PRBool nsCSSScanner::ParseString(nsresult& aErrorCode, PRInt32 aStop,
nsCSSToken& aToken)
{
aToken.mIdent.SetLength(0);
aToken.mType = eCSSToken_String;
aToken.mSymbol = PRUnichar(aStop); // remember how it's quoted
for (;;) {
if (EatNewline(aErrorCode)) {
aToken.mType = eCSSToken_Error;
#ifdef CSS_REPORT_PARSE_ERRORS
ReportUnexpectedToken(aToken, "SEUnterminatedString");
#endif
return PR_TRUE;
}
PRInt32 ch = Read(aErrorCode);
if (ch < 0) {
return PR_FALSE;
}
if (ch == aStop) {
break;
}
if (ch == CSS_ESCAPE) {
ParseAndAppendEscape(aErrorCode, aToken.mIdent);
}
else if (0 < ch) {
aToken.mIdent.Append(ch);
}
}
return PR_TRUE;
}
PRBool nsCSSScanner::ParseEOLComment(nsresult& aErrorCode, nsCSSToken& aToken)
{
nsString& ident = aToken.mIdent;
ident.SetLength(0);
for (;;) {
if (EatNewline(aErrorCode)) {
break;
}
PRInt32 ch = Read(aErrorCode);
if (ch < 0) {
break;
}
#ifdef COLLECT_WHITESPACE
ident.Append(PRUnichar(ch));
#endif
}
aToken.mType = eCSSToken_WhiteSpace;
return PR_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;
}
}
