From 06b900642342afb680aab3b36fba9247e1e29aa9 Mon Sep 17 00:00:00 2001 From: pdanner Date: Fri, 26 Nov 2010 12:02:47 +0000 Subject: Placeholder-image handling git-svn-id: https://joinup.ec.europa.eu/svn/pdf-as/trunk@613 7b5415b0-85f9-ee4d-85bd-d5d0c3b42d1c --- .../placeholder/SignaturePlaceholderExtractor.java | 299 +++++++++++++++++++++ 1 file changed, 299 insertions(+) create mode 100644 src/main/java/at/gv/egiz/pdfas/placeholder/SignaturePlaceholderExtractor.java (limited to 'src/main/java/at/gv/egiz/pdfas/placeholder/SignaturePlaceholderExtractor.java') diff --git a/src/main/java/at/gv/egiz/pdfas/placeholder/SignaturePlaceholderExtractor.java b/src/main/java/at/gv/egiz/pdfas/placeholder/SignaturePlaceholderExtractor.java new file mode 100644 index 0000000..50c3281 --- /dev/null +++ b/src/main/java/at/gv/egiz/pdfas/placeholder/SignaturePlaceholderExtractor.java @@ -0,0 +1,299 @@ +package at.gv.egiz.pdfas.placeholder; + +import java.awt.geom.AffineTransform; +import java.awt.geom.NoninvertibleTransformException; +import java.awt.image.BufferedImage; +import java.io.IOException; +import java.io.InputStream; +import java.util.Hashtable; +import java.util.Iterator; +import java.util.List; +import java.util.Map; +import java.util.Vector; + +import org.apache.commons.logging.Log; +import org.apache.commons.logging.LogFactory; +import org.apache.pdfbox.cos.COSName; +import org.apache.pdfbox.exceptions.WrappedIOException; +import org.apache.pdfbox.pdmodel.PDDocument; +import org.apache.pdfbox.pdmodel.PDPage; +import org.apache.pdfbox.pdmodel.graphics.xobject.PDXObject; +import org.apache.pdfbox.pdmodel.graphics.xobject.PDXObjectImage; +import org.apache.pdfbox.util.Matrix; +import org.apache.pdfbox.util.PDFOperator; +import org.apache.pdfbox.util.PDFStreamEngine; +import org.apache.pdfbox.util.ResourceLoader; + +import at.gv.egiz.pdfas.api.sign.SignParameters; +import at.gv.egiz.pdfas.exceptions.ErrorCode; +import at.gv.egiz.pdfas.exceptions.framework.PlaceholderExtractionException; +import at.gv.egiz.pdfas.utils.PDFASUtils; +import at.knowcenter.wag.egov.egiz.exceptions.PDFDocumentException; +import at.knowcenter.wag.egov.egiz.pdf.TablePos; + +import com.google.zxing.BarcodeFormat; +import com.google.zxing.BinaryBitmap; +import com.google.zxing.DecodeHintType; +import com.google.zxing.LuminanceSource; +import com.google.zxing.MultiFormatReader; +import com.google.zxing.NotFoundException; +import com.google.zxing.ReaderException; +import com.google.zxing.Result; +import com.google.zxing.client.j2se.BufferedImageLuminanceSource; +import com.google.zxing.common.HybridBinarizer; + +/** + * Extract all relevant information from a placeholder image. + * + * @author exthex + * + */ +public class SignaturePlaceholderExtractor extends PDFStreamEngine { + /** + * The log. + */ + private static Log log = LogFactory.getLog(SignaturePlaceholderExtractor.class); + + private List placeholders = new Vector(); + private int currentPage = 0; + + private SignaturePlaceholderExtractor(String placeholderId, int placeholderMatchMode) throws IOException { + super(ResourceLoader.loadProperties("at/gv/egiz/pdfas/placeholder/pdfbox-reader.properties", + true)); + } + + /** + * Search the document for placeholder images and possibly included + * additional info.
+ * Searches only for the first placeholder page after page from top. + * + * @param inputStream + * @return all available info from the first found placeholder. + * @throws PDFDocumentException if the document could not be read. + * @throws PlaceholderExtractionException if STRICT matching mode was requested and no suitable placeholder could be found. + */ + public static SignaturePlaceholderData extract(InputStream inputStream, String placeholderId, int matchMode) + throws PDFDocumentException, PlaceholderExtractionException { + SignaturePlaceholderContext.setSignaturePlaceholderData(null); + PDDocument doc = null; + try + { + try { + doc = PDDocument.load(inputStream); + } catch (IOException e) { + throw new PDFDocumentException(ErrorCode.DOCUMENT_CANNOT_BE_READ, e); + } + PDFASUtils.checkDocumentPermissions(doc); + + SignaturePlaceholderExtractor extractor; + try + { + extractor = new SignaturePlaceholderExtractor(placeholderId, matchMode); + } catch (IOException e2) { + throw new PDFDocumentException(ErrorCode.DOCUMENT_CANNOT_BE_READ, e2); + } + List pages = doc.getDocumentCatalog().getAllPages(); + Iterator iter = pages.iterator(); + int pageNr = 0; + while (iter.hasNext()) { + pageNr++; + PDPage page = (PDPage) iter.next(); + try { + extractor.setCurrentPage(pageNr); + extractor.processStream( page, page.findResources(), page.getContents().getStream() ); + SignaturePlaceholderData ret = matchPlaceholderPage(extractor.placeholders, placeholderId, matchMode); + if (ret != null){ + SignaturePlaceholderContext.setSignaturePlaceholderData(ret); + return ret; + } + } catch (IOException e1) { + throw new PDFDocumentException(ErrorCode.DOCUMENT_CANNOT_BE_READ, e1); + } + + } + if (extractor.placeholders.size() > 0){ + SignaturePlaceholderData ret = matchPlaceholderDocument(extractor.placeholders, placeholderId, matchMode); + SignaturePlaceholderContext.setSignaturePlaceholderData(ret); + return ret; + } + return null; + }finally{ + if (doc != null) + try { + doc.close(); + } catch (IOException e) { + log.debug("Could not close document.", e); + } + } + + } + + private static SignaturePlaceholderData matchPlaceholderDocument( + List placeholders, String placeholderId, int matchMode) throws PlaceholderExtractionException { + + if (matchMode == SignParameters.PLACEHOLDER_MATCH_MODE_STRICT) + throw new PlaceholderExtractionException(ErrorCode.SIGNATURE_PLACEHOLDER_EXTRACTION_FAILED, "no suitable placeholder found and STRICT matching mode requested."); + + if (placeholders.size() == 0) + return null; + + for (int i = 0; i < placeholders.size(); i++) + { + SignaturePlaceholderData spd = (SignaturePlaceholderData)placeholders.get(i); + if (spd.getId() == null) + return spd; + } + + if (matchMode == SignParameters.PLACEHOLDER_MATCH_MODE_LENIENT) + return (SignaturePlaceholderData)placeholders.get(0); + + return null; + } + + private static SignaturePlaceholderData matchPlaceholderPage(List placeholders, + String placeholderId, int matchMode) { + if (placeholders.size() == 0) + return null; + for (int i = 0; i < placeholders.size(); i++) + { + SignaturePlaceholderData data = (SignaturePlaceholderData)placeholders.get(i); + if (placeholderId != null && placeholderId.equals(data.getId())) + return data; + if (placeholderId == null && data.getId() == null) + return data; + } + return null; + } + + private void setCurrentPage(int pageNr) { + this.currentPage = pageNr; + } + + protected void processOperator( PDFOperator operator, List arguments ) throws IOException + { + String operation = operator.getOperation(); + if( operation.equals( "Do" ) ) + { + COSName objectName = (COSName)arguments.get( 0 ); + Map xobjects = getResources().getXObjects(); + PDXObject xobject = (PDXObject)xobjects.get( objectName.getName() ); + if( xobject instanceof PDXObjectImage ) + { + try + { + PDXObjectImage image = (PDXObjectImage)xobject; + SignaturePlaceholderData data = checkImage(image); + if (data != null) + { + PDPage page = getCurrentPage(); + Matrix ctm = getGraphicsState().getCurrentTransformationMatrix(); + double rotationInRadians = (page.findRotation() * Math.PI)/180; + + AffineTransform rotation = new AffineTransform(); + rotation.setToRotation( rotationInRadians ); + AffineTransform rotationInverse = rotation.createInverse(); + Matrix rotationInverseMatrix = new Matrix(); + rotationInverseMatrix.setFromAffineTransform( rotationInverse ); + Matrix rotationMatrix = new Matrix(); + rotationMatrix.setFromAffineTransform( rotation ); + + Matrix unrotatedCTM = ctm.multiply( rotationInverseMatrix ); + + float x = unrotatedCTM.getXPosition() - 1F; + float y = unrotatedCTM.getYPosition() + unrotatedCTM.getYScale() + 1F; + float w = unrotatedCTM.getXScale() + 2F; + + String posString = "p:" + currentPage + ";x:" + x + ";y:" + y + ";w:" + w; + try + { + data.setTablePos(new TablePos(posString)); + data.setPlaceholderName(objectName.getName()); + placeholders.add(data); + } catch (PDFDocumentException e) { + throw new WrappedIOException(e); + } + } + } + catch( NoninvertibleTransformException e ) + { + throw new WrappedIOException( e ); + } + } + } + else + { + super.processOperator( operator, arguments ); + } + } + + /** + * Checks an image if it is a placeholder for a signature. + * + * @param image + * @return + * @throws IOException + */ + private SignaturePlaceholderData checkImage(PDXObjectImage image) throws IOException { + BufferedImage bimg = image.getRGBImage(); + if(bimg.getHeight() < 10 || bimg.getWidth() < 10) + return null; + + LuminanceSource source = new BufferedImageLuminanceSource(bimg); + BinaryBitmap bitmap = new BinaryBitmap(new HybridBinarizer(source)); + Result result; + long before = System.currentTimeMillis(); + try { + Hashtable hints = new Hashtable(); + Vector formats = new Vector(); + formats.add(BarcodeFormat.QR_CODE); + hints.put(DecodeHintType.POSSIBLE_FORMATS, formats); + result = new MultiFormatReader().decode(bitmap, hints); + String text = result.getText(); + String profile = null; + String type = null; + String sigKey = null; + String id = null; + if (text != null && text.startsWith("PDF-AS-POS")) { + String[] data = text.split(";"); + if (data.length > 1) { + for (int i = 1; i < data.length; i++) { + String kvPair = data[i]; + String[] kv = kvPair.split("="); + if (kv.length != 2) { + log.debug("invalid parameter in placeholder data: " + kvPair); + } else { + if (kv[0].equalsIgnoreCase(SignaturePlaceholderData.ID_KEY)) { + id = kv[1]; + } else if (kv[0].equalsIgnoreCase(SignaturePlaceholderData.PROFILE_KEY)) { + profile = kv[1]; + } else if (kv[0] + .equalsIgnoreCase(SignaturePlaceholderData.SIG_KEY_KEY)) { + sigKey = kv[1]; + } else if (kv[0] + .equalsIgnoreCase(SignaturePlaceholderData.TYPE_KEY)) { + type = kv[1]; + } + } + } + } + return new SignaturePlaceholderData(profile, type, sigKey, id); + } + } catch (ReaderException re) { + if (log.isDebugEnabled()) { + log.debug("Could not decode - not a placeholder. needed: " + + (System.currentTimeMillis() - before)); + } + if (!(re instanceof NotFoundException)){ + if (log.isInfoEnabled()) { + log.info("Failed to decode image", re); + } + } + } catch(ArrayIndexOutOfBoundsException e){ + if (log.isInfoEnabled()) { + log.info("Failed to decode image. Probably a zxing bug", e); + } + } + return null; + } + +} -- cgit v1.2.3