Github user patrickmcgloin commented on a diff in the pull request:
https://github.com/apache/spark/pull/21671#discussion_r199319246
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/json/JacksonGenerator.scala
---
@@ -163,6 +175,22 @@ private[sql] class JacksonGenerator(
s"with the type of $dataType to JSON.")
}
+ private def makeReader(dataType: DataType): ValueReader = dataType match
{
+
+ case TimestampType =>
+ (row: SpecializedGetters, ordinal: Int) =>
+
options.timestampFormat.format(DateTimeUtils.toJavaTimestamp(row.getLong(ordinal)))
+
+ case DateType =>
+ (row: SpecializedGetters, ordinal: Int) =>
+
options.dateFormat.format(DateTimeUtils.toJavaDate(row.getInt(ordinal)))
+
+ case _ =>
+ (row: SpecializedGetters, ordinal: Int) =>
+ row.get(ordinal, dataType).toString
--- End diff --
I considered this the default case and thought it would be covered by
existing Unit Tests. Which types should I add tests for? Int, String, Long,
etc. All of the children of AtomicType?
Nested types is interesting. I will add some tests for keys with a
Seq[Int], Map[String, Int] to see that they work. If there was something else
you were thinking of please let me know.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
