n3nash commented on a change in pull request #2379: URL: https://github.com/apache/hudi/pull/2379#discussion_r554301213
########## File path: hudi-client/hudi-client-common/src/main/java/org/apache/hudi/client/AbstractHoodieWriteClient.java ########## @@ -711,12 +711,30 @@ public void rollbackInflightCompaction(HoodieInstant inflightInstant, HoodieTabl table.getActiveTimeline().revertCompactionInflightToRequested(inflightInstant); } + /** + * Get inflight time line exclude compaction and clustering. + * @param table + * @return + */ + private HoodieTimeline getInflightTimelineExcludeCompactionAndClustering(HoodieTable<T, I, K, O> table) { Review comment: @lw309637554 Can you add this API to the HoodieTimeline ? You can name it `filterPendingExcludingCompactionAndReplace` for now. ########## File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/HoodieClusteringJob.java ########## @@ -0,0 +1,174 @@ +/* + * 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.hudi.utilities; + +import com.beust.jcommander.JCommander; +import com.beust.jcommander.Parameter; +import org.apache.avro.Schema; +import org.apache.hadoop.fs.FileSystem; +import org.apache.hadoop.fs.Path; +import org.apache.hudi.client.SparkRDDWriteClient; +import org.apache.hudi.client.WriteStatus; +import org.apache.hudi.common.config.TypedProperties; +import org.apache.hudi.common.fs.FSUtils; +import org.apache.hudi.common.table.HoodieTableMetaClient; +import org.apache.hudi.common.table.TableSchemaResolver; +import org.apache.hudi.common.util.Option; +import org.apache.hudi.exception.HoodieException; +import org.apache.log4j.LogManager; +import org.apache.log4j.Logger; +import org.apache.spark.api.java.JavaRDD; +import org.apache.spark.api.java.JavaSparkContext; +import org.jetbrains.annotations.TestOnly; + +import java.io.Serializable; +import java.util.ArrayList; +import java.util.List; + +public class HoodieClusteringJob { + + private static final Logger LOG = LogManager.getLogger(HoodieClusteringJob.class); + private final Config cfg; + private transient FileSystem fs; + private TypedProperties props; + private final JavaSparkContext jsc; + + public HoodieClusteringJob(JavaSparkContext jsc, Config cfg) { + this.cfg = cfg; + this.jsc = jsc; + this.props = cfg.propsFilePath == null + ? UtilHelpers.buildProperties(cfg.configs) + : readConfigFromFileSystem(jsc, cfg); + } + + private TypedProperties readConfigFromFileSystem(JavaSparkContext jsc, Config cfg) { + final FileSystem fs = FSUtils.getFs(cfg.basePath, jsc.hadoopConfiguration()); + + return UtilHelpers + .readConfig(fs, new Path(cfg.propsFilePath), cfg.configs) + .getConfig(); + } + + public static class Config implements Serializable { + @Parameter(names = {"--base-path", "-sp"}, description = "Base path for the table", required = true) + public String basePath = null; + @Parameter(names = {"--table-name", "-tn"}, description = "Table name", required = true) + public String tableName = null; + @Parameter(names = {"--instant-time", "-it"}, description = "Clustering Instant time, only need when cluster. " + + "And schedule clustering can generate it.", required = false) + public String clusteringInstantTime = null; + @Parameter(names = {"--parallelism", "-pl"}, description = "Parallelism for hoodie insert", required = false) + public int parallelism = 1; + @Parameter(names = {"--spark-master", "-ms"}, description = "Spark master", required = false) + public String sparkMaster = null; + @Parameter(names = {"--spark-memory", "-sm"}, description = "spark memory to use", required = true) + public String sparkMemory = null; + @Parameter(names = {"--retry", "-rt"}, description = "number of retries", required = false) + public int retry = 0; + + @Parameter(names = {"--schedule", "-sc"}, description = "Schedule clustering") + public Boolean runSchedule = false; + @Parameter(names = {"--help", "-h"}, help = true) + public Boolean help = false; + + @Parameter(names = {"--props"}, description = "path to properties file on localfs or dfs, with configurations for " + + "hoodie client for clustering") + public String propsFilePath = null; + + @Parameter(names = {"--hoodie-conf"}, description = "Any configuration that can be set in the properties file " + + "(using the CLI parameter \"--props\") can also be passed command line using this parameter. This can be repeated", + splitter = IdentitySplitter.class) + public List<String> configs = new ArrayList<>(); + } + + public static void main(String[] args) { + final Config cfg = new Config(); + JCommander cmd = new JCommander(cfg, null, args); + if (cfg.help || args.length == 0 || (!cfg.runSchedule && cfg.clusteringInstantTime == null)) { + cmd.usage(); + System.exit(1); + } + final JavaSparkContext jsc = UtilHelpers.buildSparkContext("clustering-" + cfg.tableName, cfg.sparkMaster, cfg.sparkMemory); + HoodieClusteringJob clusteringJob = new HoodieClusteringJob(jsc, cfg); + int result = clusteringJob.cluster(cfg.retry); + String resultMsg = String.format("Clustering with basePath: %s, tableName: %s, runSchedule: %s", + cfg.basePath, cfg.tableName, cfg.runSchedule); + if (result == -1) { + LOG.error(resultMsg + " failed"); + } else { + LOG.info(resultMsg + " success"); + } + jsc.stop(); + } + + public int cluster(int retry) { Review comment: This code about retry is repeated in `HoodieCompactor` as well, is there a way to reuse this using some kinds of Utils class ? ########## File path: hudi-utilities/src/test/java/org/apache/hudi/utilities/functional/TestHoodieDeltaStreamer.java ########## @@ -672,14 +681,72 @@ public void testInlineClustering() throws Exception { cfg.configs.add(String.format("%s=false", HoodieCompactionConfig.AUTO_CLEAN_PROP)); cfg.configs.add(String.format("%s=%s", HoodieClusteringConfig.INLINE_CLUSTERING_PROP, "true")); cfg.configs.add(String.format("%s=%s", HoodieClusteringConfig.INLINE_CLUSTERING_MAX_COMMIT_PROP, "2")); - - deltaStreamerTestRunner(cfg, (r) -> { + HoodieDeltaStreamer ds = new HoodieDeltaStreamer(cfg, jsc); + deltaStreamerTestRunner(ds, cfg, (r) -> { HoodieTableMetaClient metaClient = new HoodieTableMetaClient(this.dfs.getConf(), tableBasePath, true); int pendingReplaceSize = metaClient.getActiveTimeline().filterPendingReplaceTimeline().getInstants().toArray().length; int completeReplaceSize = metaClient.getActiveTimeline().getCompletedReplaceTimeline().getInstants().toArray().length; LOG.info("PendingReplaceSize=" + pendingReplaceSize + ",completeReplaceSize = " + completeReplaceSize); return completeReplaceSize > 0; }); + HoodieTableMetaClient metaClient = new HoodieTableMetaClient(this.dfs.getConf(), tableBasePath, true); + assertEquals(1, metaClient.getActiveTimeline().getCompletedReplaceTimeline().getInstants().toArray().length); + } + + private HoodieClusteringJob.Config buildHoodieClusteringUtilConfig(String basePath, + String clusteringInstantTime, boolean runSchedule) { + HoodieClusteringJob.Config config = new HoodieClusteringJob.Config(); + config.basePath = basePath; + config.clusteringInstantTime = clusteringInstantTime; + config.runSchedule = runSchedule; + config.propsFilePath = dfsBasePath + "/clusteringjob.properties"; + return config; + } + + @Test + public void testHoodieAsyncClusteringJob() throws Exception { Review comment: Why is this specific test case in `TestHoodieDeltaStreamer` ? Is the intention to just test if clustering works with deltastreamer ? ########## File path: hudi-utilities/src/test/java/org/apache/hudi/utilities/functional/TestHoodieDeltaStreamer.java ########## @@ -672,14 +681,72 @@ public void testInlineClustering() throws Exception { cfg.configs.add(String.format("%s=false", HoodieCompactionConfig.AUTO_CLEAN_PROP)); cfg.configs.add(String.format("%s=%s", HoodieClusteringConfig.INLINE_CLUSTERING_PROP, "true")); cfg.configs.add(String.format("%s=%s", HoodieClusteringConfig.INLINE_CLUSTERING_MAX_COMMIT_PROP, "2")); - - deltaStreamerTestRunner(cfg, (r) -> { + HoodieDeltaStreamer ds = new HoodieDeltaStreamer(cfg, jsc); + deltaStreamerTestRunner(ds, cfg, (r) -> { HoodieTableMetaClient metaClient = new HoodieTableMetaClient(this.dfs.getConf(), tableBasePath, true); int pendingReplaceSize = metaClient.getActiveTimeline().filterPendingReplaceTimeline().getInstants().toArray().length; int completeReplaceSize = metaClient.getActiveTimeline().getCompletedReplaceTimeline().getInstants().toArray().length; LOG.info("PendingReplaceSize=" + pendingReplaceSize + ",completeReplaceSize = " + completeReplaceSize); return completeReplaceSize > 0; }); + HoodieTableMetaClient metaClient = new HoodieTableMetaClient(this.dfs.getConf(), tableBasePath, true); + assertEquals(1, metaClient.getActiveTimeline().getCompletedReplaceTimeline().getInstants().toArray().length); + } + + private HoodieClusteringJob.Config buildHoodieClusteringUtilConfig(String basePath, + String clusteringInstantTime, boolean runSchedule) { + HoodieClusteringJob.Config config = new HoodieClusteringJob.Config(); + config.basePath = basePath; + config.clusteringInstantTime = clusteringInstantTime; + config.runSchedule = runSchedule; + config.propsFilePath = dfsBasePath + "/clusteringjob.properties"; + return config; + } + + @Test + public void testHoodieAsyncClusteringJob() throws Exception { Review comment: Why is this specific test case in `TestHoodieDeltaStreamer` ? Is the intention to just test if clustering works with deltastreamer ? I'm worried this is polluting the tests in hoodie delta streamer, is this test case to be refactored after deltastreamer natively supports clustering ? ########## File path: hudi-utilities/src/test/java/org/apache/hudi/utilities/functional/TestHoodieDeltaStreamer.java ########## @@ -672,14 +681,72 @@ public void testInlineClustering() throws Exception { cfg.configs.add(String.format("%s=false", HoodieCompactionConfig.AUTO_CLEAN_PROP)); cfg.configs.add(String.format("%s=%s", HoodieClusteringConfig.INLINE_CLUSTERING_PROP, "true")); cfg.configs.add(String.format("%s=%s", HoodieClusteringConfig.INLINE_CLUSTERING_MAX_COMMIT_PROP, "2")); - - deltaStreamerTestRunner(cfg, (r) -> { + HoodieDeltaStreamer ds = new HoodieDeltaStreamer(cfg, jsc); + deltaStreamerTestRunner(ds, cfg, (r) -> { HoodieTableMetaClient metaClient = new HoodieTableMetaClient(this.dfs.getConf(), tableBasePath, true); int pendingReplaceSize = metaClient.getActiveTimeline().filterPendingReplaceTimeline().getInstants().toArray().length; int completeReplaceSize = metaClient.getActiveTimeline().getCompletedReplaceTimeline().getInstants().toArray().length; LOG.info("PendingReplaceSize=" + pendingReplaceSize + ",completeReplaceSize = " + completeReplaceSize); return completeReplaceSize > 0; }); + HoodieTableMetaClient metaClient = new HoodieTableMetaClient(this.dfs.getConf(), tableBasePath, true); + assertEquals(1, metaClient.getActiveTimeline().getCompletedReplaceTimeline().getInstants().toArray().length); + } + + private HoodieClusteringJob.Config buildHoodieClusteringUtilConfig(String basePath, + String clusteringInstantTime, boolean runSchedule) { + HoodieClusteringJob.Config config = new HoodieClusteringJob.Config(); + config.basePath = basePath; + config.clusteringInstantTime = clusteringInstantTime; + config.runSchedule = runSchedule; + config.propsFilePath = dfsBasePath + "/clusteringjob.properties"; + return config; + } + + @Test + public void testHoodieAsyncClusteringJob() throws Exception { Review comment: Why is this specific test case in `TestHoodieDeltaStreamer` ? Is the intention to just test if clustering works with deltastreamer ? I'm worried this is polluting the tests in hoodie delta streamer, is this test case to be refactored after deltastreamer natively supports clustering ? If yes, can we open a ticket for the follow up.. ---------------------------------------------------------------- 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: us...@infra.apache.org