Basically, I need to manipulate a date but keep its existing format. I'm looking for a library/algorithm that does something akin to this:

Input: Jan 6

Output: %b %w

Is it doable?

CodePudding user response：

There's no need for a library. If you are thinking to "convert" the date string to its possible date format, you can use a function called strptime, and "reverse lookup" for a date string's possible date format like this:

import datetime as dt

mydates='''\
Jan 6
February 19
1 December
20 Sep
'''

matched_dateformats_list=[]

# You can expand this to include more variations and combinations of dateformats
dateformats = ('%b %d','%d %b', '%B %d','%d %B')

for date in mydates.splitlines():
    for dateformat in dateformats:
        try:
           # Check to see if strptime errors out, this way we know if the dateformate matches the date string format
           dateobject = dt.datetime.strptime(date, dateformat)

           # Append successful match
           matched_dateformats_list.append((date, dateformat)) 
           break
        except:
            # If it's not a match, skip, don't append, and continue to the next one
           continue

# Print the successful matches (date and the date format)
for (i,j) in matched_dateformats_list:
    print (i   " -> "   j)

Output:

Jan 6 -> %b %d
February 19 -> %B %d
1 December -> %d %B
20 Sep -> %d %b

The Python datetime formats can be found here:

https://docs.python.org/3/library/datetime.html

There's a limitation though. If both your month and day is in numeric format this won't work.

CodePudding user response：

it was quite challenging to make it with regex, hope it will work:

from re import sub

dates = ['Jan 6','February 19','1 December','20 Sep']

def func(m):
    return f"%d %{['b','B'][bool(m[3])]}" if m[1] else f"%{['b','B'][bool(m[3])]} %d"

for date in dates:
    print(sub(r'(\d )?\s?([a-zA-Z]{3})([a-z] )?\s?(\d )?',func, date))

>>> out
'''
%b %d
%B %d
%d %B
%d %b