PdfContentsTokenizer::PdfContentsTokenizer( PdfCanvas* pCanvas )
: PdfTokenizer()
{
PdfObject* pContents = pCanvas->GetContents();
if( pContents && pContents->IsArray() )
{
PdfArray& a = pContents->GetArray();
for ( PdfArray::iterator it = a.begin(); it != a.end() ; ++it )
{
if ( !(*it).IsReference() )
{
PODOFO_RAISE_ERROR_INFO( ePdfError_InvalidDataType, "/Contents array contained non-references" );
}
m_lstContents.push_back( pContents->GetOwner()->GetObject( (*it).GetReference() ) );
}
}
else if ( pContents && pContents->HasStream() )
{
m_lstContents.push_back( pContents );
}
else
{
PODOFO_RAISE_ERROR_INFO( ePdfError_InvalidDataType, "Page /Contents not stream or array of streams" );
}
if( m_lstContents.size() )
{
SetCurrentContentsStream( m_lstContents.front() );
m_lstContents.pop_front();
}
}
bool PdfContentsTokenizer::GetNextToken( const char*& pszToken , EPdfTokenType* peType )
{
bool result = PdfTokenizer::GetNextToken(pszToken, peType);
while (!result) {
if( !m_lstContents.size() )
return false;
SetCurrentContentsStream( m_lstContents.front() );
m_lstContents.pop_front();
result = PdfTokenizer::GetNextToken(pszToken, peType);
}
return result;
}
bool PdfContentsTokenizer::ReadNext( EPdfContentsType& reType, const char*& rpszKeyword, PdfVariant & rVariant )
{
EPdfTokenType eTokenType;
EPdfDataType eDataType;
const char* pszToken;
// While officially the keyword pointer is undefined if not needed, it
// costs us practically nothing to zero it (in case someone fails to check
// the return value and/or reType). Do so. We won't nullify the variant
// since that has a real cost.
//rpszKeyword = 0;
// If we've run out of data in this stream and there's another one to read,
// switch to reading the next stream.
//if( m_device.Device() && m_device.Device()->Eof() && m_lstContents.size() )
//{
// SetCurrentContentsStream( m_lstContents.front() );
// m_lstContents.pop_front();
//}
bool gotToken = this->GetNextToken( pszToken, &eTokenType );
if ( !gotToken )
{
if ( m_lstContents.size() )
{
// We ran out of tokens in this stream. Switch to the next stream
// and try again.
SetCurrentContentsStream( m_lstContents.front() );
m_lstContents.pop_front();
return ReadNext( reType, rpszKeyword, rVariant );
}
else
{
// No more content stream tokens to read.
return false;
}
}
eDataType = this->DetermineDataType( pszToken, eTokenType, rVariant );
// asume we read a variant unless we discover otherwise later.
reType = ePdfContentsType_Variant;
switch( eDataType )
{
case ePdfDataType_Null:
case ePdfDataType_Bool:
case ePdfDataType_Number:
case ePdfDataType_Real:
// the data was already read into rVariant by the DetermineDataType function
break;
case ePdfDataType_Reference:
{
// references are invalid in content streams
PODOFO_RAISE_ERROR_INFO( ePdfError_InvalidDataType, "references are invalid in content streams" );
break;
}
case ePdfDataType_Dictionary:
this->ReadDictionary( rVariant, NULL );
break;
case ePdfDataType_Array:
this->ReadArray( rVariant, NULL );
break;
case ePdfDataType_String:
this->ReadString( rVariant, NULL );
break;
case ePdfDataType_HexString:
this->ReadHexString( rVariant, NULL );
break;
case ePdfDataType_Name:
this->ReadName( rVariant );
break;
case ePdfDataType_Unknown:
case ePdfDataType_RawData:
default:
// Assume we have a keyword
reType = ePdfContentsType_Keyword;
rpszKeyword = pszToken;
break;
}
return true;
}
PdfContentsTokenizer::PdfContentsTokenizer( PdfCanvas* pCanvas )
: PdfTokenizer(), m_readingInlineImgData(false)
{
if( !pCanvas )
{
PODOFO_RAISE_ERROR( ePdfError_InvalidHandle );
}
PdfObject* pContents = pCanvas->GetContents();
if( pContents && pContents->IsArray() )
{
PdfArray& a = pContents->GetArray();
for ( PdfArray::iterator it = a.begin(); it != a.end() ; ++it )
{
if ( !(*it).IsReference() )
{
PODOFO_RAISE_ERROR_INFO( ePdfError_InvalidDataType, "/Contents array contained non-references" );
}
if ( !pContents->GetOwner()->GetObject( (*it).GetReference() ) )
{
// some damaged PDFs may have dangling references
PODOFO_RAISE_ERROR_INFO( ePdfError_InvalidDataType, "/Contents array NULL reference" );
}
m_lstContents.push_back( pContents->GetOwner()->GetObject( (*it).GetReference() ) );
}
}
else if ( pContents && pContents->HasStream() )
{
m_lstContents.push_back( pContents );
}
else if ( pContents && pContents->IsDictionary() )
{
m_lstContents.push_back( pContents );
PdfError::LogMessage(eLogSeverity_Information,
"PdfContentsTokenizer: found canvas-dictionary without stream => empty page");
// OC 18.09.2010 BugFix: Found an empty page in a PDF document:
// 103 0 obj
// <<
// /Type /Page
// /MediaBox [ 0 0 595 842 ]
// /Parent 3 0 R
// /Resources <<
// /ProcSet [ /PDF ]
// >>
// /Rotate 0
// >>
// endobj
}
else
{
PODOFO_RAISE_ERROR_INFO( ePdfError_InvalidDataType, "Page /Contents not stream or array of streams" );
}
if( m_lstContents.size() )
{
SetCurrentContentsStream( m_lstContents.front() );
m_lstContents.pop_front();
}
}