List of usage examples for com.itextpdf.text.pdf PdfDictionary getAsNumber
public PdfNumber getAsNumber(final PdfName key)
PdfObject as a PdfNumber, resolving indirect references. From source file:mkl.testarea.itext5.pdfcleanup.StrictPdfCleanUpProcessor.java
License:Open Source License
/** * Deletes redact annotations from the page and substitutes them with either OverlayText or RO object if it's needed. *//* www . j a v a2 s. c o m*/ private void deleteRedactAnnots(int pageNum) throws IOException, DocumentException { Set<String> indirRefs = redactAnnotIndirRefs.get(pageNum); if (indirRefs == null || indirRefs.isEmpty()) { return; } PdfReader reader = pdfStamper.getReader(); PdfContentByte canvas = pdfStamper.getOverContent(pageNum); PdfDictionary pageDict = reader.getPageN(pageNum); PdfArray annotsArray = pageDict.getAsArray(PdfName.ANNOTS); // j is for access annotRect (i can be decreased, so we need to store additional index, // indicating current position in ANNOTS array in case if we don't remove anything for (int i = 0, j = 0; i < annotsArray.size(); ++i, ++j) { PdfIndirectReference annotIndRef = annotsArray.getAsIndirectObject(i); PdfDictionary annotDict = annotsArray.getAsDict(i); if (indirRefs.contains(annotIndRef.toString()) || indirRefs.contains(getParentIndRefStr(annotDict))) { PdfStream formXObj = annotDict.getAsStream(PdfName.RO); PdfString overlayText = annotDict.getAsString(PdfName.OVERLAYTEXT); if (fillCleanedArea && formXObj != null) { PdfArray rectArray = annotDict.getAsArray(PdfName.RECT); Rectangle annotRect = new Rectangle(rectArray.getAsNumber(0).floatValue(), rectArray.getAsNumber(1).floatValue(), rectArray.getAsNumber(2).floatValue(), rectArray.getAsNumber(3).floatValue()); insertFormXObj(canvas, pageDict, formXObj, clippingRects.get(j), annotRect); } else if (fillCleanedArea && overlayText != null && overlayText.toUnicodeString().length() > 0) { drawOverlayText(canvas, clippingRects.get(j), overlayText, annotDict.getAsString(PdfName.DA), annotDict.getAsNumber(PdfName.Q), annotDict.getAsBoolean(PdfName.REPEAT)); } annotsArray.remove(i--); // array size is changed, so we need to decrease i } } if (annotsArray.size() == 0) { pageDict.remove(PdfName.ANNOTS); } }
From source file:org.gmdev.pdftrick.utils.CustomExtraImgReader.java
License:Open Source License
/** * Read a png image with if all other method fails * @param ref//from ww w.j a v a 2 s.c om * @param resultFile * @return The BufferedImage obj * @throws IOException * @throws ImageReadException */ public static BufferedImage readIndexedPNG(int ref, String resultFile) throws IOException, ImageReadException { PdfReader reader = new PdfReader(resultFile); PRStream stream = (PRStream) reader.getPdfObject(ref); PdfDictionary dic = stream; byte[] content = PdfReader.getStreamBytesRaw(stream); int width = dic.getAsNumber(PdfName.WIDTH).intValue(); int height = dic.getAsNumber(PdfName.HEIGHT).intValue(); int pngBitDepth = dic.getAsNumber(PdfName.BITSPERCOMPONENT).intValue(); PdfObject colorspace = dic.getDirectObject(PdfName.COLORSPACE); PdfArray decode = dic.getAsArray(PdfName.DECODE); PdfArray carray = (PdfArray) colorspace; PdfObject id2 = carray.getDirectObject(3); byte[] palette = null; if (id2 instanceof PdfString) { palette = ((PdfString) id2).getBytes(); } else if (id2 instanceof PRStream) { palette = PdfReader.getStreamBytes(((PRStream) id2)); } Map<PdfName, FilterHandlers.FilterHandler> handlers = new HashMap<PdfName, FilterHandlers.FilterHandler>( FilterHandlers.getDefaultFilterHandlers()); byte[] imageBytes = PdfReader.decodeBytes(content, dic, handlers); int stride = (width * pngBitDepth + 7) / 8; ByteArrayOutputStream ms = new ByteArrayOutputStream(); PngWriter png = new PngWriter(ms); if (decode != null) { if (pngBitDepth == 1) { // if the decode array is 1,0, then we need to invert the image if (decode.getAsNumber(0).intValue() == 1 && decode.getAsNumber(1).intValue() == 0) { int len = imageBytes.length; for (int t = 0; t < len; ++t) { imageBytes[t] ^= 0xff; } } else { // if the decode array is 0,1, do nothing. It's possible that the array could be 0,0 or 1,1 - but that would be silly, so we'll just ignore that case } } else { // todo: add decode transformation for other depths } } int pngColorType = 0; png.writeHeader(width, height, pngBitDepth, pngColorType); if (palette != null) { png.writePalette(palette); } png.writeData(imageBytes, stride); png.writeEnd(); imageBytes = ms.toByteArray(); InputStream in = new ByteArrayInputStream(imageBytes); ImageInputStream ima_stream = ImageIO.createImageInputStream(in); BufferedImage buffImg = null; BufferedImage buffPic = ImageIO.read(ima_stream); // check if image contains a mask image ... experimental for this type of image BufferedImage buffMask = null; PRStream maskStream = (PRStream) dic.getAsStream(PdfName.SMASK); if (maskStream != null) { PdfImageObject maskImage = new PdfImageObject(maskStream); buffMask = maskImage.getBufferedImage(); Image img = PdfTrickUtils.TransformGrayToTransparency(buffMask); buffImg = PdfTrickUtils.ApplyTransparency(buffPic, img); } else { buffImg = buffPic; } reader.close(); ms.close(); in.close(); return buffImg; }
From source file:uk.bl.wa.tika.parser.pdf.itext.PDFParser.java
License:Apache License
private static void extractMetadata(PdfReader reader, Metadata metadata) { try {//from w ww .jav a 2 s. c o m HashMap<String, String> map = reader.getInfo(); // Clone the PDF info: for (String key : map.keySet()) { metadata.set(key.toLowerCase(), map.get(key)); } // Add other data of interest: metadata.set("pdf:version", "1." + reader.getPdfVersion()); metadata.set("pdf:numPages", "" + reader.getNumberOfPages()); metadata.set("pdf:cryptoMode", "" + getCryptoModeAsString(reader)); metadata.set("pdf:openedWithFullPermissions", "" + reader.isOpenedWithFullPermissions()); metadata.set("pdf:encrypted", "" + reader.isEncrypted()); metadata.set("pdf:metadataEncrypted", "" + reader.isMetadataEncrypted()); metadata.set("pdf:128key", "" + reader.is128Key()); metadata.set("pdf:tampered", "" + reader.isTampered()); // Also grap XMP metadata, if present: byte[] xmpmd = reader.getMetadata(); if (xmpmd != null) { // This is standard Tika code for parsing standard stuff from the XMP: JempboxExtractor extractor = new JempboxExtractor(metadata); extractor.parse(new ByteArrayInputStream(xmpmd)); // This is custom XMP-handling code: XMPMetadata xmp = XMPMetadata.load(new ByteArrayInputStream(xmpmd)); // There is a special class for grabbing data in the PDF schema - not sure it will add much here: // Could parse xmp:CreatorTool and pdf:Producer etc. etc. out of here. //XMPSchemaPDF pdfxmp = xmp.getPDFSchema(); // Added a PDF/A schema class: xmp.addXMLNSMapping(XMPSchemaPDFA.NAMESPACE, XMPSchemaPDFA.class); XMPSchemaPDFA pdfaxmp = (XMPSchemaPDFA) xmp.getSchemaByClass(XMPSchemaPDFA.class); if (pdfaxmp != null) { metadata.set("pdfaid:part", pdfaxmp.getPart()); metadata.set("pdfaid:conformance", pdfaxmp.getConformance()); String version = "A-" + pdfaxmp.getPart() + pdfaxmp.getConformance().toLowerCase(); //metadata.set("pdfa:version", version ); metadata.set("pdf:version", version); } } // Attempt to determine Adobe extension level: PdfDictionary extensions = reader.getCatalog().getAsDict(PdfName.EXTENSIONS); if (extensions != null) { PdfDictionary adobeExt = extensions.getAsDict(PdfName.ADBE); if (adobeExt != null) { PdfName baseVersion = adobeExt.getAsName(PdfName.BASEVERSION); int el = adobeExt.getAsNumber(PdfName.EXTENSIONLEVEL).intValue(); metadata.set("pdf:version", baseVersion.toString().substring(1) + " Adobe Extension Level " + el); } } // Ensure the normalised metadata are mapped in: if (map.get("Title") != null) metadata.set(Metadata.TITLE, map.get("Title")); if (map.get("Author") != null) metadata.set(Metadata.AUTHOR, map.get("Author")); } catch (Exception e) { System.err.println("PDFParser.extractMetadata() caught Exception: " + e.getMessage()); e.printStackTrace(); } }