Column names when raise a typeerror in pandas dataframe

Question

I'm comparing 2 dataframes by their row names and I want to get the name of the rows that don't match example:

check = sum(df1.index!=df2.index)
if check:
    raise TypeError("Not Match")
else:
    print("All OK")

When the index don't match I want to print the first instance of the row that didn't match. I've tried using a print statement as shown below but it doesn't execute.

    check = sum(df1.index!=df2.index)
if check:
    print(df1.index)
    raise TypeError("Not Match")
else:
    print("All OK")

Answer 1

The statement -

df1.index != df2.index

Returns a numpy array of bools . Let's retain this for now. It may need a slight change to work -

m = df1.index.values != df2.index.values

To get the counts of True (non-matches) in m , call sum -

if m.sum():
    raise TypeError(...)

To get the first instance of no-match, index with m :

print(df1[m].iloc[0])

Thanks to MaxU, here's a nicer solution using argmax -

print(df1[m.argmax()])