| Git Clone URL: | https://aur.archlinux.org/python-datalad.git (read-only, click to copy) |
|---|---|
| Package Base: | python-datalad |
| Description: | Keep code, data, containers under control with git and git-annex |
| Upstream URL: | https://github.com/datalad/datalad |
| Licenses: | MIT |
| Submitter: | liamtimms |
| Maintainer: | liamtimms (AKSoo, alerque) |
| Last Packager: | alerque |
| Votes: | 5 |
| Popularity: | 0.000019 |
| First Submitted: | 2021-05-02 19:57 (UTC) |
| Last Updated: | 2024-11-19 15:49 (UTC) |
Not sure what's going on with the SHAs, but they are different than the upstream releases. Maybe they change it without updating a minor version?
In any case case, for 1.0.2, this is the valid SHA256 '0f5f9a85c2f1cc03b7b1ffb83418b754972f7c3b9ac6f0bc1f7d4ec997538b82'
@alerque Your checksums are always different from what I (and several others judging from comments) get. How exactly are you getting your tarballs?
@AKSoo The original checksum was correct. If it changed that means upstream retagged the release, which should be considered suspicious. Do not just update to "fix" it until you know what happened and why (e.g. contact upstream, diff the tarballs, etc.) and leave a paper trail about it. Also update the pkgrel so that builds become reproducible again.
At October 04 2023, sha256 is not updated: pkgver=0.19.3 pkgrel=1
sha256sum results: 51ede60bf4bb4e4aa72c6aaf2f8b0ad5c988b3766f7a10c293668e43f153d582
But PKGBUILD shows: sha256sums=('460a1fa7e9997304ce34a1adc72eca3809e355182f5e40c66d9ca88514471aa3')
Note I added checks per packaging recommendations and they do pass (which is nice to know all the system dependencies do the job) but the test suite is very slow. If you want to avoid the time when installing this --nocheck is a reasonable thing to use for this PKGBUILD.
PSA: Like most of the PKGBUILDs that I (co-)maintain, I host prebuilt packages for this in my user repository and all its dependencies for those who wish to install it using pacman without messing around with building from the AUR. Issue reports or contributions are welcome either in comments below or via this GitHub repository.
Can we please overhaul this using PEP517 build tooling as described in the Arch Python package guidelines? If you don't know how or can't be bothered feel free to add me as a co-maintainer. For context I'm one of the maintainers on the git-annex package in [extra].
As of Mon Jul 25 07:03:30 PM CEST 2022, version 0.17.1-1 gives the following error when building:
==> Making package: python-datalad 0.17.1-1 (Mon 25 Jul 2022 07:01:21 PM CEST)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
-> Found datalad-0.17.1.tar.gz
==> Validating source files with sha512sums...
datalad-0.17.1.tar.gz ... FAILED
==> ERROR: One or more files did not pass the validity check!
-> error making: python-datalad
0.15.6-1 working on Arch, thanks for the speedy update!
Pinned Comments
alerque commented on 2023-06-29 12:58 (UTC)
PSA: Like most of the PKGBUILDs that I (co-)maintain, I host prebuilt packages for this in my user repository and all its dependencies for those who wish to install it using
pacmanwithout messing around with building from the AUR. Issue reports or contributions are welcome either in comments below or via this GitHub repository.