Converting Excel dates in 5 number format for python

pandas_date_handling_from_excel.py

import datetime
# When dumping to csv excel puts date in 5 digit float format (e.g. 39856.0)
# function converts series of dates to date string.

def date_to_string(digit_date):
    new_form = []
    for d in digit_date:
        try:
            x = datetime.date(1899,12,30) + datetime.timedelta(days=float(d))
            new_form.append(x.strftime('%Y-%m-%d'))
        except:
            new_form.append('n/a')
    return new_form
    
    
# convert dates in a pandas dataframe column to strings
df['column_name'] = date_to_string(df['column_name'])

# convert to pandas datetime format
# n/a values become NaT
df['column_name'] = pd.to_datetime(df['column_name'], errors='coerce')

I have this date. First part, before ":" is 5 digits format. Can you help me to parse last part? Is hour, minute, seconds but I don't know how to parse. Example: CREATION_DATE=42433:6382323727