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.assertEquals;
20  import static org.junit.jupiter.api.Assertions.assertThrows;
21  
22  import org.junit.jupiter.api.BeforeAll;
23  import org.junit.jupiter.api.Test;
24  import org.junit.jupiter.params.ParameterizedTest;
25  import org.junit.jupiter.params.provider.MethodSource;
26  
27  /**
28   * Tests {@link JaroWinklerSimilarity}.
29   */
30  class JaroWinklerSimilarityTest {
31  
32      private static JaroWinklerSimilarity similarity;
33  
34      @BeforeAll
35      public static void setUp() {
36          similarity = new JaroWinklerSimilarity();
37      }
38  
39      /**
40       * Wraps the string in a custom {@link CharSequence}. This ensures that using the {@link Object#equals(Object)} method on the input CharSequence to test for
41       * equality will fail.
42       *
43       * @param string the string
44       * @return the char sequence
45       */
46      private static CharSequence wrap(final String string) {
47          return new CharSequence() {
48  
49              @Override
50              public char charAt(final int index) {
51                  return string.charAt(index);
52              }
53  
54              @Override
55              public boolean equals(final Object obj) {
56                  return string.equals(obj);
57              }
58  
59              @Override
60              public int hashCode() {
61                  return string.hashCode();
62              }
63  
64              @Override
65              public int length() {
66                  return string.length();
67              }
68  
69              @Override
70              public CharSequence subSequence(final int start, final int end) {
71                  return string.subSequence(start, end);
72              }
73  
74              @Override
75              public String toString() {
76                  return string;
77              }
78          };
79      }
80  
81      @ParameterizedTest
82      @MethodSource("org.apache.commons.text.similarity.SimilarityInputTest#similarityInputsEquals()")
83      void testGetJaroWinklerSimilarity(final Class<?> cls) {
84          assertEquals(1d, similarity.apply(SimilarityInputTest.build(cls, ""), SimilarityInputTest.build(cls, "")), 0.00001d);
85          assertEquals(1d, similarity.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "foo")), 0.00001d);
86          assertEquals(0.94166d, similarity.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "foo ")), 0.00001d);
87          assertEquals(0.90666d, similarity.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "foo  ")), 0.00001d);
88          assertEquals(0.86666d, similarity.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, " foo ")), 0.00001d);
89          assertEquals(0.51111d, similarity.apply(SimilarityInputTest.build(cls, "foo"), SimilarityInputTest.build(cls, "  foo")), 0.00001d);
90          assertEquals(0.92499d, similarity.apply(SimilarityInputTest.build(cls, "frog"), SimilarityInputTest.build(cls, "fog")), 0.00001d);
91          assertEquals(0.0d, similarity.apply(SimilarityInputTest.build(cls, "fly"), SimilarityInputTest.build(cls, "ant")), 0.00000000000000000001d);
92          assertEquals(0.44166d, similarity.apply(SimilarityInputTest.build(cls, "elephant"), SimilarityInputTest.build(cls, "hippo")), 0.00001d);
93          assertEquals(0.90666d, similarity.apply(SimilarityInputTest.build(cls, "ABC Corporation"), SimilarityInputTest.build(cls, "ABC Corp")), 0.00001d);
94          assertEquals(0.95251d,
95                  similarity.apply(SimilarityInputTest.build(cls, "D N H Enterprises Inc"), SimilarityInputTest.build(cls, "D & H Enterprises, Inc.")), 0.00001d);
96          assertEquals(0.942d, similarity.apply(SimilarityInputTest.build(cls, "My Gym Children's Fitness Center"),
97                  SimilarityInputTest.build(cls, "My Gym. Childrens Fitness")), 0.00001d);
98          assertEquals(0.898018d, similarity.apply(SimilarityInputTest.build(cls, "PENNSYLVANIA"), SimilarityInputTest.build(cls, "PENNCISYLVNIA")), 0.00001d);
99          assertEquals(0.971428d, similarity.apply(SimilarityInputTest.build(cls, "/opt/software1"), SimilarityInputTest.build(cls, "/opt/software2")), 0.00001d);
100         assertEquals(0.941666d, similarity.apply(SimilarityInputTest.build(cls, "aaabcd"), SimilarityInputTest.build(cls, "aaacdb")), 0.00001d);
101         assertEquals(0.911111d, similarity.apply(SimilarityInputTest.build(cls, "John Horn"), SimilarityInputTest.build(cls, "John Hopkins")), 0.00001d);
102     }
103 
104     @Test
105     void testGetJaroWinklerSimilarity_NullNull() {
106         assertThrows(IllegalArgumentException.class, () -> similarity.apply((String) null, null));
107     }
108 
109     @Test
110     void testGetJaroWinklerSimilarity_NullString() {
111         assertThrows(IllegalArgumentException.class, () -> similarity.apply(null, "clear"));
112     }
113 
114     @Test
115     void testGetJaroWinklerSimilarity_StringNull() {
116         assertThrows(IllegalArgumentException.class, () -> similarity.apply(" ", null));
117     }
118 
119     @Test
120     void testGetJaroWinklerSimilarity_StringString() {
121         assertEquals(1d, similarity.apply(wrap(""), ""), 0.00001d);
122         assertEquals(1d, similarity.apply(wrap("foo"), "foo"), 0.00001d);
123         assertEquals(0.94166d, similarity.apply(wrap("foo"), "foo "), 0.00001d);
124         assertEquals(0.90666d, similarity.apply(wrap("foo"), "foo  "), 0.00001d);
125         assertEquals(0.86666d, similarity.apply(wrap("foo"), " foo "), 0.00001d);
126         assertEquals(0.51111d, similarity.apply(wrap("foo"), "  foo"), 0.00001d);
127         assertEquals(0.92499d, similarity.apply(wrap("frog"), "fog"), 0.00001d);
128         assertEquals(0.0d, similarity.apply(wrap("fly"), "ant"), 0.00000000000000000001d);
129         assertEquals(0.44166d, similarity.apply(wrap("elephant"), "hippo"), 0.00001d);
130         assertEquals(0.90666d, similarity.apply(wrap("ABC Corporation"), "ABC Corp"), 0.00001d);
131         assertEquals(0.95251d, similarity.apply(wrap("D N H Enterprises Inc"), "D & H Enterprises, Inc."), 0.00001d);
132         assertEquals(0.942d, similarity.apply(wrap("My Gym Children's Fitness Center"), "My Gym. Childrens Fitness"), 0.00001d);
133         assertEquals(0.898018d, similarity.apply(wrap("PENNSYLVANIA"), "PENNCISYLVNIA"), 0.00001d);
134         assertEquals(0.971428d, similarity.apply(wrap("/opt/software1"), "/opt/software2"), 0.00001d);
135         assertEquals(0.941666d, similarity.apply(wrap("aaabcd"), "aaacdb"), 0.00001d);
136         assertEquals(0.911111d, similarity.apply(wrap("John Horn"), "John Hopkins"), 0.00001d);
137     }
138 
139 }