getSets method returns timeout if the process of getting all the sets from the oai server is not finished within 20 seconds.

This commit is contained in:
Katerina 2023-11-20 14:36:46 +02:00
parent 59fdd724e9
commit 88a57a8b85
4 changed files with 126 additions and 101 deletions

View File

@ -14,6 +14,7 @@ import eu.dnetlib.validatorapi.repositories.ValidationResultRepository;
import eu.dnetlib.validatorapi.routes.FairOaiPmhRoute2;
import eu.dnetlib.validatorapi.routes.SimpleOaiPmhRoute;
import org.apache.camel.CamelContext;
import org.apache.camel.Exchange;
import org.apache.camel.ProducerTemplate;
import org.apache.camel.builder.RouteBuilder;
import org.apache.http.HttpStatus;
@ -150,7 +151,6 @@ public class ValidationController {
//System.out.println("Final validation job "+ validationJob.hashCode());
validationJob.recordsTested = record;
validationJob.score = resultSum / record;
//TODO uncomment
log.info("Saving validation job " + validationJob.recordsTested);
validationJobRepository.save(validationJob);
}
@ -225,11 +225,14 @@ public class ValidationController {
if (fairProfile == null) {
oaiPmhRouteBuilder =
new SimpleOaiPmhRoute(oaiEndpoint, profile, validationJob, numberOfRecords, uuid.toString(),
validationResultRepository, summaryValidationJobRepository);
validationJobRepository, validationResultRepository, summaryValidationJobRepository);
// new SimpleOaiPmhRoute(oaiEndpoint, profile, validationJob, numberOfRecords, uuid.toString());
} else {
oaiPmhRouteBuilder =
new FairOaiPmhRoute2(oaiEndpoint, profile, validationJob, numberOfRecords, uuid.toString());
/* new FairOaiPmhRoute2(oaiEndpoint, profile, validationJob, numberOfRecords, uuid.toString(),
validationJobRepository, validationResultRepository, summaryValidationJobRepository);*/
}
@ -258,14 +261,20 @@ public class ValidationController {
camelContext.addRoutes(oaiSetListRoute);
}*/
@RequestMapping(value = {"/getSets"}, method = RequestMethod.GET, produces = MediaType.APPLICATION_JSON_VALUE)
public String getSets(@RequestParam(name = "baseUrl", defaultValue = "http://repositorium.sdum.uminho.pt/oai/request") String baseURL //not in use now
@RequestMapping(value = {"/getSets"}, method = RequestMethod.GET, produces = MediaType.APPLICATION_JSON_VALUE)
public ResponseEntity<String> getSets(@RequestParam(name = "baseUrl", defaultValue = "http://repositorium.sdum.uminho.pt/oai/request") String baseURL //not in use now
) throws Exception {
ProducerTemplate producerTemplate = camelContext.createProducerTemplate();
String response = producerTemplate.requestBodyAndHeader("direct:getResponse", null, "endpoint", "oaipmh://"+ baseURL + "?verb=ListSets", String.class);
/*String response = producerTemplate.requestBodyAndHeader("direct:getResponse", null, "endpoint", "oaipmh://"+ baseURL + "?verb=ListSets", String.class);*/
Exchange message = producerTemplate.request("direct:getResponse", exchange -> {
exchange.getMessage().setHeader("endpoint", "oaipmh://" + baseURL + "?verb=ListSets");
});
return response;
return ResponseEntity
//TODO: maybe something better if header is not set (null and mot 200)
.status(message.getIn().getHeader(Exchange.HTTP_RESPONSE_CODE, 200, Integer.class))
.body(message.getIn().getBody(String.class));
/*git git
System.out.println(camelContext.getRoutes());

View File

@ -0,0 +1,90 @@
package eu.dnetlib.validatorapi.processors;
import org.apache.camel.Exchange;
import org.apache.camel.Processor;
import org.apache.camel.support.DefaultMessage;
import org.apache.http.HttpStatus;
import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.json.JSONObject;
import org.json.XML;
import org.w3c.dom.Document;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.xml.sax.InputSource;
import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.transform.OutputKeys;
import javax.xml.transform.Transformer;
import javax.xml.transform.TransformerException;
import javax.xml.transform.TransformerFactory;
import javax.xml.transform.dom.DOMSource;
import javax.xml.transform.stream.StreamResult;
import java.io.StringReader;
import java.io.StringWriter;
import java.util.ArrayList;
import java.util.List;
import java.util.concurrent.TimeoutException;
public class SetsProcessor implements Processor {
private final Logger log = LogManager.getLogger(this.getClass());
@Override
public void process(Exchange exchange) throws Exception {
List<DefaultMessage> messages = exchange.getIn().getBody(ArrayList.class);
//If the body is null a timeout has occurred.
if (exchange.getIn().getBody() == null)
throw new TimeoutException();
List<String> ListSetsXml = exchange.getIn().getBody(ArrayList.class);
List<String> desiredJsonParts = new ArrayList<>();
String xmlSets = "";
// Extract the desired <set> elements from each XML string
for (String xml : ListSetsXml) {
List<String> extractedSets = extractSets(xml);
for (String setXml : extractedSets)
xmlSets += setXml;
}
JSONObject jsonObject = XML.toJSONObject(xmlSets);
String jsonString = jsonObject.toString();
exchange.getIn().setHeader(Exchange.HTTP_RESPONSE_CODE, HttpStatus.SC_OK);
exchange.getIn().setBody(jsonString);
}
private List<String> extractSets(String xml) {
List<String> setXmls = new ArrayList<>();
try {
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
DocumentBuilder builder = factory.newDocumentBuilder();
InputSource inputSource = new InputSource(new StringReader(xml));
Document document = builder.parse(inputSource);
NodeList setElements = document.getElementsByTagName("set");
for (int i = 0; i < setElements.getLength(); i++) {
Node setElement = setElements.item(i);
String setXml = nodeToString(setElement);
setXmls.add(setXml);
}
} catch (Exception e) {
// TODO: consider if Ineed to handle any exceptions that occur during
// XML parsing or processing and not only log them
log.error("Could not parse the xml ", e);
}
return setXmls;
}
private String nodeToString(Node node) throws TransformerException {
StringWriter sw = new StringWriter();
TransformerFactory tf = TransformerFactory.newInstance();
Transformer transformer = tf.newTransformer();
transformer.setOutputProperty(OutputKeys.OMIT_XML_DECLARATION, "yes");
transformer.transform(new DOMSource(node), new StreamResult(sw));
return sw.toString();
}
}

View File

@ -0,0 +1,2 @@
package eu.dnetlib.validatorapi.processors;public class SumUpNCleanProcessor {
}

View File

@ -1,32 +1,21 @@
package eu.dnetlib.validatorapi.routes;
import eu.dnetlib.validatorapi.processors.SetsProcessor;
import org.apache.camel.Exchange;
import org.apache.camel.TypeConversionException;
import org.apache.camel.builder.RouteBuilder;
import org.apache.camel.impl.engine.DefaultShutdownStrategy;
import org.apache.http.HttpStatus;
import org.apache.http.client.ClientProtocolException;
import org.springframework.stereotype.Component;
import org.w3c.dom.Document;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.xml.sax.InputSource;
import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.transform.OutputKeys;
import javax.xml.transform.Transformer;
import javax.xml.transform.TransformerException;
import javax.xml.transform.TransformerFactory;
import javax.xml.transform.dom.DOMSource;
import javax.xml.transform.stream.StreamResult;
import java.io.StringReader;
import java.io.StringWriter;
import java.net.UnknownHostException;
import java.util.ArrayList;
import java.util.List;
import java.util.concurrent.TimeoutException;
@Component
public class OaiSetListRoute extends RouteBuilder {
@Override
public void configure() throws Exception {
@ -56,88 +45,23 @@ public class OaiSetListRoute extends RouteBuilder {
.handled(true)
.end();
from("direct:getResponse")
.log("\n\n\n LIST ")
.end();
onException(TimeoutException.class)
.handled(true)
.process(exchange -> {
exchange.getMessage().setHeader(Exchange.HTTP_RESPONSE_CODE, HttpStatus.SC_REQUEST_TIMEOUT);
exchange.getMessage().setHeader(Exchange.CONTENT_TYPE, "application/json");
exchange.getIn().setBody("{\"error_code\":\"408\",\"message\":\"Unable to retrieve sets. Server response takes too long.\"}");
});
/*from("direct:getResponse")
from("direct:getResponse")
.routeId("myRoute")
.process(exchange -> {
String endpoint = exchange.getIn().getHeader("endpoint", String.class);
exchange.getIn().setHeader("dynamicEndpoint", endpoint);
})
.recipientList(header("dynamicEndpoint"))
.process(exchange -> {
List<String> ListSetsXml = exchange.getIn().getBody(ArrayList.class);
List<String> desiredJsonParts = new ArrayList<>();
String xmlSets = "";
// Extract the desired <set> elements from each XML string
for (String xml : ListSetsXml) {
List<String> extractedSets = extractSets(xml);
for (String setXml : extractedSets)
xmlSets += setXml;
}
JSONObject jsonObject = XML.toJSONObject(xmlSets);
String jsonString = jsonObject.toString();
exchange.getIn().setBody(jsonString);
})
.end();*/
}
/*from("direct:processSets")
.process(exchange -> {
// Access the individual response from the ArrayList
ArrayList<?> responseList = exchange.getIn().getBody(ArrayList.class);
InputStream responseStream = new ByteArrayInputStream(responseList.get(0).toString().getBytes());
// Set the converted InputStream as the new body
exchange.getIn().setBody(responseStream);
}).log("${body}")
.split(xpath("//*[local-name()='set']")).log("${body}")
// .log("${body}")
.aggregate(constant(true), new ListSetAggregationStrategy())
.completionPredicate(exchangeProperty(Exchange.AGGREGATED_SIZE).isEqualTo(header("CamelSplitSize")))
.to("direct:processAggregatedSets")
.log("${body}");*/
/*
from("direct:processAggregatedSets")
.process(exchange -> {
String aggregatedXml = exchange.getIn().getBody(ArrayList.class) + "";
JSONObject jsonObject = XML.toJSONObject(aggregatedXml);
String jsonString = jsonObject.toString();
exchange.getIn().setBody(jsonString);
}).to("seda:result");*/
//}
private List<String> extractSets(String xml) {
List<String> setXmls = new ArrayList<>();
try {
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
DocumentBuilder builder = factory.newDocumentBuilder();
InputSource inputSource = new InputSource(new StringReader(xml));
Document document = builder.parse(inputSource);
NodeList setElements = document.getElementsByTagName("set");
for (int i = 0; i < setElements.getLength(); i++) {
Node setElement = setElements.item(i);
String setXml = nodeToString(setElement);
setXmls.add(setXml);
}
} catch (Exception e) {
// Handle any exceptions that occur during XML parsing or processing
log.error(e.getMessage());
}
return setXmls;
}
private String nodeToString(Node node) throws TransformerException {
StringWriter sw = new StringWriter();
TransformerFactory tf = TransformerFactory.newInstance();
Transformer transformer = tf.newTransformer();
transformer.setOutputProperty(OutputKeys.OMIT_XML_DECLARATION, "yes");
transformer.transform(new DOMSource(node), new StreamResult(sw));
return sw.toString();
}).recipientList(header("dynamicEndpoint"))
.parallelProcessing()
.timeout(20000)
.process(new SetsProcessor())
.end();
}
}