You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From Chirs Jordan: "I'm not sure how familiar you are with the rust library hifitime. We're using it in Marlu (and onwards) and it's wonderful; you can represent timestamps with nanosecond precision (hifitime::Epoch) and the struct methods allow you to convert to whatever (GPS, MJD, JD etc.). So, it might make sense to eventually use this as the single time-keeping method in the mwalib context structs, rather than having multiple time formats and things like seconds and milliseconds. Time resolutions etc. can be represented by hifitime::Duration, and this has the advantage of tracking units internally. I guess the trickiest part would be FFI; it's probably not worth exposing hifitime types to FFI callers, but you could convert to whatever you needed at the boundary. Just some food for thought."
The text was updated successfully, but these errors were encountered:
From Chirs Jordan: "I'm not sure how familiar you are with the rust library hifitime. We're using it in Marlu (and onwards) and it's wonderful; you can represent timestamps with nanosecond precision (hifitime::Epoch) and the struct methods allow you to convert to whatever (GPS, MJD, JD etc.). So, it might make sense to eventually use this as the single time-keeping method in the mwalib context structs, rather than having multiple time formats and things like seconds and milliseconds. Time resolutions etc. can be represented by hifitime::Duration, and this has the advantage of tracking units internally. I guess the trickiest part would be FFI; it's probably not worth exposing hifitime types to FFI callers, but you could convert to whatever you needed at the boundary. Just some food for thought."
The text was updated successfully, but these errors were encountered: