bool
Reader::decodeUnicodeCodePoint( Token &token,
Location ¤t,
Location end,
unsigned int &unicode )
{
if ( !decodeUnicodeEscapeSequence( token, current, end, unicode ) )
return false;
if (unicode >= 0xD800 && unicode <= 0xDBFF)
{
// surrogate pairs
if (end - current < 6)
return addError( "additional six characters expected to parse unicode surrogate pair.", token, current );
unsigned int surrogatePair;
if (*(current++) == '\\' && *(current++)== 'u')
{
if (decodeUnicodeEscapeSequence( token, current, end, surrogatePair ))
{
unicode = 0x10000 + ((unicode & 0x3FF) << 10) + (surrogatePair & 0x3FF);
}
else
return false;
}
else
return addError( "expecting another \\u token to begin the second half of a unicode surrogate pair", token, current );
}
return true;
}
bool
Reader::decodeString (Token &token, std::string &decoded)
{
Location current = token.start_ + 1; // skip '"'
Location end = token.end_ - 1; // do not include '"'
decoded.reserve (long (end - current));
while (current != end)
{
char c = *current++;
if (expect_false (c == '"'))
break;
else if (expect_false (c == '\\'))
{
if (expect_false (current == end))
return addError ("Empty escape sequence in string", token, current);
char escape = *current++;
switch (escape)
{
case '"':
case '/':
case '\\': decoded += escape; break;
case 'b': decoded += '\010'; break;
case 't': decoded += '\011'; break;
case 'n': decoded += '\012'; break;
case 'f': decoded += '\014'; break;
case 'r': decoded += '\015'; break;
case 'u':
{
unsigned unicode;
if (!decodeUnicodeEscapeSequence (token, current, end, unicode))
return false;
// @todo encode unicode as utf8.
// @todo remember to alter the writer too.
}
break;
default:
return addError ("Bad escape sequence in string", token, current);
}
}
else
{
decoded += c;
}
}
return true;
}
bool
Reader::decodeString( Token &token, std::string &decoded )
{
decoded.reserve( token.end_ - token.start_ - 2 );
Location current = token.start_ + 1; // skip '"'
Location end = token.end_ - 1; // do not include '"'
while ( current != end )
{
Char c = *current++;
if ( c == '"' )
break;
else if ( c == '\\' )
{
if ( current == end )
return addError( "Empty escape sequence in string", token, current );
Char escape = *current++;
switch ( escape )
{
case '"': decoded += '"'; break;
case '/': decoded += '/'; break;
case '\\': decoded += '\\'; break;
case 'b': decoded += '\b'; break;
case 'f': decoded += '\f'; break;
case 'n': decoded += '\n'; break;
case 'r': decoded += '\r'; break;
case 't': decoded += '\t'; break;
case 'u':
{
unsigned int unicode;
if ( !decodeUnicodeEscapeSequence( token, current, end, unicode ) )
return false;
// @todo encode unicode as utf8.
// @todo remember to alter the writer too.
}
break;
default:
return addError( "Bad escape sequence in string", token, current );
}
}
else
{
decoded += c;
}
}
return true;
}