Python : Checking the length of pandas column value for if condition

Question

I have a python program which process multiple files. Each file have customer id column based on department. Some file has 8 digit customer id and some have 9 digit customer id. I need to do this

if length of customer id column value is == 8:
    input_file['customer_id'] = input_file['customer_id'].str[0:2] + '-' + input_file['customer_id'].str[2:8]
if length of customer id column value is == 9:
    input_file['customer_id'] = 'K' + '-' + input_file['customer_id'].str[0:3] + '-' + input_file['customer_id'].str[3:8]

Input

id cusotmerid
1  89898988
2  898989889

Output

id cusotmerid
1  89-898988
2  K-898-989889

How can I achieve it. Unable to find anything which can do this

Joe Ferndz · Accepted Answer · 2020-09-30 00:26:29Z

3

You can also do something like this.

You can use pd.Series.map with len built-in to find out the length of the column value. With that you can determine how to assign the value.

Using astype(str) converts the numeric value to string so you can do string concatenation.

input_file.loc[input_file['cusotmerid'].astype(str).map(len) == 8, 'new_customer_id'] = input_file['cusotmerid'].astype(str).str[:2]+ '-' + input_file['cusotmerid'].astype(str).str[2:]

input_file.loc[input_file['cusotmerid'].astype(str).map(len) == 9, 'new_customer_id'] = 'K-' + input_file['cusotmerid'].astype(str).str[:3] + '-' + input_file['cusotmerid'].astype(str).str[3:]

This should take are of the new value assignments.

The output of this is:

   cusotmerid new_customer_id
0    89898988       89-898988
1   898989889    K-898-989889

edited Sep 30, 2020 at 0:26

answered Sep 29, 2020 at 23:56

Joe Ferndz

8,5282 gold badges15 silver badges37 bronze badges

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

3 Comments

David Erickson Over a year ago

I use .apply(lambda x: len(x) == all of the time. .map(len) is MUCH better. + 1

Joe Ferndz Over a year ago

apply() is a bit slower than inbuilt functions. so i try to use what's already available.

David Erickson Over a year ago

exactly, I wasn't aware of this one. Apply should be a last resort if there are no available methods.

David Erickson · Accepted Answer · 2020-09-29 23:53:02Z

you can use np.select. In order to check the length of the string, you have to first make sure that the column's format is a string, hence .astype(str). Then, you can use .apply(lambda x: len(x) == condition) to to return a result based on the condition:

import numpy as np
input_file['cusotmerid'] = input_file['cusotmerid'].astype(str)
input_file['cusotmerid'] = np.select([input_file['cusotmerid'].apply(lambda x: len(x) == 8),
                                     input_file['cusotmerid'].apply(lambda x: len(x) == 9)],
                                     [input_file['cusotmerid'].str[0:2] + '-' + input_file['cusotmerid'].str[2:8],
                                     'K' + '-' + input_file['cusotmerid'].str[0:3] + '-' + input_file['cusotmerid'].str[3:9]],
                                      input_file['cusotmerid'])
input_file

    id  cusotmerid
0   1   89-898988
1   2   K-898-989889

It may be easier to break the np.select statement down into conditions and results. The 3 parameters I pass are conditions, results, and default value if no conditions are met.

input_file['cusotmerid'] = input_file['cusotmerid'].astype(str)

c1 = input_file['cusotmerid'].apply(lambda x: len(x) == 8)
c2 = input_file['cusotmerid'].apply(lambda x: len(x) == 9)

conditions = [c1, c2]

r1 = input_file['cusotmerid'].str[0:2] + '-' + input_file['cusotmerid'].str[2:8]
r2 = 'K' + '-' + input_file['cusotmerid'].str[0:3] + '-' + input_file['cusotmerid'].str[3:9]

results = [r1,r2]

input_file['cusotmerid'] = np.select(conditions, results, input_file['cusotmerid'])
input_file

    id  cusotmerid
0   1   89-898988
1   2   K-898-989889

Collectives™ on Stack Overflow

Python : Checking the length of pandas column value for if condition

2 Answers 2

3 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related