a package immediately when it has had an OOD flag for a while (maybe 2 weeks or a month) and an orphan request.

There may be various reasons for why a package is outdated for a longer time. Maybe upstream has changed their build system in a way that requires significant downstream changes, and it may thus not be trivially updatable within a few weeks (especially if there are other dependencies that are also blocked—see e.g. the language ecosystems like Python that need a lot of coordination between the Arch maintainers everytime there's a new version).

I should not have to wait another multiple weeks to update a PKGBUILD that has already been outdated for weeks.

You can modify a PKGBUILD right away, though. Nobody forces you to push those changes back to the AUR immediately. Of course it's nice if efforts are centralised a bit, but the reality is often a bit more complicated.


pkgshackscfgblag

Offline

#3 2022-01-21 10:11:52

WorMzy
Forum Moderator
From: Scotland
Registered: 2010-06-16
Posts: 11,845
Website

Re: How about automating orphaning?

Current policy is orphan requests are immediately granted if the package has been flagged for 180 days: https://wiki.archlinux.org/title/AUR_su … s#Requests

If you want to discuss this with the people who actually have control over the policy (i.e. the Trusted Users) raise it on the aur-general mailing list -- very few TUs frequent the support forum.


Sakura:-
Mobo: MSI MAG X570S TORPEDO MAX // Processor: AMD Ryzen 9 5950X @4.9GHz // GFX: AMD Radeon RX 5700 XT // RAM: 32GB (4x 8GB) Corsair DDR4 (@ 3000MHz) // Storage: 1x 3TB HDD, 6x 1TB SSD, 2x 120GB SSD, 1x 275GB M2 SSD

Making lemonade from lemons since 2015.

Offline

#4 2022-01-21 14:02:09

Trilby
Inspector Parrot
Registered: 2011-11-29
Posts: 29,523
Website

Re: How about automating orphaning?

ayekat wrote:

People appear to flag packages as out-of-date and file orphan requests for all sorts of nonsense reasons.

This paired with the fact that there doesn't seem to be anyway to remove an improper OOD flag would highlight a real problem in the proposed approach.  A bad actor could readily take over any package with a false OOD flag and an orphan request.


"UNIX is simple and coherent..." - Dennis Ritchie, "GNU's Not UNIX" -  Richard Stallman

Offline

#5 2022-01-21 15:21:14

Scimmia
Fellow
Registered: 2012-09-01
Posts: 11,544

Re: How about automating orphaning?

The maintainer can unflag it

Offline

#6 2022-01-21 16:09:46

Trilby
Inspector Parrot
Registered: 2011-11-29
Posts: 29,523
Website

Re: How about automating orphaning?

Without having to "update" the PKGBUILD?  If so, I was unaware of that and I take back most of my previous point (still false OOD flags could still be a nuisance).


"UNIX is simple and coherent..." - Dennis Ritchie, "GNU's Not UNIX" -  Richard Stallman

Offline

#7 2022-01-23 13:35:42

Lone_Wolf
Member
From: Netherlands, Europe
Registered: 2005-10-04
Posts: 11,911

Re: How about automating orphaning?

Yes, maintainers (possibly co-maintainers also) can rest the ood-flag from the webinterface .

about annoying false ood flags :

Try maintaining a VCS package in AUR .
It's very rare that out-of-date is valid for a VCS package, but that has almost no effect on people submitting them.

I'm often tempted to file a request to remove the ability to ood a VCS completely from AUR but never have sofar.


Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.


(A works at time B)  && (time C > time B ) ≠  (A works at time C)

Offline

Board footer

Powered by FluxBB