May 2025 - Arch-dev-public - lists.archlinux.org

Switch to a versioned packaging scheme for wlroots
by Robin Candau 31 May '25

31 May '25

Hi, Wlroots is one of the packages for which we need to keep multiple versions around to ensure compatibility with reverse dependencies. At the moment, we have a wlroots0.17 package [1] (for the few packages that still require it, most notably qtile [2]) and a "regular" wlroots package [3] that currently points to 0.18. We are evaluating bumping the wlroots package to 0.19, but that requires the introduction of a new wlroots0.18 package to preserve compatibility with reverse dependencies that are not yet compatible with 0.19 (similar to what was done with the wlroots0.17 package). This method makes new wlroots releases difficult to triage as we have to release both the wlroots package bump and the new wlroots0.18 package at the same time (to avoid breaking changes regarding reverse dependencies) while ensuring that we defined proper "conflicts" to avoid conflicting files issues during upgrade [4] (since we can't control in which order packages will be upgraded on users' system). For instance, the new wlroots0.18 package would likely need a "conflicts" array that looks like this; in order to avoid files collision with the "regular" wlroots package until it gets updated to 0.19 on users' system, and also to avoid an unexpected conflict with the wlroots0.17 package. ``` conflicts=( "wlroots<0.19" "wlroots>=0.18" ) ``` While the above _should_ work, this is a quirky dance that will eventually get more and more complex over time (e.g. if we have to accumulate more versions of wlroots in the future). As such, I would like to switch to a versioned packaging scheme for wlroots. This means dropping the "regular" wlroots package and solely rely on versioned packages like: wlroots0.17, wlroots0.18 & wlroots0.19 (similar to e.g. gtk2, gtk3, gtk4 in a sense). Wlroots already adopted a versioned so/lib naming scheme so there is no file collision between e.g. wlroots0.18 and wlroots0.19 (meaning we won't need the above "conflicts" array dance anymore). Technically speaking, that means introducing a new wlroots0.18 package that conflicts and replaces the current "regular" wlroots package (to ensure people make the switch without file conflicting issue) and a new wlroots0.19 package. We could then drop the "regular" wlroots package from the repo and keep moving with versioned packages only in the future (e.g. wlroots0.20, wlroots0.21, and so on...). Reverse dependencies can then explicitly point to the version they need (similar to how it's done currently with wlroots0.17). I'm aware that this may be unconventional regarding how we usually do things in Arch Linux, but I think it's fair to say that current way we have to deal with new wlroots releases is confusing and causes more harm than good. I think the above suggestion is a practical and relevant solution for such specific case. I created a related task todo list [5] that I will tackle today and put to staging. I intend to leave it in staging for a week or two in order to leave some time for people to react and / or raise any concerns I may have missed about switching to a versioned packaging scheme. Thanks :) [1] https://archlinux.org/packages/extra/x86_64/wlroots0.17/ [2] https://github.com/qtile/qtile/blob/master/docs/manual/wayland.rst?plain=1#… [3] https://archlinux.org/packages/extra/x86_64/wlroots/ [4] https://gitlab.archlinux.org/archlinux/packaging/packages/wlroots0.17/-/iss… [5] https://archlinux.org/todo/switch-to-a-versioned-packaging-scheme-for-wlroo… -- Regards, Robin Candau / Antiz

2 4

Amending RFC40 to remove custom 0BSD license
by Morten Linderud 27 May '25

27 May '25

Yo, Today I noticed that the "License package sources" RFC contained an amended 0BSD license that added a two paragraph exception for patch files and other auxiliary files. The purpose of this change is to ensure the license is not covering other files in the repository that the author can't license from the upstream. See: https://rfc.archlinux.page/0040-license-package-sources/ While this is a practical problem that needs to be solved, we should not be doing that through additional text in a FSF- and OSI approved license. This essentially makes it a custom license that is not really going to detected as 0BSD from external sources, and runs against the original goal of removing legal uncertainty. As the change, and by extension the problem itself, is not mentioned in the text it came as a surprise to me that it was done. What I think is more proper is to remove these two lines from the proposed license file, and move this to a separate RFC that would cover a use of the REUSE specification, or SPDX license identifiers. This would serve the same purpose as the Debian `copyright` files, while also being standardized. I have written a proposed amendment to the text that I hope people find okay. https://gitlab.archlinux.org/archlinux/rfcs/-/merge_requests/49 *Please note that any licenses already added to the repository needs to be amended.* My goal is to write up a RFC for the REUSE/SPDX part of this before the current 3 month timeline where we'll start adding licenses to ensure we don't prolong the process. If people are curious how this would look like, I annotated the `usd` package as an example. https://gitlab.archlinux.org/foxboron/usd/-/tree/morten/reuse See the spec for more details: https://reuse.software/spec-3.2/ Cheers! -- Morten Linderud PGP: 9C02FF419FECBE16

6 11

pulling mariadb-lts from extra-testing
by Christian Hesse 23 May '25

23 May '25

Hello everbody, I have just pulled mariadb-lts (11.4.6-1) from extra-testing. Several upstream bug reports indicate that there is a serious performance regression, though the exact cause it not yet known. Another release is expected really soon, possibly even in the following week. -- main(a){char*c=/* Schoene Gruesse */"B?IJj;MEH" "CX:;",b;for(a/* Best regards my address: */=0;b=c[a++];) putchar(b-1/(/* Chris cc -ox -xc - && ./x */b/42*2-3)*42);}

1 2

IRC Channel for mirror related discussions now available
by pitastrudl 19 May '25

19 May '25

Hi all, We've created, some time ago already, a new channel on libera.chat for any Arch Linux mirror / mirrorlist related questions and chat. This is just another communication channel for all things related. It is still advised to report any issues to https://gitlab.archlinux.org/archlinux/arch-mirrors/-/issues for easier visibility and tracking of progress. Thank you for all your contributions, past and future! Channel name: #archlinux-mirrors Regards, Arun

1 0

Docker Open Source Community Program renewed for 2025
by Levente Polyak 13 May '25

13 May '25

Hi all, I've prolonged our Docker Open Source Community Program for another year for 2025. We've also received the approval and confirmation of the application. One of the benefits is, that this allows for our images to have relaxed pull limits. This is a yearly renewal. The next renewal will be popping up around March 2026. https://hub.docker.com/repository/docker/archlinux/archlinux Cheers, Levente

1 0

On dropping Redis
by Bert Peters 13 May '25

13 May '25

Hi all, A while ago we as a distro decided to stop supporting Redis, due to their licence change [1], and move to Valkey. Through a combined effort, we removed all direct dependencies on redis, replacing them with vault and patching until that worked. The announcement was posted, and a deadline was set. This may or may not have caused Redis to reconsider their license change, and have announced another relicencing, this time to the AGPL [2] [3]. With that change, I personally believe there is no longer a reason to remove redis from [extra], and keep it in the repos as-is. Redis is almost but not quite compatible with Valkey, so dropping it without good cause would be a disservice to our community. Now, I don't want to make light of the harm that Redis inc initially wrought on the open source community with their license change, or waste the work that was done to make everything work with Valkey. Pushback like this is what caused the license change. As such, I propose we continue to use Valkey as the implementation for all purposes that don't strictly require Redis, and maintain Redis simply as a package for our commynity's convenience. That way, should the licensing change again in the future, we do not have a similar amount of work ahead of us. This seems to me the Arch way: pragmatic and user central. Now I know that this is not a universally shared opinion, so please consider this email an invitation for discussion on what we should be doing here. Cheers, Bert [1]: https://lists.archlinux.org/archives/list/arch-dev-public@lists.archlinux.o… [2]: https://antirez.com/news/151 [3]: https://redis.io/blog/agplv3/

8 10

RFC: Introducing a process for sponsorships
by gromit 06 May '25

06 May '25

Hello everyone, A new RFC (request for comment) has been opened here: https://gitlab.archlinux.org/archlinux/rfcs/-/merge_requests/55 Please visit the above link for discussion. Summary: This RFC introduces a process which gives sponsorships a clear point of contact, introduces fitting measures of transparency and specifies a set of guidelines for such engagements with external entities.

1 0

Sunsetting epapirus-icon-theme
by Felix Yan 05 May '25

05 May '25

Hi, As per upstream suggested in the release note [1], epapirus-icon-theme is no longer provided in [extra] now. Users are suggested to use Papirus / Papirus-Dark from now on. [1] https://github.com/PapirusDevelopmentTeam/papirus-icon-theme/releases/tag/2… -- Regards, Felix Yan

1 0

RFC: Establish default build flags for Fortran
by Jakub Klinkovský 04 May '25

04 May '25

A new RFC (request for comment) has been opened here: https://gitlab.archlinux.org/archlinux/rfcs/-/merge_requests/54 Please visit the above link for discussion. Summary: This RFC establishes default build flags (`FFLAGS`, `FCFLAGS`, `DEBUG_FFLAGS`) for compiling Fortran code in Arch Linux packages, aligning them with the security hardening standards defined for C.

1 0