ByteString CPDF_StreamParser::ReadHexString() {
if (!PositionIsInBounds())
return ByteString();
std::ostringstream buf;
bool bFirst = true;
int code = 0;
while (PositionIsInBounds()) {
int ch = m_pBuf[m_Pos++];
if (ch == '>')
break;
if (!std::isxdigit(ch))
continue;
int val = FXSYS_HexCharToInt(ch);
if (bFirst) {
code = val * 16;
} else {
code += val;
buf << static_cast<uint8_t>(code);
}
bFirst = !bFirst;
}
if (!bFirst)
buf << static_cast<char>(code);
if (buf.tellp() <= 0)
return ByteString();
return ByteString(
buf.str().c_str(),
std::min(static_cast<size_t>(buf.tellp()), kMaxStringLength));
}
CFX_ByteString CPDF_StreamParser::ReadHexString() {
if (!PositionIsInBounds())
return CFX_ByteString();
CFX_ByteTextBuf buf;
bool bFirst = true;
int code = 0;
while (PositionIsInBounds()) {
int ch = m_pBuf[m_Pos++];
if (ch == '>')
break;
if (!std::isxdigit(ch))
continue;
int val = FXSYS_toHexDigit(ch);
if (bFirst) {
code = val * 16;
} else {
code += val;
buf.AppendByte((uint8_t)code);
}
bFirst = !bFirst;
}
if (!bFirst)
buf.AppendChar((char)code);
if (buf.GetLength() > MAX_STRING_LENGTH)
return CFX_ByteString(buf.GetBuffer(), MAX_STRING_LENGTH);
return buf.GetByteString();
}
CFX_ByteString CPDF_StreamParser::ReadHexString() {
if (!PositionIsInBounds())
return CFX_ByteString();
int ch = m_pBuf[m_Pos++];
CFX_ByteTextBuf buf;
FX_BOOL bFirst = TRUE;
int code = 0;
while (1) {
if (ch == '>') {
break;
}
if (ch >= '0' && ch <= '9') {
if (bFirst) {
code = (ch - '0') * 16;
} else {
code += ch - '0';
buf.AppendChar((char)code);
}
bFirst = !bFirst;
} else if (ch >= 'A' && ch <= 'F') {
if (bFirst) {
code = (ch - 'A' + 10) * 16;
} else {
code += ch - 'A' + 10;
buf.AppendChar((char)code);
}
bFirst = !bFirst;
} else if (ch >= 'a' && ch <= 'f') {
if (bFirst) {
code = (ch - 'a' + 10) * 16;
} else {
code += ch - 'a' + 10;
buf.AppendChar((char)code);
}
bFirst = !bFirst;
}
if (!PositionIsInBounds())
break;
ch = m_pBuf[m_Pos++];
}
if (!bFirst) {
buf.AppendChar((char)code);
}
if (buf.GetLength() > MAX_STRING_LENGTH) {
return CFX_ByteString(buf.GetBuffer(), MAX_STRING_LENGTH);
}
return buf.GetByteString();
}
CFX_ByteString CPDF_StreamParser::ReadString() {
if (!PositionIsInBounds())
return CFX_ByteString();
int ch = m_pBuf[m_Pos++];
CFX_ByteTextBuf buf;
int parlevel = 0;
int status = 0, iEscCode = 0;
while (1) {
switch (status) {
case 0:
if (ch == ')') {
if (parlevel == 0) {
if (buf.GetLength() > MAX_STRING_LENGTH) {
return CFX_ByteString(buf.GetBuffer(), MAX_STRING_LENGTH);
}
return buf.GetByteString();
}
parlevel--;
buf.AppendChar(')');
} else if (ch == '(') {
parlevel++;
buf.AppendChar('(');
} else if (ch == '\\') {
status = 1;
} else {
buf.AppendChar((char)ch);
}
break;
case 1:
if (ch >= '0' && ch <= '7') {
iEscCode = ch - '0';
status = 2;
break;
}
if (ch == 'n') {
buf.AppendChar('\n');
} else if (ch == 'r') {
buf.AppendChar('\r');
} else if (ch == 't') {
buf.AppendChar('\t');
} else if (ch == 'b') {
buf.AppendChar('\b');
} else if (ch == 'f') {
buf.AppendChar('\f');
} else if (ch == '\r') {
status = 4;
break;
} else if (ch == '\n') {
} else {
buf.AppendChar(ch);
}
status = 0;
break;
case 2:
if (ch >= '0' && ch <= '7') {
iEscCode = iEscCode * 8 + ch - '0';
status = 3;
} else {
buf.AppendChar(iEscCode);
status = 0;
continue;
}
break;
case 3:
if (ch >= '0' && ch <= '7') {
iEscCode = iEscCode * 8 + ch - '0';
buf.AppendChar(iEscCode);
status = 0;
} else {
buf.AppendChar(iEscCode);
status = 0;
continue;
}
break;
case 4:
status = 0;
if (ch != '\n') {
continue;
}
break;
}
if (!PositionIsInBounds())
break;
ch = m_pBuf[m_Pos++];
}
if (PositionIsInBounds())
ch = m_pBuf[m_Pos++];
if (buf.GetLength() > MAX_STRING_LENGTH) {
return CFX_ByteString(buf.GetBuffer(), MAX_STRING_LENGTH);
}
return buf.GetByteString();
}
void CPDF_StreamParser::GetNextWord(FX_BOOL& bIsNumber) {
m_WordSize = 0;
bIsNumber = TRUE;
if (!PositionIsInBounds())
return;
int ch = m_pBuf[m_Pos++];
while (1) {
while (PDFCharIsWhitespace(ch)) {
if (!PositionIsInBounds()) {
return;
}
ch = m_pBuf[m_Pos++];
}
if (ch != '%')
break;
while (1) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (PDFCharIsLineEnding(ch))
break;
}
}
if (PDFCharIsDelimiter(ch)) {
bIsNumber = FALSE;
m_WordBuffer[m_WordSize++] = ch;
if (ch == '/') {
while (1) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (!PDFCharIsOther(ch) && !PDFCharIsNumeric(ch)) {
m_Pos--;
return;
}
if (m_WordSize < MAX_WORD_BUFFER)
m_WordBuffer[m_WordSize++] = ch;
}
} else if (ch == '<') {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (ch == '<')
m_WordBuffer[m_WordSize++] = ch;
else
m_Pos--;
} else if (ch == '>') {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (ch == '>')
m_WordBuffer[m_WordSize++] = ch;
else
m_Pos--;
}
return;
}
while (1) {
if (m_WordSize < MAX_WORD_BUFFER)
m_WordBuffer[m_WordSize++] = ch;
if (!PDFCharIsNumeric(ch))
bIsNumber = FALSE;
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (PDFCharIsDelimiter(ch) || PDFCharIsWhitespace(ch)) {
m_Pos--;
break;
}
}
}
void CPDF_StreamParser::SkipPathObject() {
FX_DWORD command_startpos = m_Pos;
if (!PositionIsInBounds())
return;
int ch = m_pBuf[m_Pos++];
while (1) {
while (PDFCharIsWhitespace(ch)) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
}
if (!PDFCharIsNumeric(ch)) {
m_Pos = command_startpos;
return;
}
while (1) {
while (!PDFCharIsWhitespace(ch)) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
}
while (PDFCharIsWhitespace(ch)) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
}
if (PDFCharIsNumeric(ch))
continue;
FX_DWORD op_startpos = m_Pos - 1;
while (!PDFCharIsWhitespace(ch) && !PDFCharIsDelimiter(ch)) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
}
if (m_Pos - op_startpos == 2) {
int op = m_pBuf[op_startpos];
if (op == kPathOperatorSubpath || op == kPathOperatorLine ||
op == kPathOperatorCubicBezier1 ||
op == kPathOperatorCubicBezier2 ||
op == kPathOperatorCubicBezier3) {
command_startpos = m_Pos;
break;
}
} else if (m_Pos - op_startpos == 3) {
if (m_pBuf[op_startpos] == kPathOperatorRectangle[0] &&
m_pBuf[op_startpos + 1] == kPathOperatorRectangle[1]) {
command_startpos = m_Pos;
break;
}
}
m_Pos = command_startpos;
return;
}
}
}
CPDF_StreamParser::SyntaxType CPDF_StreamParser::ParseNextElement() {
if (m_pLastObj) {
m_pLastObj->Release();
m_pLastObj = nullptr;
}
m_WordSize = 0;
FX_BOOL bIsNumber = TRUE;
if (!PositionIsInBounds())
return EndOfData;
int ch = m_pBuf[m_Pos++];
while (1) {
while (PDFCharIsWhitespace(ch)) {
if (!PositionIsInBounds())
return EndOfData;
ch = m_pBuf[m_Pos++];
}
if (ch != '%')
break;
while (1) {
if (!PositionIsInBounds())
return EndOfData;
ch = m_pBuf[m_Pos++];
if (PDFCharIsLineEnding(ch))
break;
}
}
if (PDFCharIsDelimiter(ch) && ch != '/') {
m_Pos--;
m_pLastObj = ReadNextObject();
return Others;
}
while (1) {
if (m_WordSize < MAX_WORD_BUFFER)
m_WordBuffer[m_WordSize++] = ch;
if (!PDFCharIsNumeric(ch))
bIsNumber = FALSE;
if (!PositionIsInBounds())
break;
ch = m_pBuf[m_Pos++];
if (PDFCharIsDelimiter(ch) || PDFCharIsWhitespace(ch)) {
m_Pos--;
break;
}
}
m_WordBuffer[m_WordSize] = 0;
if (bIsNumber)
return Number;
if (m_WordBuffer[0] == '/')
return Name;
if (m_WordSize == 4) {
if (*(FX_DWORD*)m_WordBuffer == FXDWORD_TRUE) {
m_pLastObj = CPDF_Boolean::Create(TRUE);
return Others;
}
if (*(FX_DWORD*)m_WordBuffer == FXDWORD_NULL) {
m_pLastObj = CPDF_Null::Create();
return Others;
}
} else if (m_WordSize == 5) {
if (*(FX_DWORD*)m_WordBuffer == FXDWORD_FALS && m_WordBuffer[4] == 'e') {
m_pLastObj = CPDF_Boolean::Create(FALSE);
return Others;
}
}
return Keyword;
}
ByteString CPDF_StreamParser::ReadString() {
if (!PositionIsInBounds())
return ByteString();
uint8_t ch = m_pBuf[m_Pos++];
std::ostringstream buf;
int parlevel = 0;
int status = 0;
int iEscCode = 0;
while (1) {
switch (status) {
case 0:
if (ch == ')') {
if (parlevel == 0) {
if (buf.tellp() <= 0)
return ByteString();
return ByteString(
buf.str().c_str(),
std::min(static_cast<size_t>(buf.tellp()), kMaxStringLength));
}
parlevel--;
buf << ')';
} else if (ch == '(') {
parlevel++;
buf << '(';
} else if (ch == '\\') {
status = 1;
} else {
buf << static_cast<char>(ch);
}
break;
case 1:
if (FXSYS_IsOctalDigit(ch)) {
iEscCode = FXSYS_DecimalCharToInt(static_cast<char>(ch));
status = 2;
break;
}
if (ch == '\r') {
status = 4;
break;
}
if (ch == '\n') {
// Do nothing.
} else if (ch == 'n') {
buf << '\n';
} else if (ch == 'r') {
buf << '\r';
} else if (ch == 't') {
buf << '\t';
} else if (ch == 'b') {
buf << '\b';
} else if (ch == 'f') {
buf << '\f';
} else {
buf << static_cast<char>(ch);
}
status = 0;
break;
case 2:
if (FXSYS_IsOctalDigit(ch)) {
iEscCode =
iEscCode * 8 + FXSYS_DecimalCharToInt(static_cast<char>(ch));
status = 3;
} else {
buf << static_cast<char>(iEscCode);
status = 0;
continue;
}
break;
case 3:
if (FXSYS_IsOctalDigit(ch)) {
iEscCode =
iEscCode * 8 + FXSYS_DecimalCharToInt(static_cast<char>(ch));
buf << static_cast<char>(iEscCode);
status = 0;
} else {
buf << static_cast<char>(iEscCode);
status = 0;
continue;
}
break;
case 4:
status = 0;
if (ch != '\n')
continue;
break;
}
if (!PositionIsInBounds())
break;
ch = m_pBuf[m_Pos++];
}
if (PositionIsInBounds())
++m_Pos;
if (buf.tellp() <= 0)
return ByteString();
return ByteString(
buf.str().c_str(),
std::min(static_cast<size_t>(buf.tellp()), kMaxStringLength));
}
// TODO(npm): the following methods are almost identical in cpdf_syntaxparser
void CPDF_StreamParser::GetNextWord(bool& bIsNumber) {
m_WordSize = 0;
bIsNumber = true;
if (!PositionIsInBounds())
return;
int ch = m_pBuf[m_Pos++];
while (1) {
while (PDFCharIsWhitespace(ch)) {
if (!PositionIsInBounds()) {
return;
}
ch = m_pBuf[m_Pos++];
}
if (ch != '%')
break;
while (1) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (PDFCharIsLineEnding(ch))
break;
}
}
if (PDFCharIsDelimiter(ch)) {
bIsNumber = false;
m_WordBuffer[m_WordSize++] = ch;
if (ch == '/') {
while (1) {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (!PDFCharIsOther(ch) && !PDFCharIsNumeric(ch)) {
m_Pos--;
return;
}
if (m_WordSize < kMaxWordLength)
m_WordBuffer[m_WordSize++] = ch;
}
} else if (ch == '<') {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (ch == '<')
m_WordBuffer[m_WordSize++] = ch;
else
m_Pos--;
} else if (ch == '>') {
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (ch == '>')
m_WordBuffer[m_WordSize++] = ch;
else
m_Pos--;
}
return;
}
while (1) {
if (m_WordSize < kMaxWordLength)
m_WordBuffer[m_WordSize++] = ch;
if (!PDFCharIsNumeric(ch))
bIsNumber = false;
if (!PositionIsInBounds())
return;
ch = m_pBuf[m_Pos++];
if (PDFCharIsDelimiter(ch) || PDFCharIsWhitespace(ch)) {
m_Pos--;
break;
}
}
}
CPDF_StreamParser::SyntaxType CPDF_StreamParser::ParseNextElement() {
m_pLastObj.reset();
m_WordSize = 0;
if (!PositionIsInBounds())
return EndOfData;
int ch = m_pBuf[m_Pos++];
while (1) {
while (PDFCharIsWhitespace(ch)) {
if (!PositionIsInBounds())
return EndOfData;
ch = m_pBuf[m_Pos++];
}
if (ch != '%')
break;
while (1) {
if (!PositionIsInBounds())
return EndOfData;
ch = m_pBuf[m_Pos++];
if (PDFCharIsLineEnding(ch))
break;
}
}
if (PDFCharIsDelimiter(ch) && ch != '/') {
m_Pos--;
m_pLastObj = ReadNextObject(false, false, 0);
return Others;
}
bool bIsNumber = true;
while (1) {
if (m_WordSize < kMaxWordLength)
m_WordBuffer[m_WordSize++] = ch;
if (!PDFCharIsNumeric(ch))
bIsNumber = false;
if (!PositionIsInBounds())
break;
ch = m_pBuf[m_Pos++];
if (PDFCharIsDelimiter(ch) || PDFCharIsWhitespace(ch)) {
m_Pos--;
break;
}
}
m_WordBuffer[m_WordSize] = 0;
if (bIsNumber)
return Number;
if (m_WordBuffer[0] == '/')
return Name;
if (m_WordSize == 4) {
if (memcmp(m_WordBuffer, "true", 4) == 0) {
m_pLastObj = pdfium::MakeUnique<CPDF_Boolean>(true);
return Others;
}
if (memcmp(m_WordBuffer, "null", 4) == 0) {
m_pLastObj = pdfium::MakeUnique<CPDF_Null>();
return Others;
}
} else if (m_WordSize == 5) {
if (memcmp(m_WordBuffer, "false", 5) == 0) {
m_pLastObj = pdfium::MakeUnique<CPDF_Boolean>(false);
return Others;
}
}
return Keyword;
}
