8330590: TextInputControl: previous word fails with Bhojpuri characters

Reviewed-by: kpk, arapte

Author: Andy Goryachev

modules/javafx.controls/src/main/java/javafx/scene/control/TextInputControl.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2011, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2011, 2024, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -810,8 +810,7 @@ private void previousWord(boolean select) {
         int pos = wordIterator.preceding(Utils.clamp(0, getCaretPosition(), textLength));
 
         // Skip the non-word region, then move/select to the beginning of the word.
-        while (pos != BreakIterator.DONE &&
-               !Character.isLetterOrDigit(text.charAt(Utils.clamp(0, pos, textLength-1)))) {
+        while (pos != BreakIterator.DONE && !isLetterOrDigit(text, pos)) {
             pos = wordIterator.preceding(Utils.clamp(0, pos, textLength));
         }
 
@@ -1743,4 +1742,15 @@ public int getAnchor() {
         }
     }
 
+    private static boolean isLetterOrDigit(String text, int ix) {
+        if (ix < 0) {
+            // should not happen
+            return false;
+        } else if (ix >= text.length()) {
+            return false;
+        }
+        // ignore the case when 'c' is a high surrogate without the low surrogate
+        int c = Character.codePointAt(text, ix);
+        return Character.isLetterOrDigit(c);
+    }
 }

modules/javafx.controls/src/test/java/test/javafx/scene/control/TextInputControlTest.java

@@ -2145,4 +2145,29 @@ private void copy(String string) {
         content.putString(string);
         Clipboard.getSystemClipboard().setContent(content);
     }
+
+    @Test
+    public void previousWord_Bhojpuri() {
+        // "Bhojpuri \ud804\udca6\ud804\udcb7\ud804\udc94\ud804\udca3\ud804\udcb3\ud804\udca9\ud804\udcb2 test"
+        textInput.setText("Bhojpuri 𑂦𑂷𑂔𑂣𑂳𑂩𑂲 test");
+        textInput.end();
+        verifyCaret(28);
+        textInput.previousWord(); // at the beginning of "test"
+        verifyCaret(24);
+        textInput.previousWord(); // at the beginning of "𑂦𑂷𑂔𑂣𑂳𑂩𑂲"
+        verifyCaret(9);
+        textInput.previousWord(); // at the beginning of "Bhojpuri"
+        verifyCaret(0);
+    }
+
+    private void verifyCaret(int index) {
+        verifyCaret(index, index, false);
+    }
+
+    private void verifyCaret(int caret, int anchor, boolean reverse) {
+        assertEquals(caret, textInput.getCaretPosition());
+        assertEquals(anchor, textInput.getAnchor());
+        IndexRange sel = reverse ? new IndexRange(caret, anchor) : new IndexRange(anchor, caret);
+        assertEquals(sel, textInput.getSelection());
+    }
 }