GPKit 2.0 is out. It's a Swift package for GP orbital data: it reads OMM and TLE, six-digit catalog numbers included, propagates them with SGP4, and finds passes with look angles and Doppler. The core doesn't need Foundation, and CI tests it on macOS, Linux, Windows, Android and WebAssembly and builds it for iOS, tvOS, watchOS and visionOS.
The new part is data age. Give it older and newer sets for the same satellites and it measures how far a prediction drifts as the data gets older, by orbit class. There's no table in the package yet. That comes once there's medium-Earth and geosynchronous data to measure it from.
It's tested against gpconf, a conformance kit I also maintain. With GPKit's default settings, my copy of gpconf's provider files (fetched in September) and the SGP4 verification files, 17 of gpconf 0.9.0's 21 cases match exactly and 2 more within gpconf's tolerance. One fails on two sets (23333 and 20413) because the reference output was made with a rounded epoch and GPKit keeps the exact one, and the remaining one checks data and runs no parser, so it isn't exercised. With the other epoch reading, which the README describes, 17 match exactly, 3 within tolerance, and none fails.
One change can break code: SGP4-XP sets (ephemeris type 4) now get their own refusal, .sgp4XP, where 1.x said .notSGP4. The migration article in the docs walks through it.
.package(url: "https://github.com/hneogy/GPKit.git", from: "2.0.0")
On Apple platforms your package also needs platforms: [.macOS(.v13), .iOS(.v16)] or newer.
The repo's at github.com/hneogy/GPKit. I'd love to hear what doesn't work for you.