Function that works as a try except lambda function in a dataframe-CodePudding

I'm trying to clean the column of a dataframe so only the first name is left.

What I'm currently trying is splitting each value of the column into a list, and then gathering the [0] of the list, if the element contains a comma, then get the [1].

I try the code below and it works perfectly, except for some outlier values that only have the last name with a comma, but no first name as shown in the examples below:

What I used:

msgDFt['From Name'] = msgDFt['From Name'].str.split(' ')
msgDFt['From Name'] = msgDFt['From Name'].apply(lambda row: row[0] if ',' not in row[0] else row[1])

Now I'm aware that lambda functions do not have the feature of having a try-except, the next thing I tried was a function:

def firstNameMod(name):
    for n in name:
        if n[0] == None:
            name = 'NOT FOUND'
        elif ',' in n[0]:
            name = name[1]
        elif ',' in n[0] and n[1] == None:
            name = name[0]
        elif n[0] != False:
            name = name[0]
        
df.apply(firstNameMod(df['Name']))

This did not work because the column I'm selected is being read as a NoneType.

What I have:

    Name
0   Robert Marin
1   Katherine Ortiz
2   Sloth, Herbert
3   Perez,

What I want:

    Name
0   Robert
1   Katherine
2   Herbert
3   NaN

sample dataframe:

names = {'Name': ['Robert Marin','Katherine Ortiz', 'Sloth, Herbert','Perez,']}
df = pd.DataFrame(names)

CodePudding user response：

You can use a regex to extract your first name:

df['Name'].str.extract('(^\w (?=[^,]*$)|(?<=, )\w )')[0]

output:

0       Robert
1    Katherine
2      Herbert
3          NaN