Regex in pandas to find a match based on string in another column

Question

I have a dataframe of which this is a part.

   CodeID    Codes
0  'code1'   '[code1(a,b,c)][code2(c,d,e)][code3(e,f,g)]'   ...
1  'code2'   '[code1(a,b,c)][code2(c,d,e)][code3(e,f,g)]'   ...
2  'code3'   '[code1(a,b,c)][code2(c,d,e)][code3(e,f,g)]'   ...
...

What I'm trying to do is extract the part of the string in column Codes that matches the pattern r"\[<code in CodeID column>[^][]*\]"

Something like:

df['Code'] = df['Codes'].str.find(r"\[<code in CodeID column>[^][]*\]")

This recent question seems to imply it's not possible in a vectorised way but it's not exactly the same situation.

If it is possible, then the regex will look like r"\[<code in CodeID column>[^][]*\]" — Wiktor Stribiżew
– Wiktor Stribiżew, Commented Dec 3, 2015 at 17:55
Thanks. I'm always blind to regex and leave that part of debugging till last! — Jamie Bull
– Jamie Bull, Commented Dec 3, 2015 at 17:56

WoodChopper · Accepted Answer · 2015-12-04 07:25:08Z

2

We can certainly use string from one column to compare another like below,

In lambda expression x[0] is codeID and x[1] is codes.

import re
import pandas as pd

Out[20]: 
    CodeID                                         Codes
0  'code1'  '[code1(a,b,c)][code2(c,d,e)][code3(e,f,g)]'
1  'code2'  '[code1(a,b,c)][code2(c,d,e)][code3(e,f,g)]'
2  'code3'  '[code1(a,b,c)][code2(c,d,e)][code3(e,f,g)]'

df[['CodeID','Codes']].apply(lambda x: re.match(r"\[%s[^][]*\]"%x[0], x[1]),axis=1)
Out[21]: 
0    None
1    None
2    None
dtype: object

Well it returns None because of my bad regex skills :)

edited Dec 4, 2015 at 7:25

answered Dec 3, 2015 at 19:16

WoodChopper

4,4056 gold badges34 silver badges58 bronze badges

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

Collectives™ on Stack Overflow

Regex in pandas to find a match based on string in another column

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related