dhp-schemas/pom.xml

@@ -6,7 +6,7 @@
         <groupId>eu.dnetlib.dhp</groupId>
         <artifactId>dhp</artifactId>
         <version>1.2.4-SNAPSHOT</version>
-        <relativePath>../</relativePath>
+        <relativePath>../pom.xml</relativePath>
     </parent>
 
     <artifactId>dhp-schemas</artifactId>

dhp-schemas/src/main/java/eu/dnetlib/dhp/schema/orcid/OrcidDOI.java

@@ -0,0 +1,24 @@
+package eu.dnetlib.dhp.schema.orcid;
+
+import java.util.List;
+
+public class OrcidDOI {
+    private String doi;
+    private List<AuthorData> authors;
+
+    public String getDoi() {
+        return doi;
+    }
+
+    public void setDoi(String doi) {
+        this.doi = doi;
+    }
+
+    public List<AuthorData> getAuthors() {
+        return authors;
+    }
+
+    public void setAuthors(List<AuthorData> authors) {
+        this.authors = authors;
+    }
+}

dhp-workflows/dhp-doiboost/src/main/java/eu/dnetlib/doiboost/crossref/Crossref2Oaf.scala

@@ -200,7 +200,7 @@ case object Crossref2Oaf {
     a.setSurname(family)
     a.setFullname(s"$given $family")
     if (StringUtils.isNotBlank(orcid))
-      a.setPid(List(createSP(orcid, ORCID, PID_TYPES)).asJava)
+      a.setPid(List(createSP(orcid, ORCID, PID_TYPES, generateDataInfo())).asJava)
 
     a
   }
@@ -248,7 +248,7 @@ case object Crossref2Oaf {
 
 
     def snsfRule(award:String): String = {
-      var tmp1 = StringUtils.substringAfter(award,"_")
+      val tmp1 = StringUtils.substringAfter(award,"_")
       val tmp2 = StringUtils.substringBefore(tmp1,"/")
       logger.debug(s"From $award to $tmp2")
       tmp2
@@ -294,7 +294,7 @@ case object Crossref2Oaf {
     }
 
     def getProjectId (nsPrefix:String, targetId:String):String = {
-      "40|$nsPrefix::$targetId"
+      s"40|$nsPrefix::$targetId"
     }
 
 

dhp-workflows/dhp-doiboost/src/main/java/eu/dnetlib/doiboost/crossref/CrossrefDataset.scala

@@ -2,6 +2,7 @@ package eu.dnetlib.doiboost.crossref
 
 import eu.dnetlib.dhp.application.ArgumentApplicationParser
 import org.apache.commons.io.IOUtils
+import org.apache.hadoop.io.{IntWritable, Text}
 import org.apache.spark.SparkConf
 import org.apache.spark.sql.expressions.Aggregator
 import org.apache.spark.sql.{Dataset, Encoder, Encoders, SaveMode, SparkSession}
@@ -12,21 +13,23 @@ import org.slf4j.{Logger, LoggerFactory}
 
 object CrossrefDataset {
 
+  val logger: Logger = LoggerFactory.getLogger(SparkMapDumpIntoOAF.getClass)
 
-  def extractTimestamp(input:String): Long = {
+
+  def to_item(input:String):CrossrefDT = {
 
     implicit lazy val formats: DefaultFormats.type = org.json4s.DefaultFormats
     lazy val json: json4s.JValue = parse(input)
-
+    val ts:Long = (json \ "indexed" \ "timestamp").extract[Long]
-    (json\"indexed"\"timestamp").extractOrElse[Long](0)
+    val doi:String  = (json \ "DOI").extract[String]
+    CrossrefDT(doi, input, ts)
 
   }
 
-
   def main(args: Array[String]): Unit = {
 
 
-    val logger: Logger = LoggerFactory.getLogger(SparkMapDumpIntoOAF.getClass)
+
     val conf: SparkConf = new SparkConf()
     val parser = new ArgumentApplicationParser(IOUtils.toString(CrossrefDataset.getClass.getResourceAsStream("/eu/dnetlib/dhp/doiboost/crossref_to_dataset_params.json")))
     parser.parseArgument(args)
@@ -49,9 +52,8 @@ object CrossrefDataset {
         if (a == null)
           return b
 
-        val tb = extractTimestamp(b.json)
+
-        val ta = extractTimestamp(a.json)
+        if(a.timestamp >b.timestamp) {
-        if(ta >tb) {
           return a
         }
         b
@@ -63,9 +65,7 @@ object CrossrefDataset {
         if (a == null)
           return b
 
-        val tb = extractTimestamp(b.json)
+        if(a.timestamp >b.timestamp) {
-        val ta = extractTimestamp(a.json)
-        if(ta >tb) {
           return a
         }
         b
@@ -78,15 +78,21 @@ object CrossrefDataset {
       override def finish(reduction: CrossrefDT): CrossrefDT = reduction
     }
 
-    val sourcePath:String = parser.get("sourcePath")
+    val workingPath:String = parser.get("workingPath")
-    val targetPath:String = parser.get("targetPath")
 
-    val ds:Dataset[CrossrefDT] = spark.read.load(sourcePath).as[CrossrefDT]
 
-    ds.groupByKey(_.doi)
+    val main_ds:Dataset[CrossrefDT] = spark.read.load(s"$workingPath/crossref_ds").as[CrossrefDT]
+
+
+    val update =
+      spark.createDataset(spark.sparkContext.sequenceFile(s"$workingPath/index_update",  classOf[IntWritable], classOf[Text])
+        .map(i =>CrossrefImporter.decompressBlob(i._2.toString))
+        .map(i =>to_item(i)))
+
+    main_ds.union(update).groupByKey(_.doi)
       .agg(crossrefAggregator.toColumn)
       .map(s=>s._2)
-      .write.mode(SaveMode.Overwrite).save(targetPath)
+      .write.mode(SaveMode.Overwrite).save(s"$workingPath/crossref_ds_updated")
 
   }
 

dhp-workflows/dhp-doiboost/src/main/java/eu/dnetlib/doiboost/crossref/SparkMapDumpIntoOAF.scala

@@ -34,85 +34,21 @@ object SparkMapDumpIntoOAF {
     implicit val mapEncoderRelatons: Encoder[Relation] = Encoders.kryo[Relation]
     implicit val mapEncoderDatasets: Encoder[oaf.Dataset] = Encoders.kryo[OafDataset]
 
-    val sc = spark.sparkContext
     val targetPath = parser.get("targetPath")
     import spark.implicits._
 
-
     spark.read.load(parser.get("sourcePath")).as[CrossrefDT]
       .flatMap(k => Crossref2Oaf.convert(k.json))
       .filter(o => o != null)
       .write.mode(SaveMode.Overwrite).save(s"$targetPath/mixObject")
 
-
     val ds:Dataset[Oaf] = spark.read.load(s"$targetPath/mixObject").as[Oaf]
 
-    ds.filter(o => o.isInstanceOf[Publication]).map(o => o.asInstanceOf[Publication]).write.save(s"$targetPath/publication")
+    ds.filter(o => o.isInstanceOf[Publication]).map(o => o.asInstanceOf[Publication]).write.mode(SaveMode.Overwrite).save(s"$targetPath/crossrefPublication")
 
-    ds.filter(o => o.isInstanceOf[Relation]).map(o => o.asInstanceOf[Relation]).write.save(s"$targetPath/relation")
+    ds.filter(o => o.isInstanceOf[Relation]).map(o => o.asInstanceOf[Relation]).write.mode(SaveMode.Overwrite).save(s"$targetPath/crossrefRelation")
 
-    ds.filter(o => o.isInstanceOf[OafDataset]).map(o => o.asInstanceOf[OafDataset]).write.save(s"$targetPath/dataset")
+    ds.filter(o => o.isInstanceOf[OafDataset]).map(o => o.asInstanceOf[OafDataset]).write.mode(SaveMode.Overwrite).save(s"$targetPath/crossrefDataset")
-
-
-
-//
-//
-//
-//    sc.sequenceFile(parser.get("sourcePath"), classOf[IntWritable], classOf[Text])
-//      .map(k => k._2.toString).map(CrossrefImporter.decompressBlob)
-//      .flatMap(k => Crossref2Oaf.convert(k)).saveAsObjectFile(s"${targetPath}/mixObject")
-//
-//    val inputRDD = sc.objectFile[Oaf](s"${targetPath}/mixObject").filter(p=> p!= null)
-//
-//    val distinctPubs:RDD[Publication] = inputRDD.filter(k => k != null && k.isInstanceOf[Publication])
-//      .map(k => k.asInstanceOf[Publication]).map { p: Publication => Tuple2(p.getId, p) }.reduceByKey { case (p1: Publication, p2: Publication) =>
-//      var r = if (p1 == null) p2 else p1
-//      if (p1 != null && p2 != null) {
-//        if (p1.getLastupdatetimestamp != null && p2.getLastupdatetimestamp != null) {
-//          if (p1.getLastupdatetimestamp < p2.getLastupdatetimestamp)
-//            r = p2
-//          else
-//            r = p1
-//        } else {
-//          r = if (p1.getLastupdatetimestamp == null) p2 else p1
-//        }
-//      }
-//      r
-//    }.map(_._2)
-//
-//    val pubs:Dataset[Publication] = spark.createDataset(distinctPubs)
-//    pubs.write.mode(SaveMode.Overwrite).save(s"${targetPath}/publication")
-//
-//
-//    val distincDatasets:RDD[OafDataset] = inputRDD.filter(k => k != null && k.isInstanceOf[OafDataset])
-//      .map(k => k.asInstanceOf[OafDataset]).map(p => Tuple2(p.getId, p)).reduceByKey { case (p1: OafDataset, p2: OafDataset) =>
-//      var r = if (p1 == null) p2 else p1
-//      if (p1 != null && p2 != null) {
-//        if (p1.getLastupdatetimestamp != null && p2.getLastupdatetimestamp != null) {
-//          if (p1.getLastupdatetimestamp < p2.getLastupdatetimestamp)
-//            r = p2
-//          else
-//            r = p1
-//        } else {
-//          r = if (p1.getLastupdatetimestamp == null) p2 else p1
-//        }
-//      }
-//      r
-//    }.map(_._2)
-//
-//    spark.createDataset(distincDatasets).write.mode(SaveMode.Overwrite).save(s"${targetPath}/dataset")
-//
-//
-//
-//    val distinctRels =inputRDD.filter(k => k != null && k.isInstanceOf[Relation])
-//      .map(k => k.asInstanceOf[Relation]).map(r=> (s"${r.getSource}::${r.getTarget}",r))
-//      .reduceByKey { case (p1: Relation, p2: Relation) =>
-//        if (p1 == null) p2 else p1
-//      }.map(_._2)
-//
-//    val rels: Dataset[Relation] = spark.createDataset(distinctRels)
-//
-//    rels.write.mode(SaveMode.Overwrite).save(s"${targetPath}/relations")
   }
 
 

dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/crossref/oozie_app/workflow.xml

@@ -16,88 +16,86 @@
             <name>sparkExecutorCores</name>
             <description>number of cores used by single executor</description>
         </property>
-<!--        <property>-->
+        <property>
-<!--            <name>timestamp</name>-->
+            <name>timestamp</name>
-<!--            <description>Timestamp for incremental Harvesting</description>-->
+            <description>Timestamp for incremental Harvesting</description>
-<!--        </property>-->
+        </property>
 
     </parameters>
 
-    <start to="ExtractCrossrefToOAF"/>
+    <start to="ImportCrossRef"/>
 
 
     <kill name="Kill">
         <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
     </kill>
 
-<!--    <action name="ResetWorkingPath">-->
+    <action name="ImportCrossRef">
-<!--        <fs>-->
+        <java>
-<!--            <delete path='${workingPath}/input/crossref/index_dump'/>-->
+            <job-tracker>${jobTracker}</job-tracker>
-<!--&lt;!&ndash;            <mkdir path='${workingPath}/input/crossref'/>&ndash;&gt;-->
+            <name-node>${nameNode}</name-node>
-<!--        </fs>-->
+            <main-class>eu.dnetlib.doiboost.crossref.CrossrefImporter</main-class>
-<!--        <ok to="ImportCrossRef"/>-->
+            <arg>-t</arg><arg>${workingPath}/input/crossref/index_update</arg>
-<!--        <error to="Kill"/>-->
+            <arg>-n</arg><arg>${nameNode}</arg>
-<!--    </action>-->
+            <arg>-ts</arg><arg>${timestamp}</arg>
+        </java>
+        <ok to="GenerateDataset"/>
+        <error to="Kill"/>
+    </action>
 
-
+    <action name="GenerateDataset">
-
-<!--    <action name="ImportCrossRef">-->
-<!--        <java>-->
-<!--            <job-tracker>${jobTracker}</job-tracker>-->
-<!--            <name-node>${nameNode}</name-node>-->
-<!--            <main-class>eu.dnetlib.doiboost.crossref.CrossrefImporter</main-class>-->
-<!--            <arg>-t</arg><arg>${workingPath}/input/crossref/index_dump_1</arg>-->
-<!--            <arg>-n</arg><arg>${nameNode}</arg>-->
-<!--            <arg>-ts</arg><arg>${timestamp}</arg>-->
-<!--        </java>-->
-<!--        <ok to="End"/>-->
-<!--        <error to="Kill"/>-->
-<!--    </action>-->
-
-
-    <action name="ExtractCrossrefToOAF">
             <spark xmlns="uri:oozie:spark-action:0.2">
                 <master>yarn-cluster</master>
                 <mode>cluster</mode>
                 <name>ExtractCrossrefToOAF</name>
+                <class>eu.dnetlib.doiboost.crossref.CrossrefDataset</class>
+                <jar>dhp-doiboost-${projectVersion}.jar</jar>
+                <spark-opts>
+                    --executor-memory=${sparkExecutorMemory}
+                    --executor-cores=${sparkExecutorCores}
+                    --driver-memory=${sparkDriverMemory}
+                    --conf spark.sql.shuffle.partitions=3840
+                    ${sparkExtraOPT}
+                </spark-opts>
+                <arg>--workingPath</arg><arg>/data/doiboost/input/crossref</arg>
+                <arg>--master</arg><arg>yarn-cluster</arg>
+            </spark>
+            <ok to="RenameDataset"/>
+            <error to="Kill"/>
+    </action>
+
+    <action name="RenameDataset">
+        <fs>
+            <delete path='${workingPath}/input/crossref/crossref_ds'/>
+            <move source="${workingPath}/input/crossref/crossref_ds_updated"
+                  target="${workingPath}/input/crossref/crossref_ds"/>
+        </fs>
+        <ok to="ConvertCrossrefToOAF"/>
+        <error to="Kill"/>
+    </action>
+
+
+    <action name="ConvertCrossrefToOAF">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn-cluster</master>
+            <mode>cluster</mode>
+            <name>ConvertCrossrefToOAF</name>
             <class>eu.dnetlib.doiboost.crossref.SparkMapDumpIntoOAF</class>
             <jar>dhp-doiboost-${projectVersion}.jar</jar>
             <spark-opts>
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
+                --conf spark.sql.shuffle.partitions=3840
                 ${sparkExtraOPT}
             </spark-opts>
             <arg>--sourcePath</arg><arg>${workingPath}/input/crossref/crossref_ds</arg>
-            <arg>--targetPath</arg><arg>${workingPath}/input/crossref</arg>
+            <arg>--targetPath</arg><arg>${workingPath}/process/</arg>
             <arg>--master</arg><arg>yarn-cluster</arg>
         </spark>
         <ok to="End"/>
         <error to="Kill"/>
     </action>
 
-
-
-<!--    <action name="GenerateDataset">-->
-<!--        <spark xmlns="uri:oozie:spark-action:0.2">-->
-<!--            <master>yarn-cluster</master>-->
-<!--            <mode>cluster</mode>-->
-<!--            <name>ExtractCrossrefToOAF</name>-->
-<!--            <class>eu.dnetlib.doiboost.crossref.CrossrefDataset</class>-->
-<!--            <jar>dhp-doiboost-${projectVersion}.jar</jar>-->
-<!--            <spark-opts>-->
-<!--                &#45;&#45;executor-memory=${sparkExecutorMemory}-->
-<!--                &#45;&#45;executor-cores=${sparkExecutorCores}-->
-<!--                &#45;&#45;driver-memory=${sparkDriverMemory}-->
-<!--                ${sparkExtraOPT}-->
-<!--            </spark-opts>-->
-<!--            <arg>&#45;&#45;sourcePath</arg><arg>/data/doiboost/crossref/cr_dataset</arg>-->
-<!--            <arg>&#45;&#45;targetPath</arg><arg>/data/doiboost/crossref/crossrefDataset</arg>-->
-<!--            <arg>&#45;&#45;master</arg><arg>yarn-cluster</arg>-->
-<!--        </spark>-->
-<!--        <ok to="End"/>-->
-<!--        <error to="Kill"/>-->
-<!--    </action>-->
-
     <end name="End"/>
 </workflow-app>

dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/crossref_to_dataset_params.json

@@ -1,6 +1,5 @@
 [
-  {"paramName":"s",   "paramLongName":"sourcePath", "paramDescription": "the path of the sequencial file to read",  "paramRequired": true},
+  {"paramName":"w",   "paramLongName":"workingPath", "paramDescription": "the working dir path",                      "paramRequired": true},
-  {"paramName":"t",   "paramLongName":"targetPath", "paramDescription": "the working dir path",                      "paramRequired": true},
   {"paramName":"m",   "paramLongName":"master",     "paramDescription": "the master name",                          "paramRequired": true}
 
 ]

dhp-workflows/dhp-graph-provision/src/test/java/eu/dnetlib/dhp/oa/provision/XmlRecordFactoryTest.java

@@ -5,31 +5,23 @@ import static org.junit.jupiter.api.Assertions.*;
 
 import java.io.IOException;
 import java.io.StringReader;
-import java.util.List;
 
 import org.apache.commons.io.IOUtils;
-import org.apache.commons.lang3.StringUtils;
 import org.dom4j.Document;
 import org.dom4j.DocumentException;
 import org.dom4j.io.SAXReader;
 import org.junit.jupiter.api.Assertions;
 import org.junit.jupiter.api.Disabled;
 import org.junit.jupiter.api.Test;
-import org.mockito.Mock;
 
 import com.fasterxml.jackson.databind.ObjectMapper;
 
 import eu.dnetlib.dhp.oa.provision.model.JoinedEntity;
 import eu.dnetlib.dhp.oa.provision.utils.ContextMapper;
 import eu.dnetlib.dhp.oa.provision.utils.XmlRecordFactory;
-import eu.dnetlib.dhp.schema.oaf.Oaf;
-import eu.dnetlib.dhp.schema.oaf.OafEntity;
-import eu.dnetlib.dhp.schema.oaf.OafMapperUtils;
-import eu.dnetlib.dhp.schema.oaf.Publication;
-import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
 
 //TODO to enable it we need to update the joined_entity.json test file
-//@Disabled
+@Disabled
 public class XmlRecordFactoryTest {
 
 	private static final String otherDsTypeId = "scholarcomminfra,infospace,pubsrepository::mock,entityregistry,entityregistry::projects,entityregistry::repositories,websource";
@@ -43,27 +35,6 @@ public class XmlRecordFactoryTest {
 		JoinedEntity je = new ObjectMapper().readValue(json, JoinedEntity.class);
 		assertNotNull(je);
 
-		Document doc = buildXml(je);
-		//// TODO specific test assertion on doc
-	}
-
-	@Test
-	void testBologna() throws IOException, DocumentException {
-		final String json = IOUtils.toString(getClass().getResourceAsStream("oaf-bologna.json"));
-		Publication oaf = new ObjectMapper().readValue(json, Publication.class);
-		assertNotNull(oaf);
-		JoinedEntity je = new JoinedEntity();
-		je.setEntity(oaf);
-		assertNotNull(je);
-
-		Document doc = buildXml(je);
-		// TODO specific test assertion on doc
-
-		System.out.println(doc.asXML());
-
-	}
-
-	private Document buildXml(JoinedEntity je) throws DocumentException {
 		ContextMapper contextMapper = new ContextMapper();
 
 		XmlRecordFactory xmlRecordFactory = new XmlRecordFactory(contextMapper, false, XmlConverterJob.schemaLocation,
@@ -78,7 +49,5 @@ public class XmlRecordFactoryTest {
 		assertNotNull(doc);
 
 		// TODO add assertions based of values extracted from the XML record
-
-		return doc;
 	}
 }

dhp-workflows/pom.xml

@@ -7,7 +7,7 @@
         <groupId>eu.dnetlib.dhp</groupId>
         <artifactId>dhp</artifactId>
         <version>1.2.4-SNAPSHOT</version>
-        <relativePath>../</relativePath>
+        <relativePath>../pom.xml</relativePath>
     </parent>
 
     <artifactId>dhp-workflows</artifactId>