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

Fixes in workflow.xml

This commit is contained in:
Serafeim Chatzopoulos 2023-07-21 15:26:50 +03:00
parent db4ca43ee8
commit 2cc5b1a39b
10 changed files with 100 additions and 92 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/bipfinder/SparkAtomicActionScoreJob.java
View File

@ -9,7 +9,6 @@ import java.util.List;
import java.util.Optional;
import java.util.stream.Collectors;
import eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers.BipProjectModel;
import org.apache.commons.io.IOUtils;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.SequenceFileOutputFormat;
@ -25,8 +24,9 @@ import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers.BipResultModel;
import eu.dnetlib.dhp.actionmanager.bipmodel.BipScore;
import eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers.BipProjectModel;
import eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers.BipResultModel;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.common.HdfsSupport;
import eu.dnetlib.dhp.schema.action.AtomicAction;
@ -89,8 +89,7 @@ public class SparkAtomicActionScoreJob implements Serializable {
default:
throw new RuntimeException("Unknown target entity: " + targetEntity);
}
}
);
});
}
private static <I extends Project> void prepareProjects(SparkSession spark, String inputPath, String outputPath) {
@ -106,7 +105,8 @@ public class SparkAtomicActionScoreJob implements Serializable {
}, Encoders.bean(Project.class))
.toJavaRDD()
.map(p -> new AtomicAction(Project.class, p))
.mapToPair( aa -> new Tuple2<>(new Text(aa.getClazz().getCanonicalName()),
.mapToPair(
aa -> new Tuple2<>(new Text(aa.getClazz().getCanonicalName()),
new Text(OBJECT_MAPPER.writeValueAsString(aa))))
.saveAsHadoopFile(outputPath, Text.class, Text.class, SequenceFileOutputFormat.class);

41
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/bipmodel/score/deserializers/BipProjectModel.java
View File

@ -1,21 +1,23 @@
package eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers;
import com.opencsv.bean.CsvBindByPosition;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.oaf.KeyValue;
import eu.dnetlib.dhp.schema.oaf.utils.OafMapperUtils;
import lombok.AllArgsConstructor;
import lombok.Getter;
import lombok.NoArgsConstructor;
import lombok.Setter;
import eu.dnetlib.dhp.schema.oaf.Measure;
import static eu.dnetlib.dhp.actionmanager.Constants.*;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import static eu.dnetlib.dhp.actionmanager.Constants.*;
import com.opencsv.bean.CsvBindByPosition;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.oaf.KeyValue;
import eu.dnetlib.dhp.schema.oaf.Measure;
import eu.dnetlib.dhp.schema.oaf.utils.OafMapperUtils;
import lombok.AllArgsConstructor;
import lombok.Getter;
import lombok.NoArgsConstructor;
import lombok.Setter;
@NoArgsConstructor
@AllArgsConstructor
@ -38,32 +40,35 @@ public class BipProjectModel {
KeyValue kv = new KeyValue();
kv.setKey("score");
kv.setValue(measureValue);
kv.setDataInfo(
OafMapperUtils.dataInfo(
kv
.setDataInfo(
OafMapperUtils
.dataInfo(
false,
UPDATE_DATA_INFO_TYPE,
true,
false,
OafMapperUtils.qualifier(
OafMapperUtils
.qualifier(
UPDATE_MEASURE_BIP_CLASS_ID,
UPDATE_CLASS_NAME,
ModelConstants.DNET_PROVENANCE_ACTIONS,
ModelConstants.DNET_PROVENANCE_ACTIONS),
"")
);
""));
Measure measure = new Measure();
measure.setId(measureId);
measure.setUnit(Collections.singletonList(kv));
return measure;
}
public List<Measure> toMeasures() {
return Arrays.asList(
return Arrays
.asList(
createMeasure("numOfInfluentialResults", numOfInfluentialResults),
createMeasure("numOfPopularResults", numOfPopularResults),
createMeasure("totalImpulse", totalImpulse),
createMeasure("totalCitationCount", totalCitationCount)
);
createMeasure("totalCitationCount", totalCitationCount));
}
}

4
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/bipmodel/score/deserializers/BipResultModel.java
View File

@ -1,13 +1,13 @@
package eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers;
import eu.dnetlib.dhp.actionmanager.bipmodel.Score;
import java.io.Serializable;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import eu.dnetlib.dhp.actionmanager.bipmodel.Score;
/**
* Class that maps the model of the bipFinder! input data.
* Only needed for deserialization purposes

2
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/createunresolvedentities/PrepareBipFinder.java
View File

@ -24,8 +24,8 @@ import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers.BipResultModel;
import eu.dnetlib.dhp.actionmanager.bipmodel.BipScore;
import eu.dnetlib.dhp.actionmanager.bipmodel.score.deserializers.BipResultModel;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.common.HdfsSupport;
import eu.dnetlib.dhp.schema.common.ModelConstants;

14
dhp-workflows/dhp-aggregation/src/test/java/eu/dnetlib/dhp/actionmanager/bipfinder/SparkAtomicActionScoreJobTest.java
View File

@ -7,8 +7,6 @@ import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import eu.dnetlib.dhp.schema.oaf.KeyValue;
import eu.dnetlib.dhp.schema.oaf.Project;
import org.apache.commons.io.FileUtils;
import org.apache.hadoop.io.Text;
import org.apache.spark.SparkConf;
@ -28,6 +26,8 @@ import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.databind.ObjectMapper;
import eu.dnetlib.dhp.schema.action.AtomicAction;
import eu.dnetlib.dhp.schema.oaf.KeyValue;
import eu.dnetlib.dhp.schema.oaf.Project;
import eu.dnetlib.dhp.schema.oaf.Result;
public class SparkAtomicActionScoreJobTest {
@ -73,15 +73,16 @@ public class SparkAtomicActionScoreJobTest {
}
private void runJob(String inputPath, String outputPath, String targetEntity) throws Exception {
SparkAtomicActionScoreJob.main(
SparkAtomicActionScoreJob
.main(
new String[] {
"-isSparkSessionManaged", Boolean.FALSE.toString(),
"-inputPath", inputPath,
"-outputPath", outputPath,
"-targetEntity", targetEntity,
});
}
);
}
@Test
void testResultScores() throws Exception {
final String targetEntity = RESULT;
@ -171,7 +172,8 @@ public class SparkAtomicActionScoreJobTest {
// count that the project with id testProjectId is present
assertEquals(1, projects.filter(row -> row.getId().equals(testProjectId)).count());
projects.filter(row -> row.getId().equals(testProjectId))
projects
.filter(row -> row.getId().equals(testProjectId))
.flatMap(r -> r.getMeasures().iterator())
.foreach(m -> {
log.info(m.getId() + " " + m.getUnit());

2
dhp-workflows/dhp-impact-indicators/src/main/resources/eu/dnetlib/dhp/oa/graph/impact_indicators/job.properties
View File

@ -97,7 +97,7 @@ oozie.wf.application.path=${oozieTopWfApplicationPath}
# Path where the final output should be?
actionSetOutputPath=${workingDir}/bip_actionsets/
actionSetOutputPath=${workingDir}/bip_actionsets
# The directory to store project impact indicators
projectImpactIndicatorsOutput=${workingDir}/project_indicators

33
dhp-workflows/dhp-impact-indicators/src/main/resources/eu/dnetlib/dhp/oa/graph/impact_indicators/oozie_app/workflow.xml
View File

@ -24,20 +24,21 @@
<case to="spark-impulse">${wf:conf('resume') eq "impulse"}</case>
<case to="spark-pagerank">${wf:conf('resume') eq "pagerank"}</case>
<case to="spark-attrank">${wf:conf('resume') eq "attrank"}</case>
<!-- Removed for serafeim -->
<!-- <case to="iterative-rankings">${wf:conf('resume') eq "rankings-iterative"}</case> -->
<case to="get-file-names">${wf:conf('resume') eq "format-results"}</case>
<case to="map-openaire-to-doi">${wf:conf('resume') eq "map-ids"}</case>
<case to="map-scores-to-dois">${wf:conf('resume') eq "map-scores"}</case>
<case to="create-openaire-ranking-graph">${wf:conf('resume') eq "start"}</case>
<case to="project-impact-indicators">${wf:conf('resume') eq "projects-impact"}</case>
<!-- TODO: add action set creation here -->
<!-- Aggregation of impact scores on the project level -->
<case to="project-impact-indicators">${wf:conf('resume') eq "projects-impact"}</case>
<case to="create-actionset-for-projects">${wf:conf('resume') eq "projects-impact-actionsets"}</case>
<default to="create-openaire-ranking-graph" />
</switch>
</decision>
<!-- Script here written by Serafeim: maps openaire ids to their synonyms -->
<!-- maps openaire ids to their synonyms -->
<action name="create-openaire-ranking-graph">
<!-- This is required as a tag for spark jobs, regardless of programming language -->
<spark xmlns="uri:oozie:spark-action:0.2">
@ -479,7 +480,7 @@
<!-- Finish formatting data and end -->
<join name="join-file-formatting" to="map-openaire-to-doi"/>
<!-- Script here written by Serafeim: maps openaire ids to their synonyms -->
<!-- maps openaire ids to their synonyms -->
<action name="map-openaire-to-doi">
<!-- This is required as a tag for spark jobs, regardless of programming language -->
<spark xmlns="uri:oozie:spark-action:0.2">
@ -526,7 +527,7 @@
</action>
<!-- Script here written by Serafeim: maps openaire ids to their synonyms -->
<!-- maps openaire ids to their synonyms -->
<action name="map-scores-to-dois">
<!-- This is required as a tag for spark jobs, regardless of programming language -->
<spark xmlns="uri:oozie:spark-action:0.2">
@ -568,14 +569,14 @@
</spark>
<!-- Do this after finishing okay -->
<ok to="deleteOutputPathForActionSet" />
<!-- This is the initial code <ok to="deleteOutputPathForActionSet" /> -->
<ok to="delete-output-path-for-actionset" />
<!-- This is the initial code <ok to="delete-output-path-for-actionset" /> -->
<!-- Go there if we have an error -->
<error to="map-scores-fail" />
</action>
<action name="deleteOutputPathForActionSet">
<action name="delete-output-path-for-actionset">
<fs>
<delete path="${actionSetOutputPath}/results/"/>
<delete path="${actionSetOutputPath}/projects/"/>
@ -583,13 +584,13 @@
<mkdir path="${actionSetOutputPath}/results/"/>
<mkdir path="${actionSetOutputPath}/projects/"/>
</fs>
<ok to="createActionSetForResults"/>
<ok to="create-actionset-for-results"/>
<error to="actionset-delete-fail"/>
</action>
<action name="createActionSetForResults">
<action name="create-actionset-for-results">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>Produces the atomic action with the bip finder scores for publications</name>
<class>eu.dnetlib.dhp.actionmanager.bipfinder.SparkAtomicActionScoreJob</class>
@ -640,7 +641,7 @@
<!-- Script arguments here -->
<!-- graph data folder from which to read relations -->
<arg>${openaireDataInput}/relations</arg>
<arg>${openaireDataInput}/relation</arg>
<!-- input files with impact indicators for results -->
<arg>${nameNode}/${workingDir}/${wf:actionData('get-file-names')['pr_file']}</arg>
@ -658,16 +659,16 @@
</spark>
<!-- Do this after finishing okay -->
<ok to="createActionSetForProjects" />
<ok to="create-actionset-for-projects" />
<!-- Go there if we have an error -->
<error to="project-impact-indicators-fail" />
</action>
<action name="createActionSetForProjects">
<action name="create-actionset-for-projects">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>Produces the atomic action with the bip finder scores for projects</name>
<class>eu.dnetlib.dhp.actionmanager.bipfinder.SparkAtomicActionScoreJob</class>