Closed ChristopherRabotin closed 1 year ago
Hey, thanks for the suggestion. While hifitime
looks pretty cool, I'm not sure that there's a big win here from replacing chrono
. This library is a pretty thin wrapper around the C++ implementation of sgp4
, which itself is using second-level precision for storing the TLE epoch. Additionally, since the chrono
types are used as part of the public interface, to drop them would be a breaking change and require a major version bump which I don't think is quite justified.
However, if someone wanted to use hifitime
with this library, we could potentially add new constructors and accessor functions which use those types without dropping chrono
. It wouldn't change the internal representation of any of the timing information but it may make integrations easier.
I'm happy to review a PR if that's something you, or someone else, would like to add.
Hi there!
Sorry for jumping in out of the blue: I also work in astrodynamics and, over the past few years, I've developed
hifitime
. It's specifically for converting between different time scales. This is quite important in cislunar and interplanetary work, which is what I focus on.Anyway, I saw that you were using
DateTime<Utc>
fromchrono
to store the epoch of your orbital elements. I developedhifitime
specifically forNyx
initially (cf. theOrbit
structure), but it does everythingchrono
andtimers
do, so I've been trying to see what projects might benefit from more precision.