1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518
|
/*
* Normaliz
* Copyright (C) 2007-2022 W. Bruns, B. Ichim, Ch. Soeger, U. v. d. Ohe
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <https://www.gnu.org/licenses/>.
*
* As an exception, when this program is distributed through (i) the App Store
* by Apple Inc.; (ii) the Mac App Store by Apple Inc.; or (iii) Google Play
* by Google Inc., then that store may impose any digital rights management,
* device limits and/or redistribution restrictions that are required by its
* terms of service.
*/
#ifndef LIBNORMALIZ_DYNAMIC_BITSET_H
#define LIBNORMALIZ_DYNAMIC_BITSET_H
// uncomment to disable assert()
// #define NDEBUG
#include <cassert>
#include <climits> // for CHAR_BIT
#include <iostream>
#include <vector>
namespace libnormaliz {
// An implementation of a subset of boost::dynamic_bitset.
//
// This provides only a subset of boost::dynamic_bitset, based on
// those features Normaliz actually uses. This has the advantage
// that the resulting code is much shorter and simpler. Also, unused
// code often has bugs, while code that was never written is bug free.
//
// Among the features that are left out are:
// - custom allocators -- always use the default allocator
// - custom block/limb types -- we hardcode unsigned long long
// - initialization of bitsets with non-zero data in constructors, resize()
// - iterators
// - shift operators
// - appending, pop_back, push_back
// - conversion from/to strings
// - is_proper_subset_of
// - istream operator>>
//
// Of course any of these could be added in the future, should the need arise
class dynamic_bitset {
public:
typedef unsigned long long limb_t;
static const size_t bits_per_limb = CHAR_BIT * sizeof(limb_t);
static const size_t npos = static_cast<size_t>(-1);
// proxy object used for implementing write access to the bitset
// via operator[].
class reference {
friend class dynamic_bitset;
void operator&(); // left undefined
reference(dynamic_bitset& x, size_t pos) : _limb(x._limbs[limb_index(pos)]), _mask(bit_mask(pos)) {
assert(pos < x.size());
}
public:
operator bool() const {
return (_limb & _mask) != 0;
}
bool operator~() const {
return (_limb & _mask) == 0;
}
// b[i] = x
reference& operator=(bool x) {
set_to(x);
return *this;
}
// b[i] = b[j]
reference& operator=(const reference& x) {
set_to(x);
return *this;
}
reference& flip() {
_limb ^= _mask;
return *this;
}
reference& operator|=(bool x) {
if (x)
_limb |= _mask;
return *this;
}
reference& operator&=(bool x) {
if (!x)
_limb &= ~_mask;
return *this;
}
reference& operator^=(bool x) {
if (x)
_limb ^= _mask;
return *this;
}
reference& operator-=(bool x) {
if (x)
_limb &= ~_mask;
return *this;
}
private:
limb_t& _limb;
const limb_t _mask;
void set_to(bool x) {
if (x)
_limb |= _mask;
else
_limb &= ~_mask;
}
};
typedef bool const_reference;
// constructors (copy & move constructors are provided by the compiler)
dynamic_bitset() : _limbs(), _total_bits(0) {
}
explicit dynamic_bitset(size_t nbits) : _limbs(limbs_required(nbits)), _total_bits(nbits) {
}
//
// operations involving the size of the bitset
//
void resize(size_t nbits) {
if (nbits == _total_bits)
return;
_limbs.resize(limbs_required(nbits), 0);
_total_bits = nbits;
cleanup_last_limb();
}
void clear() {
_limbs.clear();
_total_bits = 0;
}
size_t size() const {
return _total_bits;
}
bool empty() const {
return size() == 0;
}
//
// operations for accessing or modifying individual bits
//
// set the bit at the given position to the given value
dynamic_bitset& set(size_t pos, bool val) {
return val ? set(pos) : reset(pos);
}
// set the bit at the given position to 1
dynamic_bitset& set(size_t pos) {
assert(pos < size());
_limbs[limb_index(pos)] |= bit_mask(pos);
return *this;
}
// set the bit at the given position to 0
dynamic_bitset& reset(size_t pos) {
assert(pos < size());
_limbs[limb_index(pos)] &= ~bit_mask(pos);
return *this;
}
// flip the bit at the given position
dynamic_bitset& flip(size_t pos) {
assert(pos < size());
_limbs[limb_index(pos)] ^= bit_mask(pos);
return *this;
}
// return true if the bit at the given position is 1, otherwise false
bool test(size_t pos) const {
assert(pos < size());
return (_limbs[limb_index(pos)] & bit_mask(pos)) != 0;
}
// subscript operators
const_reference operator[](size_t pos) const {
return test(pos);
}
reference operator[](size_t pos) {
assert(pos < size());
return reference(*this, pos);
}
//
// operations for accessing and modifying the whole bitset
//
// set all bits to 1
dynamic_bitset& set() {
for (limb_t& limb : _limbs) {
limb = all_ones_limb;
}
cleanup_last_limb(); // cleanup in the last limb
return *this;
}
// set all bits to 0
dynamic_bitset& reset() {
for (limb_t& limb : _limbs) {
limb = 0;
}
return *this;
}
// flip all bits
dynamic_bitset& flip() {
for (limb_t& limb : _limbs) {
limb ^= all_ones_limb;
}
cleanup_last_limb(); // cleanup in the last limb
return *this;
}
// return true if any bit is set to 1, otherwise false
bool any() const {
for (const limb_t& limb : _limbs) {
if (limb != 0)
return true;
}
return false;
}
// return true if no bit is set to 1, otherwise false
bool none() const {
return !any();
}
// return the number of bits which are set to 1
size_t count() const {
size_t sum = 0;
for (const limb_t& limb : _limbs) {
sum += popcount(limb);
}
return sum;
}
//
// in-place operators
//
dynamic_bitset& operator&=(const dynamic_bitset& rhs) {
assert(size() == rhs.size());
for (size_t i = 0; i < _limbs.size(); ++i) {
_limbs[i] &= rhs._limbs[i];
}
return *this;
}
dynamic_bitset& operator|=(const dynamic_bitset& rhs) {
assert(size() == rhs.size());
for (size_t i = 0; i < _limbs.size(); ++i) {
_limbs[i] |= rhs._limbs[i];
}
return *this;
}
dynamic_bitset& operator^=(const dynamic_bitset& rhs) {
assert(size() == rhs.size());
for (size_t i = 0; i < _limbs.size(); ++i) {
_limbs[i] ^= rhs._limbs[i];
}
return *this;
}
dynamic_bitset& operator-=(const dynamic_bitset& rhs) {
assert(size() == rhs.size());
for (size_t i = 0; i < _limbs.size(); ++i) {
_limbs[i] &= ~rhs._limbs[i];
}
return *this;
}
//
// non-modifying unary and binary operators
//
dynamic_bitset operator~() const {
dynamic_bitset result(*this);
result.flip();
return result;
}
dynamic_bitset operator&(const dynamic_bitset& rhs) const {
dynamic_bitset result(*this);
return result &= rhs;
}
dynamic_bitset operator|(const dynamic_bitset& rhs) const {
dynamic_bitset result(*this);
return result |= rhs;
}
dynamic_bitset operator^(const dynamic_bitset& rhs) const {
dynamic_bitset result(*this);
return result ^= rhs;
}
dynamic_bitset operator-(const dynamic_bitset& rhs) const {
dynamic_bitset result(*this);
return result -= rhs;
}
//
// comparison of two bitsets
//
bool operator==(const dynamic_bitset& x) const {
return _total_bits == x._total_bits && _limbs == x._limbs;
}
bool operator!=(const dynamic_bitset& x) const {
return !(*this == x);
}
bool operator<(const dynamic_bitset& x) const { // for use in std::set, std::map
if (_total_bits == x._total_bits) {
for (size_t i = 1; i <= _limbs.size(); ++i) {
if (_limbs[_limbs.size() - i] != x._limbs[_limbs.size() - i])
return (_limbs[_limbs.size() - i] < x._limbs[_limbs.size() - i]);
}
return false;
}
return _total_bits < x._total_bits;
}
// subsets
bool is_subset_of(const dynamic_bitset& x) const {
assert(size() == x.size());
for (size_t i = 0; i < _limbs.size(); ++i) {
if ((_limbs[i] & ~x._limbs[i]) != 0)
return false;
}
return true;
}
bool intersects(const dynamic_bitset& x) const {
assert(size() == x.size());
for (size_t i = 0; i < _limbs.size(); ++i) {
if ((_limbs[i] & x._limbs[i]) != 0)
return true;
}
return false;
}
//
// find functions
//
// find the index of the first set bit, or npos if no bit is set
size_t find_first() const {
for (size_t i = 0; i < _limbs.size(); ++i) {
if (_limbs[i] != 0)
return i * bits_per_limb + limb_find_first(_limbs[i]);
}
return npos;
}
size_t find_next(size_t prev) const {
if (prev == npos)
return npos;
// first bit to consider is prev + 1
++prev;
if (prev >= size())
return npos;
// first check whether the limb in which prev is contained
// contains any set bits after prev
size_t i = limb_index(prev);
limb_t limb = _limbs[i] >> (prev % bits_per_limb);
if (limb != 0)
return prev + limb_find_first(limb);
// for the remaining limbs, proceed as in find_first()
++i;
for (; i < _limbs.size(); ++i) {
if (_limbs[i] != 0)
return i * bits_per_limb + limb_find_first(_limbs[i]);
}
return npos;
}
private:
// the actual bits are stored in a vector<>, which means we don't have
// to worry about memory management
std::vector<limb_t> _limbs;
// the size of this bitset, in bits; note that this may be small than
// _limbs.size() * bits_per_limb
size_t _total_bits;
static const limb_t all_zero_limb = limb_t(0);
static const limb_t all_ones_limb = limb_t(~all_zero_limb);
static size_t limbs_required(size_t nbits) {
return (nbits + bits_per_limb - 1) / bits_per_limb;
}
static size_t limb_index(size_t pos) {
return pos / bits_per_limb;
}
static limb_t bit_mask(size_t pos) {
return limb_t(limb_t(1) << (pos % bits_per_limb));
}
// zero out unused bits in last limb
void cleanup_last_limb() {
size_t shift = _total_bits % bits_per_limb;
if (shift > 0) {
_limbs[_limbs.size() - 1] &= ~(all_ones_limb << shift);
}
}
// Find the first set bit in the given non-zero limb, were we start at the
// least significant bit (LSB) and move to the most significant (MSB).
// The behavior for zero limbs is undefined.
//
// This is an internal helper function, used by find_first and find_next.
size_t limb_find_first(limb_t limb) const {
assert(limb != 0);
// ideally we'd like to use __builtin_ctzll; but this may not be
// supported by the compiler; so we use autoconf to check for it; but
// then client code using this header may for whatever reason not have
// the autoconf macros, so this code is written in such a way that we
// always can safely fall back to the generic code.
// TODO: if we want to support a limb_t other than 'unsigned long long',
// then this code will need to be adjusted.
#if HAVE___BUILTIN_CTZLL
return __builtin_ctzll(limb);
#endif
// we rely on the compiler to optimize the following "runtime checks"
// away in the generated code
if (bits_per_limb == 64) {
// source for the following table and code:
// https://www.chessprogramming.org/BitScan#De_Bruijn_Multiplication
static const char DeBrujinPositions[64] = {
0, 1, 48, 2, 57, 49, 28, 3, 61, 58, 50, 42, 38, 29, 17, 4, 62, 55, 59, 36, 53, 51,
43, 22, 45, 39, 33, 30, 24, 18, 12, 5, 63, 47, 56, 27, 60, 41, 37, 16, 54, 35, 52, 21,
44, 32, 23, 11, 46, 26, 40, 15, 34, 20, 31, 10, 25, 14, 19, 9, 13, 8, 7, 6,
};
static const limb_t magic = 0x03f79d71b4cb0a89;
return DeBrujinPositions[((limb & -limb) * magic) >> 58];
}
else {
assert(bits_per_limb == 32);
// source for the following table and code:
// http://graphics.stanford.edu/~seander/bithacks.html#ZerosOnRightMultLookup
static const char DeBrujinPositions[32] = {
0, 1, 28, 2, 29, 14, 24, 3, 30, 22, 20, 15, 25, 17, 4, 8,
31, 27, 13, 23, 21, 19, 16, 7, 26, 12, 18, 6, 11, 5, 10, 9,
};
static const limb_t magic = 0x077cb531;
return DeBrujinPositions[((limb & -limb) * magic) >> 27];
}
}
// Count the number of set bits in the given limb.
//
// This is an internal helper function, used by count.
size_t popcount(limb_t limb) const {
// ideally we'd like to use __builtin_popcountll; but this may not be
// supported by the compiler; so we use autoconf to check for it; but
// then client code using this header may for whatever reason not have
// the autoconf macros, so this code is written in such a way that we
// always can safely fall back to the generic code.
// TODO: if we want to support a limb_t other than 'unsigned long long',
// then this code will need to be adjusted.
#if HAVE___BUILTIN_POPCOUNTLL
return __builtin_popcountll(limb);
#endif
// we rely on the compiler to optimize the following "runtime checks"
// away in the generated code
if (bits_per_limb == 64) {
limb = (limb & 0x5555555555555555L) + ((limb >> 1) & 0x5555555555555555L);
limb = (limb & 0x3333333333333333L) + ((limb >> 2) & 0x3333333333333333L);
limb = (limb + (limb >> 4)) & 0x0f0f0f0f0f0f0f0fL;
limb = (limb + (limb >> 8));
limb = (limb + (limb >> 16));
limb = (limb + (limb >> 32)) & 0x00000000000000ffL;
return limb;
}
else {
assert(bits_per_limb == 32);
limb = (limb & 0x55555555) + ((limb >> 1) & 0x55555555);
limb = (limb & 0x33333333) + ((limb >> 2) & 0x33333333);
limb = (limb + (limb >> 4)) & 0x0f0f0f0f;
limb = (limb + (limb >> 8));
limb = (limb + (limb >> 16)) & 0x000000ff;
return limb;
}
}
};
// implement output to an ostream; this is mostly for debugging,
// hence no effort was made to optimize this
static inline std::ostream& operator<<(std::ostream& os, const dynamic_bitset& bitset) {
size_t pos = bitset.size();
while (pos--) {
os << bitset[pos];
}
return os;
}
} // namespace libnormaliz
#endif /* LIBNORMALIZ_DYNAMIC_BITSET_H */
|