Merge pull request 'SDG no DOI' (#464) from sdgnodoi into beta

Reviewed-on: #464
2024-07-24 09:59:13 +02:00 · 2024-07-24 09:59:13 +02:00 · ceb210993c
parent d27e9ea50f 19806c2ae3
commit ceb210993c
6 changed files with 340 additions and 35 deletions
--- a/dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/createunresolvedentities/PrepareSDGSparkJob.java
+++ b/dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/createunresolvedentities/PrepareSDGSparkJob.java
@ -6,26 +6,23 @@ import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;

 import java.io.Serializable;
 import java.util.ArrayList;
+import java.util.Iterator;
 import java.util.List;
+import java.util.Optional;

 import org.apache.commons.io.IOUtils;
 import org.apache.spark.SparkConf;
 import org.apache.spark.api.java.function.MapFunction;
 import org.apache.spark.api.java.function.MapGroupsFunction;
-import org.apache.spark.sql.Dataset;
-import org.apache.spark.sql.Encoders;
-import org.apache.spark.sql.SaveMode;
-import org.apache.spark.sql.SparkSession;
+import org.apache.spark.sql.*;
+import org.jetbrains.annotations.NotNull;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;

-import eu.dnetlib.dhp.actionmanager.createunresolvedentities.model.SDGDataModel;
 import eu.dnetlib.dhp.application.ArgumentApplicationParser;
-import eu.dnetlib.dhp.schema.common.ModelConstants;
+import eu.dnetlib.dhp.schema.common.ModelSupport;
 import eu.dnetlib.dhp.schema.oaf.Result;
-import eu.dnetlib.dhp.schema.oaf.StructuredProperty;
 import eu.dnetlib.dhp.schema.oaf.Subject;
-import eu.dnetlib.dhp.schema.oaf.utils.OafMapperUtils;
 import eu.dnetlib.dhp.utils.DHPUtils;

 public class PrepareSDGSparkJob implements Serializable {
@ -52,42 +49,91 @@ public class PrepareSDGSparkJob implements Serializable {
 		final String outputPath = parser.get("outputPath");
 		log.info("outputPath: {}", outputPath);

+		final Boolean distributeDOI = Optional
+			.ofNullable(parser.get("distributeDoi"))
+			.map(Boolean::valueOf)
+			.orElse(Boolean.TRUE);
+
+		log.info("distribute doi {}", distributeDOI);
+
 		SparkConf conf = new SparkConf();
 		runWithSparkSession(
 			conf,
 			isSparkSessionManaged,
 			spark -> {
+				if (distributeDOI)
 					doPrepare(
 						spark,
 						sourcePath,

 						outputPath);
+				else
+					doPrepareoaid(spark, sourcePath, outputPath);
+
 			});
 	}

 	private static void doPrepare(SparkSession spark, String sourcePath, String outputPath) {
-		Dataset<SDGDataModel> sdgDataset = readPath(spark, sourcePath, SDGDataModel.class);
+		Dataset<Row> sdgDataset = spark
+			.read()
+			.format("csv")
+			.option("sep", DEFAULT_DELIMITER)
+			.option("inferSchema", "true")
+			.option("header", "true")
+			.option("quotes", "\"")
+			.load(sourcePath);

 		sdgDataset
-			.groupByKey((MapFunction<SDGDataModel, String>) r -> r.getDoi().toLowerCase(), Encoders.STRING())
-			.mapGroups((MapGroupsFunction<String, SDGDataModel, Result>) (k, it) -> {
-				Result r = new Result();
-				r.setId(DHPUtils.generateUnresolvedIdentifier(k, DOI));
-				SDGDataModel first = it.next();
-				List<Subject> sbjs = new ArrayList<>();
-				sbjs.add(getSubject(first.getSbj(), SDG_CLASS_ID, SDG_CLASS_NAME, UPDATE_SUBJECT_SDG_CLASS_ID));
-				it
-					.forEachRemaining(
-						s -> sbjs
-							.add(getSubject(s.getSbj(), SDG_CLASS_ID, SDG_CLASS_NAME, UPDATE_SUBJECT_SDG_CLASS_ID)));
-				r.setSubject(sbjs);
-
-				return r;
-			}, Encoders.bean(Result.class))
+			.groupByKey((MapFunction<Row, String>) v -> ((String) v.getAs("doi")).toLowerCase(), Encoders.STRING())
+			.mapGroups(
+				(MapGroupsFunction<String, Row, Result>) (k,
+					it) -> getResult(
+						DHPUtils
+							.generateUnresolvedIdentifier(
+								ModelSupport.entityIdPrefix.get(Result.class.getSimpleName().toLowerCase()) + "|" + k,
+								DOI),
+						it),
+				Encoders.bean(Result.class))
 			.write()
 			.mode(SaveMode.Overwrite)
 			.option("compression", "gzip")
 			.json(outputPath + "/sdg");
 	}

+	private static void doPrepareoaid(SparkSession spark, String sourcePath, String outputPath) {
+		Dataset<Row> sdgDataset = spark
+			.read()
+			.format("csv")
+			.option("sep", DEFAULT_DELIMITER)
+			.option("inferSchema", "true")
+			.option("header", "true")
+			.option("quotes", "\"")
+			.load(sourcePath);
+		;
+
+		sdgDataset
+			.groupByKey((MapFunction<Row, String>) r -> "50|" + ((String) r.getAs("oaid")), Encoders.STRING())
+			.mapGroups(
+				(MapGroupsFunction<String, Row, Result>) PrepareSDGSparkJob::getResult, Encoders.bean(Result.class))
+			.write()
+			.mode(SaveMode.Overwrite)
+			.option("compression", "gzip")
+			.json(outputPath + "/sdg");
+	}
+
+	private static @NotNull Result getResult(String id, Iterator<Row> it) {
+		Result r = new Result();
+		r.setId(id);
+		Row first = it.next();
+		List<Subject> sbjs = new ArrayList<>();
+		sbjs.add(getSubject(first.getAs("sdg"), SDG_CLASS_ID, SDG_CLASS_NAME, UPDATE_SUBJECT_SDG_CLASS_ID));
+		it
+			.forEachRemaining(
+				s -> sbjs
+					.add(getSubject(s.getAs("sdg"), SDG_CLASS_ID, SDG_CLASS_NAME, UPDATE_SUBJECT_SDG_CLASS_ID)));
+		r.setSubject(sbjs);
+
+		return r;
+	}
+
 }
--- a/dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/fosnodoi/CreateActionSetSparkJob.java
+++ b/dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/fosnodoi/CreateActionSetSparkJob.java
@ -13,9 +13,6 @@ import org.apache.hadoop.io.Text;
 import org.apache.hadoop.io.compress.GzipCodec;
 import org.apache.hadoop.mapred.SequenceFileOutputFormat;
 import org.apache.spark.SparkConf;
-import org.apache.spark.api.java.JavaPairRDD;
-import org.apache.spark.api.java.function.FilterFunction;
-import org.apache.spark.api.java.function.FlatMapFunction;
 import org.apache.spark.api.java.function.MapFunction;
 import org.apache.spark.sql.Encoders;
 import org.apache.spark.sql.SparkSession;
@ -24,13 +21,9 @@ import org.slf4j.LoggerFactory;

 import com.fasterxml.jackson.databind.ObjectMapper;

-import eu.dnetlib.dhp.actionmanager.opencitations.model.COCI;
 import eu.dnetlib.dhp.application.ArgumentApplicationParser;
 import eu.dnetlib.dhp.schema.action.AtomicAction;
-import eu.dnetlib.dhp.schema.common.ModelConstants;
-import eu.dnetlib.dhp.schema.oaf.Relation;
 import eu.dnetlib.dhp.schema.oaf.Result;
-import eu.dnetlib.dhp.schema.oaf.utils.*;
 import scala.Tuple2;

 public class CreateActionSetSparkJob implements Serializable {
--- a/dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/sdgnodoi/CreateActionSetSparkJob.java
+++ b/dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/sdgnodoi/CreateActionSetSparkJob.java
@ -0,0 +1,91 @@
+
+package eu.dnetlib.dhp.actionmanager.sdgnodoi;
+
+import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
+
+import java.io.IOException;
+import java.io.Serializable;
+import java.util.Objects;
+import java.util.Optional;
+
+import org.apache.commons.cli.ParseException;
+import org.apache.commons.io.IOUtils;
+import org.apache.hadoop.fs.Hdfs;
+import org.apache.hadoop.io.Text;
+import org.apache.hadoop.io.compress.GzipCodec;
+import org.apache.hadoop.mapred.SequenceFileOutputFormat;
+import org.apache.spark.SparkConf;
+import org.apache.spark.api.java.function.MapFunction;
+import org.apache.spark.sql.Encoders;
+import org.apache.spark.sql.SparkSession;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import com.fasterxml.jackson.databind.ObjectMapper;
+
+import eu.dnetlib.dhp.application.ArgumentApplicationParser;
+import eu.dnetlib.dhp.common.HdfsSupport;
+import eu.dnetlib.dhp.schema.action.AtomicAction;
+import eu.dnetlib.dhp.schema.oaf.Result;
+import scala.Tuple2;
+
+public class CreateActionSetSparkJob implements Serializable {
+
+	private static final Logger log = LoggerFactory.getLogger(CreateActionSetSparkJob.class);
+
+	private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
+
+	public static void main(final String[] args) throws IOException, ParseException {
+
+		final ArgumentApplicationParser parser = new ArgumentApplicationParser(
+			IOUtils
+				.toString(
+					Objects
+						.requireNonNull(
+							CreateActionSetSparkJob.class
+								.getResourceAsStream(
+									"/eu/dnetlib/dhp/actionmanager/fosnodoi/as_parameters.json"))));
+
+		parser.parseArgument(args);
+
+		Boolean isSparkSessionManaged = Optional
+			.ofNullable(parser.get("isSparkSessionManaged"))
+			.map(Boolean::valueOf)
+			.orElse(Boolean.TRUE);
+
+		log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
+
+		final String inputPath = parser.get("sourcePath");
+		log.info("inputPath {}", inputPath);
+
+		final String outputPath = parser.get("outputPath");
+		log.info("outputPath {}", outputPath);
+
+		SparkConf conf = new SparkConf();
+		runWithSparkSession(
+			conf,
+			isSparkSessionManaged,
+			spark -> {
+				HdfsSupport.remove(outputPath, spark.sparkContext().hadoopConfiguration());
+				createActionSet(spark, inputPath, outputPath);
+			});
+
+	}
+
+	private static void createActionSet(SparkSession spark, String inputPath, String outputPath) {
+		spark
+			.read()
+			.textFile(inputPath)
+			.map(
+				(MapFunction<String, Result>) value -> OBJECT_MAPPER.readValue(value, Result.class),
+				Encoders.bean(Result.class))
+			.toJavaRDD()
+			.map(p -> new AtomicAction(p.getClass(), p))
+			.mapToPair(
+				aa -> new Tuple2<>(new Text(aa.getClazz().getCanonicalName()),
+					new Text(OBJECT_MAPPER.writeValueAsString(aa))))
+			.saveAsHadoopFile(
+				outputPath, Text.class, Text.class, SequenceFileOutputFormat.class, GzipCodec.class);
+	}
+
+}
--- a/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/sdgnodoi/as_parameters.json
+++ b/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/sdgnodoi/as_parameters.json
@ -0,0 +1,20 @@
+[
+  {
+    "paramName": "sp",
+    "paramLongName": "sourcePath",
+    "paramDescription": "the zipped opencitations file",
+    "paramRequired": true
+  },
+  {
+    "paramName": "op",
+    "paramLongName": "outputPath",
+    "paramDescription": "the working path",
+    "paramRequired": true
+  },
+  {
+    "paramName": "issm",
+    "paramLongName": "isSparkSessionManaged",
+    "paramDescription": "the hdfs name node",
+    "paramRequired": false
+  }
+]
--- a/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/sdgnodoi/oozie_app/config-default.xml
+++ b/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/sdgnodoi/oozie_app/config-default.xml
@ -0,0 +1,30 @@
+<configuration>
+    <property>
+        <name>jobTracker</name>
+        <value>yarnRM</value>
+    </property>
+    <property>
+        <name>nameNode</name>
+        <value>hdfs://nameservice1</value>
+    </property>
+    <property>
+        <name>oozie.use.system.libpath</name>
+        <value>true</value>
+    </property>
+    <property>
+        <name>hiveMetastoreUris</name>
+        <value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
+    </property>
+    <property>
+        <name>hiveJdbcUrl</name>
+        <value>jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000</value>
+    </property>
+    <property>
+        <name>hiveDbName</name>
+        <value>openaire</value>
+    </property>
+    <property>
+        <name>oozie.launcher.mapreduce.user.classpath.first</name>
+        <value>true</value>
+    </property>
+</configuration>
--- a/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/sdgnodoi/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/sdgnodoi/oozie_app/workflow.xml
@ -0,0 +1,125 @@
+
+<workflow-app name="SDG no doi" xmlns="uri:oozie:workflow:0.5">
+    <parameters>
+        <property>
+            <name>sdgPath</name>
+            <description>the input path of the resources to be extended</description>
+        </property>
+        <property>
+            <name>outputPath</name>
+            <description>the path where to store the actionset</description>
+        </property>
+        <property>
+            <name>sparkDriverMemory</name>
+            <description>memory for driver process</description>
+        </property>
+        <property>
+            <name>sparkExecutorMemory</name>
+            <description>memory for individual executor</description>
+        </property>
+        <property>
+            <name>sparkExecutorCores</name>
+            <description>number of cores used by single executor</description>
+        </property>
+        <property>
+            <name>oozieActionShareLibForSpark2</name>
+            <description>oozie action sharelib for spark 2.*</description>
+        </property>
+        <property>
+            <name>spark2ExtraListeners</name>
+            <value>com.cloudera.spark.lineage.NavigatorAppListener</value>
+            <description>spark 2.* extra listeners classname</description>
+        </property>
+        <property>
+            <name>spark2SqlQueryExecutionListeners</name>
+            <value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
+            <description>spark 2.* sql query execution listeners classname</description>
+        </property>
+        <property>
+            <name>spark2YarnHistoryServerAddress</name>
+            <description>spark 2.* yarn history server address</description>
+        </property>
+        <property>
+            <name>spark2EventLogDir</name>
+            <description>spark 2.* event log dir location</description>
+        </property>
+    </parameters>
+
+    <global>
+        <job-tracker>${jobTracker}</job-tracker>
+        <name-node>${nameNode}</name-node>
+        <configuration>
+            <property>
+                <name>mapreduce.job.queuename</name>
+                <value>${queueName}</value>
+            </property>
+            <property>
+                <name>oozie.launcher.mapred.job.queue.name</name>
+                <value>${oozieLauncherQueueName}</value>
+            </property>
+            <property>
+                <name>oozie.action.sharelib.for.spark</name>
+                <value>${oozieActionShareLibForSpark2}</value>
+            </property>
+
+        </configuration>
+    </global>
+    <start to="prepareSDG"/>
+
+    <kill name="Kill">
+        <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
+    </kill>
+
+    <action name="prepareSDG">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>Produces the results from FOS</name>
+            <class>eu.dnetlib.dhp.actionmanager.createunresolvedentities.PrepareSDGSparkJob</class>
+            <jar>dhp-aggregation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-memory=${sparkExecutorMemory}
+                --executor-cores=${sparkExecutorCores}
+                --driver-memory=${sparkDriverMemory}
+                --conf spark.extraListeners=${spark2ExtraListeners}
+                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
+                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
+                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
+                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
+            </spark-opts>
+            <arg>--sourcePath</arg><arg>${sdgPath}</arg>
+            <arg>--outputPath</arg><arg>${workingDir}/prepared</arg>
+            <arg>--distributeDoi</arg><arg>false</arg>
+        </spark>
+        <ok to="produceActionSet"/>
+        <error to="Kill"/>
+    </action>
+
+
+
+    <action name="produceActionSet">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>Save the action set grouping results with the same id</name>
+            <class>eu.dnetlib.dhp.actionmanager.sdgnodoi.CreateActionSetSparkJob</class>
+            <jar>dhp-aggregation-${projectVersion}.jar</jar>
+            <spark-opts>
+                --executor-memory=${sparkExecutorMemory}
+                --executor-cores=${sparkExecutorCores}
+                --driver-memory=${sparkDriverMemory}
+                --conf spark.extraListeners=${spark2ExtraListeners}
+                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
+                --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
+                --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
+                --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
+            </spark-opts>
+            <arg>--sourcePath</arg><arg>${workingDir}/prepared/sdg</arg>
+            <arg>--outputPath</arg><arg>${outputPath}</arg>
+        </spark>
+        <ok to="End"/>
+        <error to="Kill"/>
+    </action>
+
+    <end name="End"/>
+</workflow-app>