I want to take a file's epoch timestamp float (e.g., "1661796943.8816772" from "os.stat(__file__).st_mtime"), and convert that float to what I think is called a DateTime string, and convert that DateTime string back to a float that is identical to the float I started with, i.e. from "os.stat(__file__).st_mtime".
I think my DateTime string has the precision I want (e.g., "2022,08,29,11,15,43,881677"), and I'm able to convert it to a DateTime object:
print(my_DateTime_obj)
>>2022-08-29 11:15:43.881677
But the routine I found to convert the DateTime object to an Epoch float lacks a lot of the precision my original float ("1661796943.8816772") had:
print(time.mktime(DateTime_obj.timetuple()))
>>1661796943.0
I think timetuple() is the problem, but I have not been able to figure it out.
Any tips about how to the convert the DateTime object to an Epoch float, without losing what I think is the microsecond precision, would be much appreciated.
I confess I am still a long way from understanding mktime(), timetuple() and what "structured time" really means.
use the datetime module:
import os
from datetime import datetime, timezone
import numpy as np
# file modification time, seconds since Unix epoch
unix0 = os.stat(__file__).st_mtime
# to datetime object
dt = datetime.fromtimestamp(unix0, tz=timezone.utc)
print(dt.isoformat(timespec="microseconds"))
# e.g.
# 2022-08-30T08:31:32.117021+00:00
# datetime object back to Unix time
unix1 = dt.timestamp()
# assert equal with microsecond precision
assert np.isclose(unix0, unix1, atol=1e-6)
Note: if you don't set tz=timezone.utc, the datetime object will be naive (as opposed to timezone-aware) and resemble local time. The conversion would work correctly nevertheless.