How to replace empty string in python

Question

I have a csv file and I read it into array, the original csv a 5-row, 8-column file with empty elements

       1  2         3        4    5    6   7         8     
Row 1:   '1 1'     '4 4'                  '2 2'   
Row 2:   '3'       '3'                    '3' 
Row 3:   '1 1 1 1' '1 1 1 1'              '2 2 2 2'
Row 4:   '2'       '4'                    '2' 
Row 5:   '4'       '4'                               '4'

I read it into my code:

[[nan '1 1' '4 4' nan nan nan '2 2' nan]
 [nan '3' '3' nan nan nan '3' nan]
 [nan '1 1 1 1' '1 1 1 1' nan nan nan '2 2 2 2' nan]
 [nan '2' '4' nan nan nan '2' nan]
 [nan '4' '4' nan nan nan nan '4']]

So what I want to get is to replace all empty elements into same number of -1 with other elements:

[['-1 -1' '1 1' '4 4' '-1 -1' '-1 -1' '-1 -1' '2 2' '-1 -1']
 ['-1' '3' '3' '-1' '-1' '-1' '3' '-1']
 ['-1 -1 -1 -1' '1 1 1 1' '1 1 1 1' '-1 -1 -1 -1' '-1 -1 -1 -1' '-1 -1 -1 -1' '2 2 2 2' '-1 -1 -1 -1']
 ['-1' '2' '4' '-1' '-1' '-1' '2' '-1']
 ['-1' '4' '4' '-1' '-1' '-1' '-1' '4']]

When I use re.match("\d",element), I can not get the result. So could anyone help?

duplicate of : http://stackoverflow.com/questions/1540049/replace-values-in-list-using-python — C.LECLERC, Jan 11 '16 at 11:53
@C.LECLERC, thanks for your link, but I think my question is different because I need to replace the element with a specific number of `-1`, so I need to judge how many elements in each item. — flyingmouse, Jan 11 '16 at 11:56
i might misunderstand but in the example isn't "-1 -1" a single string element ? (even if it describes 2 numeric values). — C.LECLERC, Jan 11 '16 at 11:59
@PaulRooney, in the original `csv` file, there are empty elements. When I read the file into my code, it displays as `nan` — flyingmouse, Jan 11 '16 at 11:59
@C.LECLERC, it is needed to treat them as two integers, so I can use `zip` and `split" "` to split them, but I need to get the same number of `-1` to replace the empty elements — flyingmouse, Jan 11 '16 at 12:00
Your input data is not a valid python nested list. Can you show the original csv and your parsing code? — Paul Rooney, Jan 11 '16 at 12:05
You are probably using numPy, or Pandas? That is important information :-) — Ward, Jan 11 '16 at 12:07

C.LECLERC · Accepted Answer · 2016-01-11T12:50:11.820

1

what about :

for line in csvdata:
    multiplicity = max([len(datum.split(" ")) if isinstance(datum, str) else 0 for datum in line])
    for datum in line:
        if(not isinstance(datum, str)):
            datum = " ".join(["-1"]*multiplicity)

It looks awful to me, but it should works.

edited Jan 11 '16 at 12:50

answered Jan 11 '16 at 12:10

C.LECLERC

510
3
12

There is an error because `datum.split(" ")` does not work here. – flyingmouse Jan 11 '16 at 12:31
multiplicity = max([len(datum.split(" ")) for datum in line]) AttributeError: 'float' object has no attribute 'split' – flyingmouse Jan 11 '16 at 12:36
i edited my answer, but the code is uglier than before. – C.LECLERC Jan 11 '16 at 12:40
i reedited my answer to add the NaN test condition. it assume that all values that is not a string must be replace with corresponding "-1" – C.LECLERC Jan 11 '16 at 12:51
I modified your code ,using a `temp` to store the datum and write in a newline. So I can get the result. – flyingmouse Jan 12 '16 at 02:57

score 0 · Answer 2 · answered Jan 11 '16 at 12:39

try this:

xs=[["nan '1 1' '4 4' nan nan nan '2 2' nan"],
    ["nan '3' '3' nan nan nan '3' nan"],
    ["nan '1 1 1 1' '1 1 1 1' nan nan nan '2 2 2 2' nan"],
    ["nan '2' '4' nan nan nan '2' nan"],
    ["nan '4' '4' nan nan nan nan '4'"]]

for x in xs:
   s = len(x[0].replace('nan','').replace(' ','').split("''")[0])-1
   r = ' '.join('v'*s).replace('v', '-1')
   r = "'%s'" % r
   x[0] = x[0].replace('nan', r)

score 0 · Answer 3 · answered Jan 11 '16 at 14:28

I believe you should make clear in your question you are using a library (numPy). Most of the solutions will work for Python, but as you are already using numpy, this is a better solution I believe

x = np.asarray(pd.read_csv("data/org8.csv"))
x[np.isnan(x)] = -1

How to replace empty string in python

3 Answers3