D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

[UsageCount] fixed query

This commit is contained in:
Miriam Baglioni 2023-02-10 15:50:56 +01:00
parent bb5bba51b3
commit f803530df6
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/usagestats/SparkAtomicActionUsageJob.java
View File

@ -75,17 +75,17 @@ public class SparkAtomicActionUsageJob implements Serializable {
isSparkSessionManaged, isSparkSessionManaged,
spark -> { spark -> {
removeOutputDir(spark, outputPath); removeOutputDir(spark, outputPath);
prepareData(dbname, spark, workingPath + "/usageDb", "usage_stats"); prepareData(dbname, spark, workingPath + "/usageDb", "usage_stats", "result_id");
prepareData(dbname, spark, workingPath + "/projectDb", "project_stats"); prepareData(dbname, spark, workingPath + "/projectDb", "project_stats", "id");
prepareData(dbname, spark, workingPath + "/datasourceDb", "datasource_stats"); prepareData(dbname, spark, workingPath + "/datasourceDb", "datasource_stats", "repositor_id");
writeActionSet(spark, workingPath, outputPath); writeActionSet(spark, workingPath, outputPath);
}); });
} }
private static void prepareData(String dbname, SparkSession spark, String workingPath, String tableName) { private static void prepareData(String dbname, SparkSession spark, String workingPath, String tableName, String attribute_name) {
spark spark
.sql( .sql(
"Select result_id, downloads, views " + "Select " + attribute_name + " as id, downloads, views " +
"from " + dbname + "." + tableName) "from " + dbname + "." + tableName)
.as(Encoders.bean(UsageStatsModel.class)) .as(Encoders.bean(UsageStatsModel.class))
.write() .write()