Skip to main content

Simple functions for tz-aware datetimes.

Project description

Performs common operations on datetimes with clean syntax, acting as a thin wrapper for datetime and pytz. Force timezone-aware datetimes. All operations are top-level functions: No dealing with methods from multiple modules and objects.

There are several existing modules designed to improve Python’s datetimes. Here’s why Saturn is different:

  • Uses native datetime types for compatibility and speed
  • Only one import required
  • Clean, intuitive syntax and function names. No boilerplate.
  • Operates exclusively with top-level functions; no sorting through methods from multiple objects and modules
  • Raises an exception if a func is given a tz-naive datetime as input. All datetime outputs are timezone-aware, with an easily-set TZ, defaulting to UTC.

This project doesn’t tackle the deeper problems with Python’s datetimes that can arise from timezone and localization conflicts. It aims to make working with dateimtes easier by providing a concise, consistent, and expressive API; and preventing naive datetimes from entering your code.

Saturn uses Pytz for timezones. Used as a dependency.: Pytz website Pytz is licensed under the MIT license.

… and uses code from Arrow for string formatting and parsing. Not a dependency.: Arrow website: Arrow is licensed under Apache 2.

Python 2 is unsupported.

Included functions

  • datetime: Return a timezone-aware datetime.datetime object. Created the same way as datetime.datetime, with an optional ‘tz’ argument for a timezone string. Defaults to UTC.
  • time: Same concept as datetime.time; easily create a tz-aware time.
  • now: Find current utc time; timezone-aware.
  • range_dt: Iterate over datetimes, with a customizable interval. Similar to builtin range. Lazy.
  • fix_naive: Convert a timezone-naive datetime to an aware one.
  • move_tz: Change a datetime from one timezone to another.
  • combine: Similar to datetime.datetime.combine, but always tz-aware.
  • to_str: Similar to datetime.datetime.strftime, but with a cleaner format string, from Arrow.
  • from_str: Similar to datetime.datetime.strptime, but with a cleaner format string, from Arrow. Returns date, datetime, or time objects as needed.
  • to_iso: Wrapper for datetime.datetime’s isoformat() method, as a function.
  • from_iso: Create a datetime from an isoformat string.
  • to_epoch: Wrapper for datetime.datetime’s timestamp method, as a function.
  • from_epoch: Wrapper for datetime.datetime’s from_timestamp method, as a function.
  • split: Split a datetime into date and time components. Useful because datetime’s .time() method strips timezone info.
  • add, subtract: Add or subtract to/from a datetime.
  • overlaps: Deterine if two date/time/datetime ranges overlap.
  • timedelta, date, and today are included as wrappers for their respective datetime/date classes, so you don’t need to import datetime.


pip install saturn


Create a timezone-aware datetime. If you don’t specify a ‘tz’ argument, it defaults to UTC. Works for times too:

saturn.datetime(2016, 1, 1, 16, tz='US/Eastern')
# datetime.datetime(2016, 1, 1, 16, 0, tzinfo=<DstTzInfo 'US/Eastern' EST-1 day, 19:00:00 STD>)

saturn.datetime(2016, 1, 1, 16)
# datetime.datetime(2016, 1, 1, 16, 0, tzinfo=<UTC>)

saturn.time(11, 29, 30)
# datetime.time(11, 29, 30, tzinfo=<UTC>)

Make a tz-naive datetime aware:

naive = datetime.datetime(2016, 1, 1)
saturn.fix_naive(naive, "Pacific/Midway")
# datetime.datetime(2016, 1, 1, 0, 0, tzinfo=<DstTzInfo 'Pacific/Midway' SST-1 day, 13:00:00 STD>)

Find the current datetime, in UTC:
# datetime.datetime(2016, 4, 29, 20, 36, 53, 257753, tzinfo=<UTC>)

Move from one timezone to another:

dt = saturn.datetime(2016,1,1, tz='Asia/Gaza')
# datetime.datetime(2016, 1, 1, 0, 0, tzinfo=<DstTzInfo 'Asia/Gaza' EET+2:00:00 STD>)

saturn.move_tz(dt, 'Europe/Vatican')
# datetime.datetime(2015, 12, 31, 23, 0, tzinfo=<DstTzInfo 'Europe/Vatican' CET+1:00:00 STD>

Combine a date and time into a timezone-aware datetime. If the time is already aware, the ‘tz’ argument is ignored:

date, time =, 3, 2), datetime.time(16, 30)

saturn.combine(date, time)
# datetime.datetime(2016, 3, 2, 16, 30, tzinfo=<UTC>)

saturn.combine(date, time, tz='Europe/London')
# datetime.datetime(2016, 3, 2, 16, 30, tzinfo=<DstTzInfo 'Europe/London' GMT0:00:00 STD>)

Split a datetime into date and time components; keeps tzinfo, unlike datetime.time().

dt = saturn.datetime(2016, 3, 2, 16, 30, 1, 500, tz='US/Mountain')
date, time = saturn.split(dt)
#, 3, 2)
# datetime.time(16, 30, 1, 500, tzinfo=<DstTzInfo 'US/Mountain' MST-1 day, 17:00:00 STD>)

Iterate through a range of datetimes. Valid intervals are ‘week’, ‘month’, ‘day’ ‘hour’, ‘minute’, ‘second’, ‘millisecond’, and ‘microsecond’:

start, end = saturn.datetime(2016, 1, 2, 12, 30), saturn.datetime(2016, 1, 5, 12, 30)
for dt in saturn.range_dt(start, end, interval='day'):

# 2016-01-02 12:30:00+00:00
# 2016-01-03 12:30:00+00:00
# 2016-01-04 12:30:00+00:00

for dt in saturn.range_dt(start, end, 4, interval='hour'):

# 2016-01-02 12:30:00+00:00
# 2016-01-02 16:30:00+00:00
# 2016-01-02 20:30:00+00:00
# 2016-01-05 00:30:00+00:00
# 2016-01-05 04:30:00+00:00
# 2016-01-05 08:30:00+00:00

Convert a datetime to a string. Uses format from Arrow:

saturn.to_str(, 'YYYY-MM-DD hh:mm')
# '2016-04-29 03:30'

Convert a string to a datetime. Uses format from Arrow. If the string includes a timezone, the optional tz argument is ignored:

saturn.from_str('2016-04-29 03:30', 'YYYY-MM-DD hh:mm')
# datetime.datetime(2016, 4, 29, 3, 30, tzinfo=<UTC>)

saturn.from_str('2016-04-29 03:30', 'YYYY-MM-DD hh:mm', tz='Africa/Cairo')
# datetime.datetime(2016, 4, 29, 3, 30, tzinfo=<DstTzInfo 'Africa/Cairo' EET+2:00:00 STD>)

saturn.from_str('1381685817', 'X')
# datetime.datetime(2013, 10, 13, 17, 36, 57, tzinfo=<UTC>)

Convert a datetime to an ISO-8601 string or epoch:

# '2016-04-29T20:12:05.807558+00:00'

# 1461960725.807558

Convert an ISO-8601 string or epoch to a datetime:

# datetime.datetime(2016, 4, 29, 20, 12, 05, tzinfo=<UTC>)

# datetime.datetime(2016, 4, 29, 21, 12, 5, tzinfo=<UTC>)

For details on to_str and from_str syntax, please reference Arrow’s formatting guide.

Check if a range of times overlaps.

start1 = saturn.datetime(2018, 1, 1, 9)
end1 = saturn.datetime(2018, 1, 1, 12)
start2 = saturn.datetime(2018, 1, 1, 11)
end2 = saturn.datetime(2018, 1, 1, 17)

saturn.overlaps(start1, end1, start2, end2)
# True
saturn.overlaps(start1, end1, saturn.datetime(2018, 1, 1, 13), end2)
# False

Function input and output:

datetime(year: float, month: float, day: float, hour: float=0, minute: float=0,
         second: float=0, microsecond: float=0, tzinfo=None, tz: str='UTC') -> datetime.datetime

time(hour: float, minute: float=0, second: float=0,
     microsecond: float=0, tzinfo=None, tz: str='UTC') -> datetime.time

now() -> datetime.datetime

combine(_date:, _time: _datetime.time, tz: str='UTC') -> datetime.datetime

fix_naive(dt: TimeOrDatetime, tz: str='UTC') -> datetime.datetime

to_str(dt: DateOrDatetime, str_format: str) -> str

from_str(dt_str: str, str_format: str, tz: str='UTC') -> DateOrTimeOrDatetime

to_iso(dt: DateOrDatetime) -> str

from_iso(iso_str: str, tz: str='UTC') -> datetime.datetime

to_epoch(dt: DateOrDatetime) -> float:

from_epoch(epoch: float, tz: str='UTC') -> _datetime.datetime:

move_tz(dt: datetime.datetime, tz: str) -> datetime.datetime

add(dt: datetime.datetime, days: float=0, seconds: float=0, microseconds: float=0,
    milliseconds: float=0, minutes: float=0, hours: float=0, weeks: float=0) -> datetime.datetime

subtract(dt: datetime.datetime, days: float=0, seconds: float=0, microseconds: float=0,
    milliseconds: float=0, minutes: float=0, hours: float=0, weeks: float=0) -> datetime.datetime

range_dt(start: DateOrDatetime, end: DateOrDatetime, step: int=1,
         interval: str='day') -> Iterator[datetime.datetime]

split(dt: datetime.datetime) -> Tuple[, _datetime.time]:

overlaps(start1: DateOrTimeOrDatetime, start2: DateOrTimeOrDatetime,
         end1: DateOrTimeOrDatetime, end2: DateOrTimeOrDatetime) -> bool:

Some syntax we’re dodging:

pytz.timezone('Europe/Berlin').localize(datetime.datetime(1985, 2, 1, 13, 21))

arrow.Arrow(1999, 9, 9, 9, 30,'US/Eastern'))

pytz.timezone('US/Mountain').localize(datetime.datetime.combine(date, time))


aware_time = datetime.time(aware_dt.hour, aware_dt.minute, aware_dt.second,
    aware_dt.microsecond, aware_dt.tzinfo)

Replaced by:

saturn.datetime(1985, 2, 1, 13, 21, tz='Europe/Berlin')

saturn.datetime(1999, 9, 9, 9, 30, tz='US/Eastern')

saturn.combine(date, time, 'US/Mountain')

saturn.move_tz(aware_dt, 'US/Pacific')

_, aware_time = saturn.split(aware_dt)

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for Saturn, version 0.3.5
Filename, size File type Python version Upload date Hashes
Filename, size Saturn-0.3.5-py3-none-any.whl (16.8 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size Saturn-0.3.5.tar.gz (15.2 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page