Wrap selected public programs from conda/virtualenv environments
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.



public-wrappers provides a means to wrap selected public programs from conda and/or virtualenv environments, so they may be invoked directly by users without having to explicitly activate the environment. It makes use of exec-wrappers to do the wrapping.

public-wrappers identifies only certain programs from each environment as being public, by means of a configuration file. These public programs are wrapped, using exec-wrappers.

A number of environments may be wrapped, either into separate or a common wrappers directory.


A configuration file is required. The configuration file is the first of these which exist:

  • via the command line option --config
  • via the environment variable PUBLIC_WRAPPERS_CONFIG
  • ~/.public-wrappers.toml
  • /etc/public-wrappers.toml

The configuration file simply lists the conda and virtualenv environments along with the list of their public programs to be wrapped. Each environment may have its own destination directory for wrapper scripts, or these may be shared.

Programs must be unique within each wrapper directory, or globally unique if the configuration item globally-unique-wrappers is true.

Environment variables such as $HOME or user-relative pathnames such as ~ may be used in the configuration file.

See the example configuration file. Also, it is recommended to install a profile file into /etc/profile.d. See the example profile.sh.


For example, if ~/.public-wrappers.toml exists:


Otherwise, for example:

configure-public-wrappers -c ~/example-config.toml


public-wrappers is available on PyPI.

$ virtualenv ~/virtualenvs/public-wrappers
$ source ~/virtualenvs/public-wrappers/bin/activate
(public-wrappers)$ pip install public-wrappers
(public-wrappers)$ configure-public-wrappers --help

and also on conda-forge.

$ conda install public-wrappers
$ configure-public-wrappers --help