[HUDI-2850] Clustering CLI - schedule and run command fixes to avoid …

…NumberFormatException - added java doc on how to run the clustering commands
apache · Nov 26, 2021 · 89112c7 · 89112c7
1 parent 698b881
commit 89112c7
Show file tree

Hide file tree

Showing 3 changed files with 17 additions and 2 deletions.
diff --git a/hudi-cli/src/main/java/org/apache/hudi/cli/commands/ClusteringCommand.java b/hudi-cli/src/main/java/org/apache/hudi/cli/commands/ClusteringCommand.java
@@ -40,6 +40,13 @@ public class ClusteringCommand implements CommandMarker {
 
   private static final Logger LOG = LogManager.getLogger(ClusteringCommand.class);
 
+  /**
+   * Schedule clustering table service.
+   * <p>
+   * Example:
+   * > connect --path {path to hudi table}
+   * > clustering schedule --sparkMaster local --sparkMemory 2g
+   */
   @CliCommand(value = "clustering schedule", help = "Schedule Clustering")
   public String scheduleClustering(
       @CliOption(key = "sparkMaster", unspecifiedDefaultValue = SparkUtil.DEFAULT_SPARK_MASTER, help = "Spark master") final String master,
@@ -71,6 +78,14 @@ public String scheduleClustering(
     return "Succeeded to schedule clustering for " + clusteringInstantTime;
   }
 
+  /**
+   * Run clustering table service.
+   * <p>
+   * Example:
+   * > connect --path {path to hudi table}
+   * > clustering schedule --sparkMaster local --sparkMemory 2g
+   * > clustering run --sparkMaster local --sparkMemory 2g --clusteringInstant  20211124005208
+   */
   @CliCommand(value = "clustering run", help = "Run Clustering")
   public String runClustering(
       @CliOption(key = "sparkMaster", unspecifiedDefaultValue = SparkUtil.DEFAULT_SPARK_MASTER, help = "Spark master") final String master,

diff --git a/hudi-cli/src/main/java/org/apache/hudi/cli/commands/SparkMain.java b/hudi-cli/src/main/java/org/apache/hudi/cli/commands/SparkMain.java
@@ -80,11 +80,11 @@ enum SparkCommand {
   }
 
   public static void main(String[] args) throws Exception {
+    ValidationUtils.checkArgument(args.length >= 4);
     final String commandString = args[0];
     LOG.info("Invoking SparkMain: " + commandString);
     final SparkCommand cmd = SparkCommand.valueOf(commandString);
 
-    ValidationUtils.checkArgument(args.length >= 4);
     JavaSparkContext jsc = SparkUtil.initJavaSparkConf("hoodie-cli-" + commandString,
         Option.of(args[1]), Option.of(args[2]));
 

diff --git a/hudi-utilities/src/main/java/org/apache/hudi/utilities/HoodieClusteringJob.java b/hudi-utilities/src/main/java/org/apache/hudi/utilities/HoodieClusteringJob.java
@@ -59,7 +59,7 @@ public class HoodieClusteringJob {
   public HoodieClusteringJob(JavaSparkContext jsc, Config cfg) {
     this.cfg = cfg;
     this.jsc = jsc;
-    this.props = (cfg.propsFilePath == null || cfg.propsFilePath.isEmpty())
+    this.props = StringUtils.isNullOrEmpty(cfg.propsFilePath)
         ? UtilHelpers.buildProperties(cfg.configs)
         : readConfigFromFileSystem(jsc, cfg);
   }