如何在列中的单词和标点符号之间添加空格？

Question

I have a column (string) in a dataframe with multiple spaces between words and punctuation.
I need to:

1. Add space between punctuation
2. Remove duplicated spaces

Punctuation I am looking for is /+- .

My dataframe:

col A
'this/is a+ string'
'this+is+a    string'

The output I expect:

col B
'this / is a + string'
'this + is + a string'

Answer 1

The way I solved this is in two steps: first, add space between pontuation, then check to see if there are any continuous spaces. For first step I used a function called punctuation_space to pass as "repl" argument to re.sub() .

import re

def punctuation_space(match_obj):
    """ return whatever matched surrounded by spaces """

    return ' ' + match_obj.group() + ' '

def fn(string):

    # first step
    string = re.sub(r'[+/-]', punctuation_space, string)

    # second step
    return re.sub(r' {2,}', ' ', string)

To check the above code:

import pandas as pd
original_col = ['this/is a+ string', 'this+is+a    string']

s = pd.Series(original_col)
print(s)
print(s.apply(fn))

Output:

0      this/is a+ string
1    this+is+a    string
dtype: object
0    this / is a + string
1    this + is + a string
dtype: object

Answer 2

you can try:

df['col A'] = df['col A'].apply(lambda y: " ".join((re.sub(r'([+/-])', lambda x: ' ' + x.group()+' ' , y)).split()) , 1)

OR:

df['col A'] = df['col A'].str.replace(r'([+/-])',  lambda x: ' ' + x.group()+' ', regex=True).apply(lambda x: ' '.join(x.split()))