boneanxs commented on code in PR #6046:
URL: https://github.com/apache/hudi/pull/6046#discussion_r946549917
##########
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/run/strategy/MultipleSparkJobExecutionStrategy.java:
##########
@@ -131,6 +161,53 @@ public abstract HoodieData<WriteStatus>
performClusteringWithRecordsRDD(final Ho
final
Map<String, String> strategyParams, final Schema schema,
final
List<HoodieFileGroupId> fileGroupIdList, final boolean preserveHoodieMetadata);
+ protected HoodieData<WriteStatus> performRowWrite(Dataset<Row> inputRecords,
Map<String, String> parameters) {
+ String uuid = UUID.randomUUID().toString();
+ parameters.put(HoodieWriteConfig.BULKINSERT_ROW_IDENTIFY_ID.key(), uuid);
+ try {
+ inputRecords.write()
+ .format("hudi")
+ .options(JavaConverters.mapAsScalaMapConverter(parameters).asScala())
+ .mode(SaveMode.Append)
+ .save(getWriteConfig().getBasePath());
+ List<WriteStatus> writeStatusList =
HoodieInternalWriteStatusCoordinator.get().getWriteStatuses(uuid)
+ .stream()
+ .map(internalWriteStatus -> {
+ WriteStatus status = new WriteStatus(
+ internalWriteStatus.isTrackSuccessRecords(),
internalWriteStatus.getFailureFraction());
+ status.setFileId(internalWriteStatus.getFileId());
+ status.setTotalRecords(internalWriteStatus.getTotalRecords());
+ status.setPartitionPath(internalWriteStatus.getPartitionPath());
+ status.setStat(internalWriteStatus.getStat());
+ return status;
+ }).collect(Collectors.toList());
+ return getEngineContext().parallelize(writeStatusList);
+ } finally {
+ HoodieInternalWriteStatusCoordinator.get().removeStatuses(uuid);
+ }
+ }
+
+ protected Map<String, String> buildHoodieRowParameters(int numOutputGroups,
String instantTime, Map<String, String> strategyParams, boolean
preserveHoodieMetadata) {
+ HashMap<String, String> params = new HashMap<>();
+ HoodieWriteConfig writeConfig = getWriteConfig();
+ params.put(HoodieWriteConfig.BULKINSERT_PARALLELISM_VALUE.key(),
String.valueOf(numOutputGroups));
+ params.put(HoodieWriteConfig.BULKINSERT_ROW_AUTO_COMMIT.key(),
String.valueOf(false));
Review Comment:
This is a new `AUTO_COMMIT` config specifically for `BULKINSERT`, I want to
use `hoodie.auto.commit` at first, but I found it is set to false when building
HoodieConfig(not sure why),
https://github.com/apache/hudi/pull/1834#discussion_r907976925 so if using
`hoodie.auto.commit` here, could cause lots tests failed as using bulk_insert
will not auto commit by default. So I introduce a new configure to bypass this.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
