statoolinfosweb/src/fr/devinsy/statoolinfos/core/StatoolInfos.java

/*
 * Copyright (C) 2020 Christian Pierre MOMON <christian@momon.org>
 * 
 * This file is part of StatoolInfos, simple service statistics tool.
 * 
 * StatoolInfos is free software: you can redistribute it and/or modify
 * it under the terms of the GNU Affero General Public License as
 * published by the Free Software Foundation, either version 3 of the
 * License, or (at your option) any later version.
 * 
 * StatoolInfos is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU Affero General Public License for more details.
 * 
 * You should have received a copy of the GNU Affero General Public License
 * along with StatoolInfos.  If not, see <http://www.gnu.org/licenses/>.
 */
package fr.devinsy.statoolinfos.core;

import java.io.File;
import java.io.IOException;
import java.net.URL;
import java.time.LocalDateTime;

import org.apache.commons.lang3.StringUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import fr.devinsy.statoolinfos.htmlize.Htmlizer;
import fr.devinsy.statoolinfos.properties.PathProperties;
import fr.devinsy.statoolinfos.properties.PathProperty;
import fr.devinsy.statoolinfos.properties.PathPropertyList;
import fr.devinsy.statoolinfos.properties.PathPropertyUtils;

/**
 * The Class StatoolInfos.
 */
public class StatoolInfos
{
    private static Logger logger = LoggerFactory.getLogger(StatoolInfos.class);

    /**
     * Builds the.
     *
     * @param configurationFile
     *            the input
     * @throws StatoolInfosException
     *             the statool infos exception
     * @throws IOException
     *             Signals that an I/O exception has occurred.
     */
    public static void build(final File configurationFile) throws StatoolInfosException, IOException
    {
        logger.info("Build {}", configurationFile.getAbsolutePath());

        Configuration configuration = Factory.loadConfiguration(configurationFile);
        logger.info("Build input setting: {}", configuration.getBuildInput());
        logger.info("Build directory setting: {}", configuration.getBuildDirectoryPath());

        File inputFile = configuration.getBuildInput();
        File buildDirectory = configuration.getBuildDirectory();
        if (inputFile == null)
        {
            throw new StatoolInfosException("Input is undefined.");
        }
        else if (!inputFile.exists())
        {
            throw new StatoolInfosException("Input does not exist.");
        }
        else if (!inputFile.isFile())
        {
            throw new StatoolInfosException("Input is not a file.");
        }
        else if (buildDirectory == null)
        {
            throw new StatoolInfosException("Build directory is undefined.");
        }
        else if (!buildDirectory.exists())
        {
            throw new StatoolInfosException("Build directory does not exist.");
        }
        else if (!buildDirectory.isDirectory())
        {
            throw new StatoolInfosException("Build directory is not a directory.");
        }
        else
        {
            // Build file section.
            PathProperties target = new PathPropertyList();
            target.put("file.class", configuration.get("conf.class"));
            target.put("file.generator", "StatoolInfos");
            target.put("file.datetime", LocalDateTime.now().toString());
            target.put("file.protocol", configuration.get("conf.protocol"));

            // Load input properties.
            PathProperties input = PathPropertyUtils.load(inputFile);

            // Add input properties with file section ones.
            target.add(input);

            // Save the build properties.
            File targetFile = new File(buildDirectory, configurationFile.getName());
            PathPropertyUtils.save(targetFile, target);
        }
    }

    /**
     * Clear.
     *
     * @param configurationFile
     *            the input
     * @throws StatoolInfosException
     *             the statool infos exception
     * @throws IOException
     *             Signals that an I/O exception has occurred.
     */
    public static void clear(final File configurationFile) throws StatoolInfosException, IOException
    {
        Configuration configuration = Factory.loadConfiguration(configurationFile);

        {
            logger.info("Build directory setting: {}", configuration.getBuildDirectoryPath());

            String path = configuration.getBuildDirectoryPath();
            if (StringUtils.isBlank(path))
            {
                logger.warn("Undefined build directory.");
            }
            else if (!new File(path).exists())
            {
                logger.warn("Build directory does not exist: {}.", path);
            }
            else
            {
                File buildDirectory = configuration.getBuildDirectory();

                for (File file : buildDirectory.listFiles())
                {
                    if ((file.isFile()) && (StringUtils.endsWithAny(file.getName(), ".properties")))
                    {
                        logger.info("Deleting " + file.getName());
                        file.delete();
                    }
                }
            }
        }

        {
            logger.info("Cache setting: {}", configuration.getCrawlCachePath());

            String path = configuration.getCrawlCachePath();
            if (StringUtils.isBlank(path))
            {
                logger.warn("Undefined crawl cache.");
            }
            else if (!new File(path).exists())
            {
                logger.warn("Crawl cache does not exist: {}.", path);
            }
            else
            {
                CrawlCache cache = configuration.getCrawlCache();
                cache.clear();
            }
        }

        {
            logger.info("Htmlize directory setting: {}", configuration.getHtmlizeDirectoryPath());

            String htmlDirectoryPath = configuration.getHtmlizeDirectoryPath();
            if (StringUtils.isBlank(htmlDirectoryPath))
            {
                logger.warn("Undefined htmlize directory.");
            }
            else if (!new File(htmlDirectoryPath).exists())
            {
                logger.warn("Htmlize directory does not exist: {}.", htmlDirectoryPath);
            }
            else
            {
                File htmlizeDirectory = configuration.getHtmlizeDirectory();

                for (File file : htmlizeDirectory.listFiles())
                {
                    if ((file.isFile()) && (StringUtils.endsWithAny(file.getName(), ".properties", ".js", ".html", ".ico", ".css", ".jpg", ".xhtml")))
                    {
                        logger.info("Deleting " + file.getName());
                        file.delete();
                    }
                }
            }
        }
    }

    /**
     * Crawl.
     *
     * @param configurationFile
     *            the input
     * @throws StatoolInfosException
     *             the statool infos exception
     * @throws IOException
     *             Signals that an I/O exception has occurred.
     */
    public static void crawl(final File configurationFile) throws StatoolInfosException, IOException
    {
        Configuration configuration = Factory.loadConfiguration(configurationFile);

        logger.info("Crawl input setting: {}", configuration.getCrawlInputPath());
        logger.info("Crawl cache setting: {}", configuration.getCrawlCachePath());

        CrawlCache cache = configuration.getCrawlCache();

        PathProperties input = PathPropertyUtils.load(configuration.getCrawlInput());

        cache.storeQuietly(input.getURL("federation.logo"));
        cache.storeQuietly(input.getURL("organization.logo"));

        PathProperties subs = input.getByPrefix("subs");
        for (PathProperty property : subs)
        {
            URL url = new URL(property.getValue());
            crawl(url, cache);
        }
    }

    /**
     * Crawl.
     *
     * @param url
     *            the input
     * @param cache
     *            the cache
     * @throws StatoolInfosException
     *             the statool infos exception
     * @throws IOException
     *             Signals that an I/O exception has occurred.
     */
    public static void crawl(final URL url, final CrawlCache cache) throws StatoolInfosException, IOException
    {
        logger.info("Crawling " + url);

        File file = cache.store(url);
        PathProperties properties = PathPropertyUtils.load(file);

        PathProperties crawlSection = new PathPropertyList();
        crawlSection.put("crawl.crawler", "StatoolInfos");
        crawlSection.put("crawl.datetime", LocalDateTime.now().toString());
        crawlSection.put("crawl.url", url.toString());
        properties.add(crawlSection);
        cache.storeProperties(url, properties);

        cache.storeQuietly(properties.getURL("organization.logo"));
        cache.storeQuietly(properties.getURL("service.logo"));

        //
        PathProperties subs = properties.getByPrefix("subs");
        for (PathProperty property : subs)
        {
            URL subUrl = new URL(property.getValue());
            crawl(subUrl, cache);
        }
    }

    /**
     * Htmlize.
     *
     * @param configurationFile
     *            the input
     * @throws StatoolInfosException
     *             the statool infos exception
     * @throws IOException
     *             Signals that an I/O exception has occurred.
     */
    public static void htmlize(final File configurationFile) throws StatoolInfosException, IOException
    {
        Configuration configuration = Factory.loadConfiguration(configurationFile);

        logger.info("Cache setting: {}", configuration.getCrawlCachePath());
        logger.info("Htmlize input setting: {}", configuration.getHtmlizeInputPath());
        logger.info("Htmlize directory setting: {}", configuration.getHtmlizeDirectoryPath());

        File htmlizeInput = configuration.getHtmlizeInput();
        File htmlizeDirectory = configuration.getHtmlizeDirectory();
        if (htmlizeInput == null)
        {
            throw new IllegalArgumentException("Htmlize input undefined.");
        }
        else if (!htmlizeInput.exists())
        {
            throw new IllegalArgumentException("Htmlize input is missing.");
        }
        else if (htmlizeInput.isDirectory())
        {
            throw new IllegalArgumentException("Htmlize input is a directory.");
        }
        else if (htmlizeDirectory == null)
        {
            throw new IllegalArgumentException("Htmlize directory undefined.");
        }
        else if (!htmlizeDirectory.exists())
        {
            throw new IllegalArgumentException("Htmlize directory is missing.");
        }
        else if (!htmlizeDirectory.isDirectory())
        {
            throw new IllegalArgumentException("Htmlize directory is not a directory.");
        }
        else
        {
            if (configuration.isFederation())
            {
                Htmlizer.htmlizeFederation(configuration);
            }
            else if (configuration.isOrganization())
            {
                Htmlizer.htmlizeOrganisation(configuration);
            }
            else
            {
                logger.warn("No htmlize for this input: {}.", configuration.getClassName());
            }
        }
    }
}