Package org.apache.lucene.analysis

Examples of org.apache.lucene.analysis.CharFilter

80
81
82
83
84
85
86
87
88
89
90
    List<String> mappingRules = new ArrayList<>();
    mappingRules.add( "\"&uuml;\" => \"ü\"" );
    NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
    builder.add("&uuml;", "ü");
    NormalizeCharMap normMap = builder.build();
    CharFilter charStream = new MappingCharFilter( normMap, new StringReader( INPUT ) );

    // create PatternTokenizer
    Tokenizer stream = new PatternTokenizer(newAttributeFactory(), charStream, Pattern.compile("[,;/\\s]+"), -1);
    assertTokenStreamContents(stream,
        new String[] { "Günther", "Günther", "is", "here" },
View Full Code Here
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
      throw new UnsupportedOperationException("read(char[], int, int)");
    }
  };
 
  public void testWrapping() throws Exception {
    CharFilter cs = new TestRandomChains.CheckThatYouDidntReadAnythingReaderWrapper(wrappedStream);
    try {
      cs.mark(1);
      fail();
    } catch (Exception e) {
      assertEquals("mark(int)", e.getMessage());
    }
   
    try {
      cs.markSupported();
      fail();
    } catch (Exception e) {
      assertEquals("markSupported()", e.getMessage());
    }
   
    try {
      cs.read();
      fail();
    } catch (Exception e) {
      assertEquals("read()", e.getMessage());
    }
   
    try {
      cs.read(new char[0]);
      fail();
    } catch (Exception e) {
      assertEquals("read(char[])", e.getMessage());
    }
   
    try {
      cs.read(CharBuffer.wrap(new char[0]));
      fail();
    } catch (Exception e) {
      assertEquals("read(CharBuffer)", e.getMessage());
    }
   
    try {
      cs.reset();
      fail();
    } catch (Exception e) {
      assertEquals("reset()", e.getMessage());
    }
   
    try {
      cs.skip(1);
      fail();
    } catch (Exception e) {
      assertEquals("skip(long)", e.getMessage());
    }
   
    try {
      cs.correctOffset(1);
      fail();
    } catch (Exception e) {
      assertEquals("correct(int)", e.getMessage());
    }
   
    try {
      cs.close();
      fail();
    } catch (Exception e) {
      assertEquals("close()", e.getMessage());
    }
   
    try {
      cs.read(new char[0], 0, 0);
      fail();
    } catch (Exception e) {
      assertEquals("read(char[], int, int)", e.getMessage());
    }
  }
View Full Code Here
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
      throw new UnsupportedOperationException("read(char[], int, int)");
    }
  };
 
  public void testWrapping() throws Exception {
    CharFilter cs = new TestRandomChains.CheckThatYouDidntReadAnythingReaderWrapper(wrappedStream);
    try {
      cs.mark(1);
      fail();
    } catch (Exception e) {
      assertEquals("mark(int)", e.getMessage());
    }
   
    try {
      cs.markSupported();
      fail();
    } catch (Exception e) {
      assertEquals("markSupported()", e.getMessage());
    }
   
    try {
      cs.read();
      fail();
    } catch (Exception e) {
      assertEquals("read()", e.getMessage());
    }
   
    try {
      cs.read(new char[0]);
      fail();
    } catch (Exception e) {
      assertEquals("read(char[])", e.getMessage());
    }
   
    try {
      cs.read(CharBuffer.wrap(new char[0]));
      fail();
    } catch (Exception e) {
      assertEquals("read(CharBuffer)", e.getMessage());
    }
   
    try {
      cs.reset();
      fail();
    } catch (Exception e) {
      assertEquals("reset()", e.getMessage());
    }
   
    try {
      cs.skip(1);
      fail();
    } catch (Exception e) {
      assertEquals("skip(long)", e.getMessage());
    }
   
    try {
      cs.correctOffset(1);
      fail();
    } catch (Exception e) {
      assertEquals("correct(int)", e.getMessage());
    }
   
    try {
      cs.close();
      fail();
    } catch (Exception e) {
      assertEquals("close()", e.getMessage());
    }
   
    try {
      cs.read(new char[0], 0, 0);
      fail();
    } catch (Exception e) {
      assertEquals("read(char[], int, int)", e.getMessage());
    }
  }
View Full Code Here
33
34
35
36
37
38
39
40
41
public class TestJapaneseIterationMarkCharFilterFactory extends BaseTokenStreamTestCase {

  public void testIterationMarksWithKeywordTokenizer() throws IOException {
    final String text = "時々馬鹿々々しいところゞゝゝミスヾ";
    JapaneseIterationMarkCharFilterFactory filterFactory = new JapaneseIterationMarkCharFilterFactory(new HashMap<String,String>());
    CharFilter filter = filterFactory.create(new StringReader(text));
    TokenStream tokenStream = new MockTokenizer(filter, MockTokenizer.KEYWORD, false);
    assertTokenStreamContents(tokenStream, new String[]{"時時馬鹿馬鹿しいところどころミスズ"});
  }
View Full Code Here
43
44
45
46
47
48
49
50
51
52
53
  public void testIterationMarksWithJapaneseTokenizer() throws IOException {
    JapaneseTokenizerFactory tokenizerFactory = new JapaneseTokenizerFactory(new HashMap<String,String>());
    tokenizerFactory.inform(new StringMockResourceLoader(""));

    JapaneseIterationMarkCharFilterFactory filterFactory = new JapaneseIterationMarkCharFilterFactory(new HashMap<String,String>());
    CharFilter filter = filterFactory.create(
        new StringReader("時々馬鹿々々しいところゞゝゝミスヾ")
    );
    TokenStream tokenStream = tokenizerFactory.create(filter);
    assertTokenStreamContents(tokenStream, new String[]{"時時", "馬鹿馬鹿しい", "ところどころ", "ミ", "スズ"});
  }
View Full Code Here
59
60
61
62
63
64
65
66
67
68
69
    Map<String, String> filterArgs = new HashMap<>();
    filterArgs.put("normalizeKanji", "true");
    filterArgs.put("normalizeKana", "false");
    JapaneseIterationMarkCharFilterFactory filterFactory = new JapaneseIterationMarkCharFilterFactory(filterArgs);
   
    CharFilter filter = filterFactory.create(
        new StringReader("時々馬鹿々々しいところゞゝゝミスヾ")
    );
    TokenStream tokenStream = tokenizerFactory.create(filter);
    assertTokenStreamContents(tokenStream, new String[]{"時時", "馬鹿馬鹿しい", "ところ", "ゞ", "ゝ", "ゝ", "ミス", "ヾ"});
  }
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
    Map<String, String> filterArgs = new HashMap<>();
    filterArgs.put("normalizeKanji", "false");
    filterArgs.put("normalizeKana", "true");
    JapaneseIterationMarkCharFilterFactory filterFactory = new JapaneseIterationMarkCharFilterFactory(filterArgs);

    CharFilter filter = filterFactory.create(
        new StringReader("時々馬鹿々々しいところゞゝゝミスヾ")
    );
    TokenStream tokenStream = tokenizerFactory.create(filter);
    assertTokenStreamContents(tokenStream, new String[]{"時々", "馬鹿", "々", "々", "しい", "ところどころ", "ミ", "スズ"});
  }
View Full Code Here
259
260
261
262
263
264
265
266
267
268
269
  // 012345678901234567890123456789012345678
  //  a bb - ccc . --- bb a . ccc ccc bb
  //  aa b - c . --- b aa . c c b
  public void testChain() throws IOException {
    final String BLOCK = " a bb - ccc . --- bb a . ccc ccc bb";
    CharFilter cs = new PatternReplaceCharFilter( pattern("a"), "aa",
        new StringReader( BLOCK ) );
    cs = new PatternReplaceCharFilter( pattern("bb"), "b", cs );
    cs = new PatternReplaceCharFilter( pattern("ccc"), "c", cs );
    TokenStream ts = new MockTokenizer(cs, MockTokenizer.WHITESPACE, false);
    assertTokenStreamContents(ts,
View Full Code Here
80
81
82
83
84
85
86
87
88
89
90
    List<String> mappingRules = new ArrayList<>();
    mappingRules.add( "\"&uuml;\" => \"ü\"" );
    NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
    builder.add("&uuml;", "ü");
    NormalizeCharMap normMap = builder.build();
    CharFilter charStream = new MappingCharFilter( normMap, new StringReader( INPUT ) );

    // create PatternTokenizer
    TokenStream stream = new PatternTokenizer(charStream, Pattern.compile("[,;/\\s]+"), -1);
    assertTokenStreamContents(stream,
        new String[] { "Günther", "Günther", "is", "here" },
View Full Code Here
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
      throw new UnsupportedOperationException("read(char[], int, int)");
    }
  };
 
  public void testWrapping() throws Exception {
    CharFilter cs = new TestRandomChains.CheckThatYouDidntReadAnythingReaderWrapper(wrappedStream);
    try {
      cs.mark(1);
      fail();
    } catch (Exception e) {
      assertEquals("mark(int)", e.getMessage());
    }
   
    try {
      cs.markSupported();
      fail();
    } catch (Exception e) {
      assertEquals("markSupported()", e.getMessage());
    }
   
    try {
      cs.read();
      fail();
    } catch (Exception e) {
      assertEquals("read()", e.getMessage());
    }
   
    try {
      cs.read(new char[0]);
      fail();
    } catch (Exception e) {
      assertEquals("read(char[])", e.getMessage());
    }
   
    try {
      cs.read(CharBuffer.wrap(new char[0]));
      fail();
    } catch (Exception e) {
      assertEquals("read(CharBuffer)", e.getMessage());
    }
   
    try {
      cs.reset();
      fail();
    } catch (Exception e) {
      assertEquals("reset()", e.getMessage());
    }
   
    try {
      cs.skip(1);
      fail();
    } catch (Exception e) {
      assertEquals("skip(long)", e.getMessage());
    }
   
    try {
      cs.correctOffset(1);
      fail();
    } catch (Exception e) {
      assertEquals("correct(int)", e.getMessage());
    }
   
    try {
      cs.close();
      fail();
    } catch (Exception e) {
      assertEquals("close()", e.getMessage());
    }
   
    try {
      cs.read(new char[0], 0, 0);
      fail();
    } catch (Exception e) {
      assertEquals("read(char[], int, int)", e.getMessage());
    }
  }
View Full Code Here
TOP

Related Classes of org.apache.lucene.analysis.CharFilter

  • org.apache.lucene.analysis.charfilter.TestMappingCharFilter
  • org.apache.lucene.analysis.core.TestBugInSomething
  • org.apache.lucene.analysis.icu.TestICUNormalizer2CharFilter
  • org.apache.lucene.analysis.ja.TestJapaneseIterationMarkCharFilter
  • org.apache.lucene.analysis.ja.TestJapaneseIterationMarkCharFilterFactory
  • org.apache.lucene.analysis.pattern.TestPatternReplaceCharFilter
  • org.apache.lucene.analysis.pattern.TestPatternTokenizer
    • 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.