search

kennethreitz / maya

Datetimes for Humans™
MIT License
3.41k stars 199 forks source link

Include epoch microseconds in DT comparison methods. Closes #156 #157

Closed timofurrer closed 4 years ago

kennethreitz commented 6 years ago

This was originally done intentionally. Please make careful considerations before making this change.

kennethreitz commented 6 years ago

e.g. two nows generated “apparently” simultaneously, then compared equal.

I agree the microseconds are a better design decision.

kennethreitz commented 6 years ago

This is a major design change, though, and warrants a minor (or maybe even major) version bump.

timofurrer commented 6 years ago

This was originally done intentionally

What was the reason for that?

e.g. two nows generated “apparently” simultaneously, then compared equal.

Well, I hope no one bet one the that. It could easily happen that two nows generated “apparently” simultaneously would led to two datetimes with a different second and therefore not evaluate equal.

This is a major design change, though, and warrants a minor (or maybe even major) version bump.

That's right. The question even is if internally we should calculate with the total microseconds after the epoch start to avoid floating point numbers.

github-actions[bot] commented 4 years ago

This Pull Request is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 7 days