熊猫提取和替换值

Question

I'm trying to clean a large pandas Dataframe by extracting a name from a text column and replacing the value in another column. I also only want to replace values where the extraction was successful. I was able to extract the name from the "text" column but struggling to replace the value in the "name" column. Looking for some suggestions.

Example DF:

df = pd.DataFrame({'text': {0: 'John', 1: 'A girl named Susan', 2: 'A man named David'},
                   'name': {0: 'John', 1: 'girl', 2: 'man'}})

                 text  name
0                John  John
1  A girl named Susan  girl
2   A man named David   man

Extracted Names:

print(df['text'].str.extract('((?<=named\s)\w+)'))

       0
0    NaN
1  Susan
2  David

Desired Output:

                 text   name
0                John   John
1  A girl named Susan  Susan
2   A man named David  David

Answer 1

Not sure if possible with real data, but one solution is replace missing values by original values of name column:

df['name'] = df['text'].str.extract('((?<=named\s)\w+)').fillna(df['name'])
print (df)
                 text   name
0                John   John
1  A girl named Susan  Susan
2   A man named David  David