dnet-core/dnet-data-services/src/main/java/eu/dnetlib/data/collector/plugins/datasets/DatasetsByProjectIterator.java

package eu.dnetlib.data.collector.plugins.datasets;

import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.util.Iterator;
import java.util.Map;

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;

import com.google.common.collect.Maps;

/**
 * The Class DatasetsByProjectIterator.
 */
public class DatasetsByProjectIterator implements Iterable<String>, Iterator<String> {

	private static final String SPLIT_REGEX = ";";

	/** The project id key. */
	public static String PROJECT_ID_KEY = "id";

	/** The project name key. */
	public static String PROJECT_NAME_KEY = "name";

	/** The project corda id key. */
	public static String PROJECT_CORDA_ID_KEY = "corda_id";

	/** The current iterator. */
	private Iterator<String> currentIterator;

	/** The csv reader. */
	private BufferedReader csvReader;

	/** The current project. */
	private Map<String, String> currentProject;

	/** The logger. */
	private static final Log log = LogFactory.getLog(DatasetsByProjectIterator.class);

	/**
	 * Instantiates a new datasets by project iterator.
	 *
	 * @param csvInputStream
	 *            the csv input stream
	 * @throws IOException
	 *             Signals that an I/O exception has occurred.
	 */
	public DatasetsByProjectIterator(final InputStreamReader csvInputStream) throws IOException {
		this.csvReader = new BufferedReader(csvInputStream);
		this.currentProject = extractNextLine();
	}

	/*
	 * (non-Javadoc)
	 * 
	 * @see java.util.Iterator#hasNext()
	 */
	@Override
	public boolean hasNext() {
		// CASE WHEN WE REACH THE LAST ITEM ON CSV
		// OR WE HAD SOME PROBLEM ON GET NEXT CSV ITEM
		if (this.currentProject == null) { return false; }
		// IN THIS CASE WE HAVE ANOTHER DATASETS
		// FOR THE CURRENT PROJECT AND RETURN TRUE
		if (currentIterator != null && currentIterator.hasNext()) { return true; }
		// OTHERWISE WE FINISHED TO ITERATE THE CURRENT
		// SETS OF DATASETS FOR A PARTICULAR PROJECT
		// SO WE HAVE TO RETRIEVE THE NEXT ITERATOR WITH
		// ITEMS
		this.currentProject = extractNextLine();

		while (this.currentProject != null) {
			currentIterator = getNextIterator();
			// IF THE NEXT ITERATOR HAS ITEMS RETURN YES
			// OTHERWISE THE CICLE CONTINUE
			if (currentIterator.hasNext()) { return true; }
			this.currentProject = extractNextLine();
		}
		return false;

	}

	/*
	 * (non-Javadoc)
	 * 
	 * @see java.util.Iterator#next()
	 */
	@Override
	public String next() {
		return this.currentIterator.next();
	}

	/*
	 * (non-Javadoc)
	 * 
	 * @see java.util.Iterator#remove()
	 */
	@Override
	public void remove() {}

	/*
	 * (non-Javadoc)
	 * 
	 * @see java.lang.Iterable#iterator()
	 */
	@Override
	public Iterator<String> iterator() {
		if (this.currentProject != null) {
			currentIterator = getNextIterator();
			return this;
		}
		return null;

	}

	private Iterator<String> getNextIterator() {
		QueryField q = new QueryField();
		RequestField r = new RequestField();
		r.setQuery(q);
		q.getTerm().put("ft-techkeyword", this.currentProject.get(PROJECT_ID_KEY));
		return new DatasetsIterator(r, this.currentProject.get(PROJECT_CORDA_ID_KEY), null).iterator();
	}

	/**
	 * Extract next line.
	 *
	 * @return the map
	 * @throws IOException
	 *             Signals that an I/O exception has occurred.
	 */
	private Map<String, String> extractNextLine() {
		String line;
		try {
			line = this.csvReader.readLine();
		} catch (IOException e) {
			return null;
		}
		// WE REACH THE END OF THE CSV
		if (line == null) { return null; }
		log.debug("splitting line: " + line);
		String[] values = line.split(SPLIT_REGEX);
		if (values == null || values.length != 4) {
			log.error("Error on splitting line, the length must be 4");
			return null;
		}
		int id = Integer.parseInt(values[0]);
		String project_name = values[2];
		String cordaId = values[3];
		Map<String, String> splittedMap = Maps.newHashMap();
		splittedMap.put(PROJECT_CORDA_ID_KEY, cordaId);
		splittedMap.put(PROJECT_ID_KEY, "project" + id);
		splittedMap.put(PROJECT_NAME_KEY, project_name);
		log.debug(String.format("found project %s with id Corda: %s and id for API: %s", project_name, cordaId, "project" + id));
		return splittedMap;
	}
}