libkern/zlib/deflate.h

   1 /*
   2  * Copyright (c) 2008 Apple Inc. All rights reserved.
   3  *
   4  * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
   5  *
   6  * This file contains Original Code and/or Modifications of Original Code
   7  * as defined in and that are subject to the Apple Public Source License
   8  * Version 2.0 (the 'License'). You may not use this file except in
   9  * compliance with the License. The rights granted to you under the License
  10  * may not be used to create, or enable the creation or redistribution of,
  11  * unlawful or unlicensed copies of an Apple operating system, or to
  12  * circumvent, violate, or enable the circumvention or violation of, any
  13  * terms of an Apple operating system software license agreement.
  14  *
  15  * Please obtain a copy of the License at
  16  * http://www.opensource.apple.com/apsl/ and read it before using this file.
  17  *
  18  * The Original Code and all software distributed under the License are
  19  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  20  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  21  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  22  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  23  * Please see the License for the specific language governing rights and
  24  * limitations under the License.
  25  *
  26  * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
  27  */
  28 /* deflate.h -- internal compression state
  29  * Copyright (C) 1995-2004 Jean-loup Gailly
  30  * For conditions of distribution and use, see copyright notice in zlib.h
  31  */
  32
  33 /* WARNING: this file should *not* be used by applications. It is
  34    part of the implementation of the compression library and is
  35    subject to change. Applications should only use zlib.h.
  36  */
  37
  38 /* @(#) $Id$ */
  39
  40 #ifndef DEFLATE_H
  41 #define DEFLATE_H
  42
  43 #include "zutil.h"
  44
  45 /* define NO_GZIP when compiling if you want to disable gzip header and
  46    trailer creation by deflate().  NO_GZIP would be used to avoid linking in
  47    the crc code when it is not needed.  For shared libraries, gzip encoding
  48    should be left enabled. */
  49 #ifndef NO_GZIP
  50 #  define GZIP
  51 #endif
  52
  53 /* ===========================================================================
  54  * Internal compression state.
  55  */
  56
  57 #define LENGTH_CODES 29
  58 /* number of length codes, not counting the special END_BLOCK code */
  59
  60 #define LITERALS  256
  61 /* number of literal bytes 0..255 */
  62
  63 #define L_CODES (LITERALS+1+LENGTH_CODES)
  64 /* number of Literal or Length codes, including the END_BLOCK code */
  65
  66 #define D_CODES   30
  67 /* number of distance codes */
  68
  69 #define BL_CODES  19
  70 /* number of codes used to transfer the bit lengths */
  71
  72 #define HEAP_SIZE (2*L_CODES+1)
  73 /* maximum heap size */
  74
  75 #define MAX_BITS 15
  76 /* All codes must not exceed MAX_BITS bits */
  77
  78 #define INIT_STATE    42
  79 #define EXTRA_STATE   69
  80 #define NAME_STATE    73
  81 #define COMMENT_STATE 91
  82 #define HCRC_STATE   103
  83 #define BUSY_STATE   113
  84 #define FINISH_STATE 666
  85 /* Stream status */
  86
  87
  88 /* Data structure describing a single value and its code string. */
  89 typedef struct ct_data_s {
  90     union {
  91         ush  freq;       /* frequency count */
  92         ush  code;       /* bit string */
  93     } fc;
  94     union {
  95         ush  dad;        /* father node in Huffman tree */
  96         ush  len;        /* length of bit string */
  97     } dl;
  98 } FAR ct_data;
  99
 100 #define Freq fc.freq
 101 #define Code fc.code
 102 #define Dad  dl.dad
 103 #define Len  dl.len
 104
 105 typedef struct static_tree_desc_s  static_tree_desc;
 106
 107 typedef struct tree_desc_s {
 108     ct_data *dyn_tree;           /* the dynamic tree */
 109     int     max_code;            /* largest code with non zero frequency */
 110     static_tree_desc *stat_desc; /* the corresponding static tree */
 111 } FAR tree_desc;
 112
 113 typedef ush Pos;
 114 typedef Pos FAR Posf;
 115 typedef unsigned IPos;
 116
 117 /* A Pos is an index in the character window. We use short instead of int to
 118  * save space in the various tables. IPos is used only for parameter passing.
 119  */
 120
 121 typedef struct internal_state {
 122     z_streamp strm;      /* pointer back to this zlib stream */
 123     int   status;        /* as the name implies */
 124     Bytef *pending_buf;  /* output still pending */
 125     ulg   pending_buf_size; /* size of pending_buf */
 126     Bytef *pending_out;  /* next pending byte to output to the stream */
 127     uInt   pending;      /* nb of bytes in the pending buffer */
 128     int   wrap;          /* bit 0 true for zlib, bit 1 true for gzip */
 129     gz_headerp  gzhead;  /* gzip header information to write */
 130     uInt   gzindex;      /* where in extra, name, or comment */
 131     Byte  method;        /* STORED (for zip only) or DEFLATED */
 132     int   last_flush;    /* value of flush param for previous deflate call */
 133
 134                 /* used by deflate.c: */
 135
 136     uInt  w_size;        /* LZ77 window size (32K by default) */
 137     uInt  w_bits;        /* log2(w_size)  (8..16) */
 138     uInt  w_mask;        /* w_size - 1 */
 139
 140     Bytef *window;
 141     /* Sliding window. Input bytes are read into the second half of the window,
 142      * and move to the first half later to keep a dictionary of at least wSize
 143      * bytes. With this organization, matches are limited to a distance of
 144      * wSize-MAX_MATCH bytes, but this ensures that IO is always
 145      * performed with a length multiple of the block size. Also, it limits
 146      * the window size to 64K, which is quite useful on MSDOS.
 147      * To do: use the user input buffer as sliding window.
 148      */
 149
 150     ulg window_size;
 151     /* Actual size of window: 2*wSize, except when the user input buffer
 152      * is directly used as sliding window.
 153      */
 154
 155     Posf *prev;
 156     /* Link to older string with same hash index. To limit the size of this
 157      * array to 64K, this link is maintained only for the last 32K strings.
 158      * An index in this array is thus a window index modulo 32K.
 159      */
 160
 161     Posf *head; /* Heads of the hash chains or NIL. */
 162
 163     uInt  ins_h;          /* hash index of string to be inserted */
 164     uInt  hash_size;      /* number of elements in hash table */
 165     uInt  hash_bits;      /* log2(hash_size) */
 166     uInt  hash_mask;      /* hash_size-1 */
 167
 168     uInt  hash_shift;
 169     /* Number of bits by which ins_h must be shifted at each input
 170      * step. It must be such that after MIN_MATCH steps, the oldest
 171      * byte no longer takes part in the hash key, that is:
 172      *   hash_shift * MIN_MATCH >= hash_bits
 173      */
 174
 175     long block_start;
 176     /* Window position at the beginning of the current output block. Gets
 177      * negative when the window is moved backwards.
 178      */
 179
 180     uInt match_length;           /* length of best match */
 181     IPos prev_match;             /* previous match */
 182     int match_available;         /* set if previous match exists */
 183     uInt strstart;               /* start of string to insert */
 184     uInt match_start;            /* start of matching string */
 185     uInt lookahead;              /* number of valid bytes ahead in window */
 186
 187     uInt prev_length;
 188     /* Length of the best match at previous step. Matches not greater than this
 189      * are discarded. This is used in the lazy match evaluation.
 190      */
 191
 192     uInt max_chain_length;
 193     /* To speed up deflation, hash chains are never searched beyond this
 194      * length.  A higher limit improves compression ratio but degrades the
 195      * speed.
 196      */
 197
 198     uInt max_lazy_match;
 199     /* Attempt to find a better match only when the current match is strictly
 200      * smaller than this value. This mechanism is used only for compression
 201      * levels >= 4.
 202      */
 203 #   define max_insert_length  max_lazy_match
 204     /* Insert new strings in the hash table only if the match length is not
 205      * greater than this length. This saves time but degrades compression.
 206      * max_insert_length is used only for compression levels <= 3.
 207      */
 208
 209     int level;    /* compression level (1..9) */
 210     int strategy; /* favor or force Huffman coding*/
 211
 212     uInt good_match;
 213     /* Use a faster search when the previous match is longer than this */
 214
 215     int nice_match; /* Stop searching when current match exceeds this */
 216
 217                 /* used by trees.c: */
 218     /* Didn't use ct_data typedef below to supress compiler warning */
 219     struct ct_data_s dyn_ltree[HEAP_SIZE];   /* literal and length tree */
 220     struct ct_data_s dyn_dtree[2*D_CODES+1]; /* distance tree */
 221     struct ct_data_s bl_tree[2*BL_CODES+1];  /* Huffman tree for bit lengths */
 222
 223     struct tree_desc_s l_desc;               /* desc. for literal tree */
 224     struct tree_desc_s d_desc;               /* desc. for distance tree */
 225     struct tree_desc_s bl_desc;              /* desc. for bit length tree */
 226
 227     ush bl_count[MAX_BITS+1];
 228     /* number of codes at each bit length for an optimal tree */
 229
 230     int heap[2*L_CODES+1];      /* heap used to build the Huffman trees */
 231     int heap_len;               /* number of elements in the heap */
 232     int heap_max;               /* element of largest frequency */
 233     /* The sons of heap[n] are heap[2*n] and heap[2*n+1]. heap[0] is not used.
 234      * The same heap array is used to build all trees.
 235      */
 236
 237     uch depth[2*L_CODES+1];
 238     /* Depth of each subtree used as tie breaker for trees of equal frequency
 239      */
 240
 241     uchf *l_buf;          /* buffer for literals or lengths */
 242
 243     uInt  lit_bufsize;
 244     /* Size of match buffer for literals/lengths.  There are 4 reasons for
 245      * limiting lit_bufsize to 64K:
 246      *   - frequencies can be kept in 16 bit counters
 247      *   - if compression is not successful for the first block, all input
 248      *     data is still in the window so we can still emit a stored block even
 249      *     when input comes from standard input.  (This can also be done for
 250      *     all blocks if lit_bufsize is not greater than 32K.)
 251      *   - if compression is not successful for a file smaller than 64K, we can
 252      *     even emit a stored file instead of a stored block (saving 5 bytes).
 253      *     This is applicable only for zip (not gzip or zlib).
 254      *   - creating new Huffman trees less frequently may not provide fast
 255      *     adaptation to changes in the input data statistics. (Take for
 256      *     example a binary file with poorly compressible code followed by
 257      *     a highly compressible string table.) Smaller buffer sizes give
 258      *     fast adaptation but have of course the overhead of transmitting
 259      *     trees more frequently.
 260      *   - I can't count above 4
 261      */
 262
 263     uInt last_lit;      /* running index in l_buf */
 264
 265     ushf *d_buf;
 266     /* Buffer for distances. To simplify the code, d_buf and l_buf have
 267      * the same number of elements. To use different lengths, an extra flag
 268      * array would be necessary.
 269      */
 270
 271     ulg opt_len;        /* bit length of current block with optimal trees */
 272     ulg static_len;     /* bit length of current block with static trees */
 273     uInt matches;       /* number of string matches in current block */
 274     int last_eob_len;   /* bit length of EOB code for last block */
 275
 276 #ifdef DEBUG
 277     ulg compressed_len; /* total bit length of compressed file mod 2^32 */
 278     ulg bits_sent;      /* bit length of compressed data sent mod 2^32 */
 279 #endif
 280
 281     ush bi_buf;
 282     /* Output buffer. bits are inserted starting at the bottom (least
 283      * significant bits).
 284      */
 285     int bi_valid;
 286     /* Number of valid bits in bi_buf.  All bits above the last valid bit
 287      * are always zero.
 288      */
 289
 290 } FAR deflate_state;
 291
 292 /* Output a byte on the stream.
 293  * IN assertion: there is enough room in pending_buf.
 294  */
 295 #define put_byte(s, c) {s->pending_buf[s->pending++] = (c);}
 296
 297
 298 #define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1)
 299 /* Minimum amount of lookahead, except at the end of the input file.
 300  * See deflate.c for comments about the MIN_MATCH+1.
 301  */
 302
 303 #define MAX_DIST(s)  ((s)->w_size-MIN_LOOKAHEAD)
 304 /* In order to simplify the code, particularly on 16 bit machines, match
 305  * distances are limited to MAX_DIST instead of WSIZE.
 306  */
 307
 308         /* in trees.c */
 309 void _tr_init         OF((deflate_state *s));
 310 int  _tr_tally        OF((deflate_state *s, unsigned dist, unsigned lc));
 311 void _tr_flush_block  OF((deflate_state *s, charf *buf, ulg stored_len,
 312                           int eof));
 313 void _tr_align        OF((deflate_state *s));
 314 void _tr_stored_block OF((deflate_state *s, charf *buf, ulg stored_len,
 315                           int eof));
 316
 317 #define d_code(dist) \
 318    ((dist) < 256 ? _dist_code[dist] : _dist_code[256+((dist)>>7)])
 319 /* Mapping from a distance to a distance code. dist is the distance - 1 and
 320  * must not have side effects. _dist_code[256] and _dist_code[257] are never
 321  * used.
 322  */
 323
 324 #ifndef DEBUG
 325 /* Inline versions of _tr_tally for speed: */
 326
 327 #if defined(GEN_TREES_H) || !defined(STDC)
 328   extern uch _length_code[];
 329   extern uch _dist_code[];
 330 #else
 331   extern const uch _length_code[];
 332   extern const uch _dist_code[];
 333 #endif
 334
 335 # define _tr_tally_lit(s, c, flush) \
 336   { uch cc = (c); \
 337     s->d_buf[s->last_lit] = 0; \
 338     s->l_buf[s->last_lit++] = cc; \
 339     s->dyn_ltree[cc].Freq++; \
 340     flush = (s->last_lit == s->lit_bufsize-1); \
 341    }
 342 # define _tr_tally_dist(s, distance, length, flush) \
 343   { uch len = (length); \
 344     ush dist = (distance); \
 345     s->d_buf[s->last_lit] = dist; \
 346     s->l_buf[s->last_lit++] = len; \
 347     dist--; \
 348     s->dyn_ltree[_length_code[len]+LITERALS+1].Freq++; \
 349     s->dyn_dtree[d_code(dist)].Freq++; \
 350     flush = (s->last_lit == s->lit_bufsize-1); \
 351   }
 352 #else
 353 # define _tr_tally_lit(s, c, flush) flush = _tr_tally(s, 0, c)
 354 # define _tr_tally_dist(s, distance, length, flush) \
 355               flush = _tr_tally(s, distance, length)
 356 #endif
 357
 358 #endif /* DEFLATE_H */