Reverse string columns in a pandas subset dataframe

Question

I have the following dataframe.

    ID  LOC Alice   Bob  Karen
0   1   CH  9|5 6|3 4|4
1   2   ES  1|1 0|8 2|0
2   3   DE  2|4 6|6 3|1
3   4   ES  3|9 1|2 4|2

Alice and Bob columns contain string values. I want to reverse the strings in these columns conditional on the value of another column. For example, where LOC==ES, reversing the strings in the corresponding columns would look like:

    ID  LOC Alice   Bob   Karen
0   1   CH  9|5 6|3 4|4
1   2   ES  1|1 8|0 0|2
2   3   DE  2|4 6|6 3|1
3   4   ES  9|3 2|1 2|4

Is there a fast way to perform this operation on all matching rows in a csv file with thousands rows?

Thank you.

Answer 1

#cols = ['Alice','Bob']
In [17]: cols = df.columns.drop(['ID','LOC'])

In [18]: df.loc[df.LOC=='ES', cols] = df.loc[df.LOC=='ES', cols].apply(lambda x: x.str[::-1])

In [19]: df
Out[19]:
   ID LOC Alice  Bob Karen
0   1  CH   9|5  6|3   4|4
1   2  ES   1|1  8|0   0|2
2   3  DE   2|4  6|6   3|1
3   4  ES   9|3  2|1   2|4

Answer 2

Use df.loc to get your row slices, then apply string reverse [::-1] operation on the Alice and Bob columns with df.applymap .

In [533]: df.loc[df['LOC'] == 'ES', ['Alice', 'Bob']] = \
                 df.loc[df['LOC'] == 'ES', ['Alice', 'Bob']].applymap(lambda x: x[::-1])

In [534]: df
Out[534]: 
   ID LOC Alice  Bob Karen
0   1  CH   9|5  6|3   4|4
1   2  ES   1|1  8|0   2|0
2   3  DE   2|4  6|6   3|1
3   4  ES   9|3  2|1   4|2

Answer 3

You could try using .apply() as follows for your example condition where column LOC == 'ES' :

df['Alice'] = df[['LOC','Alice']].apply(lambda x: x['Alice'][::-1] if x['LOC'] == 'ES' else x['Alice'], axis=1)

Note in my answer that [::-1] is a way to reverse a string