Merge pull request #1431 from dimagi/fix-levenshtein-distance-algorithm

Fix Levenshtein distance algorithm
dimagi · Sep 4, 2024 · db0ffdd · db0ffdd
2 parents a615d84 + 5bc1663
commit db0ffdd
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 1 deletion.
diff --git a/src/main/java/org/commcare/cases/util/StringUtils.java b/src/main/java/org/commcare/cases/util/StringUtils.java
@@ -119,7 +119,7 @@ private static int LevenshteinDistance(String s0, String s1) {
         for (int j = 1; j < len1; j++) {
 
             // initial cost of skipping prefix in String s1
-            newcost[0] = j - 1;
+            newcost[0] = j;
 
             // transformation cost for each letter in s0
             for (int i = 1; i < len0; i++) {

diff --git a/src/test/java/org/commcare/util/reference/test/FuzzySearchTest.java b/src/test/java/org/commcare/util/reference/test/FuzzySearchTest.java
@@ -30,5 +30,8 @@ public void testFuzzySearch() {
         // false even though we have an exact substring match,
         // Cuz fuzzy search starts checking from 0th location.
         Assert.assertFalse(StringUtils.fuzzyMatch("Test", "CrazyTest").first);
+
+        // false since aply and cape has a difference of 3 edits add 'c', remove 'l', replace 'y' with 'e'
+        Assert.assertFalse(StringUtils.fuzzyMatch("aply", "cape").first);
     }
 }