/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
/*
* This file is part of the LibreOffice project.
*
* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/.
*
* This file incorporates work covered by the following license notice:
*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed
* with this work for additional information regarding copyright
* ownership. The ASF licenses this file to you under the Apache
* License, Version 2.0 (the "License"); you may not use this file
* except in compliance with the License. You may obtain a copy of
* the License at http://www.apache.org/licenses/LICENSE-2.0 .
*/
#include <pdfparse.hxx>
#include <boost/spirit/include/classic.hpp>
#include <boost/bind/bind.hpp>
#include <string.h>
#include <o3tl/char16_t2wchar_t.hxx>
#include <o3tl/safeint.hxx>
#include <osl/thread.h>
#include <rtl/strbuf.hxx>
#include <rtl/ustrbuf.hxx>
#include <sal/log.hxx>
#include <utility>
using namespace boost::spirit::classic;
using namespace pdfparse;
namespace {
class StringEmitContext : public EmitContext
{
OStringBuffer m_aBuf;
public:
StringEmitContext() : m_aBuf(256) {}
virtual bool write( const void* pBuf, unsigned int nLen ) noexcept override
{
m_aBuf.append( static_cast<const char*>(pBuf), nLen );
return true;
}
virtual unsigned int getCurPos() noexcept override { return m_aBuf.getLength(); }
virtual bool copyOrigBytes( unsigned int nOrigOffset, unsigned int nLen ) noexcept override
{ return (nOrigOffset+nLen < o3tl::make_unsigned(m_aBuf.getLength()) ) &&
write( m_aBuf.getStr() + nOrigOffset, nLen ); }
virtual unsigned int readOrigBytes( unsigned int nOrigOffset, unsigned int nLen, void* pBuf ) noexcept override
{
if( nOrigOffset+nLen < o3tl::make_unsigned(m_aBuf.getLength()) )
{
memcpy( pBuf, m_aBuf.getStr()+nOrigOffset, nLen );
return nLen;
}
return 0;
}
OString getString() { return m_aBuf.makeStringAndClear(); }
};
template< class iteratorT >
class PDFGrammar : public grammar< PDFGrammar<iteratorT> >
{
public:
explicit PDFGrammar( iteratorT first )
: m_fDouble( 0.0 ), m_aGlobalBegin(std::move( first )) {}
~PDFGrammar()
{
if( !m_aObjectStack.empty() )
delete m_aObjectStack.front();
}
double m_fDouble;
std::vector< unsigned int > m_aUIntStack;
std::vector< PDFEntry* > m_aObjectStack;
OString m_aErrorString;
iteratorT m_aGlobalBegin;
public:
struct pdf_string_parser
{
typedef nil_t result_t;
template <typename ScannerT>
std::ptrdiff_t
operator()(ScannerT const& scan, result_t&) const
{
std::ptrdiff_t len = 0;
int nBraceLevel = 0;
while( ! scan.at_end() )
{
char c = *scan;
if( c == ')' )
{
nBraceLevel--;
if( nBraceLevel < 0 )
break;
}
else if( c == '(' )
nBraceLevel++;
else if( c == '\\' ) // ignore escaped braces
{
++len;
++scan.first; // tdf#63054: avoid skipping spaces
if( scan.first == scan.last ) // tdf#63054: avoid skipping spaces
break;
}
++len;
++scan;
}
return scan.at_end() ? -1 : len;
}
};
template< typename ScannerT >
struct definition
{
explicit definition( const PDFGrammar<iteratorT>& rSelf )
{
using namespace boost::placeholders;
PDFGrammar<iteratorT>* pSelf = const_cast< PDFGrammar<iteratorT>* >( &rSelf );
// workaround workshop compiler: comment_p doesn't work
// comment = comment_p("%")[boost::bind(&PDFGrammar::pushComment, pSelf, _1, _2 )];
comment = lexeme_d[ (ch_p('%') >> *(~ch_p('\r') & ~ch_p('\n')) >> eol_p)[boost::bind(&PDFGrammar::pushComment, pSelf, _1, _2 )] ];
boolean = (str_p("true") | str_p("false"))[boost::bind(&PDFGrammar::pushBool, pSelf, _1, _2)];
// workaround workshop compiler: confix_p doesn't work
//stream = confix_p( "stream", *anychar_p, "endstream" )[boost::bind(&PDFGrammar::emitStream, pSelf, _1, _2 )];
stream = (str_p("stream") >> *(anychar_p - str_p("endstream")) >> str_p("endstream"))[boost::bind(&PDFGrammar::emitStream, pSelf, _1, _2 )];
name = lexeme_d[
ch_p('/')
>> (*(anychar_p-chset_p("\t\n\f\r ()<>[]{}/%")-ch_p('\0')))
[boost::bind(&PDFGrammar::pushName, pSelf, _1, _2)] ];
// workaround workshop compiler: confix_p doesn't work
//stringtype = ( confix_p("(",*anychar_p, ")") |
// confix_p("<",*xdigit_p, ">") )
// [boost::bind(&PDFGrammar::pushString,pSelf, _1, _2)];
stringtype = ( ( ch_p('(') >> functor_parser<pdf_string_parser>() >> ch_p(')') ) |
( ch_p('<') >> *xdigit_p >> ch_p('>') ) )
[boost::bind(&PDFGrammar::pushString,pSelf, _1, _2)];
null_object = str_p( "null" )[boost::bind(&PDFGrammar::pushNull, pSelf, _1, _2)];
#ifdef USE_ASSIGN_ACTOR
objectref = ( uint_p[push_back_a(pSelf->m_aUIntStack)]
>> uint_p[push_back_a(pSelf->m_aUIntStack)]
>> ch_p('R')
>> eps_p
)[boost::bind(&PDFGrammar::pushObjectRef, pSelf, _1, _2)];
#else
objectref = ( uint_p[boost::bind(&PDFGrammar::push_back_action_uint, pSelf, _1)]
>> uint_p[boost::bind(&PDFGrammar::push_back_action_uint, pSelf, _1)]
>> ch_p('R')
>> eps_p
)[boost::bind(&PDFGrammar::pushObjectRef, pSelf, _1, _2)];
#endif
#ifdef USE_ASSIGN_ACTOR
simple_type = objectref | name |
( real_p[assign_a(pSelf->m_fDouble)] >> eps_p )
[boost::bind(&PDFGrammar::pushDouble, pSelf, _1, _2)]
| stringtype | boolean | null_object;
#else
simple_type = objectref | name |
( real_p[boost::bind(&PDFGrammar::assign_action_double, pSelf, _1)] >> eps_p )
[boost::bind(&PDFGrammar::pushDouble, pSelf, _1, _2)]
| stringtype | boolean | null_object;
#endif
dict_begin = str_p( "<<" )[boost::bind(&PDFGrammar::beginDict, pSelf, _1, _2)];
dict_end = str_p( ">>" )[boost::bind(&PDFGrammar::endDict, pSelf, _1, _2)];
array_begin = str_p("[")[boost::bind(&PDFGrammar::beginArray,pSelf, _1, _2)];
array_end = str_p("]")[boost::bind(&PDFGrammar::endArray,pSelf, _1, _2)];
#ifdef USE_ASSIGN_ACTOR
object_begin= uint_p[push_back_a(pSelf->m_aUIntStack)]
>> uint_p[push_back_a(pSelf->m_aUIntStack)]
>> str_p("obj" )[boost::bind(&PDFGrammar::beginObject, pSelf, _1, _2)];
#else
object_begin= uint_p[boost::bind(&PDFGrammar::push_back_action_uint, pSelf, _1)]
>> uint_p[boost::bind(&PDFGrammar::push_back_action_uint, pSelf, _1)]
>> str_p("obj" )[boost::bind(&PDFGrammar::beginObject, pSelf, _1, _2)];
#endif
object_end = str_p( "endobj" )[boost::bind(&PDFGrammar::endObject, pSelf, _1, _2)];
xref = str_p( "xref" ) >> uint_p >> uint_p
>> lexeme_d[
+( repeat_p(10)[digit_p]
>> blank_p
>> repeat_p(5)[digit_p]
>> blank_p
>> ( ch_p('n') | ch_p('f') )
>> repeat_p(2)[space_p]
) ];
dict_element= dict_begin | comment | simple_type
| array_begin | array_end | dict_end;
object = object_begin
>> *dict_element
>> !stream
>> object_end;
trailer = str_p( "trailer" )[boost::bind(&PDFGrammar::beginTrailer,pSelf,_1,_2)]
>> *dict_element
>> str_p("startxref")
>> uint_p
>> str_p("%%EOF")[boost::bind(&PDFGrammar::endTrailer,pSelf,_1,_2)];
#ifdef USE_ASSIGN_ACTOR
pdfrule = ! (lexeme_d[
str_p( "%PDF-" )
>> uint_p[push_back_a(pSelf->m_aUIntStack)]
>> ch_p('.')
>> uint_p[push_back_a(pSelf->m_aUIntStack)]
>> *((~ch_p('\r') & ~ch_p('\n')))
>> eol_p
])[boost::bind(&PDFGrammar::haveFile,pSelf, _1, _2)]
>> *( comment | object | ( xref >> trailer ) );
#else
pdfrule = ! (lexeme_d[
str_p( "%PDF-" )
>> uint_p[boost::bind(&PDFGrammar::push_back_action_uint, pSelf, _1)]
>> ch_p('.')
>> uint_p[boost::bind(&PDFGrammar::push_back_action_uint, pSelf, _1)]
>> *(~ch_p('\r') & ~ch_p('\n'))
>> eol_p
])[boost::bind(&PDFGrammar::haveFile,pSelf, _1, _2)]
>> *( comment | object | ( xref >> trailer ) );
#endif
}
rule< ScannerT > comment, stream, boolean, name, stringtype, null_object, simple_type,
objectref, array, value, dict_element, dict_begin, dict_end,
array_begin, array_end, object, object_begin, object_end,
xref, trailer, pdfrule;
const rule< ScannerT >& start() const { return pdfrule; }
};
#ifndef USE_ASSIGN_ACTOR
void push_back_action_uint( unsigned int i )
{
m_aUIntStack.push_back( i );
}
void assign_action_double( double d )
{
m_fDouble = d;
}
#endif
[[noreturn]] static void parseError( const char* pMessage, const iteratorT& pLocation )
{
throw_( pLocation, pMessage );
}
OString iteratorToString( iteratorT first, const iteratorT& last ) const
{
OStringBuffer aStr( 32 );
while( first != last )
{
aStr.append( *first );
++first;
}
return aStr.makeStringAndClear();
}
void haveFile( const iteratorT& pBegin, SAL_UNUSED_PARAMETER iteratorT /*pEnd*/ )
{
if( m_aObjectStack.empty() )
{
PDFFile* pFile = new PDFFile();
pFile->m_nMinor = m_aUIntStack.back();
m_aUIntStack.pop_back();
pFile->m_nMajor = m_aUIntStack.back();
m_aUIntStack.pop_back();
m_aObjectStack.push_back( pFile );
}
else
parseError( "found file header in unusual place", pBegin );
}
void pushComment(const iteratorT& first, const iteratorT& last)
{
// add a comment to the current stack element
PDFComment* pComment =
new PDFComment(iteratorToString(first,last));
if( m_aObjectStack.empty() )
m_aObjectStack.push_back( new PDFPart() );
PDFContainer* pContainer = dynamic_cast<PDFContainer*>(m_aObjectStack.back());
if( pContainer == nullptr )
parseError( "comment without container", first );
pContainer->m_aSubElements.emplace_back( pComment );
}
void insertNewValue( std::unique_ptr<PDFEntry> pNewValue, const iteratorT& pPos )
{
PDFContainer* pContainer = nullptr;
const char* pMsg = nullptr;
if( ! m_aObjectStack.empty() )
{
pContainer = dynamic_cast<PDFContainer*>(m_aObjectStack.back());
if (pContainer)
{
if( dynamic_cast<PDFDict*>(pContainer) == nullptr &&
dynamic_cast<PDFArray*>(pContainer) == nullptr )
{
PDFObject* pObj = dynamic_cast<PDFObject*>(pContainer);
if( pObj )
{
if( pObj->m_pObject == nullptr )
pObj->m_pObject = pNewValue.get();
else
{
pMsg = "second value for object";
pContainer = nullptr;
}
}
else if( dynamic_cast<PDFDict*>(pNewValue.get()) )
{
PDFTrailer* pTrailer = dynamic_cast<PDFTrailer*>(pContainer);
if( pTrailer )
{
if( pTrailer->m_pDict == nullptr )
pTrailer->m_pDict = dynamic_cast<PDFDict*>(pNewValue.get());
else
pContainer = nullptr;
}
else
pContainer = nullptr;
}
else
pContainer = nullptr;
}
}
}
if( pContainer )
pContainer->m_aSubElements.emplace_back( std::move(pNewValue) );
else
{
if( ! pMsg )
{
if( dynamic_cast<PDFContainer*>(pNewValue.get()) )
pMsg = "array without container";
else
pMsg = "value without container";
}
parseError( pMsg, pPos );
}
}
void pushName(const iteratorT& first, const iteratorT& last )
{
insertNewValue( std::make_unique<PDFName>(iteratorToString(first,last)), first );
}
void pushDouble( const iteratorT& first, SAL_UNUSED_PARAMETER const iteratorT& /*last*/ )
{
insertNewValue( std::make_unique<PDFNumber>(m_fDouble), first );
}
void pushString( const iteratorT& first, const iteratorT& last )
{
insertNewValue( std::make_unique<PDFString>(iteratorToString(first,last)), first );
}
void pushBool( const iteratorT& first, const iteratorT& last )
{
insertNewValue( std::make_unique<PDFBool>( last-first == 4 ), first );
}
void pushNull( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
insertNewValue( std::make_unique<PDFNull>(), first );
}
void beginObject( const iteratorT& first, SAL_UNUSED_PARAMETER const iteratorT& /*last*/ )
{
if( m_aObjectStack.empty() )
m_aObjectStack.push_back( new PDFPart() );
unsigned int nGeneration = m_aUIntStack.back();
m_aUIntStack.pop_back();
unsigned int nObject = m_aUIntStack.back();
m_aUIntStack.pop_back();
PDFObject* pObj = new PDFObject( nObject, nGeneration );
pObj->m_nOffset = first - m_aGlobalBegin;
PDFContainer* pContainer = dynamic_cast<PDFContainer*>(m_aObjectStack.back());
if( pContainer &&
( dynamic_cast<PDFFile*>(pContainer) ||
dynamic_cast<PDFPart*>(pContainer) ) )
{
pContainer->m_aSubElements.emplace_back( pObj );
m_aObjectStack.push_back( pObj );
}
else
parseError( "object in wrong place", first );
}
void endObject( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
if( m_aObjectStack.empty() )
parseError( "endobj without obj", first );
else if( dynamic_cast<PDFObject*>(m_aObjectStack.back()) == nullptr )
parseError( "spurious endobj", first );
else
m_aObjectStack.pop_back();
}
void pushObjectRef( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
unsigned int nGeneration = m_aUIntStack.back();
m_aUIntStack.pop_back();
unsigned int nObject = m_aUIntStack.back();
m_aUIntStack.pop_back();
insertNewValue( std::make_unique<PDFObjectRef>(nObject,nGeneration), first );
}
void beginDict( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
PDFDict* pDict = new PDFDict();
pDict->m_nOffset = first - m_aGlobalBegin;
insertNewValue( std::unique_ptr<PDFEntry>(pDict), first );
// will not come here if insertion fails (exception)
m_aObjectStack.push_back( pDict );
}
void endDict( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
PDFDict* pDict = nullptr;
if( m_aObjectStack.empty() )
parseError( "dictionary end without begin", first );
else if( (pDict = dynamic_cast<PDFDict*>(m_aObjectStack.back())) == nullptr )
parseError( "spurious dictionary end", first );
else
m_aObjectStack.pop_back();
PDFEntry* pOffender = pDict->buildMap();
if( pOffender )
{
StringEmitContext aCtx;
aCtx.write( "offending dictionary element: ", 30 );
pOffender->emit( aCtx );
m_aErrorString = aCtx.getString();
parseError( m_aErrorString.getStr(), first );
}
}
void beginArray( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
PDFArray* pArray = new PDFArray();
pArray->m_nOffset = first - m_aGlobalBegin;
insertNewValue( std::unique_ptr<PDFEntry>(pArray), first );
// will not come here if insertion fails (exception)
m_aObjectStack.push_back( pArray );
}
void endArray( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
if( m_aObjectStack.empty() )
parseError( "array end without begin", first );
else if( dynamic_cast<PDFArray*>(m_aObjectStack.back()) == nullptr )
parseError( "spurious array end", first );
else
m_aObjectStack.pop_back();
}
void emitStream(const iteratorT& first, const iteratorT& last)
{
if( m_aObjectStack.empty() )
parseError( "stream without object", first );
PDFObject* pObj = dynamic_cast<PDFObject*>(m_aObjectStack.back());
if( pObj && pObj->m_pObject )
{
if( pObj->m_pStream )
parseError( "multiple streams in object", first );
PDFDict* pDict = dynamic_cast<PDFDict*>(pObj->m_pObject);
if( pDict )
{
PDFStream* pStream = new PDFStream( first - m_aGlobalBegin, last - m_aGlobalBegin, pDict );
pObj->m_pStream = pStream;
pObj->m_aSubElements.emplace_back( pStream );
}
}
else
parseError( "stream without object", first );
}
void beginTrailer( const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
if( m_aObjectStack.empty() )
m_aObjectStack.push_back( new PDFPart() );
PDFTrailer* pTrailer = new PDFTrailer();
pTrailer->m_nOffset = first - m_aGlobalBegin;
PDFContainer* pContainer = dynamic_cast<PDFContainer*>(m_aObjectStack.back());
if( pContainer &&
( dynamic_cast<PDFFile*>(pContainer) ||
dynamic_cast<PDFPart*>(pContainer) ) )
{
pContainer->m_aSubElements.emplace_back( pTrailer );
m_aObjectStack.push_back( pTrailer );
}
else
parseError( "trailer in wrong place", first );
}
void endTrailer(const iteratorT& first, SAL_UNUSED_PARAMETER iteratorT )
{
if( m_aObjectStack.empty() )
parseError( "%%EOF without trailer", first );
else if( dynamic_cast<PDFTrailer*>(m_aObjectStack.back()) == nullptr )
parseError( "spurious %%EOF", first );
else
m_aObjectStack.pop_back();
}
};
}
std::unique_ptr<PDFEntry> PDFReader::read(std::u16string_view aFileName)
{
#ifdef _WIN32
file_iterator<> file_start(std::wstring(o3tl::toW(aFileName)));
#else
file_iterator<> file_start(
std::string(OUStringToOString(aFileName, osl_getThreadTextEncoding())));
#endif
if( ! file_start )
return nullptr;
file_iterator<> file_end = file_start.make_end();
PDFGrammar< file_iterator<> > aGrammar( file_start );
try
{
#if OSL_DEBUG_LEVEL > 0
boost::spirit::classic::parse_info< file_iterator<> > aInfo =
#endif
boost::spirit::classic::parse( file_start,
file_end,
aGrammar,
boost::spirit::classic::space_p );
#if OSL_DEBUG_LEVEL > 0
SAL_INFO("sdext.pdfimport.pdfparse", "parseinfo: stop at offset = " << aInfo.stop - file_start << ", hit = " << (aInfo.hit ? "true" : "false") << ", full = " << (aInfo.full ? "true" : "false") << ", length = " << aInfo.length);
#endif
}
catch( const parser_error< const char*, file_iterator<> >& rError )
{
SAL_WARN("sdext.pdfimport.pdfparse", "parse error: " << rError.descriptor << " at buffer pos " << rError.where - file_start);
#if OSL_DEBUG_LEVEL > 0
OUStringBuffer aTmp;
unsigned int nElem = aGrammar.m_aObjectStack.size();
for( unsigned int i = 0; i < nElem; i++ )
{
aTmp.append(" ");
aTmp.appendAscii(typeid( *(aGrammar.m_aObjectStack[i]) ).name());
}
SAL_WARN("sdext.pdfimport.pdfparse", "parse error object stack: " << aTmp.makeStringAndClear());
#endif
}
std::unique_ptr<PDFEntry> pRet;
unsigned int nEntries = aGrammar.m_aObjectStack.size();
if( nEntries == 1 )
{
pRet.reset(aGrammar.m_aObjectStack.back());
aGrammar.m_aObjectStack.pop_back();
}
else if( nEntries > 1 )
{
// It is possible that there are multiple trailers, which is OK.
// But still keep the warnings, just in case.
SAL_WARN("sdext.pdfimport.pdfparse", "error got " << nEntries << " stack objects in parse");
for (;;)
{
PDFEntry* pEntry = aGrammar.m_aObjectStack.back();
aGrammar.m_aObjectStack.pop_back();
SAL_WARN("sdext.pdfimport.pdfparse", typeid(*pEntry).name());
PDFObject* pObj = dynamic_cast<PDFObject*>(pEntry);
if( pObj )
SAL_WARN("sdext.pdfimport.pdfparse", " -> object " << pObj->m_nNumber << " generation " << pObj->m_nGeneration);
if (aGrammar.m_aObjectStack.empty())
{
pRet.reset(pEntry); // The first entry references all others - see PDFGrammar dtor
break;
}
}
}
return pRet;
}
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
↑ V530 The return value of function 'append' is required to be utilized.
↑ V530 The return value of function 'appendAscii' is required to be utilized.