In hive external tables and parquet files, can we have various data types but the same column names? For example, can I have all fields in parquet files be text and define the data types in hive tables to be decimal, integer, or boolean?
For eg: parquet schema:
root
|-- empId: string (nullable = true)
|-- refrenId: string (nullable = true)
|-- position: string (nullable = true)
|-- joiningdate: string (nullable = true)
hive :
|-- empId: integer (nullable = true)
|-- refrenId: integer (nullable = true)
|-- position: string (nullable = true)
|-- joiningdate: date(nullable = true)
If this is not feasible, how may this comparable capability with various data types be supported?