lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-3458] Fix BulkInsertPartitioner generic type (#4854)

Browse Source
This commit is contained in:
Raymond Xu
2022-02-20 10:51:58 -08:00
committed by GitHub
parent 66ac1446dd
commit 0938f55a2b
10 changed files with 31 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/run/strategy/MultipleSparkJobExecutionStrategy.java
View File

@@ -18,10 +18,6 @@
package org.apache.hudi.client.clustering.run.strategy;
import org.apache.avro.Schema;
import org.apache.avro.generic.GenericRecord;
import org.apache.avro.generic.IndexedRecord;
import org.apache.hadoop.fs.Path;
import org.apache.hudi.avro.HoodieAvroUtils;
import org.apache.hudi.avro.model.HoodieClusteringGroup;
import org.apache.hudi.avro.model.HoodieClusteringPlan;
@@ -60,6 +56,11 @@ import org.apache.hudi.table.BulkInsertPartitioner;
import org.apache.hudi.table.HoodieTable;
import org.apache.hudi.table.action.HoodieWriteMetadata;
import org.apache.hudi.table.action.cluster.strategy.ClusteringExecutionStrategy;
import org.apache.avro.Schema;
import org.apache.avro.generic.GenericRecord;
import org.apache.avro.generic.IndexedRecord;
import org.apache.hadoop.fs.Path;
import org.apache.log4j.LogManager;
import org.apache.log4j.Logger;
import org.apache.spark.api.java.JavaRDD;
@@ -134,7 +135,7 @@ public abstract class MultipleSparkJobExecutionStrategy<T extends HoodieRecordPa
* @param schema Schema of the data including metadata fields.
* @return {@link RDDCustomColumnsSortPartitioner} if sort columns are provided, otherwise empty.
*/
protected Option<BulkInsertPartitioner<T>> getPartitioner(Map<String, String> strategyParams, Schema schema) {
protected Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> getPartitioner(Map<String, String> strategyParams, Schema schema) {
Option<String[]> orderByColumnsOpt =
Option.ofNullable(strategyParams.get(PLAN_STRATEGY_SORT_COLUMNS.key()))
.map(listStr -> listStr.split(","));

7
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/table/action/commit/SparkBulkInsertCommitActionExecutor.java
View File

@@ -29,6 +29,7 @@ import org.apache.hudi.exception.HoodieInsertException;
import org.apache.hudi.table.BulkInsertPartitioner;
import org.apache.hudi.table.HoodieTable;
import org.apache.hudi.table.action.HoodieWriteMetadata;
import org.apache.spark.api.java.JavaRDD;
import java.util.Map;
@@ -36,17 +37,17 @@ import java.util.Map;
public class SparkBulkInsertCommitActionExecutor<T extends HoodieRecordPayload<T>> extends BaseSparkCommitActionExecutor<T> {
private final JavaRDD<HoodieRecord<T>> inputRecordsRDD;
private final Option<BulkInsertPartitioner<T>> bulkInsertPartitioner;
private final Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner;
public SparkBulkInsertCommitActionExecutor(HoodieSparkEngineContext context, HoodieWriteConfig config, HoodieTable table,
String instantTime, JavaRDD<HoodieRecord<T>> inputRecordsRDD,
Option<BulkInsertPartitioner<T>> bulkInsertPartitioner) {
Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner) {
this(context, config, table, instantTime, inputRecordsRDD, bulkInsertPartitioner, Option.empty());
}
public SparkBulkInsertCommitActionExecutor(HoodieSparkEngineContext context, HoodieWriteConfig config, HoodieTable table,
String instantTime, JavaRDD<HoodieRecord<T>> inputRecordsRDD,
Option<BulkInsertPartitioner<T>> bulkInsertPartitioner,
Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner,
Option<Map<String, String>> extraMetadata) {
super(context, config, table, instantTime, WriteOperationType.BULK_INSERT, extraMetadata);
this.inputRecordsRDD = inputRecordsRDD;

4
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/table/action/commit/SparkBulkInsertHelper.java
View File

@@ -67,7 +67,7 @@ public class SparkBulkInsertHelper<T extends HoodieRecordPayload, R> extends Bas
final HoodieWriteConfig config,
final BaseCommitActionExecutor<T, JavaRDD<HoodieRecord<T>>, JavaRDD<HoodieKey>, JavaRDD<WriteStatus>, R> executor,
final boolean performDedupe,
final Option<BulkInsertPartitioner<T>> userDefinedBulkInsertPartitioner) {
final Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> userDefinedBulkInsertPartitioner) {
HoodieWriteMetadata result = new HoodieWriteMetadata();
//transition bulk_insert state to inflight
@@ -88,7 +88,7 @@ public class SparkBulkInsertHelper<T extends HoodieRecordPayload, R> extends Bas
HoodieTable<T, JavaRDD<HoodieRecord<T>>, JavaRDD<HoodieKey>, JavaRDD<WriteStatus>> table,
HoodieWriteConfig config,
boolean performDedupe,
Option<BulkInsertPartitioner<T>> userDefinedBulkInsertPartitioner,
Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> userDefinedBulkInsertPartitioner,
boolean useWriterSchema,
int parallelism,
WriteHandleFactory writeHandleFactory) {

10
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/table/action/commit/SparkBulkInsertPreppedCommitActionExecutor.java
View File

@@ -26,22 +26,22 @@ import org.apache.hudi.common.model.WriteOperationType;
import org.apache.hudi.common.util.Option;
import org.apache.hudi.config.HoodieWriteConfig;
import org.apache.hudi.exception.HoodieInsertException;
import org.apache.hudi.table.HoodieTable;
import org.apache.hudi.table.BulkInsertPartitioner;
import org.apache.hudi.table.HoodieTable;
import org.apache.hudi.table.action.HoodieWriteMetadata;
import org.apache.spark.api.java.JavaRDD;
public class SparkBulkInsertPreppedCommitActionExecutor<T extends HoodieRecordPayload<T>>
extends BaseSparkCommitActionExecutor<T> {
private final JavaRDD<HoodieRecord<T>> preppedInputRecordRdd;
private final Option<BulkInsertPartitioner<T>> userDefinedBulkInsertPartitioner;
private final Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> userDefinedBulkInsertPartitioner;
public SparkBulkInsertPreppedCommitActionExecutor(HoodieSparkEngineContext context,
HoodieWriteConfig config, HoodieTable table,
String instantTime, JavaRDD<HoodieRecord<T>> preppedInputRecordRdd,
Option<BulkInsertPartitioner<T>> userDefinedBulkInsertPartitioner) {
Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> userDefinedBulkInsertPartitioner) {
super(context, config, table, instantTime, WriteOperationType.BULK_INSERT);
this.preppedInputRecordRdd = preppedInputRecordRdd;
this.userDefinedBulkInsertPartitioner = userDefinedBulkInsertPartitioner;
@@ -60,4 +60,4 @@ public class SparkBulkInsertPreppedCommitActionExecutor<T extends HoodieRecordPa
}
}
}
}

6
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/table/action/deltacommit/SparkBulkInsertDeltaCommitActionExecutor.java
View File

@@ -39,17 +39,17 @@ public class SparkBulkInsertDeltaCommitActionExecutor<T extends HoodieRecordPayl
extends BaseSparkDeltaCommitActionExecutor<T> {
private final JavaRDD<HoodieRecord<T>> inputRecordsRDD;
private final Option<BulkInsertPartitioner<T>> bulkInsertPartitioner;
private final Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner;
public SparkBulkInsertDeltaCommitActionExecutor(HoodieSparkEngineContext context, HoodieWriteConfig config, HoodieTable table,
String instantTime, JavaRDD<HoodieRecord<T>> inputRecordsRDD,
Option<BulkInsertPartitioner<T>> bulkInsertPartitioner) {
Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner) {
this(context, config, table, instantTime, inputRecordsRDD, bulkInsertPartitioner, Option.empty());
}
public SparkBulkInsertDeltaCommitActionExecutor(HoodieSparkEngineContext context, HoodieWriteConfig config, HoodieTable table,
String instantTime, JavaRDD<HoodieRecord<T>> inputRecordsRDD,
Option<BulkInsertPartitioner<T>> bulkInsertPartitioner,
Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner,
Option<Map<String, String>> extraMetadata) {
super(context, config, table, instantTime, WriteOperationType.BULK_INSERT, extraMetadata);
this.inputRecordsRDD = inputRecordsRDD;

4
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/table/action/deltacommit/SparkBulkInsertPreppedDeltaCommitActionExecutor.java
View File

@@ -37,12 +37,12 @@ public class SparkBulkInsertPreppedDeltaCommitActionExecutor<T extends HoodieRec
extends BaseSparkDeltaCommitActionExecutor<T> {
private final JavaRDD<HoodieRecord<T>> preppedInputRecordRdd;
private final Option<BulkInsertPartitioner<T>> bulkInsertPartitioner;
private final Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner;
public SparkBulkInsertPreppedDeltaCommitActionExecutor(HoodieSparkEngineContext context,
HoodieWriteConfig config, HoodieTable table,
String instantTime, JavaRDD<HoodieRecord<T>> preppedInputRecordRdd,
Option<BulkInsertPartitioner<T>> bulkInsertPartitioner) {
Option<BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>>> bulkInsertPartitioner) {
super(context, config, table, instantTime, WriteOperationType.BULK_INSERT);
this.preppedInputRecordRdd = preppedInputRecordRdd;
this.bulkInsertPartitioner = bulkInsertPartitioner;