Hive comes with some useful built-in UDFs to process JSON objects.
To make Impala and Hive tables and quieries more interchangable, I am proposing porting these UDFs to be part Impala's built in functions:
Consider the following table raw_log
...and the following query:
The query should return the following results: