/* * Copyright © 2014 Cask Data, Inc. * * Licensed under the Apache License, Version 2.0 (the "License"); you may not * use this file except in compliance with the License. You may obtain a copy of * the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the * License for the specific language governing permissions and limitations under * the License. */ package co.cask.cdap.api.common; import java.io.DataOutput; import java.io.IOException; import java.io.UnsupportedEncodingException; import java.math.BigDecimal; import java.math.BigInteger; import java.nio.ByteBuffer; import java.nio.charset.Charset; import java.util.Arrays; import java.util.Comparator; import java.util.Iterator; import java.util.NavigableMap; import java.util.TreeMap; import java.util.UUID; import javax.annotation.Nullable; /** * Utility class that handles byte arrays, conversions to/from other types, * comparisons, hash code generation, manufacturing keys for HashMaps or * HashSets, etc. */ @SuppressWarnings("javadoc") public class Bytes { /** * Size of boolean in bytes. */ public static final int SIZEOF_BOOLEAN = Byte.SIZE / Byte.SIZE; /** * Size of byte in bytes. */ public static final int SIZEOF_BYTE = SIZEOF_BOOLEAN; /** * Size of char in bytes. */ public static final int SIZEOF_CHAR = Character.SIZE / Byte.SIZE; /** * Size of double in bytes. */ public static final int SIZEOF_DOUBLE = Double.SIZE / Byte.SIZE; /** * Size of float in bytes. */ public static final int SIZEOF_FLOAT = Float.SIZE / Byte.SIZE; /** * Size of int in bytes. */ public static final int SIZEOF_INT = Integer.SIZE / Byte.SIZE; /** * Size of long in bytes. */ public static final int SIZEOF_LONG = Long.SIZE / Byte.SIZE; /** * Size of short in bytes. */ public static final int SIZEOF_SHORT = Short.SIZE / Byte.SIZE; private static final char[] hexDigits = "0123456789abcdef".toCharArray(); private static final Charset UTF_8 = Charset.forName("UTF-8"); /** * Byte array comparator class. */ public static class ByteArrayComparator implements Comparator<byte []> { /** * Constructor. */ public ByteArrayComparator() { super(); } @Override public int compare(byte [] left, byte [] right) { return compareTo(left, right); } } /** * Pass this to TreeMaps where byte [] are keys. */ public static final Comparator<byte []> BYTES_COMPARATOR = new ByteArrayComparator(); /** * Put bytes at the specified byte array position. * @param tgtBytes the byte array * @param tgtOffset position in the array * @param srcBytes array to write out * @param srcOffset source offset * @param srcLength source length * @return incremented offset */ public static int putBytes(byte[] tgtBytes, int tgtOffset, byte[] srcBytes, int srcOffset, int srcLength) { System.arraycopy(srcBytes, srcOffset, tgtBytes, tgtOffset, srcLength); return tgtOffset + srcLength; } /** * Write a single byte out to the specified byte array position. * @param bytes the byte array * @param offset position in the array * @param b byte to write out * @return incremented offset */ public static int putByte(byte[] bytes, int offset, byte b) { bytes[offset] = b; return offset + 1; } /** * Returns a new byte array, copied from the passed ByteBuffer. * @param bb A ByteBuffer * @return the byte array */ public static byte[] toBytes(ByteBuffer bb) { int length = bb.remaining(); byte [] result = new byte[length]; int pos = bb.position(); bb.get(result); bb.position(pos); return result; } /** * This method will convert utf8 encoded bytes into a string. If * an UnsupportedEncodingException occurs, this method will eat it * and return null instead. * @param b Presumed UTF-8 encoded byte array. * @return String made from <code>b</code> */ public static String toString(final byte [] b) { if (b == null) { return null; } return toString(b, 0, b.length); } /** * Joins two byte arrays together using a separator. * @param b1 The first byte array. * @param sep The separator to use. * @param b2 The second byte array. */ public static String toString(final byte [] b1, String sep, final byte [] b2) { return toString(b1, 0, b1.length) + sep + toString(b2, 0, b2.length); } /** * When we encode strings, we always specify UTF8 encoding. */ public static final String UTF8_ENCODING = "UTF-8"; /** * This method will convert utf8 encoded bytes into a string. If * an UnsupportedEncodingException occurs, this method will eat it * and return null instead. * * @param b Presumed UTF-8 encoded byte array. * @param off offset into array * @param len length of utf-8 sequence * @return String made from <code>b</code> or null */ public static String toString(final byte [] b, int off, int len) { if (b == null) { return null; } if (len == 0) { return ""; } try { return new String(b, off, len, UTF8_ENCODING); } catch (UnsupportedEncodingException e) { return null; } } /** * This method will convert the remaining bytes of a UTF8 * encoded byte buffer into a string. * * @param buf Presumed UTF-8 encoded byte buffer. * @return String made from <code>buf</code> or null */ public static String toString(ByteBuffer buf) { return toString(buf, UTF_8); } /** * This method will convert the remaining bytes of an * encoded byte buffer into a string of the given charset. * * @param buf Encoded byte buffer. * @param charset Charset of the string encoded in the byte buffer. * @return String made from <code>buf</code> or null */ public static String toString(ByteBuffer buf, Charset charset) { if (buf == null) { return null; } buf.mark(); String s = charset.decode(buf).toString(); buf.reset(); return s; } /** * Write a printable representation of a byte array. * * @param b byte array * @return string * @see #toStringBinary(byte[], int, int) */ public static String toStringBinary(final byte [] b) { if (b == null) { return "null"; } return toStringBinary(b, 0, b.length); } /** * Returns a string containing each byte, in order, as a two-digit unsigned * hexadecimal number in lower case. */ public static String toHexString(byte [] bytes) { StringBuilder sb = new StringBuilder(2 * bytes.length); for (byte b : bytes) { sb.append(hexDigits[(b >> 4) & 0xf]).append(hexDigits[b & 0xf]); } return sb.toString(); } /** * Converts the given byte buffer, from its array offset to its limit, to * a string. The position and the mark are ignored. * * @param buf a byte buffer * @return a string representation of the buffer's binary contents */ public static String toStringBinary(ByteBuffer buf) { if (buf == null) { return "null"; } return toStringBinary(buf.array(), buf.arrayOffset(), buf.limit()); } /** * Write a printable representation of a byte array. Non-printable * characters are hex escaped in the format \\x%02X, eg: * \x00 \x05 etc * * @param b array to write out * @param off offset to start at * @param len length to write * @return string output */ public static String toStringBinary(final byte [] b, int off, int len) { StringBuilder result = new StringBuilder(); try { String first = new String(b, off, len, "ISO-8859-1"); for (int i = 0; i < first.length(); ++i) { int ch = first.charAt(i) & 0xFF; if ((ch >= '0' && ch <= '9') || (ch >= 'A' && ch <= 'Z') || (ch >= 'a' && ch <= 'z') || " `~!@#$%^&*()-_=+[]{}\\|;:'\",.<>/?".indexOf(ch) >= 0) { result.append(first.charAt(i)); } else { result.append(String.format("\\x%02X", ch)); } } } catch (UnsupportedEncodingException e) { } return result.toString(); } private static boolean isHexDigit(char c) { return (c >= 'A' && c <= 'F') || (c >= '0' && c <= '9'); } /** * Takes a ASCII digit in the range A-F0-9 and returns * the corresponding integer/ordinal value. * @param ch The hex digit. * @return The converted hex value as a byte. */ public static byte toBinaryFromHex(byte ch) { if (ch >= 'A' && ch <= 'F') { return (byte) ((byte) 10 + (byte) (ch - 'A')); } // else return (byte) (ch - '0'); } public static byte [] toBytesBinary(String in) { // this may be bigger than we need, but lets be safe. byte [] b = new byte[in.length()]; int size = 0; for (int i = 0; i < in.length(); ++i) { char ch = in.charAt(i); if (ch == '\\') { // begin hex escape: char next = in.charAt(i + 1); if (next != 'x') { // invalid escape sequence, ignore this one. b[size++] = (byte) ch; continue; } // ok, take next 2 hex digits. char hd1 = in.charAt(i + 2); char hd2 = in.charAt(i + 3); // they need to be A-F0-9: if (!isHexDigit(hd1) || !isHexDigit(hd2)) { // bogus escape code, ignore: continue; } // turn hex ASCII digit -> number byte d = (byte) ((toBinaryFromHex((byte) hd1) << 4) + toBinaryFromHex((byte) hd2)); b[size++] = d; i += 3; // skip 3 } else { b[size++] = (byte) ch; } } // resize: byte [] b2 = new byte[size]; System.arraycopy(b, 0, b2, 0, size); return b2; } /** * Converts a string to a UTF-8 byte array. * @param s string * @return the byte array */ public static byte[] toBytes(String s) { try { return s.getBytes(UTF8_ENCODING); } catch (UnsupportedEncodingException e) { return null; } } /** * Convert a boolean to a byte array. True becomes -1 * and false becomes 0. * * @param b value * @return <code>b</code> encoded in a byte array. */ public static byte [] toBytes(final boolean b) { return new byte[] { b ? (byte) -1 : (byte) 0 }; } /** * Reverses {@link #toBytes(boolean)}. * @param b array * @return True or false. */ public static boolean toBoolean(final byte [] b) { if (b.length != 1) { throw new IllegalArgumentException("Array has wrong size: " + b.length); } return b[0] != (byte) 0; } /** * Convert a long value to a byte array using big-endian. * * @param val value to convert * @return the byte array */ public static byte[] toBytes(long val) { byte [] b = new byte[8]; for (int i = 7; i > 0; i--) { b[i] = (byte) val; val >>>= 8; } b[0] = (byte) val; return b; } /** * Converts a byte array to a long value. Reverses * {@link #toBytes(long)} * @param bytes array * @return the long value */ public static long toLong(byte[] bytes) { return toLong(bytes, 0, SIZEOF_LONG); } /** * Converts a byte array to a long value. Assumes there will be * {@link #SIZEOF_LONG} bytes available. * * @param bytes bytes * @param offset offset * @return the long value */ public static long toLong(byte[] bytes, int offset) { return toLong(bytes, offset, SIZEOF_LONG); } /** * Converts a byte array to a long value. * * @param bytes array of bytes * @param offset offset into array * @param length length of data (must be {@link #SIZEOF_LONG}) * @return the long value * @throws IllegalArgumentException if length is not {@link #SIZEOF_LONG} or * if there's not enough room in the array at the offset indicated. */ public static long toLong(byte[] bytes, int offset, final int length) { if (length != SIZEOF_LONG || offset + length > bytes.length) { throw explainWrongLengthOrOffset(bytes, offset, length, SIZEOF_LONG); } long l = 0; for (int i = offset; i < offset + length; i++) { l <<= 8; l ^= bytes[i] & 0xFF; } return l; } private static IllegalArgumentException explainWrongLengthOrOffset(final byte[] bytes, final int offset, final int length, final int expectedLength) { String reason; if (length != expectedLength) { reason = "Wrong length: " + length + ", expected " + expectedLength; } else { reason = "offset (" + offset + ") + length (" + length + ") exceed the" + " capacity of the array: " + bytes.length; } return new IllegalArgumentException(reason); } /** * Put a long value out to the specified byte array position. * @param bytes the byte array * @param offset position in the array * @param val long to write out * @return incremented offset * @throws IllegalArgumentException if the byte array given doesn't have * enough room at the offset specified. */ public static int putLong(byte[] bytes, int offset, long val) { if (bytes.length - offset < SIZEOF_LONG) { throw new IllegalArgumentException("Not enough room to put a long at" + " offset " + offset + " in a " + bytes.length + " byte array"); } for (int i = offset + 7; i > offset; i--) { bytes[i] = (byte) val; val >>>= 8; } bytes[offset] = (byte) val; return offset + SIZEOF_LONG; } /** * Presumes float encoded as IEEE 754 floating-point "single format". * @param bytes byte array * @return Float made from passed byte array. */ public static float toFloat(byte [] bytes) { return toFloat(bytes, 0); } /** * Presumes float encoded as IEEE 754 floating-point "single format". * @param bytes array to convert * @param offset offset into array * @return Float made from passed byte array. */ public static float toFloat(byte [] bytes, int offset) { return Float.intBitsToFloat(toInt(bytes, offset, SIZEOF_INT)); } /** * Put a float value out to the specified byte array position. * @param bytes byte array * @param offset offset to write to * @param f float value * @return New offset in <code>bytes</code> */ public static int putFloat(byte [] bytes, int offset, float f) { return putInt(bytes, offset, Float.floatToRawIntBits(f)); } /** * @param f float value * @return the float represented as byte [] */ public static byte [] toBytes(final float f) { // Encode it as int return Bytes.toBytes(Float.floatToRawIntBits(f)); } /** * Return double made from passed bytes. * @param bytes byte array * @return Return double made from passed bytes. */ public static double toDouble(final byte [] bytes) { return toDouble(bytes, 0); } /** * Return double made from passed bytes. * @param bytes byte array * @param offset offset where double is * @return Return double made from passed bytes. */ public static double toDouble(final byte [] bytes, final int offset) { return Double.longBitsToDouble(toLong(bytes, offset, SIZEOF_LONG)); } /** * Put a double value out to the specified byte array position. * @param bytes byte array * @param offset offset to write to * @param d value * @return New offset into array <code>bytes</code> */ public static int putDouble(byte [] bytes, int offset, double d) { return putLong(bytes, offset, Double.doubleToLongBits(d)); } /** * Serialize a double as the IEEE 754 double format output. The resultant * array will be 8 bytes long. * * @param d value * @return the double represented as byte [] */ public static byte [] toBytes(final double d) { // Encode it as a long return Bytes.toBytes(Double.doubleToRawLongBits(d)); } /** * Convert an int value to a byte array. * @param val value * @return the byte array */ public static byte[] toBytes(int val) { byte [] b = new byte[4]; for (int i = 3; i > 0; i--) { b[i] = (byte) val; val >>>= 8; } b[0] = (byte) val; return b; } /** * Converts a byte array to an int value. * @param bytes byte array * @return the int value */ public static int toInt(byte[] bytes) { return toInt(bytes, 0, SIZEOF_INT); } /** * Converts a byte array to an int value. * @param bytes byte array * @param offset offset into array * @return the int value */ public static int toInt(byte[] bytes, int offset) { return toInt(bytes, offset, SIZEOF_INT); } /** * Converts a byte array to an int value. * @param bytes byte array * @param offset offset into array * @param length length of int (has to be {@link #SIZEOF_INT}) * @return the int value * @throws IllegalArgumentException if length is not {@link #SIZEOF_INT} or * if there's not enough room in the array at the offset indicated. */ public static int toInt(byte[] bytes, int offset, final int length) { if (length != SIZEOF_INT || offset + length > bytes.length) { throw explainWrongLengthOrOffset(bytes, offset, length, SIZEOF_INT); } int n = 0; for (int i = offset; i < (offset + length); i++) { n <<= 8; n ^= bytes[i] & 0xFF; } return n; } /** * Put an int value out to the specified byte array position. * @param bytes the byte array * @param offset position in the array * @param val int to write out * @return incremented offset * @throws IllegalArgumentException if the byte array given doesn't have * enough room at the offset specified. */ public static int putInt(byte[] bytes, int offset, int val) { if (bytes.length - offset < SIZEOF_INT) { throw new IllegalArgumentException("Not enough room to put an int at" + " offset " + offset + " in a " + bytes.length + " byte array"); } for (int i = offset + 3; i > offset; i--) { bytes[i] = (byte) val; val >>>= 8; } bytes[offset] = (byte) val; return offset + SIZEOF_INT; } /** * Convert a short value to a byte array of {@link #SIZEOF_SHORT} bytes long. * @param val value * @return the byte array */ public static byte[] toBytes(short val) { byte[] b = new byte[SIZEOF_SHORT]; b[1] = (byte) val; val >>= 8; b[0] = (byte) val; return b; } /** * Converts a byte array to a short value. * @param bytes byte array * @return the short value */ public static short toShort(byte[] bytes) { return toShort(bytes, 0, SIZEOF_SHORT); } /** * Converts a byte array to a short value. * @param bytes byte array * @param offset offset into array * @return the short value */ public static short toShort(byte[] bytes, int offset) { return toShort(bytes, offset, SIZEOF_SHORT); } /** * Converts a byte array to a short value. * @param bytes byte array * @param offset offset into array * @param length length, has to be {@link #SIZEOF_SHORT} * @return the short value * @throws IllegalArgumentException if length is not {@link #SIZEOF_SHORT} * or if there's not enough room in the array at the offset indicated. */ public static short toShort(byte[] bytes, int offset, final int length) { if (length != SIZEOF_SHORT || offset + length > bytes.length) { throw explainWrongLengthOrOffset(bytes, offset, length, SIZEOF_SHORT); } short n = 0; n ^= bytes[offset] & 0xFF; n <<= 8; n ^= bytes[offset + 1] & 0xFF; return n; } /** * This method will get a sequence of bytes from pos -> limit, * but will restore pos after. * @param buf * @return byte array */ public static byte[] getBytes(ByteBuffer buf) { int savedPos = buf.position(); byte [] newBytes = new byte[buf.remaining()]; buf.get(newBytes); buf.position(savedPos); return newBytes; } /** * Put a short value out to the specified byte array position. * @param bytes the byte array * @param offset position in the array * @param val short to write out * @return incremented offset * @throws IllegalArgumentException if the byte array given doesn't have * enough room at the offset specified. */ public static int putShort(byte[] bytes, int offset, short val) { if (bytes.length - offset < SIZEOF_SHORT) { throw new IllegalArgumentException("Not enough room to put a short at" + " offset " + offset + " in a " + bytes.length + " byte array"); } bytes[offset + 1] = (byte) val; val >>= 8; bytes[offset] = (byte) val; return offset + SIZEOF_SHORT; } /** * Convert a BigDecimal value to a byte array. * * @param val * @return the byte array */ public static byte[] toBytes(BigDecimal val) { byte[] valueBytes = val.unscaledValue().toByteArray(); byte[] result = new byte[valueBytes.length + SIZEOF_INT]; int offset = putInt(result, 0, val.scale()); putBytes(result, offset, valueBytes, 0, valueBytes.length); return result; } /** * Returns a copy of the byte representation of the given UUID. * @param uuid UUID to get the byte representation of * @return byte representation of the given UUID */ public static byte[] toBytes(UUID uuid) { ByteBuffer bb = ByteBuffer.wrap(new byte[16]); bb.putLong(uuid.getMostSignificantBits()); bb.putLong(uuid.getLeastSignificantBits()); return bb.array(); } /** * Convert the given byte representation of a UUID into a UUID. * @param bytes byte representation of a UUID * @return UUID */ public static UUID toUUID(byte [] bytes) { ByteBuffer bb = ByteBuffer.wrap(bytes); return new UUID(bb.getLong(), bb.getLong()); } /** * Converts a byte array to a BigDecimal. * * @param bytes * @return the char value */ public static BigDecimal toBigDecimal(byte[] bytes) { return toBigDecimal(bytes, 0, bytes.length); } /** * Converts a byte array to a BigDecimal value. * * @param bytes * @param offset * @param length * @return the char value */ public static BigDecimal toBigDecimal(byte[] bytes, int offset, final int length) { if (bytes == null || length < SIZEOF_INT + 1 || (offset + length > bytes.length)) { return null; } int scale = toInt(bytes, offset); byte[] tcBytes = new byte[length - SIZEOF_INT]; System.arraycopy(bytes, offset + SIZEOF_INT, tcBytes, 0, length - SIZEOF_INT); return new BigDecimal(new BigInteger(tcBytes), scale); } /** * Put a BigDecimal value out to the specified byte array position. * * @param bytes the byte array * @param offset position in the array * @param val BigDecimal to write out * @return incremented offset */ public static int putBigDecimal(byte[] bytes, int offset, BigDecimal val) { if (bytes == null) { return offset; } byte[] valueBytes = val.unscaledValue().toByteArray(); byte[] result = new byte[valueBytes.length + SIZEOF_INT]; offset = putInt(result, offset, val.scale()); return putBytes(result, offset, valueBytes, 0, valueBytes.length); } /** * Lexicographically compare two arrays. * @param left left operand * @param right right operand * @return 0 if equal, < 0 if left is less than right, etc. */ public static int compareTo(final byte [] left, final byte [] right) { return LexicographicalComparerHolder.BEST_COMPARER. compareTo(left, 0, left.length, right, 0, right.length); } /** * Lexicographically compare two arrays. * * @param buffer1 left operand * @param buffer2 right operand * @param offset1 Where to start comparing in the left buffer * @param offset2 Where to start comparing in the right buffer * @param length1 How much to compare from the left buffer * @param length2 How much to compare from the right buffer * @return 0 if equal, < 0 if left is less than right, etc. */ public static int compareTo(byte[] buffer1, int offset1, int length1, byte[] buffer2, int offset2, int length2) { return LexicographicalComparerHolder.BEST_COMPARER. compareTo(buffer1, offset1, length1, buffer2, offset2, length2); } interface Comparer<T> { int compareTo(T buffer1, int offset1, int length1, T buffer2, int offset2, int length2); } static Comparer<byte[]> lexicographicalComparerJavaImpl() { return LexicographicalComparerHolder.PureJavaComparer.INSTANCE; } static class LexicographicalComparerHolder { static final String UNSAFE_COMPARER_NAME = LexicographicalComparerHolder.class.getName() + "$UnsafeComparer"; static final Comparer<byte[]> BEST_COMPARER = getBestComparer(); /** * Returns the Unsafe-using Comparer, or falls back to the pure-Java * implementation if unable to do so. */ static Comparer<byte[]> getBestComparer() { try { Class<?> theClass = Class.forName(UNSAFE_COMPARER_NAME); // yes, UnsafeComparer does implement Comparer<byte[]> @SuppressWarnings("unchecked") Comparer<byte[]> comparer = (Comparer<byte[]>) theClass.getEnumConstants()[0]; return comparer; } catch (Throwable t) { // ensure we really catch *everything* return lexicographicalComparerJavaImpl(); } } enum PureJavaComparer implements Comparer<byte[]> { INSTANCE; @Override public int compareTo(byte[] buffer1, int offset1, int length1, byte[] buffer2, int offset2, int length2) { // Short circuit equal case if (buffer1 == buffer2 && offset1 == offset2 && length1 == length2) { return 0; } // Bring WritableComparator code local int end1 = offset1 + length1; int end2 = offset2 + length2; for (int i = offset1, j = offset2; i < end1 && j < end2; i++, j++) { int a = (buffer1[i] & 0xff); int b = (buffer2[j] & 0xff); if (a != b) { return a - b; } } return length1 - length2; } } } /** * Checks two byte arrays for equality. * @param left left operand * @param right right operand * @return True if equal */ public static boolean equals(final byte [] left, final byte [] right) { // Could use Arrays.equals? //noinspection SimplifiableConditionalExpression if (left == right) { return true; } if (left == null || right == null) { return false; } if (left.length != right.length) { return false; } if (left.length == 0) { return true; } // Since we're often comparing adjacent sorted data, // it's usual to have equal arrays except for the very last byte // so check that first if (left[left.length - 1] != right[right.length - 1]) { return false; } return compareTo(left, right) == 0; } /** * Checks segments of two byte arrays for equality. * @param left left operand * @param leftOffset offset from which to start comparison * @param leftLen length of left segment * @param right right operand * @param rightOffset offset from which to start comparison * @param rightLen length of right segment * @return True if two segments are equal */ public static boolean equals(final byte[] left, int leftOffset, int leftLen, final byte[] right, int rightOffset, int rightLen) { // short circuit case if (left == right && leftOffset == rightOffset && leftLen == rightLen) { return true; } // different lengths fast check if (leftLen != rightLen) { return false; } if (leftLen == 0) { return true; } // Since we're often comparing adjacent sorted data, // it's usual to have equal arrays except for the very last byte // so check that first if (left[leftOffset + leftLen - 1] != right[rightOffset + rightLen - 1]) { return false; } return LexicographicalComparerHolder.BEST_COMPARER. compareTo(left, leftOffset, leftLen, right, rightOffset, rightLen) == 0; } /** * Return true if the byte array on the right is a prefix of the byte * array on the left. */ public static boolean startsWith(byte[] bytes, byte[] prefix) { return bytes != null && prefix != null && bytes.length >= prefix.length && LexicographicalComparerHolder.BEST_COMPARER. compareTo(bytes, 0, prefix.length, prefix, 0, prefix.length) == 0; } /** * Compute hash for binary data. * @param b bytes to hash * @return Runs {@link #hashBytes(byte[], int)} on the * passed in array. */ public static int hashCode(final byte [] b) { return hashCode(b, b.length); } /** * Compute hash for binary data. * @param b value * @param length length of the value * @return Runs {@link #hashBytes(byte[], int)} on the * passed in array. */ public static int hashCode(final byte [] b, final int length) { return hashBytes(b, length); } /** Compute hash for binary data. */ public static int hashBytes(byte[] bytes, int offset, int length) { int hash = 1; for (int i = offset; i < offset + length; i++) { hash = (31 * hash) + bytes[i]; } return hash; } /** Compute hash for binary data. */ public static int hashBytes(byte[] bytes, int length) { return hashBytes(bytes, 0, length); } /** * Returns a hash of a byte array as an Integer that can be used as key in Maps. * @param b bytes to hash * @return A hash of <code>b</code> as an Integer that can be used as key in * Maps. */ public static Integer mapKey(final byte [] b) { return hashCode(b); } /** * Returns a hash of a byte array segment as an Integer that can be used as key in Maps. * @param b bytes to hash * @param length length to hash * @return A hash of <code>b</code> as an Integer that can be used as key in * Maps. */ public static Integer mapKey(final byte [] b, final int length) { return hashCode(b, length); } /** * Byte array of size zero. */ public static final byte [] EMPTY_BYTE_ARRAY = new byte [0]; /** * Concatenate two byte arrays. * @param a lower half * @param b upper half * @return New array that has a in lower half and b in upper half. */ public static byte [] add(final byte [] a, final byte [] b) { return add(a, b, EMPTY_BYTE_ARRAY); } /** * Concatenate three byte arrays. * @param a first third * @param b second third * @param c third third * @return New array made from a, b and c */ public static byte [] add(final byte [] a, final byte [] b, final byte [] c) { byte [] result = new byte[a.length + b.length + c.length]; System.arraycopy(a, 0, result, 0, a.length); System.arraycopy(b, 0, result, a.length, b.length); System.arraycopy(c, 0, result, a.length + b.length, c.length); return result; } /** * Returns the values from each provided array combined into a single array. * For example, {@code concat(new byte[] {a, b}, new byte[] {}, new * byte[] {c}} returns the array {@code {a, b, c}}. This method is copied from google guava library. * * @param arrays zero or more {@code byte} arrays * @return a single array containing all the values from the source arrays, in * order */ public static byte[] concat(byte[]... arrays) { int length = 0; for (byte[] array : arrays) { length += array.length; } byte[] result = new byte[length]; int pos = 0; for (byte[] array : arrays) { System.arraycopy(array, 0, result, pos, array.length); pos += array.length; } return result; } /** * Returns first <code>length</code> bytes from byte array. * @param a array * @param length amount of bytes to grab * @return First <code>length</code> bytes from <code>a</code> */ public static byte [] head(final byte [] a, final int length) { if (a.length < length) { return null; } byte [] result = new byte[length]; System.arraycopy(a, 0, result, 0, length); return result; } /** * Returns last <code>length</code> bytes from byte array. * @param a array * @param length amount of bytes to snarf * @return Last <code>length</code> bytes from <code>a</code> */ public static byte [] tail(final byte [] a, final int length) { if (a.length < length) { return null; } byte [] result = new byte[length]; System.arraycopy(a, a.length - length, result, 0, length); return result; } /** * Return a byte array with value in <code>a</code> plus <code>length</code> prepended 0 bytes. * @param a array * @param length new array size * @return Value in <code>a</code> plus <code>length</code> prepended 0 bytes */ public static byte [] padHead(final byte [] a, final int length) { byte [] padding = new byte[length]; for (int i = 0; i < length; i++) { padding[i] = 0; } return add(padding, a); } /** * Return a byte array with value in <code>a</code> plus <code>length</code> appended 0 bytes. * @param a array * @param length new array size * @return Value in <code>a</code> plus <code>length</code> appended 0 bytes */ public static byte [] padTail(final byte [] a, final int length) { byte [] padding = new byte[length]; for (int i = 0; i < length; i++) { padding[i] = 0; } return add(a, padding); } /** * Split passed range. Expensive operation relatively. Uses BigInteger math. * Useful splitting ranges for MapReduce jobs. * @param a Beginning of range * @param b End of range * @param num Number of times to split range. Pass 1 if you want to split * the range in two; i.e. one split. * @return Array of dividing values */ public static byte [][] split(final byte [] a, final byte [] b, final int num) { return split(a, b, false, num); } /** * Split passed range. Expensive operation relatively. Uses BigInteger math. * Useful splitting ranges for MapReduce jobs. * @param a Beginning of range * @param b End of range * @param inclusive Whether the end of range is prefix-inclusive or is * considered an exclusive boundary. Automatic splits are generally exclusive * and manual splits with an explicit range utilize an inclusive end of range. * @param num Number of times to split range. Pass 1 if you want to split * the range in two; i.e. one split. * @return Array of dividing values */ public static byte[][] split(final byte[] a, final byte[] b, boolean inclusive, final int num) { byte[][] ret = new byte[num + 2][]; int i = 0; Iterable<byte[]> iter = iterateOnSplits(a, b, inclusive, num); if (iter == null) { return null; } for (byte[] elem : iter) { ret[i++] = elem; } return ret; } /** * Iterate over keys within the passed range, splitting at an [a,b) boundary. */ public static Iterable<byte[]> iterateOnSplits(final byte[] a, final byte[] b, final int num) { return iterateOnSplits(a, b, false, num); } /** * Iterate over keys within the passed range. */ public static Iterable<byte[]> iterateOnSplits(final byte[] a, final byte[]b, boolean inclusive, final int num) { byte [] aPadded; byte [] bPadded; if (a.length < b.length) { aPadded = padTail(a, b.length - a.length); bPadded = b; } else if (b.length < a.length) { aPadded = a; bPadded = padTail(b, a.length - b.length); } else { aPadded = a; bPadded = b; } if (compareTo(aPadded, bPadded) >= 0) { throw new IllegalArgumentException("b <= a"); } if (num <= 0) { throw new IllegalArgumentException("num cannot be < 0"); } byte [] prependHeader = {1, 0}; final BigInteger startBI = new BigInteger(add(prependHeader, aPadded)); final BigInteger stopBI = new BigInteger(add(prependHeader, bPadded)); BigInteger diffBI = stopBI.subtract(startBI); if (inclusive) { diffBI = diffBI.add(BigInteger.ONE); } final BigInteger splitsBI = BigInteger.valueOf(num + 1); if (diffBI.compareTo(splitsBI) < 0) { return null; } final BigInteger intervalBI; try { intervalBI = diffBI.divide(splitsBI); } catch (Exception e) { return null; } final Iterator<byte[]> iterator = new Iterator<byte[]>() { private int i = -1; @Override public boolean hasNext() { return this.i < num + 1; } @Override public byte[] next() { this.i++; if (this.i == 0) { return a; } if (this.i == num + 1) { return b; } BigInteger curBI = startBI.add(intervalBI.multiply(BigInteger.valueOf(this.i))); byte [] padded = curBI.toByteArray(); if (padded[1] == 0) { padded = tail(padded, padded.length - 2); } else { padded = tail(padded, padded.length - 1); } return padded; } @Override public void remove() { throw new UnsupportedOperationException(); } }; return new Iterable<byte[]>() { @Override public Iterator<byte[]> iterator() { return iterator; } }; } /** * @param bytes array to hash * @param offset offset to start from * @param length length to hash * */ public static int hashCode(byte[] bytes, int offset, int length) { int hash = 1; for (int i = offset; i < offset + length; i++) { hash = (31 * hash) + bytes[i]; } return hash; } /** * Returns an array of byte arrays made from passed array of Text. * @param t operands * @return Array of byte arrays made from passed array of Text */ public static byte [][] toByteArrays(final String [] t) { byte [][] result = new byte[t.length][]; for (int i = 0; i < t.length; i++) { result[i] = Bytes.toBytes(t[i]); } return result; } /** * Returns an array of byte arrays where first and only entry is. * <code>column</code> * @param column operand * @return An array of byte arrays where first and only entry is * <code>column</code> */ public static byte [][] toByteArrays(final String column) { return toByteArrays(toBytes(column)); } /** * Returns an array of byte arrays where first and only entry is. * <code>column</code> * @param column operand * @return An array of byte arrays where first and only entry is * <code>column</code> */ public static byte [][] toByteArrays(final byte [] column) { byte [][] result = new byte[1][]; result[0] = column; return result; } /** * Bytewise binary increment/deincrement of long contained in byte array * on given amount. * * @param value - array of bytes containing long (length <= SIZEOF_LONG) * @param amount value will be incremented on (deincremented if negative) * @return array of bytes containing incremented long (length == SIZEOF_LONG) */ public static byte [] incrementBytes(byte[] value, long amount) { byte[] val = value; if (val.length < SIZEOF_LONG) { // Hopefully this doesn't happen too often. byte [] newvalue; if (val[0] < 0) { newvalue = new byte[]{-1, -1, -1, -1, -1, -1, -1, -1}; } else { newvalue = new byte[SIZEOF_LONG]; } System.arraycopy(val, 0, newvalue, newvalue.length - val.length, val.length); val = newvalue; } else if (val.length > SIZEOF_LONG) { throw new IllegalArgumentException("Increment Bytes - value too big: " + val.length); } if (amount == 0) { return val; } if (val[0] < 0) { return binaryIncrementNeg(val, amount); } return binaryIncrementPos(val, amount); } /* increment/deincrement for positive value */ private static byte [] binaryIncrementPos(byte [] value, long amount) { long amo = amount; int sign = 1; if (amount < 0) { amo = -amount; sign = -1; } for (int i = 0; i < value.length; i++) { int cur = ((int) amo % 256) * sign; amo = (amo >> 8); int val = value[value.length - i - 1] & 0x0ff; int total = val + cur; if (total > 255) { amo += sign; total %= 256; } else if (total < 0) { amo -= sign; } value[value.length - i - 1] = (byte) total; if (amo == 0) { return value; } } return value; } /* increment/deincrement for negative value */ private static byte [] binaryIncrementNeg(byte [] value, long amount) { long amo = amount; int sign = 1; if (amount < 0) { amo = -amount; sign = -1; } for (int i = 0; i < value.length; i++) { int cur = ((int) amo % 256) * sign; amo = (amo >> 8); int val = ((~value[value.length - i - 1]) & 0x0ff) + 1; int total = cur - val; if (total >= 0) { amo += sign; } else if (total < -256) { amo -= sign; total %= 256; } value[value.length - i - 1] = (byte) total; if (amo == 0) { return value; } } return value; } /** * Writes a string as a fixed-size field, padded with zeros. */ public static void writeStringFixedSize(final DataOutput out, String s, int size) throws IOException { byte[] b = toBytes(s); if (b.length > size) { throw new IOException("Trying to write " + b.length + " bytes (" + toStringBinary(b) + ") into a field of length " + size); } out.writeBytes(s); for (int i = 0; i < size - s.length(); ++i) { out.writeByte(0); } } /** * Returns the given prefix, incremented by one, in the form that will be suitable for prefix matching. * @param prefix the prefix to increment for the stop key * @return the stop key to use (may be null if the prefix cannot be incremented) */ // NOTE: null means "read to the end" @Nullable public static byte[] stopKeyForPrefix(byte[] prefix) { for (int i = prefix.length - 1; i >= 0; i--) { int unsigned = prefix[i] & 0xff; if (unsigned < 0xff) { byte[] stopKey = Arrays.copyOf(prefix, i + 1); stopKey[stopKey.length - 1]++; return stopKey; } } // i.e. "read to the end" return null; } /** * Creates immutable sorted map with one entry with given key and value. * @param key key of the entry * @param value value of the entry * @return instance of {@link NavigableMap} * * @deprecated This method will be removed in future release */ @Deprecated public static <T> NavigableMap<byte[], T> immutableSortedMapOf(byte[] key, T value) { // Not really immutable. However, for the sake of removing // guava usage and given this API is deprecated, it should be fine. TreeMap<byte[], T> result = new TreeMap<byte[], T>(BYTES_COMPARATOR); result.put(key, value); return result; } /** * Creates immutable sorted map with two entries with given keys and values. * @param key1 key of the first entry * @param value1 value of the first entry * @param key2 key of the second entry * @param value2 value of the second entry * @return instance of {@link NavigableMap} * * @deprecated This method will be removed in future release */ @Deprecated public static <T> NavigableMap<byte[], T> immutableSortedMapOf(byte[] key1, T value1, byte[] key2, T value2) { // Not really immutable. However, for the sake of removing // guava usage and given this API is deprecated, it should be fine. TreeMap<byte[], T> result = new TreeMap<byte[], T>(BYTES_COMPARATOR); result.put(key1, value1); result.put(key2, value2); return result; } }