D-Net

package eu.dnetlib.msro.workflows.nodes.is;

import java.io.StringReader;

import java.util.List;

import javax.xml.XMLConstants;

import javax.xml.transform.stream.StreamSource;

import javax.xml.validation.Schema;

import javax.xml.validation.SchemaFactory;

import javax.xml.validation.Validator;

import eu.dnetlib.enabling.locators.UniqueServiceLocator;

import eu.dnetlib.msro.workflows.graph.Arc;

import eu.dnetlib.msro.workflows.nodes.SimpleJobNode;

import eu.dnetlib.msro.workflows.procs.Env;

import eu.dnetlib.msro.workflows.procs.Token;

import eu.dnetlib.msro.workflows.util.ProgressProvider;

import eu.dnetlib.msro.workflows.util.WorkflowsConstants;

import eu.dnetlib.rmi.enabling.ISLookUpService;

import eu.dnetlib.rmi.manager.MSROException;

import org.apache.commons.logging.Log;

import org.apache.commons.logging.LogFactory;

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

public class ValidateProfilesJobNode extends SimpleJobNode {

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

	@Autowired

	private UniqueServiceLocator serviceLocator;

	private int total = 0;

	private int current = 0;

	@Override

	protected void beforeStart(final Token token) {

		token.setProgressProvider(new ProgressProvider() {

			@Override

			public String getProgressDescription() {

				return ValidateProfilesJobNode.this.current + " / " + ValidateProfilesJobNode.this.total;

			}

		});

	}

	@Override

	protected String execute(final Env env) throws Exception {

		final ISLookUpService lookup = this.serviceLocator.getService(ISLookUpService.class);

		final SchemaFactory schemaFactory = SchemaFactory.newInstance(XMLConstants.W3C_XML_SCHEMA_NS_URI);

		int invalidTotals = 0;

		final List<String> list = lookup.listResourceTypes();

		this.total = list.size();

		this.current = 0;

		for (final String resourceType : list) {

			int valid = 0;

			int invalid = 0;

			final String schemaSource = lookup.getResourceTypeSchema(resourceType);

			final Schema schema = schemaFactory.newSchema(new StreamSource(new StringReader(schemaSource)));

			final Validator validator = schema.newValidator();

			for (final String profile : lookup.quickSearchProfile("/RESOURCE_PROFILE[./HEADER/RESOURCE_TYPE/@value='" + resourceType + "']")) {

				try {

					validator.validate(new StreamSource(new StringReader(profile)));

					valid++;

				} catch (final Exception e) {

					invalid++;

				}

			}

			invalidTotals += invalid;

			final String message = String.format("Valid: %s, Invalid: %s, Total: %s", valid, invalid, valid + invalid);

			env.setAttribute(WorkflowsConstants.MAIN_LOG_PREFIX + resourceType, message);

			log.info("Validation of " + resourceType + " profiles: " + message);

		}

		if (invalidTotals > 0) { throw new MSROException("Validation wf has found " + invalidTotals + " invalid profiles"); }

		return Arc.DEFAULT_ARC;

	}

}