1
0

[HUDI-2987] Update all deprecated calls to new apis in HoodieRecordPayload (#4681)

This commit is contained in:
Sivabalan Narayanan
2022-02-10 19:19:33 -05:00
committed by GitHub
parent 2fe7a3a41f
commit ba4e732ba7
8 changed files with 68 additions and 21 deletions

View File

@@ -19,6 +19,7 @@
package org.apache.hudi.hadoop.realtime;
import org.apache.hudi.common.model.HoodieAvroPayload;
import org.apache.hudi.common.model.HoodiePayloadProps;
import org.apache.hudi.common.table.HoodieTableMetaClient;
import org.apache.hudi.common.table.log.LogReaderUtils;
import org.apache.hudi.exception.HoodieIOException;
@@ -39,6 +40,7 @@ import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.Properties;
import java.util.stream.Collectors;
/**
@@ -50,6 +52,7 @@ public abstract class AbstractRealtimeRecordReader {
protected final RealtimeSplit split;
protected final JobConf jobConf;
protected final boolean usesCustomPayload;
protected Properties payloadProps = new Properties();
// Schema handles
private Schema readerSchema;
private Schema writerSchema;
@@ -62,7 +65,11 @@ public abstract class AbstractRealtimeRecordReader {
LOG.info("columnIds ==> " + job.get(ColumnProjectionUtils.READ_COLUMN_IDS_CONF_STR));
LOG.info("partitioningColumns ==> " + job.get(hive_metastoreConstants.META_TABLE_PARTITION_COLUMNS, ""));
try {
this.usesCustomPayload = usesCustomPayload();
HoodieTableMetaClient metaClient = HoodieTableMetaClient.builder().setConf(jobConf).setBasePath(split.getBasePath()).build();
if (metaClient.getTableConfig().getPreCombineField() != null) {
this.payloadProps.setProperty(HoodiePayloadProps.PAYLOAD_ORDERING_FIELD_PROP_KEY, metaClient.getTableConfig().getPreCombineField());
}
this.usesCustomPayload = usesCustomPayload(metaClient);
LOG.info("usesCustomPayload ==> " + this.usesCustomPayload);
init();
} catch (IOException e) {
@@ -70,8 +77,7 @@ public abstract class AbstractRealtimeRecordReader {
}
}
private boolean usesCustomPayload() {
HoodieTableMetaClient metaClient = HoodieTableMetaClient.builder().setConf(jobConf).setBasePath(split.getBasePath()).build();
private boolean usesCustomPayload(HoodieTableMetaClient metaClient) {
return !(metaClient.getTableConfig().getPayloadClass().contains(HoodieAvroPayload.class.getName())
|| metaClient.getTableConfig().getPayloadClass().contains("org.apache.hudi.OverwriteWithLatestAvroPayload"));
}

View File

@@ -97,9 +97,9 @@ class RealtimeCompactedRecordReader extends AbstractRealtimeRecordReader
private Option<GenericRecord> buildGenericRecordwithCustomPayload(HoodieRecord record) throws IOException {
if (usesCustomPayload) {
return ((HoodieAvroRecord) record).getData().getInsertValue(getWriterSchema());
return ((HoodieAvroRecord) record).getData().getInsertValue(getWriterSchema(), payloadProps);
} else {
return ((HoodieAvroRecord) record).getData().getInsertValue(getReaderSchema());
return ((HoodieAvroRecord) record).getData().getInsertValue(getReaderSchema(), payloadProps);
}
}

View File

@@ -88,7 +88,7 @@ class RealtimeUnmergedRecordReader extends AbstractRealtimeRecordReader
.withBufferSize(this.jobConf.getInt(HoodieRealtimeConfig.MAX_DFS_STREAM_BUFFER_SIZE_PROP, HoodieRealtimeConfig.DEFAULT_MAX_DFS_STREAM_BUFFER_SIZE))
.withLogRecordScannerCallback(record -> {
// convert Hoodie log record to Hadoop AvroWritable and buffer
GenericRecord rec = (GenericRecord) record.getData().getInsertValue(getReaderSchema()).get();
GenericRecord rec = (GenericRecord) record.getData().getInsertValue(getReaderSchema(), payloadProps).get();
ArrayWritable aWritable = (ArrayWritable) HoodieRealtimeRecordReaderUtils.avroToArrayWritable(rec, getHiveSchema());
this.executor.getQueue().insertRecord(aWritable);
})