[regexp] explicitly whitelist allowed binary property classes.

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"
@@ skipping 1184 matching lines @@
                 zone);
   } else if (NameEquals(name, "Assigned")) {
     return LookupPropertyValueName(UCHAR_GENERAL_CATEGORY, "Unassigned",
                                    !negate, result, zone);
   } else {
     return false;
   }
   return true;
 }
 
+// Explicitly whitelist supported binary properties. The spec forbids supporting
+// properties outside of this set to ensure interoperability.
+bool IsSupportedBinaryProperty(UProperty property) {
+  switch (property) {
+    case UCHAR_ALPHABETIC:

[mathias, 2017/04/18 08:29:38]

Maybe add a comment here explaining that Any and ASCII are implemented
separately? Useful when cross-referencing with the list in the spec.

+    case UCHAR_ASCII_HEX_DIGIT:

[mathias, 2017/04/18 08:29:38]

Same here with `Assigned`.

+    case UCHAR_BIDI_CONTROL:
+    case UCHAR_BIDI_MIRRORED:
+    case UCHAR_CASE_IGNORABLE:
+    case UCHAR_CASED:
+    case UCHAR_CHANGES_WHEN_CASEFOLDED:
+    case UCHAR_CHANGES_WHEN_CASEMAPPED:
+    case UCHAR_CHANGES_WHEN_LOWERCASED:
+    case UCHAR_CHANGES_WHEN_NFKC_CASEFOLDED:
+    case UCHAR_CHANGES_WHEN_TITLECASED:
+    case UCHAR_CHANGES_WHEN_UPPERCASED:
+    case UCHAR_DASH:
+    case UCHAR_DEFAULT_IGNORABLE_CODE_POINT:
+    case UCHAR_DEPRECATED:
+    case UCHAR_DIACRITIC:
+    case UCHAR_EMOJI:
+    // TODO(yangguo): Uncomment this once we upgrade to ICU 60.
+    //                See https://ssl.icu-project.org/trac/ticket/13062
+    // case UCHAR_EMOJI_COMPONENT:
+    case UCHAR_EMOJI_MODIFIER_BASE:
+    case UCHAR_EMOJI_MODIFIER:
+    case UCHAR_EMOJI_PRESENTATION:
+    case UCHAR_EXTENDER:
+    case UCHAR_GRAPHEME_BASE:
+    case UCHAR_GRAPHEME_EXTEND:
+    case UCHAR_HEX_DIGIT:
+    case UCHAR_ID_CONTINUE:
+    case UCHAR_ID_START:
+    case UCHAR_IDEOGRAPHIC:
+    case UCHAR_IDS_BINARY_OPERATOR:
+    case UCHAR_IDS_TRINARY_OPERATOR:
+    case UCHAR_JOIN_CONTROL:
+    case UCHAR_LOGICAL_ORDER_EXCEPTION:
+    case UCHAR_LOWERCASE:
+    case UCHAR_MATH:
+    case UCHAR_NONCHARACTER_CODE_POINT:
+    case UCHAR_PATTERN_SYNTAX:
+    case UCHAR_PATTERN_WHITE_SPACE:
+    case UCHAR_QUOTATION_MARK:
+    case UCHAR_RADICAL:
+    case UCHAR_S_TERM:
+    case UCHAR_SOFT_DOTTED:
+    case UCHAR_TERMINAL_PUNCTUATION:
+    case UCHAR_UNIFIED_IDEOGRAPH:
+    case UCHAR_UPPERCASE:
+    case UCHAR_VARIATION_SELECTOR:
+    case UCHAR_WHITE_SPACE:
+    case UCHAR_XID_CONTINUE:
+    case UCHAR_XID_START:
+      return true;
+    default:
+      break;
+  }
+  return false;
+}
+
 }  // anonymous namespace
 
 bool RegExpParser::ParsePropertyClass(ZoneList<CharacterRange>* result,
                                       bool negate) {
   // Parse the property class as follows:
   // - In \p{name}, 'name' is interpreted
   //   - either as a general category property value name.
   //   - or as a binary property name.
   // - In \p{name=value}, 'name' is interpreted as an enumerated property name,
   //   and 'value' is interpreted as one of the available property value names.
@@ skipping 26 matching lines @@
     if (LookupPropertyValueName(UCHAR_GENERAL_CATEGORY_MASK, name, negate,
                                 result, zone())) {
       return true;
     }
     // Interpret "Any", "ASCII", and "Assigned".
     if (LookupSpecialPropertyValueName(name, result, negate, zone())) {
       return true;
     }
     // Then attempt to interpret as binary property name with value name 'Y'.
     UProperty property = u_getPropertyEnum(name);
-    if (property < UCHAR_BINARY_START) return false;
-    if (property >= UCHAR_BINARY_LIMIT) return false;
+    if (!IsSupportedBinaryProperty(property)) return false;
     if (!IsExactPropertyAlias(name, property)) return false;
     return LookupPropertyValueName(property, negate ? "N" : "Y", false, result,
                                    zone());
   } else {
     // Both property name and value name are specified. Attempt to interpret
     // the property name as enumerated property.
     const char* property_name = first_part.ToConstVector().start();
     const char* value_name = second_part.ToConstVector().start();
     UProperty property = u_getPropertyEnum(property_name);
     if (!IsExactPropertyAlias(property_name, property)) return false;
@@ skipping 608 matching lines @@
     return false;
   }
   terms_.Add(new (zone()) RegExpQuantifier(min, max, quantifier_type, atom),
              zone());
   LAST(ADD_TERM);
   return true;
 }
 
 }  // namespace internal
 }  // namespace v8