Skip to main content

Python wheel package for git-annex

Test git-annex wheel from PyPi

Why?

Git-annex is written in Haskell and plenty of installation methods are available. However, for deploying git-annex as a dependency of a Python library/application, like DataLad or AnnexRemote, system packages are a lot less flexible than Python's virtual environments, and other methods are more complex and fragile.

With git-annex being available from PyPi, versioned dependencies and deployment in application-specific environments are possible via standard means of Python packaging.

Caveats

A standard git-annex deployment is (primarily) a single binary (git-annex), and a bunch of symlinks that make this one binary fulfilled multiple roles (git-annex-shell, git-remote-annex, etc.). A Python wheel, however, is a ZIP file container with no support for symlinks, and also no support for "post-install" scripts.

In order to square this circle, the git-annex binary is wrapped via regular Python entrypoint scripts that handle calling git-annex as necessary. This delivers a cross-platform compatible wheel, but at a start-up cost (~30ms vs ~11ms on my laptop).

Installation

Get the package from PyPi, and install like any other package from PyPi.

uv users can deploy git-annex in a dedicated virtual environment via the one-liner:

uv tool install git-annex

git-annex build configuration

Git-annex is built with libmagic support.

Platform notes

Linux

The manylinux wheel is self-contained and includes copies of all libraries. It only depends on the declared GLIBC versions. The magic.mgc database is not included, and is assumed to be available on the target system. Install it separately, if needed (e.g., libmagic-mgc package).

Windows

The wheel is self-contained and includes a copy of libmagic and the magic.mgc database.

Mac

The wheel is self-contained and includes a copy of libmagic and the magic.mgc database.

Developer information

Sources

The sources for this package are available at https://github.com/psychoinformatics-de/git-annex-wheel

The repository tracks the git-annex sources as a Git submodule.

Issues

For issues related to the Python wheel packaging of git-annex, please use the tracker at https://github.com/psychoinformatics-de/git-annex-wheel/issues

How to update for a new git-annex release?

Advance the submodule ./git-annex to the new release tag.

Now adjust the package version in pyproject.toml accordingly. This version must follow the rules for Python packages.

The included (GitHub) action workflows will build a corresponding wheel and upload it to PyPi.

Acknowledgements

This work was funded, in part, by

  • Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under grant TRR 379 (546006540, Q02 project)
  • Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under grant SFB 1451 (431549029, INF project)
  • MKW-NRW: Ministerium für Kultur und Wissenschaft des Landes Nordrhein-Westfalen under the Kooperationsplattformen 2022 program, grant number: KP22-106A

Release files for git-annex 10.20250605

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distributions (wheels)

Table of built distributions (wheels) for git-annex 10.20250605
File
git_annex-10.20250605-py3-none-win_amd64.whl Python 3 none Windows x86-64 Details
git_annex-10.20250605-py3-none-manylinux_2_34_x86_64.whl Python 3 none Linux glibc 2.34+ x86-64 Details
git_annex-10.20250605-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl Python 3 none Linux glibc 2.17+ x86-64 Details
git_annex-10.20250605-py3-none-macosx_14_0_arm64.whl Python 3 none macOS 14.0+ ARM64 Details

Total release size: 108.2 MB

Release files / git_annex-10.20250605-py3-none-win_amd64.whl

Download URL git_annex-10.20250605-py3-none-win_amd64.whl
Size 26.8 MB
Tags Python 3 Windows x86-64
SHA-256 checksum
How to use checksums
397e828179808f5f030aad881726b232c5f8e1361e51df99251ec762d24e0949
BLAKE2b-256 checksum
How to use checksums
e79f55c0036f21570b2ccb659cc88cda61b5ef6ccd6c80ed56938251c5579900
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.15

Release files / git_annex-10.20250605-py3-none-manylinux_2_34_x86_64.whl

Download URL git_annex-10.20250605-py3-none-manylinux_2_34_x86_64.whl
Size 21.3 MB
Tags Linux glibc 2.34+ x86-64 Python 3
SHA-256 checksum
How to use checksums
19511d9de7dc27aae6a47a64ddbd253aa30c5ef8b09b9d79aecb3c1908b111ce
BLAKE2b-256 checksum
How to use checksums
527952f345427278f7589fd741ef4b7d92e5ab341161fca150cbc8a1566af5d2
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.15

Release files / git_annex-10.20250605-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Download URL git_annex-10.20250605-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Size 24.1 MB
Tags Linux glibc 2.17+ x86-64 Python 3
SHA-256 checksum
How to use checksums
be34127973123bc4a9a93f5007ad7323fd01c7b884bab5120eadc92fd4cb33bd
BLAKE2b-256 checksum
How to use checksums
908ac270cdb48223fc3b307f3ac6539723f02612ff8b2c2ee8db9793a17e100e
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.15

Release files / git_annex-10.20250605-py3-none-macosx_14_0_arm64.whl

Download URL git_annex-10.20250605-py3-none-macosx_14_0_arm64.whl
Size 36.0 MB
Tags Python 3 macOS 14.0+ ARM64
SHA-256 checksum
How to use checksums
a64521e3c771b5e15f686fd2670eb7ccc8b06519b46b4bbe2f0ad7f75c16fbfc
BLAKE2b-256 checksum
How to use checksums
12f6a1cc9648b564c9f56a2909495e5f92c4a87b62c69e1bccc79ffc0ec58607
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.15
Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page