2
\$\begingroup\$

I have a code that calls many times a function returning a list of the dates between two dates formatted as

('0001-01-01 00:01:00'),('0001-01-02 00:01:00'), ...

The current solution is:

import numpy as np
import time
from datetime import datetime


beg = datetime.strptime('01 01 0001', '%d %M %Y') #datetime
end = datetime.strptime('01 01 2001', '%d %M %Y')


def get_days(date1, date2):
    day_diff = (date1 - date2).days + 1
    days = [str(start_date + dt.timedelta(d)) for d in range(day_diff)]
    dates = "('" + "'),('".join(days) + "')"
    return dates

Is there a faster way to achieve this?

 #timing
 t0 = time.time()

dates_list = get_days(beg, end) #feed datetime

t1 = time.time()
total_create = t1-t0
print("list comprehension: ", total_create,'s')
\$\endgroup\$
2
\$\begingroup\$

With the list comprehension you are filling a whole list with your values, and then you are sending that list to join.

Instead of a generating a list and then sending it, you can send a generator instead: similar to the list comprehension, but generates the values on-demand. With your old approach, if you had 10000 dates you would have them all at a list; with a generator it generates one at a time, so at least you will be consuming less memory.

With a generator, you would directly do:

dates = "('" + "'),('".join(str(start_date + dt.timedelta(d)) for d in range(day_diff)) + "')"

On a side note, the parameter names date1, date2 are not very explicit; it should be clear from the names which is the start and which is the end date.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.