sabeel
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

[country propagation] fixes error 'cannot resolve countrySet given input columns: []' when there is no prepared information driving the propagation process for a given result type

This commit is contained in:
Claudio Atzori 2023-01-19 15:57:43 +01:00
parent 699736addc
commit 3800361033
1 changed files with 19 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
dhp-workflows/dhp-enrichment/src/main/java/eu/dnetlib/dhp/countrypropagation/SparkCountryPropagationJob.java
View File

@ -14,6 +14,7 @@ import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.Row;
import org.apache.spark.sql.SaveMode;
import org.apache.spark.sql.SparkSession;
import org.slf4j.Logger;
@ -84,19 +85,26 @@ public class SparkCountryPropagationJob {
Dataset<R> res = readPath(spark, sourcePath, resultClazz);
log.info("Reading prepared info: {}", preparedInfoPath);
Dataset<ResultCountrySet> prepared = spark
final Dataset<Row> preparedInfoRaw = spark
.read()
.json(preparedInfoPath)
.as(Encoders.bean(ResultCountrySet.class));
res
.joinWith(prepared, res.col("id").equalTo(prepared.col("resultId")), "left_outer")
.map(getCountryMergeFn(), Encoders.bean(resultClazz))
.write()
.option("compression", "gzip")
.mode(SaveMode.Overwrite)
.json(outputPath);
.json(preparedInfoPath);
if (!preparedInfoRaw.isEmpty()) {
final Dataset<ResultCountrySet> prepared = preparedInfoRaw.as(Encoders.bean(ResultCountrySet.class));
res
.joinWith(prepared, res.col("id").equalTo(prepared.col("resultId")), "left_outer")
.map(getCountryMergeFn(), Encoders.bean(resultClazz))
.write()
.option("compression", "gzip")
.mode(SaveMode.Overwrite)
.json(outputPath);
} else {
res
.write()
.option("compression", "gzip")
.mode(SaveMode.Overwrite)
.json(outputPath);
}
}
private static <R extends Result> MapFunction<Tuple2<R, ResultCountrySet>, R> getCountryMergeFn() {