D-Net

<?xml version="1.0" encoding="UTF-8"?>

<beans xmlns="http://www.springframework.org/schema/beans"

	xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:jaxws="http://cxf.apache.org/jaxws"

	xmlns:sec="http://cxf.apache.org/configuration/security" xmlns:wsa="http://cxf.apache.org/ws/addressing"

	xmlns:p="http://www.springframework.org/schema/p" xmlns:http="http://cxf.apache.org/transports/http/configuration"

	xmlns:t="http://dnetlib.eu/springbeans/t" xmlns:template="http://dnetlib.eu/springbeans/template"

	xmlns:util="http://www.springframework.org/schema/util"

	xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsd

                                    http://cxf.apache.org/ws/addressing http://cxf.apache.org/schemas/ws-addr-conf.xsd

                                    http://cxf.apache.org/configuration/security http://cxf.apache.org/schemas/configuration/security.xsd

                                    http://cxf.apache.org/transports/http/configuration http://cxf.apache.org/schemas/configuration/http-conf.xsd

                            http://cxf.apache.org/jaxws http://cxf.apache.org/schemas/jaxws.xsd

                            http://www.springframework.org/schema/util http://www.springframework.org/schema/util/spring-util-2.0.xsd

                            http://dnetlib.eu/springbeans/template http://dnetlib.eu/springbeans/template.xsd">

	<bean id="hbaseActionManagerCore" class="eu.dnetlib.actionmanager.hbase.HBaseActionManagerCore"

		p:actionFactory-ref="actionFactory"

		p:hbaseClient-ref="hbaseClient" />

	<bean id="hbaseClient" class="eu.dnetlib.actionmanager.hbase.HBaseClient"

		p:tableName="db_actions_michele"

		p:hbaseConf="${services.data.hbase.dm.properties}" 

		init-method="initWithEmptyTable" />		

		<!-- p:hbaseConf="classpath:/eu/dnetlib/actionmanager/test/hbase.properties" -->

	<bean id="actionFactory" class="eu.dnetlib.actionmanager.actions.ActionFactory">

		<property name="xslts">

			<util:map>

				<entry key="dmf2actions" value="classpath:/eu/dnetlib/actionmanager/xslt/dmf2actions.xslt" />

			</util:map>

		</property>

	</bean>

</beans>

#COMMON

hadoop.entry.node                               =       node1.hadoop.research-infrastructures.eu

#CORE-SITE

fs.defaultFS                                    =       hdfs://${hadoop.entry.node}:8020

io.file.buffer.size                     =       65536

io.compression.codecs                   =       org.apache.hadoop.io.compress.DefaultCodec,org.apache.hadoop.io.compress.GzipCodec,org.apache.hadoop.io.compress.BZip2Codec,org.apache.hadoop.io.compress.DeflateCodec,org.apache.hadoop.io.compress.SnappyCodec

hadoop.security.authentication  =       simple

hadoop.security.auth_to_local   =       DEFAULT

#HBASE-SITE

hbase.rootdir                                   =       ${fs.defaultFS}/hbase

hbase.client.write.buffer               =       2097152

hbase.client.pause                              =       1000

hbase.client.retries.number             =       10

hbase.client.scanner.caching    =       1

hbase.client.keyvalue.maxsize   =       10485760

hbase.security.authentication   =       simple

zookeeper.session.timeout               =       60000

zookeeper.znode.parent                  =       /hbase

zookeeper.znode.rootserver              =       root-region-server

hbase.zookeeper.quorum                  =       ${hadoop.entry.node}

hbase.zookeeper.property.clientPort     =       2182

#HDFS-SITE

dfs.https.port                                  =       50470

dfs.namenode.https-address              =       ${hadoop.entry.node}:${dfs.https.port}

dfs.namenode.http-address               =       ${hadoop.entry.node}:50070

dfs.replication                                 =       3

dfs.blocksize                                   =       134217728

dfs.client.use.datanode.hostname=       false

#MAPRED-SITE

mapreduce.jobtracker.address    =       ${hadoop.entry.node}:8021

mapred.job.tracker                              =       ${hadoop.entry.node}:8021

mapred.mapper.new-api                   =       true

mapred.reducer.new-api                  =       true

mapred.output.compress                  =       false

mapred.output.compression.type  =       BLOCK

mapred.output.compression.codec =       org.apache.hadoop.io.compress.DefaultCodec

#mapred.map.output.compression.codec    =       org.apache.hadoop.io.compress.SnappyCodec

mapred.compress.map.output              =       true

#io.sort.factor                                 =       64

#io.sort.record.percent                 =       0.05

#io.sort.spill.percent                  =       0.8

#mapred.reduce.parallel.copies  =       10

#mapred.submit.replication              =       1

#mapred.reduce.tasks                            =       8

#io.sort.mb                                             =       56

#mapred.child.java.opts                 =        -Xmx238449165

#mapred.job.reuse.jvm.num.tasks =       1

#mapred.map.tasks                               =       6

#mapred.map.tasks.speculative.execution =       false

#mapred.reduce.tasks.speculative.execution      =       false

#mapred.reduce.slowstart.completed.maps =       0.8

##mapred.local.dir                              = /data/mapred/local,/data/mapred/jt

mapred.local.dir                                = /tmp

mapred.child.java.opts                  = -Xmx512973352

#dfs.client.use.legacy.blockreader=true

<?xml version="1.0" encoding="UTF-8"?>

<record xmlns="http://www.driver-repository.eu/" xmlns:dc="http://purl.org/dc/elements/1.1/"

    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

    xmlns:dr="http://www.driver-repository.eu/namespace/dr"

    xmlns:dri="http://www.driver-repository.eu/namespace/dri">

    <header>

        <dri:objIdentifier>b91fdb84-1030-49d9-b398-813ab7cae968_UmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZXMvUmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZVR5cGU=::4d4life.eu/2009_Presentation_29301_!W__!V_1</dri:objIdentifier>

        <dri:recordIdentifier>7ce67201-b143-4ba8-aeaf-32f218778733_TURTdG9yZURTUmVzb3VyY2VzL01EU3RvcmVEU1Jlc291cmNlVHlwZQ==::4d4life.eu/2009_Presentation_29301_!W__!V_1</dri:recordIdentifier>

        <dri:dateOfCollection>2013-02-10T14:33:11Z</dri:dateOfCollection>

        <dri:mdFormat/>

        <dri:mdFormatInterpretation/>

        <dri:repositoryId>b91fdb84-1030-49d9-b398-813ab7cae968_UmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZXMvUmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZVR5cGU=</dri:repositoryId>

        <dr:objectIdentifier>b91fdb84-1030-49d9-b398-813ab7cae968_UmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZXMvUmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZVR5cGU=::4d4life.eu/2009_Presentation_29301_!W__!V_1</dr:objectIdentifier>

        <dr:dateOfCollection>2013-02-10T14:33:11Z</dr:dateOfCollection>

    </header>

    <metadata>

        <dr:CobjContentSynthesis/>

        <dr:CobjTypology>Textual</dr:CobjTypology>

        <dr:CobjModel>OAI</dr:CobjModel>

        <dr:CobjMDFormats>oai_dc</dr:CobjMDFormats>

        <dr:CobjDescriptionSynthesis/>

        <dr:aggregatorName>DRIVER</dr:aggregatorName>

        <dr:aggregatorInstitution>Bielefeld University Library</dr:aggregatorInstitution>

        <dr:repositoryName>BELIEF e-Infrastructure Digital Library</dr:repositoryName>

        <dr:repositoryLink>http://belief-dl.research-infrastructures.eu</dr:repositoryLink>

        <dr:repositoryCountry>IT</dr:repositoryCountry>

        <dr:repositoryInstitution>European Commission</dr:repositoryInstitution>

        <dc:creator>Frank Bisby, Frank</dc:creator>

        <dc:title>Distributed Dynamic Diversity Databases for Life (4D4Life)</dc:title>

        <dc:subject>Biology</dc:subject>

        <dc:subject>Natural Sciences</dc:subject>

        <dc:subject>Biodiversity</dc:subject>

        <dc:subject>Species Catalogue</dc:subject>

        <dr:CobjCategory>0000</dr:CobjCategory>

        <dc:language>eng</dc:language>

        <dc:dateAccepted/>

        <dc:identifier>http://belief-dl.research-infrastructures.eu/UIMain.jsp?handle=4d4life.eu/2009_Presentation_29301&amp;view=&amp;version=1</dc:identifier>

        <dc:publisher>4d4Life Project</dc:publisher>

        <dc:description>Brief presentation on the objectives of the D4DLife project. The

            presentation was provided in the data session of the 7th eConcertation Meeting in

            Brussels</dc:description>

    </metadata>

</record>

package eu.dnetlib.actionmanager.actions;

import java.io.StringReader;

import java.util.List;

import javax.xml.transform.Transformer;

import javax.xml.transform.TransformerException;

import org.apache.commons.codec.binary.Base64;

import org.apache.commons.logging.Log;

import org.apache.commons.logging.LogFactory;

import org.apache.hadoop.hbase.util.Bytes;

import org.dom4j.Document;

import org.dom4j.DocumentException;

import org.dom4j.Element;

import org.dom4j.io.DocumentResult;

import org.dom4j.io.DocumentSource;

import org.dom4j.io.SAXReader;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.ActionManagerConstants.ACTION_TYPE;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.common.Provenance;

import eu.dnetlib.actionmanager.rmi.ActionManagerException;

public class XsltInfoPackageAction extends AbstractMetaAction<String> {

	private final Transformer transformer;

	private static final Log log = LogFactory.getLog(XsltInfoPackageAction.class); // NOPMD by marko on 11/24/08 5:02 PM

	protected XsltInfoPackageAction(String set, Agent agent, Operation operation, String infoPackage, Transformer transformer, ActionFactory actionFactory) {

		super(set, agent, operation, infoPackage, actionFactory);

		this.transformer = transformer;

	}

	private Document applyXslt(String xml, Provenance provenance, String trust, String nsprefix) throws DocumentException, TransformerException {

		Document doc = new SAXReader().read(new StringReader(xml));

		final DocumentResult result = new DocumentResult();

		transformer.setParameter("trust", trust);

		transformer.setParameter("provenance", provenance.toString());

		transformer.setParameter("namespaceprefix", nsprefix);

		transformer.transform(new DocumentSource(doc), result);

		// System.out.println(result.getDocument().asXML());

		return result.getDocument();

	}

	@Override

	protected List<AtomicAction> calculateAtomicActions(Provenance provenance, String trust, String nsprefix) throws ActionManagerException {

		List<AtomicAction> list = Lists.newArrayList();

		try {

			Document doc = applyXslt(getMetaContent(), provenance, trust, nsprefix);

			for (Object o : doc.selectNodes("//ACTION")) {

				list.add(createAtomicAction((Element) o));

			}

		} catch (Exception e) {

			log.error("Error generating actions", e);

			throw new ActionManagerException(e);

		}

		return list;

	}

	private AtomicAction createAtomicAction(Element elem) {

		String key = elem.valueOf("./@targetKey");

		String colFamily = elem.valueOf("./@targetColumnFamily");

		String col = elem.valueOf("./@targetColumn");

		String value64 = elem.getTextTrim();

		byte[] value = value64.isEmpty() ? null : Base64.decodeBase64(value64);

		return getActionFactory().createAtomicAction(getSet(), getAgent(), key, colFamily, col, value);

	}

	@Override

	public boolean isValid() {

		// TODO Auto-generated method stub

		return true;

	}

	@Override

	public ACTION_TYPE getTypology() {

		return ACTION_TYPE.pkg;

	}

	@Override

	protected byte[] getBytesContent(String content) {

		return Bytes.toBytes(content);

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import java.util.UUID;

import org.apache.hadoop.hbase.client.Put;

import eu.dnetlib.actionmanager.ActionManagerConstants.ACTION_TYPE;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Provenance;

import eu.dnetlib.actionmanager.rmi.ActionManagerException;

import eu.dnetlib.miscutils.functional.hash.Hashing;

public abstract class AbstractAction implements Action {

	private String set;

	private String rowKey;

	private Agent agent;

	public AbstractAction(String set, Agent agent) {

		super();

		this.set = set;

		this.agent = agent;

		this.rowKey = getTypology() + "|" + Hashing.md5(agent.getId()) + "|" + UUID.randomUUID();

	}

	@Override

	abstract public List<Put> asPutOperations(String infoPackageId, Provenance provenance, String trust, String nsprefix) throws ActionManagerException;

	@Override

	abstract public boolean isValid();

	@Override

	abstract public ACTION_TYPE getTypology();

	public String getSet() {

		return set;

	}

	public void setSet(String set) {

		this.set = set;

	}

	public String getRowKey() {

		return rowKey;

	}

	public void setRowKey(String rowKey) {

		this.rowKey = rowKey;

	}

	/**

	 * @return the agent

	 */

	public Agent getAgent() {

		return agent;

	}

	/**

	 * @param agent

	 *            the agent to set

	 */

	public void setAgent(Agent agent) {

		this.agent = agent;

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import java.util.Map;

import javax.xml.transform.Transformer;

import javax.xml.transform.TransformerFactory;

import org.apache.hadoop.hbase.util.Bytes;

import org.dom4j.Document;

import org.dom4j.io.DocumentSource;

import org.dom4j.io.SAXReader;

import org.springframework.beans.factory.annotation.Required;

import org.springframework.core.io.Resource;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.rmi.ActionManagerException;

import eu.dnetlib.data.proto.RelTypeProtos.RelType;

import eu.dnetlib.data.proto.TypeProtos.Type;

import eu.dnetlib.miscutils.datetime.DateUtils;

/**

 * Factory bean to create Actions bypassing the Action Manager API

 * 

 * @author michele

 */

public class ActionFactory {

	private Map<String, Resource> xslts;

	final TransformerFactory tFactory = TransformerFactory.newInstance();

	/**

	 * Creates a list of AtomicAction(s) to handle an UPDATE. 1 body and 2 similarRel(s)

	 * 

	 * @param set

	 *            the given set.

	 * @param agent

	 *            the given agent.

	 * @param originalId

	 *            main entity identifier to be updated (already existing in the information space).

	 * @param updateId

	 *            new and unique identifier, must be the same used into "updateContent".

	 * @param entityType

	 *            one of: result, person, project, datasource, organization

	 *            {@link eu.dnetlib.data.proto.TypeProtos.Type}.

	 * @param updateContent

	 *            content to be updated over the original main entity. note that: the given trust level has to be [>.9 |

	 *            NEUTRAL | INFINITE] in order to gain visibility after the merging process.

	 * @return the list of AtomicActions.

	 */

	@Deprecated

	public List<AtomicAction> createUpdateActions(String set, Agent agent, String originalId, String updateId, Type entityType, byte[] updateContent) {

		List<AtomicAction> res = Lists.newArrayList();

		if (originalId.equals(updateId)) {

			throw new IllegalArgumentException("originalId and updateId must be different");

		}

		res.add(createAtomicAction(set, agent, updateId, entityType.toString(), "body", updateContent));

		res.add(createAtomicAction(set, agent, updateId, RelType.similarRel.toString(), originalId, Bytes.toBytes("")));

		res.add(createAtomicAction(set, agent, originalId, RelType.similarRel.toString(), updateId, Bytes.toBytes("")));

		return res;

	}

	/**

	 * Creates a list of AtomicAction(s) to handle an UPDATE. Produces one Oaf, stored in CF =

	 * [entity]:update_[timestamp]

	 * 

	 * @param set

	 *            the given set.

	 * @param agent

	 *            the given agent.

	 * @param originalId

	 *            main entity identifier to be updated (already existing in the information space).

	 * @param entityType

	 *            one of: result, person, project, datasource, organization

	 *            {@link eu.dnetlib.data.proto.TypeProtos.Type}.

	 * @param updateContent

	 *            content to be updated over the original main entity. Oaf#OafEntity#[entity]#id must be consistent with

	 *            the originalId. note that: the given trust level has to be [>.9 | NEUTRAL | INFINITE] in order to gain

	 *            visibility after the merging process.

	 * @return the list of AtomicActions.

	 */

	public List<AtomicAction> createUpdateActions(String set, Agent agent, String originalId, Type entityType, byte[] updateContent) {

		List<AtomicAction> res = Lists.newArrayList();

		res.add(createAtomicAction(set, agent, originalId, entityType.toString(), "update_" + DateUtils.now(), updateContent));

		return res;

	}

	public AtomicAction createAtomicAction(String set, Agent agent, String targetKey, String targetColFamily, String targetCol, byte[] targetContent) {

		AtomicAction action = new AtomicAction(set, agent);

		action.setTargetRowKey(targetKey);

		action.setTargetColumnFamily(targetColFamily);

		action.setTargetColumn(targetCol);

		action.setTargetValue(targetContent);

		return action;

	}

	public XsltInfoPackageAction generateInfoPackageAction(String xsltResource, String set, Agent agent, Operation operation, String infoPackage)

			throws ActionManagerException {

		Transformer transformer = prepareXsltTransformer(xsltResource);

		return new XsltInfoPackageAction(set, agent, operation, infoPackage, transformer, this);

	}

	public Transformer prepareXsltTransformer(String xsltName) throws ActionManagerException {

		try {

			if (!xslts.containsKey(xsltName)) {

				throw new ActionManagerException("XSLT " + xsltName + " not found");

			}

			final Document doc = new SAXReader().read(xslts.get(xsltName).getInputStream());

			return tFactory.newTransformer(new DocumentSource(doc));

		} catch (Exception e) {

			throw new ActionManagerException("Problem with xslt resource " + xsltName, e);

		}

	}

	public Map<String, Resource> getXslts() {

		return xslts;

	}

	@Required

	public void setXslts(Map<String, Resource> xslts) {

		this.xslts = xslts;

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import org.apache.hadoop.hbase.client.Put;

import org.apache.hadoop.hbase.util.Bytes;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.ActionManagerConstants;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.common.Provenance;

import eu.dnetlib.actionmanager.hbase.HBasePutFactory;

import eu.dnetlib.actionmanager.rmi.ActionManagerException;

abstract public class AbstractMetaAction<T> extends AbstractAction {

	private Operation operation;

	private T metaContent;

	private ActionFactory actionFactory;

	public AbstractMetaAction(String set, Agent agent, Operation operation, T metaContent, ActionFactory actionFactory) {

		super(set, agent);

		this.setOperation(operation);

		this.metaContent = metaContent;

		this.setActionFactory(actionFactory);

	}

	@Override

	public List<Put> asPutOperations(String parentId, Provenance provenance, String trust, String nsprefix) throws ActionManagerException {

		List<Put> puts = Lists.newArrayList();

		List<String> partIds = Lists.newArrayList();

		for (AtomicAction action : calculateAtomicActions(provenance, trust, nsprefix)) {

			if (action != null && action.isValid()) {

				puts.addAll(action.asPutOperations(getRowKey(), provenance, trust, nsprefix));

				partIds.add(action.getRowKey());

			}

		}

		puts.add(prepareMetaPut(partIds));

		return puts;

	}

	private Put prepareMetaPut(List<String> partIds) {

		final Put put = HBasePutFactory.createPutOperation(getRowKey(), getSet(), getAgent());

		put.add(ActionManagerConstants.OPERATION_COLFAMILY, Bytes.toBytes(getOperation().toString()), Bytes.toBytes(getOperation().toString()));

		put.add(ActionManagerConstants.ACTION_COLFAMILY, Bytes.toBytes(getTypology().toString()), getBytesContent(metaContent));

		for (String id : partIds) {

			put.add(ActionManagerConstants.RELATION_COLFAMILY, Bytes.toBytes(id), ActionManagerConstants.HASPARTS);

		}

		return put;

	}

	abstract protected List<AtomicAction> calculateAtomicActions(Provenance provenance, String trust, String nsprefix) throws ActionManagerException;

	abstract protected byte[] getBytesContent(T content);

	public Operation getOperation() {

		return operation;

	}

	public void setOperation(Operation operation) {

		this.operation = operation;

	}

	public T getMetaContent() {

		return metaContent;

	}

	public void setMetaContent(T metaContent) {

		this.metaContent = metaContent;

	}

	public ActionFactory getActionFactory() {

		return actionFactory;

	}

	public void setActionFactory(ActionFactory actionFactory) {

		this.actionFactory = actionFactory;

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import org.apache.hadoop.hbase.client.Put;

import org.apache.hadoop.hbase.util.Bytes;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.ActionManagerConstants;

import eu.dnetlib.actionmanager.ActionManagerConstants.ACTION_TYPE;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.common.Provenance;

import eu.dnetlib.actionmanager.hbase.HBasePutFactory;

public class AtomicAction extends AbstractAction {

	private String targetRowKey;

	private String targetColumnFamily;

	private String targetColumn;

	private byte[] targetValue;

	public AtomicAction(String set, Agent agent) {

		super(set, agent);

	}

	public AtomicAction(String set, Agent agent, Operation operation, String targetRowKey, String targetColumnFamily, String targetColumn,

			byte[] targetValue) {

		this(set, agent);

		this.targetRowKey = targetRowKey;

		this.targetColumnFamily = targetColumnFamily;

		this.targetColumn = targetColumn;

		this.targetValue = targetValue;

	}

	@Override

	public List<Put> asPutOperations(String parentId, Provenance provenance, String trust, String nsprefix) {

		final Put put = HBasePutFactory.createPutOperation(getRowKey(), getSet(), getAgent());

		put.add(ActionManagerConstants.SET_COLFAMILY, Bytes.toBytes(getSet()), Bytes.toBytes(getSet()));

		put.add(ActionManagerConstants.TARGET_COLFAMILY, ActionManagerConstants.TARGET_KEY_COL, Bytes.toBytes(targetRowKey));

		put.add(ActionManagerConstants.TARGET_COLFAMILY, ActionManagerConstants.TARGET_COLFAMILY_COL, Bytes.toBytes(targetColumnFamily));

		put.add(ActionManagerConstants.TARGET_COLFAMILY, ActionManagerConstants.TARGET_COL_COL, Bytes.toBytes(targetColumn));

		put.add(ActionManagerConstants.TARGET_COLFAMILY, ActionManagerConstants.TARGET_OAF_COL, targetValue);

		if (parentId != null && !parentId.isEmpty()) {

			put.add(ActionManagerConstants.RELATION_COLFAMILY, Bytes.toBytes(parentId), ActionManagerConstants.ISPARTOF);

		}

		return Lists.newArrayList(put);

	}

	@Override

	public ACTION_TYPE getTypology() {

		return ACTION_TYPE.aac;

	}

	public String getTargetRowKey() {

		return targetRowKey;

	}

	public void setTargetRowKey(String targetRowKey) {

		this.targetRowKey = targetRowKey;

	}

	public String getTargetColumnFamily() {

		return targetColumnFamily;

	}

	public void setTargetColumnFamily(String targetColumnFamily) {

		this.targetColumnFamily = targetColumnFamily;

	}

	public String getTargetColumn() {

		return targetColumn;

	}

	public void setTargetColumn(String targetColumn) {

		this.targetColumn = targetColumn;

	}

	public byte[] getTargetValue() {

		return targetValue;

	}

	public void setTargetValue(byte[] targetValue) {

		this.targetValue = targetValue;

	}

	@Override

	public boolean isValid() {

		return (getRowKey() != null && !getRowKey().isEmpty() && getSet() != null && !getSet().isEmpty() && targetRowKey != null

				&& !targetRowKey.isEmpty() && targetColumnFamily != null && !targetColumnFamily.isEmpty() && targetColumn != null

				&& !targetColumn.isEmpty() && targetValue != null && targetValue.length > 0);

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import org.apache.hadoop.hbase.client.Put;

import eu.dnetlib.actionmanager.ActionManagerConstants.ACTION_TYPE;

import eu.dnetlib.actionmanager.common.Provenance;

import eu.dnetlib.actionmanager.rmi.ActionManagerException;

public interface Action {

	ACTION_TYPE getTypology();

	List<Put> asPutOperations(String parentId, Provenance provenance, String trust, String nsprefix) throws ActionManagerException;

	boolean isValid();

}

package eu.dnetlib.actionmanager.hbase;

import org.apache.hadoop.hbase.client.Put;

import org.apache.hadoop.hbase.util.Bytes;

import eu.dnetlib.actionmanager.ActionManagerConstants;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.miscutils.datetime.DateUtils;

public class HBasePutFactory {

	public static Put createPutOperation(String key, String set, Agent agent) {

		final long date = DateUtils.now();

		final String dateS = DateUtils.calculate_ISO8601(date);

		final Put put = new Put(Bytes.toBytes(key));

		put.add(ActionManagerConstants.SET_COLFAMILY, Bytes.toBytes(set), Bytes.toBytes(set));

		put.add(ActionManagerConstants.DATE_COLFAMILY, Bytes.toBytes(Long.toString(date)), Bytes.toBytes(dateS));

		put.add(ActionManagerConstants.AGENT_COLFAMILY, ActionManagerConstants.AGENT_ID_COL, Bytes.toBytes(agent.getId()));

		put.add(ActionManagerConstants.AGENT_COLFAMILY, ActionManagerConstants.AGENT_NAME_COL, Bytes.toBytes(agent.getName()));

		put.add(ActionManagerConstants.AGENT_COLFAMILY, ActionManagerConstants.AGENT_TYPE_COL, Bytes.toBytes(agent.getType().toString()));

		return put;

	}

}

package eu.dnetlib.actionmanager;

import org.apache.hadoop.hbase.util.Bytes;

public class ActionManagerConstants {

	public static enum COLUMN_FAMILIES {

		action, rel, status, set, target, operation, date, agent

	}

	public static enum ACTION_TYPE {

		aac, //atomicAction, 

		pkg, //infoPackage, 

		all //all

	}

	public static final byte[] ACTION_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.action.toString());

	public static final byte[] RELATION_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.rel.toString());

	public static final byte[] STATUS_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.status.toString());

	public static final byte[] SET_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.set.toString());

	public static final byte[] TARGET_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.target.toString());

	public static final byte[] OPERATION_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.operation.toString());

	public static final byte[] DATE_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.date.toString());

	public static final byte[] AGENT_COLFAMILY = Bytes.toBytes(COLUMN_FAMILIES.agent.toString());

	public static final byte[] OPERATION_COL = Bytes.toBytes("operation");

	public static final byte[] TARGET_KEY_COL = Bytes.toBytes("rowKey");

	public static final byte[] TARGET_COLFAMILY_COL = Bytes.toBytes("columnFamily");

	public static final byte[] TARGET_COL_COL = Bytes.toBytes("column");

	public static final byte[] TARGET_OAF_COL = Bytes.toBytes("content");

	public static final byte[] HASPARTS = Bytes.toBytes("hasParts");

	public static final byte[] ISPARTOF = Bytes.toBytes("isPartOf");

	public static final byte[] AGENT_ID_COL = Bytes.toBytes("id");

	public static final byte[] AGENT_NAME_COL = Bytes.toBytes("name");

	public static final byte[] AGENT_TYPE_COL = Bytes.toBytes("type");

}

<?xml version="1.0" encoding="UTF-8"?>

<beansProjectDescription>

	<version>1</version>

	<pluginVersion><![CDATA[3.1.0.201210040510-RELEASE]]></pluginVersion>

	<configSuffixes>

		<configSuffix><![CDATA[xml]]></configSuffix>

	</configSuffixes>

	<enableImports><![CDATA[false]]></enableImports>

	<configs>

		<config>src/main/resources/eu/dnetlib/actionmanager/hack/applicationContext-claimed-import.xml</config>

	</configs>

	<configSets>

	</configSets>

</beansProjectDescription>

<?xml version="1.0" encoding="UTF-8"?>

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

	xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">

	<parent>

		<groupId>eu.dnetlib</groupId>

		<artifactId>dnet-parent</artifactId>

		<version>0.0.1-alpha</version>

	</parent>

	<modelVersion>4.0.0</modelVersion>

	<groupId>eu.dnetlib</groupId>

	<artifactId>dnet-actionmanager-common</artifactId>

	<packaging>jar</packaging>

	<version>1.0.3</version>

	<repositories>

		<!-- Cloudera Repositories -->

		<repository>

			<snapshots>

				<enabled>false</enabled>

			</snapshots>

			<id>cloudera-central</id>

			<name>cloundera-libs-release</name>

			<url>http://maven.research-infrastructures.eu/nexus/content/repositories/cloudera-central</url>

		</repository>

		<repository>

			<id>cloudera-snapshots</id>

			<name>cloudera-libs-snapshot</name>

			<url>http://maven.research-infrastructures.eu/nexus/content/repositories/cloudera-snapshots</url>

		</repository>

	</repositories>	

	<dependencies>

		<dependency>

			<groupId>eu.dnetlib</groupId>

			<artifactId>dnet-openaireplus-mapping-utils</artifactId>

			<version>[0.2.0]</version>

		</dependency>

		<dependency>

			<groupId>eu.dnetlib</groupId>

			<artifactId>dnet-actionmanager-api</artifactId>

			<version>[1.0]</version>

		</dependency>

		<dependency>

			<groupId>org.apache.hadoop</groupId>

			<artifactId>hadoop-common</artifactId>

			<version>${hadoop.common.version}</version>

		</dependency>				

		<dependency>

			<groupId>org.apache.hbase</groupId>

			<artifactId>hbase</artifactId>

			<version>${apache.hbase.version}</version>

			<exclusions>

				<exclusion>

					<groupId>tomcat</groupId>

					<artifactId>jasper-runtime</artifactId>

				</exclusion>

				<exclusion>

					<groupId>tomcat</groupId>

					<artifactId>jasper-compiler</artifactId>

				</exclusion>			

				<exclusion>

					<artifactId>slf4j-api</artifactId>

					<groupId>org.slf4j</groupId>

				</exclusion>

				<exclusion>

					<artifactId>slf4j-log4j12</artifactId>

					<groupId>org.slf4j</groupId>

				</exclusion>

				<exclusion>

					<artifactId>commons-lang</artifactId>

					<groupId>commons-lang</groupId>

				</exclusion>

				<exclusion>

					<artifactId>commons-httpclient</artifactId>

					<groupId>commons-httpclient</groupId>

				</exclusion>

				<exclusion>

					<artifactId>httpclient</artifactId>

					<groupId>org.apache.httpcomponents</groupId>

				</exclusion>

				<exclusion>

					<artifactId>httpcore</artifactId>

					<groupId>org.apache.httpcomponents</groupId>

				</exclusion>

			</exclusions>

		</dependency>

		<dependency>

			<groupId>junit</groupId>

			<artifactId>junit</artifactId>

			<version>${junit.version}</version>

			<scope>test</scope>

		</dependency>

	</dependencies>

</project>

<?xml version="1.0" encoding="UTF-8"?>

<beans xmlns="http://www.springframework.org/schema/beans"

	xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:jaxws="http://cxf.apache.org/jaxws"

	xmlns:sec="http://cxf.apache.org/configuration/security" xmlns:wsa="http://cxf.apache.org/ws/addressing"

	xmlns:p="http://www.springframework.org/schema/p" xmlns:http="http://cxf.apache.org/transports/http/configuration"

	xmlns:t="http://dnetlib.eu/springbeans/t" xmlns:template="http://dnetlib.eu/springbeans/template"

	xmlns:util="http://www.springframework.org/schema/util"

	xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsd

                                    http://cxf.apache.org/ws/addressing http://cxf.apache.org/schemas/ws-addr-conf.xsd

                                    http://cxf.apache.org/configuration/security http://cxf.apache.org/schemas/configuration/security.xsd

                                    http://cxf.apache.org/transports/http/configuration http://cxf.apache.org/schemas/configuration/http-conf.xsd

                            http://cxf.apache.org/jaxws http://cxf.apache.org/schemas/jaxws.xsd

                            http://www.springframework.org/schema/util http://www.springframework.org/schema/util/spring-util-2.0.xsd

                            http://dnetlib.eu/springbeans/template http://dnetlib.eu/springbeans/template.xsd">

	<bean id="hbaseActionManagerCore" class="eu.dnetlib.actionmanager.hbase.HBaseActionManagerCore"

		p:actionFactory-ref="actionFactory"

		p:hbaseClient-ref="hbaseClient" />

	<bean id="hbaseClient" class="eu.dnetlib.actionmanager.hbase.HBaseClient"

		p:tableName="db_actions_michele"

		p:hbaseConf="${services.data.hbase.dm.properties}" 

		init-method="initWithEmptyTable" />		

		<!-- p:hbaseConf="classpath:/eu/dnetlib/actionmanager/test/hbase.properties" -->

	<bean id="actionFactory" class="eu.dnetlib.actionmanager.actions.ActionFactory">

		<property name="xslts">

			<util:map>

				<entry key="dmf2actions" value="classpath:/eu/dnetlib/actionmanager/xslt/dmf2actions.xslt" />

			</util:map>

		</property>

	</bean>

</beans>

#COMMON

hadoop.entry.node                               =       node1.hadoop.research-infrastructures.eu

#CORE-SITE

fs.defaultFS                                    =       hdfs://${hadoop.entry.node}:8020

io.file.buffer.size                     =       65536

io.compression.codecs                   =       org.apache.hadoop.io.compress.DefaultCodec,org.apache.hadoop.io.compress.GzipCodec,org.apache.hadoop.io.compress.BZip2Codec,org.apache.hadoop.io.compress.DeflateCodec,org.apache.hadoop.io.compress.SnappyCodec

hadoop.security.authentication  =       simple

hadoop.security.auth_to_local   =       DEFAULT

#HBASE-SITE

hbase.rootdir                                   =       ${fs.defaultFS}/hbase

hbase.client.write.buffer               =       2097152

hbase.client.pause                              =       1000

hbase.client.retries.number             =       10

hbase.client.scanner.caching    =       1

hbase.client.keyvalue.maxsize   =       10485760

hbase.security.authentication   =       simple

zookeeper.session.timeout               =       60000

zookeeper.znode.parent                  =       /hbase

zookeeper.znode.rootserver              =       root-region-server

hbase.zookeeper.quorum                  =       ${hadoop.entry.node}

hbase.zookeeper.property.clientPort     =       2182

#HDFS-SITE

dfs.https.port                                  =       50470

dfs.namenode.https-address              =       ${hadoop.entry.node}:${dfs.https.port}

dfs.namenode.http-address               =       ${hadoop.entry.node}:50070

dfs.replication                                 =       3

dfs.blocksize                                   =       134217728

dfs.client.use.datanode.hostname=       false

#MAPRED-SITE

mapreduce.jobtracker.address    =       ${hadoop.entry.node}:8021

mapred.job.tracker                              =       ${hadoop.entry.node}:8021

mapred.mapper.new-api                   =       true

mapred.reducer.new-api                  =       true

mapred.output.compress                  =       false

mapred.output.compression.type  =       BLOCK

mapred.output.compression.codec =       org.apache.hadoop.io.compress.DefaultCodec

#mapred.map.output.compression.codec    =       org.apache.hadoop.io.compress.SnappyCodec

mapred.compress.map.output              =       true

#io.sort.factor                                 =       64

#io.sort.record.percent                 =       0.05

#io.sort.spill.percent                  =       0.8

#mapred.reduce.parallel.copies  =       10

#mapred.submit.replication              =       1

#mapred.reduce.tasks                            =       8

#io.sort.mb                                             =       56

#mapred.child.java.opts                 =        -Xmx238449165

#mapred.job.reuse.jvm.num.tasks =       1

#mapred.map.tasks                               =       6

#mapred.map.tasks.speculative.execution =       false

#mapred.reduce.tasks.speculative.execution      =       false

#mapred.reduce.slowstart.completed.maps =       0.8

##mapred.local.dir                              = /data/mapred/local,/data/mapred/jt

mapred.local.dir                                = /tmp

mapred.child.java.opts                  = -Xmx512973352

#dfs.client.use.legacy.blockreader=true

<?xml version="1.0" encoding="UTF-8"?>

<record xmlns="http://www.driver-repository.eu/" xmlns:dc="http://purl.org/dc/elements/1.1/"

    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

    xmlns:dr="http://www.driver-repository.eu/namespace/dr"

    xmlns:dri="http://www.driver-repository.eu/namespace/dri">

    <header>

        <dri:objIdentifier>b91fdb84-1030-49d9-b398-813ab7cae968_UmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZXMvUmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZVR5cGU=::4d4life.eu/2009_Presentation_29301_!W__!V_1</dri:objIdentifier>

        <dri:recordIdentifier>7ce67201-b143-4ba8-aeaf-32f218778733_TURTdG9yZURTUmVzb3VyY2VzL01EU3RvcmVEU1Jlc291cmNlVHlwZQ==::4d4life.eu/2009_Presentation_29301_!W__!V_1</dri:recordIdentifier>

        <dri:dateOfCollection>2013-02-10T14:33:11Z</dri:dateOfCollection>

        <dri:mdFormat/>

        <dri:mdFormatInterpretation/>

        <dri:repositoryId>b91fdb84-1030-49d9-b398-813ab7cae968_UmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZXMvUmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZVR5cGU=</dri:repositoryId>

        <dr:objectIdentifier>b91fdb84-1030-49d9-b398-813ab7cae968_UmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZXMvUmVwb3NpdG9yeVNlcnZpY2VSZXNvdXJjZVR5cGU=::4d4life.eu/2009_Presentation_29301_!W__!V_1</dr:objectIdentifier>

        <dr:dateOfCollection>2013-02-10T14:33:11Z</dr:dateOfCollection>

    </header>

    <metadata>

        <dr:CobjContentSynthesis/>

        <dr:CobjTypology>Textual</dr:CobjTypology>

        <dr:CobjModel>OAI</dr:CobjModel>

        <dr:CobjMDFormats>oai_dc</dr:CobjMDFormats>

        <dr:CobjDescriptionSynthesis/>

        <dr:aggregatorName>DRIVER</dr:aggregatorName>

        <dr:aggregatorInstitution>Bielefeld University Library</dr:aggregatorInstitution>

        <dr:repositoryName>BELIEF e-Infrastructure Digital Library</dr:repositoryName>

        <dr:repositoryLink>http://belief-dl.research-infrastructures.eu</dr:repositoryLink>

        <dr:repositoryCountry>IT</dr:repositoryCountry>

        <dr:repositoryInstitution>European Commission</dr:repositoryInstitution>

        <dc:creator>Frank Bisby, Frank</dc:creator>

        <dc:title>Distributed Dynamic Diversity Databases for Life (4D4Life)</dc:title>

        <dc:subject>Biology</dc:subject>

        <dc:subject>Natural Sciences</dc:subject>

        <dc:subject>Biodiversity</dc:subject>

        <dc:subject>Species Catalogue</dc:subject>

        <dr:CobjCategory>0000</dr:CobjCategory>

        <dc:language>eng</dc:language>

        <dc:dateAccepted/>

        <dc:identifier>http://belief-dl.research-infrastructures.eu/UIMain.jsp?handle=4d4life.eu/2009_Presentation_29301&amp;view=&amp;version=1</dc:identifier>

        <dc:publisher>4d4Life Project</dc:publisher>

        <dc:description>Brief presentation on the objectives of the D4DLife project. The

            presentation was provided in the data session of the 7th eConcertation Meeting in

            Brussels</dc:description>

    </metadata>

</record>

package eu.dnetlib.actionmanager.actions;

import java.io.StringReader;

import java.util.List;

import javax.xml.transform.Transformer;

import javax.xml.transform.TransformerException;

import org.apache.commons.codec.binary.Base64;

import org.apache.commons.logging.Log;

import org.apache.commons.logging.LogFactory;

import org.apache.hadoop.hbase.util.Bytes;

import org.dom4j.Document;

import org.dom4j.DocumentException;

import org.dom4j.Element;

import org.dom4j.io.DocumentResult;

import org.dom4j.io.DocumentSource;

import org.dom4j.io.SAXReader;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.ActionManagerConstants.ACTION_TYPE;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.common.Provenance;

public class XsltInfoPackageAction extends AbstractMetaAction<String> {

	private final Transformer transformer;

	private static final Log log = LogFactory.getLog(XsltInfoPackageAction.class); // NOPMD by marko on 11/24/08 5:02 PM

	protected XsltInfoPackageAction(String set, Agent agent, Operation operation, String infoPackage, Transformer transformer, ActionFactory actionFactory) {

		super(set, agent, operation, infoPackage, actionFactory);

		this.transformer = transformer;

	}

	private Document applyXslt(String xml, Provenance provenance, String trust, String nsprefix) throws DocumentException, TransformerException {

		Document doc = (new SAXReader()).read(new StringReader(xml));

		final DocumentResult result = new DocumentResult();

		transformer.setParameter("trust", trust);

		transformer.setParameter("provenance", provenance.toString());

		transformer.setParameter("namespaceprefix", nsprefix);

		transformer.transform(new DocumentSource(doc), result);

		//System.out.println(result.getDocument().asXML());

		return result.getDocument();

	}

	@Override

	protected List<AtomicAction> calculateAtomicActions(Provenance provenance, String trust, String nsprefix) {

		List<AtomicAction> list = Lists.newArrayList();

		try {

			Document doc = applyXslt(getMetaContent(), provenance, trust, nsprefix);

			for (Object o : doc.selectNodes("//ACTION")) {

				list.add(createAtomicAction((Element) o));

			}

		} catch (Exception e) {

			log.error("Error generating actions", e);

		}

		return list;

	}

	private AtomicAction createAtomicAction(Element elem) {

		String key = elem.valueOf("./@targetKey");

		String colFamily = elem.valueOf("./@targetColumnFamily");

		String col = elem.valueOf("./@targetColumn");

		String value64 = elem.getTextTrim();

		byte[] value = value64.isEmpty() ? null : Base64.decodeBase64(value64);

		return getActionFactory().createAtomicAction(getSet(), getAgent(), key, colFamily, col, value);

	}

	@Override

	public boolean isValid() {

		// TODO Auto-generated method stub

		return true;

	}

	@Override

	public ACTION_TYPE getTypology() {

		return ACTION_TYPE.pkg;

	}

	@Override

	protected byte[] getBytesContent(String content) {

		return Bytes.toBytes(content);

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import java.util.UUID;

import org.apache.hadoop.hbase.client.Put;

import com.google.common.hash.Hashing;

import eu.dnetlib.actionmanager.ActionManagerConstants.ACTION_TYPE;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Provenance;

public abstract class AbstractAction implements Action {

	private String set;

	private String rowKey;

	private Agent agent;

	public AbstractAction(String set, Agent agent) {

		super();

		this.set = set;

		this.agent = agent;

		this.rowKey = getTypology() + "|" + Hashing.md5().hashString(agent.getId()) + "|" + UUID.randomUUID();

	}

	@Override

	abstract public List<Put> asPutOperations(String infoPackageId, Provenance provenance, String trust, String nsprefix);

	@Override

	abstract public boolean isValid();

	@Override

	abstract public ACTION_TYPE getTypology();

	public String getSet() {

		return set;

	}

	public void setSet(String set) {

		this.set = set;

	}

	public String getRowKey() {

		return rowKey;

	}

	public void setRowKey(String rowKey) {

		this.rowKey = rowKey;

	}

	/**

	 * @return the agent

	 */

	public Agent getAgent() {

		return agent;

	}

	/**

	 * @param agent

	 *            the agent to set

	 */

	public void setAgent(Agent agent) {

		this.agent = agent;

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import java.util.Map;

import javax.xml.transform.Transformer;

import javax.xml.transform.TransformerFactory;

import org.apache.hadoop.hbase.util.Bytes;

import org.dom4j.Document;

import org.dom4j.io.DocumentSource;

import org.dom4j.io.SAXReader;

import org.springframework.beans.factory.annotation.Required;

import org.springframework.core.io.Resource;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.rmi.ActionManagerException;

import eu.dnetlib.data.proto.RelTypeProtos.RelType;

import eu.dnetlib.data.proto.TypeProtos.Type;

import eu.dnetlib.miscutils.datetime.DateUtils;

/**

 * Factory bean to create Actions bypassing the Action Manager API

 * 

 * @author michele

 */

public class ActionFactory {

	private Map<String, Resource> xslts;

	final TransformerFactory tFactory = TransformerFactory.newInstance();

	/**

	 * Creates a list of AtomicAction(s) to handle an UPDATE. 1 body and 2 similarRel(s)

	 * 

	 * @param set

	 *            the given set.

	 * @param agent

	 *            the given agent.

	 * @param originalId

	 *            main entity identifier to be updated (already existing in the information space).

	 * @param updateId

	 *            new and unique identifier, must be the same used into "updateContent".

	 * @param entityType

	 *            one of: result, person, project, datasource, organization

	 *            {@link eu.dnetlib.data.proto.TypeProtos.Type}.

	 * @param updateContent

	 *            content to be updated over the original main entity. note that: the given trust level has to be [>.9 |

	 *            NEUTRAL | INFINITE] in order to gain visibility after the merging process.

	 * @return the list of AtomicActions.

	 */

	@Deprecated

	public List<AtomicAction> createUpdateActions(String set, Agent agent, String originalId, String updateId, Type entityType, byte[] updateContent) {

		List<AtomicAction> res = Lists.newArrayList();

		if (originalId.equals(updateId)) {

			throw new IllegalArgumentException("originalId and updateId must be different");

		}

		res.add(createAtomicAction(set, agent, updateId, entityType.toString(), "body", updateContent));

		res.add(createAtomicAction(set, agent, updateId, RelType.similarRel.toString(), originalId, Bytes.toBytes("")));

		res.add(createAtomicAction(set, agent, originalId, RelType.similarRel.toString(), updateId, Bytes.toBytes("")));

		return res;

	}

	/**

	 * Creates a list of AtomicAction(s) to handle an UPDATE. Produces one Oaf, stored in CF =

	 * [entity]:update_[timestamp]

	 * 

	 * @param set

	 *            the given set.

	 * @param agent

	 *            the given agent.

	 * @param originalId

	 *            main entity identifier to be updated (already existing in the information space).

	 * @param entityType

	 *            one of: result, person, project, datasource, organization

	 *            {@link eu.dnetlib.data.proto.TypeProtos.Type}.

	 * @param updateContent

	 *            content to be updated over the original main entity. Oaf#OafEntity#[entity]#id must be consistent with

	 *            the originalId. note that: the given trust level has to be [>.9 | NEUTRAL | INFINITE] in order to gain

	 *            visibility after the merging process.

	 * @return the list of AtomicActions.

	 */

	public List<AtomicAction> createUpdateActions(String set, Agent agent, String originalId, Type entityType, byte[] updateContent) {

		List<AtomicAction> res = Lists.newArrayList();

		res.add(createAtomicAction(set, agent, originalId, entityType.toString(), "update_" + DateUtils.now(), updateContent));

		return res;

	}

	public AtomicAction createAtomicAction(String set, Agent agent, String targetKey, String targetColFamily, String targetCol, byte[] targetContent) {

		AtomicAction action = new AtomicAction(set, agent);

		action.setTargetRowKey(targetKey);

		action.setTargetColumnFamily(targetColFamily);

		action.setTargetColumn(targetCol);

		action.setTargetValue(targetContent);

		return action;

	}

	public XsltInfoPackageAction generateInfoPackageAction(String xsltResource, String set, Agent agent, Operation operation, String infoPackage)

			throws ActionManagerException {

		Transformer transformer = prepareXsltTransformer(xsltResource);

		return new XsltInfoPackageAction(set, agent, operation, infoPackage, transformer, this);

	}

	public Transformer prepareXsltTransformer(String xsltName) throws ActionManagerException {

		try {

			if (!xslts.containsKey(xsltName)) {

				throw new ActionManagerException("XSLT " + xsltName + " not found");

			}

			final Document doc = new SAXReader().read(xslts.get(xsltName).getInputStream());

			return tFactory.newTransformer(new DocumentSource(doc));

		} catch (Exception e) {

			throw new ActionManagerException("Problem with xslt resource " + xsltName, e);

		}

	}

	public Map<String, Resource> getXslts() {

		return xslts;

	}

	@Required

	public void setXslts(Map<String, Resource> xslts) {

		this.xslts = xslts;

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import org.apache.commons.logging.Log;

import org.apache.commons.logging.LogFactory;

import org.apache.hadoop.hbase.client.Put;

import org.apache.hadoop.hbase.util.Bytes;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.ActionManagerConstants;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.common.Provenance;

import eu.dnetlib.actionmanager.hbase.HBasePutFactory;

abstract public class AbstractMetaAction<T> extends AbstractAction {

	private Operation operation;

	private T metaContent;

	private ActionFactory actionFactory;

	private static final Log log = LogFactory.getLog(AbstractMetaAction.class); // NOPMD by marko on 11/24/08 5:02 PM

	public AbstractMetaAction(String set, Agent agent, Operation operation, T metaContent, ActionFactory actionFactory) {

		super(set, agent);

		this.setOperation(operation);

		this.metaContent = metaContent;

		this.setActionFactory(actionFactory);

	}

	@Override

	public List<Put> asPutOperations(String parentId, Provenance provenance, String trust, String nsprefix) {

		List<Put> puts = Lists.newArrayList();

		List<String> partIds = Lists.newArrayList();

		try {

			for (AtomicAction action : calculateAtomicActions(provenance, trust, nsprefix)) {

				if (action != null && action.isValid()) {

					puts.addAll(action.asPutOperations(getRowKey(), provenance, trust, nsprefix));

					partIds.add(action.getRowKey());

				}

			}

			puts.add(prepareMetaPut(partIds));

		} catch (Exception e) {

			log.error("Error generating actions", e);

		}

		return puts;

	}

	private Put prepareMetaPut(List<String> partIds) {

		final Put put = HBasePutFactory.createPutOperation(getRowKey(), getSet(), getAgent());

		put.add(ActionManagerConstants.OPERATION_COLFAMILY, Bytes.toBytes(getOperation().toString()), Bytes.toBytes(getOperation().toString()));

		put.add(ActionManagerConstants.ACTION_COLFAMILY, Bytes.toBytes(getTypology().toString()), getBytesContent(metaContent));

		for (String id : partIds) {

			put.add(ActionManagerConstants.RELATION_COLFAMILY, Bytes.toBytes(id), ActionManagerConstants.HASPARTS);

		}

		return put;

	}

	abstract protected List<AtomicAction> calculateAtomicActions(Provenance provenance, String trust, String nsprefix);

	abstract protected byte[] getBytesContent(T content);

	public Operation getOperation() {

		return operation;

	}

	public void setOperation(Operation operation) {

		this.operation = operation;

	}

	public T getMetaContent() {

		return metaContent;

	}

	public void setMetaContent(T metaContent) {

		this.metaContent = metaContent;

	}

	public ActionFactory getActionFactory() {

		return actionFactory;

	}

	public void setActionFactory(ActionFactory actionFactory) {

		this.actionFactory = actionFactory;

	}

}

package eu.dnetlib.actionmanager.actions;

import java.util.List;

import org.apache.hadoop.hbase.client.Put;

import org.apache.hadoop.hbase.util.Bytes;

import com.google.common.collect.Lists;

import eu.dnetlib.actionmanager.ActionManagerConstants;

import eu.dnetlib.actionmanager.ActionManagerConstants.ACTION_TYPE;

import eu.dnetlib.actionmanager.common.Agent;

import eu.dnetlib.actionmanager.common.Operation;

import eu.dnetlib.actionmanager.common.Provenance;

import eu.dnetlib.actionmanager.hbase.HBasePutFactory;

public class AtomicAction extends AbstractAction {

	private String targetRowKey;

	private String targetColumnFamily;

	private String targetColumn;

	private byte[] targetValue;

	public AtomicAction(String set, Agent agent) {

		super(set, agent);

	}

	public AtomicAction(String set, Agent agent, Operation operation, String targetRowKey, String targetColumnFamily, String targetColumn,

			byte[] targetValue) {

		this(set, agent);

		this.targetRowKey = targetRowKey;

		this.targetColumnFamily = targetColumnFamily;