andreas.czerniak
/
BrBETA_dnet-hadoop
forked from D-Net/dnet-hadoop
1
1
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

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
parent ce0cfd79e0 ada063ce70
commit 4da141bd7c
1 changed files with 17 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/raw/MigrateHdfsMdstoresApplication.java
View File

@ -52,9 +52,8 @@ public class MigrateHdfsMdstoresApplication extends AbstractMigrationApplication
public static void main(final String[] args) throws Exception {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(
MigrateHdfsMdstoresApplication.class
.getResourceAsStream("/eu/dnetlib/dhp/oa/graph/migrate_hdfs_mstores_parameters.json")));
.toString(MigrateHdfsMdstoresApplication.class
.getResourceAsStream("/eu/dnetlib/dhp/oa/graph/migrate_hdfs_mstores_parameters.json")));
parser.parseArgument(args);
final Boolean isSparkSessionManaged = Optional
@ -94,14 +93,21 @@ public class MigrateHdfsMdstoresApplication extends AbstractMigrationApplication
.filter(p -> HdfsSupport.exists(p, sc.hadoopConfiguration()))
.toArray(size -> new String[size]);
spark
.read()
.parquet(validPaths)
.map((MapFunction<Row, String>) r -> enrichRecord(r), Encoders.STRING())
.toJavaRDD()
.mapToPair(xml -> new Tuple2<>(new Text(UUID.randomUUID() + ":" + type), new Text(xml)))
// .coalesce(1)
.saveAsHadoopFile(outputPath, Text.class, Text.class, SequenceFileOutputFormat.class, GzipCodec.class);
if (validPaths.length > 0) {
spark
.read()
.parquet(validPaths)
.map((MapFunction<Row, String>) r -> enrichRecord(r), Encoders.STRING())
.toJavaRDD()
.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) {