Github user kiszk commented on a diff in the pull request: https://github.com/apache/spark/pull/21912#discussion_r209878827 --- Diff: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala --- @@ -735,70 +735,98 @@ class CodegenContext { } /** - * Generates code creating a [[UnsafeArrayData]]. + * Generates code creating a [[UnsafeArrayData]] or [[GenericArrayData]] based on + * given parameters. * * @param arrayName name of the array to create + * @param elementType data type of the elements in source array * @param numElements code representing the number of elements the array should contain - * @param elementType data type of the elements in the array * @param additionalErrorMessage string to include in the error message + * @param elementSize optional value which shows the size of an element of the allocated + * [[UnsafeArrayData]] or [[GenericArrayData]] + * + * @return code representing the allocation of [[ArrayData]] + * code representing a setter of an assignment for the generated array */ - def createUnsafeArray( + def createArrayData( --- End diff -- I think that it would be good to have a utility method to generate this call. WDYT? On the other hand, I agree that we have a method `setArrayElementFunc` or `setArrayElement`. Thus, we can split this method into two methods that have only one return value.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org