ts.utc(...) != ts.now() when utc set to now.

Hi there, I love your work! This is an excellent python package. I have also found ephem and pyephem to be useful, though I’m not sure I understand the difference between them all. They all work well.

I am running the following:

from skyfield.api import load
import datetime, time, re

planets = load('de421.bsp')
earth, mercury = planets['earth'], planets['mercury']

ts = load.timescale()

t1 = datetime.datetime.now()
precise_second_t1 = float(t1.strftime("%-S.%f"))

t_now = ts.now()
t_utc_now = ts.utc(t1.year, t1.month, t1.day, t1.hour, t1.minute, precise_second_t1)

astrometric1 = earth.at(t_now).observe(mercury)
astrometric2 = earth.at(t_utc_now).observe(mercury)

ra1, dec1, distance1 = astrometric1.radec()
ra2, dec2, distance2 = astrometric2.radec()

print('RA for ts.now():\t', ra1)
print('RA for ts.utc(...):\t', ra2)

These should print the same, but they are not. I am wondering if you know why this might be?

Thank you again for such an amazing package!

Ever yours, Jake Ireland.

[brandon-rhodes]

Correct! The now() routine does not attach a timezone. I’m glad you have the date working correctly now in your script, and, yes, hopefully if anyone else is confused that datetime.now() does not return UTC, they will find this issue in a web search and learn about the difference.

[jakewilliami]

Ahh, I see. That does indeed work, thank you! So to clarify before we close the “issue”, datetime.datetime.now() lacks time zone information that datetime.datetime.utcnow()? Sorry to open an issue that wasn’t actually an issue. Hopefully this will help someone as naïve as myself in the future…