spark-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [spark] cloud-fan commented on a change in pull request #25368: [SPARK-28635][SQL] create CatalogManager to track registered v2 catalogs
Date Wed, 14 Aug 2019 09:37:34 GMT
cloud-fan commented on a change in pull request #25368: [SPARK-28635][SQL] create CatalogManager
to track registered v2 catalogs
URL: https://github.com/apache/spark/pull/25368#discussion_r313788299
 
 

 ##########
 File path: sql/catalyst/src/main/scala/org/apache/spark/sql/catalog/v2/CatalogManager.scala
 ##########
 @@ -0,0 +1,100 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.catalog.v2
+
+import scala.collection.mutable
+import scala.util.control.NonFatal
+
+import org.apache.spark.internal.Logging
+import org.apache.spark.sql.internal.SQLConf
+
+/**
+ * A thread-safe manager for [[CatalogPlugin]]s. It tracks all the registered catalogs, and
allow
+ * the caller to look up a catalog by name.
+ */
+class CatalogManager(conf: SQLConf) extends Logging {
+
+  private val catalogs = mutable.HashMap.empty[String, CatalogPlugin]
+
+  def catalog(name: String): CatalogPlugin = synchronized {
+    catalogs.getOrElseUpdate(name, Catalogs.load(name, conf))
+  }
+
+  def defaultCatalog: Option[CatalogPlugin] = {
+    conf.defaultV2Catalog.flatMap { catalogName =>
+      try {
+        Some(catalog(catalogName))
+      } catch {
+        case NonFatal(e) =>
+          logError(s"Cannot load default v2 catalog: $catalogName", e)
+          None
+      }
+    }
+  }
+
+  def v2SessionCatalog: Option[CatalogPlugin] = {
+    try {
+      Some(catalog(CatalogManager.SESSION_CATALOG_NAME))
+    } catch {
+      case NonFatal(e) =>
+        logError("Cannot load v2 session catalog", e)
+        None
+    }
+  }
+
+  private def getDefaultNamespace(c: CatalogPlugin) = c match {
+    case c: SupportsNamespaces => c.defaultNamespace()
+    case _ => Array.empty[String]
+  }
+
+  private var _currentNamespace = {
+    // The builtin catalog use "default" as the default database.
 
 Review comment:
   I'm not sure we should ask the catalog implementation to do it. This will make catalog
mutable and make it harder to implement. e.g. they need to take care of the current namespace
in many methods like `loadTable`, `createTable`, etc. It also brings some risks if the catalog
implementation has bugs for tracking the current namespace.
   
   Instead, I think it's better to let Spark track the current namespace, and always extend
the table identifier with the current namespace and make it a fully qualified identifier before
calling catalog APIs. I'm fine for not tracking it in `CatalogManager`, but we should track
it somewhere in Spark.

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org


Mime
View raw message