Published an OSS Repository That Continuously Stores GitHub Repository Metrics Using GAS

Hey Devs!

Previously, I explored how we can monitor OSS activity metrics based on GitHub repository data.

Consideration of OSS Activity Metrics Based on GitHub Repository Data

Takashi Masuda for ROUTE06, Inc. ・ Dec 16 '24

#ospo #opensource #github #marketing

Building on that, I implemented a Google Apps Script (GAS) that continuously stores traffic data from GitHub repositories.

Continuously Storing Traffic Data of Multiple GitHub Repositories Using GAS

Takashi Masuda for ROUTE06, Inc. ・ Dec 23 '24

#github #googlesheets #googleappsscript

This time, while adding GAS functions to store other metrics, I have published the repository as OSS. I’d like to take this opportunity to introduce it once again.

• Published Repository
• Stored Metrics
  • 1. Stargazers
    • Objectives
    • Issues
    • Solution
  • 2. Traffic views
    • Objectives
    • Issues
    • Solution
  • 3. Traffic clones
    • Objectives
    • Issues
    • Solution
  • 4. Top referral sources
    • Objectives
    • Issues
    • Solution
• Conclusion

Published Repository

The published repository is available here.

The initial setup is summarized in the README. I will focus mainly on why I decided to store these metrics.

Stored Metrics

1. Stargazers

Objectives

• To understand the daily increase in stars.
• To correlate the days with star increases with possible reasons or initiatives.

Issues

• On GitHub’s web UI, only the total number of stars and the accounts that starred are visible.
  • Example: https://github.com/giselles-ai/giselle/stargazers
• With Star History, the counts for nearby days are rounded off, and it does not allow for note-taking.
  • Example: https://www.star-history.com/#giselles-ai/giselle&Date

Solution

I implemented the function updateSheetWithStargazers to record who starred on which day.

The graph was created as shown below. The key is to enable “Aggregate” on the X-axis. Except for the colored cells, you can freely add notes.

2. Traffic views

Objectives

• To permanently store the numerical data from the Visitors graph in the repository’s Insights > Traffic section.
  • https://github.com/[ORG]/[REPO]/graphs/traffic

Issues

• The graph does not display data older than two weeks.
• The GitHub REST API Get traffic views has the same limitation.

Solution

I implemented the function updateSheetWithLatestTrafficViews that uses this API to store the daily Views and Unique visitors in a spreadsheet.

3. Traffic clones

Objectives

• To permanently store the numerical data from the Git clones graph in the repository’s Insights > Traffic section.
  • https://github.com/[ORG]/[REPO]/graphs/traffic

Issues

• The graph does not display data older than two weeks.
• The GitHub REST API Get traffic clones has the same limitation.

Solution

I implemented the function updateSheetWithLatestTrafficClones that uses this API to store the daily Clones and Unique cloners in a spreadsheet.

4. Top referral sources

Objectives

• To permanently store the numerical data from the Referring sites graph in the repository’s Insights > Traffic section.
  • https://github.com/[ORG]/[REPO]/graphs/traffic

Issues

• The graph does not display data older than two weeks.
• The GitHub REST API Get top referral sources has the same limitation.
• Each value is the total for the past 14 days, which is not very user-friendly.

Solution

I implemented the function updateSheetWithLatestTrafficReferrers using this API to store the daily Referring sites in a spreadsheet.

To create a stacked graph, I made a pivot table of Views and Unique visitors on a separate sheet, along with their respective graphs.

While the issue of each value being a total for the past 14 days remains unsolved, it seems possible to separate them using spreadsheet functions.

Conclusion

I have published the OSS repository https://github.com/route06inc/ospo-google-apps-script that continuously stores GitHub repository metrics using GAS.

This tool is already in daily use on the repositories of Giselle and Liam, and the data is being effectively utilized.

• https://github.com/giselles-ai/giselle
• https://github.com/liam-hq/liam