Github user HyukjinKwon commented on a diff in the pull request:
https://github.com/apache/spark/pull/18875#discussion_r138058970
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/json/JacksonGenerator.scala
---
@@ -26,20 +26,53 @@ import
org.apache.spark.sql.catalyst.expressions.SpecializedGetters
import org.apache.spark.sql.catalyst.util.{ArrayData, DateTimeUtils,
MapData}
import org.apache.spark.sql.types._
+/**
+ * `JackGenerator` can only be initialized with a `StructType` or a
`MapType`.
+ * Once it is initialized with `StructType`, it can be used to write out a
struct or an array of
+ * struct. Once it is initialized with ``MapType``, it can be used to
write out a map or an array
+ * of map. An exception will be thrown if trying to write out a struct if
it is initialized with
+ * a `MapType`, and vice verse.
+ */
private[sql] class JacksonGenerator(
- schema: StructType,
+ dataType: DataType,
writer: Writer,
options: JSONOptions) {
// A `ValueWriter` is responsible for writing a field of an
`InternalRow` to appropriate
// JSON data. Here we are using `SpecializedGetters` rather than
`InternalRow` so that
// we can directly access data in `ArrayData` without the help of
`SpecificMutableRow`.
private type ValueWriter = (SpecializedGetters, Int) => Unit
+ // `JackGenerator` can only be initialized with a `StructType` or a
`MapType`.
+ dataType match {
+ case _: StructType | _: MapType =>
+ case _ => throw new UnsupportedOperationException(
+ s"`JacksonGenerator` only supports to be initialized with a
`StructType` " +
--- End diff --
This `s` could be removed too. Let's avoid backquoting classes in the error
messages.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
