public void testDefaultsKuromojiAnalysis() throws IOException {
AnalysisService analysisService = createAnalysisService();
TokenizerFactory tokenizerFactory = analysisService.tokenizer("kuromoji_tokenizer");
assertThat(tokenizerFactory, instanceOf(KuromojiTokenizerFactory.class));
TokenFilterFactory filterFactory = analysisService.tokenFilter("kuromoji_part_of_speech");
assertThat(filterFactory, instanceOf(KuromojiPartOfSpeechFilterFactory.class));
filterFactory = analysisService.tokenFilter("kuromoji_readingform");
assertThat(filterFactory, instanceOf(KuromojiReadingFormFilterFactory.class));
filterFactory = analysisService.tokenFilter("kuromoji_baseform");
assertThat(filterFactory, instanceOf(KuromojiBaseFormFilterFactory.class));
filterFactory = analysisService.tokenFilter("kuromoji_stemmer");
assertThat(filterFactory, instanceOf(KuromojiKatakanaStemmerFactory.class));
filterFactory = analysisService.tokenFilter("ja_stop");
assertThat(filterFactory, instanceOf(JapaneseStopTokenFilterFactory.class));
NamedAnalyzer analyzer = analysisService.analyzer("kuromoji");
assertThat(analyzer.analyzer(), instanceOf(JapaneseAnalyzer.class));
analyzer = analysisService.analyzer("my_analyzer");
assertThat(analyzer.analyzer(), instanceOf(CustomAnalyzer.class));
assertThat(
analyzer.analyzer().tokenStream(null, new StringReader("")),
instanceOf(JapaneseTokenizer.class));
CharFilterFactory charFilterFactory = analysisService.charFilter("kuromoji_iteration_mark");
assertThat(charFilterFactory, instanceOf(KuromojiIterationMarkCharFilterFactory.class));
}
@Test
public void testDefault() throws IOException {
Settings settings =
ImmutableSettings.settingsBuilder()
.put("index.analysis.filter.limit_default.type", "limit")
.build();
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromSettings(settings);
{
TokenFilterFactory tokenFilter = analysisService.tokenFilter("limit_default");
String source = "the quick brown fox";
String[] expected = new String[] {"the"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
{
TokenFilterFactory tokenFilter = analysisService.tokenFilter("limit");
String source = "the quick brown fox";
String[] expected = new String[] {"the"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
}
public void testKatakanaStemFilter() throws IOException {
AnalysisService analysisService = createAnalysisService();
TokenFilterFactory tokenFilter = analysisService.tokenFilter("kuromoji_stemmer");
assertThat(tokenFilter, instanceOf(KuromojiKatakanaStemmerFactory.class));
String source = "明後日パーティーに行く予定がある。図書館で資料をコピーしました。";
Tokenizer tokenizer = new JapaneseTokenizer(null, true, JapaneseTokenizer.Mode.SEARCH);
tokenizer.setReader(new StringReader(source));
// パーティー should be stemmed by default
// (min len) コピー should not be stemmed
String[] expected_tokens_katakana =
new String[] {
"明後日", "パーティ", "に", "行く", "予定", "が", "ある", "図書館", "で", "資料", "を", "コピー", "し", "まし", "た"
};
assertSimpleTSOutput(tokenFilter.create(tokenizer), expected_tokens_katakana);
tokenFilter = analysisService.tokenFilter("kuromoji_ks");
assertThat(tokenFilter, instanceOf(KuromojiKatakanaStemmerFactory.class));
tokenizer = new JapaneseTokenizer(null, true, JapaneseTokenizer.Mode.SEARCH);
tokenizer.setReader(new StringReader(source));
// パーティー should not be stemmed since min len == 6
// コピー should not be stemmed
expected_tokens_katakana =
new String[] {
"明後日", "パーティー", "に", "行く", "予定", "が", "ある", "図書館", "で", "資料", "を", "コピー", "し", "まし", "た"
};
assertSimpleTSOutput(tokenFilter.create(tokenizer), expected_tokens_katakana);
}
@Test
public void testSettings() throws IOException {
{
Settings settings =
ImmutableSettings.settingsBuilder()
.put("index.analysis.filter.limit_1.type", "limit")
.put("index.analysis.filter.limit_1.max_token_count", 3)
.put("index.analysis.filter.limit_1.consume_all_tokens", true)
.build();
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromSettings(settings);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("limit_1");
String source = "the quick brown fox";
String[] expected = new String[] {"the", "quick", "brown"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
{
Settings settings =
ImmutableSettings.settingsBuilder()
.put("index.analysis.filter.limit_1.type", "limit")
.put("index.analysis.filter.limit_1.max_token_count", 3)
.put("index.analysis.filter.limit_1.consume_all_tokens", false)
.build();
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromSettings(settings);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("limit_1");
String source = "the quick brown fox";
String[] expected = new String[] {"the", "quick", "brown"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
{
Settings settings =
ImmutableSettings.settingsBuilder()
.put("index.analysis.filter.limit_1.type", "limit")
.put("index.analysis.filter.limit_1.max_token_count", 17)
.put("index.analysis.filter.limit_1.consume_all_tokens", true)
.build();
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromSettings(settings);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("limit_1");
String source = "the quick brown fox";
String[] expected = new String[] {"the", "quick", "brown", "fox"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
}
public void testCameCaseOverride() throws IOException {
Settings settings =
Settings.builder()
.put(Environment.PATH_HOME_SETTING.getKey(), createTempDir().toString())
.build();
Settings indexSettings =
settingsBuilder()
.put(IndexMetaData.SETTING_VERSION_CREATED, Version.CURRENT)
.put("index.analysis.filter.wordDelimiter.type", "word_delimiter")
.put("index.analysis.filter.wordDelimiter.split_on_numerics", false)
.put("index.analysis.analyzer.custom_analyzer.tokenizer", "whitespace")
.putArray(
"index.analysis.analyzer.custom_analyzer.filter", "lowercase", "wordDelimiter")
.put("index.analysis.analyzer.custom_analyzer_1.tokenizer", "whitespace")
.putArray(
"index.analysis.analyzer.custom_analyzer_1.filter", "lowercase", "word_delimiter")
.build();
IndexSettings idxSettings = IndexSettingsModule.newIndexSettings("index", indexSettings);
AnalysisService analysisService =
new AnalysisRegistry(null, new Environment(settings)).build(idxSettings);
TokenFilterFactory word_delimiter = analysisService.tokenFilter("word_delimiter");
TokenFilterFactory override = analysisService.tokenFilter("wordDelimiter");
assertNotEquals(word_delimiter.name(), override.name());
assertNotSame(
analysisService.tokenFilter("wordDelimiter"),
analysisService.tokenFilter("word_delimiter"));
assertSame(
analysisService.tokenFilter("porterStem"), analysisService.tokenFilter("porter_stem"));
// unconfigured
IndexSettings idxSettings1 =
IndexSettingsModule.newIndexSettings(
"index",
settingsBuilder().put(IndexMetaData.SETTING_VERSION_CREATED, Version.CURRENT).build());
AnalysisService analysisService1 =
new AnalysisRegistry(null, new Environment(settings)).build(idxSettings1);
assertSame(
analysisService1.tokenFilter("wordDelimiter"),
analysisService1.tokenFilter("word_delimiter"));
assertSame(
analysisService1.tokenFilter("porterStem"), analysisService1.tokenFilter("porter_stem"));
}
public void testBaseFormFilterFactory() throws IOException {
AnalysisService analysisService = createAnalysisService();
TokenFilterFactory tokenFilter = analysisService.tokenFilter("kuromoji_pos");
assertThat(tokenFilter, instanceOf(KuromojiPartOfSpeechFilterFactory.class));
String source = "私は制限スピードを超える。";
String[] expected = new String[] {"私", "は", "制限", "スピード", "を"};
Tokenizer tokenizer = new JapaneseTokenizer(null, true, JapaneseTokenizer.Mode.SEARCH);
tokenizer.setReader(new StringReader(source));
assertSimpleTSOutput(tokenFilter.create(tokenizer), expected);
}
public void testJapaneseStopFilterFactory() throws IOException {
AnalysisService analysisService = createAnalysisService();
TokenFilterFactory tokenFilter = analysisService.tokenFilter("ja_stop");
assertThat(tokenFilter, instanceOf(JapaneseStopTokenFilterFactory.class));
String source = "私は制限スピードを超える。";
String[] expected = new String[] {"私", "制限", "超える"};
Tokenizer tokenizer = new JapaneseTokenizer(null, true, JapaneseTokenizer.Mode.SEARCH);
tokenizer.setReader(new StringReader(source));
assertSimpleTSOutput(tokenFilter.create(tokenizer), expected);
}
@Test
public void testHanOnly() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromClassPath(RESOURCE);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("cjk_han_only");
String source = "多くの学生が試験に落ちた。";
String[] expected = new String[] {"多", "く", "の", "学生", "が", "試験", "に", "落", "ち", "た"};
Tokenizer tokenizer = new StandardTokenizer(TEST_VERSION_CURRENT, new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
public void testReadingFormFilterFactory() throws IOException {
AnalysisService analysisService = createAnalysisService();
TokenFilterFactory tokenFilter = analysisService.tokenFilter("kuromoji_rf");
assertThat(tokenFilter, instanceOf(KuromojiReadingFormFilterFactory.class));
String source = "今夜はロバート先生と話した";
String[] expected_tokens_romaji =
new String[] {"kon'ya", "ha", "robato", "sensei", "to", "hanashi", "ta"};
Tokenizer tokenizer = new JapaneseTokenizer(null, true, JapaneseTokenizer.Mode.SEARCH);
tokenizer.setReader(new StringReader(source));
assertSimpleTSOutput(tokenFilter.create(tokenizer), expected_tokens_romaji);
tokenizer = new JapaneseTokenizer(null, true, JapaneseTokenizer.Mode.SEARCH);
tokenizer.setReader(new StringReader(source));
String[] expected_tokens_katakana = new String[] {"コンヤ", "ハ", "ロバート", "センセイ", "ト", "ハナシ", "タ"};
tokenFilter = analysisService.tokenFilter("kuromoji_readingform");
assertThat(tokenFilter, instanceOf(KuromojiReadingFormFilterFactory.class));
assertSimpleTSOutput(tokenFilter.create(tokenizer), expected_tokens_katakana);
}
public void testCaseInsensitiveMapping() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromClassPath(createTempDir(), RESOURCE);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("my_keep_filter");
assertThat(tokenFilter, instanceOf(KeepWordFilterFactory.class));
String source = "hello small world";
String[] expected = new String[] {"hello", "world"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected, new int[] {1, 2});
}
public void testFillerToken() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromClassPath(createTempDir(), RESOURCE);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("shingle_filler");
String source = "simon the sorcerer";
String[] expected = new String[] {"simon FILLER", "simon FILLER sorcerer", "FILLER sorcerer"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
TokenStream stream = new StopFilter(tokenizer, StopFilter.makeStopSet("the"));
assertTokenStreamContents(tokenFilter.create(stream), expected);
}
public void testInverseMappingNoShingles() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromClassPath(createTempDir(), RESOURCE);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("shingle_inverse");
assertThat(tokenFilter, instanceOf(ShingleTokenFilterFactory.class));
String source = "the quick";
String[] expected = new String[] {"the", "quick"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
public void testDefault() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromClassPath(createTempDir(), RESOURCE);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("shingle");
String source = "the quick brown fox";
String[] expected =
new String[] {"the", "the quick", "quick", "quick brown", "brown", "brown fox", "fox"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
public void testDefault() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromSettings(
settingsBuilder()
.put("path.home", createTempDir().toString())
.put("index.analysis.filter.my_ascii_folding.type", "asciifolding")
.build());
TokenFilterFactory tokenFilter = analysisService.tokenFilter("my_ascii_folding");
String source = "Ansprüche";
String[] expected = new String[] {"Anspruche"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
public void testPreserveOriginal() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromSettings(
settingsBuilder()
.put(Environment.PATH_HOME_SETTING.getKey(), createTempDir().toString())
.put("index.analysis.filter.my_ascii_folding.type", "asciifolding")
.put("index.analysis.filter.my_ascii_folding.preserve_original", true)
.build());
TokenFilterFactory tokenFilter = analysisService.tokenFilter("my_ascii_folding");
String source = "Ansprüche";
String[] expected = new String[] {"Anspruche", "Ansprüche"};
Tokenizer tokenizer = new WhitespaceTokenizer();
tokenizer.setReader(new StringReader(source));
assertTokenStreamContents(tokenFilter.create(tokenizer), expected);
}
public void testLoadWithoutSettings() throws IOException {
AnalysisService analysisService =
AnalysisTestsHelper.createAnalysisServiceFromClassPath(createTempDir(), RESOURCE);
TokenFilterFactory tokenFilter = analysisService.tokenFilter("keep");
Assert.assertNull(tokenFilter);
}