[regexp] fix /ui regexp desugaring for text nodes.

R=littledan@chromium.org
BUG=v8:5036

Review-Url: https://codereview.chromium.org/2005753003
Cr-Commit-Position: refs/heads/master@{#36453}

src/regexp/regexp-parser.cc

@@ -1473,14 +1473,10 @@ void RegExpBuilder::FlushTerms() {
 
 bool RegExpBuilder::NeedsDesugaringForUnicode(RegExpCharacterClass* cc) {
   if (!unicode()) return false;
-  switch (cc->standard_type()) {
-    case 's':        // white space
-    case 'w':        // ASCII word character
-    case 'd':        // ASCII digit
-      return false;  // These characters do not need desugaring.
-    default:
-      break;
-  }
+  // TODO(yangguo): we could be smarter than this. Case-insensitivity does not
+  // necessarily mean that we need to desugar. It's probably nicer to have a
+  // separate pass to figure out unicode desugarings.
+  if (ignore_case()) return true;
   ZoneList<CharacterRange>* ranges = cc->ranges(zone());
   CharacterRange::Canonicalize(ranges);
   for (int i = ranges->length() - 1; i >= 0; i--) {

test/mjsunit/regress/regress-5036.js

+// Copyright 2016 the V8 project authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+// Flags: --harmony-unicode-regexps
+
+assertEquals(["1\u212a"], /\d\w/ui.exec("1\u212a"));