replace whitespace with comma in multiline string (doc string), but keeping end-of-line

Question

I have a multiline string (and not a text file) like this:

x = '''
Index    Value         Max     Min    State
0    10    nan         nan     nan
1    20    nan         nan     nan    
2    15    nan         nan     nan     
3    25    20          10      1
4    15    25          15      2
5    10    25          15      4
6    15    20          10      3    
'''

The column white spaces are unequal.

I want to replace the whitespace with a comma, but keep the end-of-line.

So the result would look like this:

Index,Value,Max,Min,State
0,10,nan,nan,nan
1,20,nan,nan,nan    
2,15,nan,nan,nan     
3,25,20,10,1
4,15,25,15,2
5,10,25,15,4
6,15,20,10,3

...or alternatively as a pandas dataframe.

what i have tried

I can use replace('') with different spaces, but need to count the white spaces
I can use the re module (from here re.sub question ), but it converts the whole string to 1 line, where as i need to keep multiple lines (end-of-line).

Maybe [this](https://stackoverflow.com/questions/3469080/match-whitespace-but-not-newlines) can help. — MYousefi, Sep 19 '22 at 15:35
potentially, but how to implement with preserving the end-of-lines ? — D.L, Sep 19 '22 at 15:37

It_is_Chris · Accepted Answer · 2022-09-19T15:42:15.563

2

Try with StringIO

from io import StringIO
import pandas as pd


x = '''
Index    Value         Max     Min    State
0    10    nan         nan     nan
1    20    nan         nan     nan    
2    15    nan         nan     nan     
3    25    20          10      1
4    15    25          15      2
5    10    25          15      4
6    15    20          10      3    
'''

df = pd.read_csv(StringIO(x), sep='\s\s+', engine='python')

   Index  Value   Max   Min  State
0      0     10   NaN   NaN    NaN
1      1     20   NaN   NaN    NaN
2      2     15   NaN   NaN    NaN
3      3     25  20.0  10.0    1.0
4      4     15  25.0  15.0    2.0
5      5     10  25.0  15.0    4.0
6      6     15  20.0  10.0    3.0

edited Sep 19 '22 at 15:42

answered Sep 19 '22 at 15:37

It_is_Chris

13,504
2
23
41

1

and then chain with `to_csv()` :-) – Quang Hoang Sep 19 '22 at 15:39
1

this works !! presumably you need to add `import pandas as pd`... would you like to add that to the answer. – D.L Sep 19 '22 at 15:40
there were two great answers (upvoted both), but i have accepted this, just because it is a bit shorter on the line string. – D.L Sep 19 '22 at 15:53

score 2 · Answer 2 · answered Sep 19 '22 at 15:40

2

Since you tagged pandas, you can try:

out = ('\n'.join(pd.Series(x.split('\n')).str.strip().str.replace('\s+',',', regex=True)))

Output (note that there are leading and trailing blank lines because your x does):

Index,Value,Max,Min,State
0,10,nan,nan,nan
1,20,nan,nan,nan
2,15,nan,nan,nan
3,25,20,10,1
4,15,25,15,2
5,10,25,15,4
6,15,20,10,3

answered Sep 19 '22 at 15:40

Quang Hoang

146,074
10
56
74

well pointed out. i presume here, one needs to `import pandas as pd` to make out work. – D.L Sep 19 '22 at 15:43

replace whitespace with comma in multiline string (doc string), but keeping end-of-line

2 Answers2