@Override
public void transform(ArchivalUnit au, PdfDocument pdfDocument) throws PdfException {
pdfDocument.unsetModificationDate();
PdfUtil.normalizeTrailerId(pdfDocument);
pdfDocument.unsetMetadata();
PDDocumentInformation pdDocInfo = ((GtvPdfBoxDocument) pdfDocument).getPdDocumentInformation();
if (pdDocInfo.getCustomMetadataValue(GtvPdfBoxDocument.PDFDATE) != null) {
pdDocInfo.setCustomMetadataValue(GtvPdfBoxDocument.PDFDATE, null);
}
if (pdDocInfo.getCustomMetadataValue(GtvPdfBoxDocument.PDFUSER) != null) {
pdDocInfo.setCustomMetadataValue(GtvPdfBoxDocument.PDFUSER, null);
}
PdfStateMachineWorker worker = new PdfStateMachineWorker();
boolean anyXform = false;
for (PdfPage pdfPage : pdfDocument.getPages()) {
PdfTokenStream pdfTokenStream = pdfPage.getPageTokenStream();
worker.process(pdfTokenStream);
if (worker.getResult()) {
anyXform = true;
List<PdfToken> tokens = pdfTokenStream.getTokens();
// clear tokens including text markers
tokens.subList(worker.getBegin(), worker.getEnd() + 1).clear();
pdfTokenStream.setTokens(tokens);
}
}
if (log.isDebug2()) {
log.debug2("Transform: " + anyXform);
}
}
/**
* Adds a blank page.
*
* @param rect The page dimension
* @param rotation The rotation angle in degrees
* @since 2.1.5
*/
public void addPage(Rectangle rect, int rotation) {
PdfRectangle mediabox = new PdfRectangle(rect, rotation);
PageResources resources = new PageResources();
PdfPage page = new PdfPage(mediabox, new HashMap(), resources.getResources(), 0);
page.put(PdfName.TABS, getTabs());
root.addPage(page);
++currentPageNumber;
}
/**
* Rewrites the contents of the PDF page being transformed with those of the output list, if at
* least one change has been indicated.
*
* @param pdfPage A PDF page (belonging to the PDF document).
* @throws IOException if any processing error occurs.
*/
protected synchronized void writeResult(PdfPage pdfPage) throws IOException {
if (atLeastOneChange) {
logger.debug3("Writing result to PDF page");
PDStream resultStream = pdfPage.getPdfDocument().makePdStream();
OutputStream outputStream = resultStream.createOutputStream();
ContentStreamWriter tokenWriter = new ContentStreamWriter(outputStream);
tokenWriter.writeTokens(getOutputList());
pdfPage.setContents(resultStream);
} else {
logger.debug3("No result to write to PDF page");
}
}
/**
* Applies this transform to a PDF page.
*
* <p>Using the configured PDF operator processors, a PDF stream engine ({@link PDFStreamEngine}
* parent class) iterates through the page's token stream, aggregating operands and invoking the
* appropriate PDF operator processor for each PDF operator encountered. An output list of tokens
* is created. At the end of the iteration, if any PDF operator processor has indicated to this
* transform that it has changed the output list compared to the original stream, the PDF page's
* contents are replaced with those of the output list (otherwise the PDF page is unchanged).
*
* @param pdfPage A PDF page (belonging to the PDF document).
* @throws PageTransformException if the output list stack does not have exactly one output list
* at the end of the transform.
* @throws IOException if any other processing error occurs.
* @see PDFStreamEngine#processStream
* @see #reset
*/
public synchronized boolean transform(PdfPage pdfPage) throws IOException {
logger.debug3("Begin page stream transform");
// Iterate over stream
reset();
this.currentPdfPage = pdfPage;
processStream(pdfPage.getPdPage(), pdfPage.findResources(), pdfPage.getContentStream());
// Sanity check
if (listStack.size() != 1) {
String logMessage =
"Split/merge mismatch: after processing stream, list stack has size " + listStack.size();
logger.error(logMessage);
throw new PageTransformException(logMessage);
}
writeResult(pdfPage);
logger.debug2("Page stream transform result: " + atLeastOneChange);
return atLeastOneChange;
}