Using pandas.Series.str.get: what is the correct way?

Question

I am following Wes Mckinney's wonderful book to get up to speed with pandas. I however can't seem to get why pandas.Series.str.get won't work. I've looked at a few Github issues and questions on here but none seems to help.

Data

data = pd.Series({'Dave': 'dave@google.com', 'Steve': 'steve@gmail.com', 'Rob': 'rob@yahoo.com', 'Wes': np.nan}
)

Code

import pandas as pd
import re
import numpy as np
pattern = '[a-zA-Z0-9]+@.*'
matches = data.str.match(pattern)
matches.str.get(1)

The above code should work and result in something like:

Dave NaN
Rob  NaN
Steve NaN

I did use a different regex pattern than used in the book but don't think that's the issue.

ERROR:

raise AttributeError("Can only use .str accessor with string " "values!") AttributeError: Can only use .str accessor with string values

What am I missing? I am using pycharm community and python 3.6.6, pandas Version: 0.24.2 if that makes a difference.

Here's a screenshot from the book:

Answer 1

The reason you get a series containing NaNs is because matches is a boolean Series:

In[58]:
matches

Out[58]: 
Dave     True
Steve    True
Rob      True
Wes       NaN
dtype: object

So it doesn't make sense to return an element at the ordinal position in this case, hence why you get a Series of NaNs.

If you look at the example in the docs: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.str.get.html#pandas.Series.str.get

In[61]:
s = pd.Series(["String",
...               (1, 2, 3),
...               ["a", "b", "c"],
...               123,
...               -456,
...               {1: "Hello", "2": "World"}])
s

Out[61]: 
0                        String
1                     (1, 2, 3)
2                     [a, b, c]
3                           123
4                          -456
5    {1: 'Hello', '2': 'World'}
dtype: object

In[62]:
s.str.get(1)

Out[62]: 
0        t
1        2
2        b
3      NaN
4      NaN
5    Hello
dtype: object

So here it's returning the element at the ordinal position for each row, You can see that for some rows there is no 2nd element so it returns NaN.