Efficiently drop dataframe rows where *index* contains nulls (NaT)

Question

My data files infrequently contain malformed lines (think abrupt power loss). When badness occurs across a timestamp, making it uninterpretable, the resulting DataFrame.Index contains Not-a-Time (NaT) values (because I've coerced it to).

My real problem is that instances of NaT prevent the use of resample. I need to remove them, first. Unfortunately, I haven't figured out if/how to use dropna on the index itself. It's looking more and more like I need to make the index a column, operate on it, then re-make it the index. (But I don't want to do that.)

Is there an established idiom for dropping dataframe rows where the Index values are null?

Answer 1

Use

df.loc[pd.notnull(df.index)]

for newer versions you can do

df.loc[df.index.notnull()]

or

df.loc[df.index.dropna()]