[PRQ#50396] Deletion Request for petsc-free+opts
MarsSeed [1] filed a deletion request for petsc-free+opts [2]: Badly configured PKGBUILD, fails execution. VCS package in disguise, with wrong pkgname. Also, please refrain from using the '+' sign in pkgname. While it is allowed by pacman, AURweb and AUR helpers had some issues with that character earlier. (It is very rarely used, so support for it is not much tested I guess.) If you want to maintain this package, please name it properly, like 'petsc-free-plus-opts-git', and also don't add a split package with a '-bin' suffix (petsc-doc-bin). Make the docs package reflect the pkgbase name. (Using the previous example: petsc-free-plus-opts-doc- git). [1] https://aur.archlinux.org/account/MarsSeed/ [2] https://aur.archlinux.org/pkgbase/petsc-free+opts/
On 2023-11-06 13:13, notify@aur.archlinux.org wrote:
MarsSeed [1] filed a deletion request for petsc-free+opts [2]:
Badly configured PKGBUILD, fails execution. VCS package in disguise, with wrong pkgname.
Also, please refrain from using the '+' sign in pkgname. While it is allowed by pacman, AURweb and AUR helpers had some issues with that character earlier. (It is very rarely used, so support for it is not much tested I guess.)
If you want to maintain this package, please name it properly, like 'petsc-free-plus-opts-git', and also don't add a split package with a '-bin' suffix (petsc-doc-bin). Make the docs package reflect the pkgbase name. (Using the previous example: petsc-free-plus-opts-doc- git).
[1] https://aur.archlinux.org/account/MarsSeed/ [2] https://aur.archlinux.org/pkgbase/petsc-free+opts/
First, you complained about a non-existing -git suffix (a suffix means: at the end, the last part). Now you complain of a VCS in disguise. It was *_you_* who said that this package cannot have a -git suffix (never did), and now, you are suggesting that I add one to the only part of the package that does *not* use Git sources? Please, RTFM: https://wiki.archlinux.org/title/Arch_package_guidelines#Package_naming Package names can contain only alphanumeric characters and any of @, ., _, +, -. Names are not allowed to start with hyphens or dots. All letters should be lowercase. 1. This package is not in the official repo 2. It provides extra features 3. It does include at least one of these "executable(s); configuration file(s); online or offline documentation for specific software or the Arch Linux distribution as a whole; media intended to be used directly by software." 4. The part building from source does not use prebuilt binaries (no -bin suffix) 5. "> Badly configured PKGBUILD, fails execution." May be you did something wrong? (it works for me). I don't think that this mailing list is the right place for such a complaint. 6. It does provide information about past and present contributors 7. The documentation does come with *prebuilt* binaries, unless you consider a 2.5 MB PDF plain-text. If you don't like my PKGBUILD, please, don't use it and leave me alone. If you want to involve in a constructive discussion, please be polite and do it in the right place.
Request #50396 has been Accepted by muflone [1]: [Autogenerated] Accepted deletion for petsc-free+opts. [1] https://aur.archlinux.org/account/muflone/
Is there a way to challenge a package deletion? (I did not see such an option.) There were no pending requests for this package other than the deletion request. In my opinion (I'm biased, of course), the request should have been removed, not followed. This package was being maintained by yours truly, as compared to--let's say petsc-git (which I abandoned) or pygtk-object-browser. The need for the package was clarified as a pinned comment and at the top of the PKGBUILD. It was not violating any license nor anything. It was a perfectly innocent package that was not harming anyone and could be useful. It's not like I actually care if the package is deleted. I'm just curious as to why these things happen. Thanks. On 2024-04-01 14:23, notify@aur.archlinux.org wrote:
Request #50396 has been Accepted by muflone [1]:
[Autogenerated] Accepted deletion for petsc-free+opts.
Hola
This package was being maintained by yours truly, as compared to--let's say petsc-git (which I abandoned) or pygtk-object-browser. The need for the package was clarified as a pinned comment and at the top of the PKGBUILD. It was not violating any license nor anything. It was a perfectly innocent package that was not harming anyone and could be useful.
The package petsc-free+opts was removed because it contained a couple of issues: 1) it was a duplicate of petsc-git 2) the name was plain wrong: it was a git package whose name didn't ended with -git 3) the name was meaningless: +opts is not a package classification and explains nothing about its nature 4) the documentation package was also meaningless: the pkgbase don't point to a "binary" package but when it builds it produces parts of prebuilt packages and parts of build packages. 5) the pkgver variable used in the makedepends is not valid and produces errors in the case hypre was not previously prebuilt 6) the whole PKGBUILD was very chaotic, with a lot of commented lines, disabled code, empty arrays, confused instructions Regards -- Fabio Castelli aka Muflone
participants (3)
-
edgar
-
Muflone
-
notify＠aur.archlinux.org