Hi, Spark Users
I'm trying to update registered DataFrame temp table by invoke
DataFrame.registerTempTable again and again.
Assume I have DataFrame temp table "table1", below is the concurrent logic
sqlContext.table("table1").filter(***).unionAll(dummy1DF).registerTempTable("table1")
sqlContext.table("table1").filter(***).unionAll(dummy2DF).registerTempTable("table1")
And I expect after above logic completed, the final temp table "table1" contain
all the changes.
Is this a correct expectation even without proper locking system to make the
executing ordered?
Such updating logic seems not functional programming, correct?
Thanks a lot
prosp4300
