LUCENE-9782: Hunspell suggestions: split by space (but not dash) also…

… before last char (apache#2387)
ramil4icr · Feb 18, 2021 · 589eefc · 589eefc
1 parent f879c6a
commit 589eefc
Show file tree

Hide file tree

Showing 4 changed files with 5 additions and 3 deletions.
diff --git a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/ModifyingSuggester.java b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/ModifyingSuggester.java
@@ -302,12 +302,12 @@ private List<String> checkDictionaryForSplitSuggestions(String word) {
   }
 
   private void trySplitting(String word) {
-    for (int i = 1; i < word.length() - 1; i++) {
+    for (int i = 1; i < word.length(); i++) {
       String w1 = word.substring(0, i);
       String w2 = word.substring(i);
       if (checkSimpleWord(w1) && checkSimpleWord(w2)) {
         result.add(w1 + " " + w2);
-        if (shouldSplitByDash()) {
+        if (w1.length() > 1 && w2.length() > 1 && shouldSplitByDash()) {
           result.add(w1 + "-" + w2);
         }
       }

diff --git a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/sug.aff b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/sug.aff
@@ -13,6 +13,7 @@
 
 # switch off ngram suggestion for testing
 MAXNGRAMSUGS 0
+TRY -
 REP 2
 REP alot a_lot
 REP inspite in_spite

diff --git a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/sug.sug b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/sug.sug
@@ -11,5 +11,5 @@ which
 Gandhi
 McDonald
 permanent
-
+in, in a
 
diff --git a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/sug.wrong b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/sug.wrong
@@ -13,3 +13,4 @@ Mcdonald
 permqnent
 permanent-vacation
 permqnent-vacation
+ina