Github user shivaram commented on a diff in the pull request:
https://github.com/apache/spark/pull/12836#discussion_r66713462
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/RelationalGroupedDataset.scala ---
@@ -381,6 +385,50 @@ class RelationalGroupedDataset protected[sql](
def pivot(pivotColumn: String, values: java.util.List[Any]):
RelationalGroupedDataset = {
pivot(pivotColumn, values.asScala)
}
+
+ /**
+ * Applies the given serialized R function `func` to each group of data.
For each unique group,
+ * the function will be passed the group key and an iterator that
contains all of the elements in
+ * the group. The function can return an iterator containing elements of
an arbitrary type which
+ * will be returned as a new [[DataFrame]].
+ *
+ * This function does not support partial aggregation, and as a result
requires shuffling all
+ * the data in the [[Dataset]]. If an application intends to perform an
aggregation over each
+ * key, it is best to use the reduce function or an
+ * [[org.apache.spark.sql.expressions#Aggregator Aggregator]].
+ *
+ * Internally, the implementation will spill to disk if any given group
is too large to fit into
+ * memory. However, users must take care to avoid materializing the
whole iterator for a group
+ * (for example, by calling `toList`) unless they are sure that this is
possible given the memory
+ * constraints of their cluster.
+ *
+ * @since 2.0.0
+ */
+ private[sql] def flatMapGroupsInR(
+ f: Array[Byte],
+ packageNames: Array[Byte],
+ broadcastVars: Array[Object],
+ outputSchema: StructType): DataFrame = {
+ val broadcastVarObj =
broadcastVars.map(_.asInstanceOf[Broadcast[Object]])
+ val groupingNamedExpressions = groupingExprs.map(alias)
+ val groupingCols = groupingNamedExpressions.map(Column(_))
+ val groupingDataFrame = df.select(groupingCols : _*)
+ val groupingAttributes = groupingNamedExpressions.map(_.toAttribute)
+ val realOutputSchema = if (outputSchema == null)
SERIALIZED_R_DATA_SCHEMA else outputSchema
--- End diff --
If schema should not be null can we assert it to be not null on the R side
and just pass in a non-null value always ? I think for `dapply` we wanted to
support `collect` on the result of the UDF which could work even without a
schema.
The other nice way to handle this would be to do construct the binary
schema that we fall back on from the R side and pass that in (i.e. keeping all
input validation in R and just logic in scala)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
