remove one charachter from a speficic column in df pandas-CodePudding

I have a df with a column that some values are having ... and some .. and some are without dots.

 Type    range
    Mike    10..13
    Ni      3..4
    NANA    2...3
    Gi      2

desired output should look like this

Type    range
Mike    10
Mike    11
Mike    12
MIke    13    
Ni      3
Ni      4
NANA    2
NANA    3
Gi      2

So dots represnt the range of between to number ( inclusive the end number).

How am I suppsoed to do it in pandas?

CodePudding user response：

Parse str as list first and then explode:

import re
def str_to_list(s):
    if not s: return []
    nums = re.split('\.{2,3}', s)
    if len(nums) == 1:
        return nums
    return list(range(int(nums[0]), int(nums[1])   1))

df['range'] = df['range'].astype(str).map(str_to_list)
df.explode('range')

   Type range
0  Mike    10
0  Mike    11
0  Mike    12
0  Mike    13
1    Ni     3
1    Ni     4
2  NANA     2
2  NANA     3
3    Gi     2

CodePudding user response：

An approach using numpy.arange with pandas.DataFrame.explode :

out = (
        df
          .assign(range=
                          df["range"]
                            .str.replace("\. ", "-", regex=True)
                            .str.split("-")
                            .apply(lambda x: np.arange(list(map(int, x))[0], list(map(int, x))[-1] 1, 1) if len(x)>1 else x))
          .explode("range", ignore_index=True)
        )

# Output :

print(out)

   Type range
0  Mike    10
1  Mike    11
2  Mike    12
3  Mike    13
4    Ni     3
5    Ni     4
6  NANA     2
7  NANA     3
8    Gi     2