将三个单独的列转换为pandas数据框中的一个日期列

问题描述：

我希望将pandas数据框中的三列转换为单个日期列.问题在于列之一是日列.我无法将其转换为该月份和年份的确切日期.谁能帮我解决这个问题.看起来像这样:

I have three columns in a pandas dataframe that I want to convert into a single date column. The problem is that one of the columns is day column. I am not able to convert into exact date of that month and year. Can anyone please help me to solve this issue. It looks something like this:

   BirthMonth BirthYear Day
0   5           88      1st Monday
1   10          87      3rd Tuesday
2   12          87      2nd Saturday
3   1           88      1st Tuesday
4   2           88      1st Monday

答

根据您对我的第一条评论的答复，我的回答如下.我想这就是您要寻找的东西:

Based on your reply to my first comment I updated my answer as follows. I think this is what you are looking for:

import re
import time
import calendar
import numpy as np


days = ['1st Monday', '3rd Tuesday', '4th wednesday']
months = [2, 3, 5]
years = [1990, 2000, 2019]

def extract_numeric(text: str):
    return int(re.findall(r'\d+', text)[0])

def weekday_to_number(weekday: str):
    return time.strptime(weekday, "%A").tm_wday

def get_date(number: int, weekday: int, month: int, year: int) -> str:
    """ 3rd Tuesday translates to number: 3, weekday: 1 """
    firstday, n_days = calendar.monthrange(year, month)
    day_list = list(range(7)) * 6
    month_days = day_list[firstday:][:n_days]
    day = (np.where(np.array(month_days) == weekday)[0] + 1)[number - 1]
    return '{}/{}/{}'.format(day, month, year)

numbers = []
weekdays = []
for day in days:
    number, weekday = day.split()
    numbers.append(extract_numeric(number))
    weekdays.append(weekday_to_number(weekday))

dates = []
for number, weekday, month, year in zip(numbers, weekdays, months, years):
    dates.append(get_date(number, weekday, month, year))

print(dates)  # ['5/2/1990', '21/3/2000', '22/5/2019']

将三个单独的列转换为pandas数据框中的一个日期列

相关推荐