Package org.languagetool.tagging

Examples of org.languagetool.tagging.ManualTagger

80
81
82
83
84
85
86
87
88
89
90
    if (morfologik == null) {     
      final URL url = JLanguageTool.getDataBroker().getFromResourceDirAsUrl(DICT_FILENAME);
      morfologik = new DictionaryLookup(Dictionary.read(url));     
    }
    if (manualTagger == null) {
      manualTagger = new ManualTagger(JLanguageTool.getDataBroker().getFromResourceDirAsStream(USER_DICT_FILENAME));
    }
    if (compoundTokenizer == null) {
      compoundTokenizer = new GermanCompoundTokenizer();
    }
View Full Code Here
76
77
78
79
80
81
82
83
84
85
86
    if (morfologik == null) {     
      final URL url = JLanguageTool.getDataBroker().getFromResourceDirAsUrl(binaryDictPath);
      morfologik = new DictionaryLookup(Dictionary.read(url));
    }
    if (manualTagger == null && plaintextDictPath != null) {
      manualTagger = new ManualTagger(JLanguageTool.getDataBroker().getFromResourceDirAsStream(plaintextDictPath));
    }

    for (final String word : sentenceTokens) {
      final List<AnalyzedToken> l = new ArrayList<>();
      final String lowerCaseWord = word.toLowerCase(RO_LOCALE);
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
    // A manual tagger is used for closed words.  Closed words are the small
    // limited set of words in Esperanto which have no standard ending, as
    // opposed to open words which are unlimited in numbers and which follow
    // strict rules for their suffixes.
    manualTagger = new ManualTagger(JLanguageTool.getDataBroker().getFromResourceDirAsStream("/eo/manual-tagger.txt"));

    // Load set of transitive and intransitive verbs.  Files don't contain
    // verbs with suffix -iĝ or -ig since transitivity is obvious for those verbs.
    // They also don't contain verbs with prefixes mal-, ek-, re-, mis- fi- and
    // suffixes -ad, -aĉ, -et, -eg since these affixes never alter transitivity.
View Full Code Here
90
91
92
93
94
95
96
97
98
99
100
  //-- setup

  @Override
  protected void setUp() throws Exception {
    super.setUp();
    tagger = new ManualTaggerAdapter(new ManualTagger(new ByteArrayInputStream(TEST_DATA.getBytes("UTF-8"))));
    synthesizer = new ManualSynthesizerAdapter(new ManualSynthesizer(new ByteArrayInputStream(TEST_DATA.getBytes("UTF-8"))));
    languageTool = new JLanguageTool(new Demo() {
      @Override
      public String getName() {
        return "TEST";
View Full Code Here
57
58
59
60
61
62
63
64
  }

  protected void initialize() throws IOException {
    final URL url = JLanguageTool.getDataBroker().getFromResourceDirAsUrl(DICT_FILENAME);
    dictionary = Dictionary.read(url);
    manualTagger = new ManualTagger(JLanguageTool.getDataBroker().getFromResourceDirAsStream(USER_DICT_FILENAME));
    compoundTokenizer = new GermanCompoundTokenizer();
  }
View Full Code Here
64
65
66
67
68
69
70
71
72
73
74
  }

  protected void initializeIfRequired() throws IOException {
    // Lazy initialize all fields when needed and only once.
    Dictionary dict = dictionary;
    ManualTagger mTagger = manualTagger;
    GermanCompoundTokenizer gTokenizer = compoundTokenizer;
    if (dict == null || mTagger == null || gTokenizer == null) {
      synchronized (this) {
        dict = dictionary;
        mTagger = manualTagger;
View Full Code Here
79
80
81
82
83
84
85
86
87
88
89
  //-- setup

  @Override
  protected void setUp() throws Exception {
    super.setUp();
    tagger = new ManualTaggerAdapter(new ManualTagger(new ByteArrayInputStream(TEST_DATA.getBytes("UTF-8"))));
    synthesizer = new ManualSynthesizerAdapter(new ManualSynthesizer(new ByteArrayInputStream(TEST_DATA.getBytes("UTF-8"))));
    languageTool = new JLanguageTool(new Demo() {
      @Override
      public String getName() {
        return "TEST";
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
    // A manual tagger is used for closed words.  Closed words are the small
    // limited set of words in Esperanto which have no standard ending, as
    // opposed to open words which are unlimited in numbers and which follow
    // strict rules for their suffixes.
    manualTagger = new ManualTagger(JLanguageTool.getDataBroker().getFromResourceDirAsStream("/eo/manual-tagger.txt"));

    // Load set of transitive and intransitive verbs.  Files don't contain
    // verbs with suffix -iĝ or -ig since transitivity is obvious for those verbs.
    // They also don't contain verbs with prefixes mal-, ek-, re-, mis- fi- and
    // suffixes -ad, -aĉ, -et, -eg since these affixes never alter transitivity.
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
      final List<String> sentenceTokens) throws IOException {
    final List<AnalyzedTokenReadings> tokenReadings = new ArrayList<>();
    int pos = 0;
    final IStemmer morfologik = new DictionaryLookup(getDictionary());
    if (manualTagger == null && plaintextDictPath != null) {
      manualTagger = new ManualTagger(JLanguageTool.getDataBroker().getFromResourceDirAsStream(plaintextDictPath));
    }

    for (final String word : sentenceTokens) {
      final List<AnalyzedToken> l = new ArrayList<>();
      final String lowerCaseWord = word.toLowerCase(RO_LOCALE);
View Full Code Here
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
    dontTagLowercaseWithUppercase();
  }

  private void initializeIfRequired() throws IOException {
    // Lazy initialize fields when needed and only once.
    ManualTagger mTagger = manualTagger;
    if (mTagger == null) {
      synchronized (this) {
        mTagger = manualTagger;
        if (mTagger == null) {
          manualTagger = new ManualTagger(JLanguageTool.getDataBroker().getFromResourceDirAsStream(USER_DICT_FILENAME));
        }
      }
    }
  }
View Full Code Here
TOP

Related Classes of org.languagetool.tagging.ManualTagger

  • org.languagetool.rules.patterns.MatchTest
  • org.languagetool.tagging.ca.CatalanTagger
  • org.languagetool.tagging.de.GermanTagger
  • org.languagetool.tagging.eo.EsperantoTagger
  • org.languagetool.tagging.ro.RomanianTagger
  • org.languagetool.tokenizers.ManualTaggerAdapterTest
  • java.io.InputStreamReader
  • java.io.BufferedReader
  • java.util.Scanner
  • java.io.IOException
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.