[regexp] implement regexp property sequence proposal

Also-By: mathias@chromium.org
Bug: v8:7467
Change-Id: I9fd6e61f4da1097c2375f671b4801e9730f792c4
Reviewed-on: https://chromium-review.googlesource.com/1227974
Commit-Queue: Yang Guo <yangguo@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Mathias Bynens <mathias@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55991}

BUILD.gn

@@ -2360,6 +2360,8 @@ v8_source_set("v8_base") {
     "src/regexp/jsregexp-inl.h",
     "src/regexp/jsregexp.cc",
     "src/regexp/jsregexp.h",
+    "src/regexp/property-sequences.cc",
+    "src/regexp/property-sequences.h",
     "src/regexp/regexp-ast.cc",
     "src/regexp/regexp-ast.h",
     "src/regexp/regexp-macro-assembler-irregexp-inl.h",

src/bootstrapper.cc

@@ -4767,6 +4767,8 @@ void Genesis::InitializeGlobal_harmony_intl_relative_time_format() {
 
 #endif  // V8_INTL_SUPPORT
 
+void Genesis::InitializeGlobal_harmony_regexp_sequence() {}
+
 Handle<JSFunction> Genesis::CreateArrayBuffer(
     Handle<String> name, ArrayBufferKind array_buffer_kind) {
   // Create the %ArrayBufferPrototype%

src/flag-definitions.h

@@ -209,10 +209,11 @@ DEFINE_IMPLICATION(harmony_class_fields, harmony_private_fields)
 // Update bootstrapper.cc whenever adding a new feature flag.
 
 // Features that are still work in progress (behind individual flags).
-#define HARMONY_INPROGRESS_BASE(V)                                    \
-  V(harmony_do_expressions, "harmony do-expressions")                 \
-  V(harmony_class_fields, "harmony fields in class literals")         \
-  V(harmony_await_optimization, "harmony await taking 1 tick")
+#define HARMONY_INPROGRESS_BASE(V)                             \
+  V(harmony_do_expressions, "harmony do-expressions")          \
+  V(harmony_class_fields, "harmony fields in class literals")  \
+  V(harmony_await_optimization, "harmony await taking 1 tick") \
+  V(harmony_regexp_sequence, "RegExp Unicode sequence properties")
 
 #ifdef V8_INTL_SUPPORT
 #define HARMONY_INPROGRESS(V)      \

src/regexp/property-sequences.h

+// Copyright 2018 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.
+
+#ifndef V8_REGEXP_PROPERTY_SEQUENCES_H_
+#define V8_REGEXP_PROPERTY_SEQUENCES_H_
+
+#ifdef V8_INTL_SUPPORT
+
+#include "src/globals.h"
+
+namespace v8 {
+namespace internal {
+
+class UnicodePropertySequences : public AllStatic {
+ public:
+  static const uc32 kEmojiFlagSequences[];
+
+  static const uc32 kEmojiTagSequences[];
+  static const uc32 kEmojiZWJSequences[];
+};
+
+}  // namespace internal
+}  // namespace v8
+
+#endif  // V8_INTL_SUPPORT
+
+#endif  // V8_REGEXP_PROPERTY_SEQUENCES_H_

src/regexp/regexp-parser.h

@@ -176,7 +176,14 @@ class RegExpParser {
   bool ParseHexEscape(int length, uc32* value);
   bool ParseUnicodeEscape(uc32* value);
   bool ParseUnlimitedLengthHexNumber(int max_value, uc32* value);
-  bool ParsePropertyClass(ZoneList<CharacterRange>* result, bool negate);
+
+  bool ParsePropertyClassName(std::vector<char>* name_1,
+                              std::vector<char>* name_2);
+  bool AddPropertyClassRange(ZoneList<CharacterRange>* add_to, bool negate,
+                             const std::vector<char>& name_1,
+                             const std::vector<char>& name_2);
+
+  RegExpTree* GetPropertySequence(const std::vector<char>& name_1);
   RegExpTree* ParseCharacterClass(const RegExpBuilder* state);
 
   uc32 ParseOctalLiteral();

test/mjsunit/harmony/regexp-property-emoji-keycap-sequence-generated.js

+// Copyright 2018 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-regexp-sequence
+
+const re = /\p{Emoji_Keycap_Sequence}/u;
+
+assertTrue(re.test('#\uFE0F\u20E3'));
+assertTrue(re.test('9\uFE0F\u20E3'));
+assertTrue(re.test('0\uFE0F\u20E3'));
+assertTrue(re.test('1\uFE0F\u20E3'));
+assertTrue(re.test('2\uFE0F\u20E3'));
+assertTrue(re.test('3\uFE0F\u20E3'));
+assertTrue(re.test('*\uFE0F\u20E3'));
+assertTrue(re.test('5\uFE0F\u20E3'));
+assertTrue(re.test('6\uFE0F\u20E3'));
+assertTrue(re.test('7\uFE0F\u20E3'));
+assertTrue(re.test('8\uFE0F\u20E3'));
+assertTrue(re.test('4\uFE0F\u20E3'));

test/mjsunit/harmony/regexp-property-emoji-tag-sequence-generated.js

+// Copyright 2018 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-regexp-sequence
+
+const re = /\p{Emoji_Tag_Sequence}/u;
+
+assertTrue(re.test('\u{1F3F4}\u{E0067}\u{E0062}\u{E0065}\u{E006E}\u{E0067}\u{E007F}'));
+assertTrue(re.test('\u{1F3F4}\u{E0067}\u{E0062}\u{E0073}\u{E0063}\u{E0074}\u{E007F}'));
+assertTrue(re.test('\u{1F3F4}\u{E0067}\u{E0062}\u{E0077}\u{E006C}\u{E0073}\u{E007F}'));

test/mjsunit/harmony/regexp-property-invalid.js

@@ -34,3 +34,5 @@ assertThrows("/\\p{In}/u");
 assertThrows("/\\pI/u");
 assertThrows("/\\p{I}/u");
 assertThrows("/\\p{CJK}/u");
+
+assertThrows("/\\p{}/u");

test/mjsunit/harmony/regexp-property-sequence.js

+// Copyright 2018 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-regexp-sequence
+
+// Normal usage.
+assertDoesNotThrow("/\\p{Emoji_Flag_Sequence}/u");
+assertTrue(/\p{Emoji_Flag_Sequence}/u.test("\u{1F1E9}\u{1F1EA}"));
+
+assertDoesNotThrow("/\\p{Emoji_Keycap_Sequence}/u");
+assertTrue(/\p{Emoji_Keycap_Sequence}/u.test("\u0023\ufe0f\u20e3"));
+
+assertDoesNotThrow("/\\p{Emoji_Keycap_Sequence}/u");
+assertFalse(/\p{Emoji_Keycap_Sequence}/u.test("\u0022\ufe0f\u20e3"));
+
+assertDoesNotThrow("/\\p{Emoji_Modifier_Sequence}/u");
+assertTrue(/\p{Emoji_Modifier_Sequence}/u.test("\u26f9\u{1f3ff}"));
+
+assertDoesNotThrow("/\\p{Emoji_ZWJ_Sequence}/u");
+assertTrue(/\p{Emoji_ZWJ_Sequence}/u.test("\u{1F468}\u{200D}\u{1F467}"));
+
+// Without unicode flag.
+assertDoesNotThrow("/\\p{Emoji_Flag_Sequence}/");
+assertFalse(/\p{Emoji_Flag_Sequence}/.test("\u{1F1E9}\u{1F1EA}"));
+assertTrue(/\p{Emoji_Flag_Sequence}/.test("\\p{Emoji_Flag_Sequence}"));
+
+// Negated and/or inside a character class.
+assertThrows("/\\P{Emoji_Flag_Sequence}/u");
+assertThrows("/[\\p{Emoji_Flag_Sequence}]/u");
+assertThrows("/[\\P{Emoji_Flag_Sequence}]/u");
+assertThrows("/[\\w\\p{Emoji_Flag_Sequence}]/u");
+assertThrows("/[\\w\\P{Emoji_Flag_Sequence}]/u");
+
+// Two regional indicators, but not a country.
+assertFalse(/\p{Emoji_Flag_Sequence}/u.test("\u{1F1E6}\u{1F1E6}"));
+
+// ZWJ sequence as in two ZWJ elements joined by a ZWJ, but not in the list.
+assertFalse(/\p{Emoji_ZWJ_Sequence}/u.test("\u{1F467}\u{200D}\u{1F468}"));
+
+// More complex regexp
+assertEquals(
+    ["country flag: \u{1F1E6}\u{1F1F9}"],
+    /Country Flag: \p{Emoji_Flag_Sequence}/iu.exec(
+        "this is an example of a country flag: \u{1F1E6}\u{1F1F9} is Austra"));
+assertEquals(
+    ["country flag: \u{1F1E6}\u{1F1F9}", "\u{1F1E6}\u{1F1F9}"],
+    /Country Flag: (\p{Emoji_Flag_Sequence})/iu.exec(
+        "this is an example of a country flag: \u{1F1E6}\u{1F1F9} is Austra"));
+assertEquals(
+    ["country flag: \u{1F1E6}\u{1F1F9}"],
+    /Country Flag: ..(?<=\p{Emoji_Flag_Sequence})/iu.exec(
+        "this is an example of a country flag: \u{1F1E6}\u{1F1F9} is Austra"));
+assertEquals(
+    ["flag: \u{1F1E6}\u{1F1F9}", "\u{1F1E6}\u{1F1F9}"],
+    /Flag: ..(?<=(\p{Emoji_Flag_Sequence})|\p{Emoji_Keycap_Sequence})/iu.exec(
+        "this is an example of a country flag: \u{1F1E6}\u{1F1F9} is Austra"));