/* * Copyright (c) 1997, 2014, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License version 2 only, as * published by the Free Software Foundation. Oracle designates this * particular file as subject to the "Classpath" exception as provided * by Oracle in the LICENSE file that accompanied this code. * * This code is distributed in the hope that it will be useful, but WITHOUT * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License * version 2 for more details (a copy is included in the LICENSE file that * accompanied this code). * * You should have received a copy of the GNU General Public License version * 2 along with this work; if not, write to the Free Software Foundation, * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. * * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA * or visit www.oracle.com if you need additional information or have any * questions. */ package java.awt.image; import java.awt.color.ColorSpace; import java.awt.color.ICC_ColorSpace; /** * A {@code ColorModel} class that works with pixel values that * represent color and alpha information as separate samples and that * store each sample in a separate data element. This class can be * used with an arbitrary {@code ColorSpace}. The number of * color samples in the pixel values must be same as the number of * color components in the {@code ColorSpace}. There may be a * single alpha sample. *

* For those methods that use * a primitive array pixel representation of type {@code transferType}, * the array length is the same as the number of color and alpha samples. * Color samples are stored first in the array followed by the alpha * sample, if present. The order of the color samples is specified * by the {@code ColorSpace}. Typically, this order reflects the * name of the color space type. For example, for {@code TYPE_RGB}, * index 0 corresponds to red, index 1 to green, and index 2 to blue. *

* The translation from pixel sample values to color/alpha components for * display or processing purposes is based on a one-to-one correspondence of * samples to components. * Depending on the transfer type used to create an instance of * {@code ComponentColorModel}, the pixel sample values * represented by that instance may be signed or unsigned and may * be of integral type or float or double (see below for details). * The translation from sample values to normalized color/alpha components * must follow certain rules. For float and double samples, the translation * is an identity, i.e. normalized component values are equal to the * corresponding sample values. For integral samples, the translation * should be only a simple scale and offset, where the scale and offset * constants may be different for each component. The result of * applying the scale and offset constants is a set of color/alpha * component values, which are guaranteed to fall within a certain * range. Typically, the range for a color component will be the range * defined by the {@code getMinValue} and {@code getMaxValue} * methods of the {@code ColorSpace} class. The range for an * alpha component should be 0.0 to 1.0. *

* Instances of {@code ComponentColorModel} created with transfer types * {@code DataBuffer.TYPE_BYTE}, {@code DataBuffer.TYPE_USHORT}, * and {@code DataBuffer.TYPE_INT} have pixel sample values which * are treated as unsigned integral values. * The number of bits in a color or alpha sample of a pixel value might not * be the same as the number of bits for the corresponding color or alpha * sample passed to the * {@code ComponentColorModel(ColorSpace, int[], boolean, boolean, int, int)} * constructor. In * that case, this class assumes that the least significant n bits of a sample * value hold the component value, where n is the number of significant bits * for the component passed to the constructor. It also assumes that * any higher-order bits in a sample value are zero. Thus, sample values * range from 0 to 2n - 1. This class maps these sample values * to normalized color component values such that 0 maps to the value * obtained from the {@code ColorSpace's getMinValue} * method for each component and 2n - 1 maps to the value * obtained from {@code getMaxValue}. To create a * {@code ComponentColorModel} with a different color sample mapping * requires subclassing this class and overriding the * {@code getNormalizedComponents(Object, float[], int)} method. * The mapping for an alpha sample always maps 0 to 0.0 and * 2n - 1 to 1.0. *

* For instances with unsigned sample values, * the unnormalized color/alpha component representation is only * supported if two conditions hold. First, sample value 0 must * map to normalized component value 0.0 and sample value 2n - 1 * to 1.0. Second the min/max range of all color components of the * {@code ColorSpace} must be 0.0 to 1.0. In this case, the * component representation is the n least * significant bits of the corresponding sample. Thus each component is * an unsigned integral value between 0 and 2n - 1, where * n is the number of significant bits for a particular component. * If these conditions are not met, any method taking an unnormalized * component argument will throw an {@code IllegalArgumentException}. *

* Instances of {@code ComponentColorModel} created with transfer types * {@code DataBuffer.TYPE_SHORT}, {@code DataBuffer.TYPE_FLOAT}, and * {@code DataBuffer.TYPE_DOUBLE} have pixel sample values which * are treated as signed short, float, or double values. * Such instances do not support the unnormalized color/alpha component * representation, so any methods taking such a representation as an argument * will throw an {@code IllegalArgumentException} when called on one * of these instances. The normalized component values of instances * of this class have a range which depends on the transfer * type as follows: for float samples, the full range of the float data * type; for double samples, the full range of the float data type * (resulting from casting double to float); for short samples, * from approximately -maxVal to +maxVal, where maxVal is the per * component maximum value for the {@code ColorSpace} * (-32767 maps to -maxVal, 0 maps to 0.0, and 32767 maps * to +maxVal). A subclass may override the scaling for short sample * values to normalized component values by overriding the * {@code getNormalizedComponents(Object, float[], int)} method. * For float and double samples, the normalized component values are * taken to be equal to the corresponding sample values, and subclasses * should not attempt to add any non-identity scaling for these transfer * types. *

* Instances of {@code ComponentColorModel} created with transfer types * {@code DataBuffer.TYPE_SHORT}, {@code DataBuffer.TYPE_FLOAT}, and * {@code DataBuffer.TYPE_DOUBLE} * use all the bits of all sample values. Thus all color/alpha components * have 16 bits when using {@code DataBuffer.TYPE_SHORT}, 32 bits when * using {@code DataBuffer.TYPE_FLOAT}, and 64 bits when using * {@code DataBuffer.TYPE_DOUBLE}. When the * {@code ComponentColorModel(ColorSpace, int[], boolean, boolean, int, int)} * form of constructor is used with one of these transfer types, the * bits array argument is ignored. *

* It is possible to have color/alpha sample values * which cannot be reasonably interpreted as component values for rendering. * This can happen when {@code ComponentColorModel} is subclassed to * override the mapping of unsigned sample values to normalized color * component values or when signed sample values outside a certain range * are used. (As an example, specifying an alpha component as a signed * short value outside the range 0 to 32767, normalized range 0.0 to 1.0, can * lead to unexpected results.) It is the * responsibility of applications to appropriately scale pixel data before * rendering such that color components fall within the normalized range * of the {@code ColorSpace} (obtained using the {@code getMinValue} * and {@code getMaxValue} methods of the {@code ColorSpace} class) * and the alpha component is between 0.0 and 1.0. If color or alpha * component values fall outside these ranges, rendering results are * indeterminate. *

* Methods that use a single int pixel representation throw * an {@code IllegalArgumentException}, unless the number of components * for the {@code ComponentColorModel} is one and the component * value is unsigned -- in other words, a single color component using * a transfer type of {@code DataBuffer.TYPE_BYTE}, * {@code DataBuffer.TYPE_USHORT}, or {@code DataBuffer.TYPE_INT} * and no alpha. *

* A {@code ComponentColorModel} can be used in conjunction with a * {@code ComponentSampleModel}, a {@code BandedSampleModel}, * or a {@code PixelInterleavedSampleModel} to construct a * {@code BufferedImage}. * * @see ColorModel * @see ColorSpace * @see ComponentSampleModel * @see BandedSampleModel * @see PixelInterleavedSampleModel * @see BufferedImage * */ public class ComponentColorModel extends ColorModel { /** * {@code signed} is {@code true} for {@code short}, * {@code float}, and {@code double} transfer types; it * is {@code false} for {@code byte}, {@code ushort}, * and {@code int} transfer types. */ private boolean signed; // true for transfer types short, float, double // false for byte, ushort, int private boolean is_sRGB_stdScale; private boolean is_LinearRGB_stdScale; private boolean is_LinearGray_stdScale; private boolean is_ICCGray_stdScale; private byte[] tosRGB8LUT; private byte[] fromsRGB8LUT8; private short[] fromsRGB8LUT16; private byte[] fromLinearGray16ToOtherGray8LUT; private short[] fromLinearGray16ToOtherGray16LUT; private boolean needScaleInit; private boolean noUnnorm; private boolean nonStdScale; private float[] min; private float[] diffMinMax; private float[] compOffset; private float[] compScale; /** * Constructs a {@code ComponentColorModel} from the specified * parameters. Color components will be in the specified * {@code ColorSpace}. The supported transfer types are * {@code DataBuffer.TYPE_BYTE}, {@code DataBuffer.TYPE_USHORT}, * {@code DataBuffer.TYPE_INT}, * {@code DataBuffer.TYPE_SHORT}, {@code DataBuffer.TYPE_FLOAT}, * and {@code DataBuffer.TYPE_DOUBLE}. * If not null, the {@code bits} array specifies the * number of significant bits per color and alpha component and its * length should be at least the number of components in the * {@code ColorSpace} if there is no alpha * information in the pixel values, or one more than this number if * there is alpha information. When the {@code transferType} is * {@code DataBuffer.TYPE_SHORT}, {@code DataBuffer.TYPE_FLOAT}, * or {@code DataBuffer.TYPE_DOUBLE} the {@code bits} array * argument is ignored. {@code hasAlpha} indicates whether alpha * information is present. If {@code hasAlpha} is true, then * the boolean {@code isAlphaPremultiplied} * specifies how to interpret color and alpha samples in pixel values. * If the boolean is true, color samples are assumed to have been * multiplied by the alpha sample. The {@code transparency} * specifies what alpha values can be represented by this color model. * The acceptable {@code transparency} values are * {@code OPAQUE}, {@code BITMASK} or {@code TRANSLUCENT}. * The {@code transferType} is the type of primitive array used * to represent pixel values. * * @param colorSpace The {@code ColorSpace} associated * with this color model. * @param bits The number of significant bits per component. * May be null, in which case all bits of all * component samples will be significant. * Ignored if transferType is one of * {@code DataBuffer.TYPE_SHORT}, * {@code DataBuffer.TYPE_FLOAT}, or * {@code DataBuffer.TYPE_DOUBLE}, * in which case all bits of all component * samples will be significant. * @param hasAlpha If true, this color model supports alpha. * @param isAlphaPremultiplied If true, alpha is premultiplied. * @param transparency Specifies what alpha values can be represented * by this color model. * @param transferType Specifies the type of primitive array used to * represent pixel values. * * @throws IllegalArgumentException If the {@code bits} array * argument is not null, its length is less than the number of * color and alpha components, and transferType is one of * {@code DataBuffer.TYPE_BYTE}, * {@code DataBuffer.TYPE_USHORT}, or * {@code DataBuffer.TYPE_INT}. * @throws IllegalArgumentException If transferType is not one of * {@code DataBuffer.TYPE_BYTE}, * {@code DataBuffer.TYPE_USHORT}, * {@code DataBuffer.TYPE_INT}, * {@code DataBuffer.TYPE_SHORT}, * {@code DataBuffer.TYPE_FLOAT}, or * {@code DataBuffer.TYPE_DOUBLE}. * * @see ColorSpace * @see java.awt.Transparency */ public ComponentColorModel (ColorSpace colorSpace, int[] bits, boolean hasAlpha, boolean isAlphaPremultiplied, int transparency, int transferType) { super (bitsHelper(transferType, colorSpace, hasAlpha), bitsArrayHelper(bits, transferType, colorSpace, hasAlpha), colorSpace, hasAlpha, isAlphaPremultiplied, transparency, transferType); switch(transferType) { case DataBuffer.TYPE_BYTE: case DataBuffer.TYPE_USHORT: case DataBuffer.TYPE_INT: signed = false; needScaleInit = true; break; case DataBuffer.TYPE_SHORT: signed = true; needScaleInit = true; break; case DataBuffer.TYPE_FLOAT: case DataBuffer.TYPE_DOUBLE: signed = true; needScaleInit = false; noUnnorm = true; nonStdScale = false; break; default: throw new IllegalArgumentException("This constructor is not "+ "compatible with transferType " + transferType); } setupLUTs(); } /** * Constructs a {@code ComponentColorModel} from the specified * parameters. Color components will be in the specified * {@code ColorSpace}. The supported transfer types are * {@code DataBuffer.TYPE_BYTE}, {@code DataBuffer.TYPE_USHORT}, * {@code DataBuffer.TYPE_INT}, * {@code DataBuffer.TYPE_SHORT}, {@code DataBuffer.TYPE_FLOAT}, * and {@code DataBuffer.TYPE_DOUBLE}. The number of significant * bits per color and alpha component will be 8, 16, 32, 16, 32, or 64, * respectively. The number of color components will be the * number of components in the {@code ColorSpace}. There will be * an alpha component if {@code hasAlpha} is {@code true}. * If {@code hasAlpha} is true, then * the boolean {@code isAlphaPremultiplied} * specifies how to interpret color and alpha samples in pixel values. * If the boolean is true, color samples are assumed to have been * multiplied by the alpha sample. The {@code transparency} * specifies what alpha values can be represented by this color model. * The acceptable {@code transparency} values are * {@code OPAQUE}, {@code BITMASK} or {@code TRANSLUCENT}. * The {@code transferType} is the type of primitive array used * to represent pixel values. * * @param colorSpace The {@code ColorSpace} associated * with this color model. * @param hasAlpha If true, this color model supports alpha. * @param isAlphaPremultiplied If true, alpha is premultiplied. * @param transparency Specifies what alpha values can be represented * by this color model. * @param transferType Specifies the type of primitive array used to * represent pixel values. * * @throws IllegalArgumentException If transferType is not one of * {@code DataBuffer.TYPE_BYTE}, * {@code DataBuffer.TYPE_USHORT}, * {@code DataBuffer.TYPE_INT}, * {@code DataBuffer.TYPE_SHORT}, * {@code DataBuffer.TYPE_FLOAT}, or * {@code DataBuffer.TYPE_DOUBLE}. * * @see ColorSpace * @see java.awt.Transparency * @since 1.4 */ public ComponentColorModel (ColorSpace colorSpace, boolean hasAlpha, boolean isAlphaPremultiplied, int transparency, int transferType) { this(colorSpace, null, hasAlpha, isAlphaPremultiplied, transparency, transferType); } private static int bitsHelper(int transferType, ColorSpace colorSpace, boolean hasAlpha) { int numBits = DataBuffer.getDataTypeSize(transferType); int numComponents = colorSpace.getNumComponents(); if (hasAlpha) { ++numComponents; } return numBits * numComponents; } private static int[] bitsArrayHelper(int[] origBits, int transferType, ColorSpace colorSpace, boolean hasAlpha) { switch(transferType) { case DataBuffer.TYPE_BYTE: case DataBuffer.TYPE_USHORT: case DataBuffer.TYPE_INT: if (origBits != null) { return origBits; } break; default: break; } int numBits = DataBuffer.getDataTypeSize(transferType); int numComponents = colorSpace.getNumComponents(); if (hasAlpha) { ++numComponents; } int[] bits = new int[numComponents]; for (int i = 0; i < numComponents; i++) { bits[i] = numBits; } return bits; } private void setupLUTs() { // REMIND: there is potential to accelerate sRGB, LinearRGB, // LinearGray, ICCGray, and non-ICC Gray spaces with non-standard // scaling, if that becomes important // // NOTE: The is_xxx_stdScale and nonStdScale booleans are provisionally // set here when this method is called at construction time. These // variables may be set again when initScale is called later. // When setupLUTs returns, nonStdScale is true if (the transferType // is not float or double) AND (some minimum ColorSpace component // value is not 0.0 OR some maximum ColorSpace component value // is not 1.0). This is correct for the calls to // getNormalizedComponents(Object, float[], int) from initScale(). // initScale() may change the value nonStdScale based on the // return value of getNormalizedComponents() - this will only // happen if getNormalizedComponents() has been overridden by a // subclass to make the mapping of min/max pixel sample values // something different from min/max color component values. if (is_sRGB) { is_sRGB_stdScale = true; nonStdScale = false; } else if (ColorModel.isLinearRGBspace(colorSpace)) { // Note that the built-in Linear RGB space has a normalized // range of 0.0 - 1.0 for each coordinate. Usage of these // LUTs makes that assumption. is_LinearRGB_stdScale = true; nonStdScale = false; if (transferType == DataBuffer.TYPE_BYTE) { tosRGB8LUT = ColorModel.getLinearRGB8TosRGB8LUT(); fromsRGB8LUT8 = ColorModel.getsRGB8ToLinearRGB8LUT(); } else { tosRGB8LUT = ColorModel.getLinearRGB16TosRGB8LUT(); fromsRGB8LUT16 = ColorModel.getsRGB8ToLinearRGB16LUT(); } } else if ((colorSpaceType == ColorSpace.TYPE_GRAY) && (colorSpace instanceof ICC_ColorSpace) && (colorSpace.getMinValue(0) == 0.0f) && (colorSpace.getMaxValue(0) == 1.0f)) { // Note that a normalized range of 0.0 - 1.0 for the gray // component is required, because usage of these LUTs makes // that assumption. ICC_ColorSpace ics = (ICC_ColorSpace) colorSpace; is_ICCGray_stdScale = true; nonStdScale = false; fromsRGB8LUT16 = ColorModel.getsRGB8ToLinearRGB16LUT(); if (ColorModel.isLinearGRAYspace(ics)) { is_LinearGray_stdScale = true; if (transferType == DataBuffer.TYPE_BYTE) { tosRGB8LUT = ColorModel.getGray8TosRGB8LUT(ics); } else { tosRGB8LUT = ColorModel.getGray16TosRGB8LUT(ics); } } else { if (transferType == DataBuffer.TYPE_BYTE) { tosRGB8LUT = ColorModel.getGray8TosRGB8LUT(ics); fromLinearGray16ToOtherGray8LUT = ColorModel.getLinearGray16ToOtherGray8LUT(ics); } else { tosRGB8LUT = ColorModel.getGray16TosRGB8LUT(ics); fromLinearGray16ToOtherGray16LUT = ColorModel.getLinearGray16ToOtherGray16LUT(ics); } } } else if (needScaleInit) { // if transferType is byte, ushort, int, or short and we // don't already know the ColorSpace has minVlaue == 0.0f and // maxValue == 1.0f for all components, we need to check that // now and setup the min[] and diffMinMax[] arrays if necessary. nonStdScale = false; for (int i = 0; i < numColorComponents; i++) { if ((colorSpace.getMinValue(i) != 0.0f) || (colorSpace.getMaxValue(i) != 1.0f)) { nonStdScale = true; break; } } if (nonStdScale) { min = new float[numColorComponents]; diffMinMax = new float[numColorComponents]; for (int i = 0; i < numColorComponents; i++) { min[i] = colorSpace.getMinValue(i); diffMinMax[i] = colorSpace.getMaxValue(i) - min[i]; } } } } private void initScale() { // This method is called the first time any method which uses // pixel sample value to color component value scaling information // is called if the transferType supports non-standard scaling // as defined above (byte, ushort, int, and short), unless the // method is getNormalizedComponents(Object, float[], int) (that // method must be overridden to use non-standard scaling). This // method also sets up the noUnnorm boolean variable for these // transferTypes. After this method is called, the nonStdScale // variable will be true if getNormalizedComponents() maps a // sample value of 0 to anything other than 0.0f OR maps a // sample value of 2^^n - 1 (2^^15 - 1 for short transferType) // to anything other than 1.0f. Note that this can be independent // of the colorSpace min/max component values, if the // getNormalizedComponents() method has been overridden for some // reason, e.g. to provide greater dynamic range in the sample // values than in the color component values. Unfortunately, // this method can't be called at construction time, since a // subclass may still have uninitialized state that would cause // getNormalizedComponents() to return an incorrect result. needScaleInit = false; // only needs to called once if (nonStdScale || signed) { // The unnormalized form is only supported for unsigned // transferTypes and when the ColorSpace min/max values // are 0.0/1.0. When this method is called nonStdScale is // true if the latter condition does not hold. In addition, // the unnormalized form requires that the full range of // the pixel sample values map to the full 0.0 - 1.0 range // of color component values. That condition is checked // later in this method. noUnnorm = true; } else { noUnnorm = false; } float[] lowVal, highVal; switch (transferType) { case DataBuffer.TYPE_BYTE: { byte[] bpixel = new byte[numComponents]; for (int i = 0; i < numColorComponents; i++) { bpixel[i] = 0; } if (supportsAlpha) { bpixel[numColorComponents] = (byte) ((1 << nBits[numColorComponents]) - 1); } lowVal = getNormalizedComponents(bpixel, null, 0); for (int i = 0; i < numColorComponents; i++) { bpixel[i] = (byte) ((1 << nBits[i]) - 1); } highVal = getNormalizedComponents(bpixel, null, 0); } break; case DataBuffer.TYPE_USHORT: { short[] uspixel = new short[numComponents]; for (int i = 0; i < numColorComponents; i++) { uspixel[i] = 0; } if (supportsAlpha) { uspixel[numColorComponents] = (short) ((1 << nBits[numColorComponents]) - 1); } lowVal = getNormalizedComponents(uspixel, null, 0); for (int i = 0; i < numColorComponents; i++) { uspixel[i] = (short) ((1 << nBits[i]) - 1); } highVal = getNormalizedComponents(uspixel, null, 0); } break; case DataBuffer.TYPE_INT: { int[] ipixel = new int[numComponents]; for (int i = 0; i < numColorComponents; i++) { ipixel[i] = 0; } if (supportsAlpha) { ipixel[numColorComponents] = ((1 << nBits[numColorComponents]) - 1); } lowVal = getNormalizedComponents(ipixel, null, 0); for (int i = 0; i < numColorComponents; i++) { ipixel[i] = ((1 << nBits[i]) - 1); } highVal = getNormalizedComponents(ipixel, null, 0); } break; case DataBuffer.TYPE_SHORT: { short[] spixel = new short[numComponents]; for (int i = 0; i < numColorComponents; i++) { spixel[i] = 0; } if (supportsAlpha) { spixel[numColorComponents] = 32767; } lowVal = getNormalizedComponents(spixel, null, 0); for (int i = 0; i < numColorComponents; i++) { spixel[i] = 32767; } highVal = getNormalizedComponents(spixel, null, 0); } break; default: lowVal = highVal = null; // to keep the compiler from complaining break; } nonStdScale = false; for (int i = 0; i < numColorComponents; i++) { if ((lowVal[i] != 0.0f) || (highVal[i] != 1.0f)) { nonStdScale = true; break; } } if (nonStdScale) { noUnnorm = true; is_sRGB_stdScale = false; is_LinearRGB_stdScale = false; is_LinearGray_stdScale = false; is_ICCGray_stdScale = false; compOffset = new float[numColorComponents]; compScale = new float[numColorComponents]; for (int i = 0; i < numColorComponents; i++) { compOffset[i] = lowVal[i]; compScale[i] = 1.0f / (highVal[i] - lowVal[i]); } } } private int getRGBComponent(int pixel, int idx) { if (numComponents > 1) { throw new IllegalArgumentException("More than one component per pixel"); } if (signed) { throw new IllegalArgumentException("Component value is signed"); } if (needScaleInit) { initScale(); } // Since there is only 1 component, there is no alpha // Normalize the pixel in order to convert it Object opixel = null; switch (transferType) { case DataBuffer.TYPE_BYTE: { byte[] bpixel = { (byte) pixel }; opixel = bpixel; } break; case DataBuffer.TYPE_USHORT: { short[] spixel = { (short) pixel }; opixel = spixel; } break; case DataBuffer.TYPE_INT: { int[] ipixel = { pixel }; opixel = ipixel; } break; } float[] norm = getNormalizedComponents(opixel, null, 0); float[] rgb = colorSpace.toRGB(norm); return (int) (rgb[idx] * 255.0f + 0.5f); } /** * Returns the red color component for the specified pixel, scaled * from 0 to 255 in the default RGB ColorSpace, sRGB. A color conversion * is done if necessary. The pixel value is specified as an int. * The returned value will be a non pre-multiplied value. * If the alpha is premultiplied, this method divides * it out before returning the value (if the alpha value is 0, * the red value will be 0). * * @param pixel The pixel from which you want to get the red color component. * * @return The red color component for the specified pixel, as an int. * * @throws IllegalArgumentException If there is more than * one component in this {@code ColorModel}. * @throws IllegalArgumentException If the component value for this * {@code ColorModel} is signed */ public int getRed(int pixel) { return getRGBComponent(pixel, 0); } /** * Returns the green color component for the specified pixel, scaled * from 0 to 255 in the default RGB ColorSpace, sRGB. A color conversion * is done if necessary. The pixel value is specified as an int. * The returned value will be a non * pre-multiplied value. If the alpha is premultiplied, this method * divides it out before returning the value (if the alpha value is 0, * the green value will be 0). * * @param pixel The pixel from which you want to get the green color component. * * @return The green color component for the specified pixel, as an int. * * @throws IllegalArgumentException If there is more than * one component in this {@code ColorModel}. * @throws IllegalArgumentException If the component value for this * {@code ColorModel} is signed */ public int getGreen(int pixel) { return getRGBComponent(pixel, 1); } /** * Returns the blue color component for the specified pixel, scaled * from 0 to 255 in the default RGB ColorSpace, sRGB. A color conversion * is done if necessary. The pixel value is specified as an int. * The returned value will be a non * pre-multiplied value. If the alpha is premultiplied, this method * divides it out before returning the value (if the alpha value is 0, * the blue value will be 0). * * @param pixel The pixel from which you want to get the blue color component. * * @return The blue color component for the specified pixel, as an int. * * @throws IllegalArgumentException If there is more than * one component in this {@code ColorModel}. * @throws IllegalArgumentException If the component value for this * {@code ColorModel} is signed */ public int getBlue(int pixel) { return getRGBComponent(pixel, 2); } /** * Returns the alpha component for the specified pixel, scaled * from 0 to 255. The pixel value is specified as an int. * * @param pixel The pixel from which you want to get the alpha component. * * @return The alpha component for the specified pixel, as an int. * * @throws IllegalArgumentException If there is more than * one component in this {@code ColorModel}. * @throws IllegalArgumentException If the component value for this * {@code ColorModel} is signed */ public int getAlpha(int pixel) { if (supportsAlpha == false) { return 255; } if (numComponents > 1) { throw new IllegalArgumentException("More than one component per pixel"); } if (signed) { throw new IllegalArgumentException("Component value is signed"); } return (int) ((((float) pixel) / ((1< 1) { throw new IllegalArgumentException("More than one component per pixel"); } if (signed) { throw new IllegalArgumentException("Component value is signed"); } return (getAlpha(pixel) << 24) | (getRed(pixel) << 16) | (getGreen(pixel) << 8) | (getBlue(pixel) << 0); } private int extractComponent(Object inData, int idx, int precision) { // Extract component idx from inData. The precision argument // should be either 8 or 16. If it's 8, this method will return // an 8-bit value. If it's 16, this method will return a 16-bit // value for transferTypes other than TYPE_BYTE. For TYPE_BYTE, // an 8-bit value will be returned. // This method maps the input value corresponding to a // normalized ColorSpace component value of 0.0 to 0, and the // input value corresponding to a normalized ColorSpace // component value of 1.0 to 2^n - 1 (where n is 8 or 16), so // it is appropriate only for ColorSpaces with min/max component // values of 0.0/1.0. This will be true for sRGB, the built-in // Linear RGB and Linear Gray spaces, and any other ICC grayscale // spaces for which we have precomputed LUTs. boolean needAlpha = (supportsAlpha && isAlphaPremultiplied); int alp = 0; int comp; int mask = (1 << nBits[idx]) - 1; switch (transferType) { // Note: we do no clamping of the pixel data here - we // assume that the data is scaled properly case DataBuffer.TYPE_SHORT: { short sdata[] = (short[]) inData; float scalefactor = (float) ((1 << precision) - 1); if (needAlpha) { short s = sdata[numColorComponents]; if (s != (short) 0) { return (int) ((((float) sdata[idx]) / ((float) s)) * scalefactor + 0.5f); } else { return 0; } } else { return (int) ((sdata[idx] / 32767.0f) * scalefactor + 0.5f); } } case DataBuffer.TYPE_FLOAT: { float fdata[] = (float[]) inData; float scalefactor = (float) ((1 << precision) - 1); if (needAlpha) { float f = fdata[numColorComponents]; if (f != 0.0f) { return (int) (((fdata[idx] / f) * scalefactor) + 0.5f); } else { return 0; } } else { return (int) (fdata[idx] * scalefactor + 0.5f); } } case DataBuffer.TYPE_DOUBLE: { double ddata[] = (double[]) inData; double scalefactor = (double) ((1 << precision) - 1); if (needAlpha) { double d = ddata[numColorComponents]; if (d != 0.0) { return (int) (((ddata[idx] / d) * scalefactor) + 0.5); } else { return 0; } } else { return (int) (ddata[idx] * scalefactor + 0.5); } } case DataBuffer.TYPE_BYTE: byte bdata[] = (byte[])inData; comp = bdata[idx] & mask; precision = 8; if (needAlpha) { alp = bdata[numColorComponents] & mask; } break; case DataBuffer.TYPE_USHORT: short usdata[] = (short[])inData; comp = usdata[idx] & mask; if (needAlpha) { alp = usdata[numColorComponents] & mask; } break; case DataBuffer.TYPE_INT: int idata[] = (int[])inData; comp = idata[idx]; if (needAlpha) { alp = idata[numColorComponents]; } break; default: throw new UnsupportedOperationException("This method has not "+ "been implemented for transferType " + transferType); } if (needAlpha) { if (alp != 0) { float scalefactor = (float) ((1 << precision) - 1); float fcomp = ((float) comp) / ((float)mask); float invalp = ((float) ((1<>16) & 0xff; grn = (rgb>>8) & 0xff; blu = rgb & 0xff; if (needScaleInit) { initScale(); } if (signed) { // Handle SHORT, FLOAT, & DOUBLE here switch(transferType) { case DataBuffer.TYPE_SHORT: { short sdata[]; if (pixel == null) { sdata = new short[numComponents]; } else { sdata = (short[])pixel; } float factor; if (is_sRGB_stdScale || is_LinearRGB_stdScale) { factor = 32767.0f / 255.0f; if (is_LinearRGB_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; factor = 32767.0f / 65535.0f; } if (supportsAlpha) { alp = (rgb>>24) & 0xff; sdata[3] = (short) (alp * (32767.0f / 255.0f) + 0.5f); if (isAlphaPremultiplied) { factor = alp * factor * (1.0f / 255.0f); } } sdata[0] = (short) (red * factor + 0.5f); sdata[1] = (short) (grn * factor + 0.5f); sdata[2] = (short) (blu * factor + 0.5f); } else if (is_LinearGray_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; float gray = ((0.2125f * red) + (0.7154f * grn) + (0.0721f * blu)) / 65535.0f; factor = 32767.0f; if (supportsAlpha) { alp = (rgb>>24) & 0xff; sdata[1] = (short) (alp * (32767.0f / 255.0f) + 0.5f); if (isAlphaPremultiplied) { factor = alp * factor * (1.0f / 255.0f); } } sdata[0] = (short) (gray * factor + 0.5f); } else if (is_ICCGray_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; int gray = (int) ((0.2125f * red) + (0.7154f * grn) + (0.0721f * blu) + 0.5f); gray = fromLinearGray16ToOtherGray16LUT[gray] & 0xffff; factor = 32767.0f / 65535.0f; if (supportsAlpha) { alp = (rgb>>24) & 0xff; sdata[1] = (short) (alp * (32767.0f / 255.0f) + 0.5f); if (isAlphaPremultiplied) { factor = alp * factor * (1.0f / 255.0f); } } sdata[0] = (short) (gray * factor + 0.5f); } else { factor = 1.0f / 255.0f; float norm[] = new float[3]; norm[0] = red * factor; norm[1] = grn * factor; norm[2] = blu * factor; norm = colorSpace.fromRGB(norm); if (nonStdScale) { for (int i = 0; i < numColorComponents; i++) { norm[i] = (norm[i] - compOffset[i]) * compScale[i]; // REMIND: need to analyze whether this // clamping is necessary if (norm[i] < 0.0f) { norm[i] = 0.0f; } if (norm[i] > 1.0f) { norm[i] = 1.0f; } } } factor = 32767.0f; if (supportsAlpha) { alp = (rgb>>24) & 0xff; sdata[numColorComponents] = (short) (alp * (32767.0f / 255.0f) + 0.5f); if (isAlphaPremultiplied) { factor *= alp * (1.0f / 255.0f); } } for (int i = 0; i < numColorComponents; i++) { sdata[i] = (short) (norm[i] * factor + 0.5f); } } return sdata; } case DataBuffer.TYPE_FLOAT: { float fdata[]; if (pixel == null) { fdata = new float[numComponents]; } else { fdata = (float[])pixel; } float factor; if (is_sRGB_stdScale || is_LinearRGB_stdScale) { if (is_LinearRGB_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; factor = 1.0f / 65535.0f; } else { factor = 1.0f / 255.0f; } if (supportsAlpha) { alp = (rgb>>24) & 0xff; fdata[3] = alp * (1.0f / 255.0f); if (isAlphaPremultiplied) { factor *= fdata[3]; } } fdata[0] = red * factor; fdata[1] = grn * factor; fdata[2] = blu * factor; } else if (is_LinearGray_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; fdata[0] = ((0.2125f * red) + (0.7154f * grn) + (0.0721f * blu)) / 65535.0f; if (supportsAlpha) { alp = (rgb>>24) & 0xff; fdata[1] = alp * (1.0f / 255.0f); if (isAlphaPremultiplied) { fdata[0] *= fdata[1]; } } } else if (is_ICCGray_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; int gray = (int) ((0.2125f * red) + (0.7154f * grn) + (0.0721f * blu) + 0.5f); fdata[0] = (fromLinearGray16ToOtherGray16LUT[gray] & 0xffff) / 65535.0f; if (supportsAlpha) { alp = (rgb>>24) & 0xff; fdata[1] = alp * (1.0f / 255.0f); if (isAlphaPremultiplied) { fdata[0] *= fdata[1]; } } } else { float norm[] = new float[3]; factor = 1.0f / 255.0f; norm[0] = red * factor; norm[1] = grn * factor; norm[2] = blu * factor; norm = colorSpace.fromRGB(norm); if (supportsAlpha) { alp = (rgb>>24) & 0xff; fdata[numColorComponents] = alp * factor; if (isAlphaPremultiplied) { factor *= alp; for (int i = 0; i < numColorComponents; i++) { norm[i] *= factor; } } } for (int i = 0; i < numColorComponents; i++) { fdata[i] = norm[i]; } } return fdata; } case DataBuffer.TYPE_DOUBLE: { double ddata[]; if (pixel == null) { ddata = new double[numComponents]; } else { ddata = (double[])pixel; } if (is_sRGB_stdScale || is_LinearRGB_stdScale) { double factor; if (is_LinearRGB_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; factor = 1.0 / 65535.0; } else { factor = 1.0 / 255.0; } if (supportsAlpha) { alp = (rgb>>24) & 0xff; ddata[3] = alp * (1.0 / 255.0); if (isAlphaPremultiplied) { factor *= ddata[3]; } } ddata[0] = red * factor; ddata[1] = grn * factor; ddata[2] = blu * factor; } else if (is_LinearGray_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; ddata[0] = ((0.2125 * red) + (0.7154 * grn) + (0.0721 * blu)) / 65535.0; if (supportsAlpha) { alp = (rgb>>24) & 0xff; ddata[1] = alp * (1.0 / 255.0); if (isAlphaPremultiplied) { ddata[0] *= ddata[1]; } } } else if (is_ICCGray_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; int gray = (int) ((0.2125f * red) + (0.7154f * grn) + (0.0721f * blu) + 0.5f); ddata[0] = (fromLinearGray16ToOtherGray16LUT[gray] & 0xffff) / 65535.0; if (supportsAlpha) { alp = (rgb>>24) & 0xff; ddata[1] = alp * (1.0 / 255.0); if (isAlphaPremultiplied) { ddata[0] *= ddata[1]; } } } else { float factor = 1.0f / 255.0f; float norm[] = new float[3]; norm[0] = red * factor; norm[1] = grn * factor; norm[2] = blu * factor; norm = colorSpace.fromRGB(norm); if (supportsAlpha) { alp = (rgb>>24) & 0xff; ddata[numColorComponents] = alp * (1.0 / 255.0); if (isAlphaPremultiplied) { factor *= alp; for (int i = 0; i < numColorComponents; i++) { norm[i] *= factor; } } } for (int i = 0; i < numColorComponents; i++) { ddata[i] = norm[i]; } } return ddata; } } } // Handle BYTE, USHORT, & INT here //REMIND: maybe more efficient not to use int array for //DataBuffer.TYPE_USHORT and DataBuffer.TYPE_INT int intpixel[]; if (transferType == DataBuffer.TYPE_INT && pixel != null) { intpixel = (int[])pixel; } else { intpixel = new int[numComponents]; } if (is_sRGB_stdScale || is_LinearRGB_stdScale) { int precision; float factor; if (is_LinearRGB_stdScale) { if (transferType == DataBuffer.TYPE_BYTE) { red = fromsRGB8LUT8[red] & 0xff; grn = fromsRGB8LUT8[grn] & 0xff; blu = fromsRGB8LUT8[blu] & 0xff; precision = 8; factor = 1.0f / 255.0f; } else { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; precision = 16; factor = 1.0f / 65535.0f; } } else { precision = 8; factor = 1.0f / 255.0f; } if (supportsAlpha) { alp = (rgb>>24)&0xff; if (nBits[3] == 8) { intpixel[3] = alp; } else { intpixel[3] = (int) (alp * (1.0f / 255.0f) * ((1<>24) & 0xff; if (nBits[1] == 8) { intpixel[1] = alp; } else { intpixel[1] = (int) (alp * (1.0f / 255.0f) * ((1 << nBits[1]) - 1) + 0.5f); } if (isAlphaPremultiplied) { gray *= (alp * (1.0f / 255.0f)); } } intpixel[0] = (int) (gray * ((1 << nBits[0]) - 1) + 0.5f); } else if (is_ICCGray_stdScale) { red = fromsRGB8LUT16[red] & 0xffff; grn = fromsRGB8LUT16[grn] & 0xffff; blu = fromsRGB8LUT16[blu] & 0xffff; int gray16 = (int) ((0.2125f * red) + (0.7154f * grn) + (0.0721f * blu) + 0.5f); float gray = (fromLinearGray16ToOtherGray16LUT[gray16] & 0xffff) / 65535.0f; if (supportsAlpha) { alp = (rgb>>24) & 0xff; if (nBits[1] == 8) { intpixel[1] = alp; } else { intpixel[1] = (int) (alp * (1.0f / 255.0f) * ((1 << nBits[1]) - 1) + 0.5f); } if (isAlphaPremultiplied) { gray *= (alp * (1.0f / 255.0f)); } } intpixel[0] = (int) (gray * ((1 << nBits[0]) - 1) + 0.5f); } else { // Need to convert the color float[] norm = new float[3]; float factor = 1.0f / 255.0f; norm[0] = red * factor; norm[1] = grn * factor; norm[2] = blu * factor; norm = colorSpace.fromRGB(norm); if (nonStdScale) { for (int i = 0; i < numColorComponents; i++) { norm[i] = (norm[i] - compOffset[i]) * compScale[i]; // REMIND: need to analyze whether this // clamping is necessary if (norm[i] < 0.0f) { norm[i] = 0.0f; } if (norm[i] > 1.0f) { norm[i] = 1.0f; } } } if (supportsAlpha) { alp = (rgb>>24) & 0xff; if (nBits[numColorComponents] == 8) { intpixel[numColorComponents] = alp; } else { intpixel[numColorComponents] = (int) (alp * factor * ((1< 23) { // fix 4412670 - for components of 24 or more bits // some calculations done above with float precision // may lose enough precision that the integer result // overflows nBits, so we need to clamp. for (int i = 0; i < numComponents; i++) { if (intpixel[i] > ((1< 1) { throw new IllegalArgumentException("More than one component per pixel"); } if (needScaleInit) { initScale(); } if (noUnnorm) { throw new IllegalArgumentException( "This ColorModel does not support the unnormalized form"); } if (components == null) { components = new int[offset+1]; } components[offset+0] = (pixel & ((1<n - 1, where * n is the number of bits for a particular component. Normalized * components are float values between a per component minimum and * maximum specified by the {@code ColorSpace} object for this * {@code ColorModel}. An {@code IllegalArgumentException} * will be thrown if color component values for this * {@code ColorModel} are not conveniently representable in the * unnormalized form. If the * {@code components} array is {@code null}, a new array * will be allocated. The {@code components} array will * be returned. Color/alpha components are stored in the * {@code components} array starting at {@code offset} (even * if the array is allocated by this method). An * {@code ArrayIndexOutOfBoundsException} is thrown if the * {@code components} array is not {@code null} and is not * large enough to hold all the color and alpha * components (starting at {@code offset}). An * {@code IllegalArgumentException} is thrown if the * {@code normComponents} array is not large enough to hold * all the color and alpha components starting at * {@code normOffset}. * @param normComponents an array containing normalized components * @param normOffset the offset into the {@code normComponents} * array at which to start retrieving normalized components * @param components an array that receives the components from * {@code normComponents} * @param offset the index into {@code components} at which to * begin storing normalized components from * {@code normComponents} * @return an array containing unnormalized color and alpha * components. * @throws IllegalArgumentException If this * {@code ComponentColorModel} does not support the unnormalized form * @throws IllegalArgumentException if the length of * {@code normComponents} minus {@code normOffset} * is less than {@code numComponents} */ public int[] getUnnormalizedComponents(float[] normComponents, int normOffset, int[] components, int offset) { if (needScaleInit) { initScale(); } if (noUnnorm) { throw new IllegalArgumentException( "This ColorModel does not support the unnormalized form"); } return super.getUnnormalizedComponents(normComponents, normOffset, components, offset); } /** * Returns an array of all of the color/alpha components in normalized * form, given an unnormalized component array. Unnormalized components * are unsigned integral values between 0 and 2n - 1, where * n is the number of bits for a particular component. Normalized * components are float values between a per component minimum and * maximum specified by the {@code ColorSpace} object for this * {@code ColorModel}. An {@code IllegalArgumentException} * will be thrown if color component values for this * {@code ColorModel} are not conveniently representable in the * unnormalized form. If the * {@code normComponents} array is {@code null}, a new array * will be allocated. The {@code normComponents} array * will be returned. Color/alpha components are stored in the * {@code normComponents} array starting at * {@code normOffset} (even if the array is allocated by this * method). An {@code ArrayIndexOutOfBoundsException} is thrown * if the {@code normComponents} array is not {@code null} * and is not large enough to hold all the color and alpha components * (starting at {@code normOffset}). An * {@code IllegalArgumentException} is thrown if the * {@code components} array is not large enough to hold all the * color and alpha components starting at {@code offset}. * @param components an array containing unnormalized components * @param offset the offset into the {@code components} array at * which to start retrieving unnormalized components * @param normComponents an array that receives the normalized components * @param normOffset the index into {@code normComponents} at * which to begin storing normalized components * @return an array containing normalized color and alpha * components. * @throws IllegalArgumentException If this * {@code ComponentColorModel} does not support the unnormalized form */ public float[] getNormalizedComponents(int[] components, int offset, float[] normComponents, int normOffset) { if (needScaleInit) { initScale(); } if (noUnnorm) { throw new IllegalArgumentException( "This ColorModel does not support the unnormalized form"); } return super.getNormalizedComponents(components, offset, normComponents, normOffset); } /** * Returns a pixel value represented as an int in this {@code ColorModel}, * given an array of unnormalized color/alpha components. * * @param components An array of unnormalized color/alpha components. * @param offset An offset into the {@code components} array. * * @return A pixel value represented as an int. * * @throws IllegalArgumentException If there is more than one component * in this {@code ColorModel}. * @throws IllegalArgumentException If this * {@code ComponentColorModel} does not support the unnormalized form */ public int getDataElement(int[] components, int offset) { if (needScaleInit) { initScale(); } if (numComponents == 1) { if (noUnnorm) { throw new IllegalArgumentException( "This ColorModel does not support the unnormalized form"); } return components[offset+0]; } throw new IllegalArgumentException("This model returns "+ numComponents+ " elements in the pixel array."); } /** * Returns a data element array representation of a pixel in this * {@code ColorModel}, given an array of unnormalized color/alpha * components. This array can then be passed to the {@code setDataElements} * method of a {@code WritableRaster} object. * * @param components An array of unnormalized color/alpha components. * @param offset The integer offset into the {@code components} array. * @param obj The object in which to store the data element array * representation of the pixel. If {@code obj} variable is null, * a new array is allocated. If {@code obj} is not null, it must * be a primitive array of type {@code transferType}. An * {@code ArrayIndexOutOfBoundsException} is thrown if * {@code obj} is not large enough to hold a pixel value * for this {@code ColorModel}. Since * {@code ComponentColorModel} can be subclassed, subclasses * inherit the implementation of this method and if they don't * override it then they throw an exception if they use an * unsupported {@code transferType}. * * @return The data element array representation of a pixel * in this {@code ColorModel}. * * @throws IllegalArgumentException If the components array * is not large enough to hold all the color and alpha components * (starting at offset). * @throws ClassCastException If {@code obj} is not null and is not a * primitive array of type {@code transferType}. * @throws ArrayIndexOutOfBoundsException If {@code obj} is not large * enough to hold a pixel value for this {@code ColorModel}. * @throws IllegalArgumentException If this * {@code ComponentColorModel} does not support the unnormalized form * @throws UnsupportedOperationException If the transfer type of * this {@code ComponentColorModel} * is not one of the following transfer types: * {@code DataBuffer.TYPE_BYTE}, {@code DataBuffer.TYPE_USHORT}, * or {@code DataBuffer.TYPE_INT}. * * @see WritableRaster#setDataElements * @see SampleModel#setDataElements */ public Object getDataElements(int[] components, int offset, Object obj) { if (needScaleInit) { initScale(); } if (noUnnorm) { throw new IllegalArgumentException( "This ColorModel does not support the unnormalized form"); } if ((components.length-offset) < numComponents) { throw new IllegalArgumentException("Component array too small"+ " (should be "+numComponents); } switch(transferType) { case DataBuffer.TYPE_INT: { int[] pixel; if (obj == null) { pixel = new int[numComponents]; } else { pixel = (int[]) obj; } System.arraycopy(components, offset, pixel, 0, numComponents); return pixel; } case DataBuffer.TYPE_BYTE: { byte[] pixel; if (obj == null) { pixel = new byte[numComponents]; } else { pixel = (byte[]) obj; } for (int i=0; i < numComponents; i++) { pixel[i] = (byte) (components[offset+i]&0xff); } return pixel; } case DataBuffer.TYPE_USHORT: { short[] pixel; if (obj == null) { pixel = new short[numComponents]; } else { pixel = (short[]) obj; } for (int i=0; i < numComponents; i++) { pixel[i] = (short) (components[offset+i]&0xffff); } return pixel; } default: throw new UnsupportedOperationException("This method has not been "+ "implemented for transferType " + transferType); } } /** * Returns a pixel value represented as an {@code int} in this * {@code ColorModel}, given an array of normalized color/alpha * components. This method will throw an * {@code IllegalArgumentException} if pixel values for this * {@code ColorModel} are not conveniently representable as a * single {@code int}. An * {@code ArrayIndexOutOfBoundsException} is thrown if the * {@code normComponents} array is not large enough to hold all the * color and alpha components (starting at {@code normOffset}). * @param normComponents an array of normalized color and alpha * components * @param normOffset the index into {@code normComponents} at which to * begin retrieving the color and alpha components * @return an {@code int} pixel value in this * {@code ColorModel} corresponding to the specified components. * @throws IllegalArgumentException if * pixel values for this {@code ColorModel} are not * conveniently representable as a single {@code int} * @throws ArrayIndexOutOfBoundsException if * the {@code normComponents} array is not large enough to * hold all of the color and alpha components starting at * {@code normOffset} * @since 1.4 */ public int getDataElement(float[] normComponents, int normOffset) { if (numComponents > 1) { throw new IllegalArgumentException("More than one component per pixel"); } if (signed) { throw new IllegalArgumentException("Component value is signed"); } if (needScaleInit) { initScale(); } Object pixel = getDataElements(normComponents, normOffset, null); switch (transferType) { case DataBuffer.TYPE_BYTE: { byte bpixel[] = (byte[]) pixel; return bpixel[0] & 0xff; } case DataBuffer.TYPE_USHORT: { short[] uspixel = (short[]) pixel; return uspixel[0] & 0xffff; } case DataBuffer.TYPE_INT: { int[] ipixel = (int[]) pixel; return ipixel[0]; } default: throw new UnsupportedOperationException("This method has not been " + "implemented for transferType " + transferType); } } /** * Returns a data element array representation of a pixel in this * {@code ColorModel}, given an array of normalized color/alpha * components. This array can then be passed to the * {@code setDataElements} method of a {@code WritableRaster} * object. An {@code ArrayIndexOutOfBoundsException} is thrown * if the {@code normComponents} array is not large enough to hold * all the color and alpha components (starting at * {@code normOffset}). If the {@code obj} variable is * {@code null}, a new array will be allocated. If * {@code obj} is not {@code null}, it must be a primitive * array of type transferType; otherwise, a * {@code ClassCastException} is thrown. An * {@code ArrayIndexOutOfBoundsException} is thrown if * {@code obj} is not large enough to hold a pixel value for this * {@code ColorModel}. * @param normComponents an array of normalized color and alpha * components * @param normOffset the index into {@code normComponents} at which to * begin retrieving color and alpha components * @param obj a primitive data array to hold the returned pixel * @return an {@code Object} which is a primitive data array * representation of a pixel * @throws ClassCastException if {@code obj} * is not a primitive array of type {@code transferType} * @throws ArrayIndexOutOfBoundsException if * {@code obj} is not large enough to hold a pixel value * for this {@code ColorModel} or the {@code normComponents} * array is not large enough to hold all of the color and alpha * components starting at {@code normOffset} * @see WritableRaster#setDataElements * @see SampleModel#setDataElements * @since 1.4 */ public Object getDataElements(float[] normComponents, int normOffset, Object obj) { boolean needAlpha = supportsAlpha && isAlphaPremultiplied; float[] stdNormComponents; if (needScaleInit) { initScale(); } if (nonStdScale) { stdNormComponents = new float[numComponents]; for (int c = 0, nc = normOffset; c < numColorComponents; c++, nc++) { stdNormComponents[c] = (normComponents[nc] - compOffset[c]) * compScale[c]; // REMIND: need to analyze whether this // clamping is necessary if (stdNormComponents[c] < 0.0f) { stdNormComponents[c] = 0.0f; } if (stdNormComponents[c] > 1.0f) { stdNormComponents[c] = 1.0f; } } if (supportsAlpha) { stdNormComponents[numColorComponents] = normComponents[numColorComponents + normOffset]; } normOffset = 0; } else { stdNormComponents = normComponents; } switch (transferType) { case DataBuffer.TYPE_BYTE: byte[] bpixel; if (obj == null) { bpixel = new byte[numComponents]; } else { bpixel = (byte[]) obj; } if (needAlpha) { float alpha = stdNormComponents[numColorComponents + normOffset]; for (int c = 0, nc = normOffset; c < numColorComponents; c++, nc++) { bpixel[c] = (byte) ((stdNormComponents[nc] * alpha) * ((float) ((1 << nBits[c]) - 1)) + 0.5f); } bpixel[numColorComponents] = (byte) (alpha * ((float) ((1 << nBits[numColorComponents]) - 1)) + 0.5f); } else { for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { bpixel[c] = (byte) (stdNormComponents[nc] * ((float) ((1 << nBits[c]) - 1)) + 0.5f); } } return bpixel; case DataBuffer.TYPE_USHORT: short[] uspixel; if (obj == null) { uspixel = new short[numComponents]; } else { uspixel = (short[]) obj; } if (needAlpha) { float alpha = stdNormComponents[numColorComponents + normOffset]; for (int c = 0, nc = normOffset; c < numColorComponents; c++, nc++) { uspixel[c] = (short) ((stdNormComponents[nc] * alpha) * ((float) ((1 << nBits[c]) - 1)) + 0.5f); } uspixel[numColorComponents] = (short) (alpha * ((float) ((1 << nBits[numColorComponents]) - 1)) + 0.5f); } else { for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { uspixel[c] = (short) (stdNormComponents[nc] * ((float) ((1 << nBits[c]) - 1)) + 0.5f); } } return uspixel; case DataBuffer.TYPE_INT: int[] ipixel; if (obj == null) { ipixel = new int[numComponents]; } else { ipixel = (int[]) obj; } if (needAlpha) { float alpha = stdNormComponents[numColorComponents + normOffset]; for (int c = 0, nc = normOffset; c < numColorComponents; c++, nc++) { ipixel[c] = (int) ((stdNormComponents[nc] * alpha) * ((float) ((1 << nBits[c]) - 1)) + 0.5f); } ipixel[numColorComponents] = (int) (alpha * ((float) ((1 << nBits[numColorComponents]) - 1)) + 0.5f); } else { for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { ipixel[c] = (int) (stdNormComponents[nc] * ((float) ((1 << nBits[c]) - 1)) + 0.5f); } } return ipixel; case DataBuffer.TYPE_SHORT: short[] spixel; if (obj == null) { spixel = new short[numComponents]; } else { spixel = (short[]) obj; } if (needAlpha) { float alpha = stdNormComponents[numColorComponents + normOffset]; for (int c = 0, nc = normOffset; c < numColorComponents; c++, nc++) { spixel[c] = (short) (stdNormComponents[nc] * alpha * 32767.0f + 0.5f); } spixel[numColorComponents] = (short) (alpha * 32767.0f + 0.5f); } else { for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { spixel[c] = (short) (stdNormComponents[nc] * 32767.0f + 0.5f); } } return spixel; case DataBuffer.TYPE_FLOAT: float[] fpixel; if (obj == null) { fpixel = new float[numComponents]; } else { fpixel = (float[]) obj; } if (needAlpha) { float alpha = normComponents[numColorComponents + normOffset]; for (int c = 0, nc = normOffset; c < numColorComponents; c++, nc++) { fpixel[c] = normComponents[nc] * alpha; } fpixel[numColorComponents] = alpha; } else { for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { fpixel[c] = normComponents[nc]; } } return fpixel; case DataBuffer.TYPE_DOUBLE: double[] dpixel; if (obj == null) { dpixel = new double[numComponents]; } else { dpixel = (double[]) obj; } if (needAlpha) { double alpha = (double) (normComponents[numColorComponents + normOffset]); for (int c = 0, nc = normOffset; c < numColorComponents; c++, nc++) { dpixel[c] = normComponents[nc] * alpha; } dpixel[numColorComponents] = alpha; } else { for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { dpixel[c] = (double) normComponents[nc]; } } return dpixel; default: throw new UnsupportedOperationException("This method has not been "+ "implemented for transferType " + transferType); } } /** * Returns an array of all of the color/alpha components in normalized * form, given a pixel in this {@code ColorModel}. The pixel * value is specified by an array of data elements of type transferType * passed in as an object reference. If pixel is not a primitive array * of type transferType, a {@code ClassCastException} is thrown. * An {@code ArrayIndexOutOfBoundsException} is thrown if * {@code pixel} is not large enough to hold a pixel value for this * {@code ColorModel}. * Normalized components are float values between a per component minimum * and maximum specified by the {@code ColorSpace} object for this * {@code ColorModel}. If the * {@code normComponents} array is {@code null}, a new array * will be allocated. The {@code normComponents} array * will be returned. Color/alpha components are stored in the * {@code normComponents} array starting at * {@code normOffset} (even if the array is allocated by this * method). An {@code ArrayIndexOutOfBoundsException} is thrown * if the {@code normComponents} array is not {@code null} * and is not large enough to hold all the color and alpha components * (starting at {@code normOffset}). *

* This method must be overridden by a subclass if that subclass * is designed to translate pixel sample values to color component values * in a non-default way. The default translations implemented by this * class is described in the class comments. Any subclass implementing * a non-default translation must follow the constraints on allowable * translations defined there. * @param pixel the specified pixel * @param normComponents an array to receive the normalized components * @param normOffset the offset into the {@code normComponents} * array at which to start storing normalized components * @return an array containing normalized color and alpha * components. * @throws ClassCastException if {@code pixel} is not a primitive * array of type transferType * @throws ArrayIndexOutOfBoundsException if * {@code normComponents} is not large enough to hold all * color and alpha components starting at {@code normOffset} * @throws ArrayIndexOutOfBoundsException if * {@code pixel} is not large enough to hold a pixel * value for this {@code ColorModel}. * @since 1.4 */ public float[] getNormalizedComponents(Object pixel, float[] normComponents, int normOffset) { if (normComponents == null) { normComponents = new float[numComponents+normOffset]; } switch (transferType) { case DataBuffer.TYPE_BYTE: byte[] bpixel = (byte[]) pixel; for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { normComponents[nc] = ((float) (bpixel[c] & 0xff)) / ((float) ((1 << nBits[c]) - 1)); } break; case DataBuffer.TYPE_USHORT: short[] uspixel = (short[]) pixel; for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { normComponents[nc] = ((float) (uspixel[c] & 0xffff)) / ((float) ((1 << nBits[c]) - 1)); } break; case DataBuffer.TYPE_INT: int[] ipixel = (int[]) pixel; for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { normComponents[nc] = ((float) ipixel[c]) / ((float) ((1 << nBits[c]) - 1)); } break; case DataBuffer.TYPE_SHORT: short[] spixel = (short[]) pixel; for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { normComponents[nc] = ((float) spixel[c]) / 32767.0f; } break; case DataBuffer.TYPE_FLOAT: float[] fpixel = (float[]) pixel; for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { normComponents[nc] = fpixel[c]; } break; case DataBuffer.TYPE_DOUBLE: double[] dpixel = (double[]) pixel; for (int c = 0, nc = normOffset; c < numComponents; c++, nc++) { normComponents[nc] = (float) dpixel[c]; } break; default: throw new UnsupportedOperationException("This method has not been "+ "implemented for transferType " + transferType); } if (supportsAlpha && isAlphaPremultiplied) { float alpha = normComponents[numColorComponents + normOffset]; if (alpha != 0.0f) { float invAlpha = 1.0f / alpha; for (int c = normOffset; c < numColorComponents + normOffset; c++) { normComponents[c] *= invAlpha; } } } if (min != null) { // Normally (i.e. when this class is not subclassed to override // this method), the test (min != null) will be equivalent to // the test (nonStdScale). However, there is an unlikely, but // possible case, in which this method is overridden, nonStdScale // is set true by initScale(), the subclass method for some // reason calls this superclass method, but the min and // diffMinMax arrays were never initialized by setupLUTs(). In // that case, the right thing to do is follow the intended // semantics of this method, and rescale the color components // only if the ColorSpace min/max were detected to be other // than 0.0/1.0 by setupLUTs(). Note that this implies the // transferType is byte, ushort, int, or short - i.e. components // derived from float and double pixel data are never rescaled. for (int c = 0; c < numColorComponents; c++) { normComponents[c + normOffset] = min[c] + diffMinMax[c] * normComponents[c + normOffset]; } } return normComponents; } /** * Forces the raster data to match the state specified in the * {@code isAlphaPremultiplied} variable, assuming the data * is currently correctly described by this {@code ColorModel}. * It may multiply or divide the color raster data by alpha, or * do nothing if the data is in the correct state. If the data needs * to be coerced, this method also returns an instance of * this {@code ColorModel} with * the {@code isAlphaPremultiplied} flag set appropriately. * Since {@code ColorModel} can be subclassed, subclasses inherit * the implementation of this method and if they don't override it * then they throw an exception if they use an unsupported * {@code transferType}. * * @throws NullPointerException if {@code raster} is * {@code null} and data coercion is required. * @throws UnsupportedOperationException if the transfer type of * this {@code ComponentColorModel} * is not one of the supported transfer types: * {@code DataBuffer.TYPE_BYTE}, {@code DataBuffer.TYPE_USHORT}, * {@code DataBuffer.TYPE_INT}, {@code DataBuffer.TYPE_SHORT}, * {@code DataBuffer.TYPE_FLOAT}, or {@code DataBuffer.TYPE_DOUBLE}. */ public ColorModel coerceData (WritableRaster raster, boolean isAlphaPremultiplied) { if ((supportsAlpha == false) || (this.isAlphaPremultiplied == isAlphaPremultiplied)) { // Nothing to do return this; } int w = raster.getWidth(); int h = raster.getHeight(); int aIdx = raster.getNumBands() - 1; float normAlpha; int rminX = raster.getMinX(); int rY = raster.getMinY(); int rX; if (isAlphaPremultiplied) { switch (transferType) { case DataBuffer.TYPE_BYTE: { byte pixel[] = null; byte zpixel[] = null; float alphaScale = 1.0f / ((float) ((1<