Package org.pdfbox.pdmodel

Examples of org.pdfbox.pdmodel.PDDocument

401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
     *
     * @throws IOException If there is an error parsing the document.
     */
    private void addContent( Document document, InputStream is, String documentLocation ) throws IOException
    {
        PDDocument pdfDocument = null;
        try
        {
            pdfDocument = PDDocument.load( is );

            if( pdfDocument.isEncrypted() )
            {
                //Just try using the default password and move on
                pdfDocument.decrypt( "" );
            }

            //create a writer where to append the text content.
            StringWriter writer = new StringWriter();
            if( stripper == null )
            {
                stripper = new PDFTextStripper();
            }
            else
            {
                stripper.resetEngine();
            }
            stripper.writeText( pdfDocument, writer );

            // Note: the buffer to string operation is costless;
            // the char array value of the writer buffer and the content string
            // is shared as long as the buffer content is not modified, which will
            // not occur here.
            String contents = writer.getBuffer().toString();

            StringReader reader = new StringReader( contents );

            // Add the tag-stripped contents as a Reader-valued Text field so it will
            // get tokenized and indexed.
            addTextField( document, "contents", reader );

            PDDocumentInformation info = pdfDocument.getDocumentInformation();
            if( info != null )
            {
                addTextField( document, "Author", info.getAuthor() );
                addTextField( document, "CreationDate", info.getCreationDate() );
                addTextField( document, "Creator", info.getCreator() );
                addTextField( document, "Keywords", info.getKeywords() );
                addTextField( document, "ModificationDate", info.getModificationDate() );
                addTextField( document, "Producer", info.getProducer() );
                addTextField( document, "Subject", info.getSubject() );
                addTextField( document, "Title", info.getTitle() );
                addTextField( document, "Trapped", info.getTrapped() );
            }
            int summarySize = Math.min( contents.length(), 500 );
            String summary = contents.substring( 0, summarySize );
            // Add the summary as an UnIndexed field, so that it is stored and returned
            // with hit documents for display.
            addUnindexedField( document, "summary", summary );
        }
        catch( CryptographyException e )
        {
            throw new IOException( "Error decrypting document(" + documentLocation + "): " + e );
        }
        catch( InvalidPasswordException e )
        {
            //they didn't suppply a password and the default of "" was wrong.
            throw new IOException(
                "Error: The document(" + documentLocation +
                ") is encrypted and will not be indexed." );
        }
        finally
        {
            if( pdfDocument != null )
            {
                pdfDocument.close();
            }
        }
    }
View Full Code Here
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
        {
            usage();
        }
        else
        {
            PDDocument document = null;
            try
            {
                document = PDDocument.load( args[0] );
                if( document.isEncrypted() )
                {
                    try
                    {
                        document.decrypt( "" );
                    }
                    catch( InvalidPasswordException e )
                    {
                        System.err.println( "Error: Document is encrypted with a password." );
                        System.exit( 1 );
                    }
                }
                PrintImageLocations printer = new PrintImageLocations();
                List allPages = document.getDocumentCatalog().getAllPages();
                for( int i=0; i<allPages.size(); i++ )
                {
                    PDPage page = (PDPage)allPages.get( i );
                    System.out.println( "Processing page: " + i );
                    printer.processStream( page, page.findResources(), page.getContents().getStream() );
                }
            }
            finally
            {
                if( document != null )
                {
                    document.close();
                }
            }
        }
    }
View Full Code Here
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
     * @throws IOException If there is an error parsing the document.
     * @throws COSVisitorException If there is an error while copying the document.
     */
    public void doIt(String in, String out) throws IOException, COSVisitorException
    {
        PDDocument doc = null;
        try
        {
            doc = PDDocument.load( in );
            if( doc.isEncrypted() )
            {
                try
                {
                    doc.decrypt( "" );
                }
                catch( InvalidPasswordException e )
                {
                    System.err.println( "Error: The document is encrypted." );
                }
                catch( org.pdfbox.exceptions.CryptographyException e )
                {
                    e.printStackTrace();
                }
            }

            for (Iterator i = doc.getDocument().getObjects().iterator(); i.hasNext();)
            {
                COSBase base = ((COSObject) i.next()).getObject();
                if (base instanceof COSStream)
                {
                    // just kill the filters
                    COSStream cosStream = (COSStream)base;
                    cosStream.getUnfilteredStream();
                    cosStream.setFilters(null);
                }
            }
            doc.save( out );
        }
        finally
        {
            if( doc != null )
            {
                doc.close();
            }
        }
    }
View Full Code Here
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
     * @throws IOException If there is an error importing the FDF document.
     * @throws CryptographyException If there is an error decrypting the document.
     */
    public static void main(String[] args) throws IOException, CryptographyException
    {
        PDDocument pdf = null;
        try
        {
            if( args.length != 1 )
            {
                usage();
            }
            else
            {
                pdf = PDDocument.load( args[0] );
                PrintFields exporter = new PrintFields();
                if( pdf.isEncrypted() )
                {
                    try
                    {
                        pdf.decrypt( "" );
                    }
                    catch( InvalidPasswordException e )
                    {
                        System.err.println( "Error: The document is encrypted." );
                        usage();
                    }
                }
                exporter.printFields( pdf );
            }
        }
        finally
        {
            if( pdf != null )
            {
                pdf.close();
            }
        }
    }
View Full Code Here
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
        importer.importXFDF( args );
    }

    private void importXFDF( String[] args ) throws Exception
    {
        PDDocument pdf = null;
        FDFDocument fdf = null;

        try
        {
            if( args.length != 3 )
            {
                usage();
            }
            else
            {
                ImportFDF importer = new ImportFDF();
                pdf = PDDocument.load( args[0] );
                fdf = FDFDocument.loadXFDF( args[1] );

                importer.importFDF( pdf, fdf );
                pdf.save( args[2] );
                fdf.save( "tmp/outputXFDFtoPDF.fdf");
            }
        }
        finally
        {
View Full Code Here
991
992
993
994
995
996
997
998
     *
     * @throws COSVisitorException If an error occurs while generating the data.
     */
    public void write(COSDocument doc) throws COSVisitorException
    {
        PDDocument pdDoc = new PDDocument( doc );
        write( pdDoc );
    }
View Full Code Here
157
158
159
160
161
162
163
     * @return The document text.
     * @throws IOException If there is an error extracting the text.
     */
    public String getText( COSDocument doc ) throws IOException
    {
        return getText( new PDDocument( doc ) );
    }
View Full Code Here
169
170
171
172
173
174
175
     * @param outputStream The stream to write the text to.
     * @throws IOException If there is an error extracting the text.
     */
    public void writeText( COSDocument doc, Writer outputStream ) throws IOException
    {
        writeText( new PDDocument( doc ), outputStream );
    }
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
     * @throws COSVisitorException If there is an error writing the PDF.
     */
    public void createPDFFromImage( String file, String image) throws IOException, COSVisitorException
    {
        // the document
        PDDocument doc = null;
        try
        {
            doc = new PDDocument();
           
            PDPage page = new PDPage();
            doc.addPage( page );
           
            PDXObjectImage ximage = null;
            if( image.toLowerCase().endsWith( ".jpg" ) )
            {
                ximage = new PDJpeg(doc, new FileInputStream( image ) );
            }
            else if (image.toLowerCase().endsWith(".tif") || image.toLowerCase().endsWith(".tiff"))
            {
                ximage = new PDCcitt(doc, new RandomAccessFile(new File(image),"r"));
            }
            else
            {
                //BufferedImage awtImage = ImageIO.read( new File( image ) );
                //ximage = new PDPixelMap(doc, awtImage);
                throw new IOException( "Image type not supported:" + image );
            }
            PDPageContentStream contentStream = new PDPageContentStream(doc, page);
           
            contentStream.drawImage( ximage, 20, 20 );
         
            contentStream.close();
            doc.save( file );
        }
        finally
        {
            if( doc != null )
            {
                doc.close();
            }
        }
    }
View Full Code Here
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
        {
            usage();
        }
        else
        {
            PDDocument document = null;
            try
            {
                document = PDDocument.load( args[0] );
                if( document.isEncrypted() )
                {
                    throw new IOException( "Encrypted documents are not supported for this example" );
                }
                List allpages = new ArrayList();
                document.getDocumentCatalog().getPages().getAllKids(allpages);
               
                for (int i=0; i < allpages.size(); i++)
                {
                    PDPage apage = (PDPage) allpages.get(i);
                    List annotations = apage.getAnnotations();
                   
                    PDAnnotationRubberStamp rs = new PDAnnotationRubberStamp();
                    rs.setName(PDAnnotationRubberStamp.NAME_TOP_SECRET);
                    rs.setRectangle(new PDRectangle(100,100));
                    rs.setContents("A top secret note");
                   
                    annotations.add(rs);
                }
               
                document.save( args[1] );
            }
            finally
            {
                if( document != null )
                {
                    document.close();
                }
            }
        }
    }
View Full Code Here
TOP

Related Classes of org.pdfbox.pdmodel.PDDocument

  • br.com.caelum.stella.boleto.transformer.BoletoTransformerIntegrationTest
  • com.stimulus.archiva.extraction.PDFExtractor
  • eu.lsem.bakalarka.filetypeprocess.document.PdfDocumentParser
  • net.sf.jabref.util.XMPUtil
  • org.apache.jackrabbit.core.query.PdfTextFilter
  • org.apache.jackrabbit.extractor.PdfTextExtractor
  • org.apache.nutch.parse.pdf.PdfParser
  • org.exoplatform.services.document.impl.PDFDocumentReader
  • org.infoglue.cms.controllers.kernel.impl.simple.LuceneController
  • org.pdfbox.cos.COSArray
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.