Open
Description
The tool only has access to what is already on the user's system, and what is contained in the downloaded archive or pulled from github. Therefore, for the tool to accomplish this, the archive would need to contain the data that the tool should report. Some examples of how to accomplish this:
- A referrer url is tracked in a query parameter, which is then injected into the archive server-side before download.
- We create different archives based on broad classifications of referrer. The archives each contain the data (e.g. a string in a file in the repo root) that distinguishes among these broad classifications.
Alternately, the referrer at the time of download could be stored in a cookie, which the tool could then extract. I'd consider doing this only if (1) and (2) are not possible since it entails an n^3 OS/browser/browser version implementation and testing blowup in the tool.
@sfshaza2 I was referred to you for help (or routing to the right person) on the website side of things.
@jonahwilliams for thoughts.
/cc @jayoung-lee