This repository contains scripts for collecting and analyzing data about open source activity associated with a university or institution.
The vision for this repository is to build out a set of Python scripts that can effectively be utilized by open source program offices to gain a better understanding of their specific institution. See details below about specific tools.
This script is designed to gather data about GitHub accounts which mention the specified institution in the "bio" statement associated with the account and save that information in a CSV (simple-github-account-url-list-[year]-[month]-[day]-[institutionname].csv). In addition to gathering summary information about each account, it has been developed with a specific focus on gathering data about the open source activity of university research communities and uses the provided "bio" information associated with each account to make a prediction about type of affiliation with the defined university. The script is also able to gather information about the individual GitHub repositories under each account and saves information about those repositories to a separate CSV file (simple-github-repo-url-list-[year]-[month]-[day]-[institutionname].csv). In order for the script to be used successfully, important parameter information must be defined in the repository's .env file. See section below for more details about preparing the .env file.
This script creates visualizations of the CSV formatted data that is collected using github-activity-metrics-tool.py. The path to the GitHub account information CSV and GitHub repo CSV needs to be defined in the repositories .env file for the script to run successfully.
Once this repo is cloned locally, the template.env file should be renamed to just .env and the contents of the file should be edited to replace the example values that are provided in the file by default with the correct values based on the institution for which the script will be run. Take care to preserve the JSON formatting of the .env file to ensure proper functioning of the Python scripts in the repository which depend on the parameters defined in the .env file.
- githubtoken: The personalized GitHub token generated by the user of the script should be supplied here
- institutionname: The standard institution name - this will be used to name files and manage other script processes
- institutionnamepermutations: A list of all of the variations of the institution name that GitHub users might mention in their account bio statement
- institutioncity: The city that the institution is located in
- institutionemaildomain: The email address extension for the institution (e.g. "utexas.edu")
- githubaccountdetailscsvpath: Path to CSV file containing GitHub account results
- githubrepodetailscsvpath: Path to CSV file containing GitHub repository results
- resultsperpage: Number of GitHub results per page to return
- pagelimit: Number of pages of GitHub results to return
- minimumfollowers: Minimum number of followers that a GitHub account must have for it to be included in results
- minimumrepos: Minimum number of repositories that a GitHub account must have for it to be included in results
- githubrepolastupdatethresholdinmonths: An integer used to restrict repository data gathered so that only repositories updated with the specified number of months will be included in the saved CSV. This can be helpful for filtering out projects that are no longer active.
- detaillevel: "fulldetail" or "limiteddetail" - this controls whether the email, company, and bio fields will be filled in in the results CSV
- githubaccountdatacsvpathforvisualization: The file path to the specific CSV that contains GitHub account data that was gathered by the script that should be used for generating data visualizations
- githubrepodatacsvpathforvisualization: The file path to the specific CSV that contains GitHub repository data that was gathered by the script that should be used for generating data visualizations
For any questions about this repository, please contact the UT Austin Open Source Program Office at [email protected].