dnet-hadoop/dhp-workflows/dhp-usage-stats-update/src/main/java/eu/dnetlib/oa/graph/usagestats/export/PiwikDownloadLogs.java

package eu.dnetlib.oa.graph.usagestats.export;

import java.io.*;
import java.net.Authenticator;
import java.net.PasswordAuthentication;
import java.net.URL;
import java.net.URLConnection;
import java.security.cert.X509Certificate;
import java.sql.PreparedStatement;
import java.sql.ResultSet;
import java.sql.Statement;
import java.text.SimpleDateFormat;
import java.util.ArrayList;
import java.util.Calendar;
import java.util.Date;
import java.util.List;

import javax.net.ssl.HostnameVerifier;
import javax.net.ssl.HttpsURLConnection;
import javax.net.ssl.SSLContext;
import javax.net.ssl.SSLSession;
import javax.net.ssl.TrustManager;
import javax.net.ssl.X509TrustManager;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.json.simple.JSONArray;
import org.json.simple.JSONObject;
import org.json.simple.parser.JSONParser;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

/**
 * @author D. Pierrakos, S. Zoupanos
 */
public class PiwikDownloadLogs {

	private final String piwikUrl;
	private Date startDate;
	private final String tokenAuth;

	/*
	 * The Piwik's API method
	 */
	private final String APImethod = "?module=API&method=Live.getLastVisitsDetails";
	private final String format = "&format=json";

	private static final Logger logger = LoggerFactory.getLogger(PiwikDownloadLogs.class);

	public PiwikDownloadLogs(String piwikUrl, String tokenAuth) {
		this.piwikUrl = piwikUrl;
		this.tokenAuth = tokenAuth;

	}

	private String getPiwikLogUrl() {
		return "https://" + piwikUrl + "/";
	}

	private String getJson(String url) throws Exception {
		try {
			logger.info("Connecting to download the JSON: " + url);
			URL website = new URL(url);
			URLConnection connection = website.openConnection();

			// connection.setRequestProperty ("Authorization", "Basic "+encoded);
			StringBuilder response;
			try (BufferedReader in = new BufferedReader(new InputStreamReader(connection.getInputStream()))) {
				response = new StringBuilder();
				String inputLine;
				while ((inputLine = in.readLine()) != null) {
					response.append(inputLine);
//					response.append("\n");
				}
			}

//			System.out.println("response ====> " + response.toString());

			return response.toString();
		} catch (Exception e) {
			logger.error("Failed to get URL: " + e);
			throw new Exception("Failed to get URL: " + e.toString(), e);
		}
	}

	public void GetOpenAIRELogs(String repoLogsPath, String portalLogPath, String portalMatomoID) throws Exception {

		Statement statement = ConnectDB.getHiveConnection().createStatement();
		SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd");

		ResultSet rs = statement
			.executeQuery(
				"SELECT distinct piwik_id from " + ConnectDB.getStatsDBSchema()
					+ ".datasource where piwik_id is not null and piwik_id <> 0 order by piwik_id");

		// Getting all the piwikids in a list for logging reasons & limitting the list
		// to the max number of piwikids
		List<Integer> piwikIdToVisit = new ArrayList<Integer>();
		while (rs.next())
			piwikIdToVisit.add(rs.getInt(1));
		logger.info("Found the following piwikIds for download: " + piwikIdToVisit);

		if (ExecuteWorkflow.numberOfPiwikIdsToDownload > 0 &&
			ExecuteWorkflow.numberOfPiwikIdsToDownload <= piwikIdToVisit.size()) {
			logger.info("Trimming piwikIds list to the size of: " + ExecuteWorkflow.numberOfPiwikIdsToDownload);
			piwikIdToVisit = piwikIdToVisit.subList(0, ExecuteWorkflow.numberOfPiwikIdsToDownload);
		}

		logger.info("Downloading from repos with the followins piwikIds: " + piwikIdToVisit);

		// Setting the starting period
		Calendar start = (Calendar) ExecuteWorkflow.startingLogPeriod.clone();
		logger.info("Starting period for log download: " + sdf.format(start.getTime()));

		// Setting the ending period (last day of the month)
		Calendar end = (Calendar) ExecuteWorkflow.endingLogPeriod.clone();
		end.add(Calendar.MONTH, +1);
		end.add(Calendar.DAY_OF_MONTH, -1);
		logger.info("Ending period for log download: " + sdf.format(end.getTime()));

		for (int siteId : piwikIdToVisit) {

			logger.info("Now working on piwikId: " + siteId);

			PreparedStatement st = ConnectDB.DB_HIVE_CONNECTION
				.prepareStatement(
					"SELECT max(timestamp) FROM " + ConnectDB.getUsageStatsDBSchema()
						+ ".piwiklog WHERE source=? GROUP BY timestamp HAVING max(timestamp) is not null");
			st.setInt(1, siteId);

			ResultSet rs_date = st.executeQuery();
			while (rs_date.next()) {
				if (rs_date.getString(1) != null && !rs_date.getString(1).equals("null")
					&& !rs_date.getString(1).equals("")) {
					start.setTime(sdf.parse(rs_date.getString(1)));
				}
			}
			rs_date.close();

			for (Calendar currDay = (Calendar) start.clone(); currDay.before(end); currDay.add(Calendar.DATE, 1)) {
				Date date = currDay.getTime();
				logger.info("Downloading logs for repoid " + siteId + " and for " + sdf.format(date));

				String period = "&period=day&date=" + sdf.format(date);
				String outFolder = "";
				// portal siteId = 109;
				if (siteId == Integer.parseInt(portalMatomoID)) {
					outFolder = portalLogPath;
				} else {
					outFolder = repoLogsPath;
				}
				FileSystem fs = FileSystem.get(new Configuration());
				FSDataOutputStream fin = fs
					.create(new Path(outFolder + "/" + siteId + "_Piwiklog" + sdf.format((date)) + ".json"), true);

				String baseApiUrl = getPiwikLogUrl() + APImethod + "&idSite=" + siteId + period + format
					+ "&expanded=5&filter_limit=1000&token_auth=" + tokenAuth;
				String content = "";

				int i = 0;

				JSONParser parser = new JSONParser();
				while (!content.equals("[]")) {
					String apiUrl = baseApiUrl;

					if (i > 0) {
						apiUrl += "&filter_offset=" + (i * 1000);
					}

					content = getJson(apiUrl);

					JSONArray jsonArray = (JSONArray) parser.parse(content);
					for (Object aJsonArray : jsonArray) {
						JSONObject jsonObjectRaw = (JSONObject) aJsonArray;
						fin.write(jsonObjectRaw.toJSONString().getBytes());
						fin.writeChar('\n');
					}

					i++;
				}
				fin.close();
			}

		}
	}
}