Python, Pandas: What it is the best way to separate string into several columns?-CodePudding

I want to split one string into several columns however I encounter issue with str.split method.

Example: While performing str.split on column Name which consist such strings it work as expected.

data1 = {'Name': ['Alice 23', 'Philip 12', 'Krish 64', 'John 29']}
df1 = pd.DataFrame(data1)
df1
    Name
0   Alice 23
1   Philip 12
2   Krish 64
3   John 29

performing split:

df1[['Name', 'age']] = df1['Name'].str.split(' ', 1, expand=True)
df1

Name    age
0   Alice   23
1   Philip  12
2   Krish   64
3   John    29

All good, as I wanted but if I need to place other separator like || its not working properly.

data2 = {'Name': ['Alice||23', 'Philip||12', 'Krish||64', 'John||29']}
df2 = pd.DataFrame(data)
df2

    Name
0   Alice||23
1   Philip||12
2   Krish||64
3   John||29

performing split...

df2[['Name', 'age']] = df2['Name'].str.split('[||]',1,expand = True)
df3[['Name', 'age']] = df2['Name'].str.split('||',1,expand = True)

results in not what I expected

df2
    Name    age
0   Alice   |23
1   Philip  |12
2   Krish   |64
3   John    |29

df3

Name    age
0       Alice||23
1       Philip||12
2       Krish||64
3       John||29

What's the reason for this behaviour and how get expected result which is as df1?

CodePudding user response：

The issue you're hitting is that pandas defaults to assuming your string to split on is a regular expression. In regular expressions the "|" character is a special character that enables you to match either expression on the left or right of that character (e.g. you can match 'a' or 'b' with the expression '(a|b)'.

In your case, we don't want to pass a regular expression, so you can pass .str.split(…, regex=False).

>>> df2['Name'].str.split('||', regex=False, expand=True)
        0   1
0   Alice  23
1  Philip  12
2   Krish  64
3    John  29

CodePudding user response：

escape the |:

df2['Name'].str.split('\|\|',expand = True)

Output:

        0   1
0   Alice  23
1  Philip  12
2   Krish  64
3    John  29