// http://www.w3.org/TR/css3-syntax/#consume-a-name
CSSParserString CSSTokenizer::consumeName()
{
// Names without escapes get handled without allocations
for (unsigned size = 0; ; ++size) {
UChar cc = m_input.peekWithoutReplacement(size);
if (cc == '\0' || cc == '\\')
break;
if (!isNameChar(cc)) {
unsigned startOffset = m_input.offset();
m_input.advance(size);
return m_input.rangeAsCSSParserString(startOffset, size);
}
}
StringBuilder result;
while (true) {
UChar cc = consume();
if (isNameChar(cc)) {
result.append(cc);
continue;
}
if (twoCharsAreValidEscape(cc, m_input.nextInputChar())) {
result.append(consumeEscape());
continue;
}
reconsume(cc);
return registerString(result.toString());
}
}
CSSParserToken CSSTokenizer::reverseSolidus(UChar cc)
{
if (twoCharsAreValidEscape(cc, m_input.nextInputChar())) {
reconsume(cc);
return consumeIdentLikeToken();
}
return CSSParserToken(DelimiterToken, cc);
}
// http://dev.w3.org/csswg/css-syntax/#consume-the-remnants-of-a-bad-url
void CSSTokenizer::consumeBadUrlRemnants()
{
while (true) {
UChar cc = consume();
if (cc == ')' || cc == kEndOfFileMarker)
return;
if (twoCharsAreValidEscape(cc, m_input.nextInputChar()))
consumeEscape();
}
}
CSSParserToken CSSTokenizer::hash(UChar cc)
{
UChar nextChar = m_input.nextInputChar();
if (isNameChar(nextChar) || twoCharsAreValidEscape(nextChar, m_input.peek(1))) {
HashTokenType type = nextCharsAreIdentifier() ? HashTokenId : HashTokenUnrestricted;
return CSSParserToken(type, consumeName());
}
return CSSParserToken(DelimiterToken, cc);
}
// http://dev.w3.org/csswg/css-syntax/#would-start-an-identifier
bool CSSTokenizer::nextCharsAreIdentifier(UChar first)
{
UChar second = m_input.nextInputChar();
if (isNameStart(first) || twoCharsAreValidEscape(first, second))
return true;
if (first == '-')
return isNameStart(second) || second == '-' || nextTwoCharsAreValidEscape();
return false;
}
// http://www.w3.org/TR/css3-syntax/#would-start-an-identifier
bool MediaQueryTokenizer::nextCharsAreIdentifier(UChar first)
{
UChar second = m_input.nextInputChar();
if (isNameStart(first) || twoCharsAreValidEscape(first, second))
return true;
if (first == '-') {
if (isNameStart(m_input.nextInputChar()))
return true;
return nextTwoCharsAreValidEscape();
}
return false;
}
// http://dev.w3.org/csswg/css-syntax/#consume-url-token
CSSParserToken CSSTokenizer::consumeUrlToken()
{
consumeUntilNonWhitespace();
// URL tokens without escapes get handled without allocations
for (unsigned size = 0; ; size++) {
UChar cc = m_input.peekWithoutReplacement(size);
if (cc == ')') {
unsigned startOffset = m_input.offset();
m_input.advance(size + 1);
return CSSParserToken(UrlToken, m_input.rangeAsCSSParserString(startOffset, size));
}
if (cc <= ' ' || cc == '\\' || cc == '"' || cc == '\'' || cc == '(' || cc == '\x7f')
break;
}
StringBuilder result;
while (true) {
UChar cc = consume();
if (cc == ')' || cc == kEndOfFileMarker)
return CSSParserToken(UrlToken, registerString(result.toString()));
if (isHTMLSpace(cc)) {
consumeUntilNonWhitespace();
if (consumeIfNext(')') || m_input.nextInputChar() == kEndOfFileMarker)
return CSSParserToken(UrlToken, registerString(result.toString()));
break;
}
if (cc == '"' || cc == '\'' || cc == '(' || isNonPrintableCodePoint(cc))
break;
if (cc == '\\') {
if (twoCharsAreValidEscape(cc, m_input.nextInputChar())) {
result.append(consumeEscape());
continue;
}
break;
}
result.append(cc);
}
consumeBadUrlRemnants();
return CSSParserToken(BadUrlToken);
}
// http://www.w3.org/TR/css3-syntax/#consume-a-name
String MediaQueryTokenizer::consumeName()
{
// FIXME: Is this as efficient as it can be?
// The possibility of escape chars mandates a copy AFAICT.
StringBuilder result;
while (true) {
UChar cc = consume();
if (isNameChar(cc)) {
result.append(cc);
continue;
}
if (twoCharsAreValidEscape(cc, m_input.nextInputChar())) {
result.append(consumeEscape());
continue;
}
reconsume(cc);
return result.toString();
}
}
// http://dev.w3.org/csswg/css-syntax/#consume-url-token
CSSParserToken CSSTokenizer::consumeUrlToken()
{
consumeUntilNonWhitespace();
StringBuilder result;
while (true) {
UChar cc = consume();
if (cc == ')' || cc == kEndOfFileMarker) {
// The "reconsume" here deviates from the spec, but is required to avoid consuming past the EOF
if (cc == kEndOfFileMarker)
reconsume(cc);
return CSSParserToken(UrlToken, result.toString());
}
if (isHTMLSpace(cc)) {
consumeUntilNonWhitespace();
if (consumeIfNext(')') || m_input.nextInputChar() == kEndOfFileMarker)
return CSSParserToken(UrlToken, result.toString());
break;
}
if (cc == '"' || cc == '\'' || cc == '(' || isNonPrintableCodePoint(cc))
break;
if (cc == '\\') {
if (twoCharsAreValidEscape(cc, m_input.nextInputChar())) {
result.append(consumeEscape());
continue;
}
break;
}
result.append(cc);
}
consumeBadUrlRemnants();
return CSSParserToken(BadUrlToken);
}
bool CSSTokenizer::nextTwoCharsAreValidEscape()
{
return twoCharsAreValidEscape(m_input.nextInputChar(), m_input.peek(1));
}
bool MediaQueryTokenizer::nextTwoCharsAreValidEscape()
{
if (m_input.leftChars() < 1)
return false;
return twoCharsAreValidEscape(m_input.nextInputChar(), m_input.peek(1));
}
