Description
Similar to SPARK-12747 but for DecimalType.
Do we need to handle precision and scale?
I've already starting trying to work on this. I cannot see if Postgres JDBC driver handles precision and scale or just converts to default BigDecimal precision and scale.