[SPARK-24669][SQL] Invalidate tables in case of DROP DATABASE CASCADE

## What changes were proposed in this pull request? Before dropping database refresh the tables of that database, so as to refresh all cached entries associated with those tables. We follow the same when dropping a table. UT is added Closes apache#23905 from Udbhav30/SPARK-24669. Authored-by: Udbhav30 <u.agrawal30@gmail.com> Signed-off-by: Dongjoon Hyun <dhyun@apple.com> (cherry picked from commit 9bddf71) Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
kai-chi · Jul 25, 2019 · caaf93f · caaf93f
1 parent aef2585
commit caaf93f
Show file tree

Hide file tree

Showing 2 changed files with 42 additions and 1 deletion.
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/catalog/SessionCatalog.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/catalog/SessionCatalog.scala
@@ -218,6 +218,11 @@ class SessionCatalog(
     if (dbName == DEFAULT_DATABASE) {
       throw new AnalysisException(s"Can not drop default database")
     }
+    if (cascade && databaseExists(dbName)) {
+      listTables(dbName).foreach { t =>
+        invalidateCachedTable(QualifiedTableName(dbName, t.table))
+      }
+    }
     externalCatalog.dropDatabase(dbName, ignoreIfNotExists, cascade)
   }
 

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLSuite.scala
@@ -17,7 +17,7 @@
 
 package org.apache.spark.sql.execution.command
 
-import java.io.File
+import java.io.{File, PrintWriter}
 import java.net.URI
 import java.util.Locale
 
@@ -2715,4 +2715,40 @@ abstract class DDLSuite extends QueryTest with SQLTestUtils {
       }
     }
   }
+
+  test("Refresh table before drop database cascade") {
+    withTempDir { tempDir =>
+      val file1 = new File(tempDir + "/first.csv")
+      val writer1 = new PrintWriter(file1)
+      writer1.write("first")
+      writer1.close()
+
+      val file2 = new File(tempDir + "/second.csv")
+      val writer2 = new PrintWriter(file2)
+      writer2.write("second")
+      writer2.close()
+
+      withDatabase("foo") {
+        withTable("foo.first") {
+          sql("CREATE DATABASE foo")
+          sql(
+            s"""CREATE TABLE foo.first (id STRING)
+               |USING csv OPTIONS (path='${file1.toURI}')
+             """.stripMargin)
+          sql("SELECT * FROM foo.first")
+          checkAnswer(spark.table("foo.first"), Row("first"))
+
+          // Dropping the database and again creating same table with different path
+          sql("DROP DATABASE foo CASCADE")
+          sql("CREATE DATABASE foo")
+          sql(
+            s"""CREATE TABLE foo.first (id STRING)
+               |USING csv OPTIONS (path='${file2.toURI}')
+             """.stripMargin)
+          sql("SELECT * FROM foo.first")
+          checkAnswer(spark.table("foo.first"), Row("second"))
+        }
+      }
+    }
+  }
 }