Merge pull request '[Aggregator graph|master] Discard invalid records' (#245) from discard-non-wellformed into master

Reviewed-on: D-Net/dnet-hadoop#245

dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/raw/AbstractMdRecordToOafMapper.java

@@ -107,7 +107,7 @@ public abstract class AbstractMdRecordToOafMapper {
 		this.forceOriginalId = false;
 	}
 
-	public List<Oaf> processMdRecord(final String xml) throws DocumentException {
+	public List<Oaf> processMdRecord(final String xml) {
 
 		DocumentFactory.getInstance().setXPathNamespaceURIs(nsContext);
 		try {
@@ -143,7 +143,7 @@ public abstract class AbstractMdRecordToOafMapper {
 			return createOafs(doc, type, instances, collectedFrom, info, lastUpdateTimestamp);
 		} catch (DocumentException e) {
 			log.error("Error with record:\n" + xml);
-			throw e;
+			return Lists.newArrayList();
 		}
 	}
 

dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/raw/GenerateEntitiesApplication.java

@@ -16,6 +16,9 @@ import org.apache.hadoop.io.compress.GzipCodec;
 import org.apache.spark.SparkConf;
 import org.apache.spark.api.java.JavaRDD;
 import org.apache.spark.api.java.JavaSparkContext;
+import org.apache.spark.rdd.RDD;
+import org.apache.spark.sql.Encoders;
+import org.apache.spark.sql.SaveMode;
 import org.apache.spark.sql.SparkSession;
 import org.dom4j.DocumentException;
 import org.slf4j.Logger;
@@ -127,8 +130,8 @@ public class GenerateEntitiesApplication {
 						.sequenceFile(sp, Text.class, Text.class)
 						.map(k -> new Tuple2<>(k._1().toString(), k._2().toString()))
 						.map(k -> convertToListOaf(k._1(), k._2(), shouldHashId, vocs))
-						.filter(Objects::nonNull)
+						.flatMap(List::iterator)
-						.flatMap(List::iterator));
+						.filter(Objects::nonNull));
 		}
 
 		switch (mode) {
@@ -155,11 +158,11 @@ public class GenerateEntitiesApplication {
 			.saveAsTextFile(targetPath, GzipCodec.class);
 	}
 
-	private static List<Oaf> convertToListOaf(
+	public static List<Oaf> convertToListOaf(
 		final String id,
 		final String s,
 		final boolean shouldHashId,
-		final VocabularyGroup vocs) throws DocumentException {
+		final VocabularyGroup vocs) {
 		final String type = StringUtils.substringAfter(id, ":");
 
 		switch (type.toLowerCase()) {
@@ -200,8 +203,7 @@ public class GenerateEntitiesApplication {
 		try {
 			return OBJECT_MAPPER.readValue(s, clazz);
 		} catch (final Exception e) {
-			log.error("Error parsing object of class: {}", clazz);
+			log.error("Error parsing object of class: {}:\n{}", clazz, s);
-			log.error(s);
 			throw new IllegalArgumentException(e);
 		}
 	}

dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/raw/VerifyRecordsApplication.java

@@ -0,0 +1,108 @@
+
+package eu.dnetlib.dhp.oa.graph.raw;
+
+import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
+
+import java.util.Arrays;
+import java.util.List;
+import java.util.Objects;
+import java.util.Optional;
+import java.util.stream.Collectors;
+
+import org.apache.commons.io.IOUtils;
+import org.apache.hadoop.io.Text;
+import org.apache.spark.SparkConf;
+import org.apache.spark.api.java.JavaSparkContext;
+import org.apache.spark.rdd.RDD;
+import org.apache.spark.sql.Encoders;
+import org.apache.spark.sql.SaveMode;
+import org.apache.spark.sql.SparkSession;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import eu.dnetlib.dhp.application.ArgumentApplicationParser;
+import eu.dnetlib.dhp.common.HdfsSupport;
+import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
+import eu.dnetlib.dhp.schema.oaf.Oaf;
+import eu.dnetlib.dhp.utils.ISLookupClientFactory;
+import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
+import scala.Tuple2;
+
+public class VerifyRecordsApplication {
+
+	private static final Logger log = LoggerFactory.getLogger(VerifyRecordsApplication.class);
+
+	public static void main(final String[] args) throws Exception {
+		final ArgumentApplicationParser parser = new ArgumentApplicationParser(
+			IOUtils
+				.toString(
+					VerifyRecordsApplication.class
+						.getResourceAsStream("/eu/dnetlib/dhp/oa/graph/verify_records_parameters.json")));
+
+		parser.parseArgument(args);
+
+		final Boolean isSparkSessionManaged = Optional
+			.ofNullable(parser.get("isSparkSessionManaged"))
+			.map(Boolean::valueOf)
+			.orElse(Boolean.TRUE);
+		log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
+
+		final String sourcePaths = parser.get("sourcePaths");
+		log.info("sourcePaths: {}", sourcePaths);
+
+		final String invalidPath = parser.get("invalidPath");
+		log.info("invalidPath: {}", invalidPath);
+
+		final String isLookupUrl = parser.get("isLookupUrl");
+		log.info("isLookupUrl: {}", isLookupUrl);
+
+		final ISLookUpService isLookupService = ISLookupClientFactory.getLookUpService(isLookupUrl);
+		final VocabularyGroup vocs = VocabularyGroup.loadVocsFromIS(isLookupService);
+
+		final SparkConf conf = new SparkConf();
+		runWithSparkSession(conf, isSparkSessionManaged, spark -> {
+			HdfsSupport.remove(invalidPath, spark.sparkContext().hadoopConfiguration());
+			validateRecords(spark, sourcePaths, invalidPath, vocs);
+		});
+	}
+
+	private static void validateRecords(SparkSession spark, String sourcePaths, String invalidPath,
+		VocabularyGroup vocs) {
+
+		final JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
+		final List<String> existingSourcePaths = Arrays
+			.stream(sourcePaths.split(","))
+			.filter(p -> HdfsSupport.exists(p, sc.hadoopConfiguration()))
+			.collect(Collectors.toList());
+
+		log.info("Verify records in files:");
+		existingSourcePaths.forEach(log::info);
+
+		for (final String sp : existingSourcePaths) {
+			RDD<String> invalidRecords = sc
+				.sequenceFile(sp, Text.class, Text.class)
+				.map(k -> tryApplyMapping(k._1().toString(), k._2().toString(), true, vocs))
+				.filter(Objects::nonNull)
+				.rdd();
+			spark
+				.createDataset(invalidRecords, Encoders.STRING())
+				.write()
+				.mode(SaveMode.Append)
+				.option("compression", "gzip")
+				.text(invalidPath);
+		}
+	}
+
+	private static String tryApplyMapping(
+		final String id,
+		final String xmlRecord,
+		final boolean shouldHashId,
+		final VocabularyGroup vocs) {
+
+		final List<Oaf> oaf = GenerateEntitiesApplication.convertToListOaf(id, xmlRecord, shouldHashId, vocs);
+		if (Optional.ofNullable(oaf).map(List::isEmpty).orElse(false)) {
+			return xmlRecord;
+		}
+		return null;
+	}
+}

dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/raw/common/AbstractMigrationApplication.java

@@ -4,6 +4,7 @@ package eu.dnetlib.dhp.oa.graph.raw.common;
 import java.io.Closeable;
 import java.io.IOException;
 import java.util.Arrays;
+import java.util.List;
 import java.util.Set;
 import java.util.concurrent.atomic.AtomicInteger;
 import java.util.stream.Collectors;
@@ -24,8 +25,11 @@ import org.apache.http.impl.client.HttpClients;
 import com.fasterxml.jackson.core.JsonProcessingException;
 import com.fasterxml.jackson.databind.ObjectMapper;
 
+import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
+import eu.dnetlib.dhp.oa.graph.raw.OafToOafMapper;
+import eu.dnetlib.dhp.oa.graph.raw.OdfToOafMapper;
 import eu.dnetlib.dhp.schema.mdstore.MDStoreWithInfo;
-import eu.dnetlib.dhp.schema.oaf.Oaf;
+import eu.dnetlib.dhp.schema.oaf.*;
 import eu.dnetlib.dhp.utils.DHPUtils;
 
 public class AbstractMigrationApplication implements Closeable {

dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/raw_all/oozie_app/workflow.xml

@@ -446,10 +446,34 @@
     <join name="wait_import" to="fork_generate_entities"/>
 
     <fork name="fork_generate_entities">
-        <path start="GenerateEntities_claim"/>
+        <path start="VerifyRecords_claim"/>
-        <path start="GenerateEntities"/>
+        <path start="VerifyRecords"/>
     </fork>
 
+    <action name="VerifyRecords_claim">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>VerifyRecords_claim</name>
+            <class>eu.dnetlib.dhp.oa.graph.raw.VerifyRecordsApplication</class>
+            <jar>dhp-graph-mapper-${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}${spark2EventLogDir}
+            </spark-opts>
+            <arg>--sourcePaths</arg><arg>${contentPath}/db_claims,${contentPath}/oaf_claims,${contentPath}/odf_claims</arg>
+            <arg>--invalidPath</arg><arg>${workingDir}/invalid_records_claim</arg>
+            <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
+        </spark>
+        <ok to="GenerateEntities_claim"/>
+        <error to="Kill"/>
+    </action>
+
     <action name="GenerateEntities_claim">
         <spark xmlns="uri:oozie:spark-action:0.2">
             <master>yarn</master>
@@ -499,6 +523,30 @@
         <error to="Kill"/>
     </action>
 
+    <action name="VerifyRecords">
+        <spark xmlns="uri:oozie:spark-action:0.2">
+            <master>yarn</master>
+            <mode>cluster</mode>
+            <name>VerifyRecords</name>
+            <class>eu.dnetlib.dhp.oa.graph.raw.VerifyRecordsApplication</class>
+            <jar>dhp-graph-mapper-${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}${spark2EventLogDir}
+            </spark-opts>
+            <arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_records,${contentPath}/odf_records,${contentPath}/oaf_records_hdfs,${contentPath}/odf_records_hdfs,${contentPath}/oaf_records_invisible</arg>
+            <arg>--invalidPath</arg><arg>${workingDir}/invalid_records</arg>
+            <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
+        </spark>
+        <ok to="GenerateEntities"/>
+        <error to="Kill"/>
+    </action>
+
     <action name="GenerateEntities">
         <spark xmlns="uri:oozie:spark-action:0.2">
             <master>yarn</master>

dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/verify_records_parameters.json

@@ -0,0 +1,26 @@
+[
+	{
+		"paramName": "issm",
+		"paramLongName": "isSparkSessionManaged",
+		"paramDescription": "when true will stop SparkSession after job execution",
+		"paramRequired": false
+	},
+	{
+		"paramName": "s",
+		"paramLongName": "sourcePaths",
+		"paramDescription": "the HDFS source paths which contains the sequential file (comma separated)",
+		"paramRequired": true
+	},
+	{
+		"paramName": "i",
+		"paramLongName": "invalidPath",
+		"paramDescription": "the path of the invalid records file",
+		"paramRequired": false
+	},
+	{
+		"paramName": "isu",
+		"paramLongName": "isLookupUrl",
+		"paramDescription": "the url of the ISLookupService",
+		"paramRequired": true
+	}
+]

dhp-workflows/dhp-graph-mapper/src/test/java/eu/dnetlib/dhp/oa/graph/raw/MappersTest.java

@@ -12,7 +12,6 @@ import java.util.Objects;
 import java.util.Optional;
 import java.util.stream.Collectors;
 
-import kotlin.jvm.Throws;
 import org.apache.commons.io.IOUtils;
 import org.apache.commons.lang3.StringUtils;
 import org.dom4j.DocumentException;
@@ -22,7 +21,6 @@ import org.junit.jupiter.api.extension.ExtendWith;
 import org.mockito.Mock;
 import org.mockito.junit.jupiter.MockitoExtension;
 
-import com.fasterxml.jackson.core.JsonProcessingException;
 import com.fasterxml.jackson.databind.ObjectMapper;
 
 import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
@@ -928,14 +926,12 @@ class MappersTest {
 	}
 
 	@Test
-	void testNotWellFormed() throws IOException, DocumentException {
+	void testNotWellFormed() throws IOException {
 		final String xml = IOUtils
 			.toString(Objects.requireNonNull(getClass().getResourceAsStream("oaf_notwellformed.xml")));
-		final DocumentException generalEx = new DocumentException();
+		final List<Oaf> actual = new OafToOafMapper(vocs, false, true).processMdRecord(xml);
-
+		assertNotNull(actual);
-		DocumentException exception = assertThrows(DocumentException.class, () -> {
+		assertTrue(actual.isEmpty());
-			new OafToOafMapper(vocs, false, true).processMdRecord(xml);
-		});
 	}
 
 	private void assertValidId(final String id) {

dhp-workflows/dhp-graph-mapper/src/test/java/eu/dnetlib/dhp/oa/graph/raw/MigrateDbEntitiesApplicationTest.java

@@ -18,7 +18,6 @@ import java.util.Objects;
 import java.util.Optional;
 import java.util.stream.Collectors;
 
-import eu.dnetlib.dhp.schema.common.ModelConstants;
 import org.apache.commons.io.IOUtils;
 import org.apache.commons.lang3.StringUtils;
 import org.junit.jupiter.api.BeforeEach;
@@ -32,6 +31,7 @@ import com.fasterxml.jackson.core.type.TypeReference;
 import com.fasterxml.jackson.databind.ObjectMapper;
 
 import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup;
+import eu.dnetlib.dhp.schema.common.ModelConstants;
 import eu.dnetlib.dhp.schema.oaf.*;
 import eu.dnetlib.dhp.schema.oaf.utils.OafMapperUtils;