How to flatten a struct in a Spark dataframe?

This should work in Spark 1.6 or later:

df.select(df.col("data.*"))

or

df.select(df.col("data.id"), df.col("data.keyNote"), df.col("data.details"))