Description
A DataFrame is essentially a Dataset[Row]. However, to keep binary compatibility, DataFrame didn't extend from Dataset[Row] in 1.6.
In Spark 2.0, they should be unified to minimize concepts.
Attachments
Issue Links
- blocks
-
SPARK-13822 Follow-ups of DataFrame/Dataset API unification
- Resolved
- is blocked by
-
SPARK-13457 Remove DataFrame RDD operations
- Resolved
- is duplicated by
-
SPARK-13458 Datasets cannot be sorted
- Resolved
- is related to
-
SPARK-13458 Datasets cannot be sorted
- Resolved
- links to
(1 links to)