Examples with StringReader java.io.StringReader used on opensource projects

Example 86 with StringReader

use of java.io.StringReader in project CoreNLP by stanfordnlp.

the class DcorefBenchmarkSlowITest method getCorefResults.

public static Counter<String> getCorefResults(String resultsString) throws IOException {
    Counter<String> results = new ClassicCounter<>();
    BufferedReader r = new BufferedReader(new StringReader(resultsString));
    for (String line; (line = r.readLine()) != null; ) {
        Matcher m1 = MENTION_PATTERN.matcher(line);
        if (m1.matches()) {
            results.setCount(MENTION_TP, Double.parseDouble(m1.group(1)));
            results.setCount(MENTION_F1, Double.parseDouble(m1.group(2)));
        }
        Matcher m2 = MUC_PATTERN.matcher(line);
        if (m2.matches()) {
            results.setCount(MUC_TP, Double.parseDouble(m2.group(1)));
            results.setCount(MUC_F1, Double.parseDouble(m2.group(2)));
        }
        Matcher m3 = BCUBED_PATTERN.matcher(line);
        if (m3.matches()) {
            results.setCount(BCUBED_TP, Double.parseDouble(m3.group(1)));
            results.setCount(BCUBED_F1, Double.parseDouble(m3.group(2)));
        }
        Matcher m4 = CEAFM_PATTERN.matcher(line);
        if (m4.matches()) {
            results.setCount(CEAFM_TP, Double.parseDouble(m4.group(1)));
            results.setCount(CEAFM_F1, Double.parseDouble(m4.group(2)));
        }
        Matcher m5 = CEAFE_PATTERN.matcher(line);
        if (m5.matches()) {
            results.setCount(CEAFE_TP, Double.parseDouble(m5.group(1)));
            results.setCount(CEAFE_F1, Double.parseDouble(m5.group(2)));
        }
        Matcher m6 = BLANC_PATTERN.matcher(line);
        if (m6.matches()) {
            results.setCount(BLANC_F1, Double.parseDouble(m6.group(1)));
        }
        Matcher m7 = CONLL_PATTERN.matcher(line);
        if (m7.matches()) {
            results.setCount(CONLL_SCORE, Double.parseDouble(m7.group(1)));
        }
    }
    return results;
}

Example 87 with StringReader

use of java.io.StringReader in project CoreNLP by stanfordnlp.

the class SpanishTokenizerITest method testOffsetsSpacing.

public void testOffsetsSpacing() {
    // guide                 1         2         3         4          5         6         7           8         9         0         1         2         3
    // guide       0123456789012345678901234567890123456789012345678 90123456789012345678901234567 8 901234567890123456789012345678901234567890123456789012345
    String text = "  La   combinación consonántica ss es ajena a la\tortografía    castellana:   \n\n traigámosela, mandémoselos, escribámosela, comprémoselo.";
    final TokenizerFactory<CoreLabel> tf = SpanishTokenizer.coreLabelFactory();
    tf.setOptions("");
    tf.setOptions("splitAll=true");
    Tokenizer<CoreLabel> spanishTokenizer = tf.getTokenizer(new StringReader(text));
    List<CoreLabel> tokens = spanishTokenizer.tokenize();
    System.err.println(tokens);
    assertEquals(27, tokens.size());
    // assertEquals("  ", tokens.get(0).get(CoreAnnotations.BeforeAnnotation.class));
    // assertEquals("\t", tokens.get(8).get(CoreAnnotations.AfterAnnotation.class));
    assertEquals("Begin char offset", 2, (int) tokens.get(0).get(CoreAnnotations.CharacterOffsetBeginAnnotation.class));
    assertEquals("End char offset", 4, (int) tokens.get(0).get(CoreAnnotations.CharacterOffsetEndAnnotation.class));
    assertEquals("La", tokens.get(0).get(CoreAnnotations.OriginalTextAnnotation.class));
    // note: after(x) and before(x+1) are the same
    // assertEquals("   ", tokens.get(0).get(CoreAnnotations.AfterAnnotation.class));
    // assertEquals("   ", tokens.get(1).get(CoreAnnotations.BeforeAnnotation.class));
    assertEquals("escribámo", tokens.get(19).get(CoreAnnotations.OriginalTextAnnotation.class));
    assertEquals("escribamos", tokens.get(19).get(CoreAnnotations.TextAnnotation.class));
    assertEquals("Begin char offset", 108, (int) tokens.get(19).get(CoreAnnotations.CharacterOffsetBeginAnnotation.class));
    assertEquals("End char offset", 117, (int) tokens.get(19).get(CoreAnnotations.CharacterOffsetEndAnnotation.class));
    assertEquals("se", tokens.get(20).get(CoreAnnotations.OriginalTextAnnotation.class));
    assertEquals("se", tokens.get(20).get(CoreAnnotations.TextAnnotation.class));
    assertEquals("Begin char offset", 117, (int) tokens.get(20).get(CoreAnnotations.CharacterOffsetBeginAnnotation.class));
    assertEquals("End char offset", 119, (int) tokens.get(20).get(CoreAnnotations.CharacterOffsetEndAnnotation.class));
    assertEquals("la", tokens.get(21).get(CoreAnnotations.OriginalTextAnnotation.class));
    assertEquals("la", tokens.get(21).get(CoreAnnotations.TextAnnotation.class));
    assertEquals("Begin char offset", 119, (int) tokens.get(21).get(CoreAnnotations.CharacterOffsetBeginAnnotation.class));
    assertEquals("End char offset", 121, (int) tokens.get(21).get(CoreAnnotations.CharacterOffsetEndAnnotation.class));
    assertEquals(",", tokens.get(22).get(CoreAnnotations.OriginalTextAnnotation.class));
    assertEquals(",", tokens.get(22).get(CoreAnnotations.TextAnnotation.class));
    assertEquals("Begin char offset", 121, (int) tokens.get(22).get(CoreAnnotations.CharacterOffsetBeginAnnotation.class));
    assertEquals("End char offset", 122, (int) tokens.get(22).get(CoreAnnotations.CharacterOffsetEndAnnotation.class));
}

Example 88 with StringReader

use of java.io.StringReader in project CoreNLP by stanfordnlp.

the class Morphology method stem.

public String stem(String word) {
    try {
        lexer.yyreset(new StringReader(word));
        lexer.yybegin(Morpha.any);
        String wordRes = lexer.next();
        return wordRes;
    } catch (IOException e) {
        log.warning("Morphology.stem() had error on word " + word);
        return word;
    }
}

Example 89 with StringReader

use of java.io.StringReader in project CoreNLP by stanfordnlp.

the class Morphology method lemmatize.

/** Lemmatize the word, being sensitive to the tag, using the
   *  passed in lexer.
   *
   *  @param lowercase If this is true, words other than proper nouns will
   *      be changed to all lowercase.
   */
private static String lemmatize(String word, String tag, Morpha lexer, boolean lowercase) {
    boolean wordHasForbiddenChar = word.indexOf('_') >= 0 || word.indexOf(' ') >= 0 || word.indexOf('\n') >= 0;
    String quotedWord = word;
    if (wordHasForbiddenChar) {
        // choose something unlikely. Classical Vedic!
        quotedWord = quotedWord.replaceAll("_", "ᳰ");
        quotedWord = quotedWord.replaceAll(" ", "ᳱ");
        quotedWord = quotedWord.replaceAll("\n", "ᳲ");
    }
    String wordtag = quotedWord + '_' + tag;
    if (DEBUG)
        log.info("Trying to normalize |" + wordtag + '|');
    try {
        lexer.setOption(1, lowercase);
        lexer.yyreset(new StringReader(wordtag));
        lexer.yybegin(Morpha.scan);
        String wordRes = lexer.next();
        // go past tag
        lexer.next();
        if (wordHasForbiddenChar) {
            if (DEBUG)
                log.info("Restoring forbidden chars");
            wordRes = wordRes.replaceAll("ᳰ", "_");
            wordRes = wordRes.replaceAll("ᳱ", " ");
            wordRes = wordRes.replaceAll("ᳲ", "\n");
        }
        return wordRes;
    } catch (IOException e) {
        log.warning("Morphology.stem() had error on word " + word + '/' + tag);
        return word;
    }
}

Example 90 with StringReader

use of java.io.StringReader in project liquibase by liquibase.

the class StreamUtilTest method testGetReaderContents.

@Test
public void testGetReaderContents() throws IOException {
    String contents = "TEST";
    StringReader reader = new StringReader(contents);
    String result = StreamUtil.getReaderContents(reader);
    assertEquals(contents, result);
}