William1104 commented on a change in pull request #24221: [SPARK-27248][SQL]
`refreshTable` should recreate cache with same cache name and storage level
URL: https://github.com/apache/spark/pull/24221#discussion_r282959517
##########
File path:
sql/hive/src/test/scala/org/apache/spark/sql/hive/CachedTableSuite.scala
##########
@@ -361,4 +352,87 @@ class CachedTableSuite extends QueryTest with
SQLTestUtils with TestHiveSingleto
assert(spark.sharedState.cacheManager.lookupCachedData(samePlan).isDefined)
}
}
+
+ test("SPARK-27248 refreshTable should recreate cache with same cache name
and storage level") {
+
+ // This section tests when a table is cached with its qualified name but
its is refreshed with
+ // its unqualified name.
+ withTempPath { path =>
+ withTempDatabase { db =>
+ withTable(s"$db.cachedTable") {
+ withCache(s"$db.cachedTable") {
+
+ // Create table 'cachedTable' in default db for testing purpose.
+ sql(s"CREATE TABLE $db.cachedTable(key STRING) USING PARQUET
LOCATION '${path.toURI}'")
Review comment:
Sorry about this. The SQL is update to:
- s"CREATE TABLE $db.cachedTable AS SELECT 1 AS key"
- "CREATE TABLE cachedTable AS SELECT 1 AS key"
I also modified the function `withCache` in SQLTestUtils to avoid exception
masking. Before the modification, any exception throwing from finally block may
mask the exception throwing from try block. I hope the created function
`withFinallyBlock` does the right job.
----------------------------------------------------------------
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]
