161 lines
4.9 KiB
Java
161 lines
4.9 KiB
Java
|
package eu.dnetlib.ariadneplus.workflows.nodes;
|
||
|
|
||
|
import com.google.common.collect.Lists;
|
||
|
import com.google.common.collect.Maps;
|
||
|
import com.google.gson.Gson;
|
||
|
import eu.dnetlib.enabling.resultset.client.ResultSetClient;
|
||
|
import eu.dnetlib.msro.workflows.graph.Arc;
|
||
|
import eu.dnetlib.msro.workflows.nodes.AsyncJobNode;
|
||
|
import eu.dnetlib.msro.workflows.procs.Env;
|
||
|
import eu.dnetlib.msro.workflows.procs.Token;
|
||
|
import eu.dnetlib.msro.workflows.util.ResultsetProgressProvider;
|
||
|
import eu.dnetlib.msro.workflows.util.WorkflowsConstants;
|
||
|
import eu.dnetlib.rmi.common.ResultSet;
|
||
|
import eu.dnetlib.rmi.manager.MSROException;
|
||
|
import org.apache.commons.logging.Log;
|
||
|
import org.apache.commons.logging.LogFactory;
|
||
|
import org.apache.http.NameValuePair;
|
||
|
import org.apache.http.client.entity.UrlEncodedFormEntity;
|
||
|
import org.apache.http.client.methods.CloseableHttpResponse;
|
||
|
import org.apache.http.client.methods.HttpPost;
|
||
|
import org.apache.http.entity.StringEntity;
|
||
|
import org.apache.http.impl.client.CloseableHttpClient;
|
||
|
import org.apache.http.impl.client.HttpClients;
|
||
|
import org.apache.http.impl.conn.PoolingHttpClientConnectionManager;
|
||
|
import org.apache.http.message.BasicNameValuePair;
|
||
|
import org.springframework.beans.factory.annotation.Autowired;
|
||
|
|
||
|
import java.io.IOException;
|
||
|
import java.net.ConnectException;
|
||
|
import java.util.List;
|
||
|
import java.util.Map;
|
||
|
import java.util.concurrent.ExecutorService;
|
||
|
import java.util.concurrent.Executors;
|
||
|
import java.util.concurrent.Future;
|
||
|
|
||
|
|
||
|
public class EnrichGraphDBContentJobNode extends AsyncJobNode {
|
||
|
|
||
|
private static final Log log = LogFactory.getLog(EnrichGraphDBContentJobNode.class);
|
||
|
|
||
|
private String eprParam;
|
||
|
|
||
|
@Autowired
|
||
|
private ResultSetClient resultSetClient;
|
||
|
|
||
|
private String sparqlUpdateQuery;
|
||
|
private String publisherEndpoint;
|
||
|
private String datasourceInterface;
|
||
|
private String datasource;
|
||
|
|
||
|
//for parallel requests to the publisher endpoint
|
||
|
private int nThreads = 5;
|
||
|
|
||
|
@Override
|
||
|
protected String execute(final Env env) throws Exception {
|
||
|
|
||
|
int statusCode = -1;
|
||
|
String enrichResult = "noResult";
|
||
|
log.info("Publisher endpoint: " + getPublisherEndpoint());
|
||
|
log.info("Enrich Query Value: " + getSparqlUpdateQuery());
|
||
|
PoolingHttpClientConnectionManager cm = new PoolingHttpClientConnectionManager();
|
||
|
cm.setMaxTotal(nThreads);
|
||
|
CloseableHttpClient client = HttpClients.custom().setConnectionManager(cm).build();
|
||
|
|
||
|
log.info("Enrich endpoint: " + getEnrichEndpoint());
|
||
|
CloseableHttpResponse responsePOST = null;
|
||
|
try {
|
||
|
HttpPost post = new HttpPost(getEnrichEndpoint());
|
||
|
List<NameValuePair> params = Lists.newArrayList();
|
||
|
String datasourceInterfaceValue = getDatasourceInterface();
|
||
|
StringEntity entity = new StringEntity(getSparqlUpdateQuery());
|
||
|
post.setEntity(entity);
|
||
|
responsePOST = client.execute(post);
|
||
|
statusCode = responsePOST.getStatusLine().getStatusCode();
|
||
|
switch (statusCode) {
|
||
|
case 200:
|
||
|
log.info("enrich graphDB content completed");
|
||
|
break;
|
||
|
default:
|
||
|
log.error("error enriching graphDB " + responsePOST.getStatusLine().getStatusCode() + ": " + responsePOST.getStatusLine().getReasonPhrase());
|
||
|
break;
|
||
|
}
|
||
|
} catch (ConnectException ce) {
|
||
|
throw new MSROException("unable to connect to Publisher endpoint" + getEnrichEndpoint());
|
||
|
}
|
||
|
catch (IOException e) {
|
||
|
log.error("IO error enriching graphDB ", e);
|
||
|
}
|
||
|
finally{
|
||
|
if(responsePOST != null) responsePOST.close();
|
||
|
client.close();
|
||
|
cm.shutdown();
|
||
|
}
|
||
|
|
||
|
env.setAttribute(WorkflowsConstants.MAIN_LOG_PREFIX + "statusCode", Integer.toString(statusCode));
|
||
|
env.setAttribute(WorkflowsConstants.MAIN_LOG_PREFIX + "enrichResult", enrichResult);
|
||
|
|
||
|
log.info("enriching completed");
|
||
|
|
||
|
return Arc.DEFAULT_ARC;
|
||
|
}
|
||
|
|
||
|
public String getPublisherEndpoint() {
|
||
|
return publisherEndpoint;
|
||
|
}
|
||
|
|
||
|
private String getEnrichEndpoint() {
|
||
|
return publisherEndpoint.concat("/updateSparql");
|
||
|
}
|
||
|
|
||
|
public void setPublisherEndpoint(final String publisherEndpoint) {
|
||
|
this.publisherEndpoint = publisherEndpoint;
|
||
|
}
|
||
|
|
||
|
public ResultSetClient getResultSetClient() {
|
||
|
return resultSetClient;
|
||
|
}
|
||
|
|
||
|
public void setResultSetClient(final ResultSetClient resultSetClient) {
|
||
|
this.resultSetClient = resultSetClient;
|
||
|
}
|
||
|
|
||
|
public String getEprParam() {
|
||
|
return eprParam;
|
||
|
}
|
||
|
|
||
|
public void setEprParam(String eprParam) {
|
||
|
this.eprParam = eprParam;
|
||
|
}
|
||
|
|
||
|
public String getDatasourceInterface() {
|
||
|
return datasourceInterface;
|
||
|
}
|
||
|
|
||
|
|
||
|
public void setDatasourceInterface(String datasourceInterface) {
|
||
|
this.datasourceInterface = datasourceInterface;
|
||
|
}
|
||
|
|
||
|
@Override
|
||
|
protected void beforeStart(Token token) {
|
||
|
token.setProgressProvider(new ResultsetProgressProvider(token.getEnv().getAttribute(getEprParam(), ResultSet.class), this.resultSetClient));
|
||
|
}
|
||
|
|
||
|
public String getDatasource() {
|
||
|
return datasource;
|
||
|
}
|
||
|
|
||
|
public void setDatasource(String datasource) {
|
||
|
this.datasource = datasource;
|
||
|
}
|
||
|
|
||
|
public String getSparqlUpdateQuery() {
|
||
|
return sparqlUpdateQuery;
|
||
|
}
|
||
|
|
||
|
public void setSparqlUpdateQuery(String sparqlUpdateQuery) {
|
||
|
this.sparqlUpdateQuery = sparqlUpdateQuery;
|
||
|
}
|
||
|
}
|