gaborgsomogyi commented on a change in pull request #24173: [SPARK-27237][SS]
Introduce State schema validation among query restart
URL: https://github.com/apache/spark/pull/24173#discussion_r268117156
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/state/StateStore.scala
##########
@@ -477,5 +486,82 @@ object StateStore extends Logging {
None
}
}
-}
+ private class StateSchemaCompatibilityChecker(
+ providerId: StateStoreProviderId,
+ hadoopConf: Configuration) {
+
+ private val storeCpLocation = providerId.storeId.storeCheckpointLocation()
+ private val fm = CheckpointFileManager.create(storeCpLocation, hadoopConf)
+ private val schemaFileLocation = schemaFile(storeCpLocation)
+
+ fm.mkdirs(schemaFileLocation.getParent)
+
+ def check(keySchema: StructType, valueSchema: StructType): Unit = {
+ if (fm.exists(schemaFileLocation)) {
+ logDebug(s"Schema file for provider $providerId exists. Comparing with
provided schema.")
+ val (storedKeySchema, storedValueSchema) = readSchemaFile()
+
+ def typesEq(schema1: StructType, schema2: StructType): Boolean = {
+ val fieldToType: StructField => (DataType, Boolean) = f =>
(f.dataType, f.nullable)
+ (schema1.length == schema2.length) &&
+ schema1.map(fieldToType).equals(schema2.map(fieldToType))
Review comment:
Shouldn't be `==` instead of `equals`? Scala does more checks than java.
This question applies to other places too.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
