Skip to content

Latest commit

 

History

History
91 lines (68 loc) · 3.25 KB

README.rst

File metadata and controls

91 lines (68 loc) · 3.25 KB

SAS Kernel for Jupyter

What is this?

A SAS Kernel for Jupyter Notebooks

Dependencies

  • Python3.X
  • Jupyter
  • SAS 9.4 or higher – This includes SAS Viya
  • Linux OS

With the latest changes in saspy it is no longer a requirement that Jupyter and SAS be installed on the same machine. SAS and Jupyter can now communicate via passwordless ssh. This is in response to issue 11. The configuration details are located in sascfg.py

Documentation

Here is the link to the current documentation https://sassoftware.github.io/sas_kernel/

Install

To successfully use the SAS Kernel you must have each of the following: * SAS version 9.4 or above * Jupyter * Jupyter has a number of dependencies. See the subsections for steps on installing Jupyter on your system. * Python 3

Install for Anaconda Python (assuming SAS already installed)

  1. Download and install Anaconda Python (make sure you get Python3.X). If you install Anaconda without super user rights (root or sudo) then other users on the system will not be able to access the SAS kernel. A couple notes that I’ve observed:
  • The default install location is the users home directory. This is fine for a single user install I would put it in a common location (/opt) if you’re doing a system wide install
  • One of the prompts is to add the path to your environment. I recommend you want to answer ‘yes’ to that question so that all the installing user has the executables in their path. If you’re doing a system wide install (root or sudo) all the other users should add that path to their environmental variables

  1. Install sas_kernel. The sas_kernel has a dependency on saspy which is located here. In the command below I’m assuming that pip maps to python3 if that is not the case the you might need to use pip3 instead. pip install sas_kernel

  2. Verify that the sas_kernel is installed jupyter kernelspec list

    If you installed as a superuser, your output should similar to this:

    Available kernels:
  python3    /opt/Anaconda3-2.5.0/lib/python3.5/site-packages/ipykernel/resources
  sas        /usr/local/share/jupyter/kernels/sas

    If you installed as a regular user (sas in this case), your output should similar to this:

    Available kernels:
  python3    /home/sas/anaconda3/lib/python3.5/site-packages/ipykernel/resources
  sas        /home/sas/.local/share/jupyter/kernels/sas

  3. Verify SAS Executable is correct

    1. find the sascfg.py file – it is currently located in the install location (see above) `[install location]/site-