Date and time in python

In this blog, I will discuss how to convert date time column from string to pandas._libs.tslibs.timestamps.Timestamp format using pd.to_datetime or datetime.datetime format using datetime.datetime.strptime function. I will also discuss ways to change a datetime column into other formats.

At first, import pandas and datetime libraries

import datetime
import pandas as pd

1. Conversion of dates to standard format using pd.to_datetime

# Using pd.to_datetime
sample_dates = ('03/11/2021', '03-Nov-2021', 'Nov 03 21', '11/03/2021')
date1 = pd.to_datetime(sample_dates[0], format = '%d/%m/%Y')
date2 = pd.to_datetime(sample_dates[1], format = '%d-%b-%Y')
date3 = pd.to_datetime(sample_dates[2], format = '%b %d %y')
date4 = pd.to_datetime(sample_dates[3], format = '%m/%d/%Y')
print('date1 = ', date1, ', date2 = ', date2, ', date3 = ', date3, ' , date4 = ', date4)

Output:

date1 =  2021-11-03 00:00:00 , date2 =  2021-11-03 00:00:00 , date3 =  2021-11-03 00:00:00  , date4 =  2021-11-03 00:00:00

2. Conversion of dates to standard format using datetime.datetime.strptime

# Using datetime.datetime.strptime
sample_dates = ('03/11/2021', '03-Nov-2021', 'Nov 03 21', '11/03/2021')
date1 = datetime.datetime.strptime(sample_dates[0], '%d/%m/%Y')
date2 = datetime.datetime.strptime(sample_dates[1], '%d-%b-%Y')
date3 = datetime.datetime.strptime(sample_dates[2], '%b %d %y')
date4 = datetime.datetime.strptime(sample_dates[3], '%m/%d/%Y')
print('date1 = ', date1, ', date2 = ', date2, ', date3 = ', date3, ' , date4 = ', date4)

Output:

date1 =  2021-11-03 00:00:00 , date2 =  2021-11-03 00:00:00 , date3 =  2021-11-03 00:00:00  , date4 =  2021-11-03 00:00:00

3. Conversion of date time to standard format using strptime

The time is in 12 hrs format and AM/PM is used. Hence in the format argument I — for hours, M — for minutes and p — for AM/PM is used.

# Using I for 24 hrs and AM/PM
start_dt = '2/11/2022 4:00 AM'
end_dt = '2/11/2022 4:30 PM'
start_dt1 = datetime.datetime.strptime(start_dt, '%m/%d/%Y %I:%M %p')
end_dt1 = datetime.datetime.strptime(end_dt, '%m/%d/%Y %I:%M %p')
print('start time: ', start_dt1, ' , end time: ', end_dt1)

Output:

start time:  2022-02-11 04:00:00  , end time:  2022-02-11 16:30:00

The time is in 24 hrs format and AM/PM is not used. Hence in the format argument H — for hours and M — for minutes is used.

# Using H for 24 hrs and without AM/PM
start_dth = '2/11/2022 4:00'
end_dth = '2/11/2022 16:30'
start_dth1 = datetime.datetime.strptime(start_dth, '%m/%d/%Y %H:%M')
end_dth1 = datetime.datetime.strptime(end_dth, '%m/%d/%Y %H:%M')
print('start time: ', start_dt1, ' , end time: ', end_dt1)

Output:

start time:  2022-02-11 04:00:00  , end time:  2022-02-11 16:30:00

4. Changing date from one format to other

# Changing formats '03/11/2021' to '03-Nov-2021'
date_a = '03/11/2021'
date_b = datetime.datetime.strptime(date_a,'%d/%m/%Y').strftime('Day: %d-%b-%Y')
print(date_b)

Output:

Day: 03-Nov-2021

Note: While changing the format we can add additional text based on our requirement.

5. Getting days after or before a specific date

# Getting dates a few days before or after today
today = datetime.date.today()
yesterday = today - datetime.timedelta(days = 1)
tomorrow = today + datetime.timedelta(days = 1)
print("yesterday was: ",yesterday)
print("tomorrow is:", tomorrow)
print("Yesterday formatted: ",yesterday.strftime('%d-%b-%Y'))

Output:

yesterday was:  2022-06-17
tomorrow is: 2022-06-19
Yesterday formatted:  17-Jun-2022

6. Getting now in python and formatting it as per requirement

# now to add time of data update
now_date = datetime.datetime.now().date().strftime('%Y-%m-%d')
now_time = datetime.datetime.now().time().strftime('%H-%M-%S')
now_datetime = datetime.datetime.now().strftime('Day: %Y-%m-%d Time: %H-%M-%S')
print('now date = ', now_date, ' now time = ', now_time, ' now with formatting = ', now_datetime)

Output:

now date =  2022-06-18  now time =  05-15-12  now with formatting =  Day: 2022-06-18 Time: 05-15-12

7. Calculating date difference

# Date difference
sample_dt = ('03/11/2021', '03/12/2020')
date_end = pd.to_datetime(sample_dt[0], format = '%d/%m/%Y')
date_start = pd.to_datetime(sample_dt[1], format = '%d/%m/%Y')
diff_dt = date_end - date_start
diff_dt

Output:

Timedelta('335 days 00:00:00')

8. Converting datetime column in a sample data frame

# Creating a dataframe with event and date time columns
time_dict = {"Event": ['A', 'B', 'C', 'D', 'E'], "date_time" : ['6/9/2022 10:42 PM', '6/8/2022 6:22 AM', '6/10/2022 7:15 AM', '6/8/2022 10:59 PM', '6/7/2022 11:13 PM']}
dt_df = pd.DataFrame(time_dict)
dt_df

Output:

dt_df

# Extract date
dt_df['date'] = pd.to_datetime(dt_df['date_time'], format='%m/%d/%Y %I:%M %p').dt.date
# Extract time in 24 hrs format
dt_df['time_24'] = pd.to_datetime(dt_df['date_time'], format='%m/%d/%Y %I:%M %p').dt.time
# Extract time in 12 hrs format
dt_df['time_12'] = pd.to_datetime(dt_df['date_time'], format='%m/%d/%Y %I:%M %p').dt.strftime('%I:%M %p')
# Extract year
dt_df['year'] = pd.to_datetime(dt_df['date_time'], format='%m/%d/%Y %I:%M %p').dt.year
# Extract weekday
dt_df['weekday'] = pd.to_datetime(dt_df['date_time'], format='%m/%d/%Y %I:%M %p').dt.weekday
# Data frame after adding columns
dt_df

Output:

dt_df with added columns

I have published the code at Date time in python (Static report) (jetbrains.com).

Please share the blog if you like it.

References:

1. https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.to_datetime.html
2. https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior