Skip to main content

Installation script for Datalad and related components

Project description

CI Status https://codecov.io/gh/datalad/datalad-installer/branch/master/graph/badge.svg https://img.shields.io/pypi/pyversions/datalad-installer.svg MIT License

GitHub | PyPI | Issues | Changelog

datalad-installer is a script for installing Datalad, git-annex, and related components all in a single invocation. It requires no third-party Python libraries, though it does make heavy use of external packaging commands.

Installation

datalad-installer requires Python 3.6 or higher. Just use pip for Python 3 (You have pip, right?) to install it:

python3 -m pip install datalad-installer

Alternatively, download the latest version directly from <https://raw.githubusercontent.com/datalad/datalad-installer/master/src/datalad_installer.py>.

Usage

datalad-installer [<global options>] <component>[=<version>] [<options>] <component>[=<version>] [<options>] ...

datalad-installer provisions one or more components listed on the command line. Each component is either a software package (i.e., Datalad or git-annex) or an environment in which software packages can be installed. If no components are specified on the command line, the script defaults to installing the datalad component.

Global Options

-E FILE, --env-write-file FILE

Append any PATH modifications or other shell commands needed to use the new components to the given file. This option can be specified multiple times. If this option is not given, the data is written to a temporary file whose location is logged at the beginning of the program.

-l LEVEL, --log-level LEVEL

Set the log level to the given value. Possible values are “CRITICAL”, “ERROR”, “WARNING”, “INFO”, “DEBUG” (all case-insensitive) and their Python integer equivalents. [default value: INFO]

--sudo <ask|error|ok>

What to do when the script needs to run a command with sudo or privilege escalation: ask for confirmation (default), error, or run without confirmation. This is always “ok” on Windows, where the system always asks for confirmation.

-V, --version

Display the script version and exit

-h, --help

Display usage information and exit

Components

venv

Creates a Python virtual environment using python -m venv. Subsequent datalad components on the command line will be installed into this virtual environment by default if not overridden by an intervening componnent.

Options
-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to python -m venv

--path PATH

Create the virtual environment at PATH. If not specified, the environment will be created in a directory in $TMPDIR.

miniconda

Installs the latest version of Miniconda. Subsequent conda-env components on the command line will use this installation, and subsequent datalad and git-annex components will be installed using this conda by default if not overridden by an intervening component.

The Miniconda installation script is downloaded from $ANACONDA_URL/Miniconda3-latest-$OS-x86_64.{sh,exe}, where $ANACONDA_URL is taken from the environment, defaulting to https://repo.anaconda.com/miniconda.

Options
--batch

Run the Miniconda installation script in batch (noninteractive) mode. This option is always in effect when installing on Windows.

-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the Miniconda installation script.

--path PATH

Install Miniconda at PATH. If not specified, it will be installed in a directory in $TMPDIR.

--spec SPEC

Space-separated specifiers for packages to install in the Conda base environment after provisioning.

conda-env

Creates a Conda environment. If there is no preceding miniconda component on the command line, Conda must already be installed on the system, and this installation will be used to create the environment.

Subsequent datalad and git-annex components will be installed into this environment by default if not overridden by an intervening component.

Options
-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the conda create command.

-n NAME, --name NAME

The name for the new environment. If not specified, a random name will be generated.

--spec SPEC

Space-separated specifiers for packages to install in the new environment.

neurodebian

Installs & configures NeuroDebian.

Options
-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the nd-configurerepo command.

git-annex

Installs git-annex. The component takes an -m, --method option specifying the installation method to use; the supported methods are:

  • apt

  • autobuild

  • brew

  • conda (only supported on Linux)

  • datalad/git-annex:tested

  • datalad/git-annex

  • datalad/packages

  • deb-url

  • dmg

  • neurodebian

  • snapshot

If no method is specified, or if the method is set to “auto”, then the most recent component on the command line that provides a compatible installation method will be used. If there is no such component, the first supported component from the following list will be used:

  • conda

  • apt

  • neurodebian

  • brew

  • autobuild

  • datalad/packages

A specific version to install can be specified for those methods that support it by suffixing “git-annex” with “=” and the version number on the command line.

The git-annex component also accepts all options for the supported installation methods; options not belonging to whichever method ends up used will be ignored.

datalad

Installs Datalad. The component takes an -m, --method option specifying the installation method to use; the supported methods are:

  • apt

  • brew

  • conda

  • deb-url

  • pip

If no method is specified, or if the method is set to “auto”, then the most recent component on the command line that provides a compatible installation method will be used. If there is no such component, the first supported component from the following list will be used:

  • conda

  • apt

  • neurodebian

  • brew

  • autobuild

  • datalad/packages

A specific version to install can be specified for those methods that support it by suffixing “datalad” with “=” and the version number on the command line.

The datalad component also accepts all options for the supported installation methods; options not belonging to whichever method ends up used will be ignored.

Installation Methods

apt

Install with sudo apt-get install. Supports installing specific versions.

Options
--build-dep

Run sudo apt-get build-dep instead of sudo apt-get install.

-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the installation command.

autobuild

Downloads & installs the latest official build of git-annex from kitenet.net. Does not support installing specific versions.

This installation method is only supported on Linux and macOS.

brew

Install with brew (Homebrew). Does not support installing specific versions.

Options
-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the installation command.

conda

Install with conda install. Supports installing specific versions.

Options
-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the installation command.

datalad/git-annex:tested

Downloads & installs the artifact from the latest successful build of datalad/git-annex for the running OS. Does not support installing specific versions.

This installation method requires a GitHub OAuth token with appropriate permissions. It must be specified either via the GITHUB_TOKEN environment variable or as the value of the hub.oauthtoken Git config option.

Options
--install-dir DIR

Directory in which to unpack the *.deb package instead of installing it system-wide. If this contains the string {tmpdir}, it will be replaced with the path to a directory in $TMPDIR. (Linux only)

datalad/git-annex

Downloads & installs the artifact from the latest build of datalad/git-annex that produced artifacts for the running OS. Does not support installing specific versions.

This installation method requires a GitHub OAuth token with appropriate permissions. It must be specified either via the GITHUB_TOKEN environment variable or as the value of the hub.oauthtoken Git config option.

Options
--install-dir DIR

Directory in which to unpack the *.deb package instead of installing it system-wide. If this contains the string {tmpdir}, it will be replaced with the path to a directory in $TMPDIR. (Linux only)

datalad/packages

Downloads & installs the artifact from <https://datasets.datalad.org/?dir=/datalad/packages> for the running OS. Supports installing specific versions (though note that the version strings for this method tend to include Git commit information, e.g., “8.20210127+git111-gbe5a0e4b8”).

This installation method is only supported on Windows.

deb-url

Download & install a given *.deb package. Does not support installing specific versions.

Options
-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the installation command.

--install-dir DIR

Directory in which to unpack the *.deb package instead of installing it system-wide If this contains the string {tmpdir}, it will be replaced with the path to a directory in $TMPDIR. If this contains the string {version}, it will be replaced with the package’s version. (git-annex only)

--url URL

Specify the URL of the *.deb package. This option is required for this installation method.

dmg

Install git-annex to the /Applications directory from a properly-built *.dmg image. Does not support installing specific versions.

This installation method is only supported on macOS.

Options
--path PATH

Specify the path to the *.dmg image. This option is required for this installation method.

neurodebian

Install from NeuroDebian repositories with sudo apt-get install. Supports installing specific versions.

Options
--build-dep

Run sudo apt-get build-dep instead of sudo apt-get install.

-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the installation command.

pip

Install with python -m pip. Supports installing specific versions.

If a venv component is previously given on the command line, the installation will be performed in that virtual environment; otherwise, it will be performed using the same Python used to run datalad-installer.

Options
--devel

Install the given component from its GitHub repository instead of from PyPI.

-e ARGS, --extra-args ARGS

Specify extra command-line arguments to pass to the installation command.

-E EXTRAS, --extras EXTRAS

Specify (comma-separated) package extras to install.

snapshot

Downloads & installs the latest official snapshot build of git-annex from kitenet.net. Does not support installing specific versions.

This installation method is only supported on Linux and macOS.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

datalad-installer-0.5.1.tar.gz (29.8 kB view hashes)

Uploaded Source

Built Distribution

datalad_installer-0.5.1-py3-none-any.whl (22.6 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page