CSSParserToken CSSTokenizer::consumeUnicodeRange()
{
ASSERT(isASCIIHexDigit(m_input.nextInputChar()) || m_input.nextInputChar() == '?');
int lengthRemaining = 6;
UChar32 start = 0;
while (lengthRemaining && isASCIIHexDigit(m_input.nextInputChar())) {
start = start * 16 + toASCIIHexValue(consume());
--lengthRemaining;
}
UChar32 end = start;
if (lengthRemaining && consumeIfNext('?')) {
do {
start *= 16;
end = end * 16 + 0xF;
--lengthRemaining;
} while (lengthRemaining && consumeIfNext('?'));
} else if (m_input.nextInputChar() == '-' && isASCIIHexDigit(m_input.peek(1))) {
consume();
lengthRemaining = 6;
end = 0;
do {
end = end * 16 + toASCIIHexValue(consume());
--lengthRemaining;
} while (lengthRemaining && isASCIIHexDigit(m_input.nextInputChar()));
}
return CSSParserToken(UnicodeRangeToken, start, end);
}
static RGBA32 parseColorStringWithCrazyLegacyRules(const String& colorString)
{
// Per spec, only look at the first 128 digits of the string.
const size_t maxColorLength = 128;
// We'll pad the buffer with two extra 0s later, so reserve two more than the max.
Vector<char, maxColorLength+2> digitBuffer;
size_t i = 0;
// Skip a leading #.
if (colorString[0] == '#')
i = 1;
// Grab the first 128 characters, replacing non-hex characters with 0.
// Non-BMP characters are replaced with "00" due to them appearing as two "characters" in the String.
for (; i < colorString.length() && digitBuffer.size() < maxColorLength; i++) {
if (!isASCIIHexDigit(colorString[i]))
digitBuffer.append('0');
else
digitBuffer.append(colorString[i]);
}
if (!digitBuffer.size())
return Color::black;
// Pad the buffer out to at least the next multiple of three in size.
digitBuffer.append('0');
digitBuffer.append('0');
if (digitBuffer.size() < 6)
return makeRGB(toASCIIHexValue(digitBuffer[0]), toASCIIHexValue(digitBuffer[1]), toASCIIHexValue(digitBuffer[2]));
// Split the digits into three components, then search the last 8 digits of each component.
ASSERT(digitBuffer.size() >= 6);
size_t componentLength = digitBuffer.size() / 3;
size_t componentSearchWindowLength = min<size_t>(componentLength, 8);
size_t redIndex = componentLength - componentSearchWindowLength;
size_t greenIndex = componentLength * 2 - componentSearchWindowLength;
size_t blueIndex = componentLength * 3 - componentSearchWindowLength;
// Skip digits until one of them is non-zero, or we've only got two digits left in the component.
while (digitBuffer[redIndex] == '0' && digitBuffer[greenIndex] == '0' && digitBuffer[blueIndex] == '0' && (componentLength - redIndex) > 2) {
redIndex++;
greenIndex++;
blueIndex++;
}
ASSERT(redIndex + 1 < componentLength);
ASSERT(greenIndex >= componentLength);
ASSERT(greenIndex + 1 < componentLength * 2);
ASSERT(blueIndex >= componentLength * 2);
ASSERT_WITH_SECURITY_IMPLICATION(blueIndex + 1 < digitBuffer.size());
int redValue = toASCIIHexValue(digitBuffer[redIndex], digitBuffer[redIndex + 1]);
int greenValue = toASCIIHexValue(digitBuffer[greenIndex], digitBuffer[greenIndex + 1]);
int blueValue = toASCIIHexValue(digitBuffer[blueIndex], digitBuffer[blueIndex + 1]);
return makeRGB(redValue, greenValue, blueValue);
}
void quotedPrintableDecode(const char* data, size_t dataLength, Vector<char>& out)
{
out.clear();
if (!dataLength)
return;
for (size_t i = 0; i < dataLength; ++i) {
char currentCharacter = data[i];
if (currentCharacter != '=') {
out.append(currentCharacter);
continue;
}
// We are dealing with a '=xx' sequence.
if (dataLength - i < 3) {
// Unfinished = sequence, append as is.
out.append(currentCharacter);
continue;
}
char upperCharacter = data[++i];
char lowerCharacter = data[++i];
if (upperCharacter == '\r' && lowerCharacter == '\n')
continue;
if (!isASCIIHexDigit(upperCharacter) || !isASCIIHexDigit(lowerCharacter)) {
// Invalid sequence, = followed by non hex digits, just insert the characters as is.
out.append('=');
out.append(upperCharacter);
out.append(lowerCharacter);
continue;
}
out.append(static_cast<char>(toASCIIHexValue(upperCharacter, lowerCharacter)));
}
}
template <typename CharType> bool hexDigitsToHash(CharType* characters, NetworkCacheKey::HashType& hash)
{
for (unsigned i = 0; i < sizeof(hash); ++i) {
auto high = characters[2 * i];
auto low = characters[2 * i + 1];
if (!isASCIIHexDigit(high) || !isASCIIHexDigit(low))
return false;
hash[i] = toASCIIHexValue(high, low);
}
return true;
}
static inline bool parseHexColorInternal(const CharacterType* name, unsigned length, RGBA32& rgb)
{
if (length != 3 && length != 6)
return false;
unsigned value = 0;
for (unsigned i = 0; i < length; ++i) {
if (!isASCIIHexDigit(name[i]))
return false;
value <<= 4;
value |= toASCIIHexValue(name[i]);
}
if (length == 6) {
rgb = 0xFF000000 | value;
return true;
}
// #abc converts to #aabbcc
rgb = 0xFF000000
| (value & 0xF00) << 12 | (value & 0xF00) << 8
| (value & 0xF0) << 8 | (value & 0xF0) << 4
| (value & 0xF) << 4 | (value & 0xF);
return true;
}
// originally moved here from the CSS parser
bool Color::parseHexColor(const String& name, RGBA32& rgb)
{
unsigned length = name.length();
if (length != 3 && length != 6)
return false;
unsigned value = 0;
for (unsigned i = 0; i < length; ++i) {
if (!isASCIIHexDigit(name[i]))
return false;
value <<= 4;
value |= toASCIIHexValue(name[i]);
}
if (length == 6) {
rgb = 0xFF000000 | value;
return true;
}
// #abc converts to #aabbcc
rgb = 0xFF000000
| (value & 0xF00) << 12 | (value & 0xF00) << 8
| (value & 0xF0) << 8 | (value & 0xF0) << 4
| (value & 0xF) << 4 | (value & 0xF);
return true;
}
static bool parseUnicodeRange(const CharType* characters, unsigned length, UnicodeRange& range)
{
if (length < 2 || characters[0] != 'U' || characters[1] != '+')
return false;
// Parse the starting hex number (or its prefix).
unsigned startRange = 0;
unsigned startLength = 0;
const CharType* ptr = characters + 2;
const CharType* end = characters + length;
while (ptr < end) {
if (!isASCIIHexDigit(*ptr))
break;
++startLength;
if (startLength > 6)
return false;
startRange = (startRange << 4) | toASCIIHexValue(*ptr);
++ptr;
}
// Handle the case of ranges separated by "-" sign.
if (2 + startLength < length && *ptr == '-') {
if (!startLength)
return false;
// Parse the ending hex number (or its prefix).
unsigned endRange = 0;
unsigned endLength = 0;
++ptr;
while (ptr < end) {
if (!isASCIIHexDigit(*ptr))
break;
++endLength;
if (endLength > 6)
return false;
endRange = (endRange << 4) | toASCIIHexValue(*ptr);
++ptr;
}
if (!endLength)
return false;
range.first = startRange;
range.second = endRange;
return true;
}
// Handle the case of a number with some optional trailing question marks.
unsigned endRange = startRange;
while (ptr < end) {
if (*ptr != '?')
break;
++startLength;
if (startLength > 6)
return false;
startRange <<= 4;
endRange = (endRange << 4) | 0xF;
++ptr;
}
if (!startLength)
return false;
range.first = startRange;
range.second = endRange;
return true;
}
int Lexer::lex(void* p1, void* p2)
{
ASSERT(!m_error);
ASSERT(m_buffer8.isEmpty());
ASSERT(m_buffer16.isEmpty());
YYSTYPE* lvalp = static_cast<YYSTYPE*>(p1);
YYLTYPE* llocp = static_cast<YYLTYPE*>(p2);
int token = 0;
m_terminator = false;
start:
while (isWhiteSpace(m_current))
shift1();
int startOffset = currentOffset();
if (m_current == -1) {
if (!m_terminator && !m_delimited && !m_isReparsing) {
// automatic semicolon insertion if program incomplete
token = ';';
goto doneSemicolon;
}
return 0;
}
m_delimited = false;
switch (m_current) {
case '>':
if (m_next1 == '>' && m_next2 == '>') {
if (m_next3 == '=') {
shift4();
token = URSHIFTEQUAL;
break;
}
shift3();
token = URSHIFT;
break;
}
if (m_next1 == '>') {
if (m_next2 == '=') {
shift3();
token = RSHIFTEQUAL;
break;
}
shift2();
token = RSHIFT;
break;
}
if (m_next1 == '=') {
shift2();
token = GE;
break;
}
shift1();
token = '>';
break;
case '=':
if (m_next1 == '=') {
if (m_next2 == '=') {
shift3();
token = STREQ;
break;
}
shift2();
token = EQEQ;
break;
}
shift1();
token = '=';
break;
case '!':
if (m_next1 == '=') {
if (m_next2 == '=') {
shift3();
token = STRNEQ;
break;
}
shift2();
token = NE;
break;
}
shift1();
token = '!';
break;
case '<':
if (m_next1 == '!' && m_next2 == '-' && m_next3 == '-') {
// <!-- marks the beginning of a line comment (for www usage)
shift4();
goto inSingleLineComment;
}
if (m_next1 == '<') {
if (m_next2 == '=') {
shift3();
token = LSHIFTEQUAL;
break;
}
shift2();
token = LSHIFT;
break;
}
if (m_next1 == '=') {
shift2();
token = LE;
break;
}
shift1();
token = '<';
break;
case '+':
if (m_next1 == '+') {
shift2();
if (m_terminator) {
token = AUTOPLUSPLUS;
break;
}
token = PLUSPLUS;
break;
}
if (m_next1 == '=') {
shift2();
token = PLUSEQUAL;
break;
}
shift1();
token = '+';
break;
case '-':
if (m_next1 == '-') {
if (m_atLineStart && m_next2 == '>') {
shift3();
goto inSingleLineComment;
}
shift2();
if (m_terminator) {
token = AUTOMINUSMINUS;
break;
}
token = MINUSMINUS;
break;
}
if (m_next1 == '=') {
shift2();
token = MINUSEQUAL;
break;
}
shift1();
token = '-';
break;
case '*':
if (m_next1 == '=') {
shift2();
token = MULTEQUAL;
break;
}
shift1();
token = '*';
break;
case '/':
if (m_next1 == '/') {
shift2();
goto inSingleLineComment;
}
if (m_next1 == '*')
goto inMultiLineComment;
if (m_next1 == '=') {
shift2();
token = DIVEQUAL;
break;
}
shift1();
token = '/';
break;
case '&':
if (m_next1 == '&') {
shift2();
token = AND;
break;
}
if (m_next1 == '=') {
shift2();
token = ANDEQUAL;
break;
}
shift1();
token = '&';
break;
case '^':
if (m_next1 == '=') {
shift2();
token = XOREQUAL;
break;
}
shift1();
token = '^';
break;
case '%':
if (m_next1 == '=') {
shift2();
token = MODEQUAL;
break;
}
shift1();
token = '%';
break;
case '|':
if (m_next1 == '=') {
shift2();
token = OREQUAL;
break;
}
if (m_next1 == '|') {
shift2();
token = OR;
break;
}
shift1();
token = '|';
break;
case '.':
if (isASCIIDigit(m_next1)) {
record8('.');
shift1();
goto inNumberAfterDecimalPoint;
}
token = '.';
shift1();
break;
case ',':
case '~':
case '?':
case ':':
case '(':
case ')':
case '[':
case ']':
token = m_current;
shift1();
break;
case ';':
shift1();
m_delimited = true;
token = ';';
break;
case '{':
lvalp->intValue = currentOffset();
shift1();
token = OPENBRACE;
break;
case '}':
lvalp->intValue = currentOffset();
shift1();
m_delimited = true;
token = CLOSEBRACE;
break;
case '\\':
goto startIdentifierWithBackslash;
case '0':
goto startNumberWithZeroDigit;
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
goto startNumber;
case '"':
case '\'':
goto startString;
default:
if (isIdentStart(m_current))
goto startIdentifierOrKeyword;
if (isLineTerminator(m_current)) {
shiftLineTerminator();
m_atLineStart = true;
m_terminator = true;
if (lastTokenWasRestrKeyword()) {
token = ';';
goto doneSemicolon;
}
goto start;
}
goto returnError;
}
m_atLineStart = false;
goto returnToken;
startString: {
int stringQuoteCharacter = m_current;
shift1();
const UChar* stringStart = currentCharacter();
while (m_current != stringQuoteCharacter) {
// Fast check for characters that require special handling.
// Catches -1, \n, \r, \, 0x2028, and 0x2029 as efficiently
// as possible, and lets through all common ASCII characters.
if (UNLIKELY(m_current == '\\') || UNLIKELY(((static_cast<unsigned>(m_current) - 0xE) & 0x2000))) {
m_buffer16.append(stringStart, currentCharacter() - stringStart);
goto inString;
}
shift1();
}
lvalp->ident = makeIdentifier(stringStart, currentCharacter() - stringStart);
shift1();
m_atLineStart = false;
m_delimited = false;
token = STRING;
goto returnToken;
inString:
while (m_current != stringQuoteCharacter) {
if (m_current == '\\')
goto inStringEscapeSequence;
if (UNLIKELY(isLineTerminator(m_current)))
goto returnError;
if (UNLIKELY(m_current == -1))
goto returnError;
record16(m_current);
shift1();
}
goto doneString;
inStringEscapeSequence:
shift1();
if (m_current == 'x') {
shift1();
if (isASCIIHexDigit(m_current) && isASCIIHexDigit(m_next1)) {
record16(convertHex(m_current, m_next1));
shift2();
goto inString;
}
record16('x');
if (m_current == stringQuoteCharacter)
goto doneString;
goto inString;
}
if (m_current == 'u') {
shift1();
if (isASCIIHexDigit(m_current) && isASCIIHexDigit(m_next1) && isASCIIHexDigit(m_next2) && isASCIIHexDigit(m_next3)) {
record16(convertUnicode(m_current, m_next1, m_next2, m_next3));
shift4();
goto inString;
}
if (m_current == stringQuoteCharacter) {
record16('u');
goto doneString;
}
goto returnError;
}
if (isASCIIOctalDigit(m_current)) {
if (m_current >= '0' && m_current <= '3' && isASCIIOctalDigit(m_next1) && isASCIIOctalDigit(m_next2)) {
record16((m_current - '0') * 64 + (m_next1 - '0') * 8 + m_next2 - '0');
shift3();
goto inString;
}
if (isASCIIOctalDigit(m_next1)) {
record16((m_current - '0') * 8 + m_next1 - '0');
shift2();
goto inString;
}
record16(m_current - '0');
shift1();
goto inString;
}
if (isLineTerminator(m_current)) {
shiftLineTerminator();
goto inString;
}
record16(singleEscape(m_current));
shift1();
goto inString;
}
startIdentifierWithBackslash:
shift1();
if (UNLIKELY(m_current != 'u'))
goto returnError;
shift1();
if (UNLIKELY(!isASCIIHexDigit(m_current) || !isASCIIHexDigit(m_next1) || !isASCIIHexDigit(m_next2) || !isASCIIHexDigit(m_next3)))
goto returnError;
token = convertUnicode(m_current, m_next1, m_next2, m_next3);
if (UNLIKELY(!isIdentStart(token)))
goto returnError;
goto inIdentifierAfterCharacterCheck;
startIdentifierOrKeyword: {
const UChar* identifierStart = currentCharacter();
shift1();
while (isIdentPart(m_current))
shift1();
if (LIKELY(m_current != '\\')) {
lvalp->ident = makeIdentifier(identifierStart, currentCharacter() - identifierStart);
goto doneIdentifierOrKeyword;
}
m_buffer16.append(identifierStart, currentCharacter() - identifierStart);
}
do {
shift1();
if (UNLIKELY(m_current != 'u'))
goto returnError;
shift1();
if (UNLIKELY(!isASCIIHexDigit(m_current) || !isASCIIHexDigit(m_next1) || !isASCIIHexDigit(m_next2) || !isASCIIHexDigit(m_next3)))
goto returnError;
token = convertUnicode(m_current, m_next1, m_next2, m_next3);
if (UNLIKELY(!isIdentPart(token)))
goto returnError;
inIdentifierAfterCharacterCheck:
record16(token);
shift4();
while (isIdentPart(m_current)) {
record16(m_current);
shift1();
}
} while (UNLIKELY(m_current == '\\'));
goto doneIdentifier;
inSingleLineComment:
while (!isLineTerminator(m_current)) {
if (UNLIKELY(m_current == -1))
return 0;
shift1();
}
shiftLineTerminator();
m_atLineStart = true;
m_terminator = true;
if (lastTokenWasRestrKeyword())
goto doneSemicolon;
goto start;
inMultiLineComment:
shift2();
while (m_current != '*' || m_next1 != '/') {
if (isLineTerminator(m_current))
shiftLineTerminator();
else {
shift1();
if (UNLIKELY(m_current == -1))
goto returnError;
}
}
shift2();
m_atLineStart = false;
goto start;
startNumberWithZeroDigit:
shift1();
if ((m_current | 0x20) == 'x' && isASCIIHexDigit(m_next1)) {
shift1();
goto inHex;
}
if (m_current == '.') {
record8('0');
record8('.');
shift1();
goto inNumberAfterDecimalPoint;
}
if ((m_current | 0x20) == 'e') {
record8('0');
record8('e');
shift1();
goto inExponentIndicator;
}
if (isASCIIOctalDigit(m_current))
goto inOctal;
if (isASCIIDigit(m_current))
goto startNumber;
lvalp->doubleValue = 0;
goto doneNumeric;
inNumberAfterDecimalPoint:
while (isASCIIDigit(m_current)) {
record8(m_current);
shift1();
}
if ((m_current | 0x20) == 'e') {
record8('e');
shift1();
goto inExponentIndicator;
}
goto doneNumber;
inExponentIndicator:
if (m_current == '+' || m_current == '-') {
record8(m_current);
shift1();
}
if (!isASCIIDigit(m_current))
goto returnError;
do {
record8(m_current);
shift1();
} while (isASCIIDigit(m_current));
goto doneNumber;
inOctal: {
do {
record8(m_current);
shift1();
} while (isASCIIOctalDigit(m_current));
if (isASCIIDigit(m_current))
goto startNumber;
double dval = 0;
const char* end = m_buffer8.end();
for (const char* p = m_buffer8.data(); p < end; ++p) {
dval *= 8;
dval += *p - '0';
}
if (dval >= mantissaOverflowLowerBound)
dval = parseIntOverflow(m_buffer8.data(), end - m_buffer8.data(), 8);
m_buffer8.resize(0);
lvalp->doubleValue = dval;
goto doneNumeric;
}
inHex: {
do {
record8(m_current);
shift1();
} while (isASCIIHexDigit(m_current));
double dval = 0;
const char* end = m_buffer8.end();
for (const char* p = m_buffer8.data(); p < end; ++p) {
dval *= 16;
dval += toASCIIHexValue(*p);
}
if (dval >= mantissaOverflowLowerBound)
dval = parseIntOverflow(m_buffer8.data(), end - m_buffer8.data(), 16);
m_buffer8.resize(0);
lvalp->doubleValue = dval;
goto doneNumeric;
}
startNumber:
record8(m_current);
shift1();
while (isASCIIDigit(m_current)) {
record8(m_current);
shift1();
}
if (m_current == '.') {
record8('.');
shift1();
goto inNumberAfterDecimalPoint;
}
if ((m_current | 0x20) == 'e') {
record8('e');
shift1();
goto inExponentIndicator;
}
// Fall through into doneNumber.
doneNumber:
// Null-terminate string for strtod.
m_buffer8.append('\0');
lvalp->doubleValue = WTF::strtod(m_buffer8.data(), 0);
m_buffer8.resize(0);
// Fall through into doneNumeric.
doneNumeric:
// No identifiers allowed directly after numeric literal, e.g. "3in" is bad.
if (UNLIKELY(isIdentStart(m_current)))
goto returnError;
m_atLineStart = false;
m_delimited = false;
token = NUMBER;
goto returnToken;
doneSemicolon:
token = ';';
m_delimited = true;
goto returnToken;
doneIdentifier:
m_atLineStart = false;
m_delimited = false;
lvalp->ident = makeIdentifier(m_buffer16.data(), m_buffer16.size());
m_buffer16.resize(0);
token = IDENT;
goto returnToken;
doneIdentifierOrKeyword: {
m_atLineStart = false;
m_delimited = false;
m_buffer16.resize(0);
const HashEntry* entry = m_keywordTable.entry(m_globalData, *lvalp->ident);
token = entry ? entry->lexerValue() : IDENT;
goto returnToken;
}
doneString:
// Atomize constant strings in case they're later used in property lookup.
shift1();
m_atLineStart = false;
m_delimited = false;
lvalp->ident = makeIdentifier(m_buffer16.data(), m_buffer16.size());
m_buffer16.resize(0);
token = STRING;
// Fall through into returnToken.
returnToken: {
int lineNumber = m_lineNumber;
llocp->first_line = lineNumber;
llocp->last_line = lineNumber;
llocp->first_column = startOffset;
llocp->last_column = currentOffset();
m_lastToken = token;
return token;
}
returnError:
m_error = true;
return -1;
}
template<typename CharType> inline int toASCIIHexValue(CharType upperValue, CharType lowerValue)
{
DCHECK(isASCIIHexDigit(upperValue) && isASCIIHexDigit(lowerValue));
return ((toASCIIHexValue(upperValue) << 4) & 0xF0) | toASCIIHexValue(lowerValue);
}
