[GitHub] xhumanoid commented on a change in pull request #6434: [FLINK-6968] [table] Add Queryable table sink.
xhumanoid commented on a change in pull request #6434: [FLINK-6968] [table] Add Queryable table sink. URL: https://github.com/apache/flink/pull/6434#discussion_r209314950 ## File path: flink-libraries/flink-table/src/test/scala/org/apache/flink/table/runtime/stream/sink/queryable/QueryableTableSinkTest.scala ## @@ -0,0 +1,188 @@ +/* + * 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.flink.table.runtime.stream.sink.queryable + +import java.time.Duration +import java.util.concurrent.{ExecutionException, TimeUnit} + +import org.apache.flink.api.common.state.ValueStateDescriptor +import org.apache.flink.api.common.time.{Deadline, Time} +import org.apache.flink.api.common.typeinfo.{BasicTypeInfo, TypeInformation} +import org.apache.flink.api.java.typeutils.RowTypeInfo +import org.apache.flink.api.scala._ +import org.apache.flink.contrib.streaming.state.RocksDBStateBackend +import org.apache.flink.queryablestate.client.QueryableStateClient +import org.apache.flink.queryablestate.exceptions.UnknownKeyOrNamespaceException +import org.apache.flink.runtime.state.StateBackend +import org.apache.flink.streaming.api.scala.StreamExecutionEnvironment +import org.apache.flink.table.api.scala._ +import org.apache.flink.table.api.{StreamQueryConfig, TableEnvironment} +import org.apache.flink.table.runtime.harness.HarnessTestBase.TestStreamQueryConfig +import org.apache.flink.table.sinks.queryable.QueryableTableSink +import org.apache.flink.types.Row +import org.hamcrest.core.Is +import org.junit.Assert._ +import org.junit.rules.{ExpectedException, TemporaryFolder} +import org.junit.{Rule, Test} + + +class QueryableTableSinkTest extends QueryableSinkTestBase { + + private val queryConfig = new StreamQueryConfig() + queryConfig.withIdleStateRetentionTime(Time.hours(1), Time.hours(2)) + + val _tempFolder = new TemporaryFolder + @Rule + def tempFolder: TemporaryFolder = _tempFolder + + val _expectedException = ExpectedException.none() + @Rule + def expectedException: ExpectedException = _expectedException + + def getStateBackend: StateBackend = { +val dbPath = tempFolder.newFolder().getAbsolutePath +val checkpointPath = tempFolder.newFolder().toURI.toString +val backend = new RocksDBStateBackend(checkpointPath) +backend.setDbStoragePath(dbPath) +backend + } + + @Test + def testQueryableSink(): Unit = { +val env = StreamExecutionEnvironment.getExecutionEnvironment +env.setStateBackend(getStateBackend) +val tEnv = TableEnvironment.getTableEnvironment(env) + +//name, money +val data = List(("jeff", -1), ("dean", -2), ("jeff", 2), ("dean", 4)) +val source = new TestKVListSource[String, Int](data) + +// select name, sum(money) as sm from t group by name +val t = env.addSource(source).toTable(tEnv, 'name, 'money) +.groupBy("name") +.select("name, sum(money) as sm") + +val queryableSink = new QueryableTableSink("prefix", + new StreamQueryConfig().withIdleStateRetentionTime(Time.minutes(1), Time.minutes(7))) + +t.writeToSink(queryableSink) + +val clusterClient = QueryableSinkTestBase.miniClusterResource.getClusterClient +val deadline = Deadline.now.plus(Duration.ofSeconds(100)) + +val autoCancellableJob = new AutoCancellableJob(deadline, clusterClient, env.getJavaEnv) +val client = new QueryableStateClient("localhost", 9084) + Review comment: Required shutdown client in the end This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] xhumanoid commented on a change in pull request #6434: [FLINK-6968] [table] Add Queryable table sink.
xhumanoid commented on a change in pull request #6434: [FLINK-6968] [table] Add Queryable table sink. URL: https://github.com/apache/flink/pull/6434#discussion_r209300017 ## File path: flink-libraries/flink-table/src/main/scala/org/apache/flink/table/sinks/queryable/QueryableTableSink.scala ## @@ -0,0 +1,117 @@ +/* + * 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.flink.table.sinks.queryable + +import java.lang.{Boolean => JBool} + +import org.apache.flink.api.common.typeinfo.TypeInformation +import org.apache.flink.api.java.tuple.{Tuple2 => JTuple2} +import org.apache.flink.api.java.typeutils.RowTypeInfo +import org.apache.flink.streaming.api.datastream.DataStream +import org.apache.flink.table.api.StreamQueryConfig +import org.apache.flink.table.sinks.{TableSinkBase, UpsertStreamTableSink} +import org.apache.flink.types.Row + +/** + * A QueryableTableSink stores table in queryable state. + * + * This class stores table in queryable state so that users can access table data without + * dependency on external storage. + * + * Since this is only a kv storage, currently user can only do point query against it. + * + * Example: + * {{{ + * val env = ExecutionEnvironment.getExecutionEnvironment + * val tEnv = TableEnvironment.getTableEnvironment(env) + * + * val table: Table = ... + * + * val queryableTableSink: QueryableTableSink = new QueryableTableSink( + * "prefix", + * queryConfig) + * + * tEnv.writeToSink(table, queryableTableSink, config) + * }}} + * + * When program starts to run, user can access state with QueryableStateClient. + * {{{ + * val client = new QueryableStateClient(tmHostname, proxyPort) + * val data = client.getKvState( + * jobId, + * "prefix-column1", + * Row.of(1), + * new RowTypeInfo(Array(TypeInfoformation.of(classOf[Int]), Array("id")) + * stateDescriptor) + * .get(); + * + * }}} + * + * + * @param namePrefix + * @param queryConfig + */ +class QueryableTableSink( + private val namePrefix: String, + private val queryConfig: StreamQueryConfig) + extends UpsertStreamTableSink[Row] +with TableSinkBase[JTuple2[JBool, Row]] { + private var keys: Array[String] = _ + + override def setKeyFields(keys: Array[String]): Unit = { +if (keys == null) { + throw new IllegalArgumentException("keys can't be null!") +} +this.keys = keys + } + + override def setIsAppendOnly(isAppendOnly: JBool): Unit = { +if (isAppendOnly) { + throw new IllegalArgumentException("A QueryableTableSink can not be used with append-only " + +"tables as the table would grow infinitely") +} + } + + override def getRecordType: TypeInformation[Row] = new RowTypeInfo(getFieldTypes, getFieldNames) + + override def emitDataStream(dataStream: DataStream[JTuple2[JBool, Row]]): Unit = { +val keyIndices = keys.map(getFieldNames.indexOf(_)) +val keyTypes = keyIndices.map(getFieldTypes(_)) + +val keySelectorType = new RowTypeInfo(keyTypes, keys) + +val processFunction = new QueryableStateProcessFunction( + namePrefix, + queryConfig, + keys, + getFieldNames, + getFieldTypes) + +dataStream.keyBy(new RowKeySelector(keyIndices, keySelectorType)) + .process(processFunction) + } + + override protected def copy: TableSinkBase[JTuple2[JBool, Row]] = { +new QueryableTableSink(this.namePrefix, this.queryConfig) + } +} + + + + Review comment: remove block of blank lines This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] xhumanoid commented on a change in pull request #6434: [FLINK-6968] [table] Add Queryable table sink.
xhumanoid commented on a change in pull request #6434: [FLINK-6968] [table] Add Queryable table sink. URL: https://github.com/apache/flink/pull/6434#discussion_r209297911 ## File path: flink-libraries/flink-table/pom.xml ## @@ -186,6 +186,13 @@ under the License. test test-jar + + + org.apache.flink + flink-queryable-state-runtime_2.11 Review comment: replace hardcoded 2.11 with ${scala.binary.version} This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services