Python create time interval from human readable string

Question

Is there a Python library to convert human readable time interval to 2 datetimes?

Example:

Str(TimeInterval("last week")) # return a tuple (datetime(2023,11,12,0,0,0), datetime(2023,11,19,23,59,59))

Same for today, yesterday, last month, 2 months ago, year-to-date, etc.

Answer 1

As far as I know, there is no direct converter. But, with the pendulum library, you can implement it easily by using now.subtract() (as parameter days, months, years):

yesterday: now.subtract(days=1)
last week: now.subtract(weeks=1).start_of('week'), .end_of('week')
last month: now.subtract(months=1).start_of('month'), .end_of('month')
last year: now.subtract(years=1).start_of('year'), .end_of('year')

Sample Code:

import pendulum

def human_readable(interval):
    now = pendulum.now()
    if interval.lower() == 'today':
        return now.start_of('day'), now.end_of('day')
    elif interval.lower() == 'yesterday':
        yesterday = now.subtract(days=1)
        return yesterday.start_of('day'), yesterday.end_of('day')
    elif interval.lower() == 'last week':
        start_last_week = now.subtract(weeks=1).start_of('week')
        end_last_week = now.subtract(weeks=1).end_of('week')
        return start_last_week, end_last_week
    elif interval.lower() == 'last month':
        start_last_month = now.subtract(months=1).start_of('month')
        end_last_month = now.subtract(months=1).end_of('month')
        return start_last_month, end_last_month
    elif interval.lower() == '2 months ago':
        start_date = now.subtract(months=2).start_of('month')
        end_date = now.subtract(months=2).end_of('month')
        return start_date, end_date

intervals = ['last week', 'last month', '2 months ago']
for interval in intervals:
    start_date, end_date = human_readable(interval)
    print(f"{interval}: {start_date},  {end_date}")

Output:

last week: 2023-11-13T00:00:00+01:00,  2023-11-19T23:59:59.999999+01:00
last month: 2023-10-01T00:00:00+02:00,  2023-10-31T23:59:59.999999+01:00
2 months ago: 2023-09-01T00:00:00+02:00,  2023-09-30T23:59:59.999999+02:00