creating start and end date tuples in list-CodePudding

I'm pretty new to python so I need some help. The below code gets the job done, but I feel like it isn't the most efficient way to accomplish this, any input would be greatly appreciated.

The goal: 1) pass a string start and end date 2) create a list to be able to chunk a request to an API. Each end date need to be 4 hours after the start date 3) each following start date in the list needs to be 3 hours after the previous start date 4) end the list when the final start date is >= end date

Important the final number must come out as i.e. '2018-06-05T09 2018-06-05T13' or my API request won't work

Thank you all in advance!

from datetime import datetime, timedelta
import numpy as np

start='2018-06-05T00'
end='2020-11-01T23'

start_date = datetime.strptime(start, '%Y-%m-%dT%H')
end_date = datetime.strptime(end, '%Y-%m-%dT%H')

delta = timedelta(hours=3)

end_date_initial = start_date   timedelta(hours=4)

start_date_list = [start_date]
end_date_list = [end_date_initial]

while True:
    start_date  = delta
    start_date_list.append(start_date)
    if start_date >= end_date:
        break

while True:
    end_date_initial  = delta
    end_date_list.append(end_date_initial)
    if end_date_initial >= end_date:
        break

start_date_array = np.array(start_date_list)
end_date_array = np.array(end_date_list)

formated_start_date = []
formated_end_date = []

for i in range(len(start_date_array)):
    formated_start_date.append(start_date_array[i].strftime('%Y-%m-%dT%H'))

for i in range(len(end_date_array)):
    formated_end_date.append(end_date_array[i].strftime('%Y-%m-%dT%H'))

# print(start_date_array[0], end_date_array[0])
# print(formated_start_date[0], formated_end_date[0])

time = formated_start_date[0]   ' '   formated_end_date[0]

def listOfTuples(i, j):
    return list(map(lambda x, y:(x   ' '   y), i, j))

time_tuples = listOfTuples(formated_start_date, formated_end_date)

print(time_tuples[3])

CodePudding user response：

If I've understood correctly, you can do this pretty easily and still take advantage of numpy's vectorization and broadcasting capabilities:

import numpy as np


ti, tf = np.datetime64("2018-06-05T00"), np.datetime64("2020-11-01T23")

start = np.arange(ti, tf, np.timedelta64(3, "h"))
end = start   np.timedelta64(4, "h")

api_times = [" ".join(map(str, t)) for t in zip(start, end)]