new method findSimilarWords

morfologik-speller/src/main/java/morfologik/speller/Speller.java

@@ -374,6 +374,26 @@ private void addReplacement(List<CandidateData> candidates, String replacement)
     }
   }
 
+  /**
+   * Find similar words even if the original word is a correct word that exists in the dictionary
+   * 
+   * @param word The original word.
+   * @return A list of suggested candidate replacements.
+   */
+  public ArrayList<CandidateData> findSimilarWordCandidates(String word) {
+    return findReplacementCandidates(word, true);
+  }
+
+  public ArrayList<String> findSimilarWords(String word) {
+    final List<CandidateData> result = findSimilarWordCandidates(word);
+    final ArrayList<String> resultSuggestions = new ArrayList<>(result.size());
+    for (CandidateData cd : result) {
+      resultSuggestions.add(cd.getWord());
+    }
+    return resultSuggestions;
+  }
+
+
   /**
    * Find suggestions by using K. Oflazer's algorithm. See Jan Daciuk's s_fsa
    * package, spell.cc for further explanation.
@@ -391,6 +411,7 @@ public ArrayList<String> findReplacements(String word) {
     return resultSuggestions;
   }
 
+
   /**
    * Find and return suggestions by using K. Oflazer's algorithm. See Jan Daciuk's s_fsa
    * package, spell.cc for further explanation. This method is identical to
@@ -400,14 +421,18 @@ public ArrayList<String> findReplacements(String word) {
    * @return A list of suggested candidate replacements.
    */
   public ArrayList<CandidateData> findReplacementCandidates(String word) {
+    return findReplacementCandidates(word, false);
+  }
+
+  public ArrayList<CandidateData> findReplacementCandidates(String word, boolean evenIfWordInDictionary) {
     if (!dictionaryMetadata.getInputConversionPairs().isEmpty()) {
       word = DictionaryLookup.applyReplacements(word, dictionaryMetadata.getInputConversionPairs());
     }
 
     // candidate strings, including same additional data such as edit distance from the original word.
     List<CandidateData> candidates = new ArrayList<>();
 
-    if (word.length() > 0 && word.length() < MAX_WORD_LENGTH && !isInDictionary(word)) {
+    if (word.length() > 0 && word.length() < MAX_WORD_LENGTH && (!isInDictionary(word) || evenIfWordInDictionary)) {
       List<String> wordsToCheck = new ArrayList<>();
       if (replacementsTheRest != null && word.length() > 1) {
         for (final String wordChecked : getAllReplacements(word, 0, 0)) {
@@ -467,7 +492,7 @@ public ArrayList<CandidateData> findReplacementCandidates(String word) {
     for (final CandidateData cd : candidates) {
       String replaced = DictionaryLookup.applyReplacements(cd.getWord(), dictionaryMetadata.getOutputConversionPairs());
       // Add only the first occurrence of a given word.
-      if (words.add(replaced)) {
+      if (words.add(replaced) && !replaced.equals(word)) {
         result.add(new CandidateData(replaced, cd.origDistance));
       }
     }

morfologik-speller/src/test/java/morfologik/speller/SpellerTest.java

@@ -165,6 +165,21 @@ public void testFindReplacementsUsingFrequency() throws IOException {
     assertTrue(reps.get(5).equals("dist"));
     assertTrue(reps.get(6).equals("gist"));
   }
+
+  @Test
+  public void testFindSimilarWords() throws IOException {
+    final URL url = getClass().getResource("dict-with-freq.dict");
+    final Speller spell = new Speller(Dictionary.read(url));
+
+    List<String> reps = spell.findSimilarWords("fist");
+    assertTrue(reps.toString().equals("[list, mist, dist, gist, wist, hist]"));
+    reps = spell.findSimilarWords("mist");
+    assertTrue(reps.toString().equals("[list, fist, dist, gist, wist, hist]"));
+    reps = spell.findSimilarWords("Fist");
+    assertTrue(reps.toString().equals("[fist, list, mist, dist, gist, wist, hist]"));
+    reps = spell.findSimilarWords("licit");
+    assertTrue(reps.toString().equals("[list, fist, mist, dist, gist, wist, hist]"));
+  }
 
   @Test
   public void testConcurrentReplacements() throws IOException {