Examples with Tokenizer com.yahoo.prelude.query.parser.Tokenizer used on opensource projects

Example 1 with Tokenizer

use of com.yahoo.prelude.query.parser.Tokenizer in project vespa by vespa-engine.

the class TokenizerTestCase method testOutsideBMPCodepoints.

@Test
public void testOutsideBMPCodepoints() {
    Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics());
    List<?> tokens = tokenizer.tokenize("\ud841\udd47");
    assertEquals(new Token(WORD, "\ud841\udd47"), tokens.get(0));
}

Example 2 with Tokenizer

use of com.yahoo.prelude.query.parser.Tokenizer in project vespa by vespa-engine.

the class TokenizerTestCase method testOneSpecialToken.

@Test
public void testOneSpecialToken() {
    Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics());
    tokenizer.setSpecialTokens(createSpecialTokens());
    List<?> tokens = tokenizer.tokenize("c++ lovers, please apply");
    assertEquals(new Token(WORD, "c++"), tokens.get(0));
}

Example 3 with Tokenizer

use of com.yahoo.prelude.query.parser.Tokenizer in project vespa by vespa-engine.

the class TokenizerTestCase method testSingleQuoteAsWordCharacter.

@Test
public void testSingleQuoteAsWordCharacter() {
    Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics());
    tokenizer.setSpecialTokens(createSpecialTokens());
    List<?> tokens = tokenizer.tokenize("drive (to hwy88, 88) +or language:en nalle:a'a ugcapi_1 'a' 'a a'");
    assertEquals(new Token(WORD, "drive"), tokens.get(0));
    assertEquals(new Token(SPACE, " "), tokens.get(1));
    assertEquals(new Token(LBRACE, "("), tokens.get(2));
    assertEquals(new Token(WORD, "to"), tokens.get(3));
    assertEquals(new Token(SPACE, " "), tokens.get(4));
    assertEquals(new Token(WORD, "hwy88"), tokens.get(5));
    assertEquals(new Token(COMMA, ","), tokens.get(6));
    assertEquals(new Token(SPACE, " "), tokens.get(7));
    assertEquals(new Token(NUMBER, "88"), tokens.get(8));
    assertEquals(new Token(RBRACE, ")"), tokens.get(9));
    assertEquals(new Token(SPACE, " "), tokens.get(10));
    assertEquals(new Token(PLUS, "+"), tokens.get(11));
    assertEquals(new Token(WORD, "or"), tokens.get(12));
    assertEquals(new Token(SPACE, " "), tokens.get(13));
    assertEquals(new Token(WORD, "language"), tokens.get(14));
    assertEquals(new Token(COLON, ":"), tokens.get(15));
    assertEquals(new Token(WORD, "en"), tokens.get(16));
    assertEquals(new Token(SPACE, " "), tokens.get(17));
    assertEquals(new Token(WORD, "nalle"), tokens.get(18));
    assertEquals(new Token(COLON, ":"), tokens.get(19));
    assertEquals(new Token(WORD, "a'a"), tokens.get(20));
    assertEquals(new Token(SPACE, " "), tokens.get(21));
    assertEquals(new Token(WORD, "ugcapi"), tokens.get(22));
    assertEquals(new Token(UNDERSCORE, "_"), tokens.get(23));
    assertEquals(new Token(NUMBER, "1"), tokens.get(24));
    assertEquals(new Token(SPACE, " "), tokens.get(25));
    assertEquals(new Token(WORD, "'a'"), tokens.get(26));
    assertEquals(new Token(SPACE, " "), tokens.get(27));
    assertEquals(new Token(WORD, "'a"), tokens.get(28));
    assertEquals(new Token(SPACE, " "), tokens.get(29));
    assertEquals(new Token(WORD, "a'"), tokens.get(30));
}

Example 4 with Tokenizer

use of com.yahoo.prelude.query.parser.Tokenizer in project vespa by vespa-engine.

the class TokenizerTestCase method testExactMatchTokenizationEndsByColon.

@Test
public void testExactMatchTokenizationEndsByColon() {
    Index index1 = new Index("testexact1");
    index1.setExact(true, null);
    Index index2 = new Index("testexact2");
    index2.setExact(true, "()/aa*::*&");
    IndexFacts facts = new IndexFacts();
    facts.addIndex("testsd", index1);
    facts.addIndex("testsd", index2);
    Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics());
    IndexFacts.Session session = facts.newSession(Collections.emptySet(), Collections.emptySet());
    List<?> tokens = tokenizer.tokenize("normal a:b (normal testexact1:!/%#%&+-+ ) testexact2:ho_/&%&/()/aa*::*&b:", session);
    assertEquals(new Token(WORD, "normal"), tokens.get(0));
    assertEquals(new Token(SPACE, " "), tokens.get(1));
    assertEquals(new Token(WORD, "a"), tokens.get(2));
    assertEquals(new Token(COLON, ":"), tokens.get(3));
    assertEquals(new Token(WORD, "b"), tokens.get(4));
    assertEquals(new Token(SPACE, " "), tokens.get(5));
    assertEquals(new Token(LBRACE, "("), tokens.get(6));
    assertEquals(new Token(WORD, "normal"), tokens.get(7));
    assertEquals(new Token(SPACE, " "), tokens.get(8));
    assertEquals(new Token(WORD, "testexact1"), tokens.get(9));
    assertEquals(new Token(COLON, ":"), tokens.get(10));
    assertEquals(new Token(WORD, "!/%#%&+-+"), tokens.get(11));
    assertEquals(new Token(SPACE, " "), tokens.get(12));
    assertEquals(new Token(RBRACE, ")"), tokens.get(13));
    assertEquals(new Token(SPACE, " "), tokens.get(14));
    assertEquals(new Token(WORD, "testexact2"), tokens.get(15));
    assertEquals(new Token(COLON, ":"), tokens.get(16));
    assertEquals(new Token(WORD, "ho_/&%&/"), tokens.get(17));
    assertEquals(new Token(WORD, "b"), tokens.get(18));
    assertEquals(new Token(COLON, ":"), tokens.get(19));
}

Example 5 with Tokenizer

use of com.yahoo.prelude.query.parser.Tokenizer in project vespa by vespa-engine.

the class TokenizerTestCase method testTokenReplacing.

@Test
public void testTokenReplacing() {
    assertEquals("Special tokens configured", 6, defaultRegistry.getSpecialTokens("default").size());
    Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics());
    tokenizer.setSpecialTokens(defaultRegistry.getSpecialTokens("default"));
    List<?> tokens = tokenizer.tokenize("with space, c++ or .... know, not b.s.d.");
    assertEquals(new Token(WORD, "with-space"), tokens.get(0));
    assertEquals(new Token(COMMA, ","), tokens.get(1));
    assertEquals(new Token(SPACE, " "), tokens.get(2));
    assertEquals(new Token(WORD, "cpp"), tokens.get(3));
    assertEquals(new Token(SPACE, " "), tokens.get(4));
    assertEquals(new Token(WORD, "or"), tokens.get(5));
    assertEquals(new Token(SPACE, " "), tokens.get(6));
    assertEquals(new Token(WORD, "...."), tokens.get(7));
    assertEquals(new Token(SPACE, " "), tokens.get(8));
    assertEquals(new Token(WORD, "knuwww"), tokens.get(9));
    assertEquals(new Token(COMMA, ","), tokens.get(10));
    assertEquals(new Token(SPACE, " "), tokens.get(11));
    assertEquals(new Token(WORD, "not"), tokens.get(12));
    assertEquals(new Token(SPACE, " "), tokens.get(13));
    assertEquals(new Token(WORD, "b.s.d."), tokens.get(14));
    assertTrue(((Token) tokens.get(9)).isSpecial());
    assertFalse(((Token) tokens.get(12)).isSpecial());
}