CFX_ByteString PDF_NameEncode(const CFX_ByteString& orig) {
uint8_t* src_buf = (uint8_t*)orig.c_str();
int src_len = orig.GetLength();
int dest_len = 0;
int i;
for (i = 0; i < src_len; i++) {
uint8_t ch = src_buf[i];
if (ch >= 0x80 || PDFCharIsWhitespace(ch) || ch == '#' ||
PDFCharIsDelimiter(ch)) {
dest_len += 3;
} else {
dest_len++;
}
}
if (dest_len == src_len)
return orig;
CFX_ByteString res;
FX_CHAR* dest_buf = res.GetBuffer(dest_len);
dest_len = 0;
for (i = 0; i < src_len; i++) {
uint8_t ch = src_buf[i];
if (ch >= 0x80 || PDFCharIsWhitespace(ch) || ch == '#' ||
PDFCharIsDelimiter(ch)) {
dest_buf[dest_len++] = '#';
dest_buf[dest_len++] = "0123456789ABCDEF"[ch / 16];
dest_buf[dest_len++] = "0123456789ABCDEF"[ch % 16];
} else {
dest_buf[dest_len++] = ch;
}
}
dest_buf[dest_len] = 0;
res.ReleaseBuffer();
return res;
}
bool CPDF_DataAvail::GetNextToken(CFX_ByteString& token) {
uint8_t ch;
if (!GetNextChar(ch))
return false;
while (1) {
while (PDFCharIsWhitespace(ch)) {
if (!GetNextChar(ch))
return false;
}
if (ch != '%')
break;
while (1) {
if (!GetNextChar(ch))
return false;
if (PDFCharIsLineEnding(ch))
break;
}
}
uint8_t buffer[256];
uint32_t index = 0;
if (PDFCharIsDelimiter(ch)) {
buffer[index++] = ch;
if (ch == '/') {
while (1) {
if (!GetNextChar(ch))
return false;
if (!PDFCharIsOther(ch) && !PDFCharIsNumeric(ch)) {
m_Pos--;
CFX_ByteString ret(buffer, index);
token = ret;
return true;
}
if (index < sizeof(buffer))
buffer[index++] = ch;
}
} else if (ch == '<') {
if (!GetNextChar(ch))
return false;
if (ch == '<')
buffer[index++] = ch;
else
m_Pos--;
} else if (ch == '>') {
if (!GetNextChar(ch))
return false;
if (ch == '>')
buffer[index++] = ch;
else
m_Pos--;
}
CFX_ByteString ret(buffer, index);
token = ret;
return true;
}
while (1) {
if (index < sizeof(buffer))
buffer[index++] = ch;
if (!GetNextChar(ch))
return false;
if (PDFCharIsDelimiter(ch) || PDFCharIsWhitespace(ch)) {
m_Pos--;
break;
}
}
token = CFX_ByteString(buffer, index);
return true;
}
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;
}
void CPDF_SimpleParser::ParseWord(const uint8_t*& pStart,
FX_DWORD& dwSize,
int& type) {
pStart = NULL;
dwSize = 0;
type = PDFWORD_EOF;
uint8_t ch;
while (1) {
if (m_dwSize <= m_dwCurPos)
return;
ch = m_pData[m_dwCurPos++];
while (PDFCharIsWhitespace(ch)) {
if (m_dwSize <= m_dwCurPos)
return;
ch = m_pData[m_dwCurPos++];
}
if (ch != '%')
break;
while (1) {
if (m_dwSize <= m_dwCurPos)
return;
ch = m_pData[m_dwCurPos++];
if (ch == '\r' || ch == '\n')
break;
}
}
FX_DWORD start_pos = m_dwCurPos - 1;
pStart = m_pData + start_pos;
if (PDFCharIsDelimiter(ch)) {
if (ch == '/') {
while (1) {
if (m_dwSize <= m_dwCurPos)
return;
ch = m_pData[m_dwCurPos++];
if (!PDFCharIsOther(ch) && !PDFCharIsNumeric(ch)) {
m_dwCurPos--;
dwSize = m_dwCurPos - start_pos;
type = PDFWORD_NAME;
return;
}
}
} else {
type = PDFWORD_DELIMITER;
dwSize = 1;
if (ch == '<') {
if (m_dwSize <= m_dwCurPos)
return;
ch = m_pData[m_dwCurPos++];
if (ch == '<')
dwSize = 2;
else
m_dwCurPos--;
} else if (ch == '>') {
if (m_dwSize <= m_dwCurPos)
return;
ch = m_pData[m_dwCurPos++];
if (ch == '>')
dwSize = 2;
else
m_dwCurPos--;
}
}
return;
}
type = PDFWORD_NUMBER;
dwSize = 1;
while (1) {
if (!PDFCharIsNumeric(ch))
type = PDFWORD_TEXT;
if (m_dwSize <= m_dwCurPos)
return;
ch = m_pData[m_dwCurPos++];
if (PDFCharIsDelimiter(ch) || PDFCharIsWhitespace(ch)) {
m_dwCurPos--;
break;
}
dwSize++;
}
}
// 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;
}
