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

src/regexp/jsregexp.cc

 // Copyright 2012 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/jsregexp.h"
 
 #include <memory>
 
 #include "src/base/platform/platform.h"
 #include "src/compilation-cache.h"
 #include "src/elements.h"
 #include "src/execution.h"
 #include "src/factory.h"
 #include "src/isolate-inl.h"
 #include "src/messages.h"
 #include "src/ostreams.h"
 #include "src/regexp/interpreter-irregexp.h"
 #include "src/regexp/jsregexp-inl.h"
 #include "src/regexp/regexp-macro-assembler-irregexp.h"
 #include "src/regexp/regexp-macro-assembler-tracer.h"
 #include "src/regexp/regexp-macro-assembler.h"
 #include "src/regexp/regexp-parser.h"
 #include "src/regexp/regexp-stack.h"
 #include "src/runtime/runtime.h"
 #include "src/splay-tree-inl.h"
 #include "src/string-search.h"
 #include "src/unicode-decoder.h"
 
 #ifdef V8_I18N_SUPPORT
-#include "unicode/uset.h"
+#include "unicode/uniset.h"
 #include "unicode/utypes.h"
 #endif  // V8_I18N_SUPPORT
 
 #ifndef V8_INTERPRETED_REGEXP
 #if V8_TARGET_ARCH_IA32
 #include "src/regexp/ia32/regexp-macro-assembler-ia32.h"
 #elif V8_TARGET_ARCH_X64
 #include "src/regexp/x64/regexp-macro-assembler-x64.h"
 #elif V8_TARGET_ARCH_ARM64
 #include "src/regexp/arm64/regexp-macro-assembler-arm64.h"
@@ skipping 5066 matching lines @@
   return TextNode::CreateForCharacterRanges(zone, range, false, on_success);
 }
 
 
 void AddUnicodeCaseEquivalents(RegExpCompiler* compiler,
                                ZoneList<CharacterRange>* ranges) {
 #ifdef V8_I18N_SUPPORT
   // Use ICU to compute the case fold closure over the ranges.
   DCHECK(compiler->unicode());
   DCHECK(compiler->ignore_case());
-  USet* set = uset_openEmpty();
+  icu::UnicodeSet set;
   for (int i = 0; i < ranges->length(); i++) {
-    uset_addRange(set, ranges->at(i).from(), ranges->at(i).to());
+    set.add(ranges->at(i).from(), ranges->at(i).to());
   }
   ranges->Clear();
-  uset_closeOver(set, USET_CASE_INSENSITIVE);
+  set.closeOver(USET_CASE_INSENSITIVE);
   // Full case mapping map single characters to multiple characters.
   // Those are represented as strings in the set. Remove them so that
   // we end up with only simple and common case mappings.
-  uset_removeAllStrings(set);
-  int item_count = uset_getItemCount(set);
-  int item_result = 0;
-  UErrorCode ec = U_ZERO_ERROR;
+  set.removeAllStrings();
   Zone* zone = compiler->zone();
-  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);
-    ranges->Add(CharacterRange::Range(start, end), zone);
+  for (int i = 0; i < set.getRangeCount(); i++) {
+    ranges->Add(CharacterRange::Range(set.getRangeStart(i), set.getRangeEnd(i)),
+                zone);
   }
   // No errors and everything we collected have been ranges.
-  DCHECK_EQ(U_ZERO_ERROR, ec);
-  DCHECK_EQ(0, item_result);
-  uset_close(set);
 #else
   // Fallback if ICU is not included.
   CharacterRange::AddCaseEquivalents(compiler->isolate(), compiler->zone(),
                                      ranges, compiler->one_byte());
 #endif  // V8_I18N_SUPPORT
   CharacterRange::Canonicalize(ranges);
 }
 
 
 RegExpNode* RegExpCharacterClass::ToNode(RegExpCompiler* compiler,
@@ skipping 1719 matching lines @@
 
 
 void RegExpResultsCache::Clear(FixedArray* cache) {
   for (int i = 0; i < kRegExpResultsCacheSize; i++) {
     cache->set(i, Smi::kZero);
   }
 }
 
 }  // namespace internal
 }  // namespace v8