[regexp] use C++ API for unicode set over C API.

src/regexp/regexp-parser.cc

 // 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.
 
 #include "src/regexp/regexp-parser.h"
 
 #include "src/char-predicates-inl.h"
 #include "src/factory.h"
 #include "src/isolate.h"
 #include "src/objects-inl.h"
 #include "src/ostreams.h"
 #include "src/regexp/jsregexp.h"
 #include "src/utils.h"
 
 #ifdef V8_I18N_SUPPORT
-#include "unicode/uset.h"
+#include "unicode/uniset.h"
 #endif  // V8_I18N_SUPPORT
 
 namespace v8 {
 namespace internal {
 
 RegExpParser::RegExpParser(FlatStringReader* in, Handle<String>* error,
                            JSRegExp::Flags flags, Isolate* isolate, Zone* zone)
     : isolate_(isolate),
       zone_(zone),
       error_(error),
@@ skipping 1066 matching lines @@
       u_getPropertyValueEnum(property_for_lookup, property_value_name);
   if (property_value == UCHAR_INVALID_CODE) return false;
 
   // We require the property name to match exactly to one of the property value
   // aliases. However, u_getPropertyValueEnum uses loose matching.
   if (!IsExactPropertyValueAlias(property_value_name, property_for_lookup,
                                  property_value)) {
     return false;
   }
 
-  USet* set = uset_openEmpty();
   UErrorCode ec = U_ZERO_ERROR;
-  uset_applyIntPropertyValue(set, property, property_value, &ec);
-  bool success = ec == U_ZERO_ERROR && !uset_isEmpty(set);
+  icu::UnicodeSet set;
+  set.applyIntPropertyValue(property, property_value, ec);
+  bool success = ec == U_ZERO_ERROR && !set.isEmpty();
 
   if (success) {
-    uset_removeAllStrings(set);
-    if (negate) uset_complement(set);
-    int item_count = uset_getItemCount(set);
-    int item_result = 0;
-    for (int i = 0; i < item_count; i++) {
-      uc32 start = 0;
-      uc32 end = 0;
-      item_result += uset_getItem(set, i, &start, &end, nullptr, 0, &ec);
-      result->Add(CharacterRange::Range(start, end), zone);
+    set.removeAllStrings();
+    if (negate) set.complement();
+    for (int i = 0; i < set.getRangeCount(); i++) {
+      result->Add(
+          CharacterRange::Range(set.getRangeStart(i), set.getRangeEnd(i)),
+          zone);
     }
-    DCHECK_EQ(U_ZERO_ERROR, ec);
-    DCHECK_EQ(0, item_result);
   }
-  uset_close(set);
   return success;
 }
 
 template <size_t N>
 inline bool NameEquals(const char* name, const char (&literal)[N]) {
   return strncmp(name, literal, N + 1) == 0;
 }
 
 bool LookupSpecialPropertyValueName(const char* name,
                                     ZoneList<CharacterRange>* result,
@@ skipping 592 matching lines @@
     // Check for lone surrogates.
     if (from <= kTrailSurrogateEnd && to >= kLeadSurrogateStart) return true;
   }
   return false;
 }
 
 
 bool RegExpBuilder::NeedsDesugaringForIgnoreCase(uc32 c) {
 #ifdef V8_I18N_SUPPORT
   if (unicode() && ignore_case()) {
-    USet* set = uset_open(c, c);
-    uset_closeOver(set, USET_CASE_INSENSITIVE);
-    uset_removeAllStrings(set);
-    bool result = uset_size(set) > 1;
-    uset_close(set);
-    return result;
+    icu::UnicodeSet set(c, c);
+    set.closeOver(USET_CASE_INSENSITIVE);
+    set.removeAllStrings();
+    return set.size() > 1;
   }
   // In the case where ICU is not included, we act as if the unicode flag is
   // not set, and do not desugar.
 #endif  // V8_I18N_SUPPORT
   return false;
 }
 
 
 RegExpTree* RegExpBuilder::ToRegExp() {
   FlushTerms();
@@ skipping 48 matching lines @@
     return false;
   }
   terms_.Add(new (zone()) RegExpQuantifier(min, max, quantifier_type, atom),
              zone());
   LAST(ADD_TERM);
   return true;
 }
 
 }  // namespace internal
 }  // namespace v8