Pandas Delete Part of String

Question

>>> df
                       Time
    5/10/2017 (135) 01:05:03
    5/11/2017 (136) 04:05:06

Given an input date such as this in a DataFrame, how would I delete the Julian Date, (135) and (136), and remove the whitespace in the middle, so that the output looks like:

>>> df
                       Time
    5/10/2017 01:05:03
    5/11/2017 04:05:06

I've tried:

df['Time'].replace('(135)','', regex=True, inplace=True)

which outputs:

>>> df
                    Time
0  5/10/2017 () 01:05:03

I was wondering what I'm doing wrong here.

jezrael · Accepted Answer · 2017-06-08 06:13:06Z

3

You can use replace by regex:

First need escape () by \ because special chars in regex, then match all ints by \d+ and last match zero or more whitespaces after ) by \s*.

df['Time'] = df['Time'].str.replace("\(\d+\)\s*", '')
print (df)
                 Time
0  5/10/2017 01:05:03
1  5/11/2017 04:05:06

And if need convert to datetime:

df['Time'] = pd.to_datetime(df['Time'].str.replace("\(\d+\)\s*", ''))
print (df)
                 Time
0 2017-05-10 01:05:03
1 2017-05-11 04:05:06

EDIT:

In your sample are mising escaping chars \ and is possible use instead \d+ [0-9]+:

df['Time'].replace('\([0-9]+\)\s*','', regex=True, inplace=True)
print (df)
                 Time
0  5/10/2017 01:05:03
1  5/11/2017 04:05:06

edited Jun 8, 2017 at 6:13

answered Jun 8, 2017 at 5:45

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

geeb.24 Over a year ago

Awesome, that works! Could you describe the backslash and d+ values in str.replace() and how they're used, or if there's documentation on how you came up with that input? Thanks!

jezrael Over a year ago

Unfortunately regex problematic is very huge, in pandas documentation are only some samples here - it is about extracting, but similar works with str.replace.

Collectives™ on Stack Overflow

Pandas Delete Part of String

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related