Examples with PdfDocumentProcessor com.joliciel.jochre.pdf.PdfDocumentProcessor used on opensource projects

Example 1 with PdfDocumentProcessor

use of com.joliciel.jochre.pdf.PdfDocumentProcessor in project jochre by urieli.

the class TextController method onClick$btnAnalyse.

@Listen("onClick = #btnAnalyse")
public void onClick$btnAnalyse(Event event) {
    try {
        LOG.debug("onClick$btnAnalyse");
        if (currentFile != null) {
            progressBox.setVisible(true);
            lblAwaitingFile.setVisible(false);
            int startPage = txtStartPage.getValue().length() == 0 ? -1 : Integer.parseInt(txtStartPage.getValue());
            int endPage = txtEndPage.getValue().length() == 0 ? -1 : Integer.parseInt(txtEndPage.getValue());
            if (this.currentDoc != null) {
                this.currentDoc.setFileName(currentFile.getName());
                this.currentDoc.save();
                this.documentGenerator = new JochreDocumentGenerator(this.currentDoc, jochreSession);
                this.documentGenerator.requestSave(currentUser);
            } else {
                this.documentGenerator = new JochreDocumentGenerator(currentFile.getName(), "", jochreSession);
            }
            String letterModelPath = jochreSession.getLetterModelPath();
            if (letterModelPath != null) {
                MostLikelyWordChooser wordChooser = new MostLikelyWordChooser(jochreSession);
                documentGenerator.requestAnalysis(wordChooser);
            }
            this.documentHtmlGenerator = new DocumentHtmlGenerator();
            documentGenerator.addDocumentObserver(this.documentHtmlGenerator);
            String lowerCaseFileName = currentFile.getName().toLowerCase();
            Thread thread = null;
            final Set<Integer> pages;
            if (startPage < 0 && endPage < 0)
                pages = Collections.emptySet();
            else {
                if (startPage < 0)
                    startPage = 0;
                if (endPage < 0)
                    endPage = 2000;
                pages = IntStream.rangeClosed(startPage, endPage).boxed().collect(Collectors.toSet());
            }
            if (lowerCaseFileName.endsWith(".pdf")) {
                PdfDocumentProcessor pdfDocumentProcessor = new PdfDocumentProcessor(currentFile, pages, documentGenerator);
                this.progressMonitor = pdfDocumentProcessor.monitorTask();
                this.currentHtmlIndex = 0;
                thread = new Thread(pdfDocumentProcessor);
                thread.setName(currentFile.getName() + " Processor");
                progressTimer.setRunning(true);
            } else if (lowerCaseFileName.endsWith(".png") || lowerCaseFileName.endsWith(".jpg") || lowerCaseFileName.endsWith(".jpeg") || lowerCaseFileName.endsWith(".gif")) {
                ImageDocumentExtractor extractor = new ImageDocumentExtractor(currentFile, documentGenerator);
                if (startPage >= 0)
                    extractor.setPageNumber(startPage);
                this.progressMonitor = extractor.monitorTask();
                this.currentHtmlIndex = 0;
                thread = new Thread(extractor);
                thread.setName(currentFile.getName() + " Processor");
            } else {
                throw new RuntimeException("Unrecognised file extension");
            }
            thread.start();
            currentThread = thread;
            progressTimer.setRunning(true);
            btnAnalyse.setDisabled(true);
            btnInterrupt.setVisible(true);
            btnInterrupt.setDisabled(false);
        }
    } catch (Exception e) {
        LOG.error("Failure in onClick$btnAnalyse", e);
        throw new RuntimeException(e);
    }
}

Example 2 with PdfDocumentProcessor

use of com.joliciel.jochre.pdf.PdfDocumentProcessor in project jochre by urieli.

the class Jochre method doCommandAnalyse.

/**
 * Full analysis, including merge, split and letter guessing.
 *
 * @param pages
 *          the pages to process, empty means all
 */
public void doCommandAnalyse(File sourceFile, MostLikelyWordChooser wordChooser, Set<Integer> pages, List<DocumentObserver> observers, List<PdfImageObserver> imageObservers) throws IOException {
    ClassificationModel letterModel = jochreSession.getLetterModel();
    List<String> letterFeatureDescriptors = letterModel.getFeatureDescriptors();
    LetterFeatureParser letterFeatureParser = new LetterFeatureParser();
    Set<LetterFeature<?>> letterFeatures = letterFeatureParser.getLetterFeatureSet(letterFeatureDescriptors);
    LetterGuesser letterGuesser = new LetterGuesser(letterFeatures, letterModel.getDecisionMaker());
    BoundaryDetector boundaryDetector = null;
    LetterGuessObserver letterGuessObserver = null;
    if (jochreSession.getSplitModel() != null && jochreSession.getMergeModel() != null) {
        boundaryDetector = new DeterministicBoundaryDetector(jochreSession.getSplitModel(), jochreSession.getMergeModel(), jochreSession);
        OriginalShapeLetterAssigner shapeLetterAssigner = new OriginalShapeLetterAssigner();
        shapeLetterAssigner.setEvaluate(false);
        shapeLetterAssigner.setSingleLetterMethod(false);
        letterGuessObserver = shapeLetterAssigner;
    } else {
        boundaryDetector = new OriginalBoundaryDetector();
        LetterAssigner letterAssigner = new LetterAssigner();
        letterGuessObserver = letterAssigner;
    }
    ImageAnalyser analyser = new BeamSearchImageAnalyser(boundaryDetector, letterGuesser, wordChooser, jochreSession);
    analyser.addObserver(letterGuessObserver);
    JochreDocumentGenerator documentGenerator = new JochreDocumentGenerator(sourceFile.getName(), "", jochreSession);
    documentGenerator.addDocumentObserver(analyser);
    for (DocumentObserver observer : observers) documentGenerator.addDocumentObserver(observer);
    if (!sourceFile.exists())
        throw new JochreException("The file " + sourceFile.getPath() + " does not exist");
    if (sourceFile.getName().toLowerCase().endsWith(".pdf")) {
        PdfDocumentProcessor pdfDocumentProcessor = new PdfDocumentProcessor(sourceFile, pages, documentGenerator);
        for (PdfImageObserver imageObserver : imageObservers) {
            pdfDocumentProcessor.addImageObserver(imageObserver);
        }
        pdfDocumentProcessor.process();
    } else if (sourceFile.getName().toLowerCase().endsWith(".png") || sourceFile.getName().toLowerCase().endsWith(".jpg") || sourceFile.getName().toLowerCase().endsWith(".jpeg") || sourceFile.getName().toLowerCase().endsWith(".gif")) {
        ImageDocumentExtractor extractor = new ImageDocumentExtractor(sourceFile, documentGenerator);
        extractor.extractDocument();
    } else if (sourceFile.isDirectory()) {
        ImageDocumentExtractor extractor = new ImageDocumentExtractor(sourceFile, documentGenerator);
        extractor.extractDocument();
    } else {
        throw new RuntimeException("Unrecognised file extension");
    }
}

Example 3 with PdfDocumentProcessor

use of com.joliciel.jochre.pdf.PdfDocumentProcessor in project jochre by urieli.

the class Jochre method doCommandUpdateImages.

/**
 * Update the images in an existing Jochre document.
 *
 * @param filename
 *          the PDF file containing the images
 * @param docId
 *          the id of the document to update
 * @param pаges
 *          the pages to process, empty means all
 */
public void doCommandUpdateImages(String filename, int docId, Set<Integer> pages) {
    if (filename.length() == 0)
        throw new RuntimeException("Missing argument: file");
    if (docId < 0)
        throw new RuntimeException("Missing argument: docId");
    DocumentDao documentDao = DocumentDao.getInstance(jochreSession);
    JochreDocument doc = documentDao.loadJochreDocument(docId);
    if (filename.toLowerCase().endsWith(".pdf")) {
        File pdfFile = new File(filename);
        PdfDocumentProcessor pdfDocumentProcessor = new PdfDocumentProcessor(pdfFile, pages, new PdfImageUpdater(doc));
        pdfDocumentProcessor.process();
    } else {
        throw new RuntimeException("Unrecognised file extension");
    }
}

Example 4 with PdfDocumentProcessor

use of com.joliciel.jochre.pdf.PdfDocumentProcessor in project jochre by urieli.

the class Jochre method doCommandSegment.

/**
 * Segment a file, without analysing it.
 *
 * @param filename
 *          the path of the file to load
 * @param userFriendlyName
 *          a name to store against this file in the database
 * @param save
 *          should we save this file to the database?
 * @param pages
 *          the pages to process, empty means all
 */
public void doCommandSegment(String filename, String userFriendlyName, File outputDir, boolean save, Set<Integer> pages) {
    if (filename.length() == 0)
        throw new RuntimeException("Missing argument: file");
    if (userId < 0 && save)
        throw new RuntimeException("Missing argument (for save=true): userId");
    User user = null;
    if (userId >= 0) {
        SecurityDao securityDao = SecurityDao.getInstance(jochreSession);
        user = securityDao.loadUser(userId);
    }
    File file = new File(filename);
    JochreDocumentGenerator jochreDocumentGenerator = new JochreDocumentGenerator(file.getName(), userFriendlyName, jochreSession);
    if (save)
        jochreDocumentGenerator.requestSave(user);
    if (jochreDocumentGenerator.isDrawSegmentedImage()) {
        if (outputDir != null) {
            jochreDocumentGenerator.requestSegmentation(outputDir);
        }
    }
    if (filename.toLowerCase().endsWith(".pdf")) {
        PdfDocumentProcessor pdfDocumentProcessor = new PdfDocumentProcessor(file, pages, jochreDocumentGenerator);
        pdfDocumentProcessor.process();
    } else if (filename.toLowerCase().endsWith(".png") || filename.toLowerCase().endsWith(".jpg") || filename.toLowerCase().endsWith(".jpeg") || filename.toLowerCase().endsWith(".gif")) {
        ImageDocumentExtractor extractor = new ImageDocumentExtractor(file, jochreDocumentGenerator);
        extractor.extractDocument();
    } else {
        throw new RuntimeException("Unrecognised file extension");
    }
}