void
nsCSSScanner::ParseAndAppendEscape(nsString& aOutput)
{
PRInt32 ch = Peek();
if (ch < 0) {
aOutput.Append(CSS_ESCAPE);
return;
}
if (IsHexDigit(ch)) {
PRInt32 rv = 0;
int i;
for (i = 0; i < 6; i++) { // up to six digits
ch = Read();
if (ch < 0) {
// Whoops: error or premature eof
break;
}
if (!IsHexDigit(ch) && !IsWhitespace(ch)) {
Pushback(ch);
break;
} else if (IsHexDigit(ch)) {
rv = rv * 16 + HexDigitValue(ch);
} else {
NS_ASSERTION(IsWhitespace(ch), "bad control flow");
// single space ends escape
break;
}
}
if (6 == i) { // look for trailing whitespace and eat it
ch = Peek();
if (IsWhitespace(ch)) {
(void) Read();
}
}
NS_ASSERTION(rv >= 0, "How did rv become negative?");
// "[at most six hexadecimal digits following a backslash] stand
// for the ISO 10646 character with that number, which must not be
// zero. (It is undefined in CSS 2.1 what happens if a style sheet
// does contain a character with Unicode codepoint zero.)"
// -- CSS2.1 section 4.1.3
//
// Silently deleting \0 opens a content-filtration loophole (see
// bug 228856), so what we do instead is pretend the "cancels the
// meaning of special characters" rule applied.
if (rv > 0) {
AppendUCS4ToUTF16(ENSURE_VALID_CHAR(rv), aOutput);
} else {
while (i--)
aOutput.Append('0');
if (IsWhitespace(ch))
Pushback(ch);
}
return;
}
// "Any character except a hexidecimal digit can be escaped to
// remove its special meaning by putting a backslash in front"
// -- CSS1 spec section 7.1
ch = Read(); // Consume the escaped character
if ((ch > 0) && (ch != '\n')) {
aOutput.Append(ch);
}
}
/**
* If there is a valid escape sequence starting at the current read
* position, consume it, decode it, append the result to |aOutput|,
* and return true. Otherwise, consume nothing, leave |aOutput|
* unmodified, and return false. If |aInString| is true, accept the
* additional form of escape sequence allowed within string-like tokens.
*/
bool
nsCSSScanner::GatherEscape(nsString& aOutput, bool aInString)
{
MOZ_ASSERT(Peek() == '\\', "should not have been called");
int32_t ch = Peek(1);
if (ch < 0) {
// If we are in a string (or a url() containing a string), we want to drop
// the backslash on the floor. Otherwise, we want to treat it as a U+FFFD
// character.
Advance();
if (aInString) {
SetEOFCharacters(eEOFCharacters_DropBackslash);
} else {
aOutput.Append(UCS2_REPLACEMENT_CHAR);
SetEOFCharacters(eEOFCharacters_ReplacementChar);
}
return true;
}
if (IsVertSpace(ch)) {
if (aInString) {
// In strings (and in url() containing a string), escaped
// newlines are completely removed, to allow splitting over
// multiple lines.
Advance();
AdvanceLine();
return true;
}
// Outside of strings, backslash followed by a newline is not an escape.
return false;
}
if (!IsHexDigit(ch)) {
// "Any character (except a hexadecimal digit, linefeed, carriage
// return, or form feed) can be escaped with a backslash to remove
// its special meaning." -- CSS2.1 section 4.1.3
Advance(2);
if (ch == 0) {
aOutput.Append(UCS2_REPLACEMENT_CHAR);
} else {
aOutput.Append(ch);
}
return true;
}
// "[at most six hexadecimal digits following a backslash] stand
// for the ISO 10646 character with that number, which must not be
// zero. (It is undefined in CSS 2.1 what happens if a style sheet
// does contain a character with Unicode codepoint zero.)"
// -- CSS2.1 section 4.1.3
// At this point we know we have \ followed by at least one
// hexadecimal digit, therefore the escape sequence is valid and we
// can go ahead and consume the backslash.
Advance();
uint32_t val = 0;
int i = 0;
do {
val = val * 16 + HexDigitValue(ch);
i++;
Advance();
ch = Peek();
} while (i < 6 && IsHexDigit(ch));
// "Interpret the hex digits as a hexadecimal number. If this number is zero,
// or is greater than the maximum allowed codepoint, return U+FFFD
// REPLACEMENT CHARACTER" -- CSS Syntax Level 3
if (MOZ_UNLIKELY(val == 0)) {
aOutput.Append(UCS2_REPLACEMENT_CHAR);
} else {
AppendUCS4ToUTF16(ENSURE_VALID_CHAR(val), aOutput);
}
// Consume exactly one whitespace character after a
// hexadecimal escape sequence.
if (IsVertSpace(ch)) {
AdvanceLine();
} else if (IsHorzSpace(ch)) {
Advance();
}
return true;
}
void
nsCSSScanner::ParseAndAppendEscape(nsresult& aErrorCode, nsString& aOutput)
{
PRUint8* lexTable = gLexTable;
PRInt32 ch = Peek(aErrorCode);
if (ch < 0) {
aOutput.Append(CSS_ESCAPE);
return;
}
if ((ch <= 255) && ((lexTable[ch] & IS_HEX_DIGIT) != 0)) {
PRInt32 rv = 0;
int i;
for (i = 0; i < 6; i++) { // up to six digits
ch = Read(aErrorCode);
if (ch < 0) {
// Whoops: error or premature eof
break;
}
if (ch >= 256 || (lexTable[ch] & (IS_HEX_DIGIT | IS_WHITESPACE)) == 0) {
Unread();
break;
} else if ((lexTable[ch] & IS_HEX_DIGIT) != 0) {
if ((lexTable[ch] & IS_DIGIT) != 0) {
rv = rv * 16 + (ch - '0');
} else {
// Note: c&7 just keeps the low three bits which causes
// upper and lower case alphabetics to both yield their
// "relative to 10" value for computing the hex value.
rv = rv * 16 + ((ch & 0x7) + 9);
}
} else {
NS_ASSERTION((lexTable[ch] & IS_WHITESPACE) != 0, "bad control flow");
// single space ends escape
if (ch == '\r' && Peek(aErrorCode) == '\n') {
// if CR/LF, eat LF too
Read(aErrorCode);
}
break;
}
}
if (6 == i) { // look for trailing whitespace and eat it
ch = Peek(aErrorCode);
if ((0 <= ch) && (ch <= 255) &&
((lexTable[ch] & IS_WHITESPACE) != 0)) {
(void) Read(aErrorCode);
// special case: if trailing whitespace is CR/LF, eat both chars.
if (ch == '\r' && Peek(aErrorCode) == '\n') {
(void) Read(aErrorCode);
// if we hit the "\0" special case below, we'll push back
// only the '\r', but that's okay, because '\r' by itself
// is still a newline.
}
}
}
NS_ASSERTION(rv >= 0, "How did rv become negative?");
// "[at most six hexadecimal digits following a backslash] stand
// for the ISO 10646 character with that number, which must not be
// zero. (It is undefined in CSS 2.1 what happens if a style sheet
// does contain a character with Unicode codepoint zero.)"
// -- CSS2.1 section 4.1.3
//
// Silently deleting \0 opens a content-filtration loophole (see
// bug 228856), so what we do instead is pretend the "cancels the
// meaning of special characters" rule applied.
if (rv > 0) {
AppendUCS4ToUTF16(ENSURE_VALID_CHAR(rv), aOutput);
} else {
while (i--)
aOutput.Append('0');
if ((0 <= ch) && (ch <= 255) && ((lexTable[ch] & IS_WHITESPACE) != 0))
Pushback(ch);
}
return;
} else {
// "Any character except a hexidecimal digit can be escaped to
// remove its special meaning by putting a backslash in front"
// -- CSS1 spec section 7.1
if (!EatNewline(aErrorCode)) { // skip escaped newline
(void) Read(aErrorCode);
if (ch > 0) {
aOutput.Append(ch);
}
}
return;
}
}
void
nsCSSScanner::ParseAndAppendEscape(nsresult& aErrorCode, nsString& aOutput)
{
PRInt32 ch = Peek(aErrorCode);
if (ch < 0) {
aOutput.Append(CSS_ESCAPE);
return;
}
if (IsHexDigit(ch)) {
PRInt32 rv = 0;
int i;
for (i = 0; i < 6; i++) { // up to six digits
ch = Read(aErrorCode);
if (ch < 0) {
// Whoops: error or premature eof
break;
}
if (!IsHexDigit(ch) && !IsWhitespace(ch)) {
Pushback(ch);
break;
} else if (IsHexDigit(ch)) {
if (IsDigit(ch)) {
rv = rv * 16 + (ch - '0');
} else {
// Note: c&7 just keeps the low three bits which causes
// upper and lower case alphabetics to both yield their
// "relative to 10" value for computing the hex value.
rv = rv * 16 + ((ch & 0x7) + 9);
}
} else {
NS_ASSERTION(IsWhitespace(ch), "bad control flow");
// single space ends escape
break;
}
}
if (6 == i) { // look for trailing whitespace and eat it
ch = Peek(aErrorCode);
if (IsWhitespace(ch)) {
(void) Read(aErrorCode);
}
}
NS_ASSERTION(rv >= 0, "How did rv become negative?");
// "[at most six hexadecimal digits following a backslash] stand
// for the ISO 10646 character with that number, which must not be
// zero. (It is undefined in CSS 2.1 what happens if a style sheet
// does contain a character with Unicode codepoint zero.)"
// -- CSS2.1 section 4.1.3
//
// Silently deleting \0 opens a content-filtration loophole (see
// bug 228856), so what we do instead is pretend the "cancels the
// meaning of special characters" rule applied.
if (rv > 0) {
AppendUCS4ToUTF16(ENSURE_VALID_CHAR(rv), aOutput);
} else {
while (i--)
aOutput.Append('0');
if (IsWhitespace(ch))
Pushback(ch);
}
return;
} else {
// "Any character except a hexidecimal digit can be escaped to
// remove its special meaning by putting a backslash in front"
// -- CSS1 spec section 7.1
if (!EatNewline(aErrorCode)) { // skip escaped newline
(void) Read(aErrorCode);
if (ch > 0) {
aOutput.Append(ch);
}
}
return;
}
}
