In [1]:

    

import pandas as pd


    

In [2]:

    

df = pd.DataFrame({'s': ['X,Y,Z', 'X', 'XY,Y', 'Y,Z,XY']},
                  index=['a', 'b', 'c', 'd'])


    

In [3]:

    

print(df)


    

    




        s
a   X,Y,Z
b       X
c    XY,Y
d  Y,Z,XY

In [4]:

    

df['l'] = df['s'].str.split(',')
print(df)


    

    




        s           l
a   X,Y,Z   [X, Y, Z]
b       X         [X]
c    XY,Y     [XY, Y]
d  Y,Z,XY  [Y, Z, XY]

In [5]:

    

print(df.dtypes)


    

    




s    object
l    object
dtype: object

In [6]:

    

print(type(df.at['a', 's']))


    

    




<class 'str'>

In [7]:

    

print(type(df.at['a', 'l']))


    

    




<class 'list'>

In [8]:

    

print(df['s'].apply(lambda x: [s.strip() for s in x.split(',')]))


    

    




a     [X, Y, Z]
b           [X]
c       [XY, Y]
d    [Y, Z, XY]
Name: s, dtype: object

In [9]:

    

print(df['l'].apply(len))


    

    




a    3
b    1
c    2
d    3
Name: l, dtype: int64

In [10]:

    

print(df['l'].apply(sorted))


    

    




a     [X, Y, Z]
b           [X]
c       [XY, Y]
d    [XY, Y, Z]
Name: l, dtype: object

In [11]:

    

print(df['l'].apply(lambda x: ','.join(x)))


    

    




a     X,Y,Z
b         X
c      XY,Y
d    Y,Z,XY
Name: l, dtype: object

In [12]:

    

print(df['l'].apply(','.join))


    

    




a     X,Y,Z
b         X
c      XY,Y
d    Y,Z,XY
Name: l, dtype: object

In [13]:

    

print(df['l'].apply(lambda x: ','.join(sorted(x))))


    

    




a     X,Y,Z
b         X
c      XY,Y
d    XY,Y,Z
Name: l, dtype: object

In [14]:

    

df['l'].apply(lambda x: x.append('A'))
print(df)


    

    




        s              l
a   X,Y,Z   [X, Y, Z, A]
b       X         [X, A]
c    XY,Y     [XY, Y, A]
d  Y,Z,XY  [Y, Z, XY, A]

In [15]:

    

df['l'].apply(lambda x: x.remove('Z') if 'Z' in x else x)
print(df)


    

    




        s           l
a   X,Y,Z   [X, Y, A]
b       X      [X, A]
c    XY,Y  [XY, Y, A]
d  Y,Z,XY  [Y, XY, A]

In [16]:

    

print(df['l'].apply(lambda x: 'X' in x))


    

    




a     True
b     True
c    False
d    False
Name: l, dtype: bool

In [17]:

    

print(df[df['l'].apply(lambda x: 'X' in x)])


    

    




       s          l
a  X,Y,Z  [X, Y, A]
b      X     [X, A]

In [18]:

    

print(df['s'].str.contains('Z'))


    

    




a     True
b    False
c    False
d     True
Name: s, dtype: bool

In [19]:

    

print(df[df['s'].str.contains('Z')])


    

    




        s           l
a   X,Y,Z   [X, Y, A]
d  Y,Z,XY  [Y, XY, A]