/** * Copyright © 2014 Mattias Andrée (maandree@member.fsf.org) * * This program is free software: you can redistribute it and/or modify * it under the terms of the GNU Affero General Public License as published by * the Free Software Foundation, either version 3 of the License, or * (at your option) any later version. * * This program 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 Affero General Public License for more details. * * You should have received a copy of the GNU Affero General Public License * along with this program. If not, see . */ package algorithms.bits; /** * Operations on individual bits */ public class Bits { £} /** * Lookup table for the number of set bits in a byte */ private static byte[] ONES_TABLE_256 = { £(table 4 0) }; /* ONES_TABLE_256[0] = 0; * for (int i = 0; i < 256; i++) * ONES_TABLE_256[i] = (i & 1) + ONES_TABLE_256[i / 2]; */ £S=${T_S#*_} /** * Sets or clears individual bits in an integer * * @param value The value to modify * @param mask Mask of bits to modify * @param flag 1 if the bit should be set, 0 if the bit should be clears * @return The value with the bits modified */ public static £{T} set_clear(£{T} value, £{T} mask, £{T} flag) { return (£{T})(value ^ ((-flag ^ value) & mask)); } /** * Sets or clears individual bits in an integer, superscalar version * * @param value The value to modify * @param mask Mask of bits to modify * @param flag 1 if the bit should be set, 0 if the bit should be clears * @return The value with the bits modified */ public static £{T} set_clear_superscalar(£{T} value, £{T} mask, £{T} flag) { return (£{T})((value & ~mask) | (-flag & mask)); } /** * Merge bits from two values * * @param zero Integer whose bits should be kept where the mask has zeroes * @param one Integer whose bits should be kept where the mask has onces * @param mask The merge mask * @return {@code (zero & ~mask) | (one & mask)} */ public static £{T} merge(£{T} zero, £{T} one, £{T} mask) { return (£{T})(zero ^ ((£{T})(zero ^ one) & mask)); } /** * Clears the least significant bit set * * @param value The integer * @return The value with its least significant set bit cleared */ public static £{T} clearLeastSignificant(£{T} value) { return (£{T})(value & (value - 1)); } /** * Calculate the number of set bits in an integer, naïve version * * @param value The integer * @return The number of set bits */ public static £{T} ones_naïve(£{T} value) { £{T} count = 0; for (; value != 0; value >>>= 1) count += (£{T})(value & 1); return count; } /** * Calculate the number of set bits in an integer, Wegner's version * * @param value The integer * @return The number of set bits */ public static £{T} ones_wegner(£{T} value) { £{T} count = 0; for (; value != 0; count++) value &= value - 1; /* clear the least significant bit set */ return count; } /** * Calculate the number of set bits in an integer, partial lookup table version * * @param value The integer * @return The number of set bits */ public static byte ones_table(£{T} value) { £>function _ { echo "ONES_TABLE_256[(int)((value >> $1) & 255)]" ; } return (byte)((byte)(£(_ 0) + £(_ 8)) + (byte)(£(_ 16) + £(_ 24))); /* In C you can split the value by getting the address of the value and cast the pointer to char* */ } /** * Calculate the number of set bits in an integer, 64-bits instructions version * * @param value The integer * @return The number of set bits */ public static long ones_64bits(£{T} value) { long v = value, rc; £>(( $S > 1 )) && if ((v & (1L << 14L) - 1L) == v) { rc = (v * 0x200040008001L & 0x111111111111111L) % 0xF; } £>if (( $S >= 2 )); then else £>(( $S > 3 )) && if ((v & (1L << 24L) - 1L) == v) { rc = ((v & 0xFFF) * 0x1001001001001L & 0x84210842108421L) % 0x1F; rc += (((v & 0XFFF000) >> 12) * 0x1001001001001L & 0x84210842108421L) % 0x1F; } £>if (( $S > 3 )); then else £>(( $S > 4 )) && if ((v & (1L << 32L) - 1L) == v) { rc = ((v & 0xFFF) * 0x1001001001001L & 0x84210842108421L) % 0x1F; rc += (((v & 0xFFF000) >> 12) * 0x1001001001001L & 0x84210842108421L) % 0x1F; rc += ((v >> 24) * 0x1001001001001L & 0x84210842108421L) % 0x1F; } £>if (( $S > 4 )); then else { rc = ones_64bits(v & (1L << 32L) - 1L); rc += ones_64bits(v >>> 32); } £>fi;fi;fi return rc; } /** * Calculate the number of set bits in an integer, naïve parallel version * * @param value The integer * @return The number of set bits */ public static £{T} ones_parallel(£{T} value) { £{T} rc = value; rc = (£{T})(((rc >> 1) & (£{T})0xAAAAAAAAAAAAAAAAL) + (rc & (£{T})0xAAAAAAAAAAAAAAAAL)); rc = (£{T})(((rc >> 2) & (£{T})0xCCCCCCCCCCCCCCCCL) + (rc & (£{T})0xCCCCCCCCCCCCCCCCL)); rc = (£{T})(((rc >> 4) & (£{T})0xF0F0F0F0F0F0F0F0L) + (rc & (£{T})0xF0F0F0F0F0F0F0F0L)); £>(( $S > 1 )) && rc = (£{T})(((rc >> 8) & (£{T})0xFF00FF00FF00FF00L) + (rc & (£{T})0xFF00FF00FF00FF00L)); £>(( $S > 2 )) && rc = (£{T})(((rc >> 16) & (£{T})0xFFFF0000FFFF0000L) + (rc & (£{T})0xFFFF0000FFFF0000L)); £>(( $S > 4 )) && rc = (£{T})(((rc >> 32) & (£{T})0xFFFFFFFF00000000L) + (rc & (£{T})0xFFFFFFFF00000000L)); return rc; } /** * Calculate the number of set bits in an integer, optimised parallel version * * @param value The integer * @return The number of set bits */ public static £{T} ones_optimised_parallel(£{T} value) { £{T} rc = (£{T})(value - ((value >> 1) & (£{T})0x5555555555555555L)); rc = (£{T})(((rc >> 2) & (£{T})0x3333333333333333L) + (rc & (£{T})0x3333333333333333L)); rc = (£{T})(((rc >> 4) + rc) & (£{T})0x0F0F0F0F0F0F0F0FL); £>(( $S > 1 )) && rc = (£{T})(((rc >> 8) + rc) & (£{T})0x00FF00FF00FF00FFL); £>(( $S > 2 )) && rc = (£{T})(((rc >> 16) + rc) & (£{T})0x0000FFFF0000FFFFL); £>(( $S > 4 )) && rc = (£{T})(((rc >> 32) + rc) & (£{T})0x00000000FFFFFFFFL); return rc; } /** * Calculate the number of set bits in an integer, parallel–64bits-like hybrid version, probably the best version * * @param value The integer * @return The number of set bits */ public static £{T} ones_hybrid(£{T} value) { £>L1="($T)$(bc <<< "(256^$S - 1) / 3")L" £>L2="($T)$(bc <<< "(256^$S - 1) / 15 * 3")L" £>L3="($T)$(bc <<< "(256^$S - 1) / 255 * 15")L" £>L4="($T)$(bc <<< "(256^$S - 1) / 255")L" value -= (value >> 1) & £{L1}; value = (£{T})((value & £{L2}) + ((value >> 2) & £{L2})); value = (value + (value >> 4)) & £{L3}; value = (value * £{L4}) >> (($S - 1) * 8); return value; /* Only applicable upto 128 bits */ } £>done }