Examples with PdfImageVisitor com.joliciel.jochre.utils.pdf.PdfImageVisitor used on opensource projects

Example 1 with PdfImageVisitor

use of com.joliciel.jochre.utils.pdf.PdfImageVisitor in project jochre by urieli.

the class JochreIndexDirectory method getMetaData.

/**
 * The metadata contained in the PDF file.
 */
public Map<JochreIndexField, String> getMetaData() {
    if (this.metaData == null) {
        if (this.getMetaDataFile() == null) {
            PdfImageVisitor pdfMetadataReader = new PdfImageVisitor(this.getPdfFile());
            Map<String, String> pdfMetaData = pdfMetadataReader.getFields();
            pdfMetadataReader.close();
            this.metaData = new HashMap<>();
            // TODO: hack for Yiddish - need to generalize this through
            // config settings
            String bookUrl = pdfMetaData.get("Keywords");
            String title = pdfMetaData.get("Title");
            String author = pdfMetaData.get("Author");
            if (bookUrl != null && bookUrl.length() > 0) {
                this.metaData.put(JochreIndexField.url, bookUrl);
                String id = bookUrl.substring(bookUrl.lastIndexOf('/') + 1);
                this.metaData.put(JochreIndexField.id, id);
            }
            if (title != null && title.length() > 0) {
                title = DiacriticRemover.apply(title);
                this.metaData.put(JochreIndexField.titleEnglish, title);
            }
            if (author != null && author.length() > 0)
                this.metaData.put(JochreIndexField.authorEnglish, author);
        } else {
            try {
                DocumentBuilderFactory dbf = DocumentBuilderFactory.newInstance();
                DocumentBuilder db = dbf.newDocumentBuilder();
                Document dom = db.parse(this.getMetaDataFile());
                Element docElement = dom.getDocumentElement();
                XPathFactory xpf = XPathFactory.newInstance();
                XPath xp = xpf.newXPath();
                String id = xp.evaluate("/metadata/identifier/text()", docElement);
                String bookUrl = xp.evaluate("/metadata/identifier-access/text()", docElement);
                String title = xp.evaluate("/metadata/title/text()", docElement);
                String author = xp.evaluate("/metadata/creator/text()", docElement);
                String publisher = xp.evaluate("/metadata/publisher/text()", docElement);
                String date = xp.evaluate("/metadata/date/text()", docElement);
                String authorLang = xp.evaluate("/metadata/creator-alt-script/text()", docElement);
                String titleLang = xp.evaluate("/metadata/title-alt-script/text()", docElement);
                String volume = xp.evaluate("/metadata/volume/text()", docElement);
                LOG.debug("id: " + id);
                LOG.debug("bookUrl: " + bookUrl);
                LOG.debug("title: " + title);
                LOG.debug("author: " + author);
                LOG.debug("publisher: " + publisher);
                LOG.debug("date: " + date);
                LOG.debug("authorLang: " + authorLang);
                LOG.debug("titleLang: " + titleLang);
                LOG.debug("volume: " + volume);
                this.metaData = new HashMap<>();
                if (id.length() > 0)
                    this.metaData.put(JochreIndexField.id, id);
                if (bookUrl.length() > 0)
                    this.metaData.put(JochreIndexField.url, bookUrl);
                if (title.length() > 0)
                    this.metaData.put(JochreIndexField.titleEnglish, title);
                if (author.length() > 0)
                    this.metaData.put(JochreIndexField.authorEnglish, author);
                if (publisher.length() > 0)
                    this.metaData.put(JochreIndexField.publisher, publisher);
                if (date.length() > 0)
                    this.metaData.put(JochreIndexField.date, date);
                if (authorLang.length() > 0)
                    this.metaData.put(JochreIndexField.author, authorLang);
                if (titleLang.length() > 0)
                    this.metaData.put(JochreIndexField.title, titleLang);
                if (volume.length() > 0)
                    this.metaData.put(JochreIndexField.volume, volume);
            } catch (IOException e) {
                LOG.error("Failed to read metadata from  " + this.getMetaDataFile().getAbsolutePath(), e);
                throw new RuntimeException(e);
            } catch (XPathExpressionException e) {
                LOG.error("Failed to read metadata from  " + this.getMetaDataFile().getAbsolutePath(), e);
                throw new RuntimeException(e);
            } catch (ParserConfigurationException e) {
                LOG.error("Failed to read metadata from  " + this.getMetaDataFile().getAbsolutePath(), e);
                throw new RuntimeException(e);
            } catch (SAXException e) {
                LOG.error("Failed to read metadata from  " + this.getMetaDataFile().getAbsolutePath(), e);
                throw new RuntimeException(e);
            }
        }
    }
    return metaData;
}

Example 2 with PdfImageVisitor

use of com.joliciel.jochre.utils.pdf.PdfImageVisitor in project jochre by urieli.

the class PdfImageReader method readImage.

public BufferedImage readImage(int pageNumber) {
    // assuming only one image per PDF page
    Set<Integer> pages = new HashSet<>();
    pages.add(pageNumber);
    PdfImageVisitor pdfImageVisitor = new PdfImageVisitor(this.pdfFile, pages);
    PdfImageReaderInternal imageReader = new PdfImageReaderInternal();
    pdfImageVisitor.addImageObserver(imageReader);
    pdfImageVisitor.visitImages();
    BufferedImage image = imageReader.getImage();
    return image;
}

Example 3 with PdfImageVisitor

use of com.joliciel.jochre.utils.pdf.PdfImageVisitor in project jochre by urieli.

the class YiddishMetaFetcher method fetchMetaData.

public void fetchMetaData(File pdfFile, Writer writer) throws Exception {
    PdfImageVisitor pdfMetadataReader = new PdfImageVisitor(pdfFile);
    Map<String, String> metadata = pdfMetadataReader.getFields();
    pdfMetadataReader.close();
    String url = metadata.get("Keywords");
    if (url == null || !url.startsWith("http"))
        return;
    String reference = url.substring(url.lastIndexOf('/') + 1);
    URL metaUrl = new URL("https://archive.org/download/" + reference + "/" + reference + "_meta.xml");
    DocumentBuilderFactory dbf = DocumentBuilderFactory.newInstance();
    DocumentBuilder db = dbf.newDocumentBuilder();
    Document dom = db.parse(metaUrl.openStream());
    Element docElement = dom.getDocumentElement();
    XPathFactory xpf = XPathFactory.newInstance();
    XPath xp = xpf.newXPath();
    String id = xp.evaluate("/metadata/identifier/text()", docElement);
    String bookUrl = xp.evaluate("/metadata/identifier-access/text()", docElement);
    String title = xp.evaluate("/metadata/title/text()", docElement);
    String author = xp.evaluate("/metadata/creator/text()", docElement);
    String publisher = xp.evaluate("/metadata/publisher/text()", docElement);
    String date = xp.evaluate("/metadata/date/text()", docElement);
    String authorYid = xp.evaluate("/metadata/creator-alt-script/text()", docElement);
    String titleYid = xp.evaluate("/metadata/title-alt-script/text()", docElement);
    String pageCount = xp.evaluate("/metadata/imagecount/text()", docElement);
    String volume = xp.evaluate("/metadata/volume/text()", docElement);
    title = DiacriticRemover.apply(title);
    author = DiacriticRemover.apply(author);
    publisher = DiacriticRemover.apply(publisher);
    LOG.debug("id: " + id);
    LOG.debug("bookUrl: " + bookUrl);
    LOG.debug("title: " + title);
    LOG.debug("volume: " + volume);
    LOG.debug("author: " + author);
    LOG.debug("publisher: " + publisher);
    LOG.debug("date: " + date);
    LOG.debug("authorYid: " + authorYid);
    LOG.debug("titleYid: " + titleYid);
    LOG.debug("pageCount: " + pageCount);
    writer.write("<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n");
    writer.write("<metadata>\n");
    writer.write("  <identifier>" + StringEscapeUtils.escapeXml10(id) + "</identifier>\n");
    writer.write("  <identifier-access>" + StringEscapeUtils.escapeXml10(bookUrl) + "</identifier-access>\n");
    writer.write("  <title>" + StringEscapeUtils.escapeXml10(title) + "</title>\n");
    writer.write("  <volume>" + StringEscapeUtils.escapeXml10(volume) + "</volume>\n");
    writer.write("  <creator>" + StringEscapeUtils.escapeXml10(author) + "</creator>\n");
    writer.write("  <publisher>" + StringEscapeUtils.escapeXml10(publisher) + "</publisher>\n");
    writer.write("  <date>" + StringEscapeUtils.escapeXml10(date) + "</date>\n");
    writer.write("  <creator-alt-script>" + StringEscapeUtils.escapeXml10(authorYid) + "</creator-alt-script>\n");
    writer.write("  <title-alt-script>" + StringEscapeUtils.escapeXml10(titleYid) + "</title-alt-script>\n");
    writer.write("  <imagecount>" + StringEscapeUtils.escapeXml10(pageCount) + "</imagecount>\n");
    writer.write("</metadata>\n");
    writer.flush();
}

Example 4 with PdfImageVisitor

use of com.joliciel.jochre.utils.pdf.PdfImageVisitor in project jochre by urieli.

the class Jochre method doCommandExtractImages.

/**
 * Extract the images from a PDF file.
 * @param filename
 *          the path to the PDF file
 * @param pages
 */
public void doCommandExtractImages(String filename, File outputDir, Set<Integer> pages) {
    if (filename.length() == 0)
        throw new RuntimeException("Missing argument: file");
    if (filename.toLowerCase().endsWith(".pdf")) {
        File pdfFile = new File(filename);
        String baseName = this.getBaseName(pdfFile);
        List<PdfImageObserver> imageObservers = this.getImageObservers(Arrays.asList(OutputFormat.ImageExtractor), baseName, outputDir);
        PdfImageVisitor pdfImageVisitor = new PdfImageVisitor(pdfFile, pages);
        for (PdfImageObserver imageObserver : imageObservers) {
            pdfImageVisitor.addImageObserver(imageObserver);
        }
        pdfImageVisitor.visitImages();
    } else {
        throw new RuntimeException("Unrecognised file extension");
    }
}