jankatins/jupyter_environment_kernels

An Jupyter plugin to enable the automatic detection of conda environments as kernels.

Automatic Environment Kernel Detection for Jupyter

A Jupyter plugin to enable the automatic detection of environments as kernels.

This plugin looks in the directories you specify for installed environments which have Jupyter installed and lists them as kernels for Jupyter to find. This makes it easy to run one notebook instance and access kernels with access to different versions of Python or different modules seamlessly.

Installation

The plugin can be installed with:

pip install environment_kernels

To enable the plugin add the following line to your notebook config file:

c.NotebookApp.kernel_spec_manager_class = 'environment_kernels.EnvironmentKernelSpecManager'

to create a config file run:

jupyter notebook --generate-config

or run the notebook with the following argument:

--NotebookApp.kernel_spec_manager_class='environment_kernels.EnvironmentKernelSpecManager'

Search Directories for Environments

The plugin works by searching a set of directories for subdriectories which are environments with the following pattern on Linux and OS/X:

BASE_DIR/ENV_NAME/bin/ipython

and on Windows::

BASE_DIR\ENV_NAME\Scripts\ipython

The default base directories are ~/.conda/envs, ~/.virtualenvs and if the jupyter notebook is being run in the root environment conda.config.envs_dirs will be imported added to the search path, if the notebook server is run from inside a conda environment then the CONDA_ENV_DIR variable will be set and the section of the path before /env/ will be added to the search list. You can add extra directories to the search path by using the extra_env_dirs config:

c.EnvironmentKernelSpecManager.extra_env_dirs=['/opt/miniconda/envs/']

or all the automatic behavior can be overriddenby setting the env_dirs config:

c.EnvironmentKernelSpecManager.env_dirs=['/opt/miniconda/envs/']

Limiting Environments

If you want to you can also make it ignore environments with certain names:

c.EnvironmentKernelSpecManager.blacklist_envs=['testenv']

or:

--EnvironmentKernelSpecManager.blacklist_envs="['testenv']"

Or you can specify a whitelist of "allowed" environments with:

c.EnvironmentKernelSpecManager.whitelist_envs=['testenv']

or:

--EnvironmentKernelSpecManager.whitelist_envs="['testenv']"