Exercise 1

Parse the following dates and sort them by first 'date' then 'index' (== original position) and return a Series in the following form:

23   1971-07-08
19   2008-06-01
7    2009-01-01
21   2009-01-01
16   2009-02-01
4    2009-03-20
5    2009-03-20
6    2009-03-20
9    2009-03-20
10   2009-03-20
11   2009-03-20
12   2009-03-20
13   2009-03-20
14   2009-03-21
15   2009-03-22
0    2009-04-20
1    2009-04-20
17   2009-09-01
20   2009-12-01
22   2010-01-01
18   2010-10-01
3    2011-04-03
2    2014-04-20
8    2017-03-20
Name: dt, dtype: datetime64[ns]

Hints:

first column is the original position (index)
think about using pd.to_datetime()



In [1]:

    
import pandas as pd

time_values = '''I come at 04/20/2009 ...; 04/20/09 in Bremen; 4/20/14 maybe; or 4/3/2011 also possible; 
never at Mar-20-2009, but maybe later; how about Mar 20, 2009 or not; Moskow: March 20, 2009, 12:00; 
in Koeln, Mar. 20, 2009 however; Why not Mar 20 2017 ?; Clear picture for 20 Mar 2009; In the iden of 20 March 2009;
20 Mar. 2009 in Dresden; 20 March, 2009 BR; Mar 20th, 2009 Hamburg; Current date: Mar 21st, 2009; 
Mar 22nd is spring, 2009 Kiel; Feb 2009 is winter; Sep 2009; long ago in Oct 2010; 6/2008 vacation in Schwerin; 12/2009 blackout; 2009 timeout; Year 2010 the rest is history;
Soccer game at 7/8/71 in NY'''

time_s = pd.Series(time_values.split(';')).str.strip()
time_s









    Out[1]:





0                  I come at 04/20/2009 ...
1                        04/20/09 in Bremen
2                             4/20/14 maybe
3                 or 4/3/2011 also possible
4     never at Mar-20-2009, but maybe later
5             how about Mar 20, 2009 or not
6             Moskow: March 20, 2009, 12:00
7           in Koeln, Mar. 20, 2009 however
8                     Why not Mar 20 2017 ?
9             Clear picture for 20 Mar 2009
10             In the iden of 20 March 2009
11                  20 Mar. 2009 in Dresden
12                        20 March, 2009 BR
13                   Mar 20th, 2009 Hamburg
14             Current date: Mar 21st, 2009
15            Mar 22nd is spring, 2009 Kiel
16                       Feb 2009 is winter
17                                 Sep 2009
18                     long ago in Oct 2010
19              6/2008 vacation in Schwerin
20                         12/2009 blackout
21                             2009 timeout
22            Year 2010 the rest is history
23              Soccer game at 7/8/71 in NY
dtype: object



In [2]:

    
def date_sorter(df):
    # TODO your code goes here
    return

date_sorter(time_s)