BUG: Series.Count() miscounts NA values in ExtensionArrays

Code Sample, a copy-pastable example if possible

In [9]: import numpy as np
   ...: import pandas as np
   ...: from pandas.tests.extension.decimal.array import to_decimal
   ...: ser = pd.Series(to_decimal([1,2,3]))
   ...: ser[0]=np.nan
   ...: ser.count()
3

[TomAugspurger]

Should be able to change the values_from_object to `self.notna().sum()`. Interested in making a PR?

…

On Thu, Jun 13, 2019 at 1:15 PM pilkibun ***@***.***> wrote: Code Sample, a copy-pastable example if possible In [9]: import numpy as np ...: import pandas as np ...: from pandas.tests.extension.decimal.array import to_decimal ...: ser = pd.Series(to_decimal([1,2,3])) ...: ser[0]=np.nan ...: ser.count()3 — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#26835?email_source=notifications&email_token=AAKAOIXX3TRJXS4TZCDEME3P2KFCJA5CNFSM4HX4SWC2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4GZMH3YA>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAKAOIVIM7QDLKWKLEK6Z7TP2KFCJANCNFSM4HX4SWCQ> .

already have one lined up, just opened this for a whatsnew reference. we mentioned this in #26730