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;
}
CPDF_Stream* CPDF_StreamParser::ReadInlineStream(CPDF_Document* pDoc,
CPDF_Dictionary* pDict,
CPDF_Object* pCSObj,
FX_BOOL bDecode) {
if (m_Pos == m_Size)
return nullptr;
if (PDFCharIsWhitespace(m_pBuf[m_Pos]))
m_Pos++;
CFX_ByteString Decoder;
CPDF_Dictionary* pParam = nullptr;
CPDF_Object* pFilter = pDict->GetElementValue(FX_BSTRC("Filter"));
if (pFilter) {
if (CPDF_Array* pArray = pFilter->AsArray()) {
Decoder = pArray->GetString(0);
CPDF_Array* pParams = pDict->GetArray(FX_BSTRC("DecodeParms"));
if (pParams)
pParam = pParams->GetDict(0);
} else {
Decoder = pFilter->GetString();
pParam = pDict->GetDict(FX_BSTRC("DecodeParms"));
}
}
FX_DWORD width = pDict->GetInteger(FX_BSTRC("Width"));
FX_DWORD height = pDict->GetInteger(FX_BSTRC("Height"));
FX_DWORD OrigSize = 0;
if (pCSObj != NULL) {
FX_DWORD bpc = pDict->GetInteger(FX_BSTRC("BitsPerComponent"));
FX_DWORD nComponents = 1;
CPDF_ColorSpace* pCS = pDoc->LoadColorSpace(pCSObj);
if (pCS == NULL) {
nComponents = 3;
} else {
nComponents = pCS->CountComponents();
pDoc->GetPageData()->ReleaseColorSpace(pCSObj);
}
FX_DWORD pitch = width;
if (bpc && pitch > INT_MAX / bpc) {
return NULL;
}
pitch *= bpc;
if (nComponents && pitch > INT_MAX / nComponents) {
return NULL;
}
pitch *= nComponents;
if (pitch > INT_MAX - 7) {
return NULL;
}
pitch += 7;
pitch /= 8;
OrigSize = pitch;
} else {
if (width > INT_MAX - 7) {
return NULL;
}
OrigSize = ((width + 7) / 8);
}
if (height && OrigSize > INT_MAX / height) {
return NULL;
}
OrigSize *= height;
uint8_t* pData = NULL;
FX_DWORD dwStreamSize;
if (Decoder.IsEmpty()) {
if (OrigSize > m_Size - m_Pos) {
OrigSize = m_Size - m_Pos;
}
pData = FX_Alloc(uint8_t, OrigSize);
FXSYS_memcpy(pData, m_pBuf + m_Pos, OrigSize);
dwStreamSize = OrigSize;
m_Pos += OrigSize;
} else {
FX_DWORD dwDestSize = OrigSize;
dwStreamSize =
PDF_DecodeInlineStream(m_pBuf + m_Pos, m_Size - m_Pos, width, height,
Decoder, pParam, pData, dwDestSize);
if ((int)dwStreamSize < 0) {
FX_Free(pData);
return NULL;
}
if (bDecode) {
m_Pos += dwStreamSize;
dwStreamSize = dwDestSize;
if (CPDF_Array* pArray = pFilter->AsArray()) {
pArray->RemoveAt(0);
CPDF_Array* pParams = pDict->GetArray(FX_BSTRC("DecodeParms"));
if (pParams)
pParams->RemoveAt(0);
} else {
pDict->RemoveAt(FX_BSTRC("Filter"));
pDict->RemoveAt(FX_BSTRC("DecodeParms"));
}
} else {
FX_Free(pData);
FX_DWORD dwSavePos = m_Pos;
m_Pos += dwStreamSize;
while (1) {
FX_DWORD dwPrevPos = m_Pos;
CPDF_StreamParser::SyntaxType type = ParseNextElement();
if (type == CPDF_StreamParser::EndOfData) {
break;
}
if (type != CPDF_StreamParser::Keyword) {
dwStreamSize += m_Pos - dwPrevPos;
continue;
}
if (GetWordSize() == 2 && GetWordBuf()[0] == 'E' &&
GetWordBuf()[1] == 'I') {
m_Pos = dwPrevPos;
break;
}
dwStreamSize += m_Pos - dwPrevPos;
}
m_Pos = dwSavePos;
pData = FX_Alloc(uint8_t, dwStreamSize);
FXSYS_memcpy(pData, m_pBuf + m_Pos, dwStreamSize);
m_Pos += dwStreamSize;
}
}
pDict->SetAtInteger(FX_BSTRC("Length"), (int)dwStreamSize);
return new CPDF_Stream(pData, dwStreamSize, pDict);
}
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;
}
std::unique_ptr<CPDF_Stream> CPDF_StreamParser::ReadInlineStream(
CPDF_Document* pDoc,
std::unique_ptr<CPDF_Dictionary> pDict,
const CPDF_Object* pCSObj) {
if (m_Pos < m_pBuf.size() && PDFCharIsWhitespace(m_pBuf[m_Pos]))
m_Pos++;
if (m_Pos == m_pBuf.size())
return nullptr;
ByteString Decoder;
CPDF_Dictionary* pParam = nullptr;
CPDF_Object* pFilter = pDict->GetDirectObjectFor("Filter");
if (pFilter) {
if (CPDF_Array* pArray = pFilter->AsArray()) {
Decoder = pArray->GetStringAt(0);
CPDF_Array* pParams = pDict->GetArrayFor(pdfium::stream::kDecodeParms);
if (pParams)
pParam = pParams->GetDictAt(0);
} else {
Decoder = pFilter->GetString();
pParam = pDict->GetDictFor(pdfium::stream::kDecodeParms);
}
}
uint32_t width = pDict->GetIntegerFor("Width");
uint32_t height = pDict->GetIntegerFor("Height");
uint32_t OrigSize = 0;
if (pCSObj) {
uint32_t bpc = pDict->GetIntegerFor("BitsPerComponent");
uint32_t nComponents = 1;
CPDF_ColorSpace* pCS = pDoc->LoadColorSpace(pCSObj, nullptr);
if (pCS) {
nComponents = pCS->CountComponents();
pDoc->GetPageData()->ReleaseColorSpace(pCSObj);
} else {
nComponents = 3;
}
uint32_t pitch = width;
if (bpc && pitch > INT_MAX / bpc)
return nullptr;
pitch *= bpc;
if (nComponents && pitch > INT_MAX / nComponents)
return nullptr;
pitch *= nComponents;
if (pitch > INT_MAX - 7)
return nullptr;
pitch += 7;
pitch /= 8;
OrigSize = pitch;
} else {
if (width > INT_MAX - 7)
return nullptr;
OrigSize = ((width + 7) / 8);
}
if (height && OrigSize > INT_MAX / height)
return nullptr;
OrigSize *= height;
std::unique_ptr<uint8_t, FxFreeDeleter> pData;
uint32_t dwStreamSize;
if (Decoder.IsEmpty()) {
if (OrigSize > m_pBuf.size() - m_Pos)
OrigSize = m_pBuf.size() - m_Pos;
pData.reset(FX_Alloc(uint8_t, OrigSize));
auto copy_span = m_pBuf.subspan(m_Pos, OrigSize);
memcpy(pData.get(), copy_span.data(), copy_span.size());
dwStreamSize = OrigSize;
m_Pos += OrigSize;
} else {
dwStreamSize = DecodeInlineStream(m_pBuf.subspan(m_Pos), width, height,
Decoder, pParam, OrigSize);
if (static_cast<int>(dwStreamSize) < 0)
return nullptr;
uint32_t dwSavePos = m_Pos;
m_Pos += dwStreamSize;
while (1) {
uint32_t dwPrevPos = m_Pos;
CPDF_StreamParser::SyntaxType type = ParseNextElement();
if (type == CPDF_StreamParser::EndOfData)
break;
if (type != CPDF_StreamParser::Keyword) {
dwStreamSize += m_Pos - dwPrevPos;
continue;
}
if (GetWord() == "EI") {
m_Pos = dwPrevPos;
break;
}
dwStreamSize += m_Pos - dwPrevPos;
}
m_Pos = dwSavePos;
pData.reset(FX_Alloc(uint8_t, dwStreamSize));
auto copy_span = m_pBuf.subspan(m_Pos, dwStreamSize);
memcpy(pData.get(), copy_span.data(), copy_span.size());
m_Pos += dwStreamSize;
}
pDict->SetNewFor<CPDF_Number>("Length", static_cast<int>(dwStreamSize));
return pdfium::MakeUnique<CPDF_Stream>(std::move(pData), dwStreamSize,
std::move(pDict));
}