Ideal Data Gathering Scenario

[denovodavid]

• Have an executable program I can send out to people.
• The user runs the program, and they accept some permissions about data gathering and right to upload or what not.
• The program gathers information about their hardware and OS.
• The user can select a resolution to run at and maybe a few other options.
• The performance benchmark is run, and data is gathered.
• After a successful run, the program posts an issue on the GitHub repo with a specific label, and attaches the gathered data (probably a json file).
• Have a GitHub Actions Workflow that recognises the issue has been added, and creates a PR to the master branch /docs folder (for use with GitHub Pages) to add the data file to the repo so it can be loaded and visualised in the web page.

The reason to not have the program create the PR directly, is that it would have to download the repo, change the file, then create the PR. Depending on how big the repo ends up getting, this might be an easier option.

The created issue body could just be the gathered data, to make it potentially easier to handle, instead of dealing with uploading/downloading files. I'm not sure if there is a limit to the body size of an issue. It could also be in a binary format. Just had a look at supported serde-rs formats, and there's a readable Rusty Object Notation (RON) format that could be quite nice to deal with. However, the ubiquity of JSON is unfortunately too hard to pass up, especially if I actually want to load the info into JavaScript.

[denovodavid]

• Use on: issue to create workflow - https://help.github.com/en/actions/reference/events-that-trigger-workflows#webhook-events
• Use github.event context to access event payload - https://help.github.com/en/actions/reference/context-and-expression-syntax-for-github-actions#github-context
• The issue payload - https://developer.github.com/v3/activity/events/types/#issuesevent