issue #12 adopted the suggestion by yim1990 with a small change, so t…

…hat the keyword emit is lowercased as well
robert-bor · Sep 22, 2015 · 76ae822 · 76ae822
1 parent e2c5334
commit 76ae822
Show file tree

Hide file tree

Showing 2 changed files with 22 additions and 2 deletions.
diff --git a/src/main/java/org/ahocorasick/trie/Trie.java b/src/main/java/org/ahocorasick/trie/Trie.java
@@ -33,9 +33,12 @@ private void addKeyword(String keyword) {
         }
         State currentState = this.rootState;
         for (Character character : keyword.toCharArray()) {
+            if (trieConfig.isCaseInsensitive()) {
+                character = Character.toLowerCase(character);
+            }
             currentState = currentState.addState(character);
         }
-        currentState.addEmit(keyword);
+        currentState.addEmit(trieConfig.isCaseInsensitive() ? keyword.toLowerCase() : keyword);
     }
 
     public Collection<Token> tokenize(String text) {

diff --git a/src/test/java/org/ahocorasick/trie/TrieTest.java b/src/test/java/org/ahocorasick/trie/TrieTest.java
@@ -106,7 +106,24 @@ public void ushersTest() {
         checkEmit(iterator.next(), 2, 5, "hers");
     }
 
-	 @Test
+    @Test
+    public void ushersTestWithCapitalKeywords() {
+        Trie trie = Trie.builder()
+                .caseInsensitive()
+                .addKeyword("HERS")
+                .addKeyword("HIS")
+                .addKeyword("SHE")
+                .addKeyword("HE")
+                .build();
+        Collection<Emit> emits = trie.parseText("ushers");
+        assertEquals(3, emits.size()); // she @ 3, he @ 3, hers @ 5
+        Iterator<Emit> iterator = emits.iterator();
+        checkEmit(iterator.next(), 2, 3, "he");
+        checkEmit(iterator.next(), 1, 3, "she");
+        checkEmit(iterator.next(), 2, 5, "hers");
+    }
+
+    @Test
     public void ushersTestFirstMatch() {
          Trie trie = Trie.builder()
                 .addKeyword("hers")