在 pd.dataframe 系列上使用.find()

Question

I have the following df:

data = {'Org':  ['<a href="/00xO" target="_blank">Chocolate</a>'],
        'Owner': ['Charlie']
        }

df = pd.DataFrame(data)

print (df)

and when I apply the lamba function below instead of giving me 'Chocolate' it's returning 0.

df['Correct Org']=df['Org'].apply(lambda st: st[st.find(">"):st.find("<")])

I've tried adding 'str' as follows:

df['Correct Org']=df['Org'].str.apply(lambda st: st[st.find(">")+1:st.find("<")])

& get the following error:

AttributeError: 'StringMethods' object has no attribute 'apply'

Answer 1

You're getting None returned because df['Org'][0].find(">") returns 31 but df['Org'][0].find("<") returns 0. So it's not clear what st[st.find(">"):st.find("<") means. You can use bs4.BeautifulSoup to create a soup object and get the text inside a directly:

from bs4 import BeautifulSoup
df['Org'] = df['Org'].apply(lambda x: BeautifulSoup(x).text)

Output:

         Org    Owner
0  Chocolate  Charlie

Answer 2

Use BeautifulSoup for parsing html tags:

from bs4 import BeautifulSoup

df['Correct Org']=df['Org'].apply(lambda st: ','.join(BeautifulSoup(st, features="lxml").findAll(text=True)))

Answer 3

If you don't want to use BeautifulSoup , I wrote a simple function for you.

A FUNCTION FOR GETTING THE LINK TEXT

def getOrg(link):
    link = str(link)
    link = link[link.find('>'):link.find("</")]
    return link.replace(link[0], '')

FOR EXAMPLE

import pandas as pd

data = {'Org':  ['<a href="/00xO" target="_blank">Chocolate</a>'],
        'Owner': ['Charlie']
        }

df = pd.DataFrame(data)


# Function Call
getOrg(df['Org'])

OUTPUT

Chocolate