Merge branch 'master' of https://github.com/jmkeil/commons-text

apache · Aug 8, 2018 · 6a79e11 · 6a79e11
2 parents 97d6064 + 5d14854
commit 6a79e11
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 13 deletions.
diff --git a/src/main/java/org/apache/commons/text/similarity/JaroWinklerDistance.java b/src/main/java/org/apache/commons/text/similarity/JaroWinklerDistance.java
@@ -85,17 +85,17 @@ public Double apply(final CharSequence left, final CharSequence right) {
         if (m == 0) {
             return 0D;
         }
-        final double j = ((m / left.length() + m / right.length() + (m - mtp[1]) / m)) / 3;
-        final double jw = j < 0.7D ? j : j + Math.min(defaultScalingFactor, 1D / mtp[3]) * mtp[2] * (1D - j);
+        final double j = ((m / left.length() + m / right.length() + (m - (double) mtp[1] / 2) / m)) / 3;
+        final double jw = j < 0.7D ? j : j + defaultScalingFactor * mtp[2] * (1D - j);
         return jw;
     }
 
     /**
-     * This method returns the Jaro-Winkler string matches, transpositions, prefix, max array.
+     * This method returns the Jaro-Winkler string matches, half transpositions, prefix array.
      *
      * @param first the first string to be matched
      * @param second the second string to be matched
-     * @return mtp array containing: matches, transpositions, prefix, and max length
+     * @return mtp array containing: matches, half transpositions, and prefix
      */
     protected static int[] matches(final CharSequence first, final CharSequence second) {
         CharSequence max, min;
@@ -136,21 +136,21 @@ protected static int[] matches(final CharSequence first, final CharSequence seco
                 si++;
             }
         }
-        int transpositions = 0;
+        int halfTranspositions = 0;
         for (int mi = 0; mi < ms1.length; mi++) {
             if (ms1[mi] != ms2[mi]) {
-                transpositions++;
+                halfTranspositions++;
             }
         }
         int prefix = 0;
-        for (int mi = 0; mi < min.length(); mi++) {
+        for (int mi = 0; mi < Math.min(4, min.length()); mi++) {
             if (first.charAt(mi) == second.charAt(mi)) {
                 prefix++;
             } else {
                 break;
             }
         }
-        return new int[] {matches, transpositions / 2, prefix, max.length()};
+        return new int[] {matches, halfTranspositions, prefix};
     }
 
 }
diff --git a/src/test/java/org/apache/commons/text/similarity/JaroWinklerDistanceTest.java b/src/test/java/org/apache/commons/text/similarity/JaroWinklerDistanceTest.java
@@ -39,12 +39,14 @@ public void testGetJaroWinklerDistance_StringString() {
         assertEquals(0.92499d, distance.apply("frog", "fog"), 0.00001d);
         assertEquals(0.0d, distance.apply("fly", "ant"), 0.00000000000000000001d);
         assertEquals(0.44166d, distance.apply("elephant", "hippo"), 0.00001d);
-        assertEquals(0.92740d, distance.apply("ABC Corporation", "ABC Corp"), 0.00001d);
-        assertEquals(0.94580d, distance.apply("D N H Enterprises Inc", "D & H Enterprises, Inc."), 0.00001d);
-        assertEquals(0.921458d,
+        assertEquals(0.90666d, distance.apply("ABC Corporation", "ABC Corp"), 0.00001d);
+        assertEquals(0.95251d, distance.apply("D N H Enterprises Inc", "D & H Enterprises, Inc."), 0.00001d);
+        assertEquals(0.942d,
                 distance.apply("My Gym Children's Fitness Center", "My Gym. Childrens Fitness"), 0.00001d);
-        assertEquals(0.882329d, distance.apply("PENNSYLVANIA", "PENNCISYLVNIA"), 0.00001d);
-        assertEquals(0.996598d, distance.apply("/opt/software1", "/opt/software2"), 0.00001d);
+        assertEquals(0.898018d, distance.apply("PENNSYLVANIA", "PENNCISYLVNIA"), 0.00001d);
+        assertEquals(0.971428d, distance.apply("/opt/software1", "/opt/software2"), 0.00001d);
+        assertEquals(0.941666d, distance.apply("aaabcd", "aaacdb"), 0.00001d);
+        assertEquals(0.911111d, distance.apply("John Horn", "John Hopkins"), 0.00001d);
     }
 
     @Test