Integrating RStudio Connect with Python

These instructions describe how to install and integrate Python with RStudio Connect.

Once you configure Python with RStudio Connect, users will be able to publish Jupyter Notebooks or mixed R and Python content with Shiny apps, R Markdown reports, and Plumber APIs that call out to Python code using the reticulate package.

1. Install Python

Install Python, pip, and virtualenv on the RStudio Connect server following the steps to Install Python.

Warning

Due to breaking changes to the virtualenv package, Python installations must have a version of virtualenv below 20, e.g.: virtualenv<20. Incompatible versions will result in an error at startup.

Ensure that you install virtualenv in your Python environment by running the following command:

Terminal

$ sudo /opt/python/3.6.5/bin/pip install "virtualenv<20"

Info

Our recommended installation instructions for Python allow you to make multiple versions of Python available and avoid replacing existing versions of Python when updating system packages.

2. Install requirements for Python APIs and Dash applications

Python APIs and Dash apps require the libev development package to be installed:

$ sudo yum install libev-devel

$ sudo apt install libev-dev

$ sudo zypper install libev-devel

Info

Additional RPM repositories may be required to install libev-devel on CentOS 6.

3. Configure Python in RStudio Connect

Enable Python in RStudio Connect by adding a [Python] section to the RStudio Connect configuration file. This section should contain an Executable option that refers to the location of the Python executable.

For example, if you installed Python 3.6.5 on the RStudio Connect server, then your configuration file would appear as:

File: /etc/rstudio-connect/rstudio-connect.gcfg

[Python]
Enabled = true
Executable = /opt/python/3.6.5/bin/python

(Optional) Install additional versions of Python

To make additional versions of Python available alongside existing versions, repeat the above steps to install a different version of Python to other locations.

Then, add a new line to the RStudio Connect configuration file with the Executable option for each additional version of Python that you installed.

For example, if you installed Python 2.7.16, 3.6.5, and 3.7.3 on the RStudio Connect server, then your configuration file would appear as:

File: /etc/rstudio-connect/rstudio-connect.gcfg

[Python]
Enabled = true
Executable = /opt/python/2.7.16/bin/python
Executable = /opt/python/3.6.5/bin/python
Executable = /opt/python/3.7.3/bin/python

Additional Information

For more information on best practices for end users when working with Python and RStudio Connect, refer to the support article on Best Practices for using Python with RStudio Connect.