dnet-applications/apps/dnet-is-application/src/main/java/eu/dnetlib/is/importer/OldProfilesImporter.java

package eu.dnetlib.is.importer;

import java.util.Date;

import javax.transaction.Transactional;

import org.apache.commons.lang3.StringUtils;
import org.dom4j.Document;
import org.dom4j.DocumentException;
import org.dom4j.DocumentHelper;
import org.dom4j.Node;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;

import eu.dnetlib.is.resource.model.SimpleResource;
import eu.dnetlib.is.resource.repository.SimpleResourceRepository;
import eu.dnetlib.is.util.InformationServiceException;
import eu.dnetlib.is.util.ResourceValidator;
import eu.dnetlib.is.util.XmlIndenter;
import eu.dnetlib.is.vocabulary.model.Synonym;
import eu.dnetlib.is.vocabulary.model.Vocabulary;
import eu.dnetlib.is.vocabulary.model.VocabularyTerm;
import eu.dnetlib.is.vocabulary.repository.VocabularyRepository;
import eu.dnetlib.is.vocabulary.repository.VocabularyTermRepository;

@Service
public class OldProfilesImporter {

	@Autowired
	private SimpleResourceRepository simpleResourceRepository;

	@Autowired
	private VocabularyRepository vocabularyRepository;

	@Autowired
	private VocabularyTermRepository vocabularyTermRepository;

	@Autowired
	private ResourceValidator resourceValidator;

	@Transactional
	public String importSimpleResource(final String xml) throws InformationServiceException {

		try {
			final Document doc = DocumentHelper.parseText(xml);

			final String id = StringUtils.substringBefore(doc.valueOf("//RESOURCE_IDENTIFIER/@value"), "_");
			final Date now = new Date();

			final SimpleResource res = new SimpleResource();
			res.setId(id);
			res.setCreationDate(now);
			res.setModificationDate(now);

			String resContent;
			switch (doc.valueOf("//RESOURCE_TYPE/@value")) {
			case "CleanerDSResourceType":
				res.setType("cleaning_rule");
				res.setName(doc.valueOf("//CLEANER_NAME"));
				res.setDescription(doc.valueOf("//CLEANER_DESCRIPTION"));
				resContent = XmlIndenter.indent(doc.selectSingleNode("//CLEANER_RULES"));
				break;
			case "TransformationRuleDSResourceType":
				res.setName(doc.valueOf("//SCRIPT/TITLE"));
				res.setDescription("");
				if (doc.selectNodes("//*[local-name() = 'stylesheet']").size() > 0) {
					res.setType("transformation_rule_xslt");
					resContent = XmlIndenter.indent(doc.selectSingleNode("//*[local-name() = 'stylesheet']"));
				} else {
					final String code = doc.valueOf("//SCRIPT/CODE").trim();

					try {
						final Document xsltDoc = DocumentHelper.parseText(code);
						res.setType("transformation_rule_xslt");
						resContent = XmlIndenter.indent(xsltDoc);
					} catch (final DocumentException e) {
						res.setType("transformation_rule_legacy");
						resContent = code;
					}
				}
				break;
			case "HadoopJobConfigurationDSResourceType":
				res.setType("hadoop_job_configuration");
				res.setName(doc.valueOf("//HADOOP_JOB/@name"));
				res.setDescription(doc.valueOf("//HADOOP_JOB/DESCRIPTION"));
				resContent = XmlIndenter.indent(doc.selectSingleNode("//HADOOP_JOB"));
				break;
			case "DedupConfigurationDSResourceType":
				res.setType("dedup_configuration");
				res.setName(doc.valueOf("//DESCRIPTION"));
				res.setDescription(doc.valueOf("//DESCRIPTION"));
				resContent = doc.valueOf("//DEDUPLICATION");
				break;
			default:
				throw new InformationServiceException("Invalid resource type: " + doc.valueOf("//RESOURCE_TYPE/@value"));
			}

			resourceValidator.validate(res.getType(), resContent.trim());

			simpleResourceRepository.save(res);
			simpleResourceRepository.setContentById(id, resContent.trim());

			return res.getId();
		} catch (final Exception e) {
			throw new InformationServiceException("Error parsing file", e);
		}
	}

	@Transactional
	public Vocabulary importVocabulary(final String xml) throws Exception {
		final Document doc = DocumentHelper.parseText(xml);

		final Vocabulary voc = new Vocabulary();
		final String vocId = doc.valueOf("//VOCABULARY_NAME/@code");
		final String vocName = doc.valueOf("//VOCABULARY_NAME");
		final String vocDesc = doc.valueOf("//VOCABULARY_DESCRIPTION");

		voc.setId(vocId);
		voc.setName(vocName);
		voc.setDescription(vocDesc);

		vocabularyRepository.save(voc);

		for (final Node n : doc.selectNodes("//TERM")) {
			final VocabularyTerm term = new VocabularyTerm();
			term.setVocabulary(vocId);
			term.setCode(n.valueOf("@code"));
			term.setName(n.valueOf("@english_name"));
			term.setEncoding(n.valueOf("@encoding"));
			term.setSynonyms(n.selectNodes(".//SYNONYM")
				.stream()
				.map(ns -> new Synonym(ns.valueOf("@term"), ns.valueOf("@encoding")))
				.sorted()
				.distinct()
				.toArray(Synonym[]::new));

			vocabularyTermRepository.save(term);
		}

		return voc;
	}

}