org.scijava.util.Bytes Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of scijava-common Show documentation
Show all versions of scijava-common Show documentation
SciJava Common is a shared library for SciJava software. It provides a plugin framework, with an extensible mechanism for service discovery, backed by its own annotation processor, so that plugins can be loaded dynamically. It is used by downstream projects in the SciJava ecosystem, such as ImageJ and SCIFIO.
/*
* #%L
* SciJava Common shared library for SciJava software.
* %%
* Copyright (C) 2009 - 2017 Board of Regents of the University of
* Wisconsin-Madison, Broad Institute of MIT and Harvard, Max Planck
* Institute of Molecular Cell Biology and Genetics, University of
* Konstanz, and KNIME GmbH.
* %%
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright notice,
* this list of conditions and the following disclaimer in the documentation
* and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
* #L%
*/
// This class was derived from the loci.common.DataTools class of the
// Bio-Formats library, licensed according to Simplified BSD, as follows:
//
// Copyright (C) 2005 - 2015 Open Microscopy Environment:
// - Board of Regents of the University of Wisconsin-Madison
// - Glencoe Software, Inc.
// - University of Dundee
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
// 2. Redistributions in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other materials provided with the distribution.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
// AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
// ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
// LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
// CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
// SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
// INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
// CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
// ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
// POSSIBILITY OF SUCH DAMAGE.
package org.scijava.util;
/**
* Useful methods for reading, writing, decoding and converting {@code byte}s
* and {@code byte} arrays.
*
* @author Curtis Rueden
* @author Melissa Linkert
* @author Chris Allan
*/
public final class Bytes {
private Bytes() {
// NB: prevent instantiation of utility class.
}
// -- Word decoding - bytes to primitive types --
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to a {@code short}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static short toShort(final byte[] bytes, final int off, int len,
final boolean little)
{
if (bytes.length - off < len) len = bytes.length - off;
short total = 0;
for (int i = 0, ndx = off; i < len; i++, ndx++) {
total |=
(bytes[ndx] < 0 ? 256 + bytes[ndx] : (int) bytes[ndx]) << ((little ? i
: len - i - 1) * 8);
}
return total;
}
/**
* Translates up to the first 2 bytes of a {@code byte} array beyond the given
* offset to a {@code short}. If there are fewer than 2 bytes available the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static short toShort(final byte[] bytes, final int off,
final boolean little)
{
return toShort(bytes, off, 2, little);
}
/**
* Translates up to the first 2 bytes of a {@code byte} array to a
* {@code short}. If there are fewer than 2 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static short toShort(final byte[] bytes, final boolean little) {
return toShort(bytes, 0, 2, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to a {@code short}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static short toShort(final short[] bytes, final int off, int len,
final boolean little)
{
if (bytes.length - off < len) len = bytes.length - off;
short total = 0;
for (int i = 0, ndx = off; i < len; i++, ndx++) {
total |= bytes[ndx] << ((little ? i : len - i - 1) * 8);
}
return total;
}
/**
* Translates up to the first 2 bytes of a {@code byte} array beyond the given
* offset to a {@code short}. If there are fewer than 2 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static short toShort(final short[] bytes, final int off,
final boolean little)
{
return toShort(bytes, off, 2, little);
}
/**
* Translates up to the first 2 bytes of a {@code byte} array to a
* {@code short}. If there are fewer than 2 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static short toShort(final short[] bytes, final boolean little) {
return toShort(bytes, 0, 2, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to an {@code int}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static int toInt(final byte[] bytes, final int off, int len,
final boolean little)
{
if (bytes.length - off < len) len = bytes.length - off;
int total = 0;
for (int i = 0, ndx = off; i < len; i++, ndx++) {
total |=
(bytes[ndx] < 0 ? 256 + bytes[ndx] : (int) bytes[ndx]) << ((little ? i
: len - i - 1) * 8);
}
return total;
}
/**
* Translates up to the first 4 bytes of a {@code byte} array beyond the given
* offset to an {@code int}. If there are fewer than 4 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static int toInt(final byte[] bytes, final int off,
final boolean little)
{
return toInt(bytes, off, 4, little);
}
/**
* Translates up to the first 4 bytes of a {@code byte} array to an
* {@code int}. If there are fewer than 4 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static int toInt(final byte[] bytes, final boolean little) {
return toInt(bytes, 0, 4, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to an {@code int}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static int toInt(final short[] bytes, final int off, int len,
final boolean little)
{
if (bytes.length - off < len) len = bytes.length - off;
int total = 0;
for (int i = 0, ndx = off; i < len; i++, ndx++) {
total |= bytes[ndx] << ((little ? i : len - i - 1) * 8);
}
return total;
}
/**
* Translates up to the first 4 bytes of a {@code byte} array beyond the given
* offset to an {@code int}. If there are fewer than 4 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static int toInt(final short[] bytes, final int off,
final boolean little)
{
return toInt(bytes, off, 4, little);
}
/**
* Translates up to the first 4 bytes of a {@code byte} array to an
* {@code int}. If there are fewer than 4 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static int toInt(final short[] bytes, final boolean little) {
return toInt(bytes, 0, 4, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to a {@code float}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static float toFloat(final byte[] bytes, final int off, final int len,
final boolean little)
{
return Float.intBitsToFloat(toInt(bytes, off, len, little));
}
/**
* Translates up to the first 4 bytes of a {@code byte} array beyond a given
* offset to a {@code float}. If there are fewer than 4 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static float toFloat(final byte[] bytes, final int off,
final boolean little)
{
return toFloat(bytes, off, 4, little);
}
/**
* Translates up to the first 4 bytes of a {@code byte} array to a
* {@code float}. If there are fewer than 4 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static float toFloat(final byte[] bytes, final boolean little) {
return toFloat(bytes, 0, 4, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* a given offset to a {@code float}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static float toFloat(final short[] bytes, final int off,
final int len, final boolean little)
{
return Float.intBitsToFloat(toInt(bytes, off, len, little));
}
/**
* Translates up to the first 4 bytes of a {@code byte} array beyond a given
* offset to a {@code float}. If there are fewer than 4 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static float toFloat(final short[] bytes, final int off,
final boolean little)
{
return toInt(bytes, off, 4, little);
}
/**
* Translates up to the first 4 bytes of a {@code byte} array to a
* {@code float}. If there are fewer than 4 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static float toFloat(final short[] bytes, final boolean little) {
return toInt(bytes, 0, 4, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to a {@code long}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static long toLong(final byte[] bytes, final int off, int len,
final boolean little)
{
if (bytes.length - off < len) len = bytes.length - off;
long total = 0;
for (int i = 0, ndx = off; i < len; i++, ndx++) {
total |=
(bytes[ndx] < 0 ? 256L + bytes[ndx] : (long) bytes[ndx]) << ((little
? i : len - i - 1) * 8);
}
return total;
}
/**
* Translates up to the first 8 bytes of a {@code byte} array beyond the given
* offset to a {@code long}. If there are fewer than 8 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static long toLong(final byte[] bytes, final int off,
final boolean little)
{
return toLong(bytes, off, 8, little);
}
/**
* Translates up to the first 8 bytes of a {@code byte} array to a
* {@code long}. If there are fewer than 8 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static long toLong(final byte[] bytes, final boolean little) {
return toLong(bytes, 0, 8, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to a {@code long}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static long toLong(final short[] bytes, final int off, int len,
final boolean little)
{
if (bytes.length - off < len) len = bytes.length - off;
long total = 0;
for (int i = 0, ndx = off; i < len; i++, ndx++) {
total |= ((long) bytes[ndx]) << ((little ? i : len - i - 1) * 8);
}
return total;
}
/**
* Translates up to the first 8 bytes of a {@code byte} array beyond the given
* offset to a {@code long}. If there are fewer than 8 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static long toLong(final short[] bytes, final int off,
final boolean little)
{
return toLong(bytes, off, 8, little);
}
/**
* Translates up to the first 8 bytes of a {@code byte} array to a
* {@code long}. If there are fewer than 8 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static long toLong(final short[] bytes, final boolean little) {
return toLong(bytes, 0, 8, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to a {@code double}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static double toDouble(final byte[] bytes, final int off,
final int len, final boolean little)
{
return Double.longBitsToDouble(toLong(bytes, off, len, little));
}
/**
* Translates up to the first 8 bytes of a {@code byte} array beyond the given
* offset to a {@code double}. If there are fewer than 8 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static double toDouble(final byte[] bytes, final int off,
final boolean little)
{
return toDouble(bytes, off, 8, little);
}
/**
* Translates up to the first 8 bytes of a {@code byte} array to a
* {@code double}. If there are fewer than 8 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static double toDouble(final byte[] bytes, final boolean little) {
return toDouble(bytes, 0, 8, little);
}
/**
* Translates up to the first {@code len} bytes of a {@code byte} array beyond
* the given offset to a {@code double}. If there are fewer than {@code len}
* bytes available, the MSBs are all assumed to be zero (regardless of
* endianness).
*/
public static double toDouble(final short[] bytes, final int off,
final int len, final boolean little)
{
return Double.longBitsToDouble(toLong(bytes, off, len, little));
}
/**
* Translates up to the first 8 bytes of a {@code byte} array beyond the given
* offset to a {@code double}. If there are fewer than 8 bytes available, the
* MSBs are all assumed to be zero (regardless of endianness).
*/
public static double toDouble(final short[] bytes, final int off,
final boolean little)
{
return toDouble(bytes, off, 8, little);
}
/**
* Translates up to the first 8 bytes of a {@code byte} array to a
* {@code double}. If there are fewer than 8 bytes available, the MSBs are all
* assumed to be zero (regardless of endianness).
*/
public static double toDouble(final short[] bytes, final boolean little) {
return toDouble(bytes, 0, 8, little);
}
// -- Word decoding - primitive types to bytes --
/** Translates the {@code short} value into an array of two {@code byte}s. */
public static byte[] fromShort(final short value, final boolean little) {
final byte[] v = new byte[2];
unpack(value, v, 0, 2, little);
return v;
}
/** Translates the {@code int} value into an array of four {@code byte}s. */
public static byte[] fromInt(final int value, final boolean little) {
final byte[] v = new byte[4];
unpack(value, v, 0, 4, little);
return v;
}
/** Translates the {@code float} value into an array of four {@code byte}s. */
public static byte[] fromFloat(final float value, final boolean little) {
final byte[] v = new byte[4];
unpack(Float.floatToIntBits(value), v, 0, 4, little);
return v;
}
/** Translates the {@code long} value into an array of eight {@code byte}s. */
public static byte[] fromLong(final long value, final boolean little) {
final byte[] v = new byte[8];
unpack(value, v, 0, 8, little);
return v;
}
/** Translates the {@code double} value into an array of eight {@code byte}s. */
public static byte[] fromDouble(final double value, final boolean little) {
final byte[] v = new byte[8];
unpack(Double.doubleToLongBits(value), v, 0, 8, little);
return v;
}
/**
* Translates an array of {@code short} values into an array of {@code byte}
* values.
*/
public static byte[] fromShorts(final short[] values, final boolean little) {
final byte[] v = new byte[values.length * 2];
for (int i = 0; i < values.length; i++) {
unpack(values[i], v, i * 2, 2, little);
}
return v;
}
/**
* Translates an array of {@code int} values into an array of {@code byte}
* values.
*/
public static byte[] fromInts(final int[] values, final boolean little) {
final byte[] v = new byte[values.length * 4];
for (int i = 0; i < values.length; i++) {
unpack(values[i], v, i * 4, 4, little);
}
return v;
}
/**
* Translates an array of {@code float} values into an array of {@code byte}
* values.
*/
public static byte[] fromFloats(final float[] values, final boolean little) {
final byte[] v = new byte[values.length * 4];
for (int i = 0; i < values.length; i++) {
unpack(Float.floatToIntBits(values[i]), v, i * 4, 4, little);
}
return v;
}
/**
* Translates an array of {@code long} values into an array of {@code byte}
* values.
*/
public static byte[] fromLongs(final long[] values, final boolean little) {
final byte[] v = new byte[values.length * 8];
for (int i = 0; i < values.length; i++) {
unpack(values[i], v, i * 8, 8, little);
}
return v;
}
/**
* Translates an array of {@code double} values into an array of {@code byte}
* values.
*/
public static byte[] fromDoubles(final double[] values, final boolean little)
{
final byte[] v = new byte[values.length * 8];
for (int i = 0; i < values.length; i++) {
unpack(Double.doubleToLongBits(values[i]), v, i * 8, 8, little);
}
return v;
}
/**
* Translates {@code nBytes} of the given {@code long} and places the result
* in the given {@code byte} array.
*
* @throws IllegalArgumentException if the specified indices fall outside the
* buffer
*/
public static void unpack(final long value, final byte[] buf, final int ndx,
final int nBytes, final boolean little)
{
if (buf.length < ndx + nBytes) {
throw new IllegalArgumentException("Invalid indices: buf.length=" +
buf.length + ", ndx=" + ndx + ", nBytes=" + nBytes);
}
if (little) {
for (int i = 0; i < nBytes; i++) {
buf[ndx + i] = (byte) ((value >> (8 * i)) & 0xff);
}
}
else {
for (int i = 0; i < nBytes; i++) {
buf[ndx + i] = (byte) ((value >> (8 * (nBytes - i - 1))) & 0xff);
}
}
}
/**
* Converts a {@code byte} array to the appropriate 1D primitive type array.
*
* @param b Byte array to convert.
* @param bpp Denotes the number of bytes in the returned primitive type (e.g.
* if bpp == 2, we should return an array of type {@code short}).
* @param fp If set and bpp == 4 or bpp == 8, then return {@code float}s or
* {@code double}s.
* @param little Whether {@code byte} array is in little-endian order.
*/
public static Object makeArray(final byte[] b, final int bpp,
final boolean fp, final boolean little)
{
if (bpp == 1) {
return b;
}
else if (bpp == 2) {
final short[] s = new short[b.length / 2];
for (int i = 0; i < s.length; i++) {
s[i] = toShort(b, i * 2, 2, little);
}
return s;
}
else if (bpp == 4 && fp) {
final float[] f = new float[b.length / 4];
for (int i = 0; i < f.length; i++) {
f[i] = toFloat(b, i * 4, 4, little);
}
return f;
}
else if (bpp == 4) {
final int[] i = new int[b.length / 4];
for (int j = 0; j < i.length; j++) {
i[j] = toInt(b, j * 4, 4, little);
}
return i;
}
else if (bpp == 8 && fp) {
final double[] d = new double[b.length / 8];
for (int i = 0; i < d.length; i++) {
d[i] = toDouble(b, i * 8, 8, little);
}
return d;
}
else if (bpp == 8) {
final long[] l = new long[b.length / 8];
for (int i = 0; i < l.length; i++) {
l[i] = toLong(b, i * 8, 8, little);
}
return l;
}
return null;
}
/**
* Converts a {@code byte} array to the appropriate 2D primitive type array.
*
* @param b Byte array to convert.
* @param bpp Denotes the number of bytes in the returned primitive type (e.g.
* if bpp == 2, we should return an array of type {@code short}).
* @param fp If set and bpp == 4 or bpp == 8, then return {@code float}s or
* {@code double}s.
* @param little Whether {@code byte} array is in little-endian order.
* @param height The height of the output primitive array (2nd dim length).
* @return a 2D primitive array of appropriate type, dimensioned
* [height][b.length / (bpp * height)]
* @throws IllegalArgumentException if input {@code byte} array does not
* divide evenly into height pieces
*/
public static Object makeArray2D(final byte[] b, final int bpp,
final boolean fp, final boolean little, final int height)
{
if (b.length % (bpp * height) != 0) {
throw new IllegalArgumentException("Array length mismatch: " +
"b.length=" + b.length + "; bpp=" + bpp + "; height=" + height);
}
final int width = b.length / (bpp * height);
if (bpp == 1) {
final byte[][] b2 = new byte[height][width];
for (int y = 0; y < height; y++) {
final int index = width * y;
System.arraycopy(b, index, b2[y], 0, width);
}
return b2;
}
else if (bpp == 2) {
final short[][] s = new short[height][width];
for (int y = 0; y < height; y++) {
for (int x = 0; x < width; x++) {
final int index = 2 * (width * y + x);
s[y][x] = toShort(b, index, 2, little);
}
}
return s;
}
else if (bpp == 4 && fp) {
final float[][] f = new float[height][width];
for (int y = 0; y < height; y++) {
for (int x = 0; x < width; x++) {
final int index = 4 * (width * y + x);
f[y][x] = toFloat(b, index, 4, little);
}
}
return f;
}
else if (bpp == 4) {
final int[][] i = new int[height][width];
for (int y = 0; y < height; y++) {
for (int x = 0; x < width; x++) {
final int index = 4 * (width * y + x);
i[y][x] = toInt(b, index, 4, little);
}
}
return i;
}
else if (bpp == 8 && fp) {
final double[][] d = new double[height][width];
for (int y = 0; y < height; y++) {
for (int x = 0; x < width; x++) {
final int index = 8 * (width * y + x);
d[y][x] = toDouble(b, index, 8, little);
}
}
return d;
}
else if (bpp == 8) {
final long[][] l = new long[height][width];
for (int y = 0; y < height; y++) {
for (int x = 0; x < width; x++) {
final int index = 8 * (width * y + x);
l[y][x] = toLong(b, index, 8, little);
}
}
return l;
}
return null;
}
// -- Byte swapping --
public static short swap(final short x) {
return (short) ((x << 8) | ((x >> 8) & 0xFF));
}
public static char swap(final char x) {
return (char) ((x << 8) | ((x >> 8) & 0xFF));
}
public static int swap(final int x) {
return (swap((short) x) << 16) | (swap((short) (x >> 16)) & 0xFFFF);
}
public static long swap(final long x) {
return ((long) swap((int) x) << 32) | (swap((int) (x >> 32)) & 0xFFFFFFFFL);
}
public static float swap(final float x) {
return Float.intBitsToFloat(swap(Float.floatToIntBits(x)));
}
public static double swap(final double x) {
return Double.longBitsToDouble(swap(Double.doubleToLongBits(x)));
}
// -- Normalization --
/**
* Normalize the given {@code float} array so that the minimum value maps to
* 0.0 and the maximum value maps to 1.0.
*/
public static float[] normalize(final float[] data) {
final float[] rtn = new float[data.length];
// determine the finite min and max values
float min = Float.MAX_VALUE;
float max = Float.MIN_VALUE;
for (final float floatValue : data) {
if (floatValue == Float.POSITIVE_INFINITY ||
floatValue == Float.NEGATIVE_INFINITY)
{
continue;
}
if (floatValue < min) min = floatValue;
if (floatValue > max) max = floatValue;
}
// normalize infinity values
for (int i = 0; i < data.length; i++) {
if (data[i] == Float.POSITIVE_INFINITY) data[i] = max;
else if (data[i] == Float.NEGATIVE_INFINITY) data[i] = min;
}
// now normalize; min => 0.0, max => 1.0
final float range = max - min;
for (int i = 0; i < rtn.length; i++) {
rtn[i] = (data[i] - min) / range;
}
return rtn;
}
/**
* Normalize the given {@code double} array so that the minimum value maps to
* 0.0 and the maximum value maps to 1.0.
*/
public static double[] normalize(final double[] data) {
final double[] rtn = new double[data.length];
// determine the finite min and max values
double min = Double.MAX_VALUE;
double max = Double.MIN_VALUE;
for (final double doubleValue : data) {
if (doubleValue == Double.POSITIVE_INFINITY ||
doubleValue == Double.NEGATIVE_INFINITY)
{
continue;
}
if (doubleValue < min) min = doubleValue;
if (doubleValue > max) max = doubleValue;
}
// normalize infinity values
for (int i = 0; i < data.length; i++) {
if (data[i] == Double.POSITIVE_INFINITY) data[i] = max;
else if (data[i] == Double.NEGATIVE_INFINITY) data[i] = min;
}
// now normalize; min => 0.0, max => 1.0
final double range = max - min;
for (int i = 0; i < rtn.length; i++) {
rtn[i] = (data[i] - min) / range;
}
return rtn;
}
// -- Signed data conversion --
public static byte[] makeSigned(final byte[] b) {
for (int i = 0; i < b.length; i++) {
b[i] = (byte) (b[i] + 128);
}
return b;
}
public static short[] makeSigned(final short[] s) {
for (int i = 0; i < s.length; i++) {
s[i] = (short) (s[i] + 32768);
}
return s;
}
public static int[] makeSigned(final int[] i) {
for (int j = 0; j < i.length; j++) {
i[j] = (int) (i[j] + 2147483648L);
}
return i;
}
}