Merge branch 'master' of code-repo.d4science.org:D-Net/dnet-hadoop

Reviewed-on: #426

CODE_OF_CONDUCT.md

@@ -0,0 +1,43 @@
+# Contributor Code of Conduct
+
+Openness, transparency and our community-driven participatory approach guide us in our day-to-day interactions and decision-making. Our open source projects are no exception. Trust, respect, collaboration and transparency are core values we believe should live and breathe within our projects. Our community welcomes participants from around the world with different experiences, unique perspectives, and great ideas to share.
+
+## Our Pledge
+
+In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation.
+
+## Our Standards
+
+Examples of behavior that contributes to creating a positive environment include:
+
+- Using welcoming and inclusive language
+- Being respectful of differing viewpoints and experiences
+- Gracefully accepting constructive criticism
+- Attempting collaboration before conflict
+- Focusing on what is best for the community
+- Showing empathy towards other community members
+
+Examples of unacceptable behavior by participants include:
+
+- Violence, threats of violence, or inciting others to commit self-harm
+- The use of sexualized language or imagery and unwelcome sexual attention or advances
+- Trolling, intentionally spreading misinformation, insulting/derogatory comments, and personal or political attacks
+- Public or private harassment
+- Publishing others' private information, such as a physical or electronic address, without explicit permission
+- Abuse of the reporting process to intentionally harass or exclude others
+- Advocating for, or encouraging, any of the above behavior
+- Other conduct which could reasonably be considered inappropriate in a professional setting
+
+## Our Responsibilities
+
+Project maintainers are responsible for clarifying the standards of acceptable behavior and are expected to take appropriate and fair corrective action in response to any instances of unacceptable behavior.
+
+Project maintainers have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, or to ban temporarily or permanently any contributor for other behaviors that they deem inappropriate, threatening, offensive, or harmful.
+
+## Scope
+
+This Code of Conduct applies both within project spaces and in public spaces when an individual is representing the project or its community. Examples of representing a project or community include using an official project e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Representation of a project may be further defined and clarified by project maintainers.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant](https://www.contributor-covenant.org/), [version 1.4](https://www.contributor-covenant.org/version/1/4/code-of-conduct.html).

CONTRIBUTING.md

@@ -0,0 +1,10 @@
+# Contributing to D-Net Hadoop
+
+:+1::tada: First off, thanks for taking the time to contribute! :tada::+1:
+
+This project and everyone participating in it is governed by our [Code of Conduct](CODE_OF_CONDUCT.md). By participating, you are expected to uphold this code. Please report unacceptable behavior to [dnet-team@isti.cnr.it](mailto:dnet-team@isti.cnr.it).
+
+The following is a set of guidelines for contributing to this project and its packages. These are mostly guidelines, not rules, which applies to this project as a while, including all its sub-modules. 
+Use your best judgment, and feel free to propose changes to this document in a pull request.
+
+All contributions are welcome, all contributions will be considered to be contributed under the [project license](LICENSE.md).

README.md

@@ -2,6 +2,11 @@
 
 Dnet-hadoop is the project that defined all the [OOZIE workflows](https://oozie.apache.org/) for the OpenAIRE Graph construction, processing, provisioning.
 
+This project adheres to the Contributor Covenant [code of conduct](CODE_OF_CONDUCT.md).
+By participating, you are expected to uphold this code. Please report unacceptable behavior to [dnet-team@isti.cnr.it](mailto:dnet-team@isti.cnr.it).
+
+This project is licensed under the [AGPL v3 or later version](#LICENSE.md).
+
 How to build, package and run oozie workflows
 ====================
 

dhp-common/src/main/java/eu/dnetlib/dhp/common/api/context/CategorySummary.java

@@ -0,0 +1,39 @@
+
+package eu.dnetlib.dhp.common.api.context;
+
+public class CategorySummary {
+
+	private String id;
+
+	private String label;
+
+	private boolean hasConcept;
+
+	public String getId() {
+		return id;
+	}
+
+	public String getLabel() {
+		return label;
+	}
+
+	public boolean isHasConcept() {
+		return hasConcept;
+	}
+
+	public CategorySummary setId(final String id) {
+		this.id = id;
+		return this;
+	}
+
+	public CategorySummary setLabel(final String label) {
+		this.label = label;
+		return this;
+	}
+
+	public CategorySummary setHasConcept(final boolean hasConcept) {
+		this.hasConcept = hasConcept;
+		return this;
+	}
+
+}

dhp-common/src/main/java/eu/dnetlib/dhp/common/api/context/CategorySummaryList.java

@@ -0,0 +1,7 @@
+
+package eu.dnetlib.dhp.common.api.context;
+
+import java.util.ArrayList;
+
+public class CategorySummaryList extends ArrayList<CategorySummary> {
+}

dhp-common/src/main/java/eu/dnetlib/dhp/common/api/context/ConceptSummary.java

@@ -0,0 +1,52 @@
+
+package eu.dnetlib.dhp.common.api.context;
+
+import java.util.List;
+
+public class ConceptSummary {
+
+	private String id;
+
+	private String label;
+
+	public boolean hasSubConcept;
+
+	private List<ConceptSummary> concepts;
+
+	public String getId() {
+		return id;
+	}
+
+	public String getLabel() {
+		return label;
+	}
+
+	public List<ConceptSummary> getConcepts() {
+		return concepts;
+	}
+
+	public ConceptSummary setId(final String id) {
+		this.id = id;
+		return this;
+	}
+
+	public ConceptSummary setLabel(final String label) {
+		this.label = label;
+		return this;
+	}
+
+	public boolean isHasSubConcept() {
+		return hasSubConcept;
+	}
+
+	public ConceptSummary setHasSubConcept(final boolean hasSubConcept) {
+		this.hasSubConcept = hasSubConcept;
+		return this;
+	}
+
+	public ConceptSummary setConcept(final List<ConceptSummary> concepts) {
+		this.concepts = concepts;
+		return this;
+	}
+
+}

dhp-common/src/main/java/eu/dnetlib/dhp/common/api/context/ConceptSummaryList.java

@@ -0,0 +1,7 @@
+
+package eu.dnetlib.dhp.common.api.context;
+
+import java.util.ArrayList;
+
+public class ConceptSummaryList extends ArrayList<ConceptSummary> {
+}

dhp-common/src/main/java/eu/dnetlib/dhp/common/api/context/ContextSummary.java

@@ -0,0 +1,50 @@
+
+package eu.dnetlib.dhp.common.api.context;
+
+public class ContextSummary {
+
+	private String id;
+
+	private String label;
+
+	private String type;
+
+	private String status;
+
+	public String getId() {
+		return id;
+	}
+
+	public String getLabel() {
+		return label;
+	}
+
+	public String getType() {
+		return type;
+	}
+
+	public String getStatus() {
+		return status;
+	}
+
+	public ContextSummary setId(final String id) {
+		this.id = id;
+		return this;
+	}
+
+	public ContextSummary setLabel(final String label) {
+		this.label = label;
+		return this;
+	}
+
+	public ContextSummary setType(final String type) {
+		this.type = type;
+		return this;
+	}
+
+	public ContextSummary setStatus(final String status) {
+		this.status = status;
+		return this;
+	}
+
+}

dhp-common/src/main/java/eu/dnetlib/dhp/common/api/context/ContextSummaryList.java

@@ -0,0 +1,7 @@
+
+package eu.dnetlib.dhp.common.api.context;
+
+import java.util.ArrayList;
+
+public class ContextSummaryList extends ArrayList<ContextSummary> {
+}

dhp-common/src/main/java/eu/dnetlib/dhp/common/collection/HttpConnector2.java

@@ -8,10 +8,13 @@ import java.io.InputStream;
 import java.net.*;
 import java.util.List;
 import java.util.Map;
+import java.util.concurrent.TimeUnit;
 
 import org.apache.commons.io.IOUtils;
 import org.apache.commons.lang3.math.NumberUtils;
+import org.apache.commons.lang3.time.DateUtils;
 import org.apache.http.HttpHeaders;
+import org.joda.time.Instant;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
@@ -94,14 +97,16 @@ public class HttpConnector2 {
 			throw new CollectorException(msg);
 		}
 
-		log.info("Request attempt {} [{}]", retryNumber, requestUrl);
-
 		InputStream input = null;
 
+		long start = System.currentTimeMillis();
 		try {
 			if (getClientParams().getRequestDelay() > 0) {
 				backoffAndSleep(getClientParams().getRequestDelay());
 			}
+
+			log.info("Request attempt {} [{}]", retryNumber, requestUrl);
+
 			final HttpURLConnection urlConn = (HttpURLConnection) new URL(requestUrl).openConnection();
 			urlConn.setInstanceFollowRedirects(false);
 			urlConn.setReadTimeout(getClientParams().getReadTimeOut() * 1000);
@@ -115,9 +120,8 @@ public class HttpConnector2 {
 					urlConn.addRequestProperty(headerEntry.getKey(), headerEntry.getValue());
 				}
 			}
-			if (log.isDebugEnabled()) {
-				logHeaderFields(urlConn);
-			}
+
+			logHeaderFields(urlConn);
 
 			int retryAfter = obtainRetryAfter(urlConn.getHeaderFields());
 			String rateLimit = urlConn.getHeaderField(Constants.HTTPHEADER_IETF_DRAFT_RATELIMIT_LIMIT);
@@ -132,9 +136,7 @@ public class HttpConnector2 {
 			}
 
 			if (is2xx(urlConn.getResponseCode())) {
-				input = urlConn.getInputStream();
-				responseType = urlConn.getContentType();
-				return input;
+				return getInputStream(urlConn, start);
 			}
 			if (is3xx(urlConn.getResponseCode())) {
 				// REDIRECTS
@@ -144,6 +146,7 @@ public class HttpConnector2 {
 					.put(
 						REPORT_PREFIX + urlConn.getResponseCode(),
 						String.format("Moved to: %s", newUrl));
+				logRequestTime(start);
 				urlConn.disconnect();
 				if (retryAfter > 0) {
 					backoffAndSleep(retryAfter);
@@ -159,26 +162,50 @@ public class HttpConnector2 {
 						if (retryAfter > 0) {
 							log
 								.warn(
-									"{} - waiting and repeating request after suggested retry-after {} sec.",
-									requestUrl, retryAfter);
+									"waiting and repeating request after suggested retry-after {} sec for URL {}",
+									retryAfter, requestUrl);
 							backoffAndSleep(retryAfter * 1000);
 						} else {
 							log
 								.warn(
-									"{} - waiting and repeating request after default delay of {} sec.",
-									requestUrl, getClientParams().getRetryDelay());
-							backoffAndSleep(retryNumber * getClientParams().getRetryDelay() * 1000);
+									"waiting and repeating request after default delay of {} sec for URL {}",
+									getClientParams().getRetryDelay(), requestUrl);
+							backoffAndSleep(retryNumber * getClientParams().getRetryDelay());
 						}
 						report.put(REPORT_PREFIX + urlConn.getResponseCode(), requestUrl);
+
+						logRequestTime(start);
+
 						urlConn.disconnect();
+
 						return attemptDownload(requestUrl, retryNumber + 1, report);
+					case 422: // UNPROCESSABLE ENTITY
+						report.put(REPORT_PREFIX + urlConn.getResponseCode(), requestUrl);
+						log.warn("waiting and repeating request after 10 sec for URL {}", requestUrl);
+						backoffAndSleep(10000);
+						urlConn.disconnect();
+						logRequestTime(start);
+						try {
+							return getInputStream(urlConn, start);
+						} catch (IOException e) {
+							log
+								.error(
+									"server returned 422 and got IOException accessing the response body from URL {}",
+									requestUrl);
+							log.error("IOException:", e);
+							return attemptDownload(requestUrl, retryNumber + 1, report);
+						}
 					default:
+						log.error("gor error {} from URL: {}", urlConn.getResponseCode(), urlConn.getURL());
+						log.error("response message: {}", urlConn.getResponseMessage());
 						report
 							.put(
 								REPORT_PREFIX + urlConn.getResponseCode(),
 								String
 									.format(
 										"%s Error: %s", requestUrl, urlConn.getResponseMessage()));
+						logRequestTime(start);
+						urlConn.disconnect();
 						throw new CollectorException(urlConn.getResponseCode() + " error " + report);
 				}
 			}
@@ -199,13 +226,27 @@ public class HttpConnector2 {
 		}
 	}
 
+	private InputStream getInputStream(HttpURLConnection urlConn, long start) throws IOException {
+		InputStream input = urlConn.getInputStream();
+		responseType = urlConn.getContentType();
+		logRequestTime(start);
+		return input;
+	}
+
+	private static void logRequestTime(long start) {
+		log
+			.info(
+				"request time elapsed: {}sec",
+				TimeUnit.MILLISECONDS.toSeconds(System.currentTimeMillis() - start));
+	}
+
 	private void logHeaderFields(final HttpURLConnection urlConn) throws IOException {
-		log.debug("StatusCode: {}", urlConn.getResponseMessage());
+		log.info("Response: {} - {}", urlConn.getResponseCode(), urlConn.getResponseMessage());
 
 		for (Map.Entry<String, List<String>> e : urlConn.getHeaderFields().entrySet()) {
 			if (e.getKey() != null) {
 				for (String v : e.getValue()) {
-					log.debug("  key: {} - value: {}", e.getKey(), v);
+					log.info("  key: {} - value: {}", e.getKey(), v);
 				}
 			}
 		}
@@ -225,7 +266,7 @@ public class HttpConnector2 {
 		for (String key : headerMap.keySet()) {
 			if ((key != null) && key.equalsIgnoreCase(HttpHeaders.RETRY_AFTER) && (!headerMap.get(key).isEmpty())
 				&& NumberUtils.isCreatable(headerMap.get(key).get(0))) {
-				return Integer.parseInt(headerMap.get(key).get(0)) + 10;
+				return Integer.parseInt(headerMap.get(key).get(0));
 			}
 		}
 		return -1;

dhp-common/src/main/java/eu/dnetlib/dhp/oozie/RunSQLSparkJob.java

@@ -0,0 +1,77 @@
+
+package eu.dnetlib.dhp.oozie;
+
+import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
+
+import java.net.URL;
+import java.nio.charset.StandardCharsets;
+import java.util.HashMap;
+import java.util.Map;
+import java.util.Optional;
+
+import org.apache.commons.lang3.time.DurationFormatUtils;
+import org.apache.commons.text.StringSubstitutor;
+import org.apache.spark.SparkConf;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import com.google.common.io.Resources;
+
+import eu.dnetlib.dhp.application.ArgumentApplicationParser;
+
+public class RunSQLSparkJob {
+	private static final Logger log = LoggerFactory.getLogger(RunSQLSparkJob.class);
+
+	private final ArgumentApplicationParser parser;
+
+	public RunSQLSparkJob(ArgumentApplicationParser parser) {
+		this.parser = parser;
+	}
+
+	public static void main(String[] args) throws Exception {
+
+		Map<String, String> params = new HashMap<>();
+		for (int i = 0; i < args.length - 1; i++) {
+			if (args[i].startsWith("--")) {
+				params.put(args[i].substring(2), args[++i]);
+			}
+		}
+
+		/*
+		 * String jsonConfiguration = IOUtils .toString( Objects .requireNonNull( RunSQLSparkJob.class
+		 * .getResourceAsStream( "/eu/dnetlib/dhp/oozie/run_sql_parameters.json"))); final ArgumentApplicationParser
+		 * parser = new ArgumentApplicationParser(jsonConfiguration); parser.parseArgument(args);
+		 */
+
+		Boolean isSparkSessionManaged = Optional
+			.ofNullable(params.get("isSparkSessionManaged"))
+			.map(Boolean::valueOf)
+			.orElse(Boolean.TRUE);
+		log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
+
+		URL url = com.google.common.io.Resources.getResource(params.get("sql"));
+		String raw_sql = Resources.toString(url, StandardCharsets.UTF_8);
+
+		String sql = StringSubstitutor.replace(raw_sql, params);
+		log.info("sql: {}", sql);
+
+		SparkConf conf = new SparkConf();
+		conf.set("hive.metastore.uris", params.get("hiveMetastoreUris"));
+
+		runWithSparkHiveSession(
+			conf,
+			isSparkSessionManaged,
+			spark -> {
+				for (String statement : sql.split(";\\s*/\\*\\s*EOS\\s*\\*/\\s*")) {
+					log.info("executing: {}", statement);
+					long startTime = System.currentTimeMillis();
+					spark.sql(statement).show();
+					log
+						.info(
+							"executed in {}",
+							DurationFormatUtils.formatDuration(System.currentTimeMillis() - startTime, "HH:mm:ss.S"));
+				}
+			});
+	}
+
+}

dhp-common/src/main/java/eu/dnetlib/dhp/schema/oaf/utils/GraphCleaningFunctions.java

@@ -312,7 +312,8 @@ public class GraphCleaningFunctions extends CleaningFunctions {
 		}
 
 		if (value instanceof Datasource) {
-			// nothing to evaluate here
+			final Datasource d = (Datasource) value;
+			return Objects.nonNull(d.getOfficialname()) && StringUtils.isNotBlank(d.getOfficialname().getValue());
 		} else if (value instanceof Project) {
 			final Project p = (Project) value;
 			return Objects.nonNull(p.getCode()) && StringUtils.isNotBlank(p.getCode().getValue());

dhp-common/src/main/resources/eu/dnetlib/dhp/oozie/run_sql_parameters.json

@@ -0,0 +1,20 @@
+[
+  {
+    "paramName": "issm",
+    "paramLongName": "isSparkSessionManaged",
+    "paramDescription": "when true will stop SparkSession after job execution",
+    "paramRequired": false
+  },
+  {
+    "paramName": "hmu",
+    "paramLongName": "hiveMetastoreUris",
+    "paramDescription": "the hive metastore uris",
+    "paramRequired": true
+  },
+  {
+    "paramName": "sql",
+    "paramLongName": "sql",
+    "paramDescription": "sql script to execute",
+    "paramRequired": true
+  }
+]

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/AbstractClusteringFunction.java

@@ -14,9 +14,9 @@ import eu.dnetlib.pace.config.Config;
 
 public abstract class AbstractClusteringFunction extends AbstractPaceFunctions implements ClusteringFunction {
 
-	protected Map<String, Integer> params;
+	protected Map<String, Object> params;
 
-	public AbstractClusteringFunction(final Map<String, Integer> params) {
+	public AbstractClusteringFunction(final Map<String, Object> params) {
 		this.params = params;
 	}
 
@@ -27,7 +27,7 @@ public abstract class AbstractClusteringFunction extends AbstractPaceFunctions i
 		return fields
 			.stream()
 			.filter(f -> !f.isEmpty())
-			.map(this::normalize)
+			.map(s -> normalize(s))
 			.map(s -> filterAllStopWords(s))
 			.map(s -> doApply(conf, s))
 			.map(c -> filterBlacklisted(c, ngramBlacklist))
@@ -36,11 +36,24 @@ public abstract class AbstractClusteringFunction extends AbstractPaceFunctions i
 			.collect(Collectors.toCollection(HashSet::new));
 	}
 
-	public Map<String, Integer> getParams() {
+	public Map<String, Object> getParams() {
 		return params;
 	}
 
 	protected Integer param(String name) {
-		return params.get(name);
+		Object val = params.get(name);
+		if (val == null)
+			return null;
+		if (val instanceof Number) {
+			return ((Number) val).intValue();
+		}
+		return Integer.parseInt(val.toString());
+	}
+
+	protected int paramOrDefault(String name, int i) {
+		Integer res = param(name);
+		if (res == null)
+			res = i;
+		return res;
 	}
 }

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/Acronyms.java

@@ -13,7 +13,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("acronyms")
 public class Acronyms extends AbstractClusteringFunction {
 
-	public Acronyms(Map<String, Integer> params) {
+	public Acronyms(Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/ClusteringFunction.java

@@ -11,6 +11,6 @@ public interface ClusteringFunction {
 
 	public Collection<String> apply(Config config, List<String> fields);
 
-	public Map<String, Integer> getParams();
+	public Map<String, Object> getParams();
 
 }

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/ImmutableFieldValue.java

@@ -12,7 +12,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("immutablefieldvalue")
 public class ImmutableFieldValue extends AbstractClusteringFunction {
 
-	public ImmutableFieldValue(final Map<String, Integer> params) {
+	public ImmutableFieldValue(final Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/JSONListClustering.java

@@ -0,0 +1,69 @@
+
+package eu.dnetlib.pace.clustering;
+
+import java.util.Collection;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.stream.Collectors;
+
+import org.apache.commons.lang3.StringUtils;
+
+import com.jayway.jsonpath.Configuration;
+import com.jayway.jsonpath.DocumentContext;
+import com.jayway.jsonpath.JsonPath;
+import com.jayway.jsonpath.Option;
+
+import eu.dnetlib.pace.common.AbstractPaceFunctions;
+import eu.dnetlib.pace.config.Config;
+import eu.dnetlib.pace.util.MapDocumentUtil;
+
+@ClusteringClass("jsonlistclustering")
+public class JSONListClustering extends AbstractPaceFunctions implements ClusteringFunction {
+
+	private Map<String, Object> params;
+
+	public JSONListClustering(Map<String, Object> params) {
+		this.params = params;
+	}
+
+	@Override
+	public Map<String, Object> getParams() {
+		return params;
+	}
+
+	@Override
+	public Collection<String> apply(Config conf, List<String> fields) {
+		return fields
+			.stream()
+			.filter(f -> !f.isEmpty())
+			.map(s -> doApply(conf, s))
+			.filter(StringUtils::isNotBlank)
+			.collect(Collectors.toCollection(HashSet::new));
+	}
+
+	private String doApply(Config conf, String json) {
+		StringBuilder st = new StringBuilder(); // to build the string used for comparisons basing on the jpath into
+		// parameters
+		final DocumentContext documentContext = JsonPath
+			.using(Configuration.defaultConfiguration().addOptions(Option.SUPPRESS_EXCEPTIONS))
+			.parse(json);
+
+		// for each path in the param list
+		for (String key : params.keySet().stream().filter(k -> k.contains("jpath")).collect(Collectors.toList())) {
+			String path = params.get(key).toString();
+			String value = MapDocumentUtil.getJPathString(path, documentContext);
+			if (value == null || value.isEmpty())
+				value = "";
+			st.append(value);
+			st.append(" ");
+		}
+
+		st.setLength(st.length() - 1);
+
+		if (StringUtils.isBlank(st)) {
+			return "1";
+		}
+		return st.toString();
+	}
+}

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/KeywordsClustering.java

@@ -11,7 +11,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("keywordsclustering")
 public class KeywordsClustering extends AbstractClusteringFunction {
 
-	public KeywordsClustering(Map<String, Integer> params) {
+	public KeywordsClustering(Map<String, Object> params) {
 		super(params);
 	}
 
@@ -19,8 +19,8 @@ public class KeywordsClustering extends AbstractClusteringFunction {
 	protected Collection<String> doApply(final Config conf, String s) {
 
 		// takes city codes and keywords codes without duplicates
-		Set<String> keywords = getKeywords(s, conf.translationMap(), params.getOrDefault("windowSize", 4));
-		Set<String> cities = getCities(s, params.getOrDefault("windowSize", 4));
+		Set<String> keywords = getKeywords(s, conf.translationMap(), paramOrDefault("windowSize", 4));
+		Set<String> cities = getCities(s, paramOrDefault("windowSize", 4));
 
 		// list of combination to return as result
 		final Collection<String> combinations = new LinkedHashSet<String>();
@@ -28,7 +28,7 @@ public class KeywordsClustering extends AbstractClusteringFunction {
 		for (String keyword : keywordsToCodes(keywords, conf.translationMap())) {
 			for (String city : citiesToCodes(cities)) {
 				combinations.add(keyword + "-" + city);
-				if (combinations.size() >= params.getOrDefault("max", 2)) {
+				if (combinations.size() >= paramOrDefault("max", 2)) {
 					return combinations;
 				}
 			}
@@ -42,8 +42,8 @@ public class KeywordsClustering extends AbstractClusteringFunction {
 		return fields
 			.stream()
 			.filter(f -> !f.isEmpty())
-			.map(this::cleanup)
-			.map(this::normalize)
+			.map(KeywordsClustering::cleanup)
+			.map(KeywordsClustering::normalize)
 			.map(s -> filterAllStopWords(s))
 			.map(s -> doApply(conf, s))
 			.map(c -> filterBlacklisted(c, ngramBlacklist))

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/LastNameFirstInitial.java

@@ -16,7 +16,7 @@ public class LastNameFirstInitial extends AbstractClusteringFunction {
 
 	private boolean DEFAULT_AGGRESSIVE = true;
 
-	public LastNameFirstInitial(final Map<String, Integer> params) {
+	public LastNameFirstInitial(final Map<String, Object> params) {
 		super(params);
 	}
 
@@ -25,7 +25,7 @@ public class LastNameFirstInitial extends AbstractClusteringFunction {
 		return fields
 			.stream()
 			.filter(f -> !f.isEmpty())
-			.map(this::normalize)
+			.map(LastNameFirstInitial::normalize)
 			.map(s -> doApply(conf, s))
 			.map(c -> filterBlacklisted(c, ngramBlacklist))
 			.flatMap(c -> c.stream())
@@ -33,8 +33,7 @@ public class LastNameFirstInitial extends AbstractClusteringFunction {
 			.collect(Collectors.toCollection(HashSet::new));
 	}
 
-	@Override
-	protected String normalize(final String s) {
+	public static String normalize(final String s) {
 		return fixAliases(transliterate(nfd(unicodeNormalization(s))))
 			// do not compact the regexes in a single expression, would cause StackOverflowError in case of large input
 			// strings

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/LowercaseClustering.java

@@ -15,7 +15,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("lowercase")
 public class LowercaseClustering extends AbstractClusteringFunction {
 
-	public LowercaseClustering(final Map<String, Integer> params) {
+	public LowercaseClustering(final Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/NgramPairs.java

@@ -12,11 +12,11 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("ngrampairs")
 public class NgramPairs extends Ngrams {
 
-	public NgramPairs(Map<String, Integer> params) {
+	public NgramPairs(Map<String, Object> params) {
 		super(params, false);
 	}
 
-	public NgramPairs(Map<String, Integer> params, boolean sorted) {
+	public NgramPairs(Map<String, Object> params, boolean sorted) {
 		super(params, sorted);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/Ngrams.java

@@ -10,11 +10,11 @@ public class Ngrams extends AbstractClusteringFunction {
 
 	private final boolean sorted;
 
-	public Ngrams(Map<String, Integer> params) {
+	public Ngrams(Map<String, Object> params) {
 		this(params, false);
 	}
 
-	public Ngrams(Map<String, Integer> params, boolean sorted) {
+	public Ngrams(Map<String, Object> params, boolean sorted) {
 		super(params);
 		this.sorted = sorted;
 	}

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/NumAuthorsTitleSuffixPrefixChain.java

@@ -0,0 +1,113 @@
+
+package eu.dnetlib.pace.clustering;
+
+import java.util.*;
+import java.util.stream.Collectors;
+import java.util.stream.StreamSupport;
+
+import com.google.common.base.Splitter;
+import com.google.common.collect.Sets;
+
+import eu.dnetlib.pace.config.Config;
+
+@ClusteringClass("numAuthorsTitleSuffixPrefixChain")
+public class NumAuthorsTitleSuffixPrefixChain extends AbstractClusteringFunction {
+
+	public NumAuthorsTitleSuffixPrefixChain(Map<String, Object> params) {
+		super(params);
+	}
+
+	@Override
+	public Collection<String> apply(Config conf, List<String> fields) {
+
+		try {
+			int num_authors = Math.min(Integer.parseInt(fields.get(0)), 21); // SIZE threshold is 20, +1
+
+			if (num_authors > 0) {
+				return super.apply(conf, fields.subList(1, fields.size()))
+					.stream()
+					.map(s -> num_authors + "-" + s)
+					.collect(Collectors.toList());
+			}
+		} catch (NumberFormatException e) {
+			// missing or null authors array
+		}
+
+		return Collections.emptyList();
+	}
+
+	@Override
+	protected Collection<String> doApply(Config conf, String s) {
+		return suffixPrefixChain(cleanup(s), param("mod"));
+	}
+
+	private Collection<String> suffixPrefixChain(String s, int mod) {
+		// create the list of words from the string (remove short words)
+		List<String> wordsList = Arrays
+			.stream(s.split(" "))
+			.filter(si -> si.length() > 3)
+			.collect(Collectors.toList());
+
+		final int words = wordsList.size();
+		final int letters = s.length();
+
+		// create the prefix: number of words + number of letters/mod
+		String prefix = words / mod + "-";
+
+		return doSuffixPrefixChain(wordsList, prefix);
+
+	}
+
+	private Collection<String> doSuffixPrefixChain(List<String> wordsList, String prefix) {
+
+		Set<String> set = Sets.newLinkedHashSet();
+		switch (wordsList.size()) {
+			case 0:
+				break;
+			case 1:
+				set.add(wordsList.get(0));
+				break;
+			case 2:
+				set
+					.add(
+						prefix +
+							suffix(wordsList.get(0), 3) +
+							prefix(wordsList.get(1), 3));
+
+				set
+					.add(
+						prefix +
+							prefix(wordsList.get(0), 3) +
+							suffix(wordsList.get(1), 3));
+
+				break;
+			default:
+				set
+					.add(
+						prefix +
+							suffix(wordsList.get(0), 3) +
+							prefix(wordsList.get(1), 3) +
+							suffix(wordsList.get(2), 3));
+
+				set
+					.add(
+						prefix +
+							prefix(wordsList.get(0), 3) +
+							suffix(wordsList.get(1), 3) +
+							prefix(wordsList.get(2), 3));
+				break;
+		}
+
+		return set;
+
+	}
+
+	private String suffix(String s, int len) {
+		return s.substring(s.length() - len);
+	}
+
+	private String prefix(String s, int len) {
+		return s.substring(0, len);
+	}
+
+}

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/PersonClustering.java

@@ -17,11 +17,11 @@ import eu.dnetlib.pace.model.Person;
 @ClusteringClass("personClustering")
 public class PersonClustering extends AbstractPaceFunctions implements ClusteringFunction {
 
-	private Map<String, Integer> params;
+	private Map<String, Object> params;
 
 	private static final int MAX_TOKENS = 5;
 
-	public PersonClustering(final Map<String, Integer> params) {
+	public PersonClustering(final Map<String, Object> params) {
 		this.params = params;
 	}
 
@@ -77,7 +77,7 @@ public class PersonClustering extends AbstractPaceFunctions implements Clusterin
 //	}
 
 	@Override
-	public Map<String, Integer> getParams() {
+	public Map<String, Object> getParams() {
 		return params;
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/PersonHash.java

@@ -15,7 +15,7 @@ public class PersonHash extends AbstractClusteringFunction {
 
 	private boolean DEFAULT_AGGRESSIVE = false;
 
-	public PersonHash(final Map<String, Integer> params) {
+	public PersonHash(final Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/RandomClusteringFunction.java

@@ -8,7 +8,7 @@ import eu.dnetlib.pace.config.Config;
 
 public class RandomClusteringFunction extends AbstractClusteringFunction {
 
-	public RandomClusteringFunction(Map<String, Integer> params) {
+	public RandomClusteringFunction(Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/SortedNgramPairs.java

@@ -1,7 +1,10 @@
 
 package eu.dnetlib.pace.clustering;
 
-import java.util.*;
+import java.util.Collection;
+import java.util.Collections;
+import java.util.List;
+import java.util.Map;
 
 import com.google.common.base.Joiner;
 import com.google.common.base.Splitter;
@@ -12,7 +15,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("sortedngrampairs")
 public class SortedNgramPairs extends NgramPairs {
 
-	public SortedNgramPairs(Map<String, Integer> params) {
+	public SortedNgramPairs(Map<String, Object> params) {
 		super(params, false);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/SpaceTrimmingFieldValue.java

@@ -15,7 +15,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("spacetrimmingfieldvalue")
 public class SpaceTrimmingFieldValue extends AbstractClusteringFunction {
 
-	public SpaceTrimmingFieldValue(final Map<String, Integer> params) {
+	public SpaceTrimmingFieldValue(final Map<String, Object> params) {
 		super(params);
 	}
 
@@ -25,7 +25,7 @@ public class SpaceTrimmingFieldValue extends AbstractClusteringFunction {
 
 		res
 			.add(
-				StringUtils.isBlank(s) ? RandomStringUtils.random(getParams().get("randomLength"))
+				StringUtils.isBlank(s) ? RandomStringUtils.random(param("randomLength"))
 					: s.toLowerCase().replaceAll("\\s+", ""));
 
 		return res;

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/SuffixPrefix.java

@@ -12,7 +12,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("suffixprefix")
 public class SuffixPrefix extends AbstractClusteringFunction {
 
-	public SuffixPrefix(Map<String, Integer> params) {
+	public SuffixPrefix(Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/UrlClustering.java

@@ -15,12 +15,17 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("urlclustering")
 public class UrlClustering extends AbstractPaceFunctions implements ClusteringFunction {
 
-	protected Map<String, Integer> params;
+	protected Map<String, Object> params;
 
-	public UrlClustering(final Map<String, Integer> params) {
+	public UrlClustering(final Map<String, Object> params) {
 		this.params = params;
 	}
 
+	@Override
+	public Map<String, Object> getParams() {
+		return params;
+	}
+
 	@Override
 	public Collection<String> apply(final Config conf, List<String> fields) {
 		try {
@@ -35,11 +40,6 @@ public class UrlClustering extends AbstractPaceFunctions implements ClusteringFu
 		}
 	}
 
-	@Override
-	public Map<String, Integer> getParams() {
-		return null;
-	}
-
 	private URL asUrl(String value) {
 		try {
 			return new URL(value);

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/WordsStatsSuffixPrefixChain.java

@@ -11,7 +11,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("wordsStatsSuffixPrefixChain")
 public class WordsStatsSuffixPrefixChain extends AbstractClusteringFunction {
 
-	public WordsStatsSuffixPrefixChain(Map<String, Integer> params) {
+	public WordsStatsSuffixPrefixChain(Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/clustering/WordsSuffixPrefix.java

@@ -12,7 +12,7 @@ import eu.dnetlib.pace.config.Config;
 @ClusteringClass("wordssuffixprefix")
 public class WordsSuffixPrefix extends AbstractClusteringFunction {
 
-	public WordsSuffixPrefix(Map<String, Integer> params) {
+	public WordsSuffixPrefix(Map<String, Object> params) {
 		super(params);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/common/AbstractPaceFunctions.java

@@ -16,7 +16,6 @@ import org.apache.commons.lang3.StringUtils;
 import com.google.common.base.Joiner;
 import com.google.common.base.Splitter;
 import com.google.common.collect.Iterables;
-import com.google.common.collect.Lists;
 import com.google.common.collect.Sets;
 import com.ibm.icu.text.Transliterator;
 
@@ -27,7 +26,7 @@ import eu.dnetlib.pace.clustering.NGramUtils;
  *
  * @author claudio
  */
-public abstract class AbstractPaceFunctions {
+public class AbstractPaceFunctions {
 
 	// city map to be used when translating the city names into codes
 	private static Map<String, String> cityMap = AbstractPaceFunctions
@@ -62,11 +61,14 @@ public abstract class AbstractPaceFunctions {
 
 	private static Pattern hexUnicodePattern = Pattern.compile("\\\\u(\\p{XDigit}{4})");
 
-	protected String concat(final List<String> l) {
+	private static Pattern romanNumberPattern = Pattern
+		.compile("^M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$");
+
+	protected static String concat(final List<String> l) {
 		return Joiner.on(" ").skipNulls().join(l);
 	}
 
-	protected String cleanup(final String s) {
+	public static String cleanup(final String s) {
 		final String s1 = HTML_REGEX.matcher(s).replaceAll("");
 		final String s2 = unicodeNormalization(s1.toLowerCase());
 		final String s3 = nfd(s2);
@@ -82,7 +84,7 @@ public abstract class AbstractPaceFunctions {
 		return s12;
 	}
 
-	protected String fixXML(final String a) {
+	protected static String fixXML(final String a) {
 
 		return a
 			.replaceAll("&ndash;", " ")
@@ -91,7 +93,7 @@ public abstract class AbstractPaceFunctions {
 			.replaceAll("&minus;", " ");
 	}
 
-	protected boolean checkNumbers(final String a, final String b) {
+	protected static boolean checkNumbers(final String a, final String b) {
 		final String numbersA = getNumbers(a);
 		final String numbersB = getNumbers(b);
 		final String romansA = getRomans(a);
@@ -99,7 +101,7 @@ public abstract class AbstractPaceFunctions {
 		return !numbersA.equals(numbersB) || !romansA.equals(romansB);
 	}
 
-	protected String getRomans(final String s) {
+	protected static String getRomans(final String s) {
 		final StringBuilder sb = new StringBuilder();
 		for (final String t : s.split(" ")) {
 			sb.append(isRoman(t) ? t : "");
@@ -107,13 +109,12 @@ public abstract class AbstractPaceFunctions {
 		return sb.toString();
 	}
 
-	protected boolean isRoman(final String s) {
-		return s
-			.replaceAll("^M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$", "qwertyuiop")
-			.equals("qwertyuiop");
+	protected static boolean isRoman(final String s) {
+		Matcher m = romanNumberPattern.matcher(s);
+		return m.matches() && m.hitEnd();
 	}
 
-	protected String getNumbers(final String s) {
+	protected static String getNumbers(final String s) {
 		final StringBuilder sb = new StringBuilder();
 		for (final String t : s.split(" ")) {
 			sb.append(isNumber(t) ? t : "");
@@ -121,7 +122,7 @@ public abstract class AbstractPaceFunctions {
 		return sb.toString();
 	}
 
-	public boolean isNumber(String strNum) {
+	public static boolean isNumber(String strNum) {
 		if (strNum == null) {
 			return false;
 		}
@@ -147,7 +148,7 @@ public abstract class AbstractPaceFunctions {
 		}
 	}
 
-	protected String removeSymbols(final String s) {
+	protected static String removeSymbols(final String s) {
 		final StringBuilder sb = new StringBuilder();
 
 		s.chars().forEach(ch -> {
@@ -157,11 +158,11 @@ public abstract class AbstractPaceFunctions {
 		return sb.toString().replaceAll("\\s+", " ");
 	}
 
-	protected boolean notNull(final String s) {
+	protected static boolean notNull(final String s) {
 		return s != null;
 	}
 
-	protected String normalize(final String s) {
+	public static String normalize(final String s) {
 		return fixAliases(transliterate(nfd(unicodeNormalization(s))))
 			.toLowerCase()
 			// do not compact the regexes in a single expression, would cause StackOverflowError in case of large input
@@ -174,16 +175,16 @@ public abstract class AbstractPaceFunctions {
 			.trim();
 	}
 
-	public String nfd(final String s) {
+	public static String nfd(final String s) {
 		return Normalizer.normalize(s, Normalizer.Form.NFD);
 	}
 
-	public String utf8(final String s) {
+	public static String utf8(final String s) {
 		byte[] bytes = s.getBytes(StandardCharsets.UTF_8);
 		return new String(bytes, StandardCharsets.UTF_8);
 	}
 
-	public String unicodeNormalization(final String s) {
+	public static String unicodeNormalization(final String s) {
 
 		Matcher m = hexUnicodePattern.matcher(s);
 		StringBuffer buf = new StringBuffer(s.length());
@@ -195,7 +196,7 @@ public abstract class AbstractPaceFunctions {
 		return buf.toString();
 	}
 
-	protected String filterStopWords(final String s, final Set<String> stopwords) {
+	protected static String filterStopWords(final String s, final Set<String> stopwords) {
 		final StringTokenizer st = new StringTokenizer(s);
 		final StringBuilder sb = new StringBuilder();
 		while (st.hasMoreTokens()) {
@@ -208,7 +209,7 @@ public abstract class AbstractPaceFunctions {
 		return sb.toString().trim();
 	}
 
-	public String filterAllStopWords(String s) {
+	public static String filterAllStopWords(String s) {
 
 		s = filterStopWords(s, stopwords_en);
 		s = filterStopWords(s, stopwords_de);
@@ -221,7 +222,8 @@ public abstract class AbstractPaceFunctions {
 		return s;
 	}
 
-	protected Collection<String> filterBlacklisted(final Collection<String> set, final Set<String> ngramBlacklist) {
+	protected static Collection<String> filterBlacklisted(final Collection<String> set,
+		final Set<String> ngramBlacklist) {
 		final Set<String> newset = Sets.newLinkedHashSet();
 		for (final String s : set) {
 			if (!ngramBlacklist.contains(s)) {
@@ -268,7 +270,7 @@ public abstract class AbstractPaceFunctions {
 		return m;
 	}
 
-	public String removeKeywords(String s, Set<String> keywords) {
+	public static String removeKeywords(String s, Set<String> keywords) {
 
 		s = " " + s + " ";
 		for (String k : keywords) {
@@ -278,39 +280,39 @@ public abstract class AbstractPaceFunctions {
 		return s.trim();
 	}
 
-	public double commonElementsPercentage(Set<String> s1, Set<String> s2) {
+	public static double commonElementsPercentage(Set<String> s1, Set<String> s2) {
 
 		double longer = Math.max(s1.size(), s2.size());
 		return (double) s1.stream().filter(s2::contains).count() / longer;
 	}
 
 	// convert the set of keywords to codes
-	public Set<String> toCodes(Set<String> keywords, Map<String, String> translationMap) {
+	public static Set<String> toCodes(Set<String> keywords, Map<String, String> translationMap) {
 		return keywords.stream().map(s -> translationMap.get(s)).collect(Collectors.toSet());
 	}
 
-	public Set<String> keywordsToCodes(Set<String> keywords, Map<String, String> translationMap) {
+	public static Set<String> keywordsToCodes(Set<String> keywords, Map<String, String> translationMap) {
 		return toCodes(keywords, translationMap);
 	}
 
-	public Set<String> citiesToCodes(Set<String> keywords) {
+	public static Set<String> citiesToCodes(Set<String> keywords) {
 		return toCodes(keywords, cityMap);
 	}
 
-	protected String firstLC(final String s) {
+	protected static String firstLC(final String s) {
 		return StringUtils.substring(s, 0, 1).toLowerCase();
 	}
 
-	protected Iterable<String> tokens(final String s, final int maxTokens) {
+	protected static Iterable<String> tokens(final String s, final int maxTokens) {
 		return Iterables.limit(Splitter.on(" ").omitEmptyStrings().trimResults().split(s), maxTokens);
 	}
 
-	public String normalizePid(String pid) {
+	public static String normalizePid(String pid) {
 		return DOI_PREFIX.matcher(pid.toLowerCase()).replaceAll("");
 	}
 
 	// get the list of keywords into the input string
-	public Set<String> getKeywords(String s1, Map<String, String> translationMap, int windowSize) {
+	public static Set<String> getKeywords(String s1, Map<String, String> translationMap, int windowSize) {
 
 		String s = s1;
 
@@ -340,7 +342,7 @@ public abstract class AbstractPaceFunctions {
 		return codes;
 	}
 
-	public Set<String> getCities(String s1, int windowSize) {
+	public static Set<String> getCities(String s1, int windowSize) {
 		return getKeywords(s1, cityMap, windowSize);
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/model/ClusteringDef.java

@@ -18,7 +18,7 @@ public class ClusteringDef implements Serializable {
 
 	private List<String> fields;
 
-	private Map<String, Integer> params;
+	private Map<String, Object> params;
 
 	public ClusteringDef() {
 	}
@@ -43,11 +43,11 @@ public class ClusteringDef implements Serializable {
 		this.fields = fields;
 	}
 
-	public Map<String, Integer> getParams() {
+	public Map<String, Object> getParams() {
 		return params;
 	}
 
-	public void setParams(final Map<String, Integer> params) {
+	public void setParams(final Map<String, Object> params) {
 		this.params = params;
 	}
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/model/FieldDef.java

@@ -2,6 +2,7 @@
 package eu.dnetlib.pace.model;
 
 import java.io.Serializable;
+import java.util.HashSet;
 import java.util.List;
 
 import com.fasterxml.jackson.core.JsonProcessingException;
@@ -36,6 +37,16 @@ public class FieldDef implements Serializable {
 	 */
 	private int length = -1;
 
+	private HashSet<String> filter;
+
+	private boolean sorted;
+
+	public boolean isSorted() {
+		return sorted;
+	}
+
+	private String clean;
+
 	public FieldDef() {
 	}
 
@@ -91,6 +102,30 @@ public class FieldDef implements Serializable {
 		this.path = path;
 	}
 
+	public HashSet<String> getFilter() {
+		return filter;
+	}
+
+	public void setFilter(HashSet<String> filter) {
+		this.filter = filter;
+	}
+
+	public boolean getSorted() {
+		return sorted;
+	}
+
+	public void setSorted(boolean sorted) {
+		this.sorted = sorted;
+	}
+
+	public String getClean() {
+		return clean;
+	}
+
+	public void setClean(String clean) {
+		this.clean = clean;
+	}
+
 	@Override
 	public String toString() {
 		try {

dhp-pace-core/src/main/java/eu/dnetlib/pace/model/SparkDeduper.scala

@@ -5,9 +5,9 @@ import eu.dnetlib.pace.util.{BlockProcessor, SparkReporter}
 import org.apache.spark.SparkContext
 import org.apache.spark.sql.catalyst.expressions.Literal
 import org.apache.spark.sql.expressions._
-import org.apache.spark.sql.functions.{col, lit, udf}
+import org.apache.spark.sql.functions.{col, desc, expr, lit, udf}
 import org.apache.spark.sql.types._
-import org.apache.spark.sql.{Column, Dataset, Row, functions}
+import org.apache.spark.sql.{Column, Dataset, Row, SaveMode, functions}
 
 import java.util.function.Predicate
 import java.util.stream.Collectors
@@ -80,6 +80,8 @@ case class SparkDeduper(conf: DedupConfig) extends Serializable {
         .withColumn("key", functions.explode(clusterValuesUDF(cd).apply(functions.array(inputColumns: _*))))
         // Add position column having the position of the row within the set of rows having the same key value ordered by the sorting value
         .withColumn("position", functions.row_number().over(Window.partitionBy("key").orderBy(col(model.orderingFieldName), col(model.identifierFieldName))))
+       // .withColumn("count", functions.max("position").over(Window.partitionBy("key").orderBy(col(model.orderingFieldName), col(model.identifierFieldName)).rowsBetween(Window.unboundedPreceding,Window.unboundedFollowing) ))
+       // .filter("count > 1")
 
       if (df_with_clustering_keys == null)
         df_with_clustering_keys = ds
@@ -88,20 +90,44 @@ case class SparkDeduper(conf: DedupConfig) extends Serializable {
     }
 
     //TODO: analytics
+    /*df_with_clustering_keys.groupBy(col("clustering"), col("key"))
+      .agg(expr("max(count) AS size"))
+      .orderBy(desc("size"))
+      .show*/
 
     val df_with_blocks = df_with_clustering_keys
-      // filter out rows with position exceeding the maxqueuesize parameter
-      .filter(col("position").leq(conf.getWf.getQueueMaxSize))
-      .groupBy("clustering", "key")
+      // split the clustering block into smaller blocks of queuemaxsize
+      .groupBy(col("clustering"), col("key"), functions.floor(col("position").divide(lit(conf.getWf.getQueueMaxSize))))
       .agg(functions.collect_set(functions.struct(model.schema.fieldNames.map(col): _*)).as("block"))
       .filter(functions.size(new Column("block")).gt(1))
+       .union(
+        //adjacency blocks
+        df_with_clustering_keys
+          // filter out leading and trailing elements
+          .filter(col("position").gt(conf.getWf.getSlidingWindowSize/2))
+          //.filter(col("position").lt(col("count").minus(conf.getWf.getSlidingWindowSize/2)))
+          // create small blocks of records on "the border" of maxqueuesize: getSlidingWindowSize/2 elements before and after
+          .filter(
+            col("position").mod(conf.getWf.getQueueMaxSize).lt(conf.getWf.getSlidingWindowSize/2) // slice of the start of block
+            || col("position").mod(conf.getWf.getQueueMaxSize).gt(conf.getWf.getQueueMaxSize - (conf.getWf.getSlidingWindowSize/2)) //slice of the end of the block
+          )
+          .groupBy(col("clustering"), col("key"), functions.floor((col("position") + lit(conf.getWf.getSlidingWindowSize/2)).divide(lit(conf.getWf.getQueueMaxSize))))
+          .agg(functions.collect_set(functions.struct(model.schema.fieldNames.map(col): _*)).as("block"))
+          .filter(functions.size(new Column("block")).gt(1))
+      )
 
     df_with_blocks
   }
 
   def clusterValuesUDF(cd: ClusteringDef) = {
     udf[mutable.WrappedArray[String], mutable.WrappedArray[Any]](values => {
-      values.flatMap(f => cd.clusteringFunction().apply(conf, Seq(f.toString).asJava).asScala)
+      val valueList = values.flatMap {
+        case a: mutable.WrappedArray[Any] => a.map(_.toString)
+        case s: Any => Seq(s.toString)
+      }.asJava;
+
+      mutable.WrappedArray.make(cd.clusteringFunction().apply(conf, valueList).toArray())
+
     })
   }
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/model/SparkModel.scala

@@ -1,13 +1,16 @@
 package eu.dnetlib.pace.model
 
 import com.jayway.jsonpath.{Configuration, JsonPath}
+import eu.dnetlib.pace.common.AbstractPaceFunctions
 import eu.dnetlib.pace.config.{DedupConfig, Type}
 import eu.dnetlib.pace.util.MapDocumentUtil
+import org.apache.commons.lang3.StringUtils
 import org.apache.spark.sql.catalyst.encoders.RowEncoder
 import org.apache.spark.sql.catalyst.expressions.GenericRowWithSchema
 import org.apache.spark.sql.types.{DataTypes, Metadata, StructField, StructType}
 import org.apache.spark.sql.{Dataset, Row}
 
+import java.util.Locale
 import java.util.regex.Pattern
 import scala.collection.JavaConverters._
 
@@ -60,7 +63,7 @@ case class SparkModel(conf: DedupConfig) {
     values(identityFieldPosition) = MapDocumentUtil.getJPathString(conf.getWf.getIdPath, documentContext)
 
     schema.fieldNames.zipWithIndex.foldLeft(values) {
-      case ((res, (fname, index))) => {
+      case ((res, (fname, index))) =>
         val fdef = conf.getPace.getModelMap.get(fname)
 
         if (fdef != null) {
@@ -96,13 +99,52 @@ case class SparkModel(conf: DedupConfig) {
             case Type.DoubleArray =>
               MapDocumentUtil.getJPathArray(fdef.getPath, json)
           }
+
+          val filter = fdef.getFilter
+
+          if (StringUtils.isNotBlank(fdef.getClean)) {
+            res(index) = res(index) match {
+              case x: Seq[String] => x.map(clean(_, fdef.getClean)).toSeq
+              case _ => clean(res(index).toString, fdef.getClean)
+            }
+          }
+
+          if (filter != null && !filter.isEmpty) {
+            res(index) = res(index) match {
+              case x: String if filter.contains(x.toLowerCase(Locale.ROOT)) => null
+              case x: Seq[String] => x.filter(s => !filter.contains(s.toLowerCase(Locale.ROOT))).toSeq
+              case _ => res(index)
+            }
+          }
+
+          if (fdef.getSorted) {
+            res(index) = res(index) match {
+              case x: Seq[String] => x.sorted.toSeq
+              case _ => res(index)
+            }
+          }
         }
 
         res
-      }
     }
 
     new GenericRowWithSchema(values, schema)
   }
+
+  def clean(value: String, cleantype: String) : String = {
+    val res = cleantype match {
+      case "title" => AbstractPaceFunctions.cleanup(value)
+      case _ => value
+    }
+
+//    if (!res.equals(AbstractPaceFunctions.normalize(value))) {
+//      println(res)
+//      println(AbstractPaceFunctions.normalize(value))
+//      println()
+//    }
+
+    res
+  }
+
 }
 

dhp-pace-core/src/main/java/eu/dnetlib/pace/tree/AuthorsMatch.java

@@ -23,7 +23,6 @@ public class AuthorsMatch extends AbstractListComparator {
 	private String MODE; // full or surname
 	private int SIZE_THRESHOLD;
 	private String TYPE; // count or percentage
-	private int common;
 
 	public AuthorsMatch(Map<String, String> params) {
 		super(params, new com.wcohen.ss.JaroWinkler());
@@ -35,7 +34,6 @@ public class AuthorsMatch extends AbstractListComparator {
 		FULLNAME_THRESHOLD = Double.parseDouble(params.getOrDefault("fullname_th", "0.9"));
 		SIZE_THRESHOLD = Integer.parseInt(params.getOrDefault("size_th", "20"));
 		TYPE = params.getOrDefault("type", "percentage");
-		common = 0;
 	}
 
 	protected AuthorsMatch(double w, AbstractStringDistance ssalgo) {
@@ -44,22 +42,27 @@ public class AuthorsMatch extends AbstractListComparator {
 
 	@Override
 	public double compare(final List<String> a, final List<String> b, final Config conf) {
-
 		if (a.isEmpty() || b.isEmpty())
 			return -1;
 
 		if (a.size() > SIZE_THRESHOLD || b.size() > SIZE_THRESHOLD)
 			return 1.0;
 
-		List<Person> aList = a.stream().map(author -> new Person(author, false)).collect(Collectors.toList());
+		int maxMiss = Integer.MAX_VALUE;
 		List<Person> bList = b.stream().map(author -> new Person(author, false)).collect(Collectors.toList());
 
-		common = 0;
+		Double threshold = getDoubleParam("threshold");
+
+		if (threshold != null && threshold >= 0.0 && threshold <= 1.0 && a.size() == b.size()) {
+			maxMiss = (int) Math.floor((1 - threshold) * Math.max(a.size(), b.size()));
+		}
+
+		int common = 0;
 		// compare each element of List1 with each element of List2
-		for (Person p1 : aList)
+		for (int i = 0; i < a.size(); i++) {
+			Person p1 = new Person(a.get(i), false);
 
 			for (Person p2 : bList) {
-
 				// both persons are inaccurate
 				if (!p1.isAccurate() && !p2.isAccurate()) {
 					// compare just normalized fullnames
@@ -118,11 +121,15 @@ public class AuthorsMatch extends AbstractListComparator {
 					}
 
 				}
-
 			}
 
+			if (i - common > maxMiss) {
+				return 0.0;
+			}
+		}
+
 		// normalization factor to compute the score
-		int normFactor = aList.size() == bList.size() ? aList.size() : (aList.size() + bList.size() - common);
+		int normFactor = a.size() == b.size() ? a.size() : (a.size() + b.size() - common);
 
 		if (TYPE.equals("percentage")) {
 			return (double) common / normFactor;

dhp-pace-core/src/main/java/eu/dnetlib/pace/tree/InstanceTypeMatch.java

@@ -25,6 +25,7 @@ public class InstanceTypeMatch extends AbstractListComparator {
 		translationMap.put("Conference object", "*");
 		translationMap.put("Other literature type", "*");
 		translationMap.put("Unknown", "*");
+		translationMap.put("UNKNOWN", "*");
 
 		// article types
 		translationMap.put("Article", "Article");
@@ -76,5 +77,4 @@ public class InstanceTypeMatch extends AbstractListComparator {
 	protected double normalize(final double d) {
 		return d;
 	}
-
 }

dhp-pace-core/src/main/java/eu/dnetlib/pace/tree/LevensteinTitle.java

@@ -3,6 +3,7 @@ package eu.dnetlib.pace.tree;
 
 import java.util.Map;
 
+import org.apache.commons.lang3.StringUtils;
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
 
@@ -30,16 +31,25 @@ public class LevensteinTitle extends AbstractStringComparator {
 	}
 
 	@Override
-	public double distance(final String a, final String b, final Config conf) {
-		final String ca = cleanup(a);
-		final String cb = cleanup(b);
-
+	public double distance(final String ca, final String cb, final Config conf) {
 		final boolean check = checkNumbers(ca, cb);
 
 		if (check)
 			return 0.5;
 
-		return normalize(ssalgo.score(ca, cb), ca.length(), cb.length());
+		Double threshold = getDoubleParam("threshold");
+
+		// reduce Levenshtein algo complexity when target threshold is known
+		if (threshold != null && threshold >= 0.0 && threshold <= 1.0) {
+			int maxdistance = (int) Math.floor((1 - threshold) * Math.max(ca.length(), cb.length()));
+			int score = StringUtils.getLevenshteinDistance(ca, cb, maxdistance);
+			if (score == -1) {
+				return 0;
+			}
+			return normalize(score, ca.length(), cb.length());
+		} else {
+			return normalize(StringUtils.getLevenshteinDistance(ca, cb), ca.length(), cb.length());
+		}
 	}
 
 	private double normalize(final double score, final int la, final int lb) {

dhp-pace-core/src/main/java/eu/dnetlib/pace/tree/MaxLengthMatch.java

@@ -0,0 +1,29 @@
+
+package eu.dnetlib.pace.tree;
+
+import java.util.Map;
+
+import eu.dnetlib.pace.config.Config;
+import eu.dnetlib.pace.tree.support.AbstractStringComparator;
+import eu.dnetlib.pace.tree.support.ComparatorClass;
+
+@ComparatorClass("maxLengthMatch")
+public class MaxLengthMatch extends AbstractStringComparator {
+
+	private final int limit;
+
+	public MaxLengthMatch(Map<String, String> params) {
+		super(params);
+
+		limit = Integer.parseInt(params.getOrDefault("limit", "200"));
+	}
+
+	@Override
+	public double compare(String a, String b, final Config conf) {
+		return a.length() < limit && b.length() < limit ? 1.0 : -1.0;
+	}
+
+	protected String toString(final Object object) {
+		return toFirstString(object);
+	}
+}

dhp-pace-core/src/main/java/eu/dnetlib/pace/tree/support/AbstractComparator.java

@@ -127,4 +127,14 @@ public abstract class AbstractComparator<T> extends AbstractPaceFunctions implem
 		return this.weight;
 	}
 
+	public Double getDoubleParam(String name) {
+		String svalue = params.get(name);
+
+		try {
+			return Double.parseDouble(svalue);
+		} catch (Throwable t) {
+		}
+
+		return null;
+	}
 }

dhp-pace-core/src/main/java/eu/dnetlib/pace/util/BlockProcessor.java

@@ -67,8 +67,10 @@ public class BlockProcessor {
 
 	private void processRows(final List<Row> queue, final Reporter context) {
 
-		for (int pivotPos = 0; pivotPos < queue.size(); pivotPos++) {
-			final Row pivot = queue.get(pivotPos);
+		IncrementalConnectedComponents icc = new IncrementalConnectedComponents(queue.size());
+
+		for (int i = 0; i < queue.size(); i++) {
+			final Row pivot = queue.get(i);
 
 			final String idPivot = pivot.getString(identifierFieldPos); // identifier
 			final Object fieldsPivot = getJavaValue(pivot, orderFieldPos);
@@ -76,9 +78,9 @@ public class BlockProcessor {
 			final WfConfig wf = dedupConf.getWf();
 
 			if (fieldPivot != null) {
-				int i = 0;
-				for (int windowPos = pivotPos + 1; windowPos < queue.size(); windowPos++) {
-					final Row curr = queue.get(windowPos);
+				for (int j = icc.nextUnconnected(i, i + 1); j >= 0
+					&& j < queue.size(); j = icc.nextUnconnected(i, j + 1)) {
+					final Row curr = queue.get(j);
 					final String idCurr = curr.getString(identifierFieldPos); // identifier
 
 					if (mustSkip(idCurr)) {
@@ -86,7 +88,7 @@ public class BlockProcessor {
 						break;
 					}
 
-					if (++i > wf.getSlidingWindowSize()) {
+					if (wf.getSlidingWindowSize() > 0 && (j - i) > wf.getSlidingWindowSize()) {
 						break;
 					}
 
@@ -97,7 +99,9 @@ public class BlockProcessor {
 
 						final TreeProcessor treeProcessor = new TreeProcessor(dedupConf);
 
-						emitOutput(treeProcessor.compare(pivot, curr), idPivot, idCurr, context);
+						if (emitOutput(treeProcessor.compare(pivot, curr), idPivot, idCurr, context)) {
+							icc.connect(i, j);
+						}
 					}
 				}
 			}
@@ -115,7 +119,8 @@ public class BlockProcessor {
 		return null;
 	}
 
-	private void emitOutput(final boolean result, final String idPivot, final String idCurr, final Reporter context) {
+	private boolean emitOutput(final boolean result, final String idPivot, final String idCurr,
+		final Reporter context) {
 
 		if (result) {
 			if (idPivot.compareTo(idCurr) <= 0) {
@@ -127,6 +132,8 @@ public class BlockProcessor {
 		} else {
 			context.incrementCounter(dedupConf.getWf().getEntityType(), "d < " + dedupConf.getWf().getThreshold(), 1);
 		}
+
+		return result;
 	}
 
 	private boolean mustSkip(final String idPivot) {
@@ -142,5 +149,4 @@ public class BlockProcessor {
 
 		context.emit(type, from, to);
 	}
-
 }

dhp-pace-core/src/main/java/eu/dnetlib/pace/util/IncrementalConnectedComponents.java

@@ -0,0 +1,50 @@
+
+package eu.dnetlib.pace.util;
+
+import java.util.BitSet;
+
+public class IncrementalConnectedComponents {
+	final private int size;
+
+	final private BitSet[] indexes;
+
+	IncrementalConnectedComponents(int size) {
+		this.size = size;
+		this.indexes = new BitSet[size];
+	}
+
+	public void connect(int i, int j) {
+		if (indexes[i] == null) {
+			if (indexes[j] == null) {
+				indexes[i] = new BitSet(size);
+			} else {
+				indexes[i] = indexes[j];
+			}
+		} else {
+			if (indexes[j] != null && indexes[i] != indexes[j]) {
+				// merge adjacency lists for i and j
+				indexes[i].or(indexes[j]);
+			}
+		}
+
+		indexes[i].set(i);
+		indexes[i].set(j);
+		indexes[j] = indexes[i];
+	}
+
+	public int nextUnconnected(int i, int j) {
+		if (indexes[i] == null) {
+			return j;
+		}
+		int result = indexes[i].nextClearBit(j);
+
+		return (result >= size) ? -1 : result;
+	}
+
+	public BitSet getConnections(int i) {
+		if (indexes[i] == null) {
+			return null;
+		}
+		return indexes[i];
+	}
+}

dhp-pace-core/src/main/java/eu/dnetlib/pace/util/MapDocumentUtil.java

@@ -97,6 +97,8 @@ public class MapDocumentUtil {
 			Object o = json.read(jsonPath);
 			if (o instanceof String)
 				return (String) o;
+			if (o instanceof Number)
+				return (String) o.toString();
 			if (o instanceof JSONArray && ((JSONArray) o).size() > 0)
 				return (String) ((JSONArray) o).get(0);
 			return "";

dhp-pace-core/src/main/java/eu/dnetlib/pace/util/PaceResolver.java

@@ -40,7 +40,7 @@ public class PaceResolver implements Serializable {
 				Collectors.toMap(cl -> cl.getAnnotation(ComparatorClass.class).value(), cl -> (Class<Comparator>) cl));
 	}
 
-	public ClusteringFunction getClusteringFunction(String name, Map<String, Integer> params) throws PaceException {
+	public ClusteringFunction getClusteringFunction(String name, Map<String, Object> params) throws PaceException {
 		try {
 			return clusteringFunctions.get(name).getDeclaredConstructor(Map.class).newInstance(params);
 		} catch (InstantiationException | IllegalAccessException | InvocationTargetException

dhp-pace-core/src/test/java/eu/dnetlib/pace/clustering/ClusteringFunctionTest.java

@@ -15,7 +15,7 @@ import eu.dnetlib.pace.config.DedupConfig;
 
 public class ClusteringFunctionTest extends AbstractPaceTest {
 
-	private static Map<String, Integer> params;
+	private static Map<String, Object> params;
 	private static DedupConfig conf;
 
 	@BeforeAll
@@ -40,10 +40,10 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 
 	@Test
 	public void testNgram() {
-		params.put("ngramLen", 3);
-		params.put("max", 8);
-		params.put("maxPerToken", 2);
-		params.put("minNgramLen", 1);
+		params.put("ngramLen", "3");
+		params.put("max", "8");
+		params.put("maxPerToken", "2");
+		params.put("minNgramLen", "1");
 
 		final ClusteringFunction ngram = new Ngrams(params);
 
@@ -54,8 +54,8 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 
 	@Test
 	public void testNgramPairs() {
-		params.put("ngramLen", 3);
-		params.put("max", 2);
+		params.put("ngramLen", "3");
+		params.put("max", "2");
 
 		final ClusteringFunction np = new NgramPairs(params);
 
@@ -66,8 +66,8 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 
 	@Test
 	public void testSortedNgramPairs() {
-		params.put("ngramLen", 3);
-		params.put("max", 2);
+		params.put("ngramLen", "3");
+		params.put("max", "2");
 
 		final ClusteringFunction np = new SortedNgramPairs(params);
 
@@ -87,9 +87,9 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 
 	@Test
 	public void testAcronym() {
-		params.put("max", 4);
-		params.put("minLen", 1);
-		params.put("maxLen", 3);
+		params.put("max", "4");
+		params.put("minLen", "1");
+		params.put("maxLen", "3");
 
 		final ClusteringFunction acro = new Acronyms(params);
 
@@ -100,8 +100,8 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 
 	@Test
 	public void testSuffixPrefix() {
-		params.put("len", 3);
-		params.put("max", 4);
+		params.put("len", "3");
+		params.put("max", "4");
 
 		final ClusteringFunction sp = new SuffixPrefix(params);
 
@@ -109,8 +109,8 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 		System.out.println(s);
 		System.out.println(sp.apply(conf, Lists.newArrayList(s)));
 
-		params.put("len", 3);
-		params.put("max", 1);
+		params.put("len", "3");
+		params.put("max", "1");
 
 		System.out.println(sp.apply(conf, Lists.newArrayList("Framework for general-purpose deduplication")));
 	}
@@ -118,8 +118,8 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 	@Test
 	public void testWordsSuffixPrefix() {
 
-		params.put("len", 3);
-		params.put("max", 4);
+		params.put("len", "3");
+		params.put("max", "4");
 
 		final ClusteringFunction sp = new WordsSuffixPrefix(params);
 
@@ -130,7 +130,7 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 
 	@Test
 	public void testWordsStatsSuffixPrefix() {
-		params.put("mod", 10);
+		params.put("mod", "10");
 
 		final ClusteringFunction sp = new WordsStatsSuffixPrefixChain(params);
 
@@ -167,7 +167,7 @@ public class ClusteringFunctionTest extends AbstractPaceTest {
 	@Test
 	public void testFieldValue() {
 
-		params.put("randomLength", 5);
+		params.put("randomLength", "5");
 
 		final ClusteringFunction sp = new SpaceTrimmingFieldValue(params);
 

dhp-pace-core/src/test/java/eu/dnetlib/pace/util/IncrementalConnectedComponentsTest.java

@@ -0,0 +1,40 @@
+
+package eu.dnetlib.pace.util;
+
+import static org.junit.jupiter.api.Assertions.assertEquals;
+import static org.junit.jupiter.api.Assertions.assertNull;
+
+import org.junit.jupiter.api.Test;
+
+public class IncrementalConnectedComponentsTest {
+
+	@Test
+	public void transitiveClosureTest() {
+		IncrementalConnectedComponents icc = new IncrementalConnectedComponents(10);
+
+		icc.connect(0, 1);
+		icc.connect(0, 2);
+		icc.connect(0, 3);
+
+		icc.connect(1, 2);
+		icc.connect(1, 4);
+		icc.connect(1, 5);
+
+		icc.connect(6, 7);
+		icc.connect(6, 9);
+
+		assertEquals(icc.getConnections(0).toString(), "{0, 1, 2, 3, 4, 5}");
+		assertEquals(icc.getConnections(1).toString(), "{0, 1, 2, 3, 4, 5}");
+		assertEquals(icc.getConnections(2).toString(), "{0, 1, 2, 3, 4, 5}");
+		assertEquals(icc.getConnections(3).toString(), "{0, 1, 2, 3, 4, 5}");
+		assertEquals(icc.getConnections(4).toString(), "{0, 1, 2, 3, 4, 5}");
+		assertEquals(icc.getConnections(5).toString(), "{0, 1, 2, 3, 4, 5}");
+
+		assertEquals(icc.getConnections(6).toString(), "{6, 7, 9}");
+		assertEquals(icc.getConnections(7).toString(), "{6, 7, 9}");
+		assertEquals(icc.getConnections(9).toString(), "{6, 7, 9}");
+
+		assertNull(icc.getConnections(8));
+	}
+
+}

dhp-workflows/dhp-actionmanager/src/main/java/eu/dnetlib/dhp/actionmanager/promote/PromoteAction.java

@@ -0,0 +1,39 @@
+/*
+ * Copyright (c) 2024.
+ * SPDX-FileCopyrightText: © 2023 Consiglio Nazionale delle Ricerche
+ * SPDX-License-Identifier: AGPL-3.0-or-later
+ */
+
+package eu.dnetlib.dhp.actionmanager.promote;
+
+/** Encodes the Actionset promotion strategies */
+public class PromoteAction {
+
+	/** The supported actionset promotion strategies
+	 *
+	 * ENRICH: promotes only records in the actionset matching another record in the
+	 *  graph and enriches them applying the given MergeAndGet strategy
+	 * UPSERT: promotes all the records in an actionset, matching records are updated
+	 *  using the given MergeAndGet strategy, the non-matching record as inserted as they are.
+	 */
+	public enum Strategy {
+		ENRICH, UPSERT
+	}
+
+	/**
+	 * Returns the string representation of the join type implementing the given PromoteAction.
+	 *
+	 * @param strategy the strategy to be used to promote the Actionset contents
+	 * @return the join type used to implement the promotion strategy
+	 */
+	public static String joinTypeForStrategy(PromoteAction.Strategy strategy) {
+		switch (strategy) {
+			case ENRICH:
+				return "left_outer";
+			case UPSERT:
+				return "full_outer";
+			default:
+				throw new IllegalStateException("unsupported PromoteAction: " + strategy.toString());
+		}
+	}
+}

dhp-workflows/dhp-actionmanager/src/main/java/eu/dnetlib/dhp/actionmanager/promote/PromoteActionPayloadForGraphTableJob.java

@@ -67,8 +67,9 @@ public class PromoteActionPayloadForGraphTableJob {
 		String outputGraphTablePath = parser.get("outputGraphTablePath");
 		logger.info("outputGraphTablePath: {}", outputGraphTablePath);
 
-		MergeAndGet.Strategy strategy = MergeAndGet.Strategy.valueOf(parser.get("mergeAndGetStrategy").toUpperCase());
-		logger.info("strategy: {}", strategy);
+		MergeAndGet.Strategy mergeAndGetStrategy = MergeAndGet.Strategy
+			.valueOf(parser.get("mergeAndGetStrategy").toUpperCase());
+		logger.info("mergeAndGetStrategy: {}", mergeAndGetStrategy);
 
 		Boolean shouldGroupById = Optional
 			.ofNullable(parser.get("shouldGroupById"))
@@ -76,6 +77,12 @@ public class PromoteActionPayloadForGraphTableJob {
 			.orElse(true);
 		logger.info("shouldGroupById: {}", shouldGroupById);
 
+		PromoteAction.Strategy promoteActionStrategy = Optional
+			.ofNullable(parser.get("promoteActionStrategy"))
+			.map(PromoteAction.Strategy::valueOf)
+			.orElse(PromoteAction.Strategy.UPSERT);
+		logger.info("promoteActionStrategy: {}", promoteActionStrategy);
+
 		@SuppressWarnings("unchecked")
 		Class<? extends Oaf> rowClazz = (Class<? extends Oaf>) Class.forName(graphTableClassName);
 		@SuppressWarnings("unchecked")
@@ -97,7 +104,8 @@ public class PromoteActionPayloadForGraphTableJob {
 					inputGraphTablePath,
 					inputActionPayloadPath,
 					outputGraphTablePath,
-					strategy,
+					mergeAndGetStrategy,
+					promoteActionStrategy,
 					rowClazz,
 					actionPayloadClazz,
 					shouldGroupById);
@@ -124,14 +132,16 @@ public class PromoteActionPayloadForGraphTableJob {
 		String inputGraphTablePath,
 		String inputActionPayloadPath,
 		String outputGraphTablePath,
-		MergeAndGet.Strategy strategy,
+		MergeAndGet.Strategy mergeAndGetStrategy,
+		PromoteAction.Strategy promoteActionStrategy,
 		Class<G> rowClazz,
 		Class<A> actionPayloadClazz, Boolean shouldGroupById) {
 		Dataset<G> rowDS = readGraphTable(spark, inputGraphTablePath, rowClazz);
 		Dataset<A> actionPayloadDS = readActionPayload(spark, inputActionPayloadPath, actionPayloadClazz);
 
 		Dataset<G> result = promoteActionPayloadForGraphTable(
-			rowDS, actionPayloadDS, strategy, rowClazz, actionPayloadClazz, shouldGroupById)
+			rowDS, actionPayloadDS, mergeAndGetStrategy, promoteActionStrategy, rowClazz, actionPayloadClazz,
+			shouldGroupById)
 				.map((MapFunction<G, G>) value -> value, Encoders.bean(rowClazz));
 
 		saveGraphTable(result, outputGraphTablePath);
@@ -183,7 +193,8 @@ public class PromoteActionPayloadForGraphTableJob {
 	private static <G extends Oaf, A extends Oaf> Dataset<G> promoteActionPayloadForGraphTable(
 		Dataset<G> rowDS,
 		Dataset<A> actionPayloadDS,
-		MergeAndGet.Strategy strategy,
+		MergeAndGet.Strategy mergeAndGetStrategy,
+		PromoteAction.Strategy promoteActionStrategy,
 		Class<G> rowClazz,
 		Class<A> actionPayloadClazz,
 		Boolean shouldGroupById) {
@@ -195,8 +206,9 @@ public class PromoteActionPayloadForGraphTableJob {
 
 		SerializableSupplier<Function<G, String>> rowIdFn = ModelSupport::idFn;
 		SerializableSupplier<Function<A, String>> actionPayloadIdFn = ModelSupport::idFn;
-		SerializableSupplier<BiFunction<G, A, G>> mergeRowWithActionPayloadAndGetFn = MergeAndGet.functionFor(strategy);
-		SerializableSupplier<BiFunction<G, G, G>> mergeRowsAndGetFn = MergeAndGet.functionFor(strategy);
+		SerializableSupplier<BiFunction<G, A, G>> mergeRowWithActionPayloadAndGetFn = MergeAndGet
+			.functionFor(mergeAndGetStrategy);
+		SerializableSupplier<BiFunction<G, G, G>> mergeRowsAndGetFn = MergeAndGet.functionFor(mergeAndGetStrategy);
 		SerializableSupplier<G> zeroFn = zeroFn(rowClazz);
 		SerializableSupplier<Function<G, Boolean>> isNotZeroFn = PromoteActionPayloadForGraphTableJob::isNotZeroFnUsingIdOrSourceAndTarget;
 
@@ -207,6 +219,7 @@ public class PromoteActionPayloadForGraphTableJob {
 				rowIdFn,
 				actionPayloadIdFn,
 				mergeRowWithActionPayloadAndGetFn,
+				promoteActionStrategy,
 				rowClazz,
 				actionPayloadClazz);
 

dhp-workflows/dhp-actionmanager/src/main/java/eu/dnetlib/dhp/actionmanager/promote/PromoteActionPayloadFunctions.java

@@ -34,6 +34,7 @@ public class PromoteActionPayloadFunctions {
 	 * @param rowIdFn Function used to get the id of graph table row
 	 * @param actionPayloadIdFn Function used to get id of action payload instance
 	 * @param mergeAndGetFn Function used to merge graph table row and action payload instance
+	 * @param promoteActionStrategy the Actionset promotion strategy
 	 * @param rowClazz Class of graph table
 	 * @param actionPayloadClazz Class of action payload
 	 * @param <G> Type of graph table row
@@ -46,6 +47,7 @@ public class PromoteActionPayloadFunctions {
 		SerializableSupplier<Function<G, String>> rowIdFn,
 		SerializableSupplier<Function<A, String>> actionPayloadIdFn,
 		SerializableSupplier<BiFunction<G, A, G>> mergeAndGetFn,
+		PromoteAction.Strategy promoteActionStrategy,
 		Class<G> rowClazz,
 		Class<A> actionPayloadClazz) {
 		if (!isSubClass(rowClazz, actionPayloadClazz)) {
@@ -61,7 +63,7 @@ public class PromoteActionPayloadFunctions {
 			.joinWith(
 				actionPayloadWithIdDS,
 				rowWithIdDS.col("_1").equalTo(actionPayloadWithIdDS.col("_1")),
-				"full_outer")
+				PromoteAction.joinTypeForStrategy(promoteActionStrategy))
 			.map(
 				(MapFunction<Tuple2<Tuple2<String, G>, Tuple2<String, A>>, G>) value -> {
 					Optional<G> rowOpt = Optional.ofNullable(value._1()).map(Tuple2::_2);

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/promote/promote_action_payload_for_graph_table_input_parameters.json

@@ -41,6 +41,12 @@
     "paramDescription": "strategy for merging graph table objects with action payload instances, MERGE_FROM_AND_GET or SELECT_NEWER_AND_GET",
     "paramRequired": true
   },
+  {
+    "paramName": "pas",
+    "paramLongName": "promoteActionStrategy",
+    "paramDescription": "strategy for promoting the actionset contents into the graph tables, ENRICH or UPSERT (default)",
+    "paramRequired": false
+  },
   {
     "paramName": "sgid",
     "paramLongName": "shouldGroupById",

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/dataset/oozie_app/workflow.xml

@@ -115,6 +115,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Dataset</arg>
             <arg>--outputGraphTablePath</arg><arg>${workingDir}/dataset</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="DecisionPromoteResultActionPayloadForDatasetTable"/>
@@ -167,6 +168,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Result</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/dataset</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="End"/>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/datasource/oozie_app/workflow.xml

@@ -106,6 +106,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Datasource</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/datasource</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
         </spark>
         <ok to="End"/>
         <error to="Kill"/>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/organization/oozie_app/workflow.xml

@@ -106,6 +106,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Organization</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/organization</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
         </spark>
         <ok to="End"/>
         <error to="Kill"/>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/otherresearchproduct/oozie_app/workflow.xml

@@ -114,6 +114,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.OtherResearchProduct</arg>
             <arg>--outputGraphTablePath</arg><arg>${workingDir}/otherresearchproduct</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="DecisionPromoteResultActionPayloadForOtherResearchProductTable"/>
@@ -166,6 +167,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Result</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/otherresearchproduct</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="End"/>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/project/oozie_app/workflow.xml

@@ -106,6 +106,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Project</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/project</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
         </spark>
         <ok to="End"/>
         <error to="Kill"/>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/publication/oozie_app/workflow.xml

@@ -115,6 +115,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Publication</arg>
             <arg>--outputGraphTablePath</arg><arg>${workingDir}/publication</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="DecisionPromoteResultActionPayloadForPublicationTable"/>
@@ -167,6 +168,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Result</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/publication</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="End"/>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/relation/oozie_app/workflow.xml

@@ -107,6 +107,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Relation</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/relation</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
         </spark>
         <ok to="End"/>
         <error to="Kill"/>

dhp-workflows/dhp-actionmanager/src/main/resources/eu/dnetlib/dhp/actionmanager/wf/software/oozie_app/workflow.xml

@@ -114,6 +114,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Software</arg>
             <arg>--outputGraphTablePath</arg><arg>${workingDir}/software</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="DecisionPromoteResultActionPayloadForSoftwareTable"/>
@@ -166,6 +167,7 @@
             <arg>--actionPayloadClassName</arg><arg>eu.dnetlib.dhp.schema.oaf.Result</arg>
             <arg>--outputGraphTablePath</arg><arg>${outputGraphRootPath}/software</arg>
             <arg>--mergeAndGetStrategy</arg><arg>${mergeAndGetStrategy}</arg>
+            <arg>--promoteActionStrategy</arg><arg>${promoteActionStrategy}</arg>
             <arg>--shouldGroupById</arg><arg>${shouldGroupById}</arg>
         </spark>
         <ok to="End"/>

dhp-workflows/dhp-actionmanager/src/test/java/eu/dnetlib/dhp/actionmanager/promote/PromoteActionPayloadFunctionsTest.java

@@ -54,7 +54,7 @@ public class PromoteActionPayloadFunctionsTest {
 				RuntimeException.class,
 				() -> PromoteActionPayloadFunctions
 					.joinGraphTableWithActionPayloadAndMerge(
-						null, null, null, null, null, OafImplSubSub.class, OafImpl.class));
+						null, null, null, null, null, null, OafImplSubSub.class, OafImpl.class));
 		}
 
 		@Test
@@ -104,6 +104,7 @@ public class PromoteActionPayloadFunctionsTest {
 					rowIdFn,
 					actionPayloadIdFn,
 					mergeAndGetFn,
+					PromoteAction.Strategy.UPSERT,
 					OafImplSubSub.class,
 					OafImplSubSub.class)
 				.collectAsList();
@@ -183,6 +184,7 @@ public class PromoteActionPayloadFunctionsTest {
 					rowIdFn,
 					actionPayloadIdFn,
 					mergeAndGetFn,
+					PromoteAction.Strategy.UPSERT,
 					OafImplSubSub.class,
 					OafImplSub.class)
 				.collectAsList();

dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/createunresolvedentities/PrepareFOSSparkJob.java

@@ -124,8 +124,19 @@ public class PrepareFOSSparkJob implements Serializable {
 		FOSDataModel first) {
 		level1.add(first.getLevel1());
 		level2.add(first.getLevel2());
-		level3.add(first.getLevel3() + "@@" + first.getScoreL3());
-		level4.add(first.getLevel4() + "@@" + first.getScoreL4());
+		if (Optional.ofNullable(first.getLevel3()).isPresent() &&
+			!first.getLevel3().equalsIgnoreCase(NA) && !first.getLevel3().equalsIgnoreCase(NULL)
+			&& first.getLevel3() != null)
+			level3.add(first.getLevel3() + "@@" + first.getScoreL3());
+		else
+			level3.add(NULL);
+		if (Optional.ofNullable(first.getLevel4()).isPresent() &&
+			!first.getLevel4().equalsIgnoreCase(NA) &&
+			!first.getLevel4().equalsIgnoreCase(NULL) &&
+			first.getLevel4() != null)
+			level4.add(first.getLevel4() + "@@" + first.getScoreL4());
+		else
+			level4.add(NULL);
 	}
 
 }

dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/collection/CollectorWorker.java

@@ -19,6 +19,7 @@ import org.slf4j.LoggerFactory;
 import eu.dnetlib.dhp.aggregation.common.ReporterCallback;
 import eu.dnetlib.dhp.aggregation.common.ReportingJob;
 import eu.dnetlib.dhp.collection.plugin.CollectorPlugin;
+import eu.dnetlib.dhp.collection.plugin.base.BaseCollectorPlugin;
 import eu.dnetlib.dhp.collection.plugin.file.FileCollectorPlugin;
 import eu.dnetlib.dhp.collection.plugin.file.FileGZipCollectorPlugin;
 import eu.dnetlib.dhp.collection.plugin.mongodb.MDStoreCollectorPlugin;
@@ -120,6 +121,8 @@ public class CollectorWorker extends ReportingJob {
 				return new FileCollectorPlugin(fileSystem);
 			case fileGzip:
 				return new FileGZipCollectorPlugin(fileSystem);
+			case baseDump:
+				return new BaseCollectorPlugin(this.fileSystem);
 			case other:
 				final CollectorPlugin.NAME.OTHER_NAME plugin = Optional
 					.ofNullable(api.getParams().get("other_plugin_type"))

dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/collection/plugin/CollectorPlugin.java

@@ -10,7 +10,8 @@ import eu.dnetlib.dhp.common.collection.CollectorException;
 public interface CollectorPlugin {
 
 	enum NAME {
-		oai, other, rest_json2xml, file, fileGzip;
+
+		oai, other, rest_json2xml, file, fileGzip, baseDump;
 
 		public enum OTHER_NAME {
 			mdstore_mongodb_dump, mdstore_mongodb

dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/collection/plugin/base/BaseCollectorIterator.java

@@ -0,0 +1,171 @@
+
+package eu.dnetlib.dhp.collection.plugin.base;
+
+import java.io.BufferedInputStream;
+import java.io.ByteArrayInputStream;
+import java.io.InputStream;
+import java.io.StringWriter;
+import java.util.Iterator;
+import java.util.concurrent.BlockingQueue;
+import java.util.concurrent.LinkedBlockingQueue;
+
+import javax.xml.stream.XMLEventReader;
+import javax.xml.stream.XMLEventWriter;
+import javax.xml.stream.XMLInputFactory;
+import javax.xml.stream.XMLOutputFactory;
+import javax.xml.stream.events.EndElement;
+import javax.xml.stream.events.StartElement;
+import javax.xml.stream.events.XMLEvent;
+
+import org.apache.commons.compress.archivers.tar.TarArchiveEntry;
+import org.apache.commons.compress.archivers.tar.TarArchiveInputStream;
+import org.apache.commons.compress.compressors.CompressorInputStream;
+import org.apache.commons.compress.compressors.CompressorStreamFactory;
+import org.apache.commons.io.IOUtils;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import eu.dnetlib.dhp.common.aggregation.AggregatorReport;
+
+public class BaseCollectorIterator implements Iterator<String> {
+
+	private String nextElement;
+
+	private final BlockingQueue<String> queue = new LinkedBlockingQueue<>(100);
+
+	private static final Logger log = LoggerFactory.getLogger(BaseCollectorIterator.class);
+
+	private static final String END_ELEM = "__END__";
+
+	public BaseCollectorIterator(final FileSystem fs, final Path filePath, final AggregatorReport report) {
+		new Thread(() -> importHadoopFile(fs, filePath, report)).start();
+		try {
+			this.nextElement = this.queue.take();
+		} catch (final InterruptedException e) {
+			throw new RuntimeException(e);
+		}
+	}
+
+	protected BaseCollectorIterator(final String resourcePath, final AggregatorReport report) {
+		new Thread(() -> importTestFile(resourcePath, report)).start();
+		try {
+			this.nextElement = this.queue.take();
+		} catch (final InterruptedException e) {
+			throw new RuntimeException(e);
+		}
+	}
+
+	@Override
+	public synchronized boolean hasNext() {
+		return (this.nextElement != null) & !END_ELEM.equals(this.nextElement);
+	}
+
+	@Override
+	public synchronized String next() {
+		try {
+			return END_ELEM.equals(this.nextElement) ? null : this.nextElement;
+		} finally {
+			try {
+				this.nextElement = this.queue.take();
+			} catch (final InterruptedException e) {
+				throw new RuntimeException(e);
+			}
+		}
+
+	}
+
+	private void importHadoopFile(final FileSystem fs, final Path filePath, final AggregatorReport report) {
+		log.info("I start to read the TAR stream");
+
+		try (InputStream origInputStream = fs.open(filePath);
+			final TarArchiveInputStream tarInputStream = new TarArchiveInputStream(origInputStream)) {
+			importTarStream(tarInputStream, report);
+		} catch (final Throwable e) {
+			throw new RuntimeException("Error processing BASE records", e);
+		}
+	}
+
+	private void importTestFile(final String resourcePath, final AggregatorReport report) {
+		try (final InputStream origInputStream = BaseCollectorIterator.class.getResourceAsStream(resourcePath);
+			final TarArchiveInputStream tarInputStream = new TarArchiveInputStream(origInputStream)) {
+			importTarStream(tarInputStream, report);
+		} catch (final Throwable e) {
+			throw new RuntimeException("Error processing BASE records", e);
+		}
+	}
+
+	private void importTarStream(final TarArchiveInputStream tarInputStream, final AggregatorReport report) {
+		long count = 0;
+
+		final XMLInputFactory xmlInputFactory = XMLInputFactory.newInstance();
+		final XMLOutputFactory xmlOutputFactory = XMLOutputFactory.newInstance();
+
+		try {
+			TarArchiveEntry entry;
+			while ((entry = (TarArchiveEntry) tarInputStream.getNextEntry()) != null) {
+				final String name = entry.getName();
+
+				if (!entry.isDirectory() && name.contains("ListRecords") && name.endsWith(".bz2")) {
+
+					log.info("Processing file (BZIP): " + name);
+
+					final byte[] bzipData = new byte[(int) entry.getSize()];
+					IOUtils.readFully(tarInputStream, bzipData);
+
+					try (InputStream bzipIs = new ByteArrayInputStream(bzipData);
+						final BufferedInputStream bzipBis = new BufferedInputStream(bzipIs);
+						final CompressorInputStream bzipInput = new CompressorStreamFactory()
+							.createCompressorInputStream(bzipBis)) {
+
+						final XMLEventReader reader = xmlInputFactory.createXMLEventReader(bzipInput);
+
+						XMLEventWriter eventWriter = null;
+						StringWriter xmlWriter = null;
+
+						while (reader.hasNext()) {
+							final XMLEvent nextEvent = reader.nextEvent();
+
+							if (nextEvent.isStartElement()) {
+								final StartElement startElement = nextEvent.asStartElement();
+								if ("record".equals(startElement.getName().getLocalPart())) {
+									xmlWriter = new StringWriter();
+									eventWriter = xmlOutputFactory.createXMLEventWriter(xmlWriter);
+								}
+							}
+
+							if (eventWriter != null) {
+								eventWriter.add(nextEvent);
+							}
+
+							if (nextEvent.isEndElement()) {
+								final EndElement endElement = nextEvent.asEndElement();
+								if ("record".equals(endElement.getName().getLocalPart())) {
+									eventWriter.flush();
+									eventWriter.close();
+
+									this.queue.put(xmlWriter.toString());
+
+									eventWriter = null;
+									xmlWriter = null;
+									count++;
+								}
+							}
+
+						}
+					}
+				}
+			}
+
+			this.queue.put(END_ELEM); // TO INDICATE THE END OF THE QUEUE
+		} catch (final Throwable e) {
+			log.error("Error processing BASE records", e);
+			report.put(e.getClass().getName(), e.getMessage());
+			throw new RuntimeException("Error processing BASE records", e);
+		} finally {
+			log.info("Total records (written in queue): " + count);
+		}
+	}
+
+}

dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/collection/plugin/base/BaseCollectorPlugin.java

@@ -0,0 +1,159 @@
+
+package eu.dnetlib.dhp.collection.plugin.base;
+
+import java.io.IOException;
+import java.sql.SQLException;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Optional;
+import java.util.Set;
+import java.util.Spliterator;
+import java.util.Spliterators;
+import java.util.stream.Stream;
+import java.util.stream.StreamSupport;
+
+import org.apache.commons.io.IOUtils;
+import org.apache.commons.lang3.StringUtils;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.dom4j.Document;
+import org.dom4j.DocumentException;
+import org.dom4j.DocumentHelper;
+import org.dom4j.Node;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import eu.dnetlib.dhp.collection.ApiDescriptor;
+import eu.dnetlib.dhp.collection.plugin.CollectorPlugin;
+import eu.dnetlib.dhp.collection.plugin.file.AbstractSplittedRecordPlugin;
+import eu.dnetlib.dhp.common.DbClient;
+import eu.dnetlib.dhp.common.aggregation.AggregatorReport;
+import eu.dnetlib.dhp.common.collection.CollectorException;
+
+public class BaseCollectorPlugin implements CollectorPlugin {
+
+	private final FileSystem fs;
+
+	private static final Logger log = LoggerFactory.getLogger(AbstractSplittedRecordPlugin.class);
+
+	// MAPPING AND FILTERING ARE DEFINED HERE:
+	// https://docs.google.com/document/d/1Aj-ZAV11b44MCrAAUCPiS2TUlXb6PnJEu1utCMAcCOU/edit
+
+	public BaseCollectorPlugin(final FileSystem fs) {
+		this.fs = fs;
+	}
+
+	@Override
+	public Stream<String> collect(final ApiDescriptor api, final AggregatorReport report) throws CollectorException {
+		// the path of the dump file on HDFS
+		// http://oai.base-search.net/initial_load/base_oaipmh_dump-current.tar
+		// it could be downloaded from iis-cdh5-test-gw.ocean.icm.edu.pl and then copied on HDFS
+		final Path filePath = Optional
+			.ofNullable(api.getBaseUrl())
+			.map(Path::new)
+			.orElseThrow(() -> new CollectorException("missing baseUrl"));
+
+		// get the parameters for the connection to the OpenAIRE database.
+		// the database is used to obtain the list of the datasources that the plugin will collect
+		final String dbUrl = api.getParams().get("dbUrl");
+		final String dbUser = api.getParams().get("dbUser");
+		final String dbPassword = api.getParams().get("dbPassword");
+
+		// the types(comma separated, empty value for all) that the plugin will collect,
+		// the types should be expressed in the format of the normalized types of BASE (for example 1,121,...)
+		final String acceptedNormTypesString = api.getParams().get("acceptedNormTypes");
+
+		log.info("baseUrl: {}", filePath);
+		log.info("dbUrl: {}", dbUrl);
+		log.info("dbUser: {}", dbUser);
+		log.info("dbPassword: {}", "***");
+		log.info("acceptedNormTypes: {}", acceptedNormTypesString);
+
+		try {
+			if (!this.fs.exists(filePath)) {
+				throw new CollectorException("path does not exist: " + filePath);
+			}
+		} catch (final Throwable e) {
+			throw new CollectorException(e);
+		}
+
+		final Set<String> acceptedOpendoarIds = findAcceptedOpendoarIds(dbUrl, dbUser, dbPassword);
+
+		final Set<String> acceptedNormTypes = new HashSet<>();
+		if (StringUtils.isNotBlank(acceptedNormTypesString)) {
+			for (final String s : StringUtils.split(acceptedNormTypesString, ",")) {
+				if (StringUtils.isNotBlank(s)) {
+					acceptedNormTypes.add(s.trim());
+				}
+			}
+		}
+
+		final Iterator<String> iterator = new BaseCollectorIterator(this.fs, filePath, report);
+		final Spliterator<String> spliterator = Spliterators.spliteratorUnknownSize(iterator, Spliterator.ORDERED);
+		return StreamSupport
+			.stream(spliterator, false)
+			.filter(doc -> filterXml(doc, acceptedOpendoarIds, acceptedNormTypes));
+	}
+
+	private Set<String> findAcceptedOpendoarIds(final String dbUrl, final String dbUser, final String dbPassword)
+		throws CollectorException {
+		final Set<String> accepted = new HashSet<>();
+
+		try (final DbClient dbClient = new DbClient(dbUrl, dbUser, dbPassword)) {
+
+			final String sql = IOUtils
+				.toString(
+					getClass().getResourceAsStream("/eu/dnetlib/dhp/collection/plugin/base/sql/opendoar-accepted.sql"));
+
+			dbClient.processResults(sql, row -> {
+				try {
+					final String dsId = row.getString("id");
+					log.info("Accepted Datasource: " + dsId);
+					accepted.add(dsId);
+				} catch (final SQLException e) {
+					log.error("Error in SQL", e);
+					throw new RuntimeException("Error in SQL", e);
+				}
+			});
+
+		} catch (final IOException e) {
+			log.error("Error accessong SQL", e);
+			throw new CollectorException("Error accessong SQL", e);
+		}
+
+		log.info("Accepted Datasources (TOTAL): " + accepted.size());
+
+		return accepted;
+	}
+
+	protected static boolean filterXml(final String xml,
+		final Set<String> acceptedOpendoarIds,
+		final Set<String> acceptedNormTypes) {
+		try {
+
+			final Document doc = DocumentHelper.parseText(xml);
+
+			final String id = doc.valueOf("//*[local-name()='collection']/@opendoar_id").trim();
+
+			if (StringUtils.isBlank(id) || !acceptedOpendoarIds.contains("opendoar____::" + id)) {
+				return false;
+			}
+
+			if (acceptedNormTypes.isEmpty()) {
+				return true;
+			}
+
+			for (final Object s : doc.selectNodes("//*[local-name()='typenorm']")) {
+				if (acceptedNormTypes.contains(((Node) s).getText().trim())) {
+					return true;
+				}
+			}
+
+			return false;
+		} catch (final DocumentException e) {
+			log.error("Error parsing document", e);
+			throw new RuntimeException("Error parsing document", e);
+		}
+	}
+
+}

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/plugin/base/sql/base.sql

@@ -0,0 +1,114 @@
+BEGIN;
+
+INSERT INTO dsm_services(
+	_dnet_resource_identifier_, 
+	id,
+	officialname,
+	englishname,
+	namespaceprefix,
+	websiteurl,
+	logourl,
+	platform,
+	contactemail,
+	collectedfrom,
+	provenanceaction,
+	_typology_to_remove_,
+	eosc_type,
+	eosc_datasource_type,
+	research_entity_types,
+	thematic
+) VALUES (
+	'openaire____::base_search',
+	'openaire____::base_search',
+	'Bielefeld Academic Search Engine (BASE)',
+	'Bielefeld Academic Search Engine (BASE)',
+	'base_search_',
+	'https://www.base-search.net',
+	'https://www.base-search.net/about/download/logo_224x57_white.gif',
+	'BASE',
+	'openaire-helpdesk@uni-bielefeld.de',
+	'infrastruct_::openaire',
+	'user:insert',
+	'aggregator::pubsrepository::unknown',
+	'Data Source',
+	'Aggregator',
+	ARRAY['Research Products'],
+	false
+);
+
+INSERT INTO dsm_service_organization(
+	_dnet_resource_identifier_,
+	organization,
+	service
+) VALUES (
+	'fairsharing_::org::214@@openaire____::base_search',
+	'fairsharing_::org::214',
+	'openaire____::base_search'
+);
+
+INSERT INTO dsm_api(
+	_dnet_resource_identifier_,
+	id,
+	service,
+	protocol,
+	baseurl,
+	metadata_identifier_path
+) VALUES (
+	'api_________::openaire____::base_search::dump',
+	'api_________::openaire____::base_search::dump',
+	'openaire____::base_search',
+	'baseDump',
+	'/user/michele.artini/base-import/base_oaipmh_dump-current.tar',
+	'//*[local-name()=''header'']/*[local-name()=''identifier'']'
+);
+
+
+INSERT INTO dsm_apiparams(
+	_dnet_resource_identifier_, 
+	api, 
+	param, 
+	value
+) VALUES (
+	'api_________::openaire____::base_search::dump@@dbUrl',
+	'api_________::openaire____::base_search::dump',
+	'dbUrl',
+	'jdbc:postgresql://postgresql.services.openaire.eu:5432/dnet_openaireplus'
+);
+
+INSERT INTO dsm_apiparams(
+	_dnet_resource_identifier_, 
+	api, 
+	param, 
+	value
+) VALUES (
+	'api_________::openaire____::base_search::dump@@dbUser',
+	'api_________::openaire____::base_search::dump',
+	'dbUser',
+	'dnet'
+);
+
+INSERT INTO dsm_apiparams(
+	_dnet_resource_identifier_, 
+	api, 
+	param, 
+	value
+) VALUES (
+	'api_________::openaire____::base_search::dump@@dbPassword',
+	'api_________::openaire____::base_search::dump',
+	'dbPassword',
+	'***'
+);
+
+INSERT INTO dsm_apiparams(
+	_dnet_resource_identifier_, 
+	api, 
+	param, 
+	value
+) VALUES (
+	'api_________::openaire____::base_search::dump@@acceptedNormTypes',
+	'api_________::openaire____::base_search::dump',
+	'acceptedNormTypes',
+	'1,11,111,121,13,14,15,18,181,182,183,1A,6,7'
+);
+
+COMMIT;

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/plugin/base/sql/opendoar-accepted.sql

@@ -0,0 +1,9 @@
+select s.id as id 
+from dsm_services s 
+where collectedfrom = 'openaire____::opendoar' 
+and jurisdiction = 'Institutional'
+and s.id in (
+	select service from dsm_api where coalesce(compatibility_override, compatibility) = 'driver' or coalesce(compatibility_override, compatibility) = 'UNKNOWN'
+) and s.id not in (
+	select service from dsm_api where coalesce(compatibility_override, compatibility) like '%openaire%'
+);

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/plugin/base/sql/opendoar-aggregation-status.sql

@@ -0,0 +1,11 @@
+select 
+	s.id           as id, 
+	s.jurisdiction as jurisdiction, 
+	array_remove(array_agg(a.id || ' (compliance: ' || coalesce(a.compatibility_override, a.compatibility, 'UNKNOWN') || ')@@@' || coalesce(a.last_collection_total, 0)), NULL) as aggregations
+from 
+	dsm_services s 
+	join dsm_api a on (s.id = a.service) 
+where 
+	collectedfrom = 'openaire____::opendoar'
+group by 
+	s.id;

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/plugin/base/xml/base-types.vocabulary.xml

@@ -0,0 +1,180 @@
+<RESOURCE_PROFILE>
+	<HEADER>
+		<RESOURCE_IDENTIFIER value="c67911d6-9988-4a3b-b965-7d39bdd4a31d_Vm9jYWJ1bGFyeURTUmVzb3VyY2VzL1ZvY2FidWxhcnlEU1Jlc291cmNlVHlwZQ==" />
+		<RESOURCE_TYPE value="VocabularyDSResourceType" />
+		<RESOURCE_KIND value="VocabularyDSResources" />
+		<RESOURCE_URI value="" />
+		<DATE_OF_CREATION value="2024-02-13T11:15:48+00:00" />
+	</HEADER>
+	<BODY>
+		<CONFIGURATION>
+			<VOCABULARY_NAME code="base:normalized_types">base:normalized_types</VOCABULARY_NAME>
+			<VOCABULARY_DESCRIPTION>base:normalized_types</VOCABULARY_DESCRIPTION>
+			<TERMS>
+				<TERM native_name="Text" code="Text" english_name="Text" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="1" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Book" code="Book" english_name="Book" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="11" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Book part" code="Book part" english_name="Book part" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="111" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Journal/Newspaper" code="Journal/Newspaper" english_name="Journal/Newspaper" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="12" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Article contribution" code="Article contribution" english_name="Article contribution" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="121" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Other non-article" code="Other non-article" english_name="Other non-article" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="122" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Conference object" code="Conference object" english_name="Conference object" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="13" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Report" code="Report" english_name="Report" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="14" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Review" code="Review" english_name="Review" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="15" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Course material" code="Course material" english_name="Course material" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="16" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Lecture" code="Lecture" english_name="Lecture" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="17" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Thesis" code="Thesis" english_name="Thesis" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="18" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Bachelor's thesis" code="Bachelor's thesis" english_name="Bachelor's thesis" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="181" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Master's thesis" code="Master's thesis" english_name="Master's thesis" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="182" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Doctoral and postdoctoral thesis" code="Doctoral and postdoctoral thesis" english_name="Doctoral and postdoctoral thesis" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="183" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Manuscript" code="Manuscript" english_name="Manuscript" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="19" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Patent" code="Patent" english_name="Patent" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="1A" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Musical notation" code="Musical notation" english_name="Musical notation" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="2" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Map" code="Map" english_name="Map" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="3" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Audio" code="Audio" english_name="Audio" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="4" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Image/Video" code="Image/Video" english_name="Image/Video" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="5" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Still image" code="Still image" english_name="Still image" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="51" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Moving image/Video" code="Moving image/Video" english_name="Moving image/Video" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="52" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Software" code="Software" english_name="Software" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="6" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Dataset" code="Dataset" english_name="Dataset" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="7" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+				<TERM native_name="Unknown" code="Unknown" english_name="Unknown" encoding="BASE">
+					<SYNONYMS>
+						<SYNONYM term="F" encoding="BASE" />
+					</SYNONYMS>
+					<RELATIONS />
+				</TERM>
+
+			</TERMS>
+		</CONFIGURATION>
+		<STATUS>
+			<LAST_UPDATE value="2013-11-18T10:46:36Z" />
+		</STATUS>
+		<SECURITY_PARAMETERS>String</SECURITY_PARAMETERS>
+	</BODY>
+</RESOURCE_PROFILE>
+
+                

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/plugin/base/xml/base2oaf.transformationRule.xml

@@ -0,0 +1,302 @@
+<RESOURCE_PROFILE>
+	<HEADER>
+		<RESOURCE_IDENTIFIER value="" />
+		<RESOURCE_TYPE value="TransformationRuleDSResourceType" />
+		<RESOURCE_KIND value="TransformationRuleDSResources" />
+		<RESOURCE_URI value="" />
+		<DATE_OF_CREATION value="2024-03-05T11:23:00+00:00" />
+	</HEADER>
+	<BODY>
+		<CONFIGURATION>
+			<SOURCE_METADATA_FORMAT interpretation="cleaned" layout="store" name="dc" />
+			<SINK_METADATA_FORMAT name="oaf_hbase" />
+			<IMPORTED />
+			<SCRIPT>
+				<TITLE>xslt_base2oaf_hadoop</TITLE>
+				<CODE>
+					<xsl:stylesheet xmlns:oaire="http://namespace.openaire.eu/schema/oaire/" xmlns:dateCleaner="http://eu/dnetlib/transform/dateISO"
+						xmlns:base_dc="http://oai.base-search.net/base_dc/"
+						xmlns:datacite="http://datacite.org/schema/kernel-4" xmlns:dr="http://www.driver-repository.eu/namespace/dr" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+						xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:vocabulary="http://eu/dnetlib/transform/clean" xmlns:oaf="http://namespace.openaire.eu/oaf"
+						xmlns:oai="http://www.openarchives.org/OAI/2.0/" xmlns:dri="http://www.driver-repository.eu/namespace/dri" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:dc="http://purl.org/dc/elements/1.1/"
+						exclude-result-prefixes="xsl vocabulary dateCleaner base_dc" version="2.0">
+						<xsl:param name="varOfficialName" />
+						<xsl:param name="varDataSourceId" />
+						<xsl:param name="varFP7" select="'corda_______::'" />
+						<xsl:param name="varH2020" select="'corda__h2020::'" />
+						<xsl:param name="repoCode" select="substring-before(//*[local-name() = 'header']/*[local-name()='recordIdentifier'], ':')" />
+						<xsl:param name="index" select="0" />
+						<xsl:param name="transDate" select="current-dateTime()" />
+
+						<xsl:template name="terminate">
+							<xsl:message terminate="yes">
+								record is not compliant, transformation is interrupted.
+							</xsl:message>
+						</xsl:template>
+
+						<xsl:template match="/">
+							<record>
+								<xsl:apply-templates select="//*[local-name() = 'header']" />
+
+
+<!-- TO EVALUATE
+base_dc:authod_id
+base_dc:authod_id/base_dc:creator_id
+base_dc:authod_id/base_dc:creator_name
+
+example:
+
+<dc:creator>ALBU, Svetlana</dc:creator>
+
+<base_dc:authod_id>
+	<base_dc:creator_name>ALBU, Svetlana</base_dc:creator_name>
+    <base_dc:creator_id>https://orcid.org/0000-0002-8648-950X</base_dc:creator_id>
+</base_dc:authod_id>
+-->
+
+<!-- NOT USED 
+base_dc:global_id (I used oai:identifier)
+base_dc:collection/text()
+
+base_dc:continent
+base_dc:year (I used dc:date)
+dc:coverage
+dc:language (I used base_dc:lang)
+base_dc:link (I used dc:identifier)
+-->
+
+
+								<xsl:variable name="varBaseNormType" select="vocabulary:clean(//base_dc:typenorm, 'base:normalized_types')" />
+								
+
+								<metadata>
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:title" />
+										<xsl:with-param name="targetElement" select="'dc:title'" />
+									</xsl:call-template>
+
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:creator/replace(., '^(.*)\|.*$', '$1')" />
+										<xsl:with-param name="targetElement" select="'dc:creator'" />
+									</xsl:call-template>
+
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:contributor" />
+										<xsl:with-param name="targetElement" select="'dc:contributor'" />
+									</xsl:call-template>
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:description" />
+										<xsl:with-param name="targetElement" select="'dc:description'" />
+									</xsl:call-template>
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:subject" />
+										<xsl:with-param name="targetElement" select="'dc:subject'" />
+									</xsl:call-template>
+									
+									<!-- TODO: I'm not sure if this is the correct encoding -->
+									<xsl:for-each select="//base_dc:classcode|//base_dc:autoclasscode">
+										<dc:subject><xsl:value-of select="concat(@type, ':', .)" /></dc:subject>
+									</xsl:for-each>
+									<!-- END TODO -->
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:publisher" />
+										<xsl:with-param name="targetElement" select="'dc:publisher'" />
+									</xsl:call-template>
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:format" />
+										<xsl:with-param name="targetElement" select="'dc:format'" />
+									</xsl:call-template>
+									
+									<dc:type>
+										<xsl:value-of select="$varBaseNormType" />
+									</dc:type>
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:type" />
+										<xsl:with-param name="targetElement" select="'dc:type'" />
+									</xsl:call-template>
+									
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:source" />
+										<xsl:with-param name="targetElement" select="'dc:source'" />
+									</xsl:call-template>
+									
+									<dc:language>
+										<xsl:value-of select="vocabulary:clean( //base_dc:lang, 'dnet:languages')" />
+									</dc:language>
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:rights" />
+										<xsl:with-param name="targetElement" select="'dc:rights'" />
+									</xsl:call-template>
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:relation" />
+										<xsl:with-param name="targetElement" select="'dc:relation'" />
+									</xsl:call-template>
+									
+									<xsl:if test="not(//dc:identifier[starts-with(., 'http')])">
+										<xsl:call-template name="terminate" />
+									</xsl:if>
+									
+									<xsl:call-template name="allElements">
+										<xsl:with-param name="sourceElement" select="//dc:identifier[starts-with(., 'http')]" />
+										<xsl:with-param name="targetElement" select="'dc:identifier'" />
+									</xsl:call-template>
+									
+									<xsl:for-each select="//dc:relation">
+										<xsl:if test="matches(normalize-space(.), '(info:eu-repo/grantagreement/ec/fp7/)(\d\d\d\d\d\d)(.*)', 'i')">
+											<oaf:projectid>
+												<xsl:value-of select="concat($varFP7, replace(normalize-space(.), '(info:eu-repo/grantagreement/ec/fp7/)(\d\d\d\d\d\d)(.*)', '$2', 'i'))" />
+											</oaf:projectid>
+										</xsl:if>
+										<xsl:if test="matches(normalize-space(.), '(info:eu-repo/grantagreement/ec/h2020/)(\d\d\d\d\d\d)(.*)', 'i')">
+											<oaf:projectid>
+												<xsl:value-of select="concat($varH2020, replace(normalize-space(.), '(info:eu-repo/grantagreement/ec/h2020/)(\d\d\d\d\d\d)(.*)', '$2', 'i'))" />
+											</oaf:projectid>
+										</xsl:if>
+									</xsl:for-each>
+
+									<dr:CobjCategory>
+										<xsl:variable name="varCobjCategory" select="vocabulary:clean($varBaseNormType, 'dnet:publication_resource')" />
+										<xsl:variable name="varSuperType" select="vocabulary:clean($varCobjCategory, 'dnet:result_typologies')" />
+										<xsl:attribute name="type" select="$varSuperType" />
+										<xsl:value-of select="$varCobjCategory" />
+									</dr:CobjCategory>
+									
+									<oaf:accessrights>
+										<xsl:choose>
+											<xsl:when test="//base_dc:oa[.='1']">OPEN</xsl:when>
+											<xsl:when test="//base_dc:rightsnorm">
+												<xsl:value-of select="vocabulary:clean(//base_dc:rightsnorm, 'dnet:access_modes')" />
+											</xsl:when>
+											<xsl:when test="//dc:rights">
+												<xsl:value-of select="vocabulary:clean( //dc:rights, 'dnet:access_modes')" />
+											</xsl:when>
+											<xsl:otherwise>UNKNOWN</xsl:otherwise>
+										</xsl:choose>
+									</oaf:accessrights>
+									
+									<xsl:for-each select="//base_dc:doi">
+										<oaf:identifier identifierType="doi">
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>
+
+									<xsl:for-each select="distinct-values(//dc:identifier[starts-with(., 'http') and (not(contains(., '://dx.doi.org/') or contains(., '://doi.org/') or contains(., '://hdl.handle.net/')))])">
+										<oaf:identifier identifierType="url">
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>
+
+									<xsl:for-each select="distinct-values(//dc:identifier[starts-with(., 'http') and contains(., '://hdl.handle.net/')]/substring-after(., 'hdl.handle.net/'))">
+										<oaf:identifier identifierType="handle">
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>									
+
+									<xsl:for-each select="distinct-values(//dc:identifier[starts-with(., 'urn:nbn:nl:') or starts-with(., 'URN:NBN:NL:')])">
+										<oaf:identifier identifierType='urn'>
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>
+									
+									<oaf:identifier identifierType="oai-original">
+										<xsl:value-of
+											select="//*[local-name() = 'about']/*[local-name() = 'provenance']//*[local-name() = 'originDescription' and not(./*[local-name() = 'originDescription'])]/*[local-name() = 'identifier']" />
+									</oaf:identifier>
+									
+									<oaf:hostedBy>
+										<xsl:attribute name="name">
+											<xsl:value-of select="//base_dc:collname" />
+										</xsl:attribute>
+										<xsl:attribute name="id">
+											<xsl:value-of select="concat('opendoar____::', //base_dc:collection/@opendoar_id)" />
+										</xsl:attribute>
+									</oaf:hostedBy>
+									
+									<oaf:collectedFrom>
+										<xsl:attribute name="name">
+											<xsl:value-of select="$varOfficialName" />
+										</xsl:attribute>
+										<xsl:attribute name="id">
+											<xsl:value-of select="$varDataSourceId" />
+										</xsl:attribute>
+									</oaf:collectedFrom>
+									
+									<oaf:dateAccepted>
+										<xsl:value-of select="dateCleaner:dateISO( //dc:date[1] )" />
+									</oaf:dateAccepted>
+									
+									<xsl:if test="//base_dc:oa[.='1']">
+										<xsl:for-each select="//dc:relation[starts-with(., 'http')]">
+											<oaf:fulltext>
+												<xsl:value-of select="normalize-space(.)" />
+											</oaf:fulltext>
+										</xsl:for-each>
+									</xsl:if>
+									
+									<xsl:for-each select="//base_dc:collection/@ror_id">
+										<oaf:relation relType="resultOrganization"
+										 	subRelType="affiliation"
+										 	relClass="hasAuthorInstitution"
+											targetType="organization">
+											<xsl:choose>
+												<xsl:when test="contains(.,'https://ror.org/')">
+													<xsl:value-of select="concat('ror_________::', normalize-space(.))" />
+												</xsl:when>
+												<xsl:otherwise>
+													<xsl:value-of select="concat('ror_________::https://ror.org/', normalize-space(.))" />
+												</xsl:otherwise>
+											</xsl:choose>
+										</oaf:relation>										
+									</xsl:for-each>
+
+									<xsl:for-each select="//base_dc:country">
+										<oaf:country><xsl:value-of select="vocabulary:clean(., 'dnet:countries')" /></oaf:country>
+									</xsl:for-each>
+									
+								</metadata>
+								<xsl:copy-of select="//*[local-name() = 'about']" />
+							</record>
+						</xsl:template>
+
+						<xsl:template name="allElements">
+							<xsl:param name="sourceElement" />
+							<xsl:param name="targetElement" />
+							<xsl:for-each select="$sourceElement">
+								<xsl:element name="{$targetElement}">
+									<xsl:value-of select="normalize-space(.)" />
+								</xsl:element>
+							</xsl:for-each>
+						</xsl:template>
+
+						<xsl:template match="//*[local-name() = 'header']">
+							<xsl:if test="//oai:header/@status='deleted'">
+								<xsl:call-template name="terminate" />
+							</xsl:if>
+							<xsl:copy>
+								<xsl:apply-templates select="node()|@*" />
+								<xsl:element name="dr:dateOfTransformation">
+									<xsl:value-of select="$transDate" />
+								</xsl:element>
+							</xsl:copy>
+						</xsl:template>
+
+						<xsl:template match="node()|@*">
+							<xsl:copy>
+								<xsl:apply-templates select="node()|@*" />
+							</xsl:copy>
+						</xsl:template>
+					</xsl:stylesheet>
+				</CODE>
+			</SCRIPT>
+		</CONFIGURATION>
+		<STATUS />
+		<SECURITY_PARAMETERS />
+	</BODY>
+</RESOURCE_PROFILE>

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/collection/plugin/base/xml/base2odf.transformationRule.xml

@@ -0,0 +1,326 @@
+<RESOURCE_PROFILE>
+	<HEADER>
+		<RESOURCE_IDENTIFIER value="2ad0cdd9-c96c-484c-8b0e-ed56d86891fe_VHJhbnNmb3JtYXRpb25SdWxlRFNSZXNvdXJjZXMvVHJhbnNmb3JtYXRpb25SdWxlRFNSZXNvdXJjZVR5cGU=" />
+		<RESOURCE_TYPE value="TransformationRuleDSResourceType" />
+		<RESOURCE_KIND value="TransformationRuleDSResources" />
+		<RESOURCE_URI value="" />
+		<DATE_OF_CREATION value="2024-03-05T11:23:00+00:00" />
+	</HEADER>
+	<BODY>
+		<CONFIGURATION>
+			<SOURCE_METADATA_FORMAT interpretation="cleaned" layout="store" name="dc" />
+			<SINK_METADATA_FORMAT name="odf_hbase" />
+			<IMPORTED />
+			<SCRIPT>
+				<TITLE>xslt_base2odf_hadoop</TITLE>
+				<CODE>
+					<xsl:stylesheet xmlns:oaire="http://namespace.openaire.eu/schema/oaire/" xmlns:dateCleaner="http://eu/dnetlib/transform/dateISO" xmlns:base_dc="http://oai.base-search.net/base_dc/"
+						xmlns:datacite="http://datacite.org/schema/kernel-4" xmlns:dr="http://www.driver-repository.eu/namespace/dr" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+						xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:vocabulary="http://eu/dnetlib/transform/clean" xmlns:oaf="http://namespace.openaire.eu/oaf"
+						xmlns:oai="http://www.openarchives.org/OAI/2.0/" xmlns:dri="http://www.driver-repository.eu/namespace/dri" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:dc="http://purl.org/dc/elements/1.1/"
+						exclude-result-prefixes="xsl vocabulary dateCleaner base_dc" version="2.0">
+						<xsl:param name="varOfficialName" />
+						<xsl:param name="varDataSourceId" />
+						<xsl:param name="varFP7" select="'corda_______::'" />
+						<xsl:param name="varH2020" select="'corda__h2020::'" />
+						<xsl:param name="repoCode" select="substring-before(//*[local-name() = 'header']/*[local-name()='recordIdentifier'], ':')" />
+						<xsl:param name="index" select="0" />
+						<xsl:param name="transDate" select="current-dateTime()" />
+
+						<xsl:template name="terminate">
+							<xsl:message terminate="yes">
+								record is not compliant, transformation is interrupted.
+							</xsl:message>
+						</xsl:template>
+
+						<xsl:template match="/">
+							<record>
+								<xsl:apply-templates select="//*[local-name() = 'header']" />
+
+
+								<!-- NOT USED 
+									base_dc:global_id (I used oai:identifier) 
+									base_dc:collection/text() 
+									base_dc:continent 
+									dc:coverage
+									dc:source
+									dc:relation
+									dc:type (I used //base_dc:typenorm)
+									dc:language (I used base_dc:lang) 
+									base_dc:link (I used dc:identifier)
+								 -->
+
+
+								<xsl:variable name="varBaseNormType" select="vocabulary:clean(//base_dc:typenorm, 'base:normalized_types')" />
+
+
+								<metadata>
+									<datacite:resource>
+
+										<xsl:for-each select="//base_dc:doi">
+											<datacite:identifier identifierType="DOI">
+												<xsl:value-of select="." />
+											</datacite:identifier>
+										</xsl:for-each>										
+
+										<datacite:alternateIdentifiers>
+											<xsl:for-each
+												select="distinct-values(//dc:identifier[starts-with(., 'http') and (not(contains(., '://dx.doi.org/') or contains(., '://doi.org/') or contains(., '://hdl.handle.net/')))])">
+												<datacite:identifier alternateIdentifierType="url">
+													<xsl:value-of select="." />
+												</datacite:identifier>
+											</xsl:for-each>
+
+											<xsl:for-each select="distinct-values(//dc:identifier[starts-with(., 'http') and contains(., '://hdl.handle.net/')]/substring-after(., 'hdl.handle.net/'))">
+												<datacite:identifier alternateIdentifierType="handle">
+													<xsl:value-of select="." />
+												</datacite:identifier>
+											</xsl:for-each>
+
+											<xsl:for-each select="distinct-values(//dc:identifier[starts-with(., 'urn:nbn:nl:') or starts-with(., 'URN:NBN:NL:')])">
+												<datacite:identifier alternateIdentifierType='urn'>
+													<xsl:value-of select="." />
+												</datacite:identifier>
+											</xsl:for-each>
+
+											<datacite:identifier alternateIdentifierType="oai-original">
+												<xsl:value-of
+													select="//*[local-name() = 'about']/*[local-name() = 'provenance']//*[local-name() = 'originDescription' and not(./*[local-name() = 'originDescription'])]/*[local-name() = 'identifier']" />
+											</datacite:identifier>
+										</datacite:alternateIdentifiers>
+
+										<datacite:relatedIdentifiers />
+
+
+										<datacite:resourceType><xsl:value-of select="$varBaseNormType" /></datacite:resourceType>
+
+										<datacite:titles>
+											<xsl:for-each select="//dc:title">
+												<datacite:title>
+													<xsl:value-of select="normalize-space(.)" />
+												</datacite:title>
+											</xsl:for-each>
+										</datacite:titles>
+
+										<datacite:creators>
+											<xsl:for-each select="//dc:creator">
+												<xsl:variable name="author" select="normalize-space(.)" />
+												<datacite:creator>
+													<datacite:creatorName>
+														<xsl:value-of select="$author" />
+													</datacite:creatorName>
+													<xsl:for-each select="//base_dc:authod_id[normalize-space(./base_dc:creator_name) = $author]/base_dc:creator_id ">
+														<xsl:if test="contains(.,'https://orcid.org/')">
+															<nameIdentifier schemeURI="https://orcid.org/" nameIdentifierScheme="ORCID">
+																<xsl:value-of select="substring-after(., 'https://orcid.org/')" />
+															</nameIdentifier>
+														</xsl:if>
+													</xsl:for-each>
+												</datacite:creator>
+											</xsl:for-each>
+										</datacite:creators>
+
+										<datacite:contributors>
+											<xsl:for-each select="//dc:contributor">
+												<datacite:contributor>
+													<datacite:contributorName>
+														<xsl:value-of select="normalize-space(.)" />
+													</datacite:contributorName>
+												</datacite:contributor>
+											</xsl:for-each>
+										</datacite:contributors>
+
+										<datacite:descriptions>
+											<xsl:for-each select="//dc:description">
+												<datacite:description descriptionType="Abstract">
+													<xsl:value-of select="normalize-space(.)" />
+												</datacite:description>
+											</xsl:for-each>
+										</datacite:descriptions>
+
+										<datacite:subjects>
+											<xsl:for-each select="//dc:subject">
+												<datacite:subject>
+													<xsl:value-of select="normalize-space(.)" />
+												</datacite:subject>
+											</xsl:for-each>
+											
+											<xsl:for-each select="//base_dc:classcode|//base_dc:autoclasscode">
+												<datacite:subject subjectScheme="{@type}" classificationCode="{normalize-space(.)}">
+													<!-- TODO the value should be obtained by the Code -->
+													<xsl:value-of select="normalize-space(.)" />
+												</datacite:subject>
+											</xsl:for-each>
+										</datacite:subjects>
+										
+										<datacite:publisher>
+											<xsl:value-of select="normalize-space(//dc:publisher)" />
+										</datacite:publisher>
+										
+										<datacite:publicationYear>
+											<xsl:value-of select="normalize-space(//base_dc:year)" />
+										</datacite:publicationYear>
+										
+										<datacite:formats>
+											<xsl:for-each select="//dc:format">
+												<datacite:format>
+													<xsl:value-of select="normalize-space(.)" />
+												</datacite:format>
+											</xsl:for-each>
+										</datacite:formats>
+										
+										<datacite:language>
+											<xsl:value-of select="vocabulary:clean( //base_dc:lang, 'dnet:languages')" />
+										</datacite:language>
+
+										<oaf:accessrights>
+											<xsl:if test="//base_dc:oa[.='1']">
+												<datacite:rights rightsURI="http://purl.org/coar/access_right/c_abf2">open access</datacite:rights>
+											</xsl:if>
+											<xsl:for-each select="//dc:rights|//base_dc:rightsnorm">
+												<datacite:rights><xsl:value-of select="vocabulary:clean(., 'dnet:access_modes')" /></datacite:rights>	
+											</xsl:for-each>
+										</oaf:accessrights>
+
+									</datacite:resource>
+
+										<xsl:for-each select="//dc:relation">
+											<xsl:if test="matches(normalize-space(.), '(info:eu-repo/grantagreement/ec/fp7/)(\d\d\d\d\d\d)(.*)', 'i')">
+												<oaf:projectid>
+													<xsl:value-of select="concat($varFP7, replace(normalize-space(.), '(info:eu-repo/grantagreement/ec/fp7/)(\d\d\d\d\d\d)(.*)', '$2', 'i'))" />
+												</oaf:projectid>
+											</xsl:if>
+											<xsl:if test="matches(normalize-space(.), '(info:eu-repo/grantagreement/ec/h2020/)(\d\d\d\d\d\d)(.*)', 'i')">
+												<oaf:projectid>
+													<xsl:value-of select="concat($varH2020, replace(normalize-space(.), '(info:eu-repo/grantagreement/ec/h2020/)(\d\d\d\d\d\d)(.*)', '$2', 'i'))" />
+												</oaf:projectid>
+											</xsl:if>
+										</xsl:for-each>
+
+									<dr:CobjCategory>
+										<xsl:variable name="varCobjCategory" select="vocabulary:clean($varBaseNormType, 'dnet:publication_resource')" />
+										<xsl:variable name="varSuperType" select="vocabulary:clean($varCobjCategory, 'dnet:result_typologies')" />
+										<xsl:attribute name="type" select="$varSuperType" />
+										<xsl:value-of select="$varCobjCategory" />
+									</dr:CobjCategory>
+
+									<oaf:accessrights>
+										<xsl:choose>
+											<xsl:when test="//base_dc:oa[.='1']">OPEN</xsl:when>
+											<xsl:when test="//base_dc:rightsnorm">
+												<xsl:value-of select="vocabulary:clean(//base_dc:rightsnorm, 'dnet:access_modes')" />
+											</xsl:when>
+											<xsl:when test="//dc:rights">
+												<xsl:value-of select="vocabulary:clean( //dc:rights, 'dnet:access_modes')" />
+											</xsl:when>
+											<xsl:otherwise>UNKNOWN</xsl:otherwise>
+										</xsl:choose>
+									</oaf:accessrights>
+
+									<xsl:for-each select="//base_dc:doi">
+										<oaf:identifier identifierType="doi">
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>
+
+									<xsl:for-each
+										select="distinct-values(//dc:identifier[starts-with(., 'http') and ( not(contains(., '://dx.doi.org/') or contains(., '://doi.org/') or contains(., '://hdl.handle.net/')))])">
+										<oaf:identifier identifierType="url">
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>
+
+									<xsl:for-each select="distinct-values(//dc:identifier[starts-with(., 'http') and contains(., '://hdl.handle.net/')]/substring-after(., 'hdl.handle.net/'))">
+										<oaf:identifier identifierType="handle">
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>
+
+									<xsl:for-each select="distinct-values(//dc:identifier[starts-with(., 'urn:nbn:nl:') or starts-with(., 'URN:NBN:NL:')])">
+										<oaf:identifier identifierType='urn'>
+											<xsl:value-of select="." />
+										</oaf:identifier>
+									</xsl:for-each>
+
+									<oaf:identifier identifierType="oai-original">
+										<xsl:value-of
+											select="//*[local-name() = 'about']/*[local-name() = 'provenance']//*[local-name() = 'originDescription' and not(./*[local-name() = 'originDescription'])]/*[local-name() = 'identifier']" />
+									</oaf:identifier>
+
+									<oaf:hostedBy>
+										<xsl:attribute name="name">
+											<xsl:value-of select="//base_dc:collname" />
+										</xsl:attribute>
+										<xsl:attribute name="id">
+											<xsl:value-of select="concat('opendoar____::', //base_dc:collection/@opendoar_id)" />
+										</xsl:attribute>
+									</oaf:hostedBy>
+
+									<oaf:collectedFrom>
+										<xsl:attribute name="name">
+											<xsl:value-of select="$varOfficialName" />
+										</xsl:attribute>
+										<xsl:attribute name="id">
+											<xsl:value-of select="$varDataSourceId" />
+										</xsl:attribute>
+									</oaf:collectedFrom>
+
+									<oaf:dateAccepted>
+										<xsl:value-of select="dateCleaner:dateISO( //dc:date[1] )" />
+									</oaf:dateAccepted>
+
+									<xsl:if test="//base_dc:oa[.='1']">
+										<xsl:for-each select="//dc:relation[starts-with(., 'http')]">
+											<oaf:fulltext>
+												<xsl:value-of select="normalize-space(.)" />
+											</oaf:fulltext>
+										</xsl:for-each>
+									</xsl:if>
+
+									<xsl:for-each select="//base_dc:collection/@ror_id">
+										<oaf:relation relType="resultOrganization" subRelType="affiliation" relClass="hasAuthorInstitution" targetType="organization">
+											<xsl:choose>
+												<xsl:when test="contains(.,'https://ror.org/')">
+													<xsl:value-of select="concat('ror_________::', normalize-space(.))" />
+												</xsl:when>
+												<xsl:otherwise>
+													<xsl:value-of select="concat('ror_________::https://ror.org/', normalize-space(.))" />
+												</xsl:otherwise>
+											</xsl:choose>
+										</oaf:relation>
+									</xsl:for-each>
+									
+									<xsl:for-each select="//base_dc:country">
+										<oaf:country><xsl:value-of select="vocabulary:clean(., 'dnet:countries')" /></oaf:country>
+									</xsl:for-each>
+									
+								</metadata>
+								<xsl:copy-of select="//*[local-name() = 'about']" />
+							</record>
+						</xsl:template>
+
+						<xsl:template match="//*[local-name() = 'header']">
+							<xsl:if test="//oai:header/@status='deleted'">
+								<xsl:call-template name="terminate" />
+							</xsl:if>
+							<xsl:copy>
+								<xsl:apply-templates select="node()|@*" />
+								<xsl:element name="dr:dateOfTransformation">
+									<xsl:value-of select="$transDate" />
+								</xsl:element>
+							</xsl:copy>
+						</xsl:template>
+
+						<xsl:template match="node()|@*">
+							<xsl:copy>
+								<xsl:apply-templates select="node()|@*" />
+							</xsl:copy>
+						</xsl:template>
+					</xsl:stylesheet>
+				</CODE>
+			</SCRIPT>
+		</CONFIGURATION>
+		<STATUS />
+		<SECURITY_PARAMETERS />
+	</BODY>
+</RESOURCE_PROFILE>

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/datacite/hostedBy_map.json

@@ -1048,5 +1048,10 @@
   "openaire_id": "re3data_____::r3d100010399",
   "datacite_name": "ZEW Forschungsdatenzentrum",
   "official_name": "ZEW Forschungsdatenzentrum"
+ },
+ "HBP.NEUROINF": {
+  "openaire_id": "fairsharing_::2975",
+  "datacite_name": "EBRAINS",
+  "official_name": "EBRAINS"
  }
 }

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/sx/bio/db/oozie_app/workflow.xml

@@ -1,4 +1,4 @@
-    <workflow-app name="Transform_BioEntity_Workflow" xmlns="uri:oozie:workflow:0.5">
+<workflow-app name="Transform_BioEntity_Workflow" xmlns="uri:oozie:workflow:0.5">
     <parameters>
         <property>
             <name>sourcePath</name>
@@ -8,19 +8,40 @@
             <name>database</name>
             <description>the PDB Database Working Path</description>
         </property>
-
         <property>
-            <name>targetPath</name>
-            <description>the Target Working dir path</description>
+            <name>mdStoreOutputId</name>
+            <description>the identifier of the cleaned MDStore</description>
+        </property>
+        <property>
+            <name>mdStoreManagerURI</name>
+            <description>the path of the cleaned mdstore</description>
         </property>
     </parameters>
 
-    <start to="ConvertDB"/>
+    <start to="StartTransaction"/>
+
 
     <kill name="Kill">
         <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
     </kill>
 
+    <action name="StartTransaction">
+        <java>
+            <configuration>
+                <property>
+                    <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                    <value>true</value>
+                </property>
+            </configuration>
+            <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+            <arg>--action</arg><arg>NEW_VERSION</arg>
+            <arg>--mdStoreID</arg><arg>${mdStoreOutputId}</arg>
+            <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+            <capture-output/>
+        </java>
+        <ok to="ConvertDB"/>
+        <error to="RollBack"/>
+    </action>
     <action name="ConvertDB">
         <spark xmlns="uri:oozie:spark-action:0.2">
             <master>yarn</master>
@@ -41,11 +62,48 @@
             <arg>--master</arg><arg>yarn</arg>
             <arg>--dbPath</arg><arg>${sourcePath}</arg>
             <arg>--database</arg><arg>${database}</arg>
-            <arg>--targetPath</arg><arg>${targetPath}</arg>
+            <arg>--mdstoreOutputVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
         </spark>
-        <ok to="End"/>
-        <error to="Kill"/>
+        <ok to="CommitVersion"/>
+        <error to="RollBack"/>
+
     </action>
-    <end name="End"/>
+        <action name="CommitVersion">
+            <java>
+                <configuration>
+                    <property>
+                        <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                        <value>true</value>
+                    </property>
+                </configuration>
+                <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+                <arg>--action</arg><arg>COMMIT</arg>
+                <arg>--namenode</arg><arg>${nameNode}</arg>
+                <arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
+                <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+            </java>
+            <ok to="End"/>
+            <error to="Kill"/>
+        </action>
+
+        <action name="RollBack">
+            <java>
+                <configuration>
+                    <property>
+                        <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                        <value>true</value>
+                    </property>
+                </configuration>
+                <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+                <arg>--action</arg><arg>ROLLBACK</arg>
+                <arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
+                <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+            </java>
+            <ok to="Kill"/>
+            <error to="Kill"/>
+        </action>
+
+
+        <end name="End"/>
 
 </workflow-app>

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/sx/bio/ebi/baseline_to_oaf_params.json

@@ -2,7 +2,7 @@
   {"paramName":"mt",  "paramLongName":"master",         "paramDescription": "should be local or yarn",                  "paramRequired": true},
   {"paramName":"i",   "paramLongName":"isLookupUrl",    "paramDescription": "isLookupUrl",                              "paramRequired": true},
   {"paramName":"w",   "paramLongName":"workingPath",    "paramDescription": "the path of the sequencial file to read",  "paramRequired": true},
-  {"paramName":"t",   "paramLongName":"targetPath",     "paramDescription": "the oaf path ",                            "paramRequired": true},
+  {"paramName":"mo",   "paramLongName":"mdstoreOutputVersion",     "paramDescription": "the oaf path ",                            "paramRequired": true},
   {"paramName":"s",   "paramLongName":"skipUpdate",     "paramDescription": "skip update ",                             "paramRequired": false},
   {"paramName":"h",   "paramLongName":"hdfsServerUri",  "paramDescription": "the  working path ",                       "paramRequired": true}
 ]

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/sx/bio/ebi/bio_to_oaf_params.json

@@ -2,5 +2,5 @@
   {"paramName":"mt",  "paramLongName":"master",       "paramDescription": "should be local or yarn",                  "paramRequired": true},
   {"paramName":"db",  "paramLongName":"database",     "paramDescription": "should be PDB or UNIPROT",                 "paramRequired": true},
   {"paramName":"p",   "paramLongName":"dbPath",       "paramDescription": "the path of the database to transform",    "paramRequired": true},
-  {"paramName":"t",   "paramLongName":"targetPath",   "paramDescription": "the OAF target path ",                     "paramRequired": true}
+  {"paramName":"mo",   "paramLongName":"mdstoreOutputVersion",     "paramDescription": "the oaf path ",                "paramRequired": true}
 ]

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/sx/bio/ebi/ebi_to_df_params.json

@@ -1,5 +1,20 @@
 [
-  {"paramName":"mt",  "paramLongName":"master",     "paramDescription": "should be local or yarn",                  "paramRequired": true},
-  {"paramName":"s",   "paramLongName":"sourcePath","paramDescription": "the source Path",                              "paramRequired": true},
-  {"paramName":"t",   "paramLongName":"targetPath","paramDescription": "the  oaf path ",  "paramRequired": true}
+  {
+    "paramName": "mt",
+    "paramLongName": "master",
+    "paramDescription": "should be local or yarn",
+    "paramRequired": true
+  },
+  {
+    "paramName": "s",
+    "paramLongName": "sourcePath",
+    "paramDescription": "the source Path",
+    "paramRequired": true
+  },
+  {
+    "paramName": "mo",
+    "paramLongName": "mdstoreOutputVersion",
+    "paramDescription": "the oaf path ",
+    "paramRequired": true
+  }
 ]

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/sx/bio/ebi/oozie_app/workflow.xml

@@ -9,34 +9,26 @@
             <description>the Working Path</description>
         </property>
         <property>
-            <name>targetPath</name>
-            <description>the OAF MDStore Path</description>
+            <name>mdStoreOutputId</name>
+            <description>the identifier of the cleaned MDStore</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>
+            <name>mdStoreManagerURI</name>
+            <description>the path of the cleaned mdstore</description>
         </property>
         <property>
             <name>resumeFrom</name>
-            <value>DownloadEBILinks</value>
+            <value>CreateEBIDataSet</value>
             <description>node to start</description>
         </property>
     </parameters>
 
-    <start to="resume_from"/>
+    <start to="StartTransaction"/>
 
     <decision name="resume_from">
         <switch>
             <case to="DownloadEBILinks">${wf:conf('resumeFrom') eq 'DownloadEBILinks'}</case>
-            <case to="CreateEBIDataSet">${wf:conf('resumeFrom') eq 'CreateEBIDataSet'}</case>
+            <case to="StartTransaction">${wf:conf('resumeFrom') eq 'CreateEBIDataSet'}</case>
             <default to="DownloadEBILinks"/>
         </switch>
     </decision>
@@ -77,9 +69,29 @@
             <move source="${sourcePath}/ebi_links_dataset" target="${sourcePath}/ebi_links_dataset_old"/>
             <move source="${workingPath}/links_final" target="${sourcePath}/ebi_links_dataset"/>
         </fs>
-        <ok to="CreateEBIDataSet"/>
+        <ok to="StartTransaction"/>
         <error to="Kill"/>
     </action>
+
+    <action name="StartTransaction">
+        <java>
+            <configuration>
+                <property>
+                    <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                    <value>true</value>
+                </property>
+            </configuration>
+            <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+            <arg>--action</arg><arg>NEW_VERSION</arg>
+            <arg>--mdStoreID</arg><arg>${mdStoreOutputId}</arg>
+            <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+            <capture-output/>
+        </java>
+        <ok to="CreateEBIDataSet"/>
+        <error to="RollBack"/>
+    </action>
+
+
     <action name="CreateEBIDataSet">
         <spark xmlns="uri:oozie:spark-action:0.2">
             <master>yarn-cluster</master>
@@ -95,11 +107,49 @@
                 ${sparkExtraOPT}
             </spark-opts>
             <arg>--sourcePath</arg><arg>${sourcePath}/ebi_links_dataset</arg>
-            <arg>--targetPath</arg><arg>${targetPath}</arg>
+            <arg>--mdstoreOutputVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
             <arg>--master</arg><arg>yarn</arg>
         </spark>
         <ok to="End"/>
         <error to="Kill"/>
     </action>
+
+
+    <action name="CommitVersion">
+        <java>
+            <configuration>
+                <property>
+                    <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                    <value>true</value>
+                </property>
+            </configuration>
+            <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+            <arg>--action</arg><arg>COMMIT</arg>
+            <arg>--namenode</arg><arg>${nameNode}</arg>
+            <arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
+            <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+        </java>
+        <ok to="End"/>
+        <error to="Kill"/>
+    </action>
+
+    <action name="RollBack">
+        <java>
+            <configuration>
+                <property>
+                    <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                    <value>true</value>
+                </property>
+            </configuration>
+            <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+            <arg>--action</arg><arg>ROLLBACK</arg>
+            <arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
+            <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+        </java>
+        <ok to="Kill"/>
+        <error to="Kill"/>
+    </action>
+
     <end name="End"/>
+
 </workflow-app>

dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/sx/bio/pubmed/oozie_app/workflow.xml

@@ -1,4 +1,4 @@
-    <workflow-app name="Download_Transform_Pubmed_Workflow" xmlns="uri:oozie:workflow:0.5">
+<workflow-app name="Download_Transform_Pubmed_Workflow" xmlns="uri:oozie:workflow:0.5">
     <parameters>
         <property>
             <name>baselineWorkingPath</name>
@@ -9,8 +9,12 @@
             <description>The IS lookUp service endopoint</description>
         </property>
         <property>
-            <name>targetPath</name>
-            <description>The target path</description>
+            <name>mdStoreOutputId</name>
+            <description>the identifier of the cleaned MDStore</description>
+        </property>
+        <property>
+            <name>mdStoreManagerURI</name>
+            <description>the path of the cleaned mdstore</description>
         </property>
         <property>
             <name>skipUpdate</name>
@@ -19,12 +23,31 @@
         </property>
     </parameters>
 
-    <start to="ConvertDataset"/>
+    <start to="StartTransaction"/>
+
 
     <kill name="Kill">
         <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
     </kill>
 
+    <action name="StartTransaction">
+        <java>
+            <configuration>
+                <property>
+                    <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                    <value>true</value>
+                </property>
+            </configuration>
+            <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+            <arg>--action</arg><arg>NEW_VERSION</arg>
+            <arg>--mdStoreID</arg><arg>${mdStoreOutputId}</arg>
+            <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+            <capture-output/>
+        </java>
+        <ok to="ConvertDataset"/>
+        <error to="RollBack"/>
+    </action>
+
     <action name="ConvertDataset">
         <spark xmlns="uri:oozie:spark-action:0.2">
             <master>yarn</master>
@@ -43,16 +66,52 @@
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
             </spark-opts>
             <arg>--workingPath</arg><arg>${baselineWorkingPath}</arg>
-            <arg>--targetPath</arg><arg>${targetPath}</arg>
+            <arg>--mdstoreOutputVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
             <arg>--master</arg><arg>yarn</arg>
             <arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
             <arg>--hdfsServerUri</arg><arg>${nameNode}</arg>
             <arg>--skipUpdate</arg><arg>${skipUpdate}</arg>
         </spark>
+        <ok to="CommitVersion"/>
+        <error to="RollBack"/>
+    </action>
+
+    <action name="CommitVersion">
+        <java>
+            <configuration>
+                <property>
+                    <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                    <value>true</value>
+                </property>
+            </configuration>
+            <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+            <arg>--action</arg><arg>COMMIT</arg>
+            <arg>--namenode</arg><arg>${nameNode}</arg>
+            <arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
+            <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+        </java>
         <ok to="End"/>
         <error to="Kill"/>
     </action>
 
+    <action name="RollBack">
+        <java>
+            <configuration>
+                <property>
+                    <name>oozie.launcher.mapreduce.user.classpath.first</name>
+                    <value>true</value>
+                </property>
+            </configuration>
+            <main-class>eu.dnetlib.dhp.aggregation.mdstore.MDStoreActionNode</main-class>
+            <arg>--action</arg><arg>ROLLBACK</arg>
+            <arg>--mdStoreVersion</arg><arg>${wf:actionData('StartTransaction')['mdStoreVersion']}</arg>
+            <arg>--mdStoreManagerURI</arg><arg>${mdStoreManagerURI}</arg>
+        </java>
+        <ok to="Kill"/>
+        <error to="Kill"/>
+    </action>
+
+
     <end name="End"/>
 
 </workflow-app>

dhp-workflows/dhp-aggregation/src/main/scala/eu/dnetlib/dhp/sx/bio/SparkTransformBioDatabaseToOAF.scala

@@ -2,12 +2,15 @@ package eu.dnetlib.dhp.sx.bio
 
 import eu.dnetlib.dhp.application.ArgumentApplicationParser
 import eu.dnetlib.dhp.collection.CollectionUtils
+import eu.dnetlib.dhp.common.Constants.{MDSTORE_DATA_PATH, MDSTORE_SIZE_PATH}
+import eu.dnetlib.dhp.schema.mdstore.MDStoreVersion
 import eu.dnetlib.dhp.schema.oaf.Oaf
 import eu.dnetlib.dhp.sx.bio.BioDBToOAF.ScholixResolved
 import org.apache.commons.io.IOUtils
 import org.apache.spark.SparkConf
 import org.apache.spark.sql.{Encoder, Encoders, SparkSession}
 import org.slf4j.{Logger, LoggerFactory}
+import eu.dnetlib.dhp.utils.DHPUtils.{MAPPER, writeHdfsFile}
 
 object SparkTransformBioDatabaseToOAF {
 
@@ -25,8 +28,13 @@ object SparkTransformBioDatabaseToOAF {
 
     val dbPath: String = parser.get("dbPath")
     log.info("dbPath: {}", database)
-    val targetPath: String = parser.get("targetPath")
-    log.info("targetPath: {}", database)
+
+    val mdstoreOutputVersion = parser.get("mdstoreOutputVersion")
+    log.info("mdstoreOutputVersion: {}", mdstoreOutputVersion)
+
+    val cleanedMdStoreVersion = MAPPER.readValue(mdstoreOutputVersion, classOf[MDStoreVersion])
+    val outputBasePath = cleanedMdStoreVersion.getHdfsPath
+    log.info("outputBasePath: {}", outputBasePath)
 
     val spark: SparkSession =
       SparkSession
@@ -43,24 +51,28 @@ object SparkTransformBioDatabaseToOAF {
       case "UNIPROT" =>
         CollectionUtils.saveDataset(
           spark.createDataset(sc.textFile(dbPath).flatMap(i => BioDBToOAF.uniprotToOAF(i))),
-          targetPath
+          s"$outputBasePath/$MDSTORE_DATA_PATH"
         )
       case "PDB" =>
         CollectionUtils.saveDataset(
           spark.createDataset(sc.textFile(dbPath).flatMap(i => BioDBToOAF.pdbTOOaf(i))),
-          targetPath
+          s"$outputBasePath/$MDSTORE_DATA_PATH"
         )
       case "SCHOLIX" =>
         CollectionUtils.saveDataset(
           spark.read.load(dbPath).as[ScholixResolved].map(i => BioDBToOAF.scholixResolvedToOAF(i)),
-          targetPath
+          s"$outputBasePath/$MDSTORE_DATA_PATH"
         )
       case "CROSSREF_LINKS" =>
         CollectionUtils.saveDataset(
           spark.createDataset(sc.textFile(dbPath).map(i => BioDBToOAF.crossrefLinksToOaf(i))),
-          targetPath
+          s"$outputBasePath/$MDSTORE_DATA_PATH"
         )
     }
+
+    val df = spark.read.text(s"$outputBasePath/$MDSTORE_DATA_PATH")
+    val mdStoreSize = df.count
+    writeHdfsFile(spark.sparkContext.hadoopConfiguration, s"$mdStoreSize", s"$outputBasePath/$MDSTORE_SIZE_PATH")
   }
 
 }

dhp-workflows/dhp-aggregation/src/main/scala/eu/dnetlib/dhp/sx/bio/ebi/SparkCreateBaselineDataFrame.scala

@@ -2,9 +2,12 @@ package eu.dnetlib.dhp.sx.bio.ebi
 
 import eu.dnetlib.dhp.application.ArgumentApplicationParser
 import eu.dnetlib.dhp.collection.CollectionUtils
+import eu.dnetlib.dhp.common.Constants.{MDSTORE_DATA_PATH, MDSTORE_SIZE_PATH}
 import eu.dnetlib.dhp.common.vocabulary.VocabularyGroup
+import eu.dnetlib.dhp.schema.mdstore.MDStoreVersion
 import eu.dnetlib.dhp.schema.oaf.{Oaf, Result}
 import eu.dnetlib.dhp.sx.bio.pubmed._
+import eu.dnetlib.dhp.utils.DHPUtils.{MAPPER, writeHdfsFile}
 import eu.dnetlib.dhp.utils.ISLookupClientFactory
 import org.apache.commons.io.IOUtils
 import org.apache.hadoop.conf.Configuration
@@ -164,11 +167,15 @@ object SparkCreateBaselineDataFrame {
     val workingPath = parser.get("workingPath")
     log.info("workingPath: {}", workingPath)
 
-    val targetPath = parser.get("targetPath")
-    log.info("targetPath: {}", targetPath)
+    val mdstoreOutputVersion = parser.get("mdstoreOutputVersion")
+    log.info("mdstoreOutputVersion: {}", mdstoreOutputVersion)
+
+    val cleanedMdStoreVersion = MAPPER.readValue(mdstoreOutputVersion, classOf[MDStoreVersion])
+    val outputBasePath = cleanedMdStoreVersion.getHdfsPath
+    log.info("outputBasePath: {}", outputBasePath)
 
     val hdfsServerUri = parser.get("hdfsServerUri")
-    log.info("hdfsServerUri: {}", targetPath)
+    log.info("hdfsServerUri: {}", hdfsServerUri)
 
     val skipUpdate = parser.get("skipUpdate")
     log.info("skipUpdate: {}", skipUpdate)
@@ -216,8 +223,11 @@ object SparkCreateBaselineDataFrame {
         .map(a => PubMedToOaf.convert(a, vocabularies))
         .as[Oaf]
         .filter(p => p != null),
-      targetPath
+      s"$outputBasePath/$MDSTORE_DATA_PATH"
     )
 
+    val df = spark.read.text(s"$outputBasePath/$MDSTORE_DATA_PATH")
+    val mdStoreSize = df.count
+    writeHdfsFile(spark.sparkContext.hadoopConfiguration, s"$mdStoreSize", s"$outputBasePath/$MDSTORE_SIZE_PATH")
   }
 }

dhp-workflows/dhp-aggregation/src/main/scala/eu/dnetlib/dhp/sx/bio/ebi/SparkEBILinksToOaf.scala

@@ -9,6 +9,9 @@ import org.apache.commons.io.IOUtils
 import org.apache.spark.SparkConf
 import org.apache.spark.sql._
 import org.slf4j.{Logger, LoggerFactory}
+import eu.dnetlib.dhp.common.Constants.{MDSTORE_DATA_PATH, MDSTORE_SIZE_PATH}
+import eu.dnetlib.dhp.schema.mdstore.MDStoreVersion
+import eu.dnetlib.dhp.utils.DHPUtils.{MAPPER, writeHdfsFile}
 
 object SparkEBILinksToOaf {
 
@@ -32,8 +35,13 @@ object SparkEBILinksToOaf {
     import spark.implicits._
     val sourcePath = parser.get("sourcePath")
     log.info(s"sourcePath  -> $sourcePath")
-    val targetPath = parser.get("targetPath")
-    log.info(s"targetPath  -> $targetPath")
+    val mdstoreOutputVersion = parser.get("mdstoreOutputVersion")
+    log.info("mdstoreOutputVersion: {}", mdstoreOutputVersion)
+
+    val cleanedMdStoreVersion = MAPPER.readValue(mdstoreOutputVersion, classOf[MDStoreVersion])
+    val outputBasePath = cleanedMdStoreVersion.getHdfsPath
+    log.info("outputBasePath: {}", outputBasePath)
+
     implicit val PMEncoder: Encoder[Oaf] = Encoders.kryo(classOf[Oaf])
 
     val ebLinks: Dataset[EBILinkItem] = spark.read
@@ -46,7 +54,10 @@ object SparkEBILinksToOaf {
         .flatMap(j => BioDBToOAF.parse_ebi_links(j.links))
         .filter(p => BioDBToOAF.EBITargetLinksFilter(p))
         .flatMap(p => BioDBToOAF.convertEBILinksToOaf(p)),
-      targetPath
+      s"$outputBasePath/$MDSTORE_DATA_PATH"
     )
+    val df = spark.read.text(s"$outputBasePath/$MDSTORE_DATA_PATH")
+    val mdStoreSize = df.count
+    writeHdfsFile(spark.sparkContext.hadoopConfiguration, s"$mdStoreSize", s"$outputBasePath/$MDSTORE_SIZE_PATH")
   }
 }

dhp-workflows/dhp-aggregation/src/test/java/eu/dnetlib/dhp/collection/plugin/base/BaseCollectionInfo.java

@@ -0,0 +1,38 @@
+
+package eu.dnetlib.dhp.collection.plugin.base;
+
+import java.io.Serializable;
+
+public class BaseCollectionInfo implements Serializable {
+
+	private static final long serialVersionUID = 5766333937429419647L;
+
+	private String id;
+	private String opendoarId;
+	private String rorId;
+
+	public String getId() {
+		return this.id;
+	}
+
+	public void setId(final String id) {
+		this.id = id;
+	}
+
+	public String getOpendoarId() {
+		return this.opendoarId;
+	}
+
+	public void setOpendoarId(final String opendoarId) {
+		this.opendoarId = opendoarId;
+	}
+
+	public String getRorId() {
+		return this.rorId;
+	}
+
+	public void setRorId(final String rorId) {
+		this.rorId = rorId;
+	}
+
+}

dhp-workflows/dhp-aggregation/src/test/java/eu/dnetlib/dhp/collection/plugin/base/BaseCollectorIteratorTest.java

@@ -0,0 +1,184 @@
+
+package eu.dnetlib.dhp.collection.plugin.base;
+
+import static org.junit.jupiter.api.Assertions.assertEquals;
+
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.LinkedHashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Map.Entry;
+import java.util.Set;
+import java.util.concurrent.atomic.AtomicInteger;
+
+import org.apache.commons.io.IOUtils;
+import org.apache.commons.lang3.StringUtils;
+import org.apache.spark.api.java.JavaRDD;
+import org.apache.spark.api.java.JavaSparkContext;
+import org.apache.spark.sql.Dataset;
+import org.apache.spark.sql.Encoders;
+import org.apache.spark.sql.SparkSession;
+import org.dom4j.Attribute;
+import org.dom4j.Document;
+import org.dom4j.DocumentException;
+import org.dom4j.DocumentHelper;
+import org.dom4j.Element;
+import org.dom4j.Node;
+import org.junit.jupiter.api.Disabled;
+import org.junit.jupiter.api.Test;
+
+import com.fasterxml.jackson.databind.ObjectMapper;
+
+import eu.dnetlib.dhp.common.aggregation.AggregatorReport;
+
+@Disabled
+public class BaseCollectorIteratorTest {
+
+	@Test
+	void testImportFile() throws Exception {
+
+		long count = 0;
+
+		final BaseCollectorIterator iterator = new BaseCollectorIterator("base-sample.tar", new AggregatorReport());
+
+		final Map<String, Map<String, String>> collections = new HashMap<>();
+		final Map<String, AtomicInteger> fields = new HashMap<>();
+		final Set<String> types = new HashSet<>();
+
+		while (iterator.hasNext()) {
+
+			final Document record = DocumentHelper.parseText(iterator.next());
+
+			count++;
+
+			if ((count % 1000) == 0) {
+				System.out.println("# Read records: " + count);
+			}
+
+			// System.out.println(record.asXML());
+
+			for (final Object o : record.selectNodes("//*|//@*")) {
+				final String path = ((Node) o).getPath();
+
+				if (fields.containsKey(path)) {
+					fields.get(path).incrementAndGet();
+				} else {
+					fields.put(path, new AtomicInteger(1));
+				}
+
+				if (o instanceof Element) {
+					final Element n = (Element) o;
+
+					if ("collection".equals(n.getName())) {
+						final String collName = n.getText().trim();
+						if (StringUtils.isNotBlank(collName) && !collections.containsKey(collName)) {
+							final Map<String, String> collAttrs = new HashMap<>();
+							for (final Object ao : n.attributes()) {
+								collAttrs.put(((Attribute) ao).getName(), ((Attribute) ao).getValue());
+							}
+							collections.put(collName, collAttrs);
+						}
+					} else if ("type".equals(n.getName())) {
+						types.add(n.getText().trim());
+					}
+
+				}
+			}
+
+		}
+
+		final ObjectMapper mapper = new ObjectMapper();
+		for (final Entry<String, Map<String, String>> e : collections.entrySet()) {
+			System.out.println(e.getKey() + ": " + mapper.writeValueAsString(e.getValue()));
+
+		}
+
+		for (final Entry<String, AtomicInteger> e : fields.entrySet()) {
+			System.out.println(e.getKey() + ": " + e.getValue().get());
+
+		}
+
+		System.out.println("TYPES: ");
+		for (final String s : types) {
+			System.out.println(s);
+
+		}
+
+		assertEquals(30000, count);
+	}
+
+	@Test
+	public void testParquet() throws Exception {
+
+		final String xml = IOUtils.toString(getClass().getResourceAsStream("record.xml"));
+
+		final SparkSession spark = SparkSession.builder().master("local[*]").getOrCreate();
+
+		final List<BaseRecordInfo> ls = new ArrayList<>();
+
+		for (int i = 0; i < 10; i++) {
+			ls.add(extractInfo(xml));
+		}
+
+		final JavaRDD<BaseRecordInfo> rdd = JavaSparkContext
+			.fromSparkContext(spark.sparkContext())
+			.parallelize(ls);
+
+		final Dataset<BaseRecordInfo> df = spark
+			.createDataset(rdd.rdd(), Encoders.bean(BaseRecordInfo.class));
+
+		df.printSchema();
+
+		df.show(false);
+	}
+
+	private BaseRecordInfo extractInfo(final String s) {
+		try {
+			final Document record = DocumentHelper.parseText(s);
+
+			final BaseRecordInfo info = new BaseRecordInfo();
+
+			final Set<String> paths = new LinkedHashSet<>();
+			final Set<String> types = new LinkedHashSet<>();
+			final List<BaseCollectionInfo> colls = new ArrayList<>();
+
+			for (final Object o : record.selectNodes("//*|//@*")) {
+				paths.add(((Node) o).getPath());
+
+				if (o instanceof Element) {
+					final Element n = (Element) o;
+
+					final String nodeName = n.getName();
+
+					if ("collection".equals(nodeName)) {
+						final String collName = n.getText().trim();
+
+						if (StringUtils.isNotBlank(collName)) {
+							final BaseCollectionInfo coll = new BaseCollectionInfo();
+							coll.setId(collName);
+							coll.setOpendoarId(n.valueOf("@opendoar_id").trim());
+							coll.setRorId(n.valueOf("@ror_id").trim());
+							colls.add(coll);
+						}
+					} else if ("type".equals(nodeName)) {
+						types.add("TYPE: " + n.getText().trim());
+					} else if ("typenorm".equals(nodeName)) {
+						types.add("TYPE_NORM: " + n.getText().trim());
+					}
+				}
+			}
+
+			info.setId(record.valueOf("//*[local-name() = 'header']/*[local-name() = 'identifier']").trim());
+			info.getTypes().addAll(types);
+			info.getPaths().addAll(paths);
+			info.setCollections(colls);
+
+			return info;
+		} catch (final DocumentException e) {
+			throw new RuntimeException(e);
+		}
+	}
+
+}

dhp-workflows/dhp-aggregation/src/test/java/eu/dnetlib/dhp/collection/plugin/base/BaseCollectorPluginTest.java

@@ -0,0 +1,32 @@
+
+package eu.dnetlib.dhp.collection.plugin.base;
+
+import static org.junit.jupiter.api.Assertions.assertFalse;
+import static org.junit.jupiter.api.Assertions.assertTrue;
+
+import java.util.Arrays;
+import java.util.HashSet;
+import java.util.Set;
+
+import org.apache.commons.io.IOUtils;
+import org.junit.jupiter.api.Test;
+
+class BaseCollectorPluginTest {
+
+	@Test
+	void testFilterXml() throws Exception {
+		final String xml = IOUtils.toString(getClass().getResourceAsStream("record.xml"));
+
+		final Set<String> validIds = new HashSet<>(Arrays.asList("opendoar____::1234", "opendoar____::4567"));
+		final Set<String> validTypes = new HashSet<>(Arrays.asList("1", "121"));
+		final Set<String> validTypes2 = new HashSet<>(Arrays.asList("1", "11"));
+
+		assertTrue(BaseCollectorPlugin.filterXml(xml, validIds, validTypes));
+		assertTrue(BaseCollectorPlugin.filterXml(xml, validIds, new HashSet<>()));
+
+		assertFalse(BaseCollectorPlugin.filterXml(xml, new HashSet<>(), validTypes));
+		assertFalse(BaseCollectorPlugin.filterXml(xml, validIds, validTypes2));
+
+	}
+
+}

dhp-workflows/dhp-aggregation/src/test/java/eu/dnetlib/dhp/collection/plugin/base/BaseRecordInfo.java

@@ -0,0 +1,49 @@
+
+package eu.dnetlib.dhp.collection.plugin.base;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.List;
+
+public class BaseRecordInfo implements Serializable {
+
+	private static final long serialVersionUID = -8848232018350074593L;
+
+	private String id;
+	private List<BaseCollectionInfo> collections = new ArrayList<>();
+	private List<String> paths = new ArrayList<>();
+	private List<String> types = new ArrayList<>();
+
+	public String getId() {
+		return this.id;
+	}
+
+	public void setId(final String id) {
+		this.id = id;
+	}
+
+	public List<String> getPaths() {
+		return this.paths;
+	}
+
+	public void setPaths(final List<String> paths) {
+		this.paths = paths;
+	}
+
+	public List<String> getTypes() {
+		return this.types;
+	}
+
+	public void setTypes(final List<String> types) {
+		this.types = types;
+	}
+
+	public List<BaseCollectionInfo> getCollections() {
+		return this.collections;
+	}
+
+	public void setCollections(final List<BaseCollectionInfo> collections) {
+		this.collections = collections;
+	}
+
+}

dhp-workflows/dhp-aggregation/src/test/java/eu/dnetlib/dhp/collection/plugin/base/BaseTransfomationTest.java

@@ -0,0 +1,77 @@
+
+package eu.dnetlib.dhp.collection.plugin.base;
+
+import java.io.IOException;
+
+import org.apache.commons.io.IOUtils;
+import org.apache.spark.SparkConf;
+import org.apache.spark.util.LongAccumulator;
+import org.dom4j.io.SAXReader;
+import org.junit.jupiter.api.BeforeEach;
+import org.junit.jupiter.api.Test;
+import org.junit.jupiter.api.extension.ExtendWith;
+import org.mockito.junit.jupiter.MockitoExtension;
+
+import eu.dnetlib.dhp.aggregation.AbstractVocabularyTest;
+import eu.dnetlib.dhp.aggregation.common.AggregationCounter;
+import eu.dnetlib.dhp.schema.mdstore.MetadataRecord;
+import eu.dnetlib.dhp.schema.mdstore.Provenance;
+import eu.dnetlib.dhp.transformation.xslt.XSLTTransformationFunction;
+import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpException;
+
+// @Disabled
+@ExtendWith(MockitoExtension.class)
+public class BaseTransfomationTest extends AbstractVocabularyTest {
+
+	private SparkConf sparkConf;
+
+	@BeforeEach
+	public void setUp() throws IOException, ISLookUpException {
+		setUpVocabulary();
+
+		this.sparkConf = new SparkConf();
+		this.sparkConf.setMaster("local[*]");
+		this.sparkConf.set("spark.driver.host", "localhost");
+		this.sparkConf.set("spark.ui.enabled", "false");
+	}
+
+	@Test
+	void testBase2ODF() throws Exception {
+
+		final MetadataRecord mr = new MetadataRecord();
+		mr.setProvenance(new Provenance("DSID", "DSNAME", "PREFIX"));
+		mr.setBody(IOUtils.toString(getClass().getResourceAsStream("record.xml")));
+
+		final XSLTTransformationFunction tr = loadTransformationRule("xml/base2odf.transformationRule.xml");
+
+		final MetadataRecord result = tr.call(mr);
+
+		System.out.println(result.getBody());
+	}
+
+	@Test
+	void testBase2OAF() throws Exception {
+
+		final MetadataRecord mr = new MetadataRecord();
+		mr.setProvenance(new Provenance("DSID", "DSNAME", "PREFIX"));
+		mr.setBody(IOUtils.toString(getClass().getResourceAsStream("record.xml")));
+
+		final XSLTTransformationFunction tr = loadTransformationRule("xml/base2oaf.transformationRule.xml");
+
+		final MetadataRecord result = tr.call(mr);
+
+		System.out.println(result.getBody());
+	}
+
+	private XSLTTransformationFunction loadTransformationRule(final String path) throws Exception {
+		final String xslt = new SAXReader()
+				.read(this.getClass().getResourceAsStream(path))
+				.selectSingleNode("//CODE/*")
+				.asXML();
+
+		final LongAccumulator la = new LongAccumulator();
+
+		return new XSLTTransformationFunction(new AggregationCounter(la, la, la), xslt, 0, this.vocabularies);
+	}
+
+}

dhp-workflows/dhp-aggregation/src/test/resources/eu/dnetlib/dhp/collection/plugin/base/record.xml

@@ -0,0 +1,58 @@
+<record>
+  <header xmlns="http://www.openarchives.org/OAI/2.0/">
+    <identifier>ftdoajarticles:oai:doaj.org/article:e2d5b5126b2d4e479933cc7f9a9ae0c1</identifier>
+    <datestamp>2022-12-31T11:48:55Z</datestamp>
+  </header>
+  <metadata xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:base_dc="http://oai.base-search.net/base_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <base_dc:dc xsi:schemaLocation="http://oai.base-search.net/base_dc/ http://oai.base-search.net/base_dc/base_dc.xsd">
+      <base_dc:global_id>ftdoajarticles:oai:doaj.org/article:e2d5b5126b2d4e479933cc7f9a9ae0c1</base_dc:global_id>
+      <base_dc:continent>cww</base_dc:continent>
+      <base_dc:country>org</base_dc:country>
+      <base_dc:collection opendoar_id="1234" ror_id="ror1234">ftdoajarticles</base_dc:collection>
+      <base_dc:collname>TEST REPO</base_dc:collname>
+      <dc:title>Assessment of cultural heritage: the legislative and methodological framework of Russian Federation</dc:title>
+      <dc:creator>ALBU, Svetlana</dc:creator>
+      <dc:creator>LEȘAN, Anna</dc:creator>
+      <dc:subject>architectural heritage</dc:subject>
+      <dc:subject>evaluation of architectural heritage</dc:subject>
+      <dc:subject>types of values</dc:subject>
+      <dc:subject>experience of russian federation</dc:subject>
+      <dc:subject>Social Sciences</dc:subject>
+      <dc:subject>H</dc:subject>
+      <dc:description>Architectural heritage is the real estate inheritance by population of a country becoming an extremely valuable and specific category, preserving and capitalizing on those assets requires considerable effort. The state does not have sufficient means to maintain and preserve cultural heritage, as a result it is included in the civil circuit. The transfer of property right or of some partial rights over the architectural patrimony is accompanied by the necessity to estimate the value of goods. In this article, the authors examine the experience of Russian Federation (one of the largest countries with a huge architectural heritage) on the legislative framework of architectural and methodological heritage of architectural heritage assessment. The particularities of cultural assets valuation compared to other categories of real estate are examined, as well as the methodological aspects (types of values, methods applied in valuation, approaches according to the purpose of valuation) regarding the valuation of real estate with architectural value in Russian Federation.</dc:description>
+      <dc:publisher>Technical University of Moldova</dc:publisher>
+      <dc:date>2020-09-01T00:00:00Z</dc:date>
+      <base_dc:year>2020</base_dc:year>
+      <dc:type>article</dc:type>
+      <base_dc:typenorm>121</base_dc:typenorm>
+      <dc:identifier>https://doi.org/10.5281/zenodo.3971988</dc:identifier>
+      <dc:identifier>https://doaj.org/article/e2d5b5126b2d4e479933cc7f9a9ae0c1</dc:identifier>
+      <base_dc:link>https://doi.org/10.5281/zenodo.3971988</base_dc:link>
+      <dc:source>Journal of Social Sciences, Vol 3, Iss 3, Pp 134-143 (2020)</dc:source>
+      <dc:language>EN</dc:language>
+      <dc:language>FR</dc:language>
+      <dc:language>RO</dc:language>
+      <dc:relation>http://ibn.idsi.md/sites/default/files/imag_file/JSS-3-2020_134-143.pdf</dc:relation>
+      <dc:relation>https://doaj.org/toc/2587-3490</dc:relation>
+      <dc:relation>https://doaj.org/toc/2587-3504</dc:relation>
+      <dc:relation>doi:10.5281/zenodo.3971988</dc:relation>
+      <dc:relation>2587-3490</dc:relation>
+      <dc:relation>2587-3504</dc:relation>
+      <dc:relation>https://doaj.org/article/e2d5b5126b2d4e479933cc7f9a9ae0c1</dc:relation>
+      <base_dc:autoclasscode type="ddc">720</base_dc:autoclasscode>
+      <base_dc:authod_id>
+        <base_dc:creator_name>ALBU, Svetlana</base_dc:creator_name>
+        <base_dc:creator_id>https://orcid.org/0000-0002-8648-950X</base_dc:creator_id>
+      </base_dc:authod_id>
+      <base_dc:authod_id>
+        <base_dc:creator_name>LEȘAN, Anna</base_dc:creator_name>
+        <base_dc:creator_id>https://orcid.org/0000-0003-3284-0525</base_dc:creator_id>
+      </base_dc:authod_id>
+      <base_dc:doi>https://doi.org/10.5281/zenodo.3971988</base_dc:doi>
+      <base_dc:oa>1</base_dc:oa>
+      <base_dc:lang>eng</base_dc:lang>
+      <base_dc:lang>fre</base_dc:lang>
+      <base_dc:lang>rum</base_dc:lang>
+    </base_dc:dc>
+  </metadata>
+</record>

dhp-workflows/dhp-aggregation/src/test/resources/eu/dnetlib/dhp/transform/synonyms.txt

@@ -1496,4 +1496,30 @@ cnr:institutes @=@ __CDS131__ @=@ IBE - Istituto per la BioEconomia
 cnr:institutes @=@ https://ror.org/0263zy895 @=@ CDS132
 cnr:institutes @=@ https://ror.org/0263zy895 @=@ SCITEC - Istituto di Scienze e Tecnologie Chimiche \"Giulio Natta\"
 cnr:institutes @=@ __CDS133__ @=@ CDS133
-cnr:institutes @=@ __CDS133__ @=@ STEMS - Istituto di Scienze e Tecnologie per l'Energia e la Mobilità Sostenibili
+cnr:institutes @=@ __CDS133__ @=@ STEMS - Istituto di Scienze e Tecnologie per l'Energia e la Mobilità Sostenibili
+base:normalized_types @=@ Text @=@ 1
+base:normalized_types @=@ Book @=@ 11
+base:normalized_types @=@ Book part @=@ 111
+base:normalized_types @=@ Journal/Newspaper @=@ 12
+base:normalized_types @=@ Article contribution @=@ 121
+base:normalized_types @=@ Other non-article @=@ 122
+base:normalized_types @=@ Conference object @=@ 13
+base:normalized_types @=@ Report @=@ 14
+base:normalized_types @=@ Review @=@ 15
+base:normalized_types @=@ Course material @=@ 16
+base:normalized_types @=@ Lecture @=@ 17
+base:normalized_types @=@ Thesis @=@ 18
+base:normalized_types @=@ Bachelor's thesis @=@ 181
+base:normalized_types @=@ Master's thesis @=@ 182
+base:normalized_types @=@ Doctoral and postdoctoral thesis @=@ 183
+base:normalized_types @=@ Manuscript @=@ 19
+base:normalized_types @=@ Patent @=@ 1A
+base:normalized_types @=@ Musical notation @=@ 2
+base:normalized_types @=@ Map @=@ 3
+base:normalized_types @=@ Audio @=@ 4
+base:normalized_types @=@ Image/Video @=@ 5
+base:normalized_types @=@ Still image @=@ 51
+base:normalized_types @=@ Moving image/Video @=@ 52
+base:normalized_types @=@ Software @=@ 6
+base:normalized_types @=@ Dataset @=@ 7
+base:normalized_types @=@ Unknown @=@ F

dhp-workflows/dhp-aggregation/src/test/resources/eu/dnetlib/dhp/transform/terms.txt

@@ -1210,4 +1210,29 @@ cnr:institutes @=@ cnr:institutes @=@ __CDS130__ @=@ __CDS130__
 cnr:institutes @=@ cnr:institutes @=@ __CDS131__ @=@ __CDS131__
 cnr:institutes @=@ cnr:institutes @=@ https://ror.org/0263zy895 @=@ https://ror.org/0263zy895
 cnr:institutes @=@ cnr:institutes @=@ __CDS133__ @=@ __CDS133__
-
+base:normalized_types @=@ base:normalized_types @=@ Text @=@ Text
+base:normalized_types @=@ base:normalized_types @=@ Book @=@ Book
+base:normalized_types @=@ base:normalized_types @=@ Book part @=@ Book part
+base:normalized_types @=@ base:normalized_types @=@ Journal/Newspaper @=@ Journal/Newspaper
+base:normalized_types @=@ base:normalized_types @=@ Article contribution @=@ Article contribution
+base:normalized_types @=@ base:normalized_types @=@ Other non-article @=@ Other non-article
+base:normalized_types @=@ base:normalized_types @=@ Conference object @=@ Conference object
+base:normalized_types @=@ base:normalized_types @=@ Report @=@ Report
+base:normalized_types @=@ base:normalized_types @=@ Review @=@ Review
+base:normalized_types @=@ base:normalized_types @=@ Course material @=@ Course material
+base:normalized_types @=@ base:normalized_types @=@ Lecture @=@ Lecture
+base:normalized_types @=@ base:normalized_types @=@ Thesis @=@ Thesis
+base:normalized_types @=@ base:normalized_types @=@ Bachelor's thesis @=@ Bachelor's thesis
+base:normalized_types @=@ base:normalized_types @=@ Master's thesis @=@ Master's thesis
+base:normalized_types @=@ base:normalized_types @=@ Doctoral and postdoctoral thesis @=@ Doctoral and postdoctoral thesis
+base:normalized_types @=@ base:normalized_types @=@ Manuscript @=@ Manuscript
+base:normalized_types @=@ base:normalized_types @=@ Patent @=@ Patent
+base:normalized_types @=@ base:normalized_types @=@ Musical notation @=@ Musical notation
+base:normalized_types @=@ base:normalized_types @=@ Map @=@ Map
+base:normalized_types @=@ base:normalized_types @=@ Audio @=@ Audio
+base:normalized_types @=@ base:normalized_types @=@ Image/Video @=@ Image/Video
+base:normalized_types @=@ base:normalized_types @=@ Still image @=@ Still image
+base:normalized_types @=@ base:normalized_types @=@ Moving image/Video @=@ Moving image/Video
+base:normalized_types @=@ base:normalized_types @=@ Software @=@ Software
+base:normalized_types @=@ base:normalized_types @=@ Dataset @=@ Dataset
+base:normalized_types @=@ base:normalized_types @=@ Unknown @=@ Unknown

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/AbstractSparkAction.java

@@ -101,6 +101,10 @@ abstract class AbstractSparkAction implements Serializable {
 		return SparkSession.builder().config(conf).getOrCreate();
 	}
 
+	protected static SparkSession getSparkWithHiveSession(SparkConf conf) {
+		return SparkSession.builder().enableHiveSupport().config(conf).getOrCreate();
+	}
+
 	protected static <T> void save(Dataset<T> dataset, String outPath, SaveMode mode) {
 		dataset.write().option("compression", "gzip").mode(mode).json(outPath);
 	}

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/DedupRecordFactory.java

@@ -1,128 +1,212 @@
 
 package eu.dnetlib.dhp.oa.dedup;
 
-import java.lang.reflect.InvocationTargetException;
 import java.util.*;
-import java.util.stream.Collectors;
+import java.util.stream.Stream;
 
 import org.apache.commons.beanutils.BeanUtils;
 import org.apache.commons.lang3.StringUtils;
+import org.apache.spark.api.java.function.FlatMapFunction;
 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.SparkSession;
-
-import com.fasterxml.jackson.databind.DeserializationFeature;
-import com.fasterxml.jackson.databind.ObjectMapper;
-import com.google.common.collect.Lists;
+import org.apache.spark.api.java.function.ReduceFunction;
+import org.apache.spark.sql.*;
 
 import eu.dnetlib.dhp.oa.dedup.model.Identifier;
 import eu.dnetlib.dhp.oa.merge.AuthorMerger;
 import eu.dnetlib.dhp.schema.common.ModelSupport;
-import eu.dnetlib.dhp.schema.oaf.*;
+import eu.dnetlib.dhp.schema.oaf.Author;
+import eu.dnetlib.dhp.schema.oaf.DataInfo;
+import eu.dnetlib.dhp.schema.oaf.OafEntity;
+import eu.dnetlib.dhp.schema.oaf.Result;
 import scala.Tuple2;
+import scala.Tuple3;
+import scala.collection.JavaConversions;
 
 public class DedupRecordFactory {
+	public static final class DedupRecordReduceState {
+		public final String dedupId;
 
-	protected static final ObjectMapper OBJECT_MAPPER = new ObjectMapper()
-		.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false);
+		public final ArrayList<String> aliases = new ArrayList<>();
+
+		public final HashSet<String> acceptanceDate = new HashSet<>();
+
+		public OafEntity entity;
+
+		public DedupRecordReduceState(String dedupId, String id, OafEntity entity) {
+			this.dedupId = dedupId;
+			this.entity = entity;
+			if (entity == null) {
+				aliases.add(id);
+			} else {
+				if (Result.class.isAssignableFrom(entity.getClass())) {
+					Result result = (Result) entity;
+					if (result.getDateofacceptance() != null
+						&& StringUtils.isNotBlank(result.getDateofacceptance().getValue())) {
+						acceptanceDate.add(result.getDateofacceptance().getValue());
+					}
+				}
+			}
+		}
+
+		public String getDedupId() {
+			return dedupId;
+		}
+	}
+
+	private static final int MAX_ACCEPTANCE_DATE = 20;
 
 	private DedupRecordFactory() {
 	}
 
-	public static <T extends OafEntity> Dataset<T> createDedupRecord(
+	public static Dataset<OafEntity> createDedupRecord(
 		final SparkSession spark,
 		final DataInfo dataInfo,
 		final String mergeRelsInputPath,
 		final String entitiesInputPath,
-		final Class<T> clazz) {
+		final Class<OafEntity> clazz) {
 
-		long ts = System.currentTimeMillis();
+		final long ts = System.currentTimeMillis();
+		final Encoder<OafEntity> beanEncoder = Encoders.bean(clazz);
+		final Encoder<OafEntity> kryoEncoder = Encoders.kryo(clazz);
 
 		// <id, json_entity>
-		Dataset<Tuple2<String, T>> entities = spark
+		Dataset<Row> entities = spark
 			.read()
-			.textFile(entitiesInputPath)
+			.schema(Encoders.bean(clazz).schema())
+			.json(entitiesInputPath)
+			.as(beanEncoder)
 			.map(
-				(MapFunction<String, Tuple2<String, T>>) it -> {
-					T entity = OBJECT_MAPPER.readValue(it, clazz);
+				(MapFunction<OafEntity, Tuple2<String, OafEntity>>) entity -> {
 					return new Tuple2<>(entity.getId(), entity);
 				},
-				Encoders.tuple(Encoders.STRING(), Encoders.kryo(clazz)));
+				Encoders.tuple(Encoders.STRING(), kryoEncoder))
+			.selectExpr("_1 AS id", "_2 AS kryoObject");
 
 		// <source, target>: source is the dedup_id, target is the id of the mergedIn
-		Dataset<Tuple2<String, String>> mergeRels = spark
+		Dataset<Row> mergeRels = spark
 			.read()
 			.load(mergeRelsInputPath)
-			.as(Encoders.bean(Relation.class))
 			.where("relClass == 'merges'")
-			.map(
-				(MapFunction<Relation, Tuple2<String, String>>) r -> new Tuple2<>(r.getSource(), r.getTarget()),
-				Encoders.tuple(Encoders.STRING(), Encoders.STRING()));
+			.selectExpr("source as dedupId", "target as id");
 
 		return mergeRels
-			.joinWith(entities, mergeRels.col("_2").equalTo(entities.col("_1")), "inner")
+			.join(entities, JavaConversions.asScalaBuffer(Collections.singletonList("id")), "left")
+			.select("dedupId", "id", "kryoObject")
+			.as(Encoders.tuple(Encoders.STRING(), Encoders.STRING(), kryoEncoder))
 			.map(
-				(MapFunction<Tuple2<Tuple2<String, String>, Tuple2<String, T>>, Tuple2<String, T>>) value -> new Tuple2<>(
-					value._1()._1(), value._2()._2()),
-				Encoders.tuple(Encoders.STRING(), Encoders.kryo(clazz)))
+				(MapFunction<Tuple3<String, String, OafEntity>, DedupRecordReduceState>) t -> new DedupRecordReduceState(
+					t._1(), t._2(), t._3()),
+				Encoders.kryo(DedupRecordReduceState.class))
 			.groupByKey(
-				(MapFunction<Tuple2<String, T>, String>) Tuple2::_1, Encoders.STRING())
-			.mapGroups(
-				(MapGroupsFunction<String, Tuple2<String, T>, T>) (key,
-					values) -> entityMerger(key, values, ts, dataInfo, clazz),
-				Encoders.bean(clazz));
+				(MapFunction<DedupRecordReduceState, String>) DedupRecordReduceState::getDedupId, Encoders.STRING())
+			.reduceGroups(
+				(ReduceFunction<DedupRecordReduceState>) (t1, t2) -> {
+					if (t1.entity == null) {
+						t2.aliases.addAll(t1.aliases);
+						return t2;
+					}
+					if (t1.acceptanceDate.size() < MAX_ACCEPTANCE_DATE) {
+						t1.acceptanceDate.addAll(t2.acceptanceDate);
+					}
+					t1.aliases.addAll(t2.aliases);
+					t1.entity = reduceEntity(t1.entity, t2.entity);
+
+					return t1;
+				})
+			.flatMap((FlatMapFunction<Tuple2<String, DedupRecordReduceState>, OafEntity>) t -> {
+				String dedupId = t._1();
+				DedupRecordReduceState agg = t._2();
+
+				if (agg.acceptanceDate.size() >= MAX_ACCEPTANCE_DATE) {
+					return Collections.emptyIterator();
+				}
+
+				return Stream
+					.concat(
+						Stream
+							.of(agg.getDedupId())
+							.map(id -> createDedupOafEntity(id, agg.entity, dataInfo, ts)),
+						agg.aliases
+							.stream()
+							.map(id -> createMergedDedupAliasOafEntity(id, agg.entity, dataInfo, ts)))
+					.iterator();
+			}, beanEncoder);
 	}
 
-	public static <T extends OafEntity> T entityMerger(
-		String id, Iterator<Tuple2<String, T>> entities, long ts, DataInfo dataInfo, Class<T> clazz)
-		throws IllegalAccessException, InstantiationException, InvocationTargetException {
+	private static OafEntity createDedupOafEntity(String id, OafEntity base, DataInfo dataInfo, long ts) {
+		try {
+			OafEntity res = (OafEntity) BeanUtils.cloneBean(base);
+			res.setId(id);
+			res.setDataInfo(dataInfo);
+			res.setLastupdatetimestamp(ts);
+			return res;
+		} catch (Exception e) {
+			throw new RuntimeException(e);
+		}
+	}
 
-		final Comparator<Identifier<T>> idComparator = new IdentifierComparator<>();
+	private static OafEntity createMergedDedupAliasOafEntity(String id, OafEntity base, DataInfo dataInfo, long ts) {
+		try {
+			OafEntity res = createDedupOafEntity(id, base, dataInfo, ts);
+			DataInfo ds = (DataInfo) BeanUtils.cloneBean(dataInfo);
+			ds.setDeletedbyinference(true);
+			res.setDataInfo(ds);
+			return res;
+		} catch (Exception e) {
+			throw new RuntimeException(e);
+		}
+	}
 
-		final LinkedList<T> entityList = Lists
-			.newArrayList(entities)
-			.stream()
-			.map(t -> Identifier.newInstance(t._2()))
-			.sorted(idComparator)
-			.map(Identifier::getEntity)
-			.collect(Collectors.toCollection(LinkedList::new));
+	private static OafEntity reduceEntity(OafEntity entity, OafEntity duplicate) {
 
-		final T entity = clazz.newInstance();
-		final T first = entityList.removeFirst();
-
-		BeanUtils.copyProperties(entity, first);
-
-		final List<List<Author>> authors = Lists.newArrayList();
-
-		entityList
-			.forEach(
-				duplicate -> {
-					entity.mergeFrom(duplicate);
-					if (ModelSupport.isSubClass(duplicate, Result.class)) {
-						Result r1 = (Result) duplicate;
-						Optional
-							.ofNullable(r1.getAuthor())
-							.ifPresent(a -> authors.add(a));
-					}
-				});
-
-		// set authors and date
-		if (ModelSupport.isSubClass(entity, Result.class)) {
-			Optional
-				.ofNullable(((Result) entity).getAuthor())
-				.ifPresent(a -> authors.add(a));
-
-			((Result) entity).setAuthor(AuthorMerger.merge(authors));
+		if (duplicate == null) {
+			return entity;
 		}
 
-		entity.setId(id);
+		int compare = new IdentifierComparator<>()
+			.compare(Identifier.newInstance(entity), Identifier.newInstance(duplicate));
 
-		entity.setLastupdatetimestamp(ts);
-		entity.setDataInfo(dataInfo);
+		if (compare > 0) {
+			OafEntity swap = duplicate;
+			duplicate = entity;
+			entity = swap;
+		}
+
+		entity.mergeFrom(duplicate);
+
+		if (ModelSupport.isSubClass(duplicate, Result.class)) {
+			Result re = (Result) entity;
+			Result rd = (Result) duplicate;
+
+			List<List<Author>> authors = new ArrayList<>();
+			if (re.getAuthor() != null) {
+				authors.add(re.getAuthor());
+			}
+			if (rd.getAuthor() != null) {
+				authors.add(rd.getAuthor());
+			}
+
+			re.setAuthor(AuthorMerger.merge(authors));
+		}
 
 		return entity;
 	}
 
+	public static <T extends OafEntity> T entityMerger(
+		String id, Iterator<Tuple2<String, T>> entities, long ts, DataInfo dataInfo, Class<T> clazz) {
+		T base = entities.next()._2();
+
+		while (entities.hasNext()) {
+			T duplicate = entities.next()._2();
+			if (duplicate != null)
+				base = (T) reduceEntity(base, duplicate);
+		}
+
+		base.setId(id);
+		base.setDataInfo(dataInfo);
+		base.setLastupdatetimestamp(ts);
+
+		return base;
+	}
+
 }

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/IdGenerator.java

@@ -1,6 +1,7 @@
 
 package eu.dnetlib.dhp.oa.dedup;
 
+import static eu.dnetlib.dhp.utils.DHPUtils.md5;
 import static org.apache.commons.lang3.StringUtils.substringAfter;
 import static org.apache.commons.lang3.StringUtils.substringBefore;
 
@@ -14,33 +15,36 @@ import eu.dnetlib.dhp.schema.oaf.utils.PidType;
 public class IdGenerator implements Serializable {
 
 	// pick the best pid from the list (consider date and pidtype)
-	public static <T extends OafEntity> String generate(List<Identifier<T>> pids, String defaultID) {
+	public static <T extends OafEntity> String generate(List<? extends Identifier> pids, String defaultID) {
 		if (pids == null || pids.isEmpty())
 			return defaultID;
 
 		return generateId(pids);
 	}
 
-	private static <T extends OafEntity> String generateId(List<Identifier<T>> pids) {
-		Identifier<T> bp = pids
+	private static String generateId(List<? extends Identifier> pids) {
+		Identifier bp = pids
 			.stream()
 			.min(Identifier::compareTo)
 			.orElseThrow(() -> new IllegalStateException("unable to generate id"));
 
-		String prefix = substringBefore(bp.getOriginalID(), "|");
-		String ns = substringBefore(substringAfter(bp.getOriginalID(), "|"), "::");
-		String suffix = substringAfter(bp.getOriginalID(), "::");
+		return generate(bp.getOriginalID());
+	}
+
+	public static String generate(String originalId) {
+		String prefix = substringBefore(originalId, "|");
+		String ns = substringBefore(substringAfter(originalId, "|"), "::");
+		String suffix = substringAfter(originalId, "::");
 
 		final String pidType = substringBefore(ns, "_");
 		if (PidType.isValid(pidType)) {
 			return prefix + "|" + dedupify(ns) + "::" + suffix;
 		} else {
-			return prefix + "|dedup_wf_001::" + suffix;
+			return prefix + "|dedup_wf_002::" + md5(originalId); // hash the whole originalId to avoid collisions
 		}
 	}
 
 	private static String dedupify(String ns) {
-
 		StringBuilder prefix;
 		if (PidType.valueOf(substringBefore(ns, "_")) == PidType.openorgs) {
 			prefix = new StringBuilder(substringBefore(ns, "_"));
@@ -53,5 +57,4 @@ public class IdGenerator implements Serializable {
 		}
 		return prefix.substring(0, 12);
 	}
-
 }

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkCreateMergeRels.java

@@ -3,49 +3,47 @@ package eu.dnetlib.dhp.oa.dedup;
 
 import static eu.dnetlib.dhp.schema.common.ModelConstants.DNET_PROVENANCE_ACTIONS;
 import static eu.dnetlib.dhp.schema.common.ModelConstants.PROVENANCE_DEDUP;
+import static org.apache.spark.sql.functions.*;
 
 import java.io.IOException;
-import java.util.*;
-import java.util.stream.Collectors;
+import java.time.LocalDate;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Optional;
 
 import org.apache.commons.io.IOUtils;
+import org.apache.commons.lang3.StringUtils;
 import org.apache.spark.SparkConf;
-import org.apache.spark.api.java.JavaPairRDD;
-import org.apache.spark.api.java.JavaSparkContext;
 import org.apache.spark.api.java.function.FlatMapFunction;
-import org.apache.spark.api.java.function.MapFunction;
-import org.apache.spark.api.java.function.MapGroupsFunction;
-import org.apache.spark.graphx.Edge;
-import org.apache.spark.rdd.RDD;
+import org.apache.spark.sql.*;
 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.catalyst.encoders.RowEncoder;
+import org.apache.spark.sql.expressions.UserDefinedFunction;
+import org.apache.spark.sql.expressions.Window;
+import org.apache.spark.sql.expressions.WindowSpec;
+import org.apache.spark.sql.types.DataTypes;
+import org.apache.spark.sql.types.StructType;
 import org.dom4j.DocumentException;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 import org.xml.sax.SAXException;
 
-import com.google.common.collect.Lists;
 import com.google.common.hash.Hashing;
+import com.kwartile.lib.cc.ConnectedComponent;
 
 import eu.dnetlib.dhp.application.ArgumentApplicationParser;
-import eu.dnetlib.dhp.oa.dedup.graph.ConnectedComponent;
-import eu.dnetlib.dhp.oa.dedup.graph.GraphProcessor;
-import eu.dnetlib.dhp.oa.dedup.model.Identifier;
 import eu.dnetlib.dhp.schema.common.EntityType;
 import eu.dnetlib.dhp.schema.common.ModelConstants;
 import eu.dnetlib.dhp.schema.common.ModelSupport;
-import eu.dnetlib.dhp.schema.oaf.DataInfo;
-import eu.dnetlib.dhp.schema.oaf.OafEntity;
-import eu.dnetlib.dhp.schema.oaf.Qualifier;
-import eu.dnetlib.dhp.schema.oaf.Relation;
+import eu.dnetlib.dhp.schema.oaf.*;
+import eu.dnetlib.dhp.schema.oaf.utils.PidType;
 import eu.dnetlib.dhp.utils.ISLookupClientFactory;
 import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpException;
 import eu.dnetlib.enabling.is.lookup.rmi.ISLookUpService;
 import eu.dnetlib.pace.config.DedupConfig;
-import eu.dnetlib.pace.util.MapDocumentUtil;
-import scala.Tuple2;
+import scala.Tuple3;
+import scala.collection.JavaConversions;
 
 public class SparkCreateMergeRels extends AbstractSparkAction {
 
@@ -68,10 +66,12 @@ public class SparkCreateMergeRels extends AbstractSparkAction {
 		log.info("isLookupUrl {}", isLookUpUrl);
 
 		SparkConf conf = new SparkConf();
+		conf.set("hive.metastore.uris", parser.get("hiveMetastoreUris"));
 		conf.set("spark.serializer", "org.apache.spark.serializer.KryoSerializer");
+
 		conf.registerKryoClasses(ModelSupport.getOafModelClasses());
 
-		new SparkCreateMergeRels(parser, getSparkSession(conf))
+		new SparkCreateMergeRels(parser, getSparkWithHiveSession(conf))
 			.run(ISLookupClientFactory.getLookUpService(isLookUpUrl));
 	}
 
@@ -87,14 +87,15 @@ public class SparkCreateMergeRels extends AbstractSparkAction {
 			.ofNullable(parser.get("cutConnectedComponent"))
 			.map(Integer::valueOf)
 			.orElse(0);
+
+		final String pivotHistoryDatabase = parser.get("pivotHistoryDatabase");
+
 		log.info("connected component cut: '{}'", cut);
 		log.info("graphBasePath: '{}'", graphBasePath);
 		log.info("isLookUpUrl:   '{}'", isLookUpUrl);
 		log.info("actionSetId:   '{}'", actionSetId);
 		log.info("workingPath:   '{}'", workingPath);
 
-		final JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
-
 		for (DedupConfig dedupConf : getConfigurations(isLookUpService, actionSetId)) {
 			final String subEntity = dedupConf.getWf().getSubEntityValue();
 			final Class<OafEntity> clazz = ModelSupport.entityTypes.get(EntityType.valueOf(subEntity));
@@ -106,113 +107,173 @@ public class SparkCreateMergeRels extends AbstractSparkAction {
 
 			final String mergeRelPath = DedupUtility.createMergeRelPath(workingPath, actionSetId, subEntity);
 
-			// <hash(id), id>
-			JavaPairRDD<Object, String> vertexes = createVertexes(sc, graphBasePath, subEntity, dedupConf);
-
-			final RDD<Edge<String>> edgeRdd = spark
+			final Dataset<Row> simRels = spark
 				.read()
 				.load(DedupUtility.createSimRelPath(workingPath, actionSetId, subEntity))
-				.as(Encoders.bean(Relation.class))
-				.javaRDD()
-				.map(it -> new Edge<>(hash(it.getSource()), hash(it.getTarget()), it.getRelClass()))
-				.rdd();
+				.select("source", "target");
 
-			Dataset<Tuple2<String, String>> rawMergeRels = spark
-				.createDataset(
-					GraphProcessor
-						.findCCs(vertexes.rdd(), edgeRdd, maxIterations, cut)
-						.toJavaRDD()
-						.filter(k -> k.getIds().size() > 1)
-						.flatMap(this::ccToRels)
-						.rdd(),
-					Encoders.tuple(Encoders.STRING(), Encoders.STRING()));
+			UserDefinedFunction hashUDF = functions
+				.udf(
+					(String s) -> hash(s), DataTypes.LongType);
 
-			Dataset<Tuple2<String, OafEntity>> entities = spark
+			// <hash(id), id>
+			Dataset<Row> vertexIdMap = simRels
+				.selectExpr("source as id")
+				.union(simRels.selectExpr("target as id"))
+				.distinct()
+				.withColumn("vertexId", hashUDF.apply(functions.col("id")));
+
+			// transform simrels into pairs of numeric ids
+			final Dataset<Row> edges = spark
 				.read()
-				.textFile(DedupUtility.createEntityPath(graphBasePath, subEntity))
-				.map(
-					(MapFunction<String, Tuple2<String, OafEntity>>) it -> {
-						OafEntity entity = OBJECT_MAPPER.readValue(it, clazz);
-						return new Tuple2<>(entity.getId(), entity);
-					},
-					Encoders.tuple(Encoders.STRING(), Encoders.kryo(clazz)));
+				.load(DedupUtility.createSimRelPath(workingPath, actionSetId, subEntity))
+				.select("source", "target")
+				.withColumn("source", hashUDF.apply(functions.col("source")))
+				.withColumn("target", hashUDF.apply(functions.col("target")));
 
-			Dataset<Relation> mergeRels = rawMergeRels
-				.joinWith(entities, rawMergeRels.col("_2").equalTo(entities.col("_1")), "inner")
-				// <tmp_source,target>,<target,entity>
-				.map(
-					(MapFunction<Tuple2<Tuple2<String, String>, Tuple2<String, OafEntity>>, Tuple2<String, OafEntity>>) value -> new Tuple2<>(
-						value._1()._1(), value._2()._2()),
-					Encoders.tuple(Encoders.STRING(), Encoders.kryo(clazz)))
-				// <tmp_source,entity>
-				.groupByKey(
-					(MapFunction<Tuple2<String, OafEntity>, String>) Tuple2::_1, Encoders.STRING())
-				.mapGroups(
-					(MapGroupsFunction<String, Tuple2<String, OafEntity>, ConnectedComponent>) this::generateID,
-					Encoders.bean(ConnectedComponent.class))
-				// <root_id, list(target)>
+			// resolve connected components
+			// ("vertexId", "groupId")
+			Dataset<Row> cliques = ConnectedComponent
+				.runOnPairs(edges, 50, spark);
+
+			// transform "vertexId" back to its original string value
+			// groupId is kept numeric as its string value is not used
+			// ("id", "groupId")
+			Dataset<Row> rawMergeRels = cliques
+				.join(vertexIdMap, JavaConversions.asScalaBuffer(Collections.singletonList("vertexId")), "inner")
+				.drop("vertexId")
+				.distinct();
+
+			// empty dataframe if historydatabase is not used
+			Dataset<Row> pivotHistory = spark
+				.createDataset(
+					Collections.emptyList(),
+					RowEncoder
+						.apply(StructType.fromDDL("id STRING, lastUsage STRING")));
+
+			if (StringUtils.isNotBlank(pivotHistoryDatabase)) {
+				pivotHistory = spark
+					.read()
+					.table(pivotHistoryDatabase + "." + subEntity)
+					.selectExpr("id", "lastUsage");
+			}
+
+			// depending on resulttype collectefrom and dateofacceptance are evaluated differently
+			String collectedfromExpr = "false AS collectedfrom";
+			String dateExpr = "'' AS date";
+
+			if (Result.class.isAssignableFrom(clazz)) {
+				if (Publication.class.isAssignableFrom(clazz)) {
+					collectedfromExpr = "array_contains(collectedfrom.key, '" + ModelConstants.CROSSREF_ID
+						+ "') AS collectedfrom";
+				} else if (eu.dnetlib.dhp.schema.oaf.Dataset.class.isAssignableFrom(clazz)) {
+					collectedfromExpr = "array_contains(collectedfrom.key, '" + ModelConstants.DATACITE_ID
+						+ "') AS collectedfrom";
+				}
+
+				dateExpr = "dateofacceptance.value AS date";
+			}
+
+			// cap pidType at w3id as from there on they are considered equal
+			UserDefinedFunction mapPid = udf(
+				(String s) -> Math.min(PidType.tryValueOf(s).ordinal(), PidType.w3id.ordinal()), DataTypes.IntegerType);
+
+			UserDefinedFunction validDate = udf((String date) -> {
+				if (StringUtils.isNotBlank(date)
+					&& date.matches(DatePicker.DATE_PATTERN) && DatePicker.inRange(date)) {
+					return date;
+				}
+				return LocalDate.now().plusWeeks(1).toString();
+			}, DataTypes.StringType);
+
+			Dataset<Row> pivotingData = spark
+				.read()
+				.schema(Encoders.bean(clazz).schema())
+				.json(DedupUtility.createEntityPath(graphBasePath, subEntity))
+				.selectExpr(
+					"id",
+					"regexp_extract(id, '^\\\\d+\\\\|([^_]+).*::', 1) AS pidType",
+					collectedfromExpr,
+					dateExpr)
+				.withColumn("pidType", mapPid.apply(col("pidType"))) // ordinal of pid type
+				.withColumn("date", validDate.apply(col("date")));
+
+			// ordering to selected pivot id
+			WindowSpec w = Window
+				.partitionBy("groupId")
+				.orderBy(
+					col("lastUsage").desc_nulls_last(),
+					col("pidType").asc_nulls_last(),
+					col("collectedfrom").desc_nulls_last(),
+					col("date").asc_nulls_last(),
+					col("id").asc_nulls_last());
+
+			Dataset<Relation> output = rawMergeRels
+				.join(pivotHistory, JavaConversions.asScalaBuffer(Collections.singletonList("id")), "full")
+				.join(pivotingData, JavaConversions.asScalaBuffer(Collections.singletonList("id")), "left")
+				.withColumn("pivot", functions.first("id").over(w))
+				.withColumn("position", functions.row_number().over(w))
 				.flatMap(
-					(FlatMapFunction<ConnectedComponent, Relation>) cc -> ccToMergeRel(cc, dedupConf),
-					Encoders.bean(Relation.class));
+					(FlatMapFunction<Row, Tuple3<String, String, String>>) (Row r) -> {
+						String id = r.getAs("id");
+						String dedupId = IdGenerator.generate(id);
 
-			saveParquet(mergeRels, mergeRelPath, SaveMode.Overwrite);
+						String pivot = r.getAs("pivot");
+						String pivotDedupId = IdGenerator.generate(pivot);
 
+						// filter out id == pivotDedupId
+						// those are caused by claim expressed on pivotDedupId
+						// information will be merged after creating deduprecord
+						if (id.equals(pivotDedupId)) {
+							return Collections.emptyIterator();
+						}
+
+						ArrayList<Tuple3<String, String, String>> res = new ArrayList<>();
+
+						// singleton pivots have null groupId as they do not match rawMergeRels
+						if (r.isNullAt(r.fieldIndex("groupId"))) {
+							// the record is existing if it matches pivotingData
+							if (!r.isNullAt(r.fieldIndex("collectedfrom"))) {
+								// create relation with old dedup id
+								res.add(new Tuple3<>(id, dedupId, null));
+							}
+							return res.iterator();
+						}
+
+						// this was a pivot in a previous graph but it has been merged into a new group with different
+						// pivot
+						if (!r.isNullAt(r.fieldIndex("lastUsage")) && !pivot.equals(id)
+							&& !dedupId.equals(pivotDedupId)) {
+							// materialize the previous dedup record as a merge relation with the new one
+							res.add(new Tuple3<>(dedupId, pivotDedupId, null));
+						}
+
+						// add merge relations
+						if (cut <= 0 || r.<Integer> getAs("position") <= cut) {
+							res.add(new Tuple3<>(id, pivotDedupId, pivot));
+						}
+
+						return res.iterator();
+					}, Encoders.tuple(Encoders.STRING(), Encoders.STRING(), Encoders.STRING()))
+				.distinct()
+				.flatMap(
+					(FlatMapFunction<Tuple3<String, String, String>, Relation>) (Tuple3<String, String, String> r) -> {
+						String id = r._1();
+						String dedupId = r._2();
+						String pivot = r._3();
+
+						ArrayList<Relation> res = new ArrayList<>();
+						res.add(rel(pivot, dedupId, id, ModelConstants.MERGES, dedupConf));
+						res.add(rel(pivot, id, dedupId, ModelConstants.IS_MERGED_IN, dedupConf));
+
+						return res.iterator();
+					}, Encoders.bean(Relation.class));
+
+			saveParquet(output, mergeRelPath, SaveMode.Overwrite);
 		}
 	}
 
-	private <T extends OafEntity> ConnectedComponent generateID(String key, Iterator<Tuple2<String, T>> values) {
-
-		List<Identifier<T>> identifiers = Lists
-			.newArrayList(values)
-			.stream()
-			.map(v -> Identifier.newInstance(v._2()))
-			.collect(Collectors.toList());
-
-		String rootID = IdGenerator.generate(identifiers, key);
-
-		if (Objects.equals(rootID, key))
-			throw new IllegalStateException("generated default ID: " + rootID);
-
-		return new ConnectedComponent(rootID,
-			identifiers.stream().map(i -> i.getEntity().getId()).collect(Collectors.toSet()));
-	}
-
-	private JavaPairRDD<Object, String> createVertexes(JavaSparkContext sc, String graphBasePath, String subEntity,
-		DedupConfig dedupConf) {
-
-		return sc
-			.textFile(DedupUtility.createEntityPath(graphBasePath, subEntity))
-			.mapToPair(json -> {
-				String id = MapDocumentUtil.getJPathString(dedupConf.getWf().getIdPath(), json);
-				return new Tuple2<>(hash(id), id);
-			});
-	}
-
-	private Iterator<Tuple2<String, String>> ccToRels(ConnectedComponent cc) {
-		return cc
-			.getIds()
-			.stream()
-			.map(id -> new Tuple2<>(cc.getCcId(), id))
-			.iterator();
-	}
-
-	private Iterator<Relation> ccToMergeRel(ConnectedComponent cc, DedupConfig dedupConf) {
-		return cc
-			.getIds()
-			.stream()
-			.flatMap(
-				id -> {
-					List<Relation> tmp = new ArrayList<>();
-
-					tmp.add(rel(cc.getCcId(), id, ModelConstants.MERGES, dedupConf));
-					tmp.add(rel(id, cc.getCcId(), ModelConstants.IS_MERGED_IN, dedupConf));
-
-					return tmp.stream();
-				})
-			.iterator();
-	}
-
-	private Relation rel(String source, String target, String relClass, DedupConfig dedupConf) {
+	private static Relation rel(String pivot, String source, String target, String relClass, DedupConfig dedupConf) {
 
 		String entityType = dedupConf.getWf().getEntityType();
 
@@ -238,6 +299,14 @@ public class SparkCreateMergeRels extends AbstractSparkAction {
 		// TODO calculate the trust value based on the similarity score of the elements in the CC
 
 		r.setDataInfo(info);
+
+		if (pivot != null) {
+			KeyValue pivotKV = new KeyValue();
+			pivotKV.setKey("pivot");
+			pivotKV.setValue(pivot);
+
+			r.setProperties(Arrays.asList(pivotKV));
+		}
 		return r;
 	}
 

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/SparkWhitelistSimRels.java

@@ -91,18 +91,12 @@ public class SparkWhitelistSimRels extends AbstractSparkAction {
 			Dataset<Row> entities = spark
 				.read()
 				.textFile(DedupUtility.createEntityPath(graphBasePath, subEntity))
-				.repartition(numPartitions)
-				.withColumn("id", functions.get_json_object(new Column("value"), dedupConf.getWf().getIdPath()));
+				.select(functions.get_json_object(new Column("value"), dedupConf.getWf().getIdPath()).as("id"))
+				.distinct();
 
-			Dataset<Row> whiteListRels1 = whiteListRels
-				.join(entities, entities.col("id").equalTo(whiteListRels.col("from")), "inner")
-				.select("from", "to");
-
-			Dataset<Row> whiteListRels2 = whiteListRels1
-				.join(entities, whiteListRels1.col("to").equalTo(entities.col("id")), "inner")
-				.select("from", "to");
-
-			Dataset<Relation> whiteListSimRels = whiteListRels2
+			Dataset<Relation> whiteListSimRels = whiteListRels
+				.join(entities, entities.col("id").equalTo(whiteListRels.col("from")), "leftsemi")
+				.join(entities, functions.col("to").equalTo(entities.col("id")), "leftsemi")
 				.map(
 					(MapFunction<Row, Relation>) r -> DedupUtility
 						.createSimRel(r.getString(0), r.getString(1), entity),

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/graph/ConnectedComponent.java

@@ -1,100 +0,0 @@
-
-package eu.dnetlib.dhp.oa.dedup.graph;
-
-import java.io.IOException;
-import java.io.Serializable;
-import java.util.Set;
-import java.util.stream.Collectors;
-
-import org.apache.commons.lang3.StringUtils;
-import org.codehaus.jackson.annotate.JsonIgnore;
-
-import com.fasterxml.jackson.databind.ObjectMapper;
-
-import eu.dnetlib.dhp.utils.DHPUtils;
-import eu.dnetlib.pace.util.PaceException;
-
-public class ConnectedComponent implements Serializable {
-
-	private String ccId;
-	private Set<String> ids;
-
-	private static final String CONNECTED_COMPONENT_ID_PREFIX = "connect_comp";
-
-	public ConnectedComponent(Set<String> ids, final int cut) {
-		this.ids = ids;
-
-		this.ccId = createDefaultID();
-
-		if (cut > 0 && ids.size() > cut) {
-			this.ids = ids
-				.stream()
-				.filter(id -> !ccId.equalsIgnoreCase(id))
-				.limit(cut - 1)
-				.collect(Collectors.toSet());
-//			this.ids.add(ccId); ??
-		}
-	}
-
-	public ConnectedComponent(String ccId, Set<String> ids) {
-		this.ccId = ccId;
-		this.ids = ids;
-	}
-
-	public String createDefaultID() {
-		if (ids.size() > 1) {
-			final String s = getMin();
-			String prefix = s.split("\\|")[0];
-			ccId = prefix + "|" + CONNECTED_COMPONENT_ID_PREFIX + "::" + DHPUtils.md5(s);
-			return ccId;
-		} else {
-			return ids.iterator().next();
-		}
-	}
-
-	@JsonIgnore
-	public String getMin() {
-
-		final StringBuilder min = new StringBuilder();
-
-		ids
-			.forEach(
-				id -> {
-					if (StringUtils.isBlank(min.toString())) {
-						min.append(id);
-					} else {
-						if (min.toString().compareTo(id) > 0) {
-							min.setLength(0);
-							min.append(id);
-						}
-					}
-				});
-		return min.toString();
-	}
-
-	@Override
-	public String toString() {
-		ObjectMapper mapper = new ObjectMapper();
-		try {
-			return mapper.writeValueAsString(this);
-		} catch (IOException e) {
-			throw new PaceException("Failed to create Json: ", e);
-		}
-	}
-
-	public Set<String> getIds() {
-		return ids;
-	}
-
-	public void setIds(Set<String> ids) {
-		this.ids = ids;
-	}
-
-	public String getCcId() {
-		return ccId;
-	}
-
-	public void setCcId(String ccId) {
-		this.ccId = ccId;
-	}
-}

dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/graph/GraphProcessor.scala

@@ -1,37 +0,0 @@
-package eu.dnetlib.dhp.oa.dedup.graph
-
-import org.apache.spark.graphx._
-import org.apache.spark.rdd.RDD
-
-import scala.collection.JavaConversions;
-
-object GraphProcessor {
-
-  def findCCs(vertexes: RDD[(VertexId, String)], edges: RDD[Edge[String]], maxIterations: Int, cut:Int): RDD[ConnectedComponent] = {
-    val graph: Graph[String, String] = Graph(vertexes, edges).partitionBy(PartitionStrategy.RandomVertexCut) //TODO remember to remove partitionby
-    val cc = graph.connectedComponents(maxIterations).vertices
-
-    val joinResult = vertexes.leftOuterJoin(cc).map {
-      case (id, (openaireId, cc)) => {
-        if (cc.isEmpty) {
-          (id, openaireId)
-        }
-        else {
-          (cc.get, openaireId)
-        }
-      }
-    }
-    val connectedComponents = joinResult.groupByKey()
-      .map[ConnectedComponent](cc => asConnectedComponent(cc, cut))
-    connectedComponents
-  }
-
-
-
-  def asConnectedComponent(group: (VertexId, Iterable[String]), cut:Int): ConnectedComponent = {
-    val docs = group._2.toSet[String]
-    val connectedComponent = new ConnectedComponent(JavaConversions.setAsJavaSet[String](docs), cut);
-    connectedComponent
-  }
-
-}