Github user ericl commented on a diff in the pull request:
https://github.com/apache/spark/pull/14750#discussion_r86657628
--- Diff:
sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveExternalCatalog.scala ---
@@ -583,6 +633,50 @@ private[spark] class HiveExternalCatalog(conf:
SparkConf, hadoopConf: Configurat
tableWithStats.copy(properties = getOriginalTableProperties(table))
}
+ private def restoreHiveSerdeTable(table: CatalogTable): CatalogTable = {
+ val hiveTable = table.copy(
+ provider = Some(DDLUtils.HIVE_PROVIDER),
+ tracksPartitionsInCatalog = true)
+
+ val schemaFromTableProps = getSchemaFromTableProperties(table)
+ if (DataType.equalsIgnoreCaseAndNullability(schemaFromTableProps,
table.schema)) {
+ hiveTable.copy(
+ schema = schemaFromTableProps,
+ partitionColumnNames =
getPartitionColumnsFromTableProperties(table),
+ bucketSpec = getBucketSpecFromTableProperties(table))
+ } else {
+ // Hive metastore may change the table schema, e.g. schema
inference. If the table
+ // schema we read back is different(ignore case and nullability)
from the one in table
+ // properties which was written when creating table, we should
respect the table schema
+ // from hive.
+ logWarning(s"The table schema given by Hive
metastore(${table.schema.simpleString}) is " +
+ "different from the schema when this table was created by Spark
SQL" +
+ s"(${schemaFromTableProps.simpleString}). We have to trust the
table schema from Hive " +
--- End diff --
Nit: clarify that we are falling back to the hive metastore schema.
Trusting it sounds a little ambiguous.
---
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]
