compare two string columns in dataframe row-wise

Question

Problem description: i need to set a variable for each line, but only if it is within the range of a list in a second column in the same row.

Sample Dataframe:

df = pd.Dataframe({'col1': ['A', 'T' , 'P', 'Z'], 'col2': ['A, B, C', 'D, E, F' , 'G, H, I, P', 'M, N, R, ZGTR']})

i need to recieve all rows where col1 is part of col2. expected result:

col1    col2
'A'     'A, B, C'
'P'     'G, H, I, P'

My approach which returns a TypeError about Series objects being mutable and can not be hashed:

df[df['col2'].str.match(df['col1'])]

As far as i understand i have to point out somehow that the compare should be done within one row. I know itterrows would be an solution but i would prefer something without looping.

What does it mean for strings to be "within the range"? Is Z "within the range" of [M, N, R, ZGTR]? What is the criteria for inclusion? — DYZ
– DYZ, Commented Jun 19, 2020 at 7:08

jezrael · Accepted Answer · 2020-06-19 07:07:40Z

2

Use list comprehension with test by in with splitted values:

import pandas as pd

df = pd.DataFrame({'col1': ['A', 'T' , 'P', 'Z'], 
                   'col2': ['A, B, C', 'D, E, F' , 'G, H, I, P', 'M, N, R, ZGTR']})
df = df[[b in a.split(', ') for a, b in df[['col2', 'col1']].values]]
print (df)
  col1        col2
0    A     A, B, C
2    P  G, H, I, P

edited Jun 19, 2020 at 7:07

answered Jun 19, 2020 at 6:57

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

himself Over a year ago

Your solution works for the given example. Problem is that i have longer strings than given in example. I will adjust

jezrael Over a year ago

@himself - So is necessary test in splitted values by , ? Then was added split(', ')

Collectives™ on Stack Overflow

compare two string columns in dataframe row-wise

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related