计算两个日期之间的星期几小时对

Question

请考虑以下格式的星期几对小时对列表24H:

{
 'Mon': [9,23],
 'Thu': [12, 13, 14],
 'Tue': [11, 12, 14],
 'Wed': [11, 12, 13, 14]
 'Fri': [13],
 'Sat': [],
 'Sun': [],
}

和两个时间点,例如:

• 开始:

  datetime.datetime(2015, 7, 22, 17, 58, 54, 746784)
  

• 结束:

  datetime.datetime(2015, 8, 30, 10, 22, 36, 363912)
  

假设我们需要知道上面指定的每个星期几小时对的这两个日期时间(向上或向下舍入)之间有多少小时.

我怎样才能在Python中解决这个问题？我在一般的细节层面进行了探索timedelta,relativedelta但我没有发现任何与此有关的东西.

为简单起见,我们可以假设所有内容都指的是同一时区.

---

也许一个更简单的问题是专注于一个单日工作对,例如Wednesdays: 14两个任意日期之间有多少？

Answer 1

也许是这样的：

from calendar import day_abbr
from datetime import datetime, timedelta


def solve(start, end, data):
    days = list(day_abbr)
    output = dict.fromkeys(days, 0)

    while start <= end:
        day = days[start.weekday()]
        if start.hour in data[day]:
            output[day] += 1
        start = start + timedelta(minutes=60)

    return output


data = {
 'Mon': [9, 23],
 'Thu': [12, 13, 14],
 'Tue': [11, 12, 14],
 'Wed': [11, 12, 13, 14],
 'Fri': [13],
 'Sat': [],
 'Sun': [],
}

start = datetime(2015, 7, 22, 17, 58, 54, 746784)
end = datetime(2015, 8, 30, 10, 22, 36, 363912)

print solve(start, end, data)
# {'Wed': 20, 'Sun': 0, 'Fri': 6, 'Tue': 15, 'Mon': 10, 'Thu': 18, 'Sat': 0} 

获取每天每小时的计数：

from calendar import day_abbr
from collections import defaultdict
from datetime import datetime, timedelta
from pprint import pprint


def solve(start, end, data):
    days = list(day_abbr)
    output = defaultdict(lambda: defaultdict(int))

    while start <= end:
        day = days[start.weekday()]
        if start.hour in data[day]:
            output[day][start.hour] += 1
        start = start + timedelta(minutes=60)
    return {k: dict(v) for k, v in output.items()}


data = {
 'Mon': [9, 23],
 'Thu': [12, 13, 14],
 'Tue': [11, 12, 14],
 'Wed': [11, 12, 13, 14],
 'Fri': [13],
 'Sat': [],
 'Sun': [],
}

start = datetime(2015, 7, 22, 17, 58, 54, 746784)
end = datetime(2015, 8, 30, 10, 22, 36, 363912)

pprint(solve(start, end, data))
# output 
{'Fri': {13: 6},
 'Mon': {9: 5, 23: 5},
 'Thu': {12: 6, 13: 6, 14: 6},
 'Tue': {11: 5, 12: 5, 14: 5},
 'Wed': {11: 5, 12: 5, 13: 5, 14: 5}}