All Downloads are FREE. Search and download functionalities are using the official Maven repository.

org.languagetool.AnalyzedSentence Maven / Gradle / Ivy

Go to download

LanguageTool is an Open Source proofreading software for English, French, German, Polish, Romanian, and more than 20 other languages. It finds many errors that a simple spell checker cannot detect like mixing up there/their and it detects some grammar problems.

There is a newer version: 6.5
Show newest version
/* LanguageTool, a natural language style checker
 * Copyright (C) 2014 Daniel Naber, Marcin Miłkowski (http://www.languagetool.org)
 * 
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * This library 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
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301
 * USA
 */
package org.languagetool;

import org.apache.commons.lang3.StringUtils;

import java.util.*;

/**
 * A sentence that has been tokenized and analyzed.
 * 
 * @author Daniel Naber
 */
public final class AnalyzedSentence {

  private final AnalyzedTokenReadings[] tokens;
  private final AnalyzedTokenReadings[] nonBlankTokens;
  private final int[] whPositions;  // maps positions without whitespace to positions that include whitespaces
  private final Set tokenSet;
  private final Set lemmaSet;

  /**
   * Creates an AnalyzedSentence from the given {@link AnalyzedTokenReadings}. Whitespace is also a token.
   */
  public AnalyzedSentence(AnalyzedTokenReadings[] tokens) {
    this.tokens = tokens;
    int whCounter = 0;
    int nonWhCounter = 0;
    int[] mapping = new int[tokens.length + 1];
    List l = new ArrayList<>();
    for (AnalyzedTokenReadings token : tokens) {
      if (!token.isWhitespace() || token.isSentenceStart() || token.isSentenceEnd() || token.isParagraphEnd()) {
        l.add(token);
        mapping[nonWhCounter] = whCounter;
        nonWhCounter++;
      }
      whCounter++;
    }
    this.whPositions = mapping;
    this.nonBlankTokens = l.toArray(new AnalyzedTokenReadings[l.size()]);
    this.tokenSet = getTokenSet(tokens);
    this.lemmaSet = getLemmaSet(tokens);
  }

  private AnalyzedSentence(AnalyzedTokenReadings[] tokens, int[] mapping, AnalyzedTokenReadings[] nonBlankTokens) {
    this.tokens = tokens;
    this.whPositions = mapping;
    this.nonBlankTokens = nonBlankTokens;
    this.tokenSet = getTokenSet(tokens);
    this.lemmaSet = getLemmaSet(tokens);
  }

  private Set getTokenSet(AnalyzedTokenReadings[] tokens) {
    Set tokenSet = new HashSet<>();
    for (AnalyzedTokenReadings token : tokens) {
      tokenSet.add(token.getToken().toLowerCase());
    }
    return Collections.unmodifiableSet(tokenSet);
  }

  private Set getLemmaSet(AnalyzedTokenReadings[] tokens) {
    Set lemmaSet = new HashSet<>();
    for (AnalyzedTokenReadings token : tokens) {
      for (AnalyzedToken lemmaTok : token.getReadings()) {
        if (lemmaTok.getLemma() != null) {
          lemmaSet.add(lemmaTok.getLemma().toLowerCase());
        } else {
          lemmaSet.add(lemmaTok.getToken().toLowerCase());
        }
      }
    }
    return Collections.unmodifiableSet(lemmaSet);
  }

  /**
   * The method copies {@link AnalyzedSentence} and returns the copy.
   * Useful for performing local immunization (for example).
   *
   * @param sentence {@link AnalyzedSentence} to be copied
   * @return a new object which is a copy
   * @since  2.5
   */
  public AnalyzedSentence copy(AnalyzedSentence sentence) {
    AnalyzedTokenReadings[] copyTokens = new AnalyzedTokenReadings[sentence.getTokens().length];
    for (int i = 0; i < copyTokens.length; i++) {
      AnalyzedTokenReadings analyzedTokens = sentence.getTokens()[i];
      copyTokens[i] = new AnalyzedTokenReadings(analyzedTokens.getReadings(), analyzedTokens.getStartPos());
      copyTokens[i].setHistoricalAnnotations(analyzedTokens.getHistoricalAnnotations());
      copyTokens[i].setChunkTags(analyzedTokens.getChunkTags());
      if (analyzedTokens.isImmunized()) {
        copyTokens[i].immunize();
      }
      if (analyzedTokens.isIgnoredBySpeller()) {
        copyTokens[i].ignoreSpelling();
      }
      copyTokens[i].setWhitespaceBefore(sentence.getTokens()[i].isWhitespaceBefore());
    }
    return new AnalyzedSentence(copyTokens, sentence.whPositions, sentence.getTokensWithoutWhitespace());
  }

  /**
   * Returns the {@link AnalyzedTokenReadings} of the analyzed text. Whitespace
   * is also a token.
   */
  public AnalyzedTokenReadings[] getTokens() {
    // It would be better to return a clone here to make this object immutable,
    // but this would be bad for performance:
    return tokens;
  }

  /**
   * Returns the {@link AnalyzedTokenReadings} of the analyzed text, with
   * whitespace tokens removed but with the artificial SENT_START
   * token included.
   */
  public AnalyzedTokenReadings[] getTokensWithoutWhitespace() {
    return nonBlankTokens.clone();
  }

  /**
   * Get a position of a non-whitespace token in the original sentence with
   * whitespace.
   *
   * @param nonWhPosition position of a non-whitespace token
   * @return position in the original sentence.
   */
  public int getOriginalPosition(int nonWhPosition) {
    return whPositions[nonWhPosition];
  }

  @Override
  public String toString() {
    return toString(",");
  }

  /**
   * Return string representation without chunk information.
   * @since 2.3
   */
  public String toShortString(String readingDelimiter) {
    return toString(readingDelimiter, false);
  }

  /**
   * Return the original text.
   * @since 2.7
   */
  public String getText() {
    StringBuilder sb = new StringBuilder();
    for (AnalyzedTokenReadings element : tokens) {
      sb.append(element.getToken());
    }
    return sb.toString();
  }

  /**
   * Return string representation without any analysis information, just the original text.
   * @since 2.6
   */
  String toTextString() {
    return getText();
  }

  /**
   * Return string representation with chunk information.
   */
  public String toString(String readingDelimiter) {
    return toString(readingDelimiter, true);
  }

  private String toString(String readingDelimiter, boolean includeChunks) {
    StringBuilder sb = new StringBuilder();
    for (AnalyzedTokenReadings element : tokens) {
      if (!element.isWhitespace()) {
        sb.append(element.getToken());
        sb.append('[');
      }
      Iterator iterator = element.iterator();
      while (iterator.hasNext()) {
        AnalyzedToken token = iterator.next();
        String posTag = token.getPOSTag();
        if (element.isSentenceStart()) {
          sb.append("");
        } else if (JLanguageTool.SENTENCE_END_TAGNAME.equals(posTag)) {
          sb.append("");
        } else if (JLanguageTool.PARAGRAPH_END_TAGNAME.equals(posTag)) {
          sb.append("

"); } else if (posTag == null && !includeChunks) { sb.append(token.getToken()); } else { if (!element.isWhitespace()) { sb.append(token); if (iterator.hasNext()) { sb.append(readingDelimiter); } } } } if (!element.isWhitespace()) { if (includeChunks && element.getChunkTags().size() > 0) { sb.append(','); sb.append(StringUtils.join(element.getChunkTags(), "|")); } if (element.isImmunized()) { sb.append("{!}"); } sb.append(']'); } else { sb.append(' '); } } return sb.toString(); } /** * Get disambiguator actions log. */ public String getAnnotations() { StringBuilder sb = new StringBuilder(40); sb.append("Disambiguator log: \n"); for (AnalyzedTokenReadings element : tokens) { if (!element.isWhitespace() && !"".equals(element.getHistoricalAnnotations())) { sb.append(element.getHistoricalAnnotations()); sb.append('\n'); } } return sb.toString(); } /** * Get the lowercase tokens of this sentence in a set. * Used internally for performance optimization. * @since 2.4 */ public Set getTokenSet() { return tokenSet; } /** * Get the lowercase lemmas of this sentence in a set. * Used internally for performance optimization. * @since 2.5 */ public Set getLemmaSet() { return lemmaSet; } @SuppressWarnings("ControlFlowStatementWithoutBraces") @Override public boolean equals(Object o) { if (this == o) return true; if (o == null) return false; if (getClass() != o.getClass()) return false; AnalyzedSentence other = (AnalyzedSentence) o; // tokenSet and lemmaSet are a subset of tokens and don't need to be included return Arrays.equals(nonBlankTokens, other.nonBlankTokens) && Arrays.equals(tokens, other.tokens) && Arrays.equals(whPositions, other.whPositions); } @Override public int hashCode() { // tokenSet and lemmaSet are a subset of tokens and don't need to be included return Objects.hash(nonBlankTokens, tokens, whPositions); } }





© 2015 - 2024 Weber Informatics LLC | Privacy Policy