org.apache.spark.sql.execution.columnar.encoding
MutatedColumnDecoder
Related Docs:
object MutatedColumnDecoder
| package encoding
org.apache.spark.sql.execution.columnar.encoding
Decodes a column of a batch that has seen some changes (updates or deletes) by combining all delta values, delete bitmask and full column value obtained from ColumnDeltaEncoders and column encoders. Callers should provide this with the set of all deltas and delete bitmask for the column apart from the full column value, then use it like a normal decoder.
Only the first column being decoded should set the "deleteBuffer", if present, and generated code should check the return value of "next*" methods to continue if it is negative.
To create an instance, use the companion class apply method which will create a nullable or non-nullable version as appropriate.