Implement ES6 unicode escapes.

src/scanner.cc

 // Copyright 2011 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.
 
 // Features shared by parsing and pre-parsing scanners.
 
 #include <stdint.h>
 
 #include <cmath>
 
@@ skipping 53 matching lines @@
       return -1;
     }
     x = x * 16 + d;
     Advance();
   }
 
   return x;
 }
 
 
+uc32 Scanner::ScanUnlimitedLengthHexNumber(int max_value) {
+  uc32 x = 0;
+  int d = HexValue(c0_);
+  if (d < 0) {
+    return -1;
+  }
+  while (d >= 0) {
+    x = x * 16 + d;
+    if (x > max_value) return -1;
+    Advance();
+    d = HexValue(c0_);
+  }
+  return x;
+}
+
+
 // Ensure that tokens can be stored in a byte.
 STATIC_ASSERT(Token::NUM_TOKENS <= 0x100);
 
 // Table of one-character tokens, by character (0x00..0x7f only).
 static const byte one_char_tokens[] = {
   Token::ILLEGAL,
   Token::ILLEGAL,
   Token::ILLEGAL,
   Token::ILLEGAL,
   Token::ILLEGAL,
@@ skipping 582 matching lines @@
     // whether there was a line terminator in the part we skip.
     has_line_terminator_before_next_ = false;
     has_multiline_comment_before_next_ = false;
   }
   Scan();
 }
 
 
 bool Scanner::ScanEscape() {
   uc32 c = c0_;
+  uc32 c2 = -1;
   Advance();
 
   // Skip escaped newlines.
   if (c0_ >= 0 && unicode_cache_->IsLineTerminator(c)) {
     // Allow CR+LF newlines in multiline string literals.
     if (IsCarriageReturn(c) && IsLineFeed(c0_)) Advance();
     // Allow LF+CR newlines in multiline string literals.
     if (IsLineFeed(c) && IsCarriageReturn(c0_)) Advance();
     return true;
   }
 
   switch (c) {
     case '\'':  // fall through
     case '"' :  // fall through
     case '\\': break;
     case 'b' : c = '\b'; break;
     case 'f' : c = '\f'; break;
     case 'n' : c = '\n'; break;
     case 'r' : c = '\r'; break;
     case 't' : c = '\t'; break;
     case 'u' : {
-      c = ScanHexNumber(4);
+      c = ScanUnicodeEscape();
       if (c < 0) return false;
       break;
     }
     case 'v' : c = '\v'; break;
     case 'x' : {
       c = ScanHexNumber(2);
       if (c < 0) return false;
       break;
     }
     case '0' :  // fall through
     case '1' :  // fall through
     case '2' :  // fall through
     case '3' :  // fall through
     case '4' :  // fall through
     case '5' :  // fall through
     case '6' :  // fall through
     case '7' : c = ScanOctalEscape(c, 2); break;
   }
 
   // According to ECMA-262, section 7.8.4, characters not covered by the
   // above cases should be illegal, but they are commonly handled as
   // non-escaped characters by JS VMs.
   AddLiteralChar(c);
+  if (c2 >= 0) {
+    AddLiteralChar(c2);
+  }
   return true;
 }
 
 
 // Octal escapes of the forms '\0xx' and '\xxx' are not a part of
 // ECMA-262. Other JS VMs support them.
 uc32 Scanner::ScanOctalEscape(uc32 c, int length) {
   uc32 x = c - '0';
   int i = 0;
   for (; i < length; i++) {
@@ skipping 149 matching lines @@
   literal.Complete();
 
   return Token::NUMBER;
 }
 
 
 uc32 Scanner::ScanIdentifierUnicodeEscape() {
   Advance();
   if (c0_ != 'u') return -1;
   Advance();
+  return ScanUnicodeEscape();
+}
+
+
+uc32 Scanner::ScanUnicodeEscape() {
+  // Accept both \uxxxx and \u{xxxxxx}. In the latter case, the number of hex
+  // digits between { } is arbitrary. \ and u have already been read.
+  if (c0_ == '{') {
+    Advance();
+    uc32 cp = ScanUnlimitedLengthHexNumber(0x10ffff);
+    if (cp < 0) {
+      return -1;
+    }
+    if (c0_ != '}') {
+      return -1;
+    }
+    Advance();
+    return cp;
+  }
   return ScanHexNumber(4);
 }
 
 
 // ----------------------------------------------------------------------------
 // Keyword Matcher
 
 #define KEYWORDS(KEYWORD_GROUP, KEYWORD)                                     \
   KEYWORD_GROUP('b')                                                         \
   KEYWORD("break", Token::BREAK)                                             \
@@ skipping 426 matching lines @@
     }
     backing_store_.Add(static_cast<uint8_t>((one_byte_length >> 7) | 0x80u));
   }
   backing_store_.Add(static_cast<uint8_t>(one_byte_length & 0x7f));
 
   backing_store_.AddBlock(bytes);
   return backing_store_.EndSequence().start();
 }
 
 } }  // namespace v8::internal