[MINOR] Some cosmetic changes for Flink (#3503)

hudi-flink/src/main/java/org/apache/hudi/sink/bootstrap/BootstrapFunction.java

@@ -148,7 +148,7 @@ public class BootstrapFunction<I, O extends HoodieRecord>
   }
 
   /**
-   * Load all the indices of give partition path into the backup state.
+   * Loads all the indices of give partition path into the backup state.
    *
    * @param partitionPath The partition path
    */

hudi-flink/src/main/java/org/apache/hudi/sink/bootstrap/batch/BatchBootstrapFunction.java

@@ -29,7 +29,15 @@ import java.util.HashSet;
 import java.util.Set;
 
 /**
- * The function to load specify partition index from existing hoodieTable.
+ * The function to load index from existing hoodieTable.
+ *
+ * <p>This function should only be used for bounded source.
+ *
+ * <p>When a record comes in, the function firstly checks whether the partition path of the record is already loaded,
+ * if the partition is not loaded yet, loads the entire partition and sends the index records to downstream operators
+ * before it sends the input record; if the partition is loaded already, sends the input record directly.
+ *
+ * <p>The input records should shuffle by the partition path to avoid repeated loading.
  */
 public class BatchBootstrapFunction<I, O extends HoodieRecord>
     extends BootstrapFunction<I, O> {
@@ -61,5 +69,4 @@ public class BatchBootstrapFunction<I, O extends HoodieRecord>
     // send the trigger record
     out.collect((O) value);
   }
-
 }

hudi-flink/src/main/java/org/apache/hudi/sink/utils/Pipelines.java

@@ -85,7 +85,22 @@ public class Pipelines {
         .name("clean_commits");
   }
 
-  public static DataStream<HoodieRecord> bootstrap(Configuration conf, RowType rowType, int defaultParallelism, DataStream<RowData> dataStream) {
+  public static DataStream<HoodieRecord> bootstrap(
+      Configuration conf,
+      RowType rowType,
+      int defaultParallelism,
+      DataStream<RowData> dataStream,
+      boolean bounded) {
+    return bounded
+        ? boundedBootstrap(conf, rowType, defaultParallelism, dataStream)
+        : streamBootstrap(conf, rowType, defaultParallelism, dataStream);
+  }
+
+  private static DataStream<HoodieRecord> streamBootstrap(
+      Configuration conf,
+      RowType rowType,
+      int defaultParallelism,
+      DataStream<RowData> dataStream) {
     DataStream<HoodieRecord> dataStream1 = rowDataToHoodieRecord(conf, rowType, dataStream);
 
     if (conf.getBoolean(FlinkOptions.INDEX_BOOTSTRAP_ENABLED)) {
@@ -101,8 +116,11 @@ public class Pipelines {
     return dataStream1;
   }
 
-  public static DataStream<HoodieRecord> batchBootstrap(Configuration conf, RowType rowType, int defaultParallelism, DataStream<RowData> dataStream) {
-    // shuffle and sort by partition keys
+  private static DataStream<HoodieRecord> boundedBootstrap(
+      Configuration conf,
+      RowType rowType,
+      int defaultParallelism,
+      DataStream<RowData> dataStream) {
     final String[] partitionFields = FilePathUtils.extractPartitionKeys(conf);
     if (partitionFields.length > 0) {
       RowDataKeyGen rowDataKeyGen = RowDataKeyGen.instance(conf, rowType);

hudi-flink/src/main/java/org/apache/hudi/table/HoodieTableSink.java

@@ -77,9 +77,9 @@ public class HoodieTableSink implements DynamicTableSink, SupportsPartitioning,
 
       // default parallelism
       int parallelism = dataStream.getExecutionConfig().getParallelism();
-      final DataStream<HoodieRecord> dataStream1 = context.isBounded()
-          ? Pipelines.batchBootstrap(conf, rowType, parallelism, dataStream)
-          : Pipelines.bootstrap(conf, rowType, parallelism, dataStream);
+
+      // bootstrap
+      final DataStream<HoodieRecord> dataStream1 = Pipelines.bootstrap(conf, rowType, parallelism, dataStream, context.isBounded());
 
       // write pipeline
       DataStream<Object> pipeline = Pipelines.hoodieStreamWrite(conf, parallelism, dataStream1);