How to Manipulate Date And Time in Python Like a Boss

 Commonly used DateTime functions with examples

When using Python, we all have the moment of trying to figure out how to “manipulate” the date and time.

This is a cheat sheet I created that covers the most commonly used DateTime related functions in Python. If you want something simple but practical with examples, take a look!

We’ll cover all 7 popular topics below:

• Today’s date and time in different formats
• String to date conversion
• The Difference in DateTime calculation
• Datetime plus/minus a specific time
• Datetime comparison
• Time zones settings
• Unix timestamp / Epoch time calculation

Today’s date and time in different formats

Let’s warm-up from the most basic. Below is the code that prints the current year, month, date, hour, minute, seconds, and milliseconds.

In: from datetime import datetime    
    d = datetime.now() #today's datetime
    d
Out:datetime.datetime(2019, 12, 22, 13, 14, 18, 193898)

This is handy information, but often we only need part of it. We can print the different parts below.

In:  print(d.weekday()) #day of week - Monday is 0 and Sunday is 6
     print(d.year)
     print(d.month)
     print(d.day)
     print(d.hour)
     print(d.minute)
     print(d.second)
Out: 6
     2019
     12
     22
     13
     14
     18

Moreover, specific formats of the date/time might be needed. We can use the below list to customize different date formats. This can also be considered a way of converting date to string. The full list can be found here.

In:  print(d.strftime("%A %d/%m/%Y")) # date to string
Out: Sunday 22/12/2019

String to date conversion

Below are examples showing two popular strings being converted to date format:

In:  date_string = '2016-02-01 12:00PM'
     print(datetime.strptime(date_string, '%Y-%m-%d %I:%M%p'))
Out: 2016-02-01 12:00:00
In:  date_string = '02/01/2016'
     d2 = datetime.strptime(date_string, '%m/%d/%Y')
     print(d2)
Out: 2016-02-01 00:00:00

The Difference in DateTime calculation

The example below prints the difference in days (for example, today and February 1st, 2016):

In:  from datetime import timedelta     
     
     d = datetime.now()
     date_string = '2/01/2016'
     d2 = datetime.strptime(date_string, '%m/%d/%Y')
     
     print(d - d2)
Out: 1420 days, 13:18:27.386763

We can also only print the difference between two date-times in days, weeks, or years, etc.

In:  date_diff = (d - d2)/timedelta(days=1)
     print('date_diff = {} days'.format(date_diff))
Out: date_diff = 1420.5544836430902 days
In:  date_diff = (d - d2)/timedelta(weeks=1)
     print('date_diff = {} weeks'.format(date_diff))
Out: date_diff = 202.93635480615575 weeks
In:  date_diff = (d - d2)/timedelta(days=365)
     print('date_diff = {} years'.format(date_diff))
Out: date_diff = 3.8919300921728497 years

Datetime plus/minus a specific time

Let’s do some “time travel” by different time intervals of seconds, minutes, hours, days, weeks, or years:

In:  print(d + timedelta(seconds=1)) # today + one second     
     print(d + timedelta(minutes=1)) # today + one minute     
     print(d + timedelta(hours=1)) # today + one hour  
     print(d + timedelta(days=1)) # today + one day
     print(d + timedelta(weeks=1)) # today + one week
     print(d + timedelta(days=1)*365) # today + one year
Out: 2019-12-22 13:18:28.386763
     2019-12-22 13:19:27.386763
     2019-12-22 14:18:27.386763
     2019-12-23 13:18:27.386763
     2019-12-29 13:18:27.386763
     2020-12-21 13:18:27.386763

Datetime comparison

The comparisons between dates are straightforward with usual comparison symbols:

In:  print(d < (d2 +(timedelta(days=365*6)))) # d is no more than 6 years (assume each year has 365 days) after d2?
     print(d > (d2 +(timedelta(weeks=52*6)))) # d is more than 6 years (assume each year has 52 weeks) after d2?
     print(d != d2) # d2 is not the same date as d?
     print(d == d2) # d2 is the same date as d?
Out: True
     False
     True
     False

Time zones settings

And we can also compare time in different time zones, such as Toronto and Shanghai:

In:  import pytz
     timezone = pytz.timezone("America/Toronto")
     dtz = timezone.localize(d)
     print(dtz.tzinfo)
     print(dtz)
Out: America/Toronto
     2019-12-22 13:18:27.386763-05:00
In:  shanghai_dt = dtz.astimezone(pytz.timezone("Asia/Shanghai"))
     print(shanghai_dt)
Out: 2019-12-23 02:18:27.386763+08:00
In:  (dtz - shanghai_dt)/timedelta(days=1) # same datetimes
Out: 0.0

If you are interested in the whole list of different time zones. Using the below code can print them:

In:  for tz in pytz.all_timezones:
        print(tz)
Out: Africa/Abidjan
     Africa/Accra
     Africa/Addis_Ababa
(Only showing the first three as examples)

Unix timestamp / Epoch time calculation

Unix timestamps are commonly used for files in operating systems. Often they show up in datasets as well.

First, we can get the current Unix timestamp:

In:  from datetime import timezone
     dt_now = datetime.now(timezone.utc)
     print(dt_now)
     print(dt_now.tzinfo)
     print(dt_now.timestamp()) # the unix timestamp.
Out: 2019-12-22 18:21:28.681380+00:00
     UTC
     1577038888.68138

Also, we can convert Unix timestamp to DateTime format:

In:  utc_timestamp = 1377050861.206272
     unix_ts_dt = datetime.fromtimestamp(utc_timestamp, timezone.utc)
     print(unix_ts_dt)
     print(unix_ts_dt.astimezone(pytz.timezone("America/Toronto")))
     print(unix_ts_dt.astimezone(pytz.timezone("Asia/Shanghai")))
Out: 2013-08-21 02:07:41.206272+00:00
     2013-08-20 22:07:41.206272-04:00
     2013-08-21 10:07:41.206272+08:00