1 /* $OpenBSD: zopen.c,v 1.7 2001/11/19 19:02:13 mpech Exp $ */ 2 /* $NetBSD: zopen.c,v 1.5 1995/03/26 09:44:53 glass Exp $ */ 3 4 /*- 5 * Copyright (c) 1985, 1986, 1992, 1993 6 * The Regents of the University of California. All rights reserved. 7 * 8 * This code is derived from software contributed to Berkeley by 9 * Diomidis Spinellis and James A. Woods, derived from original 10 * work by Spencer Thomas and Joseph Orost. 11 * 12 * Redistribution and use in source and binary forms, with or without 13 * modification, are permitted provided that the following conditions 14 * are met: 15 * 1. Redistributions of source code must retain the above copyright 16 * notice, this list of conditions and the following disclaimer. 17 * 2. Redistributions in binary form must reproduce the above copyright 18 * notice, this list of conditions and the following disclaimer in the 19 * documentation and/or other materials provided with the distribution. 20 * 3. All advertising materials mentioning features or use of this software 21 * must display the following acknowledgement: 22 * This product includes software developed by the University of 23 * California, Berkeley and its contributors. 24 * 4. Neither the name of the University nor the names of its contributors 25 * may be used to endorse or promote products derived from this software 26 * without specific prior written permission. 27 * 28 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 29 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 30 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 31 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 32 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 33 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 34 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 35 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 37 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 38 * SUCH DAMAGE. 39 */ 40 41 #if defined(LIBC_SCCS) && !defined(lint) 42 #if 0 43 static char sccsid[] = "@(#)zopen.c 8.1 (Berkeley) 6/27/93"; 44 #else 45 static char rcsid[] = "$OpenBSD: zopen.c,v 1.7 2001/11/19 19:02:13 mpech Exp $"; 46 #endif 47 #endif /* LIBC_SCCS and not lint */ 48 49 /*- 50 * fcompress.c - File compression ala IEEE Computer, June 1984. 51 * 52 * Compress authors: 53 * Spencer W. Thomas (decvax!utah-cs!thomas) 54 * Jim McKie (decvax!mcvax!jim) 55 * Steve Davies (decvax!vax135!petsd!peora!srd) 56 * Ken Turkowski (decvax!decwrl!turtlevax!ken) 57 * James A. Woods (decvax!ihnp4!ames!jaw) 58 * Joe Orost (decvax!vax135!petsd!joe) 59 * 60 * Cleaned up and converted to library returning I/O streams by 61 * Diomidis Spinellis <dds@doc.ic.ac.uk>. 62 * 63 * zopen(filename, mode, bits) 64 * Returns a FILE * that can be used for read or write. The modes 65 * supported are only "r" and "w". Seeking is not allowed. On 66 * reading the file is decompressed, on writing it is compressed. 67 * The output is compatible with compress(1) with 16 bit tables. 68 * Any file produced by compress(1) can be read. 69 */ 70 71 #include <sys/param.h> 72 #include <sys/stat.h> 73 74 #include <ctype.h> 75 #include <errno.h> 76 #include <signal.h> 77 #include <stdio.h> 78 #include <stdlib.h> 79 #include <string.h> 80 #include <unistd.h> 81 #include <fcntl.h> 82 #include "compress.h" 83 84 #define BITS 16 /* Default bits. */ 85 #define HSIZE 69001 /* 95% occupancy */ 86 #define ZBUFSIZ 8192 /* I/O buffer size */ 87 88 /* A code_int must be able to hold 2**BITS values of type int, and also -1. */ 89 typedef long code_int; 90 typedef long count_int; 91 92 static u_char z_magic[] = 93 {'\037', '\235'}; /* 1F 9D */ 94 95 #define BIT_MASK 0x1f /* Defines for third byte of header. */ 96 #define BLOCK_MASK 0x80 97 98 /* 99 * Masks 0x40 and 0x20 are free. I think 0x20 should mean that there is 100 * a fourth header byte (for expansion). 101 */ 102 #define INIT_BITS 9 /* Initial number of bits/code. */ 103 104 #define MAXCODE(n_bits) ((1 << (n_bits)) - 1) 105 106 struct s_zstate { 107 int zs_fd; /* File stream for I/O */ 108 char zs_mode; /* r or w */ 109 enum { 110 S_START, S_MIDDLE, S_EOF 111 } zs_state; /* State of computation */ 112 int zs_n_bits; /* Number of bits/code. */ 113 int zs_maxbits; /* User settable max # bits/code. */ 114 code_int zs_maxcode; /* Maximum code, given n_bits. */ 115 code_int zs_maxmaxcode; /* Should NEVER generate this code. */ 116 count_int zs_htab [HSIZE]; 117 u_short zs_codetab [HSIZE]; 118 code_int zs_hsize; /* For dynamic table sizing. */ 119 code_int zs_free_ent; /* First unused entry. */ 120 /* 121 * Block compression parameters -- after all codes are used up, 122 * and compression rate changes, start over. 123 */ 124 int zs_block_compress; 125 int zs_clear_flg; 126 long zs_ratio; 127 count_int zs_checkpoint; 128 long zs_in_count; /* Length of input. */ 129 long zs_bytes_out; /* Length of compressed output. */ 130 long zs_out_count; /* # of codes output (for debugging).*/ 131 u_char zs_buf[ZBUFSIZ]; /* I/O buffer */ 132 u_char *zs_bp; /* Current I/O window in the zs_buf */ 133 int zs_offset; /* Number of bits in the zs_buf */ 134 union { 135 struct { 136 long zs_fcode; 137 code_int zs_ent; 138 code_int zs_hsize_reg; 139 int zs_hshift; 140 } w; /* Write paramenters */ 141 struct { 142 u_char *zs_stackp, *zs_ebp; 143 int zs_finchar; 144 code_int zs_code, zs_oldcode, zs_incode; 145 int zs_size; 146 } r; /* Read parameters */ 147 } u; 148 }; 149 150 /* Definitions to retain old variable names */ 151 #define zs_fcode u.w.zs_fcode 152 #define zs_ent u.w.zs_ent 153 #define zs_hsize_reg u.w.zs_hsize_reg 154 #define zs_hshift u.w.zs_hshift 155 #define zs_stackp u.r.zs_stackp 156 #define zs_finchar u.r.zs_finchar 157 #define zs_code u.r.zs_code 158 #define zs_oldcode u.r.zs_oldcode 159 #define zs_incode u.r.zs_incode 160 #define zs_size u.r.zs_size 161 #define zs_ebp u.r.zs_ebp 162 163 /* 164 * To save much memory, we overlay the table used by compress() with those 165 * used by decompress(). The tab_prefix table is the same size and type as 166 * the codetab. The tab_suffix table needs 2**BITS characters. We get this 167 * from the beginning of htab. The output stack uses the rest of htab, and 168 * contains characters. There is plenty of room for any possible stack 169 * (stack used to be 8000 characters). 170 */ 171 172 #define htabof(i) zs->zs_htab[i] 173 #define codetabof(i) zs->zs_codetab[i] 174 175 #define tab_prefixof(i) codetabof(i) 176 #define tab_suffixof(i) ((u_char *)(zs->zs_htab))[i] 177 #define de_stack ((u_char *)&tab_suffixof(1 << BITS)) 178 179 #define CHECK_GAP 10000 /* Ratio check interval. */ 180 181 /* 182 * the next two codes should not be changed lightly, as they must not 183 * lie within the contiguous general code space. 184 */ 185 #define FIRST 257 /* First free entry. */ 186 #define CLEAR 256 /* Table clear output code. */ 187 188 static int cl_block __P((struct s_zstate *)); 189 static void cl_hash __P((struct s_zstate *, register count_int)); 190 static code_int getcode __P((struct s_zstate *)); 191 static int output __P((struct s_zstate *, code_int)); 192 193 /*- 194 * Algorithm from "A Technique for High Performance Data Compression", 195 * Terry A. Welch, IEEE Computer Vol 17, No 6 (June 1984), pp 8-19. 196 * 197 * Algorithm: 198 * Modified Lempel-Ziv method (LZW). Basically finds common 199 * substrings and replaces them with a variable size code. This is 200 * deterministic, and can be done on the fly. Thus, the decompression 201 * procedure needs no input table, but tracks the way the table was built. 202 */ 203 204 /*- 205 * compress write 206 * 207 * Algorithm: use open addressing double hashing (no chaining) on the 208 * prefix code / next character combination. We do a variant of Knuth's 209 * algorithm D (vol. 3, sec. 6.4) along with G. Knott's relatively-prime 210 * secondary probe. Here, the modular division first probe is gives way 211 * to a faster exclusive-or manipulation. Also do block compression with 212 * an adaptive reset, whereby the code table is cleared when the compression 213 * ratio decreases, but after the table fills. The variable-length output 214 * codes are re-sized at this point, and a special CLEAR code is generated 215 * for the decompressor. Late addition: construct the table according to 216 * file size for noticeable speed improvement on small files. Please direct 217 * questions about this implementation to ames!jaw. 218 */ 219 int 220 zwrite(cookie, wbp, num) 221 void *cookie; 222 const char *wbp; 223 int num; 224 { 225 code_int i; 226 int c, disp; 227 struct s_zstate *zs; 228 const u_char *bp; 229 u_char tmp; 230 int count; 231 232 zs = cookie; 233 count = num; 234 bp = (u_char *)wbp; 235 switch (zs->zs_state) { 236 case S_EOF: 237 return 0; 238 case S_START: 239 zs->zs_state = S_MIDDLE; 240 241 zs->zs_maxmaxcode = 1L << zs->zs_maxbits; 242 if (write(zs->zs_fd, z_magic, sizeof(z_magic)) != 243 sizeof(z_magic)) 244 return (-1); 245 tmp = (u_char)(zs->zs_maxbits | zs->zs_block_compress); 246 if (write(zs->zs_fd, &tmp, sizeof(tmp)) != sizeof(tmp)) 247 return (-1); 248 249 zs->zs_bp = zs->zs_buf; 250 zs->zs_offset = 0; 251 zs->zs_bytes_out = 3; /* Includes 3-byte header mojo. */ 252 zs->zs_out_count = 0; 253 zs->zs_clear_flg = 0; 254 zs->zs_ratio = 0; 255 zs->zs_in_count = 1; 256 zs->zs_checkpoint = CHECK_GAP; 257 zs->zs_maxcode = MAXCODE(zs->zs_n_bits = INIT_BITS); 258 zs->zs_free_ent = ((zs->zs_block_compress) ? FIRST : 256); 259 260 zs->zs_ent = *bp++; 261 --count; 262 263 zs->zs_hshift = 0; 264 for (zs->zs_fcode = (long)zs->zs_hsize; zs->zs_fcode < 65536L; 265 zs->zs_fcode *= 2L) 266 zs->zs_hshift++; 267 /* Set hash code range bound. */ 268 zs->zs_hshift = 8 - zs->zs_hshift; 269 270 zs->zs_hsize_reg = zs->zs_hsize; 271 /* Clear hash table. */ 272 cl_hash(zs, (count_int)zs->zs_hsize_reg); 273 274 case S_MIDDLE: 275 for (i = 0; count-- > 0;) { 276 c = *bp++; 277 zs->zs_in_count++; 278 zs->zs_fcode = (long)(((long)c << zs->zs_maxbits) + 279 zs->zs_ent); 280 /* Xor hashing. */ 281 i = ((c << zs->zs_hshift) ^ zs->zs_ent); 282 283 if (htabof(i) == zs->zs_fcode) { 284 zs->zs_ent = codetabof(i); 285 continue; 286 } else if ((long)htabof(i) < 0) /* Empty slot. */ 287 goto nomatch; 288 /* Secondary hash (after G. Knott). */ 289 disp = zs->zs_hsize_reg - i; 290 if (i == 0) 291 disp = 1; 292 probe: if ((i -= disp) < 0) 293 i += zs->zs_hsize_reg; 294 295 if (htabof(i) == zs->zs_fcode) { 296 zs->zs_ent = codetabof(i); 297 continue; 298 } 299 if ((long)htabof(i) >= 0) 300 goto probe; 301 nomatch: if (output(zs, (code_int) zs->zs_ent) == -1) 302 return (-1); 303 zs->zs_out_count++; 304 zs->zs_ent = c; 305 if (zs->zs_free_ent < zs->zs_maxmaxcode) { 306 /* code -> hashtable */ 307 codetabof(i) = zs->zs_free_ent++; 308 htabof(i) = zs->zs_fcode; 309 } else if ((count_int)zs->zs_in_count >= 310 zs->zs_checkpoint && zs->zs_block_compress) { 311 if (cl_block(zs) == -1) 312 return (-1); 313 } 314 } 315 } 316 return (num); 317 } 318 319 int 320 zclose(cookie) 321 void *cookie; 322 { 323 struct s_zstate *zs; 324 int rval; 325 326 zs = cookie; 327 if (zs->zs_mode == 'w') { /* Put out the final code. */ 328 if (output(zs, (code_int) zs->zs_ent) == -1) { 329 (void)close(zs->zs_fd); 330 free(zs); 331 return (-1); 332 } 333 zs->zs_out_count++; 334 if (output(zs, (code_int) - 1) == -1) { 335 (void)close(zs->zs_fd); 336 free(zs); 337 return (-1); 338 } 339 } 340 rval = close(zs->zs_fd); 341 free(zs); 342 return (rval); 343 } 344 345 /*- 346 * Output the given code. 347 * Inputs: 348 * code: A n_bits-bit integer. If == -1, then EOF. This assumes 349 * that n_bits =< (long)wordsize - 1. 350 * Outputs: 351 * Outputs code to the file. 352 * Assumptions: 353 * Chars are 8 bits long. 354 * Algorithm: 355 * Maintain a BITS character long buffer (so that 8 codes will 356 * fit in it exactly). Use the VAX insv instruction to insert each 357 * code in turn. When the buffer fills up empty it and start over. 358 */ 359 360 static const u_char lmask[9] = 361 {0xff, 0xfe, 0xfc, 0xf8, 0xf0, 0xe0, 0xc0, 0x80, 0x00}; 362 static const u_char rmask[9] = 363 {0x00, 0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff}; 364 365 static int 366 output(zs, ocode) 367 struct s_zstate *zs; 368 code_int ocode; 369 { 370 int bits; 371 372 if (ocode >= 0) { 373 int r_off; 374 u_char *bp; 375 376 /* Get to the first byte. */ 377 bp = zs->zs_bp + (zs->zs_offset >> 3); 378 r_off = zs->zs_offset & 7; 379 bits = zs->zs_n_bits; 380 381 /* 382 * Since ocode is always >= 8 bits, only need to mask the first 383 * hunk on the left. 384 */ 385 *bp = (*bp & rmask[r_off]) | ((ocode << r_off) & lmask[r_off]); 386 bp++; 387 bits -= (8 - r_off); 388 ocode >>= 8 - r_off; 389 /* Get any 8 bit parts in the middle (<=1 for up to 16 bits) */ 390 if (bits >= 8) { 391 *bp++ = ocode; 392 ocode >>= 8; 393 bits -= 8; 394 } 395 /* Last bits. */ 396 if (bits) 397 *bp = ocode; 398 zs->zs_offset += zs->zs_n_bits; 399 if (zs->zs_offset == (zs->zs_n_bits << 3)) { 400 zs->zs_bp += zs->zs_n_bits; 401 zs->zs_offset = 0; 402 } 403 /* 404 * If the next entry is going to be too big for the ocode size, 405 * then increase it, if possible. 406 */ 407 if (zs->zs_free_ent > zs->zs_maxcode || 408 (zs->zs_clear_flg > 0)) { 409 /* 410 * Write the whole buffer, because the input side won't 411 * discover the size increase until after it has read it 412 */ 413 if (zs->zs_offset > 0) { 414 zs->zs_bp += zs->zs_n_bits; 415 zs->zs_offset = 0; 416 } 417 418 if (zs->zs_clear_flg) { 419 zs->zs_maxcode = 420 MAXCODE(zs->zs_n_bits = INIT_BITS); 421 zs->zs_clear_flg = 0; 422 } else { 423 zs->zs_n_bits++; 424 if (zs->zs_n_bits == zs->zs_maxbits) 425 zs->zs_maxcode = zs->zs_maxmaxcode; 426 else 427 zs->zs_maxcode = 428 MAXCODE(zs->zs_n_bits); 429 } 430 } 431 432 if (zs->zs_bp + zs->zs_n_bits > &zs->zs_buf[ZBUFSIZ]) { 433 bits = zs->zs_bp - zs->zs_buf; 434 if (write(zs->zs_fd, zs->zs_buf, bits) != bits) 435 return (-1); 436 zs->zs_bytes_out += bits; 437 if (zs->zs_offset > 0) 438 fprintf (stderr, "zs_offset != 0\n"); 439 zs->zs_bp = zs->zs_buf; 440 } 441 } else { 442 /* At EOF, write the rest of the buffer. */ 443 if (zs->zs_offset > 0) 444 zs->zs_bp += (zs->zs_offset + 7) / 8; 445 if (zs->zs_bp > zs->zs_buf) { 446 bits = zs->zs_bp - zs->zs_buf; 447 if (write(zs->zs_fd, zs->zs_buf, bits) != bits) 448 return (-1); 449 zs->zs_bytes_out += bits; 450 } 451 zs->zs_offset = 0; 452 zs->zs_bp = zs->zs_buf; 453 } 454 return (0); 455 } 456 457 /* 458 * Decompress read. This routine adapts to the codes in the file building 459 * the "string" table on-the-fly; requiring no table to be stored in the 460 * compressed file. The tables used herein are shared with those of the 461 * compress() routine. See the definitions above. 462 */ 463 int 464 zread(cookie, rbp, num) 465 void *cookie; 466 char *rbp; 467 int num; 468 { 469 u_int count; 470 struct s_zstate *zs; 471 u_char *bp, header[3]; 472 473 if (num == 0) 474 return (0); 475 476 zs = cookie; 477 count = num; 478 bp = (u_char *)rbp; 479 switch (zs->zs_state) { 480 case S_START: 481 zs->zs_state = S_MIDDLE; 482 zs->zs_bp = zs->zs_buf; 483 break; 484 case S_MIDDLE: 485 goto middle; 486 case S_EOF: 487 goto eof; 488 } 489 490 /* Check the magic number */ 491 if (read(zs->zs_fd, header, sizeof(header)) != sizeof(header) || 492 memcmp(header, z_magic, sizeof(z_magic)) != 0) { 493 errno = EFTYPE; 494 return (-1); 495 } 496 zs->zs_maxbits = header[2]; /* Set -b from file. */ 497 zs->zs_block_compress = zs->zs_maxbits & BLOCK_MASK; 498 zs->zs_maxbits &= BIT_MASK; 499 zs->zs_maxmaxcode = 1L << zs->zs_maxbits; 500 if (zs->zs_maxbits > BITS) { 501 errno = EFTYPE; 502 return (-1); 503 } 504 /* As above, initialize the first 256 entries in the table. */ 505 zs->zs_maxcode = MAXCODE(zs->zs_n_bits = INIT_BITS); 506 for (zs->zs_code = 255; zs->zs_code >= 0; zs->zs_code--) { 507 tab_prefixof(zs->zs_code) = 0; 508 tab_suffixof(zs->zs_code) = (u_char) zs->zs_code; 509 } 510 zs->zs_free_ent = zs->zs_block_compress ? FIRST : 256; 511 512 zs->zs_finchar = zs->zs_oldcode = getcode(zs); 513 if (zs->zs_oldcode == -1) /* EOF already? */ 514 return (0); /* Get out of here */ 515 516 /* First code must be 8 bits = char. */ 517 *bp++ = (u_char)zs->zs_finchar; 518 count--; 519 zs->zs_stackp = de_stack; 520 521 while ((zs->zs_code = getcode(zs)) > -1) { 522 523 if ((zs->zs_code == CLEAR) && zs->zs_block_compress) { 524 for (zs->zs_code = 255; zs->zs_code >= 0; 525 zs->zs_code--) 526 tab_prefixof(zs->zs_code) = 0; 527 zs->zs_clear_flg = 1; 528 zs->zs_free_ent = FIRST - 1; 529 if ((zs->zs_code = getcode(zs)) == -1) /* O, untimely death! */ 530 break; 531 } 532 zs->zs_incode = zs->zs_code; 533 534 /* Special case for KwKwK string. */ 535 if (zs->zs_code >= zs->zs_free_ent) { 536 *zs->zs_stackp++ = zs->zs_finchar; 537 zs->zs_code = zs->zs_oldcode; 538 } 539 540 /* Generate output characters in reverse order. */ 541 while (zs->zs_code >= 256) { 542 *zs->zs_stackp++ = tab_suffixof(zs->zs_code); 543 zs->zs_code = tab_prefixof(zs->zs_code); 544 } 545 *zs->zs_stackp++ = zs->zs_finchar = tab_suffixof(zs->zs_code); 546 547 /* And put them out in forward order. */ 548 middle: do { 549 if (count-- == 0) 550 return (num); 551 *bp++ = *--zs->zs_stackp; 552 } while (zs->zs_stackp > de_stack); 553 554 /* Generate the new entry. */ 555 if ((zs->zs_code = zs->zs_free_ent) < zs->zs_maxmaxcode) { 556 tab_prefixof(zs->zs_code) = (u_short) zs->zs_oldcode; 557 tab_suffixof(zs->zs_code) = zs->zs_finchar; 558 zs->zs_free_ent = zs->zs_code + 1; 559 } 560 561 /* Remember previous code. */ 562 zs->zs_oldcode = zs->zs_incode; 563 } 564 zs->zs_state = S_EOF; 565 eof: return (num - count); 566 } 567 568 /*- 569 * Read one code from the standard input. If EOF, return -1. 570 * Inputs: 571 * stdin 572 * Outputs: 573 * code or -1 is returned. 574 */ 575 static code_int 576 getcode(zs) 577 struct s_zstate *zs; 578 { 579 code_int gcode; 580 int r_off, bits; 581 u_char *bp; 582 583 if (zs->zs_clear_flg > 0 || zs->zs_offset >= zs->zs_size || 584 zs->zs_free_ent > zs->zs_maxcode) { 585 586 zs->zs_bp += zs->zs_n_bits; 587 /* 588 * If the next entry will be too big for the current gcode 589 * size, then we must increase the size. This implies reading 590 * a new buffer full, too. 591 */ 592 if (zs->zs_free_ent > zs->zs_maxcode) { 593 zs->zs_n_bits++; 594 if (zs->zs_n_bits == zs->zs_maxbits) /* Won't get any bigger now. */ 595 zs->zs_maxcode = zs->zs_maxmaxcode; 596 else 597 zs->zs_maxcode = MAXCODE(zs->zs_n_bits); 598 } 599 if (zs->zs_clear_flg > 0) { 600 zs->zs_maxcode = MAXCODE(zs->zs_n_bits = INIT_BITS); 601 zs->zs_clear_flg = 0; 602 } 603 604 /* fill the buffer up to the neck */ 605 if (zs->zs_bp + zs->zs_n_bits > zs->zs_ebp) { 606 for (bp = zs->zs_buf; zs->zs_bp < zs->zs_ebp; 607 *bp++ = *zs->zs_bp++); 608 if ((bits = read(zs->zs_fd, bp, ZBUFSIZ - 609 (bp - zs->zs_buf))) < 0) 610 return -1; 611 zs->zs_bp = zs->zs_buf; 612 zs->zs_ebp = bp + bits; 613 } 614 zs->zs_offset = 0; 615 zs->zs_size = MIN(zs->zs_n_bits, zs->zs_ebp - zs->zs_bp); 616 if (zs->zs_size == 0) 617 return -1; 618 /* Round size down to integral number of codes. */ 619 zs->zs_size = (zs->zs_size << 3) - (zs->zs_n_bits - 1); 620 } 621 622 bp = zs->zs_bp; 623 r_off = zs->zs_offset; 624 bits = zs->zs_n_bits; 625 626 /* Get to the first byte. */ 627 bp += (r_off >> 3); 628 r_off &= 7; 629 630 /* Get first part (low order bits). */ 631 gcode = (*bp++ >> r_off); 632 bits -= (8 - r_off); 633 r_off = 8 - r_off; /* Now, roffset into gcode word. */ 634 635 /* Get any 8 bit parts in the middle (<=1 for up to 16 bits). */ 636 if (bits >= 8) { 637 gcode |= *bp++ << r_off; 638 r_off += 8; 639 bits -= 8; 640 } 641 642 /* High order bits. */ 643 gcode |= (*bp & rmask[bits]) << r_off; 644 zs->zs_offset += zs->zs_n_bits; 645 646 return (gcode); 647 } 648 649 static int 650 cl_block(zs) /* Table clear for block compress. */ 651 struct s_zstate *zs; 652 { 653 long rat; 654 655 zs->zs_checkpoint = zs->zs_in_count + CHECK_GAP; 656 657 if (zs->zs_in_count > 0x007fffff) { /* Shift will overflow. */ 658 rat = zs->zs_bytes_out >> 8; 659 if (rat == 0) /* Don't divide by zero. */ 660 rat = 0x7fffffff; 661 else 662 rat = zs->zs_in_count / rat; 663 } else 664 rat = (zs->zs_in_count << 8) / zs->zs_bytes_out; /* 8 fractional bits. */ 665 if (rat > zs->zs_ratio) 666 zs->zs_ratio = rat; 667 else { 668 zs->zs_ratio = 0; 669 cl_hash(zs, (count_int) zs->zs_hsize); 670 zs->zs_free_ent = FIRST; 671 zs->zs_clear_flg = 1; 672 if (output(zs, (code_int) CLEAR) == -1) 673 return (-1); 674 } 675 return (0); 676 } 677 678 static void 679 cl_hash(zs, cl_hsize) /* Reset code table. */ 680 struct s_zstate *zs; 681 count_int cl_hsize; 682 { 683 count_int *htab_p; 684 long i, m1; 685 686 m1 = -1; 687 htab_p = zs->zs_htab + cl_hsize; 688 i = cl_hsize - 16; 689 do { /* Might use Sys V memset(3) here. */ 690 *(htab_p - 16) = m1; 691 *(htab_p - 15) = m1; 692 *(htab_p - 14) = m1; 693 *(htab_p - 13) = m1; 694 *(htab_p - 12) = m1; 695 *(htab_p - 11) = m1; 696 *(htab_p - 10) = m1; 697 *(htab_p - 9) = m1; 698 *(htab_p - 8) = m1; 699 *(htab_p - 7) = m1; 700 *(htab_p - 6) = m1; 701 *(htab_p - 5) = m1; 702 *(htab_p - 4) = m1; 703 *(htab_p - 3) = m1; 704 *(htab_p - 2) = m1; 705 *(htab_p - 1) = m1; 706 htab_p -= 16; 707 } while ((i -= 16) >= 0); 708 for (i += 16; i > 0; i--) 709 *--htab_p = m1; 710 } 711 712 FILE * 713 zopen(name, mode, bits) 714 const char *name; 715 const char *mode; 716 int bits; 717 { 718 int fd; 719 void *cookie; 720 if ((fd = open(name, (*mode=='r'? O_RDONLY:O_WRONLY|O_CREAT), 721 S_IRUSR|S_IWUSR|S_IRGRP|S_IROTH)) == -1) 722 return NULL; 723 if ((cookie = z_open(fd, mode, bits)) == NULL) { 724 close(fd); 725 return NULL; 726 } 727 return funopen(cookie, (*mode == 'r'?zread:NULL), 728 (*mode == 'w'?zwrite:NULL), NULL, zclose); 729 } 730 731 void * 732 z_open(fd, mode, bits) 733 int fd; 734 const char *mode; 735 int bits; 736 { 737 struct s_zstate *zs; 738 739 if ((mode[0] != 'r' && mode[0] != 'w') || mode[1] != '\0' || 740 bits < 0 || bits > BITS) { 741 errno = EINVAL; 742 return (NULL); 743 } 744 745 if ((zs = calloc(1, sizeof(struct s_zstate))) == NULL) 746 return (NULL); 747 748 /* User settable max # bits/code. */ 749 zs->zs_maxbits = bits ? bits : BITS; 750 /* Should NEVER generate this code. */ 751 zs->zs_maxmaxcode = 1 << zs->zs_maxbits; 752 zs->zs_hsize = HSIZE; /* For dynamic table sizing. */ 753 zs->zs_free_ent = 0; /* First unused entry. */ 754 zs->zs_block_compress = BLOCK_MASK; 755 zs->zs_clear_flg = 0; 756 zs->zs_ratio = 0; 757 zs->zs_checkpoint = CHECK_GAP; 758 zs->zs_in_count = 1; /* Length of input. */ 759 zs->zs_out_count = 0; /* # of codes output (for debugging).*/ 760 zs->zs_state = S_START; 761 zs->zs_offset = 0; 762 zs->zs_size = 0; 763 zs->zs_mode = mode[0]; 764 zs->zs_bp = zs->zs_ebp = zs->zs_buf; 765 766 zs->zs_fd = fd; 767 return zs; 768 } 769 770 int 771 z_check_header(fd, sb, ofn) 772 int fd; 773 struct stat *sb; 774 const char *ofn; 775 { 776 int f; 777 u_char buf[sizeof(z_magic)]; 778 off_t off = lseek(fd, 0, SEEK_CUR); 779 780 f = (read(fd, buf, sizeof(buf)) == sizeof(buf) && 781 !memcmp(buf, z_magic, sizeof(buf))); 782 783 lseek (fd, off, SEEK_SET); 784 785 return f; 786 } 787