yihua commented on code in PR #12395:
URL: https://github.com/apache/hudi/pull/12395#discussion_r1866471426
##########
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/run/strategy/MultipleSparkJobExecutionStrategy.java:
##########
@@ -263,14 +286,21 @@ private CompletableFuture<HoodieData<WriteStatus>>
runClusteringForGroupAsyncAsR
String instantTime,
ExecutorService clusteringExecutorService) {
return CompletableFuture.supplyAsync(() -> {
- JavaSparkContext jsc =
HoodieSparkEngineContext.getSparkContext(getEngineContext());
- Dataset<Row> inputRecords = readRecordsForGroupAsRow(jsc,
clusteringGroup, instantTime);
- Schema readerSchema = HoodieAvroUtils.addMetadataFields(new
Schema.Parser().parse(getWriteConfig().getSchema()));
+ JavaSparkContext jsc =
HoodieSparkEngineContext.getSparkContext(getEngineContext()); // incase of
MIT, config.getSchema may not contain the full table schema
+ Schema tableSchemaWithMetaFields = null;
+ try {
+ tableSchemaWithMetaFields = HoodieAvroUtils.addMetadataFields(new
TableSchemaResolver(getHoodieTable().getMetaClient()).getTableAvroSchema(false),
Review Comment:
`getTableAvroSchema(true)` does the same?
##########
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/run/strategy/MultipleSparkJobExecutionStrategy.java:
##########
@@ -420,64 +450,133 @@ writeConfig, new StoragePath(bootstrapFilePath)),
partitionFields,
*/
private Dataset<Row> readRecordsForGroupAsRow(JavaSparkContext jsc,
HoodieClusteringGroup
clusteringGroup,
- String instantTime) {
+ String instantTime,
+ Schema
tableSchemaWithMetaFields) {
List<ClusteringOperation> clusteringOps =
clusteringGroup.getSlices().stream()
.map(ClusteringOperation::create).collect(Collectors.toList());
- boolean hasLogFiles = clusteringOps.stream().anyMatch(op ->
op.getDeltaFilePaths().size() > 0);
- SQLContext sqlContext = new SQLContext(jsc.sc());
-
- StoragePath[] baseFilePaths = clusteringOps
- .stream()
- .map(op -> {
- ArrayList<String> readPaths = new ArrayList<>();
- // NOTE: for bootstrap tables, only need to handle data file path
(which is the skeleton file) because
- // HoodieBootstrapRelation takes care of stitching if there is
bootstrap path for the skeleton file.
- if (op.getDataFilePath() != null) {
- readPaths.add(op.getDataFilePath());
+
+ if
(getWriteConfig().getBooleanOrDefault(HoodieReaderConfig.FILE_GROUP_READER_ENABLED))
{
Review Comment:
To clarify MDT does not run clustering so this is fine, i.e., only data
table goes through this flow.
--
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]
