SASpy

This page is dedicated to Tom Boardman, whose passion for SAS and knowledge sharing continues to inspire.

SASpy

This page demonstrates how to connect Python to SAS using the SASPy library. We will show how to configure the connection, establish a session, and submit SAS code from Python.

The following packages are required:

Package Comment
saspy Main Python library for connecting to SAS.
pandas For handling dataframes (optional, but common).

Below is an example SASPy configuration block:

SASPy Configuration Example 


# The names of the SAS configurations to use
SAS_config_names = ["iomcom"]

# Configuration block for the IOM connection
iomcom = {
    "iomhost": "your_sas_server_name_or_ip",
    "iomport": 8591,
    "iomuser": "your_username",
    "iompwd": "your_password",
    "provider": "sas.iomprovider"
}

# Configuration options applied globally to the SAS connection
SAS_config_options = {
    "lock_down": True,
    "verbose": True
}

After defining parameters, you can establish a connection and submit SAS code. Here is a class-based example:

Python SAS Session Example 


import saspy

class SASSession:
    def __init__(self):
        """Initializes the SAS session using the predefined configuration."""
        self.sas = None
        self._initialize_session()

    def _initialize_session(self):
        try:
            self.sas = saspy.SASsession()
            print("SAS session initialized successfully.")
        except Exception as e:
            print(f"Error initializing SAS session: {e}")
            self.sas = None

    def submit(self, sas_code):
        """Submits SAS code to the session and returns the result."""
        if self.sas:
            result = self.sas.submit(sas_code)
            sas_log = result.get('LOG', 'No LOG output')
            sas_lst = result.get('LST', 'No LST output')
            return {'LOG': sas_log, 'LST': sas_lst}
        else:
            print("SAS session is not initialized.")
            return None

    def close(self):
        """Closes the SAS session."""
        if self.sas:
            self.sas._endsas()
            print("SAS session closed.")
        else:
            print("SAS session is not open.")

# Example usage
if __name__ == "__main__":
    sas_session = SASSession()
    result = sas_session.submit("proc print data=sashelp.class; run;")
    if result:
        print("SAS LOG:", result['LOG'])
        print("SAS OUTPUT (LST):", result['LST'])
    sas_session.close()

For more details, see the SASPy documentation.

This guide walks you through creating a lightweight Conda environment with Python 3.12.3 on Ubuntu 24, suitable for low-RAM laptops and basic data analysis tasks with Microsoft-friendly packages.

Conda Environment File
name: light-data-env
channels:
  - defaults
  - conda-forge
dependencies:
  - python=3.13.5
  - pandas
  - numpy
  - matplotlib
  - seaborn
  - openpyxl
  - xlrd
  - requests
  - jupyterlab
  - statsmodels
  - scikit-learn
  - pip
  - pip:
      - pywin32
      - selenium
Step 1: Install Miniconda
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh
Step 2: Create the Environment
conda env create -f light-data-env.yml
Step 3: Activate the Environment
conda activate light-data-env
Step 4: Launch Jupyter Lab
jupyter lab
Notes
  • Python 3.12.3 is stable with all included packages as of mid-2024.
  • xlwings on Linux supports Excel file handling but not live Excel application control (Windows-only feature).
Tip: For an even lighter setup, you can:
  • Replace jupyterlab with notebook for a faster interface.
  • Remove scikit-learn if machine learning is not required.
  • Omit xlwings if you do not need advanced Excel automation.

This guide helps you install essential VS Code extensions for Python data analysis, Conda environment management, and Microsoft-friendly workflows.

Essential Extensions
  • Python - ms-python.python
  • Jupyter - ms-toolsai.jupyter
  • Pylance - ms-python.vscode-pylance
  • Conda - ms-python.conda
  • Jupyter Keymap - ms-toolsai.jupyter-keymap
  • Jupyter Renderers - ms-toolsai.jupyter-renderers
  • Code Runner - formulahendry.code-runner
  • Path Intellisense - christian-kohler.path-intellisense
  • Bracket Pair Colorizer 2 - coenraads.bracket-pair-colorizer-2
  • Excel Viewer - GrapeCity.gc-excelviewer
  • GitLens - eamodio.gitlens
Bash Script for Quick Installation

#!/bin/bash
# VS Code Essential Extensions Installer for Windows (PowerShell)

Write-Output "Installing VS Code Extensions..."

# Python Core
code --install-extension ms-python.python
code --install-extension ms-toolsai.jupyter
code --install-extension ms-python.vscode-pylance

# Jupyter Helpers
code --install-extension ms-toolsai.jupyter-keymap
code --install-extension ms-toolsai.jupyter-renderers

# Productivity Tools
code --install-extension formulahendry.code-runner
code --install-extension christian-kohler.path-intellisense

# Data Science Extensions
code --install-extension ms-toolsai.vscode-jupyter-slideshow
code --install-extension ms-python.anaconda-extension-pack
code --install-extension njpwerner.autodocstring
code --install-extension kevinrose.vsc-python-indent
code --install-extension msrvida.vscode-sanddance
code --install-extension ms-toolsai.datawrangler
code --install-extension ms-toolsai.jupyter-renderers
code --install-extension ms-toolsai.jupyter-keymap

# PHP Extensions
code --install-extension felixfbecker.php-debug
code --install-extension felixfbecker.php-intellisense
code --install-extension bmewburn.vscode-intelephense-client
code --install-extension mehedidracula.php-namespace-resolver
code --install-extension neilbrayfield.php-docblocker

# General Web Development
code --install-extension esbenp.prettier-vscode
code --install-extension dbaeumer.vscode-eslint

# Version Control
code --install-extension eamodio.gitlens

Write-Output "VS Code Extensions installed successfully."
Pause

# Version Control
code --install-extension eamodio.gitlens

Write-Output "VS Code Extensions installed successfully."
Pause
Usage
Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass
    .\vscode-setup.ps1
Tip: Ensure the code command is available in your terminal. In VS Code, press Ctrl+Shift+P and run Shell Command: Install 'code' command in PATH if it's not already set.