Github user yhuai commented on a diff in the pull request:
https://github.com/apache/spark/pull/13025#discussion_r62699825
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/command/ddl.scala ---
@@ -489,9 +486,83 @@ private[sql] object DDLUtils {
case _ =>
})
}
+
def isTablePartitioned(table: CatalogTable): Boolean = {
- table.partitionColumns.size > 0 ||
+ table.partitionColumns.nonEmpty ||
table.properties.contains("spark.sql.sources.schema.numPartCols")
}
-}
+ def getSchemaFromTableProperties(metadata: CatalogTable):
Option[StructType] = {
+ getSchemaFromTableProperties(metadata.properties)
+ }
+
+ // A persisted data source table may not store its schema in the
catalog. In this case, its schema
+ // will be inferred at runtime when the table is referenced.
+ def getSchemaFromTableProperties(props: Map[String, String]):
Option[StructType] = {
+ require(isDatasourceTable(props))
+
+ val schemaParts = for {
+ numParts <- props.get("spark.sql.sources.schema.numParts").toSeq
+ index <- 0 until numParts.toInt
+ } yield props.getOrElse(
+ s"spark.sql.sources.schema.part.$index",
+ throw new AnalysisException(
+ s"Corrupted schema in catalog: $numParts parts expected, but part
$index is missing."
+ )
+ )
+
+ if (schemaParts.isEmpty) {
+ None
+ } else {
+
Some(DataType.fromJson(schemaParts.mkString).asInstanceOf[StructType])
+ }
+ }
--- End diff --
I will make a commit to handle `spark.sql.sources.schema`.
---
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]
