Note: This repository won't be updated anymore, please head on over to nim-fp-suite, which has the same data but with a slightly different naming scheme, and in the archive folder. This repo will still exist just in case someone actually relies on it.
This is a repository that uses automation to constantly retrieve false positive data about windows nim binaries.
In this repository, there lies a json/ folder filled with even more folders!
Every single folder inside json/ contains VirusTotal analysis results for a list of different nim versions at a specific date.
The date format is this: yyyy-MM-dd-hh:mm:ss with maybe an extra -NUM if the date itself is already taken (Fx. when there are leap seconds and so on)
So if you see a folder named json-2024-04-27-17:45:02 then you'll know that this specific scan began in the April 27th 2024, with the time being 17:45.
Inside the folders, you'll find a bunch of json files, these are the pure VirusTotal API results just saved as a file.
Every single file is named after this format: version_architecture.json (Note: This might change if I decide to scan Linux binaries in addition to Windows ones, so, y'know, be wary.)
So if you see a file named 2.0.4_x64.json then it's the analysis results for Nim 2.0.4 64 bit windows release.
Also, you might have noticed another file in the root named latest, this file contains the path to the latest scan results.
Why not? Collecting this data will allow us to see which antivirus companies constantly flag Nim binaries as malware. Because: surprise, surprise, they can flag binaries repeatedly, even after they've been reported, which is why reporting false positives is a laborious process that drains people's sanity.
Dealing with false positives easily is why I made nim-fp-suite, it's a tool designed to autogenerate email content that can be submitted to AV companies directly and nim-fp-data is a pure hobby project that serves no purpose other than to archive historical data showing which AV is least or most likely to flag Nim.
And we might be able to make lovely graphs in the near future! hopefully showing a decline! Data is beautiful and this dataset serves a bit of a practical purpose.