1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one or more
3 * contributor license agreements. See the NOTICE file distributed with
4 * this work for additional information regarding copyright ownership.
5 * The ASF licenses this file to You under the Apache License, Version 2.0
6 * (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 *
9 * http://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17 package org.apache.pdfbox.pdmodel;
18
19 import org.apache.commons.logging.Log;
20 import org.apache.commons.logging.LogFactory;
21 import org.apache.pdfbox.cos.COSArray;
22 import org.apache.pdfbox.cos.COSBase;
23 import org.apache.pdfbox.cos.COSDictionary;
24 import org.apache.pdfbox.cos.COSName;
25 import org.apache.pdfbox.cos.COSNumber;
26 import org.apache.pdfbox.cos.COSStream;
27
28 import org.apache.pdfbox.pdfviewer.PageDrawer;
29 import org.apache.pdfbox.pdmodel.common.COSArrayList;
30 import org.apache.pdfbox.pdmodel.common.COSObjectable;
31 import org.apache.pdfbox.pdmodel.common.PDMetadata;
32 import org.apache.pdfbox.pdmodel.common.PDRectangle;
33 import org.apache.pdfbox.pdmodel.common.PDStream;
34 import org.apache.pdfbox.pdmodel.interactive.action.PDPageAdditionalActions;
35 import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotation;
36 import org.apache.pdfbox.pdmodel.interactive.pagenavigation.PDThreadBead;
37
38 import java.awt.Color;
39 import java.awt.Dimension;
40 import java.awt.Graphics;
41 import java.awt.Graphics2D;
42 import java.awt.image.BufferedImage;
43 import java.awt.image.ImagingOpException;
44 import java.awt.print.PageFormat;
45 import java.awt.print.Printable;
46 import java.awt.print.PrinterException;
47 import java.awt.print.PrinterIOException;
48 import java.io.IOException;
49
50 import java.util.ArrayList;
51 import java.util.Calendar;
52 import java.util.GregorianCalendar;
53 import java.util.List;
54
55 /**
56 * This represents a single page in a PDF document.
57 *
58 * @author <a href="mailto:ben@benlitchfield.com">Ben Litchfield</a>
59 * @version $Revision: 1.29 $
60 */
61 public class PDPage implements COSObjectable, Printable
62 {
63
64 /**
65 * Log instance.
66 */
67 private static final Log log = LogFactory.getLog(PDPage.class);
68
69 private static final int DEFAULT_USER_SPACE_UNIT_DPI = 72;
70
71 private COSDictionary page;
72
73 /**
74 * A page size of LETTER or 8.5x11.
75 */
76 public static final PDRectangle PAGE_SIZE_LETTER = new PDRectangle( 612, 792 );
77 /**
78 * A page size of A0 Paper.
79 */
80 public static final PDRectangle PAGE_SIZE_A0 = new PDRectangle( 2383, 3370 );
81 /**
82 * A page size of A1 Paper.
83 */
84 public static final PDRectangle PAGE_SIZE_A1 = new PDRectangle( 1685, 2383 );
85 /**
86 * A page size of A2 Paper.
87 */
88 public static final PDRectangle PAGE_SIZE_A2 = new PDRectangle( 1192, 1685 );
89 /**
90 * A page size of A3 Paper.
91 */
92 public static final PDRectangle PAGE_SIZE_A3 = new PDRectangle( 843, 1192 );
93 /**
94 * A page size of A4 Paper.
95 */
96 public static final PDRectangle PAGE_SIZE_A4 = new PDRectangle( 596, 843 );
97 /**
98 * A page size of A5 Paper.
99 */
100 public static final PDRectangle PAGE_SIZE_A5 = new PDRectangle( 421, 596 );
101 /**
102 * A page size of A6 Paper.
103 */
104 public static final PDRectangle PAGE_SIZE_A6 = new PDRectangle( 298, 421 );
105
106
107 /**
108 * Creates a new instance of PDPage with a size of 8.5x11.
109 */
110 public PDPage()
111 {
112 page = new COSDictionary();
113 page.setItem( COSName.TYPE, COSName.PAGE );
114 setMediaBox( PAGE_SIZE_LETTER );
115 }
116
117 /**
118 * Creates a new instance of PDPage.
119 *
120 * @param pageDic The existing page dictionary.
121 */
122 public PDPage( COSDictionary pageDic )
123 {
124 page = pageDic;
125 }
126
127 /**
128 * Convert this standard java object to a COS object.
129 *
130 * @return The cos object that matches this Java object.
131 */
132 public COSBase getCOSObject()
133 {
134 return page;
135 }
136
137 /**
138 * This will get the underlying dictionary that this class acts on.
139 *
140 * @return The underlying dictionary for this class.
141 */
142 public COSDictionary getCOSDictionary()
143 {
144 return page;
145 }
146
147
148 /**
149 * This is the parent page node. The parent is a required element of the
150 * page. This will be null until this page is added to the document.
151 *
152 * @return The parent to this page.
153 */
154 public PDPageNode getParent()
155 {
156 if( parent == null){
157 COSDictionary parentDic = (COSDictionary)page.getDictionaryObject( "Parent", "P" );
158 if( parentDic != null )
159 {
160 parent = new PDPageNode( parentDic );
161 }
162 }
163 return parent;
164 }
165
166 private PDPageNode parent = null;
167
168 /**
169 * This will set the parent of this page.
170 *
171 * @param parent The parent to this page node.
172 */
173 public void setParent( PDPageNode parent )
174 {
175 this.parent = parent;
176 page.setItem( COSName.PARENT, parent.getDictionary() );
177 }
178
179 /**
180 * This will update the last modified time for the page object.
181 */
182 public void updateLastModified()
183 {
184 page.setDate( "LastModified", new GregorianCalendar() );
185 }
186
187 /**
188 * This will get the date that the content stream was last modified. This
189 * may return null.
190 *
191 * @return The date the content stream was last modified.
192 *
193 * @throws IOException If there is an error accessing the date information.
194 */
195 public Calendar getLastModified() throws IOException
196 {
197 return page.getDate( "LastModified" );
198 }
199
200 /**
201 * This will get the resources at this page and not look up the hierarchy.
202 * This attribute is inheritable, and findResources() should probably used.
203 * This will return null if no resources are available at this level.
204 *
205 * @return The resources at this level in the hierarchy.
206 */
207 public PDResources getResources()
208 {
209 PDResources retval = null;
210 COSDictionary resources = (COSDictionary)page.getDictionaryObject( COSName.RESOURCES );
211 if( resources != null )
212 {
213 retval = new PDResources( resources );
214 }
215 return retval;
216 }
217
218 /**
219 * This will find the resources for this page by looking up the hierarchy until
220 * it finds them.
221 *
222 * @return The resources at this level in the hierarchy.
223 */
224 public PDResources findResources()
225 {
226 PDResources retval = getResources();
227 PDPageNode parent = getParent();
228 if( retval == null && parent != null )
229 {
230 retval = parent.findResources();
231 }
232 return retval;
233 }
234
235 /**
236 * This will set the resources for this page.
237 *
238 * @param resources The new resources for this page.
239 */
240 public void setResources( PDResources resources )
241 {
242 page.setItem( COSName.RESOURCES, resources );
243 }
244
245 /**
246 * A rectangle, expressed
247 * in default user space units, defining the boundaries of the physical
248 * medium on which the page is intended to be displayed or printed
249 *
250 * This will get the MediaBox at this page and not look up the hierarchy.
251 * This attribute is inheritable, and findMediaBox() should probably used.
252 * This will return null if no MediaBox are available at this level.
253 *
254 * @return The MediaBox at this level in the hierarchy.
255 */
256 public PDRectangle getMediaBox()
257 {
258 if( mediaBox == null){
259 COSArray array = (COSArray)page.getDictionaryObject( COSName.MEDIA_BOX );
260 if( array != null )
261 {
262 mediaBox = new PDRectangle( array );
263 }
264 }
265 return mediaBox;
266 }
267
268 private PDRectangle mediaBox = null;
269
270 /**
271 * This will find the MediaBox for this page by looking up the hierarchy until
272 * it finds them.
273 *
274 * @return The MediaBox at this level in the hierarchy.
275 */
276 public PDRectangle findMediaBox()
277 {
278 PDRectangle retval = getMediaBox();
279 if( retval == null && getParent() != null )
280 {
281 retval = getParent().findMediaBox();
282 }
283 return retval;
284 }
285
286 /**
287 * This will set the mediaBox for this page.
288 *
289 * @param mediaBox The new mediaBox for this page.
290 */
291 public void setMediaBox( PDRectangle mediaBox )
292 {
293 this.mediaBox = mediaBox;
294 if( mediaBox == null )
295 {
296 page.removeItem( COSName.MEDIA_BOX );
297 }
298 else
299 {
300 page.setItem( COSName.MEDIA_BOX, mediaBox.getCOSArray() );
301 }
302 }
303
304 /**
305 * A rectangle, expressed in default user space units,
306 * defining the visible region of default user space. When the page is displayed
307 * or printed, its contents are to be clipped (cropped) to this rectangle
308 * and then imposed on the output medium in some implementationdefined
309 * manner
310 *
311 * This will get the CropBox at this page and not look up the hierarchy.
312 * This attribute is inheritable, and findCropBox() should probably used.
313 * This will return null if no CropBox is available at this level.
314 *
315 * @return The CropBox at this level in the hierarchy.
316 */
317 public PDRectangle getCropBox()
318 {
319 PDRectangle retval = null;
320 COSArray array = (COSArray)page.getDictionaryObject( COSName.CROP_BOX);
321 if( array != null )
322 {
323 retval = new PDRectangle( array );
324 }
325 return retval;
326 }
327
328 /**
329 * This will find the CropBox for this page by looking up the hierarchy until
330 * it finds them.
331 *
332 * @return The CropBox at this level in the hierarchy.
333 */
334 public PDRectangle findCropBox()
335 {
336 PDRectangle retval = getCropBox();
337 PDPageNode parent = getParent();
338 if( retval == null && parent != null )
339 {
340 retval = findParentCropBox( parent );
341 }
342
343 //default value for cropbox is the media box
344 if( retval == null )
345 {
346 retval = findMediaBox();
347 }
348 return retval;
349 }
350
351 /**
352 * This will search for a crop box in the parent and return null if it is not
353 * found. It will NOT default to the media box if it cannot be found.
354 *
355 * @param node The node
356 */
357 private PDRectangle findParentCropBox( PDPageNode node )
358 {
359 PDRectangle rect = node.getCropBox();
360 PDPageNode parent = node.getParent();
361 if( rect == null && parent != null )
362 {
363 rect = findParentCropBox( parent );
364 }
365 return rect;
366 }
367
368 /**
369 * This will set the CropBox for this page.
370 *
371 * @param cropBox The new CropBox for this page.
372 */
373 public void setCropBox( PDRectangle cropBox )
374 {
375 if( cropBox == null )
376 {
377 page.removeItem( COSName.CROP_BOX );
378 }
379 else
380 {
381 page.setItem( COSName.CROP_BOX, cropBox.getCOSArray() );
382 }
383 }
384
385 /**
386 * A rectangle, expressed in default user space units, defining
387 * the region to which the contents of the page should be clipped
388 * when output in a production environment. The default is the CropBox.
389 *
390 * @return The BleedBox attribute.
391 */
392 public PDRectangle getBleedBox()
393 {
394 PDRectangle retval = null;
395 COSArray array = (COSArray)page.getDictionaryObject( COSName.BLEED_BOX );
396 if( array != null )
397 {
398 retval = new PDRectangle( array );
399 }
400 else
401 {
402 retval = findCropBox();
403 }
404 return retval;
405 }
406
407 /**
408 * This will set the BleedBox for this page.
409 *
410 * @param bleedBox The new BleedBox for this page.
411 */
412 public void setBleedBox( PDRectangle bleedBox )
413 {
414 if( bleedBox == null )
415 {
416 page.removeItem( COSName.BLEED_BOX );
417 }
418 else
419 {
420 page.setItem( COSName.BLEED_BOX, bleedBox.getCOSArray() );
421 }
422 }
423
424 /**
425 * A rectangle, expressed in default user space units, defining
426 * the intended dimensions of the finished page after trimming.
427 * The default is the CropBox.
428 *
429 * @return The TrimBox attribute.
430 */
431 public PDRectangle getTrimBox()
432 {
433 PDRectangle retval = null;
434 COSArray array = (COSArray)page.getDictionaryObject( COSName.TRIM_BOX );
435 if( array != null )
436 {
437 retval = new PDRectangle( array );
438 }
439 else
440 {
441 retval = findCropBox();
442 }
443 return retval;
444 }
445
446 /**
447 * This will set the TrimBox for this page.
448 *
449 * @param trimBox The new TrimBox for this page.
450 */
451 public void setTrimBox( PDRectangle trimBox )
452 {
453 if( trimBox == null )
454 {
455 page.removeItem( COSName.TRIM_BOX );
456 }
457 else
458 {
459 page.setItem( COSName.TRIM_BOX, trimBox.getCOSArray() );
460 }
461 }
462
463 /**
464 * A rectangle, expressed in default user space units, defining
465 * the extent of the page's meaningful content (including potential
466 * white space) as intended by the page's creator The default isthe CropBox.
467 *
468 * @return The ArtBox attribute.
469 */
470 public PDRectangle getArtBox()
471 {
472 PDRectangle retval = null;
473 COSArray array = (COSArray)page.getDictionaryObject( COSName.ART_BOX );
474 if( array != null )
475 {
476 retval = new PDRectangle( array );
477 }
478 else
479 {
480 retval = findCropBox();
481 }
482 return retval;
483 }
484
485 /**
486 * This will set the ArtBox for this page.
487 *
488 * @param artBox The new ArtBox for this page.
489 */
490 public void setArtBox( PDRectangle artBox )
491 {
492 if( artBox == null )
493 {
494 page.removeItem( COSName.ART_BOX );
495 }
496 else
497 {
498 page.setItem( COSName.ART_BOX, artBox.getCOSArray() );
499 }
500 }
501
502
503 //todo BoxColorInfo
504 //todo Contents
505
506 /**
507 * A value representing the rotation. This will be null if not set at this level
508 * The number of degrees by which the page should
509 * be rotated clockwise when displayed or printed. The value must be a multiple
510 * of 90.
511 *
512 * This will get the rotation at this page and not look up the hierarchy.
513 * This attribute is inheritable, and findRotation() should probably used.
514 * This will return null if no rotation is available at this level.
515 *
516 * @return The rotation at this level in the hierarchy.
517 */
518 public Integer getRotation()
519 {
520 Integer retval = null;
521 COSNumber value = (COSNumber)page.getDictionaryObject( COSName.ROTATE );
522 if( value != null )
523 {
524 retval = new Integer( value.intValue() );
525 }
526 return retval;
527 }
528
529 /**
530 * This will find the rotation for this page by looking up the hierarchy until
531 * it finds them.
532 *
533 * @return The rotation at this level in the hierarchy.
534 */
535 public int findRotation()
536 {
537 int retval = 0;
538 Integer rotation = getRotation();
539 if( rotation != null )
540 {
541 retval = rotation.intValue();
542 }
543 else
544 {
545 PDPageNode parent = getParent();
546 if( parent != null )
547 {
548 retval = parent.findRotation();
549 }
550 }
551
552 return retval;
553 }
554
555 /**
556 * This will set the rotation for this page.
557 *
558 * @param rotation The new rotation for this page.
559 */
560 public void setRotation( int rotation )
561 {
562 page.setInt( COSName.ROTATE, rotation );
563 }
564
565 /**
566 * This will get the contents of the PDF Page, in the case that the contents
567 * of the page is an array then then the entire array of streams will be
568 * be wrapped and appear as a single stream.
569 *
570 * @return The page content stream.
571 *
572 * @throws IOException If there is an error obtaining the stream.
573 */
574 public PDStream getContents() throws IOException
575 {
576 return PDStream.createFromCOS( page.getDictionaryObject( COSName.CONTENTS ) );
577 }
578
579 /**
580 * This will set the contents of this page.
581 *
582 * @param contents The new contents of the page.
583 */
584 public void setContents( PDStream contents )
585 {
586 page.setItem( COSName.CONTENTS, contents );
587 }
588
589 /**
590 * This will get a list of PDThreadBead objects, which are article threads in the
591 * document. This will return an empty list of there are no thread beads.
592 *
593 * @return A list of article threads on this page.
594 */
595 public List getThreadBeads()
596 {
597 COSArray beads = (COSArray)page.getDictionaryObject( COSName.B );
598 if( beads == null )
599 {
600 beads = new COSArray();
601 }
602 List pdObjects = new ArrayList();
603 for( int i=0; i<beads.size(); i++)
604 {
605 COSDictionary beadDic = (COSDictionary)beads.getObject( i );
606 PDThreadBead bead = null;
607 //in some cases the bead is null
608 if( beadDic != null )
609 {
610 bead = new PDThreadBead( beadDic );
611 }
612 pdObjects.add( bead );
613 }
614 return new COSArrayList(pdObjects, beads);
615
616 }
617
618 /**
619 * This will set the list of thread beads.
620 *
621 * @param beads A list of PDThreadBead objects or null.
622 */
623 public void setThreadBeads( List beads )
624 {
625 page.setItem( COSName.B, COSArrayList.converterToCOSArray( beads ) );
626 }
627
628 /**
629 * Get the metadata that is part of the document catalog. This will
630 * return null if there is no meta data for this object.
631 *
632 * @return The metadata for this object.
633 */
634 public PDMetadata getMetadata()
635 {
636 PDMetadata retval = null;
637 COSStream stream = (COSStream)page.getDictionaryObject( COSName.METADATA );
638 if( stream != null )
639 {
640 retval = new PDMetadata( stream );
641 }
642 return retval;
643 }
644
645 /**
646 * Set the metadata for this object. This can be null.
647 *
648 * @param meta The meta data for this object.
649 */
650 public void setMetadata( PDMetadata meta )
651 {
652 page.setItem( COSName.METADATA, meta );
653 }
654
655 /**
656 * Convert this page to an output image with 8 bits per pixel and the double
657 * default screen resolution.
658 *
659 * @return A graphical representation of this page.
660 *
661 * @throws IOException If there is an error drawing to the image.
662 */
663 public BufferedImage convertToImage() throws IOException
664 {
665 //note we are doing twice as many pixels because
666 //the default size is not really good resolution,
667 //so create an image that is twice the size
668 //and let the client scale it down.
669 return convertToImage(8, 2 * DEFAULT_USER_SPACE_UNIT_DPI);
670 }
671
672 /**
673 * Convert this page to an output image.
674 *
675 * @param imageType the image type (see {@link BufferedImage}.TYPE_*)
676 * @param resolution the resolution in dpi (dots per inch)
677 * @return A graphical representation of this page.
678 *
679 * @throws IOException If there is an error drawing to the image.
680 */
681 public BufferedImage convertToImage(int imageType, int resolution) throws IOException
682 {
683 PDRectangle mBox = findMediaBox();
684 float widthPt = mBox.getWidth();
685 float heightPt = mBox.getHeight();
686 float scaling = resolution / (float)DEFAULT_USER_SPACE_UNIT_DPI;
687 int widthPx = Math.round(widthPt * scaling);
688 int heightPx = Math.round(heightPt * scaling);
689 //TODO The following reduces accuracy. It should really be a Dimension2D.Float.
690 Dimension pageDimension = new Dimension( (int)widthPt, (int)heightPt );
691
692 BufferedImage retval = new BufferedImage( widthPx, heightPx, imageType );
693 Graphics2D graphics = (Graphics2D)retval.getGraphics();
694 graphics.setBackground( Color.WHITE );
695 graphics.clearRect( 0, 0, retval.getWidth(), retval.getHeight() );
696 graphics.scale( scaling, scaling );
697 PageDrawer drawer = new PageDrawer();
698 drawer.drawPage( graphics, this, pageDimension );
699
700 //TODO This could be done directly by manipulating the transformation matrix before painting.
701 //That could result in a better image quality.
702 try
703 {
704 int rotation = findRotation();
705 if (rotation == 90 || rotation == 270)
706 {
707 int w = retval.getWidth();
708 int h = retval.getHeight();
709 BufferedImage rotatedImg = new BufferedImage(w, h, retval.getType());
710 Graphics2D g = rotatedImg.createGraphics();
711 g.rotate(Math.toRadians(rotation), w/2, h/2);
712 g.drawImage(retval, null, 0, 0);
713 }
714 }
715 catch (ImagingOpException e)
716 {
717 log.warn("Unable to rotate page image", e);
718 }
719
720 return retval;
721 }
722
723 /**
724 * Get the page actions.
725 *
726 * @return The Actions for this Page
727 */
728 public PDPageAdditionalActions getActions()
729 {
730 COSDictionary addAct = (COSDictionary) page.getDictionaryObject(COSName.AA);
731 if (addAct == null)
732 {
733 addAct = new COSDictionary();
734 page.setItem(COSName.AA, addAct);
735 }
736 return new PDPageAdditionalActions(addAct);
737 }
738
739 /**
740 * Set the page actions.
741 *
742 * @param actions The actions for the page.
743 */
744 public void setActions( PDPageAdditionalActions actions )
745 {
746 page.setItem( COSName.AA, actions );
747 }
748
749 /**
750 * This will return a list of the Annotations for this page.
751 *
752 * @return List of the PDAnnotation objects.
753 *
754 * @throws IOException If there is an error while creating the annotations.
755 */
756 public List getAnnotations() throws IOException
757 {
758 COSArrayList retval = null;
759 COSArray annots = (COSArray)page.getDictionaryObject(COSName.ANNOTS);
760 if (annots == null)
761 {
762 annots = new COSArray();
763 page.setItem(COSName.ANNOTS, annots);
764 retval = new COSArrayList(new ArrayList(), annots);
765 }
766 else
767 {
768 List actuals = new ArrayList();
769
770 for (int i=0; i < annots.size(); i++)
771 {
772 COSBase item = annots.getObject(i);
773 actuals.add( PDAnnotation.createAnnotation( item ) );
774 }
775 retval = new COSArrayList(actuals, annots);
776 }
777 return retval;
778 }
779
780 /**
781 * This will set the list of annotations.
782 *
783 * @param annots The new list of annotations.
784 */
785 public void setAnnotations( List annots )
786 {
787 page.setItem( COSName.ANNOTS, COSArrayList.converterToCOSArray( annots ) );
788 }
789
790 /**
791 * {@inheritDoc}
792 */
793 public int print(Graphics graphics, PageFormat pageFormat, int pageIndex)
794 throws PrinterException
795 {
796 int retval = Printable.PAGE_EXISTS;
797 try
798 {
799 PageDrawer drawer = new PageDrawer();
800 PDRectangle cropBox = findCropBox();
801 drawer.drawPage( graphics, this, cropBox.createDimension() );
802 }
803 catch( IOException io )
804 {
805 throw new PrinterIOException( io );
806 }
807 return retval;
808 }
809
810 /**
811 * {@inheritDoc}
812 */
813 public boolean equals( Object other )
814 {
815 return other instanceof PDPage && ((PDPage)other).getCOSObject() == this.getCOSObject();
816 }
817
818 /**
819 * {@inheritDoc}
820 */
821 public int hashCode()
822 {
823 return this.getCOSDictionary().hashCode();
824 }
825 }