Modifier and Type	Class and Description
private static class	LosslessFactory.PredictorEncoder

Modifier and Type	Field and Description
pack-priv static final boolean	USE_PREDICTOR_ENCODER Internal, only for benchmark purpose

Access	Constructor and Description
private	LosslessFactory()

Modifier and Type	Method and Description
private static PDImageXObject	createFromGrayImage(BufferedImage image, PDDocument document)
public static PDImageXObject	Returns: a new image XObject createFromImage(PDDocument the document where the image will be created document, BufferedImage the BufferedImage to embed image) Creates a new lossless encoded image XObject from a BufferedImage.
private static PDImageXObject	createFromRGBImage(BufferedImage image, PDDocument document)
private static boolean	isGrayImage(BufferedImage image)
pack-priv static PDImageXObject	Returns: the newly created PDImageXObject with the data compressed. prepareImageXObject(PDDocument The document. document, byte[] array with data. byteArray, int the image width width, int the image height height, int the bits per component bitsPerComponent, PDColorSpace the color space initColorSpace) Create a PDImageXObject using the Flate filter.

USE_PREDICTOR_ENCODER	back to summary
pack-priv static final boolean USE_PREDICTOR_ENCODER Internal, only for benchmark purpose

LosslessFactory	back to summary
private LosslessFactory()

Method Detail

createFromGrayImage	back to summary
private static PDImageXObject createFromGrayImage(BufferedImage image, PDDocument document) throws IOException

createFromImage back to summary

createFromImage	back to summary
public static PDImageXObject createFromImage(PDDocument document, BufferedImage image) throws IOException Creates a new lossless encoded image XObject from a BufferedImage. New for advanced users from 2.0.12 on: If you created your image with a non standard ICC colorspace, it will be preserved. (If you load images in java using ImageIO then no need to read this segment) However a new colorspace will be created for each image. So if you create a PDF with several such images, consider replacing the colorspace with a common object to save space. This is done with `PDImageXObject#getColorSpace()` and `PDImageXObject.setColorSpace()` Parameters document:PDDocument the document where the image will be created image:BufferedImage the BufferedImage to embed Returns:PDImageXObject a new image XObject Exceptions IOException: if something goes wrong

public static PDImageXObject createFromImage(PDDocument document, BufferedImage image) throws IOException

Creates a new lossless encoded image XObject from a BufferedImage.

New for advanced users from 2.0.12 on:
If you created your image with a non standard ICC colorspace, it will be preserved. (If you load images in java using ImageIO then no need to read this segment) However a new colorspace will be created for each image. So if you create a PDF with several such images, consider replacing the colorspace with a common object to save space. This is done with PDImageXObject#getColorSpace() and PDImageXObject.setColorSpace()

Parameters

document:PDDocument: the document where the image will be created
image:BufferedImage: the BufferedImage to embed

Returns:PDImageXObject

a new image XObject

Exceptions

IOException:: if something goes wrong

createFromRGBImage	back to summary
private static PDImageXObject createFromRGBImage(BufferedImage image, PDDocument document) throws IOException

isGrayImage	back to summary
private static boolean isGrayImage(BufferedImage image)

prepareImageXObject	back to summary
pack-priv static PDImageXObject prepareImageXObject(PDDocument document, byte[] byteArray, int width, int height, int bitsPerComponent, PDColorSpace initColorSpace) throws IOException Create a PDImageXObject using the Flate filter. Parameters document:PDDocument The document. byteArray:byte[] array with data. width:int the image width height:int the image height bitsPerComponent:int the bits per component initColorSpace:PDColorSpace the color space Returns:PDImageXObject the newly created PDImageXObject with the data compressed.

prepareImageXObject

back to summary

pack-priv static PDImageXObject prepareImageXObject(PDDocument document, byte[] byteArray, int width, int height, int bitsPerComponent, PDColorSpace initColorSpace) throws IOException

Create a PDImageXObject using the Flate filter.

Parameters

document:PDDocument: The document.
byteArray:byte[]: array with data.
width:int: the image width
height:int: the image height
bitsPerComponent:int: the bits per component
initColorSpace:PDColorSpace: the color space

Returns:PDImageXObject

the newly created PDImageXObject with the data compressed.

org.apache.pdfbox.pdmodel.graphics.image back to summary

private Class LosslessFactory.PredictorEncoder

Class Inheritance

extends Object

java.lang.Object
org.apache.pdfbox.pdmodel.graphics.image.LosslessFactory.PredictorEncoder

Field Summary

Modifier and Type	Field and Description
pack-priv final byte[]	alphaImageData
pack-priv final byte[]	aValues
pack-priv final byte[]	bValues
private final int	bytesPerComponent
private final int	bytesPerPixel
private final int	componentsPerPixel
pack-priv final byte[]	cValues
private final byte[]	dataRawRowAverage
private final byte[]	dataRawRowNone
private final byte[]	dataRawRowPaeth
private final byte[]	dataRawRowSub
private final byte[]	dataRawRowUp
private final PDDocument	document
pack-priv final boolean	hasAlpha
private final int	height
private final BufferedImage	image
pack-priv final int	imageType
pack-priv final byte[]	tmpResultValues
private final int	transferType
private final int	width
pack-priv final byte[]	xValues

Constructor Summary

Access	Constructor and Description
pack-priv	PredictorEncoder(PDDocument document, BufferedImage image) Initialize the encoder and set all final fields

Method Summary

Modifier and Type	Method and Description
private byte[]	Returns: the "best" row encoding of the row encodings chooseDataRowToWrite() We look which row encoding is the "best" one, ie.
private void	copyImageBytes(byte[] transferRow, int indexInTranferRow, byte[] targetValues, byte[] alphaImageData, int alphaPtr)
private void	copyIntToBytes(int[] transferRow, int indexInTranferRow, byte[] targetValues, byte[] alphaImageData, int alphaPtr)
private static void	copyShortsToBytes(short[] transferRow, int indexInTranferRow, byte[] targetValues, byte[] alphaImageData, int alphaPtr)
pack-priv PDImageXObject	Returns: the image or null if it is not possible to encoded the image (e.g. not supported raster format etc.) encode() Tries to compress the image using a predictor.
private static long	estCompressSum(byte[] dataRawRowSub)
private static byte	pngFilterAverage(int x, int a, int b)
private static byte	pngFilterPaeth(int x, int a, int b, int c)
private static byte	pngFilterSub(int x, int a)
private static byte	pngFilterUp(int x, int b)
private PDImageXObject	preparePredictorPDImage(ByteArrayOutputStream stream, int bitsPerComponent)

Inherited from java.lang.Object:: clone equals finalize getClass hashCode notify notifyAll toString wait wait wait

Field Detail

alphaImageData	back to summary
pack-priv final byte[] alphaImageData

aValues	back to summary
pack-priv final byte[] aValues

bValues	back to summary
pack-priv final byte[] bValues

bytesPerComponent	back to summary
private final int bytesPerComponent

bytesPerPixel	back to summary
private final int bytesPerPixel

componentsPerPixel	back to summary
private final int componentsPerPixel

cValues	back to summary
pack-priv final byte[] cValues

dataRawRowAverage	back to summary
private final byte[] dataRawRowAverage

dataRawRowNone	back to summary
private final byte[] dataRawRowNone

dataRawRowPaeth	back to summary
private final byte[] dataRawRowPaeth

dataRawRowSub	back to summary
private final byte[] dataRawRowSub

dataRawRowUp	back to summary
private final byte[] dataRawRowUp

document	back to summary
private final PDDocument document

hasAlpha	back to summary
pack-priv final boolean hasAlpha

height	back to summary
private final int height

image	back to summary
private final BufferedImage image

imageType	back to summary
pack-priv final int imageType

tmpResultValues	back to summary
pack-priv final byte[] tmpResultValues

transferType	back to summary
private final int transferType

width	back to summary
private final int width

xValues	back to summary
pack-priv final byte[] xValues

Constructor Detail

PredictorEncoder	back to summary
pack-priv PredictorEncoder(PDDocument document, BufferedImage image) Initialize the encoder and set all final fields

Method Detail

chooseDataRowToWrite	back to summary
private byte[] chooseDataRowToWrite() We look which row encoding is the "best" one, ie. has the lowest sum. We don't implement anything fancier to choose the right row encoding. This is just the recommend algorithm in the spec. The get the perfect encoding you would need to do a brute force check how all the different encoded rows compress in the zip stream together. You have would have to check 5*image-height permutations... Returns:byte[] the "best" row encoding of the row encodings

chooseDataRowToWrite

back to summary

private byte[] chooseDataRowToWrite()

We look which row encoding is the "best" one, ie. has the lowest sum. We don't implement anything fancier to choose the right row encoding. This is just the recommend algorithm in the spec. The get the perfect encoding you would need to do a brute force check how all the different encoded rows compress in the zip stream together. You have would have to check 5*image-height permutations...

Returns:byte[]: the "best" row encoding of the row encodings

copyImageBytes	back to summary
private void copyImageBytes(byte[] transferRow, int indexInTranferRow, byte[] targetValues, byte[] alphaImageData, int alphaPtr)

copyIntToBytes	back to summary
private void copyIntToBytes(int[] transferRow, int indexInTranferRow, byte[] targetValues, byte[] alphaImageData, int alphaPtr)

copyShortsToBytes	back to summary
private static void copyShortsToBytes(short[] transferRow, int indexInTranferRow, byte[] targetValues, byte[] alphaImageData, int alphaPtr)

encode	back to summary
pack-priv PDImageXObject encode() throws IOException Tries to compress the image using a predictor. Returns:PDImageXObject the image or null if it is not possible to encoded the image (e.g. not supported raster format etc.)

encode

back to summary

pack-priv PDImageXObject encode() throws IOException

Tries to compress the image using a predictor.

Returns:PDImageXObject: the image or null if it is not possible to encoded the image (e.g. not supported raster format etc.)

estCompressSum	back to summary
private static long estCompressSum(byte[] dataRawRowSub)

pngFilterAverage	back to summary
private static byte pngFilterAverage(int x, int a, int b)

pngFilterPaeth	back to summary
private static byte pngFilterPaeth(int x, int a, int b, int c)

pngFilterSub	back to summary
private static byte pngFilterSub(int x, int a)

pngFilterUp	back to summary
private static byte pngFilterUp(int x, int b)

preparePredictorPDImage	back to summary
private PDImageXObject preparePredictorPDImage(ByteArrayOutputStream stream, int bitsPerComponent) throws IOException

public final Class LosslessFactory

Nested and Inner Type Summary

Field Summary

Constructor Summary

Method Summary

Field Detail

Constructor Detail

Method Detail

private Class LosslessFactory.PredictorEncoder

Field Summary

Constructor Summary

Method Summary

Field Detail

Constructor Detail

Method Detail