diff options
author | Jon Bratseth <bratseth@gmail.com> | 2021-05-05 21:18:50 +0200 |
---|---|---|
committer | Jon Bratseth <bratseth@gmail.com> | 2021-05-05 21:18:50 +0200 |
commit | 346c5a523fec4ec7cecb373e480e30846f5e33e5 (patch) | |
tree | d8788eab6bb91d0b06a15d50f93f19d9d42ba96c /container-search/src/test/java/com/yahoo/prelude/query/parser | |
parent | a2c9cd4bc04f1a3eaa31524b3970b96be5c2eda9 (diff) |
Revert "Merge pull request #17754 from vespa-engine/revert-17747-bratseth/special-tokens-take-2"
This reverts commit a2c9cd4bc04f1a3eaa31524b3970b96be5c2eda9, reversing
changes made to 8c61a373af0066fbdf1cca354c24b197c7347321.
Diffstat (limited to 'container-search/src/test/java/com/yahoo/prelude/query/parser')
4 files changed, 61 insertions, 89 deletions
diff --git a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParseTestCase.java b/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParseTestCase.java index 6afea895f3a..cef8ae1751c 100644 --- a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParseTestCase.java +++ b/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParseTestCase.java @@ -18,16 +18,14 @@ import com.yahoo.prelude.query.PhraseSegmentItem; import com.yahoo.prelude.query.PrefixItem; import com.yahoo.prelude.query.RankItem; import com.yahoo.prelude.query.SubstringItem; -import com.yahoo.prelude.query.SubstringItem; import com.yahoo.prelude.query.SuffixItem; import com.yahoo.prelude.query.TaggableItem; import com.yahoo.prelude.query.WordItem; -import com.yahoo.prelude.query.parser.SpecialTokens; +import com.yahoo.language.process.SpecialTokens; import com.yahoo.prelude.query.parser.TestLinguistics; import com.yahoo.search.Query; import org.junit.Test; -import java.util.Collections; import java.util.Iterator; import static org.junit.Assert.assertEquals; @@ -1639,7 +1637,7 @@ public class ParseTestCase { @Test public void testNonSpecialTokenParsing() { - ParsingTester customTester = new ParsingTester(new SpecialTokens("default")); + ParsingTester customTester = new ParsingTester(SpecialTokens.empty()); customTester.assertParsed("OR c or c with (AND tcp ip)", "c# or c++ with tcp/ip", Query.Type.ANY); } diff --git a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParsingTester.java b/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParsingTester.java index 17155fff5de..fd7e4cbe0e6 100644 --- a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParsingTester.java +++ b/container-search/src/test/java/com/yahoo/prelude/query/parser/test/ParsingTester.java @@ -11,8 +11,8 @@ import com.yahoo.prelude.IndexFacts; import com.yahoo.prelude.IndexModel; import com.yahoo.prelude.query.Item; import com.yahoo.prelude.query.NullItem; -import com.yahoo.prelude.query.parser.SpecialTokenRegistry; -import com.yahoo.prelude.query.parser.SpecialTokens; +import com.yahoo.language.process.SpecialTokenRegistry; +import com.yahoo.language.process.SpecialTokens; import com.yahoo.search.Query; import com.yahoo.search.config.IndexInfoConfig; import com.yahoo.search.query.parser.Parsable; @@ -20,6 +20,9 @@ import com.yahoo.search.query.parser.Parser; import com.yahoo.search.query.parser.ParserEnvironment; import com.yahoo.search.query.parser.ParserFactory; +import java.util.ArrayList; +import java.util.List; + import static org.junit.Assert.assertEquals; import static org.junit.Assert.assertNotNull; import static org.junit.Assert.assertTrue; @@ -32,7 +35,7 @@ import static org.junit.Assert.assertTrue; public class ParsingTester { private static final Linguistics linguistics = new SimpleLinguistics(); - private IndexFacts indexFacts; + private final IndexFacts indexFacts; private SpecialTokenRegistry tokenRegistry; public ParsingTester() { @@ -49,11 +52,10 @@ public class ParsingTester { public ParsingTester(IndexFacts indexFacts, SpecialTokens specialTokens) { indexFacts.freeze(); - specialTokens.freeze(); this.indexFacts = indexFacts; tokenRegistry = new SpecialTokenRegistry(); - tokenRegistry.addSpecialTokens(specialTokens); + tokenRegistry = new SpecialTokenRegistry(List.of(specialTokens)); } /** @@ -72,13 +74,13 @@ public class ParsingTester { * This can be used to add new tokens and passing the resulting special tokens to the constructor of this. */ public static SpecialTokens createSpecialTokens() { - SpecialTokens tokens = new SpecialTokens("default"); - tokens.addSpecialToken("c++", null); - tokens.addSpecialToken(".net", "dotnet"); - tokens.addSpecialToken("tcp/ip", null); - tokens.addSpecialToken("c#", null); - tokens.addSpecialToken("special-token-fs","firstsecond"); - return tokens; + List<SpecialTokens.Token> tokens = new ArrayList<>(); + tokens.add(new SpecialTokens.Token("c++")); + tokens.add(new SpecialTokens.Token(".net", "dotnet")); + tokens.add(new SpecialTokens.Token("tcp/ip")); + tokens.add(new SpecialTokens.Token("c#")); + tokens.add(new SpecialTokens.Token("special-token-fs","firstsecond")); + return new SpecialTokens("default", tokens); } /** diff --git a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/TokenizerTestCase.java b/container-search/src/test/java/com/yahoo/prelude/query/parser/test/TokenizerTestCase.java index aa2e9dbcf75..e10fbd71c72 100644 --- a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/TokenizerTestCase.java +++ b/container-search/src/test/java/com/yahoo/prelude/query/parser/test/TokenizerTestCase.java @@ -6,12 +6,13 @@ import com.yahoo.prelude.Index; import com.yahoo.prelude.IndexFacts; import com.yahoo.prelude.IndexModel; import com.yahoo.prelude.SearchDefinition; -import com.yahoo.prelude.query.parser.SpecialTokenRegistry; -import com.yahoo.prelude.query.parser.SpecialTokens; +import com.yahoo.language.process.SpecialTokenRegistry; +import com.yahoo.language.process.SpecialTokens; import com.yahoo.prelude.query.parser.Token; import com.yahoo.prelude.query.parser.Tokenizer; import org.junit.Test; +import java.util.ArrayList; import java.util.Collections; import java.util.List; @@ -39,13 +40,11 @@ import static org.junit.Assert.assertTrue; */ public class TokenizerTestCase { - private SpecialTokenRegistry defaultRegistry = new SpecialTokenRegistry("file:src/test/java/com/yahoo/prelude/query/parser/test/replacingtokens.cfg"); - @Test public void testPlainTokenization() { Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens(createSpecialTokens()); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("default")); List<?> tokens = tokenizer.tokenize("drive (to hwy88, 88) +or language:en ugcapi_1 & &a"); assertEquals(new Token(WORD, "drive"), tokens.get(0)); @@ -87,7 +86,7 @@ public class TokenizerTestCase { public void testOneSpecialToken() { Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens(createSpecialTokens()); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("default")); List<?> tokens = tokenizer.tokenize("c++ lovers, please apply"); assertEquals(new Token(WORD, "c++"), tokens.get(0)); @@ -97,7 +96,7 @@ public class TokenizerTestCase { public void testSpecialTokenCombination() { Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens(createSpecialTokens()); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("default")); List<?> tokens = tokenizer.tokenize("c#, c++ or .net know, not tcp/ip"); assertEquals(new Token(WORD, "c#"), tokens.get(0)); @@ -123,10 +122,9 @@ public class TokenizerTestCase { */ @Test public void testSpecialTokenCJK() { - assertEquals("Special tokens configured", 6, defaultRegistry.getSpecialTokens("default").size()); Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); tokenizer.setSubstringSpecialTokens(true); - tokenizer.setSpecialTokens(defaultRegistry.getSpecialTokens("default")); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("replacing")); List<?> tokens = tokenizer.tokenize("fooc#bar,c++with spacebarknowknowknow,knowknownot know"); assertEquals(new Token(WORD, "foo"), tokens.get(0)); @@ -151,7 +149,7 @@ public class TokenizerTestCase { public void testSpecialTokenCaseInsensitive() { Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens(createSpecialTokens()); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("default")); List<?> tokens = tokenizer.tokenize("The AS/400 is great"); assertEquals(new Token(WORD, "The"), tokens.get(0)); @@ -167,7 +165,7 @@ public class TokenizerTestCase { public void testSpecialTokenNonMatch() { Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens(createSpecialTokens()); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("default")); List<?> tokens = tokenizer.tokenize("c++ c+ aS/400 i/o .net i/ooo ap.net"); assertEquals(new Token(WORD, "c++"), tokens.get(0)); @@ -190,18 +188,9 @@ public class TokenizerTestCase { @Test public void testSpecialTokenConfigurationDefault() { - String tokenFile = "file:src/test/java/com/yahoo/prelude/query/parser/test/specialtokens.cfg"; - - SpecialTokenRegistry r = new SpecialTokenRegistry(tokenFile); - assertEquals("Special tokens configured", 6, - r.getSpecialTokens("default").size()); - assertEquals("Special tokens configured", 4, - r.getSpecialTokens("other").size()); - Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens( - r.getSpecialTokens("default")); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("default")); List<?> tokens = tokenizer.tokenize( "with space, c++ or .... know, not b.s.d."); @@ -224,18 +213,9 @@ public class TokenizerTestCase { @Test public void testSpecialTokenConfigurationOther() { - String tokenFile = "file:src/test/java/com/yahoo/prelude/query/parser/test/specialtokens.cfg"; - - SpecialTokenRegistry r = new SpecialTokenRegistry(tokenFile); - assertEquals("Special tokens configured", 6, - r.getSpecialTokens("default").size()); - assertEquals("Special tokens configured", 4, - r.getSpecialTokens("other").size()); - Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens( - r.getSpecialTokens("other")); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("other")); List<?> tokens = tokenizer.tokenize( "with space,!!!*** [huh] or ------ " + "know, &&&%%% b.s.d."); @@ -267,26 +247,9 @@ public class TokenizerTestCase { } @Test - public void testSpecialTokenConfigurationMissing() { - String tokenFile = "file:source/bogus/specialtokens.cfg"; - - SpecialTokenRegistry r = new SpecialTokenRegistry(tokenFile); - - Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - - tokenizer.setSpecialTokens(r.getSpecialTokens("other")); - List<?> tokens = tokenizer.tokenize("c++"); - - assertEquals(new Token(WORD, "c"), tokens.get(0)); - assertEquals(new Token(PLUS, "+"), tokens.get(1)); - assertEquals(new Token(PLUS, "+"), tokens.get(2)); - } - - @Test public void testTokenReplacing() { - assertEquals("Special tokens configured", 6, defaultRegistry.getSpecialTokens("default").size()); Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens(defaultRegistry.getSpecialTokens("default")); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("replacing")); List<?> tokens = tokenizer.tokenize("with space, c++ or .... know, not b.s.d."); assertEquals(new Token(WORD, "with-space"), tokens.get(0)); @@ -745,7 +708,7 @@ public class TokenizerTestCase { public void testSingleQuoteAsWordCharacter() { Tokenizer tokenizer = new Tokenizer(new SimpleLinguistics()); - tokenizer.setSpecialTokens(createSpecialTokens()); + tokenizer.setSpecialTokens(createSpecialTokens().getSpecialTokens("default")); 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)); @@ -781,17 +744,38 @@ public class TokenizerTestCase { assertEquals(new Token(WORD, "a'"), tokens.get(30)); } - private SpecialTokens createSpecialTokens() { - SpecialTokens tokens = new SpecialTokens("default"); - - tokens.addSpecialToken("c+", null); - tokens.addSpecialToken("c++", null); - tokens.addSpecialToken(".net", null); - tokens.addSpecialToken("tcp/ip", null); - tokens.addSpecialToken("i/o", null); - tokens.addSpecialToken("c#", null); - tokens.addSpecialToken("AS/400", null); - return tokens; + private SpecialTokenRegistry createSpecialTokens() { + List<SpecialTokens.Token> tokens = new ArrayList<>(); + tokens.add(new SpecialTokens.Token("c+")); + tokens.add(new SpecialTokens.Token("c++")); + tokens.add(new SpecialTokens.Token(".net")); + tokens.add(new SpecialTokens.Token("tcp/ip")); + tokens.add(new SpecialTokens.Token("i/o")); + tokens.add(new SpecialTokens.Token("c#")); + tokens.add(new SpecialTokens.Token("AS/400")); + tokens.add(new SpecialTokens.Token("....")); + tokens.add(new SpecialTokens.Token("b.s.d.")); + tokens.add(new SpecialTokens.Token("with space")); + tokens.add(new SpecialTokens.Token("dvd\\xB1r")); + SpecialTokens defaultTokens = new SpecialTokens("default", tokens); + + tokens = new ArrayList<>(); + tokens.add(new SpecialTokens.Token("[huh]")); + tokens.add(new SpecialTokens.Token("&&&%%%")); + tokens.add(new SpecialTokens.Token("------")); + tokens.add(new SpecialTokens.Token("!!!***")); + SpecialTokens otherTokens = new SpecialTokens("other", tokens); + + tokens = new ArrayList<>(); + tokens.add(new SpecialTokens.Token("....")); + tokens.add(new SpecialTokens.Token("c++", "cpp")); + tokens.add(new SpecialTokens.Token("b.s.d.")); + tokens.add(new SpecialTokens.Token("with space", "with-space")); + tokens.add(new SpecialTokens.Token("c#")); + tokens.add(new SpecialTokens.Token("know", "knuwww")); + SpecialTokens replacingTokens = new SpecialTokens("replacing", tokens); + + return new SpecialTokenRegistry(List.of(defaultTokens, otherTokens, replacingTokens)); } } diff --git a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/replacingtokens.cfg b/container-search/src/test/java/com/yahoo/prelude/query/parser/test/replacingtokens.cfg deleted file mode 100644 index 6a189de0164..00000000000 --- a/container-search/src/test/java/com/yahoo/prelude/query/parser/test/replacingtokens.cfg +++ /dev/null @@ -1,12 +0,0 @@ -tokenlist[1] -tokenlist[0].name default -tokenlist[0].tokens[6] -tokenlist[0].tokens[0].token .... -tokenlist[0].tokens[1].token c++ -tokenlist[0].tokens[1].replace cpp -tokenlist[0].tokens[2].token b.s.d. -tokenlist[0].tokens[3].token with space -tokenlist[0].tokens[3].replace with-space -tokenlist[0].tokens[4].token c# -tokenlist[0].tokens[5].token know -tokenlist[0].tokens[5].replace knuwww |