KoFilter::ConversionStatus PdfImport::convert(const QByteArray& from, const QByteArray& to)
{
debugPdf << "to:" << to << " from:" << from;
if (from != "application/pdf" || to != "image/svg+xml") {
return KoFilter::NotImplemented;
}
// read config file
globalParams = new GlobalParams();
if (! globalParams)
return KoFilter::NotImplemented;
GooString * fname = new GooString(QFile::encodeName(m_chain->inputFile()).data());
PDFDoc * pdfDoc = new PDFDoc(fname, 0, 0, 0);
if (! pdfDoc) {
delete globalParams;
return KoFilter::StupidError;
}
if (! pdfDoc->isOk()) {
delete globalParams;
delete pdfDoc;
return KoFilter::StupidError;
}
double hDPI = 72.0;
double vDPI = 72.0;
int firstPage = 1;
int lastPage = pdfDoc->getNumPages();
debugPdf << "converting pages" << firstPage << "-" << lastPage;
SvgOutputDev * dev = new SvgOutputDev(m_chain->outputFile());
if (dev->isOk()) {
int rotate = 0;
GBool useMediaBox = gTrue;
GBool crop = gFalse;
GBool printing = gFalse;
pdfDoc->displayPages(dev, firstPage, lastPage, hDPI, vDPI, rotate, useMediaBox, crop, printing);
dev->dumpContent();
}
debugPdf << "wrote file to" << m_chain->outputFile();
delete dev;
delete pdfDoc;
delete globalParams;
globalParams = 0;
// check for memory leaks
Object::memCheck(stderr);
return KoFilter::OK;
}
int main(int argc, char *argv[])
{
// parse args
bool ok = parseArgs(argDesc, &argc, argv);
if (!ok || argc < 2 || argc > 3 || printHelp) {
fprintf(stderr, "pdftoipe version %s\n", PDFTOIPE_VERSION);
printUsage("pdftoipe", "<PDF-file> [<XML-file>]", argDesc);
return 1;
}
GooString *fileName = new GooString(argv[1]);
globalParams = new GlobalParams();
if (quiet)
globalParams->setErrQuiet(quiet);
GooString *ownerPW, *userPW;
if (ownerPassword[0]) {
ownerPW = new GooString(ownerPassword);
} else {
ownerPW = 0;
}
if (userPassword[0]) {
userPW = new GooString(userPassword);
} else {
userPW = 0;
}
// open PDF file
PDFDoc *doc = new PDFDoc(fileName, ownerPW, userPW);
delete userPW;
delete ownerPW;
if (!doc->isOk())
return 1;
// construct XML file name
std::string xmlFileName;
if (argc == 3) {
xmlFileName = argv[2];
} else {
const char *p = fileName->c_str() + fileName->getLength() - 4;
if (!strcmp(p, ".pdf") || !strcmp(p, ".PDF")) {
xmlFileName = std::string(fileName->c_str(),
fileName->getLength() - 4);
} else {
xmlFileName = fileName->c_str();
}
xmlFileName += ".ipe";
}
// get page range
if (firstPage < 1)
firstPage = 1;
if (lastPage < 1 || lastPage > doc->getNumPages())
lastPage = doc->getNumPages();
// write XML file
XmlOutputDev *xmlOut =
new XmlOutputDev(xmlFileName, doc->getXRef(),
doc->getCatalog(), firstPage, lastPage);
// tell output device about text handling
xmlOut->setTextHandling(math, notext, literal, mergeLevel, unicodeLevel);
int exitCode = 2;
if (xmlOut->isOk()) {
doc->displayPages(xmlOut, firstPage, lastPage,
// double hDPI, double vDPI, int rotate,
// bool useMediaBox, bool crop, bool printing,
72.0, 72.0, 0, false, false, false);
exitCode = 0;
}
if (xmlOut->hasUnicode()) {
fprintf(stderr, "The document contains Unicode (non-ASCII) text.\n");
if (unicodeLevel <= 1)
fprintf(stderr, "Unknown Unicode characters were replaced by [U+XXX].\n");
else
fprintf(stderr, "UTF-8 was set as document encoding in the preamble.\n");
}
// clean up
delete xmlOut;
delete doc;
delete globalParams;
return exitCode;
}
indri::parse::UnparsedDocument* indri::parse::PDFDocumentExtractor::nextDocument() {
if( !_documentPath.length() )
return 0;
PDFDoc* doc = 0;
TextOutputDev* textOut = 0;
GString* gfilename = new GString(_documentPath.c_str());
doc = new PDFDoc( gfilename );
// if the doc is not ok, or ok to copy, it
// will be a document of length 0.
if( doc->isOk() && doc->okToCopy() ) {
void* stream = &_documentTextBuffer;
textOut = new TextOutputDev( buffer_write, stream, gFalse, gFalse);
if ( textOut->isOk() ) {
int firstPage = 1;
int lastPage = doc->getNumPages();
double hDPI=72.0;
double vDPI=72.0;
int rotate=0;
GBool useMediaBox=gFalse;
GBool crop=gTrue;
GBool printing=gFalse;
if(doc->readMetadata()!=NULL)
{
GString rawMetaData = doc->readMetadata();
GString preparedMetaData="";
//zoek <rdf:RDF en eindig bij </rdf:RDF>!!
for(int x=0; x<rawMetaData.getLength(); x++) {
if(rawMetaData.getChar(x)!='?' && rawMetaData.getChar(x)!=':') {
//skip characters which the XMLReader doesn't understand
preparedMetaData.append(rawMetaData.getChar(x));
}
}
std::string metaData(preparedMetaData.getCString());
int startbegin = metaData.find("<rdf");
int stopend = metaData.find(">", metaData.rfind("</rdf") );
metaData = metaData.substr(startbegin, (stopend-startbegin)+1 );
indri::xml::XMLReader reader;
try {
std::auto_ptr<indri::xml::XMLNode> result( reader.read( metaData.c_str() ) );
appendPdfMetaData( result.get() );
} catch( lemur::api::Exception& e ) {
LEMUR_RETHROW( e, "Had trouble reading PDF metadata" );
}
if( _author.length()>0 || _title.length()>0 )
{
std::string createdPdfHeader;
createdPdfHeader="<head>\n";
if(_title.length()>0) {
createdPdfHeader+="<title>";
createdPdfHeader+=_title;
createdPdfHeader+="</title>\n";
}
if(_author.length()>0) {
createdPdfHeader+="<author>";
createdPdfHeader+=_author;
createdPdfHeader+="</author>\n";
}
createdPdfHeader+="</head>\n";
char *metastream = _documentTextBuffer.write( createdPdfHeader.length()+1 );
strcpy(metastream, createdPdfHeader.c_str());
}
}
doc->displayPages(textOut, firstPage, lastPage, hDPI, vDPI, rotate, useMediaBox, crop, printing);
}
}
delete textOut;
delete doc;
_unparsedDocument.textLength = _documentTextBuffer.position();
_unparsedDocument.contentLength = _unparsedDocument.textLength ? _documentTextBuffer.position() - 1 : 0 ; // no null 0 if text is empty.
char* docnoPoint = _documentTextBuffer.write( _documentPath.length()+1 );
strcpy( docnoPoint, _documentPath.c_str() );
_unparsedDocument.text = _documentTextBuffer.front();
_unparsedDocument.content = _documentTextBuffer.front();
_unparsedDocument.metadata.clear();
indri::parse::MetadataPair pair;
pair.key = "path";
pair.value = docnoPoint;
pair.valueLength = _documentPath.length()+1;
_unparsedDocument.metadata.push_back( pair );
_docnostring.assign(_documentPath.c_str() );
cleanDocno();
pair.value = _docnostring.c_str();
pair.valueLength = _docnostring.length()+1;
pair.key = "docno";
_unparsedDocument.metadata.push_back( pair );
_documentPath = "";
return &_unparsedDocument;
}