[2725] | 1 | /*
|
---|
| 2 | * bzip2 is written by Julian Seward <jseward@bzip.org>.
|
---|
| 3 | * Adapted for busybox by Denys Vlasenko <vda.linux@googlemail.com>.
|
---|
| 4 | * See README and LICENSE files in this directory for more information.
|
---|
| 5 | */
|
---|
| 6 |
|
---|
| 7 | /*-------------------------------------------------------------*/
|
---|
| 8 | /*--- Compression machinery (not incl block sorting) ---*/
|
---|
| 9 | /*--- compress.c ---*/
|
---|
| 10 | /*-------------------------------------------------------------*/
|
---|
| 11 |
|
---|
| 12 | /* ------------------------------------------------------------------
|
---|
| 13 | This file is part of bzip2/libbzip2, a program and library for
|
---|
| 14 | lossless, block-sorting data compression.
|
---|
| 15 |
|
---|
| 16 | bzip2/libbzip2 version 1.0.4 of 20 December 2006
|
---|
| 17 | Copyright (C) 1996-2006 Julian Seward <jseward@bzip.org>
|
---|
| 18 |
|
---|
| 19 | Please read the WARNING, DISCLAIMER and PATENTS sections in the
|
---|
| 20 | README file.
|
---|
| 21 |
|
---|
| 22 | This program is released under the terms of the license contained
|
---|
| 23 | in the file LICENSE.
|
---|
| 24 | ------------------------------------------------------------------ */
|
---|
| 25 |
|
---|
| 26 | /* CHANGES
|
---|
| 27 | * 0.9.0 -- original version.
|
---|
| 28 | * 0.9.0a/b -- no changes in this file.
|
---|
| 29 | * 0.9.0c -- changed setting of nGroups in sendMTFValues()
|
---|
| 30 | * so as to do a bit better on small files
|
---|
| 31 | */
|
---|
| 32 |
|
---|
| 33 | /* #include "bzlib_private.h" */
|
---|
| 34 |
|
---|
| 35 | /*---------------------------------------------------*/
|
---|
| 36 | /*--- Bit stream I/O ---*/
|
---|
| 37 | /*---------------------------------------------------*/
|
---|
| 38 |
|
---|
| 39 | /*---------------------------------------------------*/
|
---|
| 40 | static
|
---|
| 41 | void BZ2_bsInitWrite(EState* s)
|
---|
| 42 | {
|
---|
| 43 | s->bsLive = 0;
|
---|
| 44 | s->bsBuff = 0;
|
---|
| 45 | }
|
---|
| 46 |
|
---|
| 47 |
|
---|
| 48 | /*---------------------------------------------------*/
|
---|
| 49 | static NOINLINE
|
---|
| 50 | void bsFinishWrite(EState* s)
|
---|
| 51 | {
|
---|
| 52 | while (s->bsLive > 0) {
|
---|
| 53 | s->zbits[s->numZ] = (uint8_t)(s->bsBuff >> 24);
|
---|
| 54 | s->numZ++;
|
---|
| 55 | s->bsBuff <<= 8;
|
---|
| 56 | s->bsLive -= 8;
|
---|
| 57 | }
|
---|
| 58 | }
|
---|
| 59 |
|
---|
| 60 |
|
---|
| 61 | /*---------------------------------------------------*/
|
---|
| 62 | static
|
---|
| 63 | /* Helps only on level 5, on other levels hurts. ? */
|
---|
[3232] | 64 | #if CONFIG_BZIP2_FAST >= 5
|
---|
[2725] | 65 | ALWAYS_INLINE
|
---|
| 66 | #endif
|
---|
| 67 | void bsW(EState* s, int32_t n, uint32_t v)
|
---|
| 68 | {
|
---|
| 69 | while (s->bsLive >= 8) {
|
---|
| 70 | s->zbits[s->numZ] = (uint8_t)(s->bsBuff >> 24);
|
---|
| 71 | s->numZ++;
|
---|
| 72 | s->bsBuff <<= 8;
|
---|
| 73 | s->bsLive -= 8;
|
---|
| 74 | }
|
---|
| 75 | s->bsBuff |= (v << (32 - s->bsLive - n));
|
---|
| 76 | s->bsLive += n;
|
---|
| 77 | }
|
---|
| 78 |
|
---|
| 79 |
|
---|
| 80 | /*---------------------------------------------------*/
|
---|
| 81 | static
|
---|
| 82 | void bsPutU32(EState* s, unsigned u)
|
---|
| 83 | {
|
---|
| 84 | bsW(s, 8, (u >> 24) & 0xff);
|
---|
| 85 | bsW(s, 8, (u >> 16) & 0xff);
|
---|
| 86 | bsW(s, 8, (u >> 8) & 0xff);
|
---|
| 87 | bsW(s, 8, u & 0xff);
|
---|
| 88 | }
|
---|
| 89 |
|
---|
| 90 |
|
---|
| 91 | /*---------------------------------------------------*/
|
---|
| 92 | static
|
---|
| 93 | void bsPutU16(EState* s, unsigned u)
|
---|
| 94 | {
|
---|
| 95 | bsW(s, 8, (u >> 8) & 0xff);
|
---|
| 96 | bsW(s, 8, u & 0xff);
|
---|
| 97 | }
|
---|
| 98 |
|
---|
| 99 |
|
---|
| 100 | /*---------------------------------------------------*/
|
---|
| 101 | /*--- The back end proper ---*/
|
---|
| 102 | /*---------------------------------------------------*/
|
---|
| 103 |
|
---|
| 104 | /*---------------------------------------------------*/
|
---|
| 105 | static
|
---|
| 106 | void makeMaps_e(EState* s)
|
---|
| 107 | {
|
---|
| 108 | int i;
|
---|
| 109 | s->nInUse = 0;
|
---|
| 110 | for (i = 0; i < 256; i++) {
|
---|
| 111 | if (s->inUse[i]) {
|
---|
| 112 | s->unseqToSeq[i] = s->nInUse;
|
---|
| 113 | s->nInUse++;
|
---|
| 114 | }
|
---|
| 115 | }
|
---|
| 116 | }
|
---|
| 117 |
|
---|
| 118 |
|
---|
| 119 | /*---------------------------------------------------*/
|
---|
| 120 | static NOINLINE
|
---|
| 121 | void generateMTFValues(EState* s)
|
---|
| 122 | {
|
---|
| 123 | uint8_t yy[256];
|
---|
| 124 | int32_t i, j;
|
---|
| 125 | int32_t zPend;
|
---|
| 126 | int32_t wr;
|
---|
| 127 | int32_t EOB;
|
---|
| 128 |
|
---|
| 129 | /*
|
---|
| 130 | * After sorting (eg, here),
|
---|
| 131 | * s->arr1[0 .. s->nblock-1] holds sorted order,
|
---|
| 132 | * and
|
---|
| 133 | * ((uint8_t*)s->arr2)[0 .. s->nblock-1]
|
---|
| 134 | * holds the original block data.
|
---|
| 135 | *
|
---|
| 136 | * The first thing to do is generate the MTF values,
|
---|
| 137 | * and put them in ((uint16_t*)s->arr1)[0 .. s->nblock-1].
|
---|
| 138 | *
|
---|
| 139 | * Because there are strictly fewer or equal MTF values
|
---|
| 140 | * than block values, ptr values in this area are overwritten
|
---|
| 141 | * with MTF values only when they are no longer needed.
|
---|
| 142 | *
|
---|
| 143 | * The final compressed bitstream is generated into the
|
---|
| 144 | * area starting at &((uint8_t*)s->arr2)[s->nblock]
|
---|
| 145 | *
|
---|
| 146 | * These storage aliases are set up in bzCompressInit(),
|
---|
| 147 | * except for the last one, which is arranged in
|
---|
| 148 | * compressBlock().
|
---|
| 149 | */
|
---|
| 150 | uint32_t* ptr = s->ptr;
|
---|
| 151 | uint8_t* block = s->block;
|
---|
| 152 | uint16_t* mtfv = s->mtfv;
|
---|
| 153 |
|
---|
| 154 | makeMaps_e(s);
|
---|
| 155 | EOB = s->nInUse+1;
|
---|
| 156 |
|
---|
| 157 | for (i = 0; i <= EOB; i++)
|
---|
| 158 | s->mtfFreq[i] = 0;
|
---|
| 159 |
|
---|
| 160 | wr = 0;
|
---|
| 161 | zPend = 0;
|
---|
| 162 | for (i = 0; i < s->nInUse; i++)
|
---|
| 163 | yy[i] = (uint8_t) i;
|
---|
| 164 |
|
---|
| 165 | for (i = 0; i < s->nblock; i++) {
|
---|
| 166 | uint8_t ll_i;
|
---|
| 167 | AssertD(wr <= i, "generateMTFValues(1)");
|
---|
| 168 | j = ptr[i] - 1;
|
---|
| 169 | if (j < 0)
|
---|
| 170 | j += s->nblock;
|
---|
| 171 | ll_i = s->unseqToSeq[block[j]];
|
---|
| 172 | AssertD(ll_i < s->nInUse, "generateMTFValues(2a)");
|
---|
| 173 |
|
---|
| 174 | if (yy[0] == ll_i) {
|
---|
| 175 | zPend++;
|
---|
| 176 | } else {
|
---|
| 177 | if (zPend > 0) {
|
---|
| 178 | zPend--;
|
---|
| 179 | while (1) {
|
---|
| 180 | if (zPend & 1) {
|
---|
| 181 | mtfv[wr] = BZ_RUNB; wr++;
|
---|
| 182 | s->mtfFreq[BZ_RUNB]++;
|
---|
| 183 | } else {
|
---|
| 184 | mtfv[wr] = BZ_RUNA; wr++;
|
---|
| 185 | s->mtfFreq[BZ_RUNA]++;
|
---|
| 186 | }
|
---|
| 187 | if (zPend < 2) break;
|
---|
| 188 | zPend = (uint32_t)(zPend - 2) / 2;
|
---|
| 189 | /* bbox: unsigned div is easier */
|
---|
| 190 | };
|
---|
| 191 | zPend = 0;
|
---|
| 192 | }
|
---|
| 193 | {
|
---|
| 194 | register uint8_t rtmp;
|
---|
| 195 | register uint8_t* ryy_j;
|
---|
| 196 | register uint8_t rll_i;
|
---|
| 197 | rtmp = yy[1];
|
---|
| 198 | yy[1] = yy[0];
|
---|
| 199 | ryy_j = &(yy[1]);
|
---|
| 200 | rll_i = ll_i;
|
---|
| 201 | while (rll_i != rtmp) {
|
---|
| 202 | register uint8_t rtmp2;
|
---|
| 203 | ryy_j++;
|
---|
| 204 | rtmp2 = rtmp;
|
---|
| 205 | rtmp = *ryy_j;
|
---|
| 206 | *ryy_j = rtmp2;
|
---|
| 207 | };
|
---|
| 208 | yy[0] = rtmp;
|
---|
| 209 | j = ryy_j - &(yy[0]);
|
---|
| 210 | mtfv[wr] = j+1;
|
---|
| 211 | wr++;
|
---|
| 212 | s->mtfFreq[j+1]++;
|
---|
| 213 | }
|
---|
| 214 | }
|
---|
| 215 | }
|
---|
| 216 |
|
---|
| 217 | if (zPend > 0) {
|
---|
| 218 | zPend--;
|
---|
| 219 | while (1) {
|
---|
| 220 | if (zPend & 1) {
|
---|
| 221 | mtfv[wr] = BZ_RUNB;
|
---|
| 222 | wr++;
|
---|
| 223 | s->mtfFreq[BZ_RUNB]++;
|
---|
| 224 | } else {
|
---|
| 225 | mtfv[wr] = BZ_RUNA;
|
---|
| 226 | wr++;
|
---|
| 227 | s->mtfFreq[BZ_RUNA]++;
|
---|
| 228 | }
|
---|
| 229 | if (zPend < 2)
|
---|
| 230 | break;
|
---|
| 231 | zPend = (uint32_t)(zPend - 2) / 2;
|
---|
| 232 | /* bbox: unsigned div is easier */
|
---|
| 233 | };
|
---|
| 234 | zPend = 0;
|
---|
| 235 | }
|
---|
| 236 |
|
---|
| 237 | mtfv[wr] = EOB;
|
---|
| 238 | wr++;
|
---|
| 239 | s->mtfFreq[EOB]++;
|
---|
| 240 |
|
---|
| 241 | s->nMTF = wr;
|
---|
| 242 | }
|
---|
| 243 |
|
---|
| 244 |
|
---|
| 245 | /*---------------------------------------------------*/
|
---|
| 246 | #define BZ_LESSER_ICOST 0
|
---|
| 247 | #define BZ_GREATER_ICOST 15
|
---|
| 248 |
|
---|
| 249 | static NOINLINE
|
---|
| 250 | void sendMTFValues(EState* s)
|
---|
| 251 | {
|
---|
| 252 | int32_t v, t, i, j, gs, ge, totc, bt, bc, iter;
|
---|
| 253 | int32_t nSelectors, alphaSize, minLen, maxLen, selCtr;
|
---|
[3232] | 254 | int32_t nGroups;
|
---|
[2725] | 255 |
|
---|
| 256 | /*
|
---|
| 257 | * uint8_t len[BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE];
|
---|
| 258 | * is a global since the decoder also needs it.
|
---|
| 259 | *
|
---|
| 260 | * int32_t code[BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE];
|
---|
| 261 | * int32_t rfreq[BZ_N_GROUPS][BZ_MAX_ALPHA_SIZE];
|
---|
| 262 | * are also globals only used in this proc.
|
---|
| 263 | * Made global to keep stack frame size small.
|
---|
| 264 | */
|
---|
| 265 | #define code sendMTFValues__code
|
---|
| 266 | #define rfreq sendMTFValues__rfreq
|
---|
| 267 | #define len_pack sendMTFValues__len_pack
|
---|
| 268 |
|
---|
| 269 | uint16_t cost[BZ_N_GROUPS];
|
---|
| 270 | int32_t fave[BZ_N_GROUPS];
|
---|
| 271 |
|
---|
| 272 | uint16_t* mtfv = s->mtfv;
|
---|
| 273 |
|
---|
| 274 | alphaSize = s->nInUse + 2;
|
---|
| 275 | for (t = 0; t < BZ_N_GROUPS; t++)
|
---|
| 276 | for (v = 0; v < alphaSize; v++)
|
---|
| 277 | s->len[t][v] = BZ_GREATER_ICOST;
|
---|
| 278 |
|
---|
| 279 | /*--- Decide how many coding tables to use ---*/
|
---|
| 280 | AssertH(s->nMTF > 0, 3001);
|
---|
| 281 | if (s->nMTF < 200) nGroups = 2; else
|
---|
| 282 | if (s->nMTF < 600) nGroups = 3; else
|
---|
| 283 | if (s->nMTF < 1200) nGroups = 4; else
|
---|
| 284 | if (s->nMTF < 2400) nGroups = 5; else
|
---|
| 285 | nGroups = 6;
|
---|
| 286 |
|
---|
| 287 | /*--- Generate an initial set of coding tables ---*/
|
---|
| 288 | {
|
---|
| 289 | int32_t nPart, remF, tFreq, aFreq;
|
---|
| 290 |
|
---|
| 291 | nPart = nGroups;
|
---|
| 292 | remF = s->nMTF;
|
---|
| 293 | gs = 0;
|
---|
| 294 | while (nPart > 0) {
|
---|
| 295 | tFreq = remF / nPart;
|
---|
| 296 | ge = gs - 1;
|
---|
| 297 | aFreq = 0;
|
---|
| 298 | while (aFreq < tFreq && ge < alphaSize-1) {
|
---|
| 299 | ge++;
|
---|
| 300 | aFreq += s->mtfFreq[ge];
|
---|
| 301 | }
|
---|
| 302 |
|
---|
| 303 | if (ge > gs
|
---|
| 304 | && nPart != nGroups && nPart != 1
|
---|
| 305 | && ((nGroups - nPart) % 2 == 1) /* bbox: can this be replaced by x & 1? */
|
---|
| 306 | ) {
|
---|
| 307 | aFreq -= s->mtfFreq[ge];
|
---|
| 308 | ge--;
|
---|
| 309 | }
|
---|
| 310 |
|
---|
| 311 | for (v = 0; v < alphaSize; v++)
|
---|
| 312 | if (v >= gs && v <= ge)
|
---|
| 313 | s->len[nPart-1][v] = BZ_LESSER_ICOST;
|
---|
| 314 | else
|
---|
| 315 | s->len[nPart-1][v] = BZ_GREATER_ICOST;
|
---|
| 316 |
|
---|
| 317 | nPart--;
|
---|
| 318 | gs = ge + 1;
|
---|
| 319 | remF -= aFreq;
|
---|
| 320 | }
|
---|
| 321 | }
|
---|
| 322 |
|
---|
| 323 | /*
|
---|
| 324 | * Iterate up to BZ_N_ITERS times to improve the tables.
|
---|
| 325 | */
|
---|
| 326 | for (iter = 0; iter < BZ_N_ITERS; iter++) {
|
---|
| 327 | for (t = 0; t < nGroups; t++)
|
---|
| 328 | fave[t] = 0;
|
---|
| 329 |
|
---|
| 330 | for (t = 0; t < nGroups; t++)
|
---|
| 331 | for (v = 0; v < alphaSize; v++)
|
---|
| 332 | s->rfreq[t][v] = 0;
|
---|
| 333 |
|
---|
[3232] | 334 | #if CONFIG_BZIP2_FAST >= 5
|
---|
[2725] | 335 | /*
|
---|
| 336 | * Set up an auxiliary length table which is used to fast-track
|
---|
| 337 | * the common case (nGroups == 6).
|
---|
| 338 | */
|
---|
| 339 | if (nGroups == 6) {
|
---|
| 340 | for (v = 0; v < alphaSize; v++) {
|
---|
| 341 | s->len_pack[v][0] = (s->len[1][v] << 16) | s->len[0][v];
|
---|
| 342 | s->len_pack[v][1] = (s->len[3][v] << 16) | s->len[2][v];
|
---|
| 343 | s->len_pack[v][2] = (s->len[5][v] << 16) | s->len[4][v];
|
---|
| 344 | }
|
---|
| 345 | }
|
---|
| 346 | #endif
|
---|
| 347 | nSelectors = 0;
|
---|
| 348 | totc = 0;
|
---|
| 349 | gs = 0;
|
---|
| 350 | while (1) {
|
---|
| 351 | /*--- Set group start & end marks. --*/
|
---|
| 352 | if (gs >= s->nMTF)
|
---|
| 353 | break;
|
---|
| 354 | ge = gs + BZ_G_SIZE - 1;
|
---|
| 355 | if (ge >= s->nMTF)
|
---|
| 356 | ge = s->nMTF-1;
|
---|
| 357 |
|
---|
| 358 | /*
|
---|
| 359 | * Calculate the cost of this group as coded
|
---|
| 360 | * by each of the coding tables.
|
---|
| 361 | */
|
---|
| 362 | for (t = 0; t < nGroups; t++)
|
---|
| 363 | cost[t] = 0;
|
---|
[3232] | 364 | #if CONFIG_BZIP2_FAST >= 5
|
---|
[2725] | 365 | if (nGroups == 6 && 50 == ge-gs+1) {
|
---|
| 366 | /*--- fast track the common case ---*/
|
---|
| 367 | register uint32_t cost01, cost23, cost45;
|
---|
| 368 | register uint16_t icv;
|
---|
| 369 | cost01 = cost23 = cost45 = 0;
|
---|
| 370 | #define BZ_ITER(nn) \
|
---|
| 371 | icv = mtfv[gs+(nn)]; \
|
---|
| 372 | cost01 += s->len_pack[icv][0]; \
|
---|
| 373 | cost23 += s->len_pack[icv][1]; \
|
---|
| 374 | cost45 += s->len_pack[icv][2];
|
---|
| 375 | BZ_ITER(0); BZ_ITER(1); BZ_ITER(2); BZ_ITER(3); BZ_ITER(4);
|
---|
| 376 | BZ_ITER(5); BZ_ITER(6); BZ_ITER(7); BZ_ITER(8); BZ_ITER(9);
|
---|
| 377 | BZ_ITER(10); BZ_ITER(11); BZ_ITER(12); BZ_ITER(13); BZ_ITER(14);
|
---|
| 378 | BZ_ITER(15); BZ_ITER(16); BZ_ITER(17); BZ_ITER(18); BZ_ITER(19);
|
---|
| 379 | BZ_ITER(20); BZ_ITER(21); BZ_ITER(22); BZ_ITER(23); BZ_ITER(24);
|
---|
| 380 | BZ_ITER(25); BZ_ITER(26); BZ_ITER(27); BZ_ITER(28); BZ_ITER(29);
|
---|
| 381 | BZ_ITER(30); BZ_ITER(31); BZ_ITER(32); BZ_ITER(33); BZ_ITER(34);
|
---|
| 382 | BZ_ITER(35); BZ_ITER(36); BZ_ITER(37); BZ_ITER(38); BZ_ITER(39);
|
---|
| 383 | BZ_ITER(40); BZ_ITER(41); BZ_ITER(42); BZ_ITER(43); BZ_ITER(44);
|
---|
| 384 | BZ_ITER(45); BZ_ITER(46); BZ_ITER(47); BZ_ITER(48); BZ_ITER(49);
|
---|
| 385 | #undef BZ_ITER
|
---|
| 386 | cost[0] = cost01 & 0xffff; cost[1] = cost01 >> 16;
|
---|
| 387 | cost[2] = cost23 & 0xffff; cost[3] = cost23 >> 16;
|
---|
| 388 | cost[4] = cost45 & 0xffff; cost[5] = cost45 >> 16;
|
---|
| 389 |
|
---|
| 390 | } else
|
---|
| 391 | #endif
|
---|
| 392 | {
|
---|
| 393 | /*--- slow version which correctly handles all situations ---*/
|
---|
| 394 | for (i = gs; i <= ge; i++) {
|
---|
| 395 | uint16_t icv = mtfv[i];
|
---|
| 396 | for (t = 0; t < nGroups; t++)
|
---|
| 397 | cost[t] += s->len[t][icv];
|
---|
| 398 | }
|
---|
| 399 | }
|
---|
| 400 | /*
|
---|
| 401 | * Find the coding table which is best for this group,
|
---|
| 402 | * and record its identity in the selector table.
|
---|
| 403 | */
|
---|
| 404 | /*bc = 999999999;*/
|
---|
| 405 | /*bt = -1;*/
|
---|
| 406 | bc = cost[0];
|
---|
| 407 | bt = 0;
|
---|
| 408 | for (t = 1 /*0*/; t < nGroups; t++) {
|
---|
| 409 | if (cost[t] < bc) {
|
---|
| 410 | bc = cost[t];
|
---|
| 411 | bt = t;
|
---|
| 412 | }
|
---|
| 413 | }
|
---|
| 414 | totc += bc;
|
---|
| 415 | fave[bt]++;
|
---|
| 416 | s->selector[nSelectors] = bt;
|
---|
| 417 | nSelectors++;
|
---|
| 418 |
|
---|
| 419 | /*
|
---|
| 420 | * Increment the symbol frequencies for the selected table.
|
---|
| 421 | */
|
---|
| 422 | /* 1% faster compress. +800 bytes */
|
---|
[3232] | 423 | #if CONFIG_BZIP2_FAST >= 4
|
---|
[2725] | 424 | if (nGroups == 6 && 50 == ge-gs+1) {
|
---|
| 425 | /*--- fast track the common case ---*/
|
---|
| 426 | #define BZ_ITUR(nn) s->rfreq[bt][mtfv[gs + (nn)]]++
|
---|
| 427 | BZ_ITUR(0); BZ_ITUR(1); BZ_ITUR(2); BZ_ITUR(3); BZ_ITUR(4);
|
---|
| 428 | BZ_ITUR(5); BZ_ITUR(6); BZ_ITUR(7); BZ_ITUR(8); BZ_ITUR(9);
|
---|
| 429 | BZ_ITUR(10); BZ_ITUR(11); BZ_ITUR(12); BZ_ITUR(13); BZ_ITUR(14);
|
---|
| 430 | BZ_ITUR(15); BZ_ITUR(16); BZ_ITUR(17); BZ_ITUR(18); BZ_ITUR(19);
|
---|
| 431 | BZ_ITUR(20); BZ_ITUR(21); BZ_ITUR(22); BZ_ITUR(23); BZ_ITUR(24);
|
---|
| 432 | BZ_ITUR(25); BZ_ITUR(26); BZ_ITUR(27); BZ_ITUR(28); BZ_ITUR(29);
|
---|
| 433 | BZ_ITUR(30); BZ_ITUR(31); BZ_ITUR(32); BZ_ITUR(33); BZ_ITUR(34);
|
---|
| 434 | BZ_ITUR(35); BZ_ITUR(36); BZ_ITUR(37); BZ_ITUR(38); BZ_ITUR(39);
|
---|
| 435 | BZ_ITUR(40); BZ_ITUR(41); BZ_ITUR(42); BZ_ITUR(43); BZ_ITUR(44);
|
---|
| 436 | BZ_ITUR(45); BZ_ITUR(46); BZ_ITUR(47); BZ_ITUR(48); BZ_ITUR(49);
|
---|
| 437 | #undef BZ_ITUR
|
---|
| 438 | gs = ge + 1;
|
---|
| 439 | } else
|
---|
| 440 | #endif
|
---|
| 441 | {
|
---|
| 442 | /*--- slow version which correctly handles all situations ---*/
|
---|
| 443 | while (gs <= ge) {
|
---|
| 444 | s->rfreq[bt][mtfv[gs]]++;
|
---|
| 445 | gs++;
|
---|
| 446 | }
|
---|
| 447 | /* already is: gs = ge + 1; */
|
---|
| 448 | }
|
---|
| 449 | }
|
---|
| 450 |
|
---|
| 451 | /*
|
---|
| 452 | * Recompute the tables based on the accumulated frequencies.
|
---|
| 453 | */
|
---|
| 454 | /* maxLen was changed from 20 to 17 in bzip2-1.0.3. See
|
---|
| 455 | * comment in huffman.c for details. */
|
---|
| 456 | for (t = 0; t < nGroups; t++)
|
---|
| 457 | BZ2_hbMakeCodeLengths(s, &(s->len[t][0]), &(s->rfreq[t][0]), alphaSize, 17 /*20*/);
|
---|
| 458 | }
|
---|
| 459 |
|
---|
| 460 | AssertH(nGroups < 8, 3002);
|
---|
| 461 | AssertH(nSelectors < 32768 && nSelectors <= (2 + (900000 / BZ_G_SIZE)), 3003);
|
---|
| 462 |
|
---|
| 463 | /*--- Compute MTF values for the selectors. ---*/
|
---|
| 464 | {
|
---|
| 465 | uint8_t pos[BZ_N_GROUPS], ll_i, tmp2, tmp;
|
---|
| 466 |
|
---|
| 467 | for (i = 0; i < nGroups; i++)
|
---|
| 468 | pos[i] = i;
|
---|
| 469 | for (i = 0; i < nSelectors; i++) {
|
---|
| 470 | ll_i = s->selector[i];
|
---|
| 471 | j = 0;
|
---|
| 472 | tmp = pos[j];
|
---|
| 473 | while (ll_i != tmp) {
|
---|
| 474 | j++;
|
---|
| 475 | tmp2 = tmp;
|
---|
| 476 | tmp = pos[j];
|
---|
| 477 | pos[j] = tmp2;
|
---|
| 478 | };
|
---|
| 479 | pos[0] = tmp;
|
---|
| 480 | s->selectorMtf[i] = j;
|
---|
| 481 | }
|
---|
| 482 | };
|
---|
| 483 |
|
---|
| 484 | /*--- Assign actual codes for the tables. --*/
|
---|
| 485 | for (t = 0; t < nGroups; t++) {
|
---|
| 486 | minLen = 32;
|
---|
| 487 | maxLen = 0;
|
---|
| 488 | for (i = 0; i < alphaSize; i++) {
|
---|
| 489 | if (s->len[t][i] > maxLen) maxLen = s->len[t][i];
|
---|
| 490 | if (s->len[t][i] < minLen) minLen = s->len[t][i];
|
---|
| 491 | }
|
---|
| 492 | AssertH(!(maxLen > 17 /*20*/), 3004);
|
---|
| 493 | AssertH(!(minLen < 1), 3005);
|
---|
| 494 | BZ2_hbAssignCodes(&(s->code[t][0]), &(s->len[t][0]), minLen, maxLen, alphaSize);
|
---|
| 495 | }
|
---|
| 496 |
|
---|
| 497 | /*--- Transmit the mapping table. ---*/
|
---|
| 498 | {
|
---|
| 499 | /* bbox: optimized a bit more than in bzip2 */
|
---|
| 500 | int inUse16 = 0;
|
---|
| 501 | for (i = 0; i < 16; i++) {
|
---|
| 502 | if (sizeof(long) <= 4) {
|
---|
| 503 | inUse16 = inUse16*2 +
|
---|
| 504 | ((*(uint32_t*)&(s->inUse[i * 16 + 0])
|
---|
| 505 | | *(uint32_t*)&(s->inUse[i * 16 + 4])
|
---|
| 506 | | *(uint32_t*)&(s->inUse[i * 16 + 8])
|
---|
| 507 | | *(uint32_t*)&(s->inUse[i * 16 + 12])) != 0);
|
---|
| 508 | } else { /* Our CPU can do better */
|
---|
| 509 | inUse16 = inUse16*2 +
|
---|
| 510 | ((*(uint64_t*)&(s->inUse[i * 16 + 0])
|
---|
| 511 | | *(uint64_t*)&(s->inUse[i * 16 + 8])) != 0);
|
---|
| 512 | }
|
---|
| 513 | }
|
---|
| 514 |
|
---|
| 515 | bsW(s, 16, inUse16);
|
---|
| 516 |
|
---|
| 517 | inUse16 <<= (sizeof(int)*8 - 16); /* move 15th bit into sign bit */
|
---|
| 518 | for (i = 0; i < 16; i++) {
|
---|
| 519 | if (inUse16 < 0) {
|
---|
| 520 | unsigned v16 = 0;
|
---|
| 521 | for (j = 0; j < 16; j++)
|
---|
| 522 | v16 = v16*2 + s->inUse[i * 16 + j];
|
---|
| 523 | bsW(s, 16, v16);
|
---|
| 524 | }
|
---|
| 525 | inUse16 <<= 1;
|
---|
| 526 | }
|
---|
| 527 | }
|
---|
| 528 |
|
---|
| 529 | /*--- Now the selectors. ---*/
|
---|
| 530 | bsW(s, 3, nGroups);
|
---|
| 531 | bsW(s, 15, nSelectors);
|
---|
| 532 | for (i = 0; i < nSelectors; i++) {
|
---|
| 533 | for (j = 0; j < s->selectorMtf[i]; j++)
|
---|
| 534 | bsW(s, 1, 1);
|
---|
| 535 | bsW(s, 1, 0);
|
---|
| 536 | }
|
---|
| 537 |
|
---|
| 538 | /*--- Now the coding tables. ---*/
|
---|
| 539 | for (t = 0; t < nGroups; t++) {
|
---|
| 540 | int32_t curr = s->len[t][0];
|
---|
| 541 | bsW(s, 5, curr);
|
---|
| 542 | for (i = 0; i < alphaSize; i++) {
|
---|
| 543 | while (curr < s->len[t][i]) { bsW(s, 2, 2); curr++; /* 10 */ };
|
---|
| 544 | while (curr > s->len[t][i]) { bsW(s, 2, 3); curr--; /* 11 */ };
|
---|
| 545 | bsW(s, 1, 0);
|
---|
| 546 | }
|
---|
| 547 | }
|
---|
| 548 |
|
---|
| 549 | /*--- And finally, the block data proper ---*/
|
---|
| 550 | selCtr = 0;
|
---|
| 551 | gs = 0;
|
---|
| 552 | while (1) {
|
---|
| 553 | if (gs >= s->nMTF)
|
---|
| 554 | break;
|
---|
| 555 | ge = gs + BZ_G_SIZE - 1;
|
---|
| 556 | if (ge >= s->nMTF)
|
---|
| 557 | ge = s->nMTF-1;
|
---|
| 558 | AssertH(s->selector[selCtr] < nGroups, 3006);
|
---|
| 559 |
|
---|
| 560 | /* Costs 1300 bytes and is _slower_ (on Intel Core 2) */
|
---|
| 561 | #if 0
|
---|
| 562 | if (nGroups == 6 && 50 == ge-gs+1) {
|
---|
| 563 | /*--- fast track the common case ---*/
|
---|
| 564 | uint16_t mtfv_i;
|
---|
| 565 | uint8_t* s_len_sel_selCtr = &(s->len[s->selector[selCtr]][0]);
|
---|
| 566 | int32_t* s_code_sel_selCtr = &(s->code[s->selector[selCtr]][0]);
|
---|
| 567 | #define BZ_ITAH(nn) \
|
---|
| 568 | mtfv_i = mtfv[gs+(nn)]; \
|
---|
| 569 | bsW(s, s_len_sel_selCtr[mtfv_i], s_code_sel_selCtr[mtfv_i])
|
---|
| 570 | BZ_ITAH(0); BZ_ITAH(1); BZ_ITAH(2); BZ_ITAH(3); BZ_ITAH(4);
|
---|
| 571 | BZ_ITAH(5); BZ_ITAH(6); BZ_ITAH(7); BZ_ITAH(8); BZ_ITAH(9);
|
---|
| 572 | BZ_ITAH(10); BZ_ITAH(11); BZ_ITAH(12); BZ_ITAH(13); BZ_ITAH(14);
|
---|
| 573 | BZ_ITAH(15); BZ_ITAH(16); BZ_ITAH(17); BZ_ITAH(18); BZ_ITAH(19);
|
---|
| 574 | BZ_ITAH(20); BZ_ITAH(21); BZ_ITAH(22); BZ_ITAH(23); BZ_ITAH(24);
|
---|
| 575 | BZ_ITAH(25); BZ_ITAH(26); BZ_ITAH(27); BZ_ITAH(28); BZ_ITAH(29);
|
---|
| 576 | BZ_ITAH(30); BZ_ITAH(31); BZ_ITAH(32); BZ_ITAH(33); BZ_ITAH(34);
|
---|
| 577 | BZ_ITAH(35); BZ_ITAH(36); BZ_ITAH(37); BZ_ITAH(38); BZ_ITAH(39);
|
---|
| 578 | BZ_ITAH(40); BZ_ITAH(41); BZ_ITAH(42); BZ_ITAH(43); BZ_ITAH(44);
|
---|
| 579 | BZ_ITAH(45); BZ_ITAH(46); BZ_ITAH(47); BZ_ITAH(48); BZ_ITAH(49);
|
---|
| 580 | #undef BZ_ITAH
|
---|
| 581 | gs = ge+1;
|
---|
| 582 | } else
|
---|
| 583 | #endif
|
---|
| 584 | {
|
---|
| 585 | /*--- slow version which correctly handles all situations ---*/
|
---|
| 586 | /* code is bit bigger, but moves multiply out of the loop */
|
---|
| 587 | uint8_t* s_len_sel_selCtr = &(s->len [s->selector[selCtr]][0]);
|
---|
| 588 | int32_t* s_code_sel_selCtr = &(s->code[s->selector[selCtr]][0]);
|
---|
| 589 | while (gs <= ge) {
|
---|
| 590 | bsW(s,
|
---|
| 591 | s_len_sel_selCtr[mtfv[gs]],
|
---|
| 592 | s_code_sel_selCtr[mtfv[gs]]
|
---|
| 593 | );
|
---|
| 594 | gs++;
|
---|
| 595 | }
|
---|
| 596 | /* already is: gs = ge+1; */
|
---|
| 597 | }
|
---|
| 598 | selCtr++;
|
---|
| 599 | }
|
---|
| 600 | AssertH(selCtr == nSelectors, 3007);
|
---|
| 601 | #undef code
|
---|
| 602 | #undef rfreq
|
---|
| 603 | #undef len_pack
|
---|
| 604 | }
|
---|
| 605 |
|
---|
| 606 |
|
---|
| 607 | /*---------------------------------------------------*/
|
---|
| 608 | static
|
---|
| 609 | void BZ2_compressBlock(EState* s, int is_last_block)
|
---|
| 610 | {
|
---|
| 611 | if (s->nblock > 0) {
|
---|
| 612 | BZ_FINALISE_CRC(s->blockCRC);
|
---|
| 613 | s->combinedCRC = (s->combinedCRC << 1) | (s->combinedCRC >> 31);
|
---|
| 614 | s->combinedCRC ^= s->blockCRC;
|
---|
| 615 | if (s->blockNo > 1)
|
---|
| 616 | s->numZ = 0;
|
---|
| 617 |
|
---|
| 618 | BZ2_blockSort(s);
|
---|
| 619 | }
|
---|
| 620 |
|
---|
| 621 | s->zbits = &((uint8_t*)s->arr2)[s->nblock];
|
---|
| 622 |
|
---|
| 623 | /*-- If this is the first block, create the stream header. --*/
|
---|
| 624 | if (s->blockNo == 1) {
|
---|
| 625 | BZ2_bsInitWrite(s);
|
---|
| 626 | /*bsPutU8(s, BZ_HDR_B);*/
|
---|
| 627 | /*bsPutU8(s, BZ_HDR_Z);*/
|
---|
| 628 | /*bsPutU8(s, BZ_HDR_h);*/
|
---|
| 629 | /*bsPutU8(s, BZ_HDR_0 + s->blockSize100k);*/
|
---|
| 630 | bsPutU32(s, BZ_HDR_BZh0 + s->blockSize100k);
|
---|
| 631 | }
|
---|
| 632 |
|
---|
| 633 | if (s->nblock > 0) {
|
---|
| 634 | /*bsPutU8(s, 0x31);*/
|
---|
| 635 | /*bsPutU8(s, 0x41);*/
|
---|
| 636 | /*bsPutU8(s, 0x59);*/
|
---|
| 637 | /*bsPutU8(s, 0x26);*/
|
---|
| 638 | bsPutU32(s, 0x31415926);
|
---|
| 639 | /*bsPutU8(s, 0x53);*/
|
---|
| 640 | /*bsPutU8(s, 0x59);*/
|
---|
| 641 | bsPutU16(s, 0x5359);
|
---|
| 642 |
|
---|
| 643 | /*-- Now the block's CRC, so it is in a known place. --*/
|
---|
| 644 | bsPutU32(s, s->blockCRC);
|
---|
| 645 |
|
---|
| 646 | /*
|
---|
| 647 | * Now a single bit indicating (non-)randomisation.
|
---|
| 648 | * As of version 0.9.5, we use a better sorting algorithm
|
---|
| 649 | * which makes randomisation unnecessary. So always set
|
---|
| 650 | * the randomised bit to 'no'. Of course, the decoder
|
---|
| 651 | * still needs to be able to handle randomised blocks
|
---|
| 652 | * so as to maintain backwards compatibility with
|
---|
| 653 | * older versions of bzip2.
|
---|
| 654 | */
|
---|
| 655 | bsW(s, 1, 0);
|
---|
| 656 |
|
---|
| 657 | bsW(s, 24, s->origPtr);
|
---|
| 658 | generateMTFValues(s);
|
---|
| 659 | sendMTFValues(s);
|
---|
| 660 | }
|
---|
| 661 |
|
---|
| 662 | /*-- If this is the last block, add the stream trailer. --*/
|
---|
| 663 | if (is_last_block) {
|
---|
| 664 | /*bsPutU8(s, 0x17);*/
|
---|
| 665 | /*bsPutU8(s, 0x72);*/
|
---|
| 666 | /*bsPutU8(s, 0x45);*/
|
---|
| 667 | /*bsPutU8(s, 0x38);*/
|
---|
| 668 | bsPutU32(s, 0x17724538);
|
---|
| 669 | /*bsPutU8(s, 0x50);*/
|
---|
| 670 | /*bsPutU8(s, 0x90);*/
|
---|
| 671 | bsPutU16(s, 0x5090);
|
---|
| 672 | bsPutU32(s, s->combinedCRC);
|
---|
| 673 | bsFinishWrite(s);
|
---|
| 674 | }
|
---|
| 675 | }
|
---|
| 676 |
|
---|
| 677 |
|
---|
| 678 | /*-------------------------------------------------------------*/
|
---|
| 679 | /*--- end compress.c ---*/
|
---|
| 680 | /*-------------------------------------------------------------*/
|
---|