deflate.c
Go to the documentation of this file.
1 /* deflate.c -- compress data using the deflation algorithm
2  * Copyright (C) 1995-2005 Jean-loup Gailly.
3  * For conditions of distribution and use, see copyright notice in zlib.h
4  */
5 
6 /*
7  * ALGORITHM
8  *
9  * The "deflation" process depends on being able to identify portions
10  * of the input text which are identical to earlier input (within a
11  * sliding window trailing behind the input currently being processed).
12  *
13  * The most straightforward technique turns out to be the fastest for
14  * most input files: try all possible matches and select the longest.
15  * The key feature of this algorithm is that insertions into the string
16  * dictionary are very simple and thus fast, and deletions are avoided
17  * completely. Insertions are performed at each input character, whereas
18  * string matches are performed only when the previous match ends. So it
19  * is preferable to spend more time in matches to allow very fast string
20  * insertions and avoid deletions. The matching algorithm for small
21  * strings is inspired from that of Rabin & Karp. A brute force approach
22  * is used to find longer strings when a small match has been found.
23  * A similar algorithm is used in comic (by Jan-Mark Wams) and freeze
24  * (by Leonid Broukhis).
25  * A previous version of this file used a more sophisticated algorithm
26  * (by Fiala and Greene) which is guaranteed to run in linear amortized
27  * time, but has a larger average cost, uses more memory and is patented.
28  * However the F&G algorithm may be faster for some highly redundant
29  * files if the parameter max_chain_length (described below) is too large.
30  *
31  * ACKNOWLEDGEMENTS
32  *
33  * The idea of lazy evaluation of matches is due to Jan-Mark Wams, and
34  * I found it in 'freeze' written by Leonid Broukhis.
35  * Thanks to many people for bug reports and testing.
36  *
37  * REFERENCES
38  *
39  * Deutsch, L.P.,"DEFLATE Compressed Data Format Specification".
40  * Available in http://www.ietf.org/rfc/rfc1951.txt
41  *
42  * A description of the Rabin and Karp algorithm is given in the book
43  * "Algorithms" by R. Sedgewick, Addison-Wesley, p252.
44  *
45  * Fiala,E.R., and Greene,D.H.
46  * Data Compression with Finite Windows, Comm.ACM, 32,4 (1989) 490-595
47  *
48  */
49 
50 /* @(#) $Id$ */
51 
52 #include "deflate.h"
53 
54 const char deflate_copyright[] =
55  " deflate 1.2.3 Copyright 1995-2005 Jean-loup Gailly ";
56 /*
57  If you use the zlib library in a product, an acknowledgment is welcome
58  in the documentation of your product. If for some reason you cannot
59  include such an acknowledgment, I would appreciate that you keep this
60  copyright string in the executable of your product.
61  */
62 
63 /* ===========================================================================
64  * Function prototypes.
65  */
66 typedef enum {
67  need_more, /* block not completed, need more input or more output */
68  block_done, /* block flush performed */
69  finish_started, /* finish started, need only more output at next deflate */
70  finish_done /* finish done, accept no more input or output */
71 } block_state;
72 
73 typedef block_state (*compress_func) OF((deflate_state *s, int flush));
74 /* Compression function. Returns the block state after the call. */
75 
79 #ifndef FASTEST
81 #endif
82 local void lm_init OF((deflate_state *s));
84 local void flush_pending OF((z_streamp strm));
85 local int read_buf OF((z_streamp strm, Bytef *buf, unsigned size));
86 #ifndef FASTEST
87 #ifdef ASMV
88  void match_init OF((void)); /* asm code initialization */
89  uInt longest_match OF((deflate_state *s, IPos cur_match));
90 #else
91 local uInt longest_match OF((deflate_state *s, IPos cur_match));
92 #endif
93 #endif
95 
96 #ifdef DEBUG
98  int length));
99 #endif
100 
101 /* ===========================================================================
102  * Local data
103  */
104 
105 #define NIL 0
106 /* Tail of hash chains */
107 
108 #ifndef TOO_FAR
109 # define TOO_FAR 4096
110 #endif
111 /* Matches of length 3 are discarded if their distance exceeds TOO_FAR */
112 
113 #define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1)
114 /* Minimum amount of lookahead, except at the end of the input file.
115  * See deflate.c for comments about the MIN_MATCH+1.
116  */
117 
118 /* Values for max_lazy_match, good_match and max_chain_length, depending on
119  * the desired pack level (0..9). The values given below have been tuned to
120  * exclude worst case performance for pathological files. Better values may be
121  * found for specific files.
122  */
123 typedef struct config_s {
124  ush good_length; /* reduce lazy search above this match length */
125  ush max_lazy; /* do not perform lazy search above this match length */
126  ush nice_length; /* quit search above this match length */
128  compress_func func;
129 } config;
130 
131 #ifdef FASTEST
132 local const config configuration_table[2] = {
133 /* good lazy nice chain */
134 /* 0 */ {0, 0, 0, 0, deflate_stored}, /* store only */
135 /* 1 */ {4, 4, 8, 4, deflate_fast}}; /* max speed, no lazy matches */
136 #else
137 local const config configuration_table[10] = {
138 /* good lazy nice chain */
139 /* 0 */ {0, 0, 0, 0, deflate_stored}, /* store only */
140 /* 1 */ {4, 4, 8, 4, deflate_fast}, /* max speed, no lazy matches */
141 /* 2 */ {4, 5, 16, 8, deflate_fast},
142 /* 3 */ {4, 6, 32, 32, deflate_fast},
143 
144 /* 4 */ {4, 4, 16, 16, deflate_slow}, /* lazy matches */
145 /* 5 */ {8, 16, 32, 32, deflate_slow},
146 /* 6 */ {8, 16, 128, 128, deflate_slow},
147 /* 7 */ {8, 32, 128, 256, deflate_slow},
148 /* 8 */ {32, 128, 258, 1024, deflate_slow},
149 /* 9 */ {32, 258, 258, 4096, deflate_slow}}; /* max compression */
150 #endif
151 
152 /* Note: the deflate() code requires max_lazy >= MIN_MATCH and max_chain >= 4
153  * For deflate_fast() (levels <= 3) good is ignored and lazy has a different
154  * meaning.
155  */
156 
157 #define EQUAL 0
158 /* result of memcmp for equal strings */
159 
160 #ifndef NO_DUMMY_DECL
161 struct static_tree_desc_s {int dummy;}; /* for buggy compilers */
162 #endif
163 
164 /* ===========================================================================
165  * Update a hash value with the given input byte
166  * IN assertion: all calls to to UPDATE_HASH are made with consecutive
167  * input characters, so that a running hash key can be computed from the
168  * previous key instead of complete recalculation each time.
169  */
170 #define UPDATE_HASH(s,h,c) (h = (((h)<<s->hash_shift) ^ (c)) & s->hash_mask)
171 
172 
173 /* ===========================================================================
174  * Insert string str in the dictionary and set match_head to the previous head
175  * of the hash chain (the most recent string with same hash key). Return
176  * the previous length of the hash chain.
177  * If this file is compiled with -DFASTEST, the compression level is forced
178  * to 1, and no hash chains are maintained.
179  * IN assertion: all calls to to INSERT_STRING are made with consecutive
180  * input characters and the first MIN_MATCH bytes of str are valid
181  * (except for the last MIN_MATCH-1 bytes of the input file).
182  */
183 #ifdef FASTEST
184 #define INSERT_STRING(s, str, match_head) \
185  (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \
186  match_head = s->head[s->ins_h], \
187  s->head[s->ins_h] = (Pos)(str))
188 #else
189 #define INSERT_STRING(s, str, match_head) \
190  (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \
191  match_head = s->prev[(str) & s->w_mask] = s->head[s->ins_h], \
192  s->head[s->ins_h] = (Pos)(str))
193 #endif
194 
195 /* ===========================================================================
196  * Initialize the hash table (avoiding 64K overflow for 16 bit systems).
197  * prev[] will be initialized on the fly.
198  */
199 #define CLEAR_HASH(s) \
200  s->head[s->hash_size-1] = NIL; \
201  zmemzero((Bytef *)s->head, (unsigned)(s->hash_size-1)*sizeof(*s->head));
202 
203 /* ========================================================================= */
204 int ZEXPORT deflateInit_(strm, level, version, stream_size)
205  z_streamp strm;
206  int level;
207  const char *version;
208  int stream_size;
209 {
210  return deflateInit2_(strm, level, Z_DEFLATED, MAX_WBITS, DEF_MEM_LEVEL,
211  Z_DEFAULT_STRATEGY, version, stream_size);
212  /* To do: ignore strm->next_in if we use it as window */
213 }
214 
215 /* ========================================================================= */
216 int ZEXPORT deflateInit2_(strm, level, method, windowBits, memLevel, strategy,
217  version, stream_size)
218  z_streamp strm;
219  int level;
220  int method;
221  int windowBits;
222  int memLevel;
223  int strategy;
224  const char *version;
225  int stream_size;
226 {
227  deflate_state *s;
228  int wrap = 1;
229  static const char my_version[] = ZLIB_VERSION;
230 
231  ushf *overlay;
232  /* We overlay pending_buf and d_buf+l_buf. This works since the average
233  * output size for (length,distance) codes is <= 24 bits.
234  */
235 
236  if (version == Z_NULL || version[0] != my_version[0] ||
237  stream_size != sizeof(z_stream)) {
238  return Z_VERSION_ERROR;
239  }
240  if (strm == Z_NULL) return Z_STREAM_ERROR;
241 
242  strm->msg = Z_NULL;
243  if (strm->zalloc == (alloc_func)0) {
244  strm->zalloc = zcalloc;
245  strm->opaque = (voidpf)0;
246  }
247  if (strm->zfree == (free_func)0) strm->zfree = zcfree;
248 
249 #ifdef FASTEST
250  if (level != 0) level = 1;
251 #else
252  if (level == Z_DEFAULT_COMPRESSION) level = 6;
253 #endif
254 
255  if (windowBits < 0) { /* suppress zlib wrapper */
256  wrap = 0;
257  windowBits = -windowBits;
258  }
259 #ifdef GZIP
260  else if (windowBits > 15) {
261  wrap = 2; /* write gzip wrapper instead */
262  windowBits -= 16;
263  }
264 #endif
265  if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || method != Z_DEFLATED ||
266  windowBits < 8 || windowBits > 15 || level < 0 || level > 9 ||
267  strategy < 0 || strategy > Z_FIXED) {
268  return Z_STREAM_ERROR;
269  }
270  if (windowBits == 8) windowBits = 9; /* until 256-byte window bug fixed */
271  s = (deflate_state *) ZALLOC(strm, 1, sizeof(deflate_state));
272  if (s == Z_NULL) return Z_MEM_ERROR;
273  strm->state = (struct internal_state FAR *)s;
274  s->strm = strm;
275 
276  s->wrap = wrap;
277  s->gzhead = Z_NULL;
278  s->w_bits = windowBits;
279  s->w_size = 1 << s->w_bits;
280  s->w_mask = s->w_size - 1;
281 
282  s->hash_bits = memLevel + 7;
283  s->hash_size = 1 << s->hash_bits;
284  s->hash_mask = s->hash_size - 1;
285  s->hash_shift = ((s->hash_bits+MIN_MATCH-1)/MIN_MATCH);
286 
287  s->window = (Bytef *) ZALLOC(strm, s->w_size, 2*sizeof(Byte));
288  s->prev = (Posf *) ZALLOC(strm, s->w_size, sizeof(Pos));
289  s->head = (Posf *) ZALLOC(strm, s->hash_size, sizeof(Pos));
290 
291  s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */
292 
293  overlay = (ushf *) ZALLOC(strm, s->lit_bufsize, sizeof(ush)+2);
294  s->pending_buf = (uchf *) overlay;
295  s->pending_buf_size = (ulg)s->lit_bufsize * (sizeof(ush)+2L);
296 
297  if (s->window == Z_NULL || s->prev == Z_NULL || s->head == Z_NULL ||
298  s->pending_buf == Z_NULL) {
299  s->status = FINISH_STATE;
300  strm->msg = (char*)ERR_MSG(Z_MEM_ERROR);
301  deflateEnd (strm);
302  return Z_MEM_ERROR;
303  }
304  s->d_buf = overlay + s->lit_bufsize/sizeof(ush);
305  s->l_buf = s->pending_buf + (1+sizeof(ush))*s->lit_bufsize;
306 
307  s->level = level;
308  s->strategy = strategy;
309  s->method = (Byte)method;
310 
311  return deflateReset(strm);
312 }
313 
314 /* ========================================================================= */
316  z_streamp strm;
318  uInt dictLength;
319 {
320  deflate_state *s;
321  uInt length = dictLength;
322  uInt n;
323  IPos hash_head = 0;
324 
325  if (strm == Z_NULL || strm->state == Z_NULL || dictionary == Z_NULL ||
326  strm->state->wrap == 2 ||
327  (strm->state->wrap == 1 && strm->state->status != INIT_STATE))
328  return Z_STREAM_ERROR;
329 
330  s = strm->state;
331  if (s->wrap)
332  strm->adler = adler32(strm->adler, dictionary, dictLength);
333 
334  if (length < MIN_MATCH) return Z_OK;
335  if (length > MAX_DIST(s)) {
336  length = MAX_DIST(s);
337  dictionary += dictLength - length; /* use the tail of the dictionary */
338  }
339  zmemcpy(s->window, dictionary, length);
340  s->strstart = length;
341  s->block_start = (long)length;
342 
343  /* Insert all strings in the hash table (except for the last two bytes).
344  * s->lookahead stays null, so s->ins_h will be recomputed at the next
345  * call of fill_window.
346  */
347  s->ins_h = s->window[0];
348  UPDATE_HASH(s, s->ins_h, s->window[1]);
349  for (n = 0; n <= length - MIN_MATCH; n++) {
350  INSERT_STRING(s, n, hash_head);
351  }
352  if (hash_head) hash_head = 0; /* to make compiler happy */
353  return Z_OK;
354 }
355 
356 /* ========================================================================= */
358  z_streamp strm;
359 {
360  deflate_state *s;
361 
362  if (strm == Z_NULL || strm->state == Z_NULL ||
363  strm->zalloc == (alloc_func)0 || strm->zfree == (free_func)0) {
364  return Z_STREAM_ERROR;
365  }
366 
367  strm->total_in = strm->total_out = 0;
368  strm->msg = Z_NULL; /* use zfree if we ever allocate msg dynamically */
369  strm->data_type = Z_UNKNOWN;
370 
371  s = (deflate_state *)strm->state;
372  s->pending = 0;
373  s->pending_out = s->pending_buf;
374 
375  if (s->wrap < 0) {
376  s->wrap = -s->wrap; /* was made negative by deflate(..., Z_FINISH); */
377  }
378  s->status = s->wrap ? INIT_STATE : BUSY_STATE;
379  strm->adler =
380 #ifdef GZIP
381  s->wrap == 2 ? crc32(0L, Z_NULL, 0) :
382 #endif
383  adler32(0L, Z_NULL, 0);
384  s->last_flush = Z_NO_FLUSH;
385 
386  _tr_init(s);
387  lm_init(s);
388 
389  return Z_OK;
390 }
391 
392 /* ========================================================================= */
394  z_streamp strm;
396 {
397  if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
398  if (strm->state->wrap != 2) return Z_STREAM_ERROR;
399  strm->state->gzhead = head;
400  return Z_OK;
401 }
402 
403 /* ========================================================================= */
405  z_streamp strm;
406  int bits;
407  int value;
408 {
409  if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
410  strm->state->bi_valid = bits;
411  strm->state->bi_buf = (ush)(value & ((1 << bits) - 1));
412  return Z_OK;
413 }
414 
415 /* ========================================================================= */
417  z_streamp strm;
418  int level;
419  int strategy;
420 {
421  deflate_state *s;
422  compress_func func;
423  int err = Z_OK;
424 
425  if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
426  s = strm->state;
427 
428 #ifdef FASTEST
429  if (level != 0) level = 1;
430 #else
431  if (level == Z_DEFAULT_COMPRESSION) level = 6;
432 #endif
433  if (level < 0 || level > 9 || strategy < 0 || strategy > Z_FIXED) {
434  return Z_STREAM_ERROR;
435  }
436  func = configuration_table[s->level].func;
437 
438  if (func != configuration_table[level].func && strm->total_in != 0) {
439  /* Flush the last buffer: */
440  err = deflate(strm, Z_PARTIAL_FLUSH);
441  }
442  if (s->level != level) {
443  s->level = level;
444  s->max_lazy_match = configuration_table[level].max_lazy;
445  s->good_match = configuration_table[level].good_length;
446  s->nice_match = configuration_table[level].nice_length;
447  s->max_chain_length = configuration_table[level].max_chain;
448  }
449  s->strategy = strategy;
450  return err;
451 }
452 
453 /* ========================================================================= */
455  z_streamp strm;
456  int good_length;
457  int max_lazy;
458  int nice_length;
459  int max_chain;
460 {
461  deflate_state *s;
462 
463  if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
464  s = strm->state;
465  s->good_match = good_length;
467  s->nice_match = nice_length;
469  return Z_OK;
470 }
471 
472 /* =========================================================================
473  * For the default windowBits of 15 and memLevel of 8, this function returns
474  * a close to exact, as well as small, upper bound on the compressed size.
475  * They are coded as constants here for a reason--if the #define's are
476  * changed, then this function needs to be changed as well. The return
477  * value for 15 and 8 only works for those exact settings.
478  *
479  * For any setting other than those defaults for windowBits and memLevel,
480  * the value returned is a conservative worst case for the maximum expansion
481  * resulting from using fixed blocks instead of stored blocks, which deflate
482  * can emit on compressed data for some combinations of the parameters.
483  *
484  * This function could be more sophisticated to provide closer upper bounds
485  * for every combination of windowBits and memLevel, as well as wrap.
486  * But even the conservative upper bound of about 14% expansion does not
487  * seem onerous for output buffer allocation.
488  */
490  z_streamp strm;
491  uLong sourceLen;
492 {
493  deflate_state *s;
494  uLong destLen;
495 
496  /* conservative upper bound */
497  destLen = sourceLen +
498  ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 11;
499 
500  /* if can't get parameters, return conservative bound */
501  if (strm == Z_NULL || strm->state == Z_NULL)
502  return destLen;
503 
504  /* if not default parameters, return conservative bound */
505  s = strm->state;
506  if (s->w_bits != 15 || s->hash_bits != 8 + 7)
507  return destLen;
508 
509  /* default settings: return tight bound for that case */
510  return compressBound(sourceLen);
511 }
512 
513 /* =========================================================================
514  * Put a short in the pending buffer. The 16-bit value is put in MSB order.
515  * IN assertion: the stream state is correct and there is enough room in
516  * pending_buf.
517  */
518 local void putShortMSB (s, b)
519  deflate_state *s;
520  uInt b;
521 {
522  put_byte(s, (Byte)(b >> 8));
523  put_byte(s, (Byte)(b & 0xff));
524 }
525 
526 /* =========================================================================
527  * Flush as much pending output as possible. All deflate() output goes
528  * through this function so some applications may wish to modify it
529  * to avoid allocating a large strm->next_out buffer and copying into it.
530  * (See also read_buf()).
531  */
533  z_streamp strm;
534 {
535  unsigned len = strm->state->pending;
536 
537  if (len > strm->avail_out) len = strm->avail_out;
538  if (len == 0) return;
539 
540  zmemcpy(strm->next_out, strm->state->pending_out, len);
541  strm->next_out += len;
542  strm->state->pending_out += len;
543  strm->total_out += len;
544  strm->avail_out -= len;
545  strm->state->pending -= len;
546  if (strm->state->pending == 0) {
547  strm->state->pending_out = strm->state->pending_buf;
548  }
549 }
550 
551 /* ========================================================================= */
552 int ZEXPORT deflate (strm, flush)
553  z_streamp strm;
554  int flush;
555 {
556  int old_flush; /* value of flush param for previous deflate call */
557  deflate_state *s;
558 
559  if (strm == Z_NULL || strm->state == Z_NULL ||
560  flush > Z_FINISH || flush < 0) {
561  return Z_STREAM_ERROR;
562  }
563  s = strm->state;
564 
565  if (strm->next_out == Z_NULL ||
566  (strm->next_in == Z_NULL && strm->avail_in != 0) ||
567  (s->status == FINISH_STATE && flush != Z_FINISH)) {
568  ERR_RETURN(strm, Z_STREAM_ERROR);
569  }
570  if (strm->avail_out == 0) ERR_RETURN(strm, Z_BUF_ERROR);
571 
572  s->strm = strm; /* just in case */
573  old_flush = s->last_flush;
574  s->last_flush = flush;
575 
576  /* Write the header */
577  if (s->status == INIT_STATE) {
578 #ifdef GZIP
579  if (s->wrap == 2) {
580  strm->adler = crc32(0L, Z_NULL, 0);
581  put_byte(s, 31);
582  put_byte(s, 139);
583  put_byte(s, 8);
584  if (s->gzhead == NULL) {
585  put_byte(s, 0);
586  put_byte(s, 0);
587  put_byte(s, 0);
588  put_byte(s, 0);
589  put_byte(s, 0);
590  put_byte(s, s->level == 9 ? 2 :
591  (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
592  4 : 0));
593  put_byte(s, OS_CODE);
594  s->status = BUSY_STATE;
595  }
596  else {
597  put_byte(s, (s->gzhead->text ? 1 : 0) +
598  (s->gzhead->hcrc ? 2 : 0) +
599  (s->gzhead->extra == Z_NULL ? 0 : 4) +
600  (s->gzhead->name == Z_NULL ? 0 : 8) +
601  (s->gzhead->comment == Z_NULL ? 0 : 16)
602  );
603  put_byte(s, (Byte)(s->gzhead->time & 0xff));
604  put_byte(s, (Byte)((s->gzhead->time >> 8) & 0xff));
605  put_byte(s, (Byte)((s->gzhead->time >> 16) & 0xff));
606  put_byte(s, (Byte)((s->gzhead->time >> 24) & 0xff));
607  put_byte(s, s->level == 9 ? 2 :
608  (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ?
609  4 : 0));
610  put_byte(s, s->gzhead->os & 0xff);
611  if (s->gzhead->extra != NULL) {
612  put_byte(s, s->gzhead->extra_len & 0xff);
613  put_byte(s, (s->gzhead->extra_len >> 8) & 0xff);
614  }
615  if (s->gzhead->hcrc)
616  strm->adler = crc32(strm->adler, s->pending_buf,
617  s->pending);
618  s->gzindex = 0;
619  s->status = EXTRA_STATE;
620  }
621  }
622  else
623 #endif
624  {
625  uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8;
626  uInt level_flags;
627 
628  if (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2)
629  level_flags = 0;
630  else if (s->level < 6)
631  level_flags = 1;
632  else if (s->level == 6)
633  level_flags = 2;
634  else
635  level_flags = 3;
636  header |= (level_flags << 6);
637  if (s->strstart != 0) header |= PRESET_DICT;
638  header += 31 - (header % 31);
639 
640  s->status = BUSY_STATE;
641  putShortMSB(s, header);
642 
643  /* Save the adler32 of the preset dictionary: */
644  if (s->strstart != 0) {
645  putShortMSB(s, (uInt)(strm->adler >> 16));
646  putShortMSB(s, (uInt)(strm->adler & 0xffff));
647  }
648  strm->adler = adler32(0L, Z_NULL, 0);
649  }
650  }
651 #ifdef GZIP
652  if (s->status == EXTRA_STATE) {
653  if (s->gzhead->extra != NULL) {
654  uInt beg = s->pending; /* start of bytes to update crc */
655 
656  while (s->gzindex < (s->gzhead->extra_len & 0xffff)) {
657  if (s->pending == s->pending_buf_size) {
658  if (s->gzhead->hcrc && s->pending > beg)
659  strm->adler = crc32(strm->adler, s->pending_buf + beg,
660  s->pending - beg);
661  flush_pending(strm);
662  beg = s->pending;
663  if (s->pending == s->pending_buf_size)
664  break;
665  }
666  put_byte(s, s->gzhead->extra[s->gzindex]);
667  s->gzindex++;
668  }
669  if (s->gzhead->hcrc && s->pending > beg)
670  strm->adler = crc32(strm->adler, s->pending_buf + beg,
671  s->pending - beg);
672  if (s->gzindex == s->gzhead->extra_len) {
673  s->gzindex = 0;
674  s->status = NAME_STATE;
675  }
676  }
677  else
678  s->status = NAME_STATE;
679  }
680  if (s->status == NAME_STATE) {
681  if (s->gzhead->name != NULL) {
682  uInt beg = s->pending; /* start of bytes to update crc */
683  int val;
684 
685  do {
686  if (s->pending == s->pending_buf_size) {
687  if (s->gzhead->hcrc && s->pending > beg)
688  strm->adler = crc32(strm->adler, s->pending_buf + beg,
689  s->pending - beg);
690  flush_pending(strm);
691  beg = s->pending;
692  if (s->pending == s->pending_buf_size) {
693  val = 1;
694  break;
695  }
696  }
697  val = s->gzhead->name[s->gzindex++];
698  put_byte(s, val);
699  } while (val != 0);
700  if (s->gzhead->hcrc && s->pending > beg)
701  strm->adler = crc32(strm->adler, s->pending_buf + beg,
702  s->pending - beg);
703  if (val == 0) {
704  s->gzindex = 0;
705  s->status = COMMENT_STATE;
706  }
707  }
708  else
709  s->status = COMMENT_STATE;
710  }
711  if (s->status == COMMENT_STATE) {
712  if (s->gzhead->comment != NULL) {
713  uInt beg = s->pending; /* start of bytes to update crc */
714  int val;
715 
716  do {
717  if (s->pending == s->pending_buf_size) {
718  if (s->gzhead->hcrc && s->pending > beg)
719  strm->adler = crc32(strm->adler, s->pending_buf + beg,
720  s->pending - beg);
721  flush_pending(strm);
722  beg = s->pending;
723  if (s->pending == s->pending_buf_size) {
724  val = 1;
725  break;
726  }
727  }
728  val = s->gzhead->comment[s->gzindex++];
729  put_byte(s, val);
730  } while (val != 0);
731  if (s->gzhead->hcrc && s->pending > beg)
732  strm->adler = crc32(strm->adler, s->pending_buf + beg,
733  s->pending - beg);
734  if (val == 0)
735  s->status = HCRC_STATE;
736  }
737  else
738  s->status = HCRC_STATE;
739  }
740  if (s->status == HCRC_STATE) {
741  if (s->gzhead->hcrc) {
742  if (s->pending + 2 > s->pending_buf_size)
743  flush_pending(strm);
744  if (s->pending + 2 <= s->pending_buf_size) {
745  put_byte(s, (Byte)(strm->adler & 0xff));
746  put_byte(s, (Byte)((strm->adler >> 8) & 0xff));
747  strm->adler = crc32(0L, Z_NULL, 0);
748  s->status = BUSY_STATE;
749  }
750  }
751  else
752  s->status = BUSY_STATE;
753  }
754 #endif
755 
756  /* Flush as much pending output as possible */
757  if (s->pending != 0) {
758  flush_pending(strm);
759  if (strm->avail_out == 0) {
760  /* Since avail_out is 0, deflate will be called again with
761  * more output space, but possibly with both pending and
762  * avail_in equal to zero. There won't be anything to do,
763  * but this is not an error situation so make sure we
764  * return OK instead of BUF_ERROR at next call of deflate:
765  */
766  s->last_flush = -1;
767  return Z_OK;
768  }
769 
770  /* Make sure there is something to do and avoid duplicate consecutive
771  * flushes. For repeated and useless calls with Z_FINISH, we keep
772  * returning Z_STREAM_END instead of Z_BUF_ERROR.
773  */
774  } else if (strm->avail_in == 0 && flush <= old_flush &&
775  flush != Z_FINISH) {
776  ERR_RETURN(strm, Z_BUF_ERROR);
777  }
778 
779  /* User must not provide more input after the first FINISH: */
780  if (s->status == FINISH_STATE && strm->avail_in != 0) {
781  ERR_RETURN(strm, Z_BUF_ERROR);
782  }
783 
784  /* Start a new block or continue the current one.
785  */
786  if (strm->avail_in != 0 || s->lookahead != 0 ||
787  (flush != Z_NO_FLUSH && s->status != FINISH_STATE)) {
788  block_state bstate;
789 
790  bstate = (*(configuration_table[s->level].func))(s, flush);
791 
792  if (bstate == finish_started || bstate == finish_done) {
793  s->status = FINISH_STATE;
794  }
795  if (bstate == need_more || bstate == finish_started) {
796  if (strm->avail_out == 0) {
797  s->last_flush = -1; /* avoid BUF_ERROR next call, see above */
798  }
799  return Z_OK;
800  /* If flush != Z_NO_FLUSH && avail_out == 0, the next call
801  * of deflate should use the same flush parameter to make sure
802  * that the flush is complete. So we don't have to output an
803  * empty block here, this will be done at next call. This also
804  * ensures that for a very small output buffer, we emit at most
805  * one empty block.
806  */
807  }
808  if (bstate == block_done) {
809  if (flush == Z_PARTIAL_FLUSH) {
810  _tr_align(s);
811  } else { /* FULL_FLUSH or SYNC_FLUSH */
812  _tr_stored_block(s, (char*)0, 0L, 0);
813  /* For a full flush, this empty block will be recognized
814  * as a special marker by inflate_sync().
815  */
816  if (flush == Z_FULL_FLUSH) {
817  CLEAR_HASH(s); /* forget history */
818  }
819  }
820  flush_pending(strm);
821  if (strm->avail_out == 0) {
822  s->last_flush = -1; /* avoid BUF_ERROR at next call, see above */
823  return Z_OK;
824  }
825  }
826  }
827  Assert(strm->avail_out > 0, "bug2");
828 
829  if (flush != Z_FINISH) return Z_OK;
830  if (s->wrap <= 0) return Z_STREAM_END;
831 
832  /* Write the trailer */
833 #ifdef GZIP
834  if (s->wrap == 2) {
835  put_byte(s, (Byte)(strm->adler & 0xff));
836  put_byte(s, (Byte)((strm->adler >> 8) & 0xff));
837  put_byte(s, (Byte)((strm->adler >> 16) & 0xff));
838  put_byte(s, (Byte)((strm->adler >> 24) & 0xff));
839  put_byte(s, (Byte)(strm->total_in & 0xff));
840  put_byte(s, (Byte)((strm->total_in >> 8) & 0xff));
841  put_byte(s, (Byte)((strm->total_in >> 16) & 0xff));
842  put_byte(s, (Byte)((strm->total_in >> 24) & 0xff));
843  }
844  else
845 #endif
846  {
847  putShortMSB(s, (uInt)(strm->adler >> 16));
848  putShortMSB(s, (uInt)(strm->adler & 0xffff));
849  }
850  flush_pending(strm);
851  /* If avail_out is zero, the application will call deflate again
852  * to flush the rest.
853  */
854  if (s->wrap > 0) s->wrap = -s->wrap; /* write the trailer only once! */
855  return s->pending != 0 ? Z_OK : Z_STREAM_END;
856 }
857 
858 /* ========================================================================= */
860  z_streamp strm;
861 {
862  int status;
863 
864  if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR;
865 
866  status = strm->state->status;
867  if (status != INIT_STATE &&
868  status != EXTRA_STATE &&
869  status != NAME_STATE &&
870  status != COMMENT_STATE &&
871  status != HCRC_STATE &&
872  status != BUSY_STATE &&
873  status != FINISH_STATE) {
874  return Z_STREAM_ERROR;
875  }
876 
877  /* Deallocate in reverse order of allocations: */
878  TRY_FREE(strm, strm->state->pending_buf);
879  TRY_FREE(strm, strm->state->head);
880  TRY_FREE(strm, strm->state->prev);
881  TRY_FREE(strm, strm->state->window);
882 
883  ZFREE(strm, strm->state);
884  strm->state = Z_NULL;
885 
886  return status == BUSY_STATE ? Z_DATA_ERROR : Z_OK;
887 }
888 
889 /* =========================================================================
890  * Copy the source state to the destination state.
891  * To simplify the source, this is not supported for 16-bit MSDOS (which
892  * doesn't have enough memory anyway to duplicate compression states).
893  */
894 int ZEXPORT deflateCopy (dest, source)
895  z_streamp dest;
896  z_streamp source;
897 {
898 #ifdef MAXSEG_64K
899  return Z_STREAM_ERROR;
900 #else
901  deflate_state *ds;
902  deflate_state *ss;
903  ushf *overlay;
904 
905 
906  if (source == Z_NULL || dest == Z_NULL || source->state == Z_NULL) {
907  return Z_STREAM_ERROR;
908  }
909 
910  ss = source->state;
911 
912  zmemcpy(dest, source, sizeof(z_stream));
913 
914  ds = (deflate_state *) ZALLOC(dest, 1, sizeof(deflate_state));
915  if (ds == Z_NULL) return Z_MEM_ERROR;
916  dest->state = (struct internal_state FAR *) ds;
917  zmemcpy(ds, ss, sizeof(deflate_state));
918  ds->strm = dest;
919 
920  ds->window = (Bytef *) ZALLOC(dest, ds->w_size, 2*sizeof(Byte));
921  ds->prev = (Posf *) ZALLOC(dest, ds->w_size, sizeof(Pos));
922  ds->head = (Posf *) ZALLOC(dest, ds->hash_size, sizeof(Pos));
923  overlay = (ushf *) ZALLOC(dest, ds->lit_bufsize, sizeof(ush)+2);
924  ds->pending_buf = (uchf *) overlay;
925 
926  if (ds->window == Z_NULL || ds->prev == Z_NULL || ds->head == Z_NULL ||
927  ds->pending_buf == Z_NULL) {
928  deflateEnd (dest);
929  return Z_MEM_ERROR;
930  }
931  /* following zmemcpy do not work for 16-bit MSDOS */
932  zmemcpy(ds->window, ss->window, ds->w_size * 2 * sizeof(Byte));
933  zmemcpy(ds->prev, ss->prev, ds->w_size * sizeof(Pos));
934  zmemcpy(ds->head, ss->head, ds->hash_size * sizeof(Pos));
936 
937  ds->pending_out = ds->pending_buf + (ss->pending_out - ss->pending_buf);
938  ds->d_buf = overlay + ds->lit_bufsize/sizeof(ush);
939  ds->l_buf = ds->pending_buf + (1+sizeof(ush))*ds->lit_bufsize;
940 
941  ds->l_desc.dyn_tree = ds->dyn_ltree;
942  ds->d_desc.dyn_tree = ds->dyn_dtree;
943  ds->bl_desc.dyn_tree = ds->bl_tree;
944 
945  return Z_OK;
946 #endif /* MAXSEG_64K */
947 }
948 
949 /* ===========================================================================
950  * Read a new buffer from the current input stream, update the adler32
951  * and total number of bytes read. All deflate() input goes through
952  * this function so some applications may wish to modify it to avoid
953  * allocating a large strm->next_in buffer and copying from it.
954  * (See also flush_pending()).
955  */
957  z_streamp strm;
958  Bytef *buf;
959  unsigned size;
960 {
961  unsigned len = strm->avail_in;
962 
963  if (len > size) len = size;
964  if (len == 0) return 0;
965 
966  strm->avail_in -= len;
967 
968  if (strm->state->wrap == 1) {
969  strm->adler = adler32(strm->adler, strm->next_in, len);
970  }
971 #ifdef GZIP
972  else if (strm->state->wrap == 2) {
973  strm->adler = crc32(strm->adler, strm->next_in, len);
974  }
975 #endif
976  zmemcpy(buf, strm->next_in, len);
977  strm->next_in += len;
978  strm->total_in += len;
979 
980  return (int)len;
981 }
982 
983 /* ===========================================================================
984  * Initialize the "longest match" routines for a new zlib stream
985  */
986 local void lm_init (s)
987  deflate_state *s;
988 {
989  s->window_size = (ulg)2L*s->w_size;
990 
991  CLEAR_HASH(s);
992 
993  /* Set the default configuration parameters:
994  */
995  s->max_lazy_match = configuration_table[s->level].max_lazy;
996  s->good_match = configuration_table[s->level].good_length;
997  s->nice_match = configuration_table[s->level].nice_length;
998  s->max_chain_length = configuration_table[s->level].max_chain;
999 
1000  s->strstart = 0;
1001  s->block_start = 0L;
1002  s->lookahead = 0;
1003  s->match_length = s->prev_length = MIN_MATCH-1;
1004  s->match_available = 0;
1005  s->ins_h = 0;
1006 #ifndef FASTEST
1007 #ifdef ASMV
1008  match_init(); /* initialize the asm code */
1009 #endif
1010 #endif
1011 }
1012 
1013 #ifndef FASTEST
1014 /* ===========================================================================
1015  * Set match_start to the longest match starting at the given string and
1016  * return its length. Matches shorter or equal to prev_length are discarded,
1017  * in which case the result is equal to prev_length and match_start is
1018  * garbage.
1019  * IN assertions: cur_match is the head of the hash chain for the current
1020  * string (strstart) and its distance is <= MAX_DIST, and prev_length >= 1
1021  * OUT assertion: the match length is not greater than s->lookahead.
1022  */
1023 #ifndef ASMV
1024 /* For 80x86 and 680x0, an optimized version will be provided in match.asm or
1025  * match.S. The code will be functionally equivalent.
1026  */
1027 local uInt longest_match(s, cur_match)
1028  deflate_state *s;
1029  IPos cur_match; /* current match */
1030 {
1031  unsigned chain_length = s->max_chain_length;/* max hash chain length */
1032  register Bytef *scan = s->window + s->strstart; /* current string */
1033  register Bytef *match; /* matched string */
1034  register int len; /* length of current match */
1035  int best_len = s->prev_length; /* best match length so far */
1036  int nice_match = s->nice_match; /* stop if match long enough */
1037  IPos limit = s->strstart > (IPos)MAX_DIST(s) ?
1038  s->strstart - (IPos)MAX_DIST(s) : NIL;
1039  /* Stop when cur_match becomes <= limit. To simplify the code,
1040  * we prevent matches with the string of window index 0.
1041  */
1042  Posf *prev = s->prev;
1043  uInt wmask = s->w_mask;
1044 
1045 #ifdef UNALIGNED_OK
1046  /* Compare two bytes at a time. Note: this is not always beneficial.
1047  * Try with and without -DUNALIGNED_OK to check.
1048  */
1049  register Bytef *strend = s->window + s->strstart + MAX_MATCH - 1;
1050  register ush scan_start = *(ushf*)scan;
1051  register ush scan_end = *(ushf*)(scan+best_len-1);
1052 #else
1053  register Bytef *strend = s->window + s->strstart + MAX_MATCH;
1054  register Byte scan_end1 = scan[best_len-1];
1055  register Byte scan_end = scan[best_len];
1056 #endif
1057 
1058  /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16.
1059  * It is easy to get rid of this optimization if necessary.
1060  */
1061  Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever");
1062 
1063  /* Do not waste too much time if we already have a good match: */
1064  if (s->prev_length >= s->good_match) {
1065  chain_length >>= 2;
1066  }
1067  /* Do not look for matches beyond the end of the input. This is necessary
1068  * to make deflate deterministic.
1069  */
1070  if ((uInt)nice_match > s->lookahead) nice_match = s->lookahead;
1071 
1072  Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead");
1073 
1074  do {
1075  Assert(cur_match < s->strstart, "no future");
1076  match = s->window + cur_match;
1077 
1078  /* Skip to next match if the match length cannot increase
1079  * or if the match length is less than 2. Note that the checks below
1080  * for insufficient lookahead only occur occasionally for performance
1081  * reasons. Therefore uninitialized memory will be accessed, and
1082  * conditional jumps will be made that depend on those values.
1083  * However the length of the match is limited to the lookahead, so
1084  * the output of deflate is not affected by the uninitialized values.
1085  */
1086 #if (defined(UNALIGNED_OK) && MAX_MATCH == 258)
1087  /* This code assumes sizeof(unsigned short) == 2. Do not use
1088  * UNALIGNED_OK if your compiler uses a different size.
1089  */
1090  if (*(ushf*)(match+best_len-1) != scan_end ||
1091  *(ushf*)match != scan_start) continue;
1092 
1093  /* It is not necessary to compare scan[2] and match[2] since they are
1094  * always equal when the other bytes match, given that the hash keys
1095  * are equal and that HASH_BITS >= 8. Compare 2 bytes at a time at
1096  * strstart+3, +5, ... up to strstart+257. We check for insufficient
1097  * lookahead only every 4th comparison; the 128th check will be made
1098  * at strstart+257. If MAX_MATCH-2 is not a multiple of 8, it is
1099  * necessary to put more guard bytes at the end of the window, or
1100  * to check more often for insufficient lookahead.
1101  */
1102  Assert(scan[2] == match[2], "scan[2]?");
1103  scan++, match++;
1104  do {
1105  } while (*(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
1106  *(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
1107  *(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
1108  *(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
1109  scan < strend);
1110  /* The funny "do {}" generates better code on most compilers */
1111 
1112  /* Here, scan <= window+strstart+257 */
1113  Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan");
1114  if (*scan == *match) scan++;
1115 
1116  len = (MAX_MATCH - 1) - (int)(strend-scan);
1117  scan = strend - (MAX_MATCH-1);
1118 
1119 #else /* UNALIGNED_OK */
1120 
1121  if (match[best_len] != scan_end ||
1122  match[best_len-1] != scan_end1 ||
1123  *match != *scan ||
1124  *++match != scan[1]) continue;
1125 
1126  /* The check at best_len-1 can be removed because it will be made
1127  * again later. (This heuristic is not always a win.)
1128  * It is not necessary to compare scan[2] and match[2] since they
1129  * are always equal when the other bytes match, given that
1130  * the hash keys are equal and that HASH_BITS >= 8.
1131  */
1132  scan += 2, match++;
1133  Assert(*scan == *match, "match[2]?");
1134 
1135  /* We check for insufficient lookahead only every 8th comparison;
1136  * the 256th check will be made at strstart+258.
1137  */
1138  do {
1139  } while (*++scan == *++match && *++scan == *++match &&
1140  *++scan == *++match && *++scan == *++match &&
1141  *++scan == *++match && *++scan == *++match &&
1142  *++scan == *++match && *++scan == *++match &&
1143  scan < strend);
1144 
1145  Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan");
1146 
1147  len = MAX_MATCH - (int)(strend - scan);
1148  scan = strend - MAX_MATCH;
1149 
1150 #endif /* UNALIGNED_OK */
1151 
1152  if (len > best_len) {
1153  s->match_start = cur_match;
1154  best_len = len;
1155  if (len >= nice_match) break;
1156 #ifdef UNALIGNED_OK
1157  scan_end = *(ushf*)(scan+best_len-1);
1158 #else
1159  scan_end1 = scan[best_len-1];
1160  scan_end = scan[best_len];
1161 #endif
1162  }
1163  } while ((cur_match = prev[cur_match & wmask]) > limit
1164  && --chain_length != 0);
1165 
1166  if ((uInt)best_len <= s->lookahead) return (uInt)best_len;
1167  return s->lookahead;
1168 }
1169 #endif /* ASMV */
1170 #endif /* FASTEST */
1171 
1172 /* ---------------------------------------------------------------------------
1173  * Optimized version for level == 1 or strategy == Z_RLE only
1174  */
1176  deflate_state *s;
1177  IPos cur_match; /* current match */
1178 {
1179  register Bytef *scan = s->window + s->strstart; /* current string */
1180  register Bytef *match; /* matched string */
1181  register int len; /* length of current match */
1182  register Bytef *strend = s->window + s->strstart + MAX_MATCH;
1183 
1184  /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16.
1185  * It is easy to get rid of this optimization if necessary.
1186  */
1187  Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever");
1188 
1189  Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead");
1190 
1191  Assert(cur_match < s->strstart, "no future");
1192 
1193  match = s->window + cur_match;
1194 
1195  /* Return failure if the match length is less than 2:
1196  */
1197  if (match[0] != scan[0] || match[1] != scan[1]) return MIN_MATCH-1;
1198 
1199  /* The check at best_len-1 can be removed because it will be made
1200  * again later. (This heuristic is not always a win.)
1201  * It is not necessary to compare scan[2] and match[2] since they
1202  * are always equal when the other bytes match, given that
1203  * the hash keys are equal and that HASH_BITS >= 8.
1204  */
1205  scan += 2, match += 2;
1206  Assert(*scan == *match, "match[2]?");
1207 
1208  /* We check for insufficient lookahead only every 8th comparison;
1209  * the 256th check will be made at strstart+258.
1210  */
1211  do {
1212  } while (*++scan == *++match && *++scan == *++match &&
1213  *++scan == *++match && *++scan == *++match &&
1214  *++scan == *++match && *++scan == *++match &&
1215  *++scan == *++match && *++scan == *++match &&
1216  scan < strend);
1217 
1218  Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan");
1219 
1220  len = MAX_MATCH - (int)(strend - scan);
1221 
1222  if (len < MIN_MATCH) return MIN_MATCH - 1;
1223 
1224  s->match_start = cur_match;
1225  return (uInt)len <= s->lookahead ? (uInt)len : s->lookahead;
1226 }
1227 
1228 #ifdef DEBUG
1229 /* ===========================================================================
1230  * Check that the match at match_start is indeed a match.
1231  */
1232 local void check_match(s, start, match, length)
1233  deflate_state *s;
1234  IPos start, match;
1235  int length;
1236 {
1237  /* check that the match is indeed a match */
1238  if (zmemcmp(s->window + match,
1239  s->window + start, length) != EQUAL) {
1240  fprintf(stderr, " start %u, match %u, length %d\n",
1241  start, match, length);
1242  do {
1243  fprintf(stderr, "%c%c", s->window[match++], s->window[start++]);
1244  } while (--length != 0);
1245  z_error("invalid match");
1246  }
1247  if (z_verbose > 1) {
1248  fprintf(stderr,"\\[%d,%d]", start-match, length);
1249  do { putc(s->window[start++], stderr); } while (--length != 0);
1250  }
1251 }
1252 #else
1253 # define check_match(s, start, match, length)
1254 #endif /* DEBUG */
1255 
1256 /* ===========================================================================
1257  * Fill the window when the lookahead becomes insufficient.
1258  * Updates strstart and lookahead.
1259  *
1260  * IN assertion: lookahead < MIN_LOOKAHEAD
1261  * OUT assertions: strstart <= window_size-MIN_LOOKAHEAD
1262  * At least one byte has been read, or avail_in == 0; reads are
1263  * performed for at least two bytes (required for the zip translate_eol
1264  * option -- not supported here).
1265  */
1267  deflate_state *s;
1268 {
1269  register unsigned n, m;
1270  register Posf *p;
1271  unsigned more; /* Amount of free space at the end of the window. */
1272  uInt wsize = s->w_size;
1273 
1274  do {
1275  more = (unsigned)(s->window_size -(ulg)s->lookahead -(ulg)s->strstart);
1276 
1277  /* Deal with !@#$% 64K limit: */
1278  if (sizeof(int) <= 2) {
1279  if (more == 0 && s->strstart == 0 && s->lookahead == 0) {
1280  more = wsize;
1281 
1282  } else if (more == (unsigned)(-1)) {
1283  /* Very unlikely, but possible on 16 bit machine if
1284  * strstart == 0 && lookahead == 1 (input done a byte at time)
1285  */
1286  more--;
1287  }
1288  }
1289 
1290  /* If the window is almost full and there is insufficient lookahead,
1291  * move the upper half to the lower one to make room in the upper half.
1292  */
1293  if (s->strstart >= wsize+MAX_DIST(s)) {
1294 
1295  zmemcpy(s->window, s->window+wsize, (unsigned)wsize);
1296  s->match_start -= wsize;
1297  s->strstart -= wsize; /* we now have strstart >= MAX_DIST */
1298  s->block_start -= (long) wsize;
1299 
1300  /* Slide the hash table (could be avoided with 32 bit values
1301  at the expense of memory usage). We slide even when level == 0
1302  to keep the hash table consistent if we switch back to level > 0
1303  later. (Using level 0 permanently is not an optimal usage of
1304  zlib, so we don't care about this pathological case.)
1305  */
1306  /* %%% avoid this when Z_RLE */
1307  n = s->hash_size;
1308  p = &s->head[n];
1309  do {
1310  m = *--p;
1311  *p = (Pos)(m >= wsize ? m-wsize : NIL);
1312  } while (--n);
1313 
1314  n = wsize;
1315 #ifndef FASTEST
1316  p = &s->prev[n];
1317  do {
1318  m = *--p;
1319  *p = (Pos)(m >= wsize ? m-wsize : NIL);
1320  /* If n is not on any hash chain, prev[n] is garbage but
1321  * its value will never be used.
1322  */
1323  } while (--n);
1324 #endif
1325  more += wsize;
1326  }
1327  if (s->strm->avail_in == 0) return;
1328 
1329  /* If there was no sliding:
1330  * strstart <= WSIZE+MAX_DIST-1 && lookahead <= MIN_LOOKAHEAD - 1 &&
1331  * more == window_size - lookahead - strstart
1332  * => more >= window_size - (MIN_LOOKAHEAD-1 + WSIZE + MAX_DIST-1)
1333  * => more >= window_size - 2*WSIZE + 2
1334  * In the BIG_MEM or MMAP case (not yet supported),
1335  * window_size == input_size + MIN_LOOKAHEAD &&
1336  * strstart + s->lookahead <= input_size => more >= MIN_LOOKAHEAD.
1337  * Otherwise, window_size == 2*WSIZE so more >= 2.
1338  * If there was sliding, more >= WSIZE. So in all cases, more >= 2.
1339  */
1340  Assert(more >= 2, "more < 2");
1341 
1342  n = read_buf(s->strm, s->window + s->strstart + s->lookahead, more);
1343  s->lookahead += n;
1344 
1345  /* Initialize the hash value now that we have some input: */
1346  if (s->lookahead >= MIN_MATCH) {
1347  s->ins_h = s->window[s->strstart];
1348  UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]);
1349 #if MIN_MATCH != 3
1350  Call UPDATE_HASH() MIN_MATCH-3 more times
1351 #endif
1352  }
1353  /* If the whole input has less than MIN_MATCH bytes, ins_h is garbage,
1354  * but this is not important since only literal bytes will be emitted.
1355  */
1356 
1357  } while (s->lookahead < MIN_LOOKAHEAD && s->strm->avail_in != 0);
1358 }
1359 
1360 /* ===========================================================================
1361  * Flush the current block, with given end-of-file flag.
1362  * IN assertion: strstart is set to the end of the current match.
1363  */
1364 #define FLUSH_BLOCK_ONLY(s, eof) { \
1365  _tr_flush_block(s, (s->block_start >= 0L ? \
1366  (charf *)&s->window[(unsigned)s->block_start] : \
1367  (charf *)Z_NULL), \
1368  (ulg)((long)s->strstart - s->block_start), \
1369  (eof)); \
1370  s->block_start = s->strstart; \
1371  flush_pending(s->strm); \
1372  Tracev((stderr,"[FLUSH]")); \
1373 }
1374 
1375 /* Same but force premature exit if necessary. */
1376 #define FLUSH_BLOCK(s, eof) { \
1377  FLUSH_BLOCK_ONLY(s, eof); \
1378  if (s->strm->avail_out == 0) return (eof) ? finish_started : need_more; \
1379 }
1380 
1381 /* ===========================================================================
1382  * Copy without compression as much as possible from the input stream, return
1383  * the current block state.
1384  * This function does not insert new strings in the dictionary since
1385  * uncompressible data is probably not useful. This function is used
1386  * only for the level=0 compression option.
1387  * NOTE: this function should be optimized to avoid extra copying from
1388  * window to pending_buf.
1389  */
1391  deflate_state *s;
1392  int flush;
1393 {
1394  /* Stored blocks are limited to 0xffff bytes, pending_buf is limited
1395  * to pending_buf_size, and each stored block has a 5 byte header:
1396  */
1397  ulg max_block_size = 0xffff;
1398  ulg max_start;
1399 
1400  if (max_block_size > s->pending_buf_size - 5) {
1401  max_block_size = s->pending_buf_size - 5;
1402  }
1403 
1404  /* Copy as much as possible from input to output: */
1405  for (;;) {
1406  /* Fill the window as much as possible: */
1407  if (s->lookahead <= 1) {
1408 
1409  Assert(s->strstart < s->w_size+MAX_DIST(s) ||
1410  s->block_start >= (long)s->w_size, "slide too late");
1411 
1412  fill_window(s);
1413  if (s->lookahead == 0 && flush == Z_NO_FLUSH) return need_more;
1414 
1415  if (s->lookahead == 0) break; /* flush the current block */
1416  }
1417  Assert(s->block_start >= 0L, "block gone");
1418 
1419  s->strstart += s->lookahead;
1420  s->lookahead = 0;
1421 
1422  /* Emit a stored block if pending_buf will be full: */
1423  max_start = s->block_start + max_block_size;
1424  if (s->strstart == 0 || (ulg)s->strstart >= max_start) {
1425  /* strstart == 0 is possible when wraparound on 16-bit machine */
1426  s->lookahead = (uInt)(s->strstart - max_start);
1427  s->strstart = (uInt)max_start;
1428  FLUSH_BLOCK(s, 0);
1429  }
1430  /* Flush if we may have to slide, otherwise block_start may become
1431  * negative and the data will be gone:
1432  */
1433  if (s->strstart - (uInt)s->block_start >= MAX_DIST(s)) {
1434  FLUSH_BLOCK(s, 0);
1435  }
1436  }
1437  FLUSH_BLOCK(s, flush == Z_FINISH);
1438  return flush == Z_FINISH ? finish_done : block_done;
1439 }
1440 
1441 /* ===========================================================================
1442  * Compress as much as possible from the input stream, return the current
1443  * block state.
1444  * This function does not perform lazy evaluation of matches and inserts
1445  * new strings in the dictionary only for unmatched strings or for short
1446  * matches. It is used only for the fast compression options.
1447  */
1449  deflate_state *s;
1450  int flush;
1451 {
1452  IPos hash_head = NIL; /* head of the hash chain */
1453  int bflush; /* set if current block must be flushed */
1454 
1455  for (;;) {
1456  /* Make sure that we always have enough lookahead, except
1457  * at the end of the input file. We need MAX_MATCH bytes
1458  * for the next match, plus MIN_MATCH bytes to insert the
1459  * string following the next match.
1460  */
1461  if (s->lookahead < MIN_LOOKAHEAD) {
1462  fill_window(s);
1463  if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) {
1464  return need_more;
1465  }
1466  if (s->lookahead == 0) break; /* flush the current block */
1467  }
1468 
1469  /* Insert the string window[strstart .. strstart+2] in the
1470  * dictionary, and set hash_head to the head of the hash chain:
1471  */
1472  if (s->lookahead >= MIN_MATCH) {
1473  INSERT_STRING(s, s->strstart, hash_head);
1474  }
1475 
1476  /* Find the longest match, discarding those <= prev_length.
1477  * At this point we have always match_length < MIN_MATCH
1478  */
1479  if (hash_head != NIL && s->strstart - hash_head <= MAX_DIST(s)) {
1480  /* To simplify the code, we prevent matches with the string
1481  * of window index 0 (in particular we have to avoid a match
1482  * of the string with itself at the start of the input file).
1483  */
1484 #ifdef FASTEST
1485  if ((s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) ||
1486  (s->strategy == Z_RLE && s->strstart - hash_head == 1)) {
1487  s->match_length = longest_match_fast (s, hash_head);
1488  }
1489 #else
1490  if (s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) {
1491  s->match_length = longest_match (s, hash_head);
1492  } else if (s->strategy == Z_RLE && s->strstart - hash_head == 1) {
1493  s->match_length = longest_match_fast (s, hash_head);
1494  }
1495 #endif
1496  /* longest_match() or longest_match_fast() sets match_start */
1497  }
1498  if (s->match_length >= MIN_MATCH) {
1499  check_match(s, s->strstart, s->match_start, s->match_length);
1500 
1501  _tr_tally_dist(s, s->strstart - s->match_start,
1502  s->match_length - MIN_MATCH, bflush);
1503 
1504  s->lookahead -= s->match_length;
1505 
1506  /* Insert new strings in the hash table only if the match length
1507  * is not too large. This saves time but degrades compression.
1508  */
1509 #ifndef FASTEST
1510  if (s->match_length <= s->max_insert_length &&
1511  s->lookahead >= MIN_MATCH) {
1512  s->match_length--; /* string at strstart already in table */
1513  do {
1514  s->strstart++;
1515  INSERT_STRING(s, s->strstart, hash_head);
1516  /* strstart never exceeds WSIZE-MAX_MATCH, so there are
1517  * always MIN_MATCH bytes ahead.
1518  */
1519  } while (--s->match_length != 0);
1520  s->strstart++;
1521  } else
1522 #endif
1523  {
1524  s->strstart += s->match_length;
1525  s->match_length = 0;
1526  s->ins_h = s->window[s->strstart];
1527  UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]);
1528 #if MIN_MATCH != 3
1529  Call UPDATE_HASH() MIN_MATCH-3 more times
1530 #endif
1531  /* If lookahead < MIN_MATCH, ins_h is garbage, but it does not
1532  * matter since it will be recomputed at next deflate call.
1533  */
1534  }
1535  } else {
1536  /* No match, output a literal byte */
1537  Tracevv((stderr,"%c", s->window[s->strstart]));
1538  _tr_tally_lit (s, s->window[s->strstart], bflush);
1539  s->lookahead--;
1540  s->strstart++;
1541  }
1542  if (bflush) FLUSH_BLOCK(s, 0);
1543  }
1544  FLUSH_BLOCK(s, flush == Z_FINISH);
1545  return flush == Z_FINISH ? finish_done : block_done;
1546 }
1547 
1548 #ifndef FASTEST
1549 /* ===========================================================================
1550  * Same as above, but achieves better compression. We use a lazy
1551  * evaluation for matches: a match is finally adopted only if there is
1552  * no better match at the next window position.
1553  */
1555  deflate_state *s;
1556  int flush;
1557 {
1558  IPos hash_head = NIL; /* head of hash chain */
1559  int bflush; /* set if current block must be flushed */
1560 
1561  /* Process the input block. */
1562  for (;;) {
1563  /* Make sure that we always have enough lookahead, except
1564  * at the end of the input file. We need MAX_MATCH bytes
1565  * for the next match, plus MIN_MATCH bytes to insert the
1566  * string following the next match.
1567  */
1568  if (s->lookahead < MIN_LOOKAHEAD) {
1569  fill_window(s);
1570  if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) {
1571  return need_more;
1572  }
1573  if (s->lookahead == 0) break; /* flush the current block */
1574  }
1575 
1576  /* Insert the string window[strstart .. strstart+2] in the
1577  * dictionary, and set hash_head to the head of the hash chain:
1578  */
1579  if (s->lookahead >= MIN_MATCH) {
1580  INSERT_STRING(s, s->strstart, hash_head);
1581  }
1582 
1583  /* Find the longest match, discarding those <= prev_length.
1584  */
1585  s->prev_length = s->match_length, s->prev_match = s->match_start;
1586  s->match_length = MIN_MATCH-1;
1587 
1588  if (hash_head != NIL && s->prev_length < s->max_lazy_match &&
1589  s->strstart - hash_head <= MAX_DIST(s)) {
1590  /* To simplify the code, we prevent matches with the string
1591  * of window index 0 (in particular we have to avoid a match
1592  * of the string with itself at the start of the input file).
1593  */
1594  if (s->strategy != Z_HUFFMAN_ONLY && s->strategy != Z_RLE) {
1595  s->match_length = longest_match (s, hash_head);
1596  } else if (s->strategy == Z_RLE && s->strstart - hash_head == 1) {
1597  s->match_length = longest_match_fast (s, hash_head);
1598  }
1599  /* longest_match() or longest_match_fast() sets match_start */
1600 
1601  if (s->match_length <= 5 && (s->strategy == Z_FILTERED
1602 #if TOO_FAR <= 32767
1603  || (s->match_length == MIN_MATCH &&
1604  s->strstart - s->match_start > TOO_FAR)
1605 #endif
1606  )) {
1607 
1608  /* If prev_match is also MIN_MATCH, match_start is garbage
1609  * but we will ignore the current match anyway.
1610  */
1611  s->match_length = MIN_MATCH-1;
1612  }
1613  }
1614  /* If there was a match at the previous step and the current
1615  * match is not better, output the previous match:
1616  */
1617  if (s->prev_length >= MIN_MATCH && s->match_length <= s->prev_length) {
1618  uInt max_insert = s->strstart + s->lookahead - MIN_MATCH;
1619  /* Do not insert strings in hash table beyond this. */
1620 
1621  check_match(s, s->strstart-1, s->prev_match, s->prev_length);
1622 
1623  _tr_tally_dist(s, s->strstart -1 - s->prev_match,
1624  s->prev_length - MIN_MATCH, bflush);
1625 
1626  /* Insert in hash table all strings up to the end of the match.
1627  * strstart-1 and strstart are already inserted. If there is not
1628  * enough lookahead, the last two strings are not inserted in
1629  * the hash table.
1630  */
1631  s->lookahead -= s->prev_length-1;
1632  s->prev_length -= 2;
1633  do {
1634  if (++s->strstart <= max_insert) {
1635  INSERT_STRING(s, s->strstart, hash_head);
1636  }
1637  } while (--s->prev_length != 0);
1638  s->match_available = 0;
1639  s->match_length = MIN_MATCH-1;
1640  s->strstart++;
1641 
1642  if (bflush) FLUSH_BLOCK(s, 0);
1643 
1644  } else if (s->match_available) {
1645  /* If there was no match at the previous position, output a
1646  * single literal. If there was a match but the current match
1647  * is longer, truncate the previous match to a single literal.
1648  */
1649  Tracevv((stderr,"%c", s->window[s->strstart-1]));
1650  _tr_tally_lit(s, s->window[s->strstart-1], bflush);
1651  if (bflush) {
1652  FLUSH_BLOCK_ONLY(s, 0);
1653  }
1654  s->strstart++;
1655  s->lookahead--;
1656  if (s->strm->avail_out == 0) return need_more;
1657  } else {
1658  /* There is no previous match to compare with, wait for
1659  * the next step to decide.
1660  */
1661  s->match_available = 1;
1662  s->strstart++;
1663  s->lookahead--;
1664  }
1665  }
1666  Assert (flush != Z_NO_FLUSH, "no flush?");
1667  if (s->match_available) {
1668  Tracevv((stderr,"%c", s->window[s->strstart-1]));
1669  _tr_tally_lit(s, s->window[s->strstart-1], bflush);
1670  s->match_available = 0;
1671  }
1672  FLUSH_BLOCK(s, flush == Z_FINISH);
1673  return flush == Z_FINISH ? finish_done : block_done;
1674 }
1675 #endif /* FASTEST */
1676 
1677 #if 0
1678 /* ===========================================================================
1679  * For Z_RLE, simply look for runs of bytes, generate matches only of distance
1680  * one. Do not maintain a hash table. (It will be regenerated if this run of
1681  * deflate switches away from Z_RLE.)
1682  */
1683 local block_state deflate_rle(s, flush)
1684  deflate_state *s;
1685  int flush;
1686 {
1687  int bflush; /* set if current block must be flushed */
1688  uInt run; /* length of run */
1689  uInt max; /* maximum length of run */
1690  uInt prev; /* byte at distance one to match */
1691  Bytef *scan; /* scan for end of run */
1692 
1693  for (;;) {
1694  /* Make sure that we always have enough lookahead, except
1695  * at the end of the input file. We need MAX_MATCH bytes
1696  * for the longest encodable run.
1697  */
1698  if (s->lookahead < MAX_MATCH) {
1699  fill_window(s);
1700  if (s->lookahead < MAX_MATCH && flush == Z_NO_FLUSH) {
1701  return need_more;
1702  }
1703  if (s->lookahead == 0) break; /* flush the current block */
1704  }
1705 
1706  /* See how many times the previous byte repeats */
1707  run = 0;
1708  if (s->strstart > 0) { /* if there is a previous byte, that is */
1709  max = s->lookahead < MAX_MATCH ? s->lookahead : MAX_MATCH;
1710  scan = s->window + s->strstart - 1;
1711  prev = *scan++;
1712  do {
1713  if (*scan++ != prev)
1714  break;
1715  } while (++run < max);
1716  }
1717 
1718  /* Emit match if have run of MIN_MATCH or longer, else emit literal */
1719  if (run >= MIN_MATCH) {
1720  check_match(s, s->strstart, s->strstart - 1, run);
1721  _tr_tally_dist(s, 1, run - MIN_MATCH, bflush);
1722  s->lookahead -= run;
1723  s->strstart += run;
1724  } else {
1725  /* No match, output a literal byte */
1726  Tracevv((stderr,"%c", s->window[s->strstart]));
1727  _tr_tally_lit (s, s->window[s->strstart], bflush);
1728  s->lookahead--;
1729  s->strstart++;
1730  }
1731  if (bflush) FLUSH_BLOCK(s, 0);
1732  }
1733  FLUSH_BLOCK(s, flush == Z_FINISH);
1734  return flush == Z_FINISH ? finish_done : block_done;
1735 }
1736 #endif
uInt good_match
Definition: deflate.h:185
#define INIT_STATE
Definition: deflate.h:51
uInt hash_bits
Definition: deflate.h:138
uLong ZEXPORT compressBound(uLong sourceLen)
Definition: compress.c:75
int ZEXPORT deflateParams(z_streamp strm, int level, int strategy)
Definition: deflate.c:416
void zcfree(voidpf opaque, voidpf ptr)
Definition: zutil.c:310
Byte FAR * voidpf
Definition: zconf.h:283
ush FAR ushf
Definition: zutil.h:50
struct config_s config
#define INSERT_STRING(s, str, match_head)
Definition: deflate.c:189
#define check_match(s, start, match, length)
Definition: deflate.c:1253
local void flush_pending(z_streamp strm)
Definition: deflate.c:532
Bytef * window
Definition: deflate.h:113
#define HCRC_STATE
Definition: deflate.h:55
#define TOO_FAR
Definition: deflate.c:109
unsigned char Byte
Definition: zconf.h:261
void zmemcpy(Bytef *dest, const Bytef *source, uInt len)
Definition: zutil.c:149
#define Z_NO_FLUSH
Definition: zlib.h:162
#define MIN_MATCH
Definition: zutil.h:81
uInt max_lazy_match
Definition: deflate.h:171
#define PRESET_DICT
Definition: zutil.h:85
uInt hash_mask
Definition: deflate.h:139
#define FLUSH_BLOCK(s, eof)
Definition: deflate.c:1376
#define Z_UNKNOWN
Definition: zlib.h:199
block_state
Definition: deflate.c:66
#define Z_PARTIAL_FLUSH
Definition: zlib.h:163
ush Pos
Definition: deflate.h:86
uInt hash_shift
Definition: deflate.h:141
#define Assert(cond, msg)
Definition: zutil.h:252
png_uint_32 size
Definition: png.h:1521
int version
Definition: jpeglib.h:901
local uInt longest_match(deflate_state *s, IPos cur_match)
Definition: deflate.c:1027
png_voidp int value
Definition: png.h:2113
void _tr_init(deflate_state *s)
Definition: trees.c:382
#define UPDATE_HASH(s, h, c)
Definition: deflate.c:170
#define MAX_MEM_LEVEL
Definition: zconf.h:138
Posf * prev
Definition: deflate.h:128
ush good_length
Definition: deflate.c:124
struct ct_data_s dyn_ltree[HEAP_SIZE]
Definition: deflate.h:192
unsigned long uLong
Definition: zconf.h:264
int ZEXPORT deflateInit2_(z_streamp strm, int level, int method, int windowBits, int memLevel, int strategy, const char *version, int stream_size)
Definition: deflate.c:216
ct_data * dyn_tree
Definition: deflate.h:81
Bytef * pending_out
Definition: deflate.h:99
uLong ZEXPORT deflateBound(z_streamp strm, uLong sourceLen)
Definition: deflate.c:489
int ZEXPORT deflateEnd(z_streamp strm)
Definition: deflate.c:859
#define Z_FILTERED
Definition: zlib.h:189
int ZEXPORT deflatePrime(z_streamp strm, int bits, int value)
Definition: deflate.c:404
gz_header FAR * gz_headerp
Definition: zlib.h:126
#define Z_STREAM_ERROR
Definition: zlib.h:174
const char deflate_copyright[]
Definition: deflate.c:54
png_bytep png_bytep png_size_t length
Definition: png.h:1541
local int read_buf(z_streamp strm, Bytef *buf, unsigned size)
Definition: deflate.c:956
gz_headerp gzhead
Definition: deflate.h:102
int zmemcmp(Bytef *s1, const Bytef *s2, uInt len) const
Definition: zutil.c:160
#define Z_FIXED
Definition: zlib.h:192
Byte FAR Bytef
Definition: zconf.h:270
long b
Definition: jpegint.h:371
uInt lookahead
Definition: deflate.h:158
#define _tr_tally_lit(s, c, flush)
Definition: deflate.h:308
#define MAX_DIST(s)
Definition: deflate.h:276
#define ERR_MSG(err)
Definition: zutil.h:56
#define BUSY_STATE
Definition: deflate.h:56
#define ZALLOC(strm, items, size)
Definition: zutil.h:264
ulg window_size
Definition: deflate.h:123
void _tr_stored_block(deflate_state *s, charf *buf, ulg stored_len, int eof)
Definition: trees.c:867
#define DEF_MEM_LEVEL
Definition: zutil.h:70
#define Z_FINISH
Definition: zlib.h:166
local block_state deflate_stored(deflate_state *s, int flush)
Definition: deflate.c:1390
z_streamp strm
Definition: deflate.h:95
unsigned long ZEXPORT crc32(unsigned long crc, const unsigned char FAR *buf, unsigned len)
Definition: crc32.c:219
struct tree_desc_s bl_desc
Definition: deflate.h:198
struct tree_desc_s d_desc
Definition: deflate.h:197
Posf * head
Definition: deflate.h:134
voidpf zcalloc(voidpf opaque, unsigned items, unsigned size)
Definition: zutil.c:300
int ZEXPORT deflateCopy(z_streamp dest, z_streamp source)
Definition: deflate.c:894
int ZEXPORT deflateTune(z_streamp strm, int good_length, int max_lazy, int nice_length, int max_chain)
Definition: deflate.c:454
void _tr_align(deflate_state *s)
Definition: trees.c:892
#define NIL
Definition: deflate.c:105
int method
Definition: png.h:1847
#define Z_DEFLATED
Definition: zlib.h:202
int strategy
Definition: png.h:1932
#define Z_DATA_ERROR
Definition: zlib.h:175
compress_func func
Definition: deflate.c:128
local uInt longest_match_fast(deflate_state *s, IPos cur_match)
Definition: deflate.c:1175
unsigned short ush
Definition: zutil.h:49
ush max_chain
Definition: deflate.c:127
#define ZFREE(strm, addr)
Definition: zutil.h:266
local block_state deflate_fast(deflate_state *s, int flush)
Definition: deflate.c:1448
#define Z_STREAM_END
Definition: zlib.h:171
#define EQUAL
Definition: deflate.c:157
#define MAX_MATCH
Definition: zutil.h:82
uchf * l_buf
Definition: deflate.h:214
Pos FAR Posf
Definition: deflate.h:87
def run(tree, args)
int val
Definition: jpeglib.h:956
block_state compress_func OF((deflate_state *s, int flush))
Definition: deflate.c:73
#define put_byte(s, c)
Definition: deflate.h:268
#define Z_MEM_ERROR
Definition: zlib.h:176
#define COMMENT_STATE
Definition: deflate.h:54
Bytef * pending_buf
Definition: deflate.h:97
png_bytep buf
Definition: png.h:2729
uInt lit_bufsize
Definition: deflate.h:216
#define CLEAR_HASH(s)
Definition: deflate.c:199
local const config configuration_table[10]
Definition: deflate.c:137
#define NAME_STATE
Definition: deflate.h:53
#define _tr_tally_dist(s, distance, length, flush)
Definition: deflate.h:315
#define TRY_FREE(s, p)
Definition: zutil.h:267
png_size_t start
Definition: png.h:1496
int ZEXPORT deflate(z_streamp strm, int flush)
Definition: deflate.c:552
ushf * d_buf
Definition: deflate.h:238
#define FLUSH_BLOCK_ONLY(s, eof)
Definition: deflate.c:1364
long block_start
Definition: deflate.h:148
ush max_lazy
Definition: deflate.c:125
#define ZLIB_VERSION
Definition: zlib.h:40
struct ct_data_s dyn_dtree[2 *D_CODES+1]
Definition: deflate.h:193
uInt gzindex
Definition: deflate.h:103
local void fill_window(deflate_state *s)
Definition: deflate.c:1266
struct tree_desc_s l_desc
Definition: deflate.h:196
#define FINISH_STATE
Definition: deflate.h:57
#define Z_HUFFMAN_ONLY
Definition: zlib.h:190
#define MIN_LOOKAHEAD
Definition: deflate.c:113
uLong ZEXPORT adler32(uLong adler, const Bytef *buf, uInt len)
Definition: adler32.c:57
typedef int
Definition: png.h:1113
uInt max_chain_length
Definition: deflate.h:165
#define Z_VERSION_ERROR
Definition: zlib.h:178
int nice_match
Definition: deflate.h:188
local void putShortMSB(deflate_state *s, uInt b)
Definition: deflate.c:518
int level
Definition: png.h:1926
#define Z_BUF_ERROR
Definition: zlib.h:177
uInt pending
Definition: deflate.h:100
unsigned long ulg
Definition: zutil.h:51
int ZEXPORT deflateSetDictionary(z_streamp strm, const Bytef *dictionary, uInt dictLength)
Definition: deflate.c:315
uInt strstart
Definition: deflate.h:156
#define ERR_RETURN(strm, err)
Definition: zutil.h:58
int ZEXPORT deflateSetHeader(z_streamp strm, gz_headerp head)
Definition: deflate.c:393
ulg pending_buf_size
Definition: deflate.h:98
#define Z_OK
Definition: zlib.h:170
local void lm_init(deflate_state *s)
Definition: deflate.c:986
#define EXTRA_STATE
Definition: deflate.h:52
#define local
Definition: crc32.c:31
int ZEXPORT deflateInit_(z_streamp strm, int level, const char *version, int stream_size)
Definition: deflate.c:204
local block_state deflate_slow(deflate_state *s, int flush)
Definition: deflate.c:1554
#define Z_DEFAULT_STRATEGY
Definition: zlib.h:193
#define Z_NULL
Definition: zlib.h:205
#define Z_FULL_FLUSH
Definition: zlib.h:165
#define const
Definition: zconf.h:124
z_stream FAR * z_streamp
Definition: zlib.h:103
int ZEXPORT deflateReset(z_streamp strm)
Definition: deflate.c:357
int last_flush
Definition: deflate.h:105
uch FAR uchf
Definition: zutil.h:48
#define MAX_WBITS
Definition: zconf.h:148
#define ZEXPORT
Definition: zconf.h:250
ush nice_length
Definition: deflate.c:126
struct ct_data_s bl_tree[2 *BL_CODES+1]
Definition: deflate.h:194
static int max(int a, int b)
uInt hash_size
Definition: deflate.h:137
#define OS_CODE
Definition: zutil.h:169
#define Z_DEFAULT_COMPRESSION
Definition: zlib.h:186
unsigned int uInt
Definition: zconf.h:263
const char dictionary[]
#define Tracevv(x)
Definition: zutil.h:255
unsigned IPos
Definition: deflate.h:88
#define Z_RLE
Definition: zlib.h:191


openhrp3
Author(s): AIST, General Robotix Inc., Nakamura Lab of Dept. of Mechano Informatics at University of Tokyo
autogenerated on Sat Apr 13 2019 02:14:21