rdblue commented on a change in pull request #1849: URL: https://github.com/apache/iceberg/pull/1849#discussion_r535706991
########## File path: api/src/main/java/org/apache/iceberg/catalog/SupportsCatalogTransactions.java ########## @@ -0,0 +1,137 @@ +/* + * 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.iceberg.catalog; + +import java.util.Collections; +import java.util.Set; + +/** + * Catalog methods for working with catalog-level transactions. + * + * <p>Catalog implementations are not required to support catalog-level transactional state. + * If they do, they may support one or more {@code IsolationLevel}s and one or more + * {@code LockingMode}s. + */ +public interface SupportsCatalogTransactions { + + /** + * The level of isolation for a transaction. + */ + enum IsolationLevel { + + /** + * Only read committed data. Reading the same table multiple times + * may result in different committed reads for each read. + */ + READ_COMMITTED, + + /** + * Only read committed data. Reading the same table multiple times will + * result in the same view of those tables. Review comment: I was talking with Jacques on Slack and I think the difference between this "repeatable read" and the "snapshot" that we use for single-table transactions is that with multiple tables, snapshot implies that there is a point in time when all tables are in the starting state. Catalogs may not guarantee that there is happened-before relationship across tables, so repeatable read is used instead of snapshot. However if I understand correctly, the same write guarantees are used for both "repeatable read" and "snapshot": there is no check that newer commits have not occurred that would change the result of the operation. ---------------------------------------------------------------- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
