D-Net

package eu.dnetlib.parthenos.virtuoso;

import java.io.IOException;

import java.io.OutputStream;

import java.io.StringWriter;

import java.util.HashMap;

import java.util.Iterator;

import java.util.List;

import java.util.Map;

import com.google.common.collect.Iterators;

import com.google.common.collect.Lists;

import eu.dnetlib.parthenos.CRM;

import eu.dnetlib.parthenos.publisher.ParthenosPublisherException;

import freemarker.template.Configuration;

import freemarker.template.Template;

import freemarker.template.TemplateException;

import org.apache.commons.io.IOUtils;

import org.apache.commons.lang3.StringUtils;

import org.apache.commons.logging.Log;

import org.apache.commons.logging.LogFactory;

import org.apache.http.HttpEntity;

import org.apache.http.HttpHeaders;

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.impl.client.CloseableHttpClient;

import org.apache.http.impl.client.HttpClients;

import org.apache.http.message.BasicNameValuePair;

import org.apache.http.util.EntityUtils;

import org.apache.jena.query.ResultSet;

import org.apache.jena.sparql.engine.http.QueryEngineHTTP;

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

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

import org.springframework.http.HttpStatus;

import org.springframework.web.bind.annotation.*;

/**

 * Created by Alessia Bardi on 31/01/2018.

 * Read-only API for virtuoso.

 *

 * //TODO: error handling (http://www.springboottutorial.com/spring-boot-exception-handling-for-rest-services)

 * //TODO: pagination

 * //TODO swagger documentation?

 *

 * @author Alessia Bardi

 */

@RestController

public class VirtuosoReadAPI {

	private static final Log log = LogFactory.getLog(VirtuosoReadAPI.class);

	@Value("${virtuoso.sparqlurl}")

	private String sparqlUrl;

	@Value("${virtuoso.pwd}")

	private String username;

	@Value("${virtuoso.pwd}")

	private String password;

	@Value("${virtuoso.uri.base.default}")

	private String defaultBaseURI;

	@Value("${virtuoso.connectionstring")

	private String virtuosoConnectionString;

	@Autowired

	private Configuration freemarkerConfig;

	@RequestMapping(value = "/virtuoso/apiSubjectsWithType", produces = { "application/json" }, method = RequestMethod.GET)

	public List<String> getSubjectsForApiWithType(@RequestParam final String api, @RequestParam final String typeNamespace, @RequestParam final String typeName, @RequestParam final int limit, @RequestParam final int offset){

		String fullTypeName = typeNamespace + typeName;

		log.debug(String.format("Getting subjects of type %s for API %s limit %d offset %d", fullTypeName, api, limit, offset));

		//if I add the ORDER BY the query is too slow: let's hope we are not getting the same subjects over and over again

		String queryForSubjectsTemplate = "DEFINE input:inference 'parthenos_rules' SELECT DISTINCT ?s WHERE { GRAPH ?g {?s a <%s> .} . GRAPH <provenance> {?g <dnetcollectedFrom> <%s> .}} LIMIT %d OFFSET %d";

		String q = String.format(queryForSubjectsTemplate, fullTypeName, defaultBaseURI+api, limit, offset);

		final QueryEngineHTTP serviceRequest = new QueryEngineHTTP(sparqlUrl, q);

		ResultSet subjects = serviceRequest.execSelect();

		Iterator<String> s = Iterators.transform(subjects, qs -> qs.getResource("s").getURI());

		List<String> res = Lists.newArrayList(s);

		serviceRequest.close();

		return res;

	}

	/**

	 * Returns the paginated list of resource URI used as subjects in the context of a given api. All subject URIs are returned, except from CRM.E55_Type and CRM.E41_Appellation.

	 * TODO: probably we can remove this, as it is used only for testing.

	 * @param api

	 * @param limit

	 * @param offset

	 * @return a page of URIs that are subjects in the context of a given api

	 *

	 */

	@Deprecated

	@RequestMapping(value = "/virtuoso/apiSubjects", produces = { "application/json" }, method = RequestMethod.GET)

	public List<String> getSubjectsForApi(@RequestParam final String api, @RequestParam final int limit, @RequestParam final int offset){

		log.debug(String.format("Getting subjects for API %s, limit %d offset %d", api, limit, offset));

		//if I add the ORDER BY the query is too slow: let's hope we are not getting the same subjects over and over again

		String queryForSubjectsTemplate = "DEFINE input:inference 'parthenos_rules' SELECT DISTINCT ?s WHERE { GRAPH ?g {?s a ?t . FILTER (?t != <%s> && ?t != <%s>)} . GRAPH <provenance> {?g <dnetcollectedFrom> <%s> .}} LIMIT %d OFFSET %d ";

		String q = String.format(queryForSubjectsTemplate, CRM.E55_Type.getURI(), CRM.E41_Appellation.getURI(), defaultBaseURI+api, limit, offset);

		final QueryEngineHTTP serviceRequest = new QueryEngineHTTP(sparqlUrl, q);

		ResultSet subjects = serviceRequest.execSelect();

		Iterator<String> s = Iterators.transform(subjects, qs -> qs.getResource("s").getURI());

		List<String> res = Lists.newArrayList(s);

		serviceRequest.close();

		return res;

	}

	@RequestMapping(value = "/virtuoso/subject", produces = { "application/rdf+xml", "application/xml" }, method = RequestMethod.GET)

	@ResponseStatus(value = HttpStatus.OK)

	public void getSubject(@RequestParam final String subjectURL, @RequestParam final String typeName, @RequestParam final String timeoutMs, final OutputStream responseStream)

			throws IOException, TemplateException, ParthenosPublisherException {

		String templateName = typeName+".sparql";

		Template temp = freemarkerConfig.getTemplate(templateName);

		Map<String, String> values = new HashMap<>();

		values.put("subjectURL", subjectURL);

		StringWriter sw = new StringWriter();

		temp.process(values, sw);

		String q = sw.toString();

		log.debug("Querying for "+subjectURL+" with query "+templateName);

		sendConstructResponse(q, timeoutMs, responseStream);

	}

	protected void sendConstructResponse(final String query, final String timeoutMs, final OutputStream responseStream) throws IOException, ParthenosPublisherException {

		String res = executeSparqlPost(query, timeoutMs);

		IOUtils.write(res, responseStream);

	}

	protected String executeSparqlPost(final String query, final String timeoutMs) throws IOException, ParthenosPublisherException {

		try (CloseableHttpClient httpclient = HttpClients.createDefault()) {

			HttpPost httpPost = new HttpPost(sparqlUrl);

			httpPost.setHeader(HttpHeaders.ACCEPT, "application/rdf+xml");

			List<NameValuePair> nvps = Lists.newArrayList();

			nvps.add(new BasicNameValuePair("query", query));

			if (StringUtils.isNotBlank(timeoutMs)) {

				nvps.add(new BasicNameValuePair("timeout", timeoutMs));

			}

			httpPost.setEntity(new UrlEncodedFormEntity(nvps));

			try (CloseableHttpResponse response2 = httpclient.execute(httpPost)) {

				HttpEntity entity2 = response2.getEntity();

				String res = IOUtils.toString(entity2.getContent());

				EntityUtils.consume(entity2);

				int statusCode = response2.getStatusLine().getStatusCode();

				switch(statusCode){

				case 200:

					return res;

				case 504:

					String msg = String.format("ERROR 504 on query %s", query);

					throw new ParthenosPublisherException(msg);

				default:

					String errorMessage = String.format("ERROR HTTP STATUS CODE %d, REASON PHRASE: %s\n ERROR BODY: %s", statusCode, response2.getStatusLine(), res);

					log.error(errorMessage);

					throw new ParthenosPublisherException(errorMessage);

				}

			}

		}

	}

}