View Javadoc
1   /*
2    * Licensed to the Apache Software Foundation (ASF) under one or more
3    * contributor license agreements.  See the NOTICE file distributed with
4    * this work for additional information regarding copyright ownership.
5    * The ASF licenses this file to You under the Apache License, Version 2.0
6    * (the "License"); you may not use this file except in compliance with
7    * the License.  You may obtain a copy of the License at
8    *
9    *      https://www.apache.org/licenses/LICENSE-2.0
10   *
11   * Unless required by applicable law or agreed to in writing, software
12   * distributed under the License is distributed on an "AS IS" BASIS,
13   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14   * See the License for the specific language governing permissions and
15   * limitations under the License.
16   */
17  package org.apache.commons.text.similarity;
18  
19  import static org.junit.jupiter.api.Assertions.assertArrayEquals;
20  import static org.junit.jupiter.api.Assertions.assertEquals;
21  import static org.junit.jupiter.api.Assertions.assertThrows;
22  
23  import org.junit.jupiter.api.BeforeAll;
24  import org.junit.jupiter.api.Test;
25  import org.junit.jupiter.params.ParameterizedTest;
26  import org.junit.jupiter.params.provider.MethodSource;
27  
28  /**
29   * Tests {@link JaroWinklerDistance}.
30   */
31  class JaroWinklerDistanceTest {
32  
33      private static JaroWinklerDistance distance;
34  
35      @BeforeAll
36      public static void setUp() {
37          distance = new JaroWinklerDistance();
38      }
39  
40      @Test
41      void testApply_NullSimilarityInput() {
42          assertThrows(IllegalArgumentException.class, () -> distance.apply(null, new SimilarityCharacterInput("a")));
43      }
44  
45      @Test
46      void testApply_SimilarityInputNull() {
47          assertThrows(IllegalArgumentException.class, () -> distance.apply(new SimilarityCharacterInput("a"), null));
48      }
49  
50      @ParameterizedTest
51      @MethodSource("org.apache.commons.text.similarity.SimilarityInputTest#similarityInputsEquals()")
52      void testGetJaroWinklerDistance(final Class<?> cls) {
53          assertEquals(0.07501d, distance.apply(SimilarityInputTest.build(cls, "frog"), SimilarityInputTest.build(cls, "fog")), 0.00001d);
54          assertEquals(1.0d, distance.apply(SimilarityInputTest.build(cls, "fly"), SimilarityInputTest.build(cls, "ant")), 0.00000000000000000001d);
55          assertEquals(0.55834d, distance.apply(SimilarityInputTest.build(cls, "elephant"), SimilarityInputTest.build(cls, "hippo")), 0.00001d);
56          assertEquals(0.09334d, distance.apply(SimilarityInputTest.build(cls, "ABC Corporation"), SimilarityInputTest.build(cls, "ABC Corp")), 0.00001d);
57          assertEquals(0.04749d,
58                  distance.apply(SimilarityInputTest.build(cls, "D N H Enterprises Inc"), SimilarityInputTest.build(cls, "D & H Enterprises, Inc.")), 0.00001d);
59          assertEquals(0.058d,
60                  distance.apply(SimilarityInputTest.build(cls, "My Gym Children's Fitness Center"), SimilarityInputTest.build(cls, "My Gym. Childrens Fitness")),
61                  0.00001d);
62          assertEquals(0.101982d, distance.apply(SimilarityInputTest.build(cls, "PENNSYLVANIA"), SimilarityInputTest.build(cls, "PENNCISYLVNIA")), 0.00001d);
63          assertEquals(0.028572d, distance.apply(SimilarityInputTest.build(cls, "/opt/software1"), SimilarityInputTest.build(cls, "/opt/software2")), 0.00001d);
64          assertEquals(0.058334d, distance.apply(SimilarityInputTest.build(cls, "aaabcd"), SimilarityInputTest.build(cls, "aaacdb")), 0.00001d);
65          assertEquals(0.088889d, distance.apply(SimilarityInputTest.build(cls, "John Horn"), SimilarityInputTest.build(cls, "John Hopkins")), 0.00001d);
66          assertEquals(0d, distance.apply(SimilarityInputTest.build(cls, ""), SimilarityInputTest.build(cls, "")), 0.00001d);
67          assertEquals(0d, distance.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "foo")), 0.00001d);
68          assertEquals(1 - 0.94166d, distance.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "foo ")), 0.00001d);
69          assertEquals(1 - 0.90666d, distance.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "foo  ")), 0.00001d);
70          assertEquals(1 - 0.86666d, distance.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, " foo ")), 0.00001d);
71          assertEquals(1 - 0.51111d, distance.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "  foo")), 0.00001d);
72      }
73  
74      @Test
75      void testGetJaroWinklerDistance_NullNull() {
76          assertThrows(IllegalArgumentException.class, () -> distance.apply((String) null, null));
77      }
78  
79      @Test
80      void testGetJaroWinklerDistance_NullString() {
81          assertThrows(IllegalArgumentException.class, () -> distance.apply(null, "clear"));
82      }
83  
84      @Test
85      void testGetJaroWinklerDistance_StringNull() {
86          assertThrows(IllegalArgumentException.class, () -> distance.apply(" ", null));
87      }
88  
89      @Test
90      void testGetJaroWinklerDistance_StringString() {
91          assertEquals(0.07501d, distance.apply("frog", "fog"), 0.00001d);
92          assertEquals(1.0d, distance.apply("fly", "ant"), 0.00000000000000000001d);
93          assertEquals(0.55834d, distance.apply("elephant", "hippo"), 0.00001d);
94          assertEquals(0.09334d, distance.apply("ABC Corporation", "ABC Corp"), 0.00001d);
95          assertEquals(0.04749d, distance.apply("D N H Enterprises Inc", "D & H Enterprises, Inc."), 0.00001d);
96          assertEquals(0.058d, distance.apply("My Gym Children's Fitness Center", "My Gym. Childrens Fitness"), 0.00001d);
97          assertEquals(0.101982d, distance.apply("PENNSYLVANIA", "PENNCISYLVNIA"), 0.00001d);
98          assertEquals(0.028572d, distance.apply("/opt/software1", "/opt/software2"), 0.00001d);
99          assertEquals(0.058334d, distance.apply("aaabcd", "aaacdb"), 0.00001d);
100         assertEquals(0.088889d, distance.apply("John Horn", "John Hopkins"), 0.00001d);
101         assertEquals(0d, distance.apply("", ""), 0.00001d);
102         assertEquals(0d, distance.apply("foo", "foo"), 0.00001d);
103         assertEquals(1 - 0.94166d, distance.apply("foo", "foo "), 0.00001d);
104         assertEquals(1 - 0.90666d, distance.apply("foo", "foo  "), 0.00001d);
105         assertEquals(1 - 0.86666d, distance.apply("foo", " foo "), 0.00001d);
106         assertEquals(1 - 0.51111d, distance.apply("foo", "  foo"), 0.00001d);
107     }
108 
109     @Test
110     void testMatches() {
111         assertArrayEquals(new int[]{2, 0, 2}, JaroWinklerSimilarity.matches("ab", "aba"));
112     }
113 
114     @SuppressWarnings("deprecation")
115     @Test
116     void testMatchesDeprecated() {
117         assertArrayEquals(new int[]{2, 0, 2}, JaroWinklerDistance.matches("ab", "aba"));
118     }
119 
120 }