Skip to content
This repository has been archived by the owner on Jan 30, 2019. It is now read-only.
/ conda-env Public archive

Provides a unified interface to dealing with Conda environments.

License

Notifications You must be signed in to change notification settings

conda-archive/conda-env

Repository files navigation

This repository is deprecated. All code herein now rides alongside conda proper. https://github.com/conda/conda

conda-env

Provides the conda env interface to Conda environments.

Installing

To install conda env with conda, run the following command in your root environment:

$ conda install -c conda conda-env

Usage

All of the usage is documented via the --help flag.

$ conda env --help
usage: conda-env [-h] {create,export,list,remove} ...

positional arguments:
  {attach,create,export,list,remove,upload,update}
    attach              Embeds information describing your conda environment
                        into the notebook metadata
    create              Create an environment based on an environment file
    export              Export a given environment
    list                List the Conda environments
    remove              Remove an environment
    upload              Upload an environment to anaconda.org
    update              Updates the current environment based on environment
                        file

optional arguments:
  -h, --help            show this help message and exit

environment.yml

conda-env allows creating environments using the environment.yml specification file. This allows you to specify a name, channels to use when creating the environment, and the dependencies. For example, to create an environment named stats with numpy and pandas create an environment.yml file with this as the contents:

name: stats
dependencies:
  - numpy
  - pandas

Then run this from the command line:

$ conda env create
Fetching package metadata: ...
Solving package specifications: .Linking packages ...
[      COMPLETE      ] |#################################################| 100%
#
# To activate this environment, use:
# $ source activate numpy
#
# To deactivate this environment, use:
# $ source deactivate
#

Your output might vary a little bit depending on whether you have the packages in your local package cache.

You can explicitly provide an environment spec file using -f or --file and the name of the file you would like to use.

The default channels can be excluded by adding nodefaults to the list of channels. This is equivalent to passing the --override-channels option to most conda commands, and is like defaults in the .condarc channel configuration but with the reverse logic.

Environment file example

name: stats
channels:
  - javascript
dependencies:
  - python=3.4   # or 2.7 if you are feeling nostalgic
  - bokeh=0.9.2
  - numpy=1.9.*
  - nodejs=0.10.*
  - flask
  - pip:
    - Flask-Testing

Recommendation: Always create your environment.yml file by hand.