Example usage for org.apache.pdfbox.pdmodel PDDocument close

List of usage examples for org.apache.pdfbox.pdmodel PDDocument close

  1. HOME
  2. Java
  3. org.apache.pdfbox
  4. org.apache.pdfbox.pdmodel.*
  5. PDDocument
  6. close

Introduction

In this page you can find the example usage for org.apache.pdfbox.pdmodel PDDocument close.

Prototype

@Override
public void close() throws IOException

Source Link

Document

This will close the underlying COSDocument object.

Usage

From source file:com.joowon.returnA.classifier.EbsBookCrawler.java

License:Open Source License

public void run() throws IOException {
    File destination = new File("/Users/Joowon/Desktop");
    File bookFolder = new File(getClass().getClassLoader().getResource("book").getFile());
    for (File book : bookFolder.listFiles()) {
        String outputName = destination.getPath() + "/" + book.getName().replace(".pdf", ".txt");
        PDDocument document = PDDocument.load(book);
        String text = "";
        for (PDPage page : document.getPages()) {
            text += new PdfTextExtractor(page)
                    .addRegion(0, 0, (int) page.getMediaBox().getWidth(), (int) page.getMediaBox().getHeight())
                    .extract();//w w w. ja  va 2s  .c  o  m
        }

        new TxtWriter(outputName).write(text);

        document.close();
    }
}

From source file:com.jt.tool.pdf.CreateBookmarks.java

License:Apache License

public static void createBookmark(String srcFile, String targetFile, String reg) throws Exception {
    PDDocument document = null;
    try {/*ww w  .  java  2  s  .co m*/
        document = PDDocument.load(new File(srcFile));
        if (document.isEncrypted()) {
            System.err.println("Error: Cannot add bookmarks to encrypted document.");
            System.exit(1);
        }
        PDDocumentOutline outline = new PDDocumentOutline();
        document.getDocumentCatalog().setDocumentOutline(outline);
        PDOutlineItem pagesOutline = new PDOutlineItem();
        pagesOutline.setTitle("All Pages");
        //            outline.appendChild(pagesOutline);
        List pages = new ArrayList();
        //                    document.getDocumentCatalog().getAllPages();
        for (int i = 12; i < pages.size(); i++) {
            String pageText = getPageText(document, i + 1, 0);
            String[] strings = matchTitle(pageText, reg);
            if (makeBookmark(strings)) {
                PDPage page = (PDPage) pages.get(i);
                PDPageFitWidthDestination dest = new PDPageFitWidthDestination();
                dest.setPage(page);
                PDOutlineItem bookmark = new PDOutlineItem();
                bookmark.setDestination(dest);
                bookmark.setTitle(strings[0]);
                //                    pagesOutline.appendChild(bookmark);
                System.out.println("add " + strings[0]);
            }
        }
        pagesOutline.openNode();
        outline.openNode();
        document.save(targetFile);
    } finally {
        if (document != null) {
            document.close();
        }
    }
}

From source file:com.lanacion.adminsiteln.services.PdfIndexerService.PdfIndexerService.java

/**
 * Metodos privados para la indexacin//  w  ww.  j  av  a 2 s  .  c  om
 */
private String pdftoText(String fileName, int pagina) {

    PDFParser parser;
    String parsedText = null;
    ;
    PDFTextStripper pdfStripper = null;
    //pdfStripper.setStartPage(0);
    //pdfStripper.setEndPage(0);
    PDDocument pdDoc = null;
    COSDocument cosDoc = null;
    File file = new File(fileName);
    if (!file.isFile()) {
        System.err.println("File " + fileName + " does not exist.");
        return null;
    }
    try {
        parser = new PDFParser(new FileInputStream(file));
    } catch (IOException e) {
        System.err.println("Unable to open PDF Parser. " + e.getMessage());
        return null;
    }
    try {
        parser.parse();
        cosDoc = parser.getDocument();
        pdfStripper = new PDFTextStripper();
        pdDoc = new PDDocument(cosDoc);
        pdfStripper.setStartPage(pagina);
        pdfStripper.setEndPage(pagina);
        parsedText = pdfStripper.getText(pdDoc);
    } catch (Exception e) {
        System.err.println("An exception occured in parsing the PDF Document." + e.getMessage());
    } finally {
        try {
            if (cosDoc != null) {
                cosDoc.close();
            }
            if (pdDoc != null) {
                pdDoc.close();
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    return parsedText;

}

From source file:com.lanacion.adminsiteln.services.PdfIndexerService.PdfIndexerService.java

private int pdfgetPages(String fileName) {

    int numero_paginas = 0;
    PDFParser parser;/*from w ww  . j a v  a 2  s .  c  o  m*/
    String parsedText = null;
    ;
    PDFTextStripper pdfStripper = null;
    //pdfStripper.setStartPage(0);
    //pdfStripper.setEndPage(0);
    PDDocument pdDoc = null;
    COSDocument cosDoc = null;
    File file = new File(fileName);
    if (!file.isFile()) {
        System.err.println("File " + fileName + " does not exist.");
        return 0;
    }
    try {
        parser = new PDFParser(new FileInputStream(file));
    } catch (IOException e) {
        System.err.println("Unable to open PDF Parser. " + e.getMessage());
        return 0;
    }
    try {
        parser.parse();
        cosDoc = parser.getDocument();
        pdfStripper = new PDFTextStripper();
        pdDoc = new PDDocument(cosDoc);
        numero_paginas = pdDoc.getNumberOfPages();
    } catch (Exception e) {
        System.err.println("An exception occured in parsing the PDF Document." + e.getMessage());
    } finally {
        try {
            if (cosDoc != null) {
                cosDoc.close();
            }
            if (pdDoc != null) {
                pdDoc.close();
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    return numero_paginas;
}

From source file:com.liferay.portal.util.LuceneFields.java

License:Open Source License

public static Field getFile(String field, File file, String fileExt) throws IOException {

    fileExt = fileExt.toLowerCase();/*from  w w w  .ja  va  2  s.  c  o  m*/

    FileInputStream fis = new FileInputStream(file);
    Reader reader = new BufferedReader(new InputStreamReader(fis));

    String text = null;

    if (fileExt.equals(".doc")) {
        try {
            WordDocument wordDocument = new WordDocument(fis);

            StringWriter stringWriter = new StringWriter();

            wordDocument.writeAllText(stringWriter);

            text = stringWriter.toString();

            stringWriter.close();
        } catch (Exception e) {
            _log.error(e.getMessage());
        }
    } else if (fileExt.equals(".htm") || fileExt.equals(".html")) {
        try {
            DefaultStyledDocument dsd = new DefaultStyledDocument();

            HTMLEditorKit htmlEditorKit = new HTMLEditorKit();
            htmlEditorKit.read(reader, dsd, 0);

            text = dsd.getText(0, dsd.getLength());
        } catch (Exception e) {
            _log.error(e.getMessage());
        }
    } else if (fileExt.equals(".pdf")) {
        try {
            PDFParser parser = new PDFParser(fis);
            parser.parse();

            PDDocument pdDoc = parser.getPDDocument();

            StringWriter stringWriter = new StringWriter();

            PDFTextStripper stripper = new PDFTextStripper();
            stripper.setLineSeparator("\n");
            stripper.writeText(pdDoc, stringWriter);

            text = stringWriter.toString();

            stringWriter.close();
            pdDoc.close();
        } catch (Exception e) {
            _log.error(e.getMessage());
        }
    } else if (fileExt.equals(".rtf")) {
        try {
            DefaultStyledDocument dsd = new DefaultStyledDocument();

            RTFEditorKit rtfEditorKit = new RTFEditorKit();
            rtfEditorKit.read(reader, dsd, 0);

            text = dsd.getText(0, dsd.getLength());
        } catch (Exception e) {
            _log.error(e.getMessage());
        }
    } else if (fileExt.equals(".xls")) {
        try {
            XLSTextStripper stripper = new XLSTextStripper(fis);

            text = stripper.getText();
        } catch (Exception e) {
            _log.error(e.getMessage());
        }
    }

    if (text != null) {
        return new Field(field, text, Field.Store.YES, Field.Index.NOT_ANALYZED);
    } else {
        return new Field(field, reader);
    }
}

From source file:com.liferay.portlet.documentlibrary.util.LiferayPDFBoxConverter.java

License:Open Source License

public void generateImagesPB() throws Exception {
    PDDocument pdDocument = null;

    try {//from www  .  jav  a 2 s  .  c  o m
        pdDocument = PDDocument.load(_inputFile);

        PDDocumentCatalog pdDocumentCatalog = pdDocument.getDocumentCatalog();

        PDFRenderer pdfRenderer = new PDFRenderer(pdDocument);

        PDPageTree pdPages = pdDocumentCatalog.getPages();

        for (int i = 0; i < pdPages.getCount(); i++) {
            if (_generateThumbnail && (i == 0)) {
                _generateImagesPB(pdfRenderer, i, _thumbnailFile, _thumbnailExtension);
            }

            if (!_generatePreview) {
                break;
            }

            _generateImagesPB(pdfRenderer, i, _previewFiles[i], _extension);
        }
    } finally {
        if (pdDocument != null) {
            pdDocument.close();
        }
    }
}

From source file:com.liferay.portlet.documentlibrary.util.PDFProcessorImpl.java

License:Open Source License

private void _generateImagesPB(FileVersion fileVersion, InputStream inputStream) throws Exception {

    boolean generatePreview = _isGeneratePreview(fileVersion);
    boolean generateThumbnail = _isGenerateThumbnail(fileVersion);

    PDDocument pdDocument = null;

    try {//ww  w . j  a v  a 2 s.c  o m
        pdDocument = PDDocument.load(inputStream);

        PDDocumentCatalog pdDocumentCatalog = pdDocument.getDocumentCatalog();

        List<PDPage> pdPages = pdDocumentCatalog.getAllPages();

        for (int i = 0; i < pdPages.size(); i++) {
            PDPage pdPage = pdPages.get(i);

            if (generateThumbnail && (i == 0)) {
                _generateImagesPB(fileVersion, pdPage, i);

                if (_log.isInfoEnabled()) {
                    _log.info("PDFBox generated a thumbnail for " + fileVersion.getFileVersionId());
                }
            }

            if (!generatePreview) {
                break;
            }

            _generateImagesPB(fileVersion, pdPage, i + 1);
        }

        if (_log.isInfoEnabled() && generatePreview) {
            _log.info("PDFBox generated " + getPreviewFileCount(fileVersion) + " preview pages for "
                    + fileVersion.getFileVersionId());
        }
    } finally {
        if (pdDocument != null) {
            pdDocument.close();
        }
    }
}

From source file:com.lp.server.system.ejbfac.SystemFacBean.java

License:Open Source License

public byte[][] konvertierePDFFileInEinzelneBilder(String pdfFile, int resolution) {
    PDDocument document = null;
    byte[][] oBilder = null;
    try {// w  w  w.  j  a v  a  2  s .c  om
        document = PDDocument.load(pdfFile);
        List pages = document.getDocumentCatalog().getAllPages();

        oBilder = new byte[pages.size()][];

        for (int i = 0; i < pages.size(); i++) {
            PDPage page = (PDPage) pages.get(i);
            BufferedImage image = page.convertToImage(BufferedImage.TYPE_INT_RGB, resolution);
            oBilder[i] = Helper.imageToByteArray(image);
        }
    } catch (IOException e) {
        e.printStackTrace();
        throw new EJBExceptionLP(EJBExceptionLP.FEHLER, e.getMessage());

    } finally {
        if (document != null) {

            try {
                document.close();
            } catch (IOException e) {
                e.printStackTrace();
                throw new EJBExceptionLP(EJBExceptionLP.FEHLER, e.getMessage());

            }
        }

    }
    return oBilder;
}

From source file:com.mirth.connect.connectors.doc.DocumentDispatcher.java

License:Open Source License

private void encryptPDF(InputStream inputStream, OutputStream outputStream, String password) throws Exception {
    PDDocument document = null;

    try {/*from w w w .  j  a va  2 s .com*/
        document = PDDocument.load(inputStream);

        AccessPermission accessPermission = new AccessPermission();
        accessPermission.setCanAssembleDocument(false);
        accessPermission.setCanExtractContent(true);
        accessPermission.setCanExtractForAccessibility(false);
        accessPermission.setCanFillInForm(false);
        accessPermission.setCanModify(false);
        accessPermission.setCanModifyAnnotations(false);
        accessPermission.setCanPrint(true);
        accessPermission.setCanPrintDegraded(true);

        String ownerPassword = System.currentTimeMillis() + "+" + Runtime.getRuntime().freeMemory() + "+"
                + (ownerPasswordSeq++);
        StandardProtectionPolicy policy = new StandardProtectionPolicy(ownerPassword, password,
                accessPermission);
        policy.setEncryptionKeyLength(128);
        document.protect(policy);

        document.save(outputStream);
    } catch (Exception e) {
        throw e;
    } finally {
        if (document != null) {
            document.close();
        }
    }
}

From source file:com.mycompany.mavenproject4.NewMain.java

/**
 * @param args the command line arguments
 *//*  w ww.j  a  v  a 2s  .  c o m*/
public static void main(String[] args) {
    PDDocument doc;
    doc = new PDDocument();
    doc.addPage(new PDPage());
    try {
        doc.save("Empty PDF.pdf");
        doc.close();
    } catch (Exception io) {
        System.out.println(io);
    }
}