eu.fbk.twm.wiki.xmldump.WikipediaFirstSentenceExtractor Maven / Gradle / Ivy

Go to download
/*
 * Copyright (2013) Fondazione Bruno Kessler (http://www.fbk.eu/)
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *   http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package eu.fbk.twm.wiki.xmldump;

import de.tudarmstadt.ukp.wikipedia.parser.*;
import eu.fbk.twm.utils.CharacterTable;
import eu.fbk.twm.utils.Defaults;
import eu.fbk.twm.utils.ExtractorParameters;
import eu.fbk.twm.utils.WikipediaExtractor;
import eu.fbk.twm.wiki.xmldump.util.WikiMarkupParser;
import org.apache.commons.cli.*;
import org.apache.log4j.Logger;
import org.apache.log4j.PropertyConfigurator;

import java.io.*;
import java.util.Date;
import java.util.List;
import java.util.Locale;

//todo: check if it can be done in preprocessing
public class WikipediaFirstSentenceExtractor extends AbstractWikipediaExtractor implements WikipediaExtractor {
	/**
	 * Define a static logger variable so that it references the
	 * Logger instance named WikipediaAbstractExtractor.
	 */
	static Logger logger = Logger.getLogger(WikipediaFirstSentenceExtractor.class.getName());

	private PrintWriter abstractWriter;

	public WikipediaFirstSentenceExtractor(int numThreads, int numPages, Locale locale) {
		super(numThreads, numPages, locale);
	}

	@Override
	public void start(ExtractorParameters extractorParameters) {
		try {
			abstractWriter = new PrintWriter(new BufferedWriter(new OutputStreamWriter(new FileOutputStream(extractorParameters.getWikipediaFirstSentenceFileName()), "UTF-8")));
		} catch (IOException e) {
			logger.error(e);
		}

		startProcess(extractorParameters.getWikipediaXmlFileName());
	}

	@Override
	public void filePage(String text, String title, int wikiID) {
		//To change body of implemented methods use File | Settings | File Templates.
	}

	@Override
	public void categoryPage(String text, String title, int wikiID) {
	}

	@Override
	public void contentPage(String text, String title, int wikiID) {
		try {
			String s = getLinksForFirstSentence(text, title);
			synchronized (this) {
				abstractWriter.print(s);
			}

		} catch (IOException e) {
			logger.error(e);
		}
	}

	/**
	 * Returns the section titles. From this...
	 */
	public String getLinksForFirstSentence(String text, String title) throws IOException {

		//todo: add Bio for Italian

		StringBuilder sb = new StringBuilder();
		WikiMarkupParser wikiMarkupParser = WikiMarkupParser.getInstance();
		String[] prefixes = {imagePrefix, filePrefix};
		ParsedPage parsedPage = wikiMarkupParser.parsePage(text, prefixes);

		try {
			List sectionList = parsedPage.getSections();
			if (sectionList.size() > 0) {
				Section section = sectionList.get(0);

				List paragraphs = section.getParagraphs();
				Paragraph firstPar = new Paragraph();

				parLoop:
				for (Paragraph p : paragraphs) {
					if (p.getText().trim().length() == 0) {
						continue;
					}

					List spans = p.getFormatSpans(Content.FormatType.ITALIC);
					for (Span s : spans) {
						if (s.getStart() == 0) {
							continue parLoop;
						}
					}

					if (p.getText().trim().startsWith(":")) {
						continue;
					}

					firstPar = p;
					break;
				}

				List links = firstPar.getLinks();
				for (Link l : links) {
					String target = l.getTarget();
					if (target.startsWith("#")) {
						continue;
					}
					String[] parts = target.split("#");
					if (parts.length > 1) {
						target = parts[0];
					}
					target = normalizePageName(target);

					sb.append(title);
					sb.append(CharacterTable.HORIZONTAL_TABULATION);
					sb.append(target);
					sb.append(CharacterTable.LINE_FEED);
				}

//					sb.append(title);
//					sb.append(CharacterTable.HORIZONTAL_TABULATION);
//					sb.append(section.getText().replace(CharacterTable.LINE_FEED, CharacterTable.SPACE).trim());
//					sb.append(CharacterTable.LINE_FEED);
			}
		} catch (Exception e) {
			logger.warn(e.getMessage());
		}

		return sb.toString();
	}

	@Override
	public void disambiguationPage(String text, String title, int wikiID) {
		//To change body of implemented methods use File | Settings | File Templates.
	}

	@Override
	public void templatePage(String text, String title, int wikiID) {
		//To change body of implemented methods use File | Settings | File Templates.
	}

	@Override
	public void redirectPage(String text, String title, int wikiID) {
		//To change body of implemented methods use File | Settings | File Templates.
	}

	@Override
	public void portalPage(String text, String title, int wikiID) {
		//To change body of implemented methods use File | Settings | File Templates.
	}

	@Override
	public void projectPage(String text, String title, int wikiID) {
		//To change body of implemented methods use File | Settings | File Templates.
	}

	@Override
	public void endProcess() {
		super.endProcess();
		abstractWriter.flush();
		abstractWriter.close();
	}

	public static void main(String args[]) throws IOException {
		String logConfig = System.getProperty("log-config");
		if (logConfig == null) {
			logConfig = "configuration/log-config.txt";
		}

		PropertyConfigurator.configure(logConfig);

		Options options = new Options();
		try {
			Option wikipediaDumpOpt = OptionBuilder.withArgName("file").hasArg().withDescription("wikipedia xml dump file").isRequired().withLongOpt("wikipedia-dump").create("d");
			Option outputDirOpt = OptionBuilder.withArgName("dir").hasArg().withDescription("output directory in which to store output files").isRequired().withLongOpt("output-dir").create("o");
			Option numThreadOpt = OptionBuilder.withArgName("int").hasArg().withDescription("number of threads (default " + Defaults.DEFAULT_THREADS_NUMBER
                    + ")").withLongOpt("num-threads").create("t");
			Option numPageOpt = OptionBuilder.withArgName("int").hasArg().withDescription("number of pages to process (default all)").withLongOpt("num-pages").create("p");
			Option notificationPointOpt = OptionBuilder.withArgName("int").hasArg().withDescription("receive notification every n pages (default " + Defaults.DEFAULT_NOTIFICATION_POINT
                    + ")").withLongOpt("notification-point").create("n");
			Option baseDirOpt = OptionBuilder.withDescription("if set, use the output folder as base dir").withLongOpt("base-dir").create();

			options.addOption("h", "help", false, "print this message");
			options.addOption("v", "version", false, "output version information and exit");


			options.addOption(wikipediaDumpOpt);
			options.addOption(outputDirOpt);
			options.addOption(numThreadOpt);
			options.addOption(numPageOpt);
			options.addOption(notificationPointOpt);
			options.addOption(baseDirOpt);

			CommandLineParser parser = new PosixParser();
			CommandLine line = parser.parse(options, args);


			int numThreads = Defaults.DEFAULT_THREADS_NUMBER;
			if (line.hasOption("num-threads")) {
				numThreads = Integer.parseInt(line.getOptionValue("num-threads"));
			}

			int numPages = Defaults.DEFAULT_NUM_PAGES;
			if (line.hasOption("num-pages")) {
				numPages = Integer.parseInt(line.getOptionValue("num-pages"));
			}

			int notificationPoint = Defaults.DEFAULT_NOTIFICATION_POINT;
			if (line.hasOption("notification-point")) {
				notificationPoint = Integer.parseInt(line.getOptionValue("notification-point"));
			}

			ExtractorParameters extractorParameters;
			if (line.hasOption("base-dir")) {
				extractorParameters = new ExtractorParameters(line.getOptionValue("wikipedia-dump"), line.getOptionValue("output-dir"), true);
			}
			else {
				extractorParameters = new ExtractorParameters(line.getOptionValue("wikipedia-dump"), line.getOptionValue("output-dir"));

			}
			logger.debug(extractorParameters);

			logger.debug("extracting abstracts (" + extractorParameters.getWikipediaAbstractFileName() + ")...");
			WikipediaFirstSentenceExtractor wikipediaAbstractExtractor = new WikipediaFirstSentenceExtractor(numThreads, numPages, extractorParameters.getLocale());
			wikipediaAbstractExtractor.setNotificationPoint(notificationPoint);
			wikipediaAbstractExtractor.start(extractorParameters);

			logger.info("extraction ended " + new Date());

		} catch (ParseException e) {
			// oops, something went wrong
			System.out.println("Parsing failed: " + e.getMessage() + "\n");
			HelpFormatter formatter = new HelpFormatter();
			formatter.printHelp(400, "java -cp dist/thewikimachine.jar org.fbk.cit.hlt.thewikimachine.xmldump.WikipediaAbstractExtractor", "\n", options, "\n", true);
		}
	}
}