lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-2789] Flink batch upsert for non partitioned table does not work (#4028)

Browse Source
This commit is contained in:
Danny Chan
2021-11-18 13:59:03 +08:00
committed by GitHub
parent 2d3f2a3275
commit 71a2ae0fd6
4 changed files with 25 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
hudi-flink/src/main/java/org/apache/hudi/configuration/OptionsResolver.java
View File

@@ -20,6 +20,7 @@ package org.apache.hudi.configuration;
import org.apache.hudi.common.model.DefaultHoodieRecordPayload;
import org.apache.hudi.common.model.WriteOperationType;
import org.apache.hudi.table.format.FilePathUtils;
import org.apache.flink.configuration.Configuration;
@@ -92,4 +93,11 @@ public class OptionsResolver {
final String strategy = conf.getString(FlinkOptions.COMPACTION_TRIGGER_STRATEGY).toLowerCase(Locale.ROOT);
return FlinkOptions.TIME_ELAPSED.equals(strategy) || FlinkOptions.NUM_OR_TIME.equals(strategy);
}
/**
* Returns whether the table is partitioned.
*/
public static boolean isPartitionedTable(Configuration conf) {
return FilePathUtils.extractPartitionKeys(conf).length > 0;
}
}

21
hudi-flink/src/main/java/org/apache/hudi/sink/utils/Pipelines.java
View File

@@ -20,6 +20,7 @@ package org.apache.hudi.sink.utils;
import org.apache.hudi.common.model.HoodieRecord;
import org.apache.hudi.configuration.FlinkOptions;
import org.apache.hudi.configuration.OptionsResolver;
import org.apache.hudi.sink.CleanFunction;
import org.apache.hudi.sink.StreamWriteOperator;
import org.apache.hudi.sink.append.AppendWriteOperator;
@@ -129,10 +130,10 @@ public class Pipelines {
final boolean globalIndex = conf.getBoolean(FlinkOptions.INDEX_GLOBAL_ENABLED);
if (overwrite) {
return rowDataToHoodieRecord(conf, rowType, dataStream);
} else if (bounded && !globalIndex) {
} else if (bounded && !globalIndex && OptionsResolver.isPartitionedTable(conf)) {
return boundedBootstrap(conf, rowType, defaultParallelism, dataStream);
} else {
return streamBootstrap(conf, rowType, defaultParallelism, dataStream);
return streamBootstrap(conf, rowType, defaultParallelism, dataStream, bounded);
}
}
@@ -140,10 +141,11 @@ public class Pipelines {
Configuration conf,
RowType rowType,
int defaultParallelism,
DataStream<RowData> dataStream) {
DataStream<RowData> dataStream,
boolean bounded) {
DataStream<HoodieRecord> dataStream1 = rowDataToHoodieRecord(conf, rowType, dataStream);
if (conf.getBoolean(FlinkOptions.INDEX_BOOTSTRAP_ENABLED)) {
if (conf.getBoolean(FlinkOptions.INDEX_BOOTSTRAP_ENABLED) || bounded) {
dataStream1 = dataStream1
.transform(
"index_bootstrap",
@@ -161,13 +163,10 @@ public class Pipelines {
RowType rowType,
int defaultParallelism,
DataStream<RowData> dataStream) {
final String[] partitionFields = FilePathUtils.extractPartitionKeys(conf);
if (partitionFields.length > 0) {
RowDataKeyGen rowDataKeyGen = RowDataKeyGen.instance(conf, rowType);
// shuffle by partition keys
dataStream = dataStream
.keyBy(rowDataKeyGen::getPartitionPath);
}
final RowDataKeyGen rowDataKeyGen = RowDataKeyGen.instance(conf, rowType);
// shuffle by partition keys
dataStream = dataStream
.keyBy(rowDataKeyGen::getPartitionPath);
return rowDataToHoodieRecord(conf, rowType, dataStream)
.transform(