Cleaning rows of special characters and creating dataframe columns

Question

Below is my Dataframe format consisting of 2 columns (one is index and other is the data to be cleaned)

 0  ans({'A,B,C,Bad_QoS,Sort'})
 1  ans({'A,B,D,QoS_Miss1,Sort'})

1. I want to remove special characters
2. create a data frame for all comma separated items.

I have managed to first remove ans from all rows using:

ds_[col2] = ds_.replace('ans', '', regex=True)

> 0 ({'A,B,C,Bad_QoS,Sort'})
> 1 ({'A,B,D,QoS_Miss1,Sort'})

Then I try to apply replace regex, see below:

ds_['col2'] = ds_['col2'].str.replace( r' \(\{\' | \'\}\) ', '', regex=True)
ds_['col2'] 

I get no errors, but no changes.

How could I clean these characters and also create a data frame like below:

col1 col2 col3 col4  col5  col6 

0    A    B    C    Bad_Qos  Sort
1    A    B    D    Qos_Miss1 Sort

Answer 1

You can first split on ', and then on , and then remove unnecessary columns:

first

df[['col_21','col_22','col_23']] = df['col2'].str.split("'",expand=True)

and

df[['col_2','col_3','col_4','col_5','col_6']] = df['col_22'].str.split(",",expand=True)

finally remove rest.