Merge branch 'stable_ids' of https://code-repo.d4science.org/D-Net/dnet-hadoop into stable_ids

This commit is contained in:
Miriam Baglioni 2021-06-14 13:41:02 +02:00
commit 4da141bd7c
1 changed files with 17 additions and 11 deletions

View File

@ -52,8 +52,7 @@ public class MigrateHdfsMdstoresApplication extends AbstractMigrationApplication
public static void main(final String[] args) throws Exception {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(
MigrateHdfsMdstoresApplication.class
.toString(MigrateHdfsMdstoresApplication.class
.getResourceAsStream("/eu/dnetlib/dhp/oa/graph/migrate_hdfs_mstores_parameters.json")));
parser.parseArgument(args);
@ -94,6 +93,7 @@ public class MigrateHdfsMdstoresApplication extends AbstractMigrationApplication
.filter(p -> HdfsSupport.exists(p, sc.hadoopConfiguration()))
.toArray(size -> new String[size]);
if (validPaths.length > 0) {
spark
.read()
.parquet(validPaths)
@ -102,6 +102,12 @@ public class MigrateHdfsMdstoresApplication extends AbstractMigrationApplication
.mapToPair(xml -> new Tuple2<>(new Text(UUID.randomUUID() + ":" + type), new Text(xml)))
// .coalesce(1)
.saveAsHadoopFile(outputPath, Text.class, Text.class, SequenceFileOutputFormat.class, GzipCodec.class);
} else {
spark.emptyDataFrame()
.toJavaRDD()
.mapToPair(xml -> new Tuple2<>(new Text(), new Text()))
.saveAsHadoopFile(outputPath, Text.class, Text.class, SequenceFileOutputFormat.class, GzipCodec.class);
}
}
private static String enrichRecord(final Row r) {