Java Code Examples for org.apache.pdfbox.cos.COSDictionary#getItem()

/**
 * Returns the resource with the given name and kind as an indirect object, or null.
 */
private COSObject getIndirect(COSName kind, COSName name)
{
    COSDictionary dict = (COSDictionary)resources.getDictionaryObject(kind);
    if (dict == null)
    {
        return null;
    }
    COSBase base = dict.getItem(name);
    if (base instanceof COSObject)
    {
        return (COSObject)base;
    }
    // not an indirect object. Resource may have been added at runtime.
    return null;
}
 

private void addExcludedToList(COSName[] excludeObjects, COSDictionary dict, final Set<Long> parsedObjects)
{
    if (excludeObjects != null)
    {
        for (COSName objName : excludeObjects)
        {
            COSBase baseObj = dict.getItem(objName);
            if (baseObj instanceof COSObject)
            {
                parsedObjects.add(getObjectId((COSObject) baseObj));
            }
        }
    }
}
 

/**
 * Returns the base state for optional content groups.
 * @return the base state
 */
public BaseState getBaseState()
{
    COSDictionary d = getD();
    COSName name = (COSName)d.getItem(COSName.BASE_STATE);
    return BaseState.valueOf(name);
}
 

private void assignAcroFormDefaultResource(PDAcroForm acroForm, COSDictionary newDict)
{
    // read and set/update AcroForm default resource dictionary /DR if available
    COSBase newBase = newDict.getDictionaryObject(COSName.DR);
    if (newBase instanceof COSDictionary)
    {
        COSDictionary newDR = (COSDictionary) newBase;
        PDResources defaultResources = acroForm.getDefaultResources();
        if (defaultResources == null)
        {
            acroForm.getCOSObject().setItem(COSName.DR, newDR);
            newDR.setDirect(true);
            newDR.setNeedToBeUpdated(true);            
        }
        else
        {
            COSDictionary oldDR = defaultResources.getCOSObject();
            COSBase newXObjectBase = newDR.getItem(COSName.XOBJECT);
            COSBase oldXObjectBase = oldDR.getItem(COSName.XOBJECT);
            if (newXObjectBase instanceof COSDictionary &&
                oldXObjectBase instanceof COSDictionary)
            {
                ((COSDictionary) oldXObjectBase).addAll((COSDictionary) newXObjectBase);
                oldDR.setNeedToBeUpdated(true);
            }
        }
    }
}
 

/**
 * This will decrypt a dictionary.
 *
 * @param dictionary The dictionary to decrypt.
 * @param objNum The object number.
 * @param genNum The object generation number.
 *
 * @throws IOException If there is an error creating a new string.
 */
private void decryptDictionary(COSDictionary dictionary, long objNum, long genNum) throws IOException
{
    if (dictionary.getItem(COSName.CF) != null)
    {
        // PDFBOX-2936: avoid orphan /CF dictionaries found in US govt "I-" files
        return;
    }
    COSBase type = dictionary.getDictionaryObject(COSName.TYPE);
    boolean isSignature = COSName.SIG.equals(type) || COSName.DOC_TIME_STAMP.equals(type) ||
            // PDFBOX-4466: /Type is optional, see
            // https://ec.europa.eu/cefdigital/tracker/browse/DSS-1538
            (dictionary.getDictionaryObject(COSName.CONTENTS) instanceof COSString && 
             dictionary.getDictionaryObject(COSName.BYTERANGE) instanceof COSArray);
    for (Map.Entry<COSName, COSBase> entry : dictionary.entrySet())
    {
        if (isSignature && COSName.CONTENTS.equals(entry.getKey()))
        {
            // do not decrypt the signature contents string
            continue;
        }
        COSBase value = entry.getValue();
        // within a dictionary only the following kind of COS objects have to be decrypted
        if (value instanceof COSString || value instanceof COSArray || value instanceof COSDictionary)
        {
            decrypt(value, objNum, genNum);
        }
    }
}
 

/**
 * <a href="https://stackoverflow.com/questions/45812696/pdfbox-delete-comment-maintain-strikethrough">
 * PDFBox delete comment maintain strikethrough
 * </a>
 * <br/>
 * <a href="https://expirebox.com/files/3d955e6df4ca5874c38dbf92fc43b5af.pdf">
 * only_fields.pdf
 * </a>
 * <a href="https://file.io/DTvqhC">
 * (alternative download)
 * </a>
 * <p>
 * The OP only wanted the comment removed, not the strike-through. Thus, we must
 * not remove the annotation but merely the comment building attributes.
 * </p>
 */
@Test
public void testRemoveLikeStephanImproved() throws IOException {
    final COSName POPUP = COSName.getPDFName("Popup");
    try (InputStream resource = getClass().getResourceAsStream("only_fields.pdf")) {
        PDDocument document = Loader.loadPDF(resource);
        List<PDAnnotation> annotations = new ArrayList<>();
        PDPageTree allPages = document.getDocumentCatalog().getPages();

        List<COSObjectable> objectsToRemove = new ArrayList<>();

        for (int i = 0; i < allPages.getCount(); i++) {
            PDPage page = allPages.get(i);
            annotations = page.getAnnotations();

            for (PDAnnotation annotation : annotations) {
                if ("StrikeOut".equals(annotation.getSubtype()))
                {
                    COSDictionary annotationDict = annotation.getCOSObject();
                    COSBase popup = annotationDict.getItem(POPUP);
                    annotationDict.removeItem(POPUP);
                    annotationDict.removeItem(COSName.CONTENTS); // plain text comment
                    annotationDict.removeItem(COSName.RC);       // rich text comment
                    annotationDict.removeItem(COSName.T);        // author

                    if (popup != null)
                        objectsToRemove.add(popup);
                }
            }

            annotations.removeAll(objectsToRemove);
        }

        document.save(new File(RESULT_FOLDER, "only_fields-removeImproved.pdf"));
    }
}
 

/**
 * <a href="https://stackoverflow.com/questions/53420344/ho-to-reduce-the-size-of-merged-pdf-a1-b-files-with-pdfbox-or-other-java-library">
 * Ho to reduce the size of merged PDF A1/b Files with pdfbox or other java library
 * </a>
 * <br/>
 * <a href="https://datentransfer.sparkassenverlag.de/my/transfers/5q8eskgne52npemx8kid7728zk1hq3f993dfat8his">
 * dummy.pdf
 * </a>
 * <p>
 * This is the code the OP himself posted as his solution. This only works if
 * (a) all font programs embedded for the same name indeed are identical, and
 * if (b) all fonts to consider are in the immediate page resources, not the
 * resources of some referred to xobject or pattern. If those conditions are
 * fulfilled, though, it very likely is much faster than the approach in
 * {@link #optimize(PDDocument)}. For the example file provided by the OP,
 * its result is nearly as small.
 * </p>
 */
@Test
public void testOptimizeLikeSchowaveDummy() throws IOException {
    try (   InputStream resource = getClass().getResourceAsStream("dummy.pdf")  ) {
        PDDocument doc = Loader.loadPDF(resource);

        Map<String, COSBase> fontFileCache = new HashMap<>();
        for (int pageNumber = 0; pageNumber < doc.getNumberOfPages(); pageNumber++) {
            final PDPage page = doc.getPage(pageNumber);
            COSDictionary pageDictionary = (COSDictionary) page.getResources().getCOSObject().getDictionaryObject(COSName.FONT);
            for (COSName currentFont : pageDictionary.keySet()) {
                COSDictionary fontDictionary = (COSDictionary) pageDictionary.getDictionaryObject(currentFont);
                for (COSName actualFont : fontDictionary.keySet()) {
                    COSBase actualFontDictionaryObject = fontDictionary.getDictionaryObject(actualFont);
                    if (actualFontDictionaryObject instanceof COSDictionary) {
                        COSDictionary fontFile = (COSDictionary) actualFontDictionaryObject;
                        if (fontFile.getItem(COSName.FONT_NAME) instanceof COSName) {
                            COSName fontName = (COSName) fontFile.getItem(COSName.FONT_NAME);
                            fontFileCache.computeIfAbsent(fontName.getName(), key -> fontFile.getItem(COSName.FONT_FILE2));
                            fontFile.setItem(COSName.FONT_FILE2, fontFileCache.get(fontName.getName()));
                        }
                    }
                }
            }
        }

        doc.save(new File(RESULT_FOLDER, "dummy-optimized-like-schowave.pdf"));
    }
}
 

/**
 * Read the trailer information and provide a COSDictionary containing the trailer information.
 * 
 * @return a COSDictionary containing the trailer information
 * @throws IOException if something went wrong
 */
protected COSDictionary retrieveTrailer() throws IOException
{
    COSDictionary trailer = null;
    boolean rebuildTrailer = false;
    try
    {
        // parse startxref
        // TODO FDF files don't have a startxref value, so that rebuildTrailer is triggered
        long startXRefOffset = getStartxrefOffset();
        if (startXRefOffset > -1)
        {
            trailer = parseXref(startXRefOffset);
        }
        else
        {
            rebuildTrailer = isLenient();
        }
    }
    catch (IOException exception)
    {
        if (isLenient())
        {
            rebuildTrailer = true;
        }
        else
        {
            throw exception;
        }
    }
    // check if the trailer contains a Root object
    if (trailer != null && trailer.getItem(COSName.ROOT) == null)
    {
        rebuildTrailer = isLenient();
    }
    if (rebuildTrailer)
    {
        trailer = rebuildTrailer();
    }
    else
    {
        // prepare decryption if necessary
        prepareDecryption();
        if (bfSearchCOSObjectKeyOffsets != null && !bfSearchCOSObjectKeyOffsets.isEmpty())
        {
            bfSearchForObjStreams();
        }
    }
    return trailer;
}
 

/**
 * Update the Pg and Obj references to the new (merged) page.
 *
 * @param parentTreeEntry
 * @param objMapping mapping between old and new references
 */
private void updatePageReferences(PDFCloneUtility cloner,
        COSDictionary parentTreeEntry, Map<COSDictionary, COSDictionary> objMapping)
        throws IOException
{
    COSDictionary pageDict = parentTreeEntry.getCOSDictionary(COSName.PG);
    if (objMapping.containsKey(pageDict))
    {
        parentTreeEntry.setItem(COSName.PG, objMapping.get(pageDict));
    }
    COSBase obj = parentTreeEntry.getDictionaryObject(COSName.OBJ);
    if (obj instanceof COSDictionary)
    {
        COSDictionary objDict = (COSDictionary) obj;
        if (objMapping.containsKey(objDict))
        {
            parentTreeEntry.setItem(COSName.OBJ, objMapping.get(objDict));
        }
        else
        {
            // PDFBOX-3999: clone objects that are not in mapping to make sure that
            // these don't remain attached to the source document
            COSBase item = parentTreeEntry.getItem(COSName.OBJ);
            if (item instanceof COSObject)
            {
                LOG.debug("clone potential orphan object in structure tree: " + item +
                        ", Type: " + objDict.getNameAsString(COSName.TYPE) +
                        ", Subtype: " + objDict.getNameAsString(COSName.SUBTYPE) +
                        ", T: " + objDict.getNameAsString(COSName.T));
            }
            else
            {
                // don't display in full because of stack overflow
                LOG.debug("clone potential orphan object in structure tree" +
                        ", Type: " + objDict.getNameAsString(COSName.TYPE) +
                        ", Subtype: " + objDict.getNameAsString(COSName.SUBTYPE) +
                        ", T: " + objDict.getNameAsString(COSName.T));
            }
            parentTreeEntry.setItem(COSName.OBJ, cloner.cloneForNewDocument(obj));
        }
    }
    COSBase kSubEntry = parentTreeEntry.getDictionaryObject(COSName.K);
    if (kSubEntry instanceof COSArray)
    {
        updatePageReferences(cloner, (COSArray) kSubEntry, objMapping);
    }
    else if (kSubEntry instanceof COSDictionary)
    {
        updatePageReferences(cloner, (COSDictionary) kSubEntry, objMapping);
    }
}