lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-1470] Use the latest writer schema, when reading from existing parquet files in the hudi-test-suite (#2344)

Browse Source
This commit is contained in:
Balajee Nagasubramaniam
2020-12-18 03:18:52 -08:00
committed by GitHub
parent 8b5d6f9430
commit 5388c7f7a3
1 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
hudi-integ-test/src/main/java/org/apache/hudi/integ/testsuite/reader/DFSHoodieDatasetInputReader.java
View File

@@ -52,6 +52,7 @@ import org.apache.hudi.common.util.ParquetReaderIterator;
import org.apache.hudi.common.util.ValidationUtils;
import org.apache.hudi.config.HoodieMemoryConfig;
import org.apache.parquet.avro.AvroParquetReader;
import org.apache.parquet.avro.AvroReadSupport;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
@@ -243,6 +244,9 @@ public class DFSHoodieDatasetInputReader extends DFSDeltaInputReader {
private Iterator<IndexedRecord> readParquetOrLogFiles(FileSlice fileSlice) throws IOException {
if (fileSlice.getBaseFile().isPresent()) {
// Read the parquet files using the latest writer schema.
Schema schema = new Schema.Parser().parse(schemaStr);
AvroReadSupport.setAvroReadSchema(metaClient.getHadoopConf(), HoodieAvroUtils.addMetadataFields(schema));
Iterator<IndexedRecord> itr =
new ParquetReaderIterator<IndexedRecord>(AvroParquetReader.<IndexedRecord>builder(new
Path(fileSlice.getBaseFile().get().getPath())).withConf(metaClient.getHadoopConf()).build());