diff options
Diffstat (limited to 'src/Common/zlib/deflate.c')
-rw-r--r-- | src/Common/zlib/deflate.c | 108 |
1 files changed, 78 insertions, 30 deletions
diff --git a/src/Common/zlib/deflate.c b/src/Common/zlib/deflate.c index 1ec76144..799fb93c 100644 --- a/src/Common/zlib/deflate.c +++ b/src/Common/zlib/deflate.c @@ -1,90 +1,90 @@ /* deflate.c -- compress data using the deflation algorithm - * Copyright (C) 1995-2017 Jean-loup Gailly and Mark Adler + * Copyright (C) 1995-2022 Jean-loup Gailly and Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h */ /* * ALGORITHM * * The "deflation" process depends on being able to identify portions * of the input text which are identical to earlier input (within a * sliding window trailing behind the input currently being processed). * * The most straightforward technique turns out to be the fastest for * most input files: try all possible matches and select the longest. * The key feature of this algorithm is that insertions into the string * dictionary are very simple and thus fast, and deletions are avoided * completely. Insertions are performed at each input character, whereas * string matches are performed only when the previous match ends. So it * is preferable to spend more time in matches to allow very fast string * insertions and avoid deletions. The matching algorithm for small * strings is inspired from that of Rabin & Karp. A brute force approach * is used to find longer strings when a small match has been found. * A similar algorithm is used in comic (by Jan-Mark Wams) and freeze * (by Leonid Broukhis). * A previous version of this file used a more sophisticated algorithm * (by Fiala and Greene) which is guaranteed to run in linear amortized * time, but has a larger average cost, uses more memory and is patented. * However the F&G algorithm may be faster for some highly redundant * files if the parameter max_chain_length (described below) is too large. * * ACKNOWLEDGEMENTS * * The idea of lazy evaluation of matches is due to Jan-Mark Wams, and * I found it in 'freeze' written by Leonid Broukhis. * Thanks to many people for bug reports and testing. * * REFERENCES * * Deutsch, L.P.,"DEFLATE Compressed Data Format Specification". * Available in http://tools.ietf.org/html/rfc1951 * * A description of the Rabin and Karp algorithm is given in the book * "Algorithms" by R. Sedgewick, Addison-Wesley, p252. * * Fiala,E.R., and Greene,D.H. * Data Compression with Finite Windows, Comm.ACM, 32,4 (1989) 490-595 * */ /* @(#) $Id$ */ #include "deflate.h" const char deflate_copyright[] = - " deflate 1.2.11 Copyright 1995-2017 Jean-loup Gailly and Mark Adler "; + " deflate 1.2.12 Copyright 1995-2022 Jean-loup Gailly and Mark Adler "; /* If you use the zlib library in a product, an acknowledgment is welcome in the documentation of your product. If for some reason you cannot include such an acknowledgment, I would appreciate that you keep this copyright string in the executable of your product. */ /* =========================================================================== * Function prototypes. */ typedef enum { need_more, /* block not completed, need more input or more output */ block_done, /* block flush performed */ finish_started, /* finish started, need only more output at next deflate */ finish_done /* finish done, accept no more input or output */ } block_state; typedef block_state (*compress_func) OF((deflate_state *s, int flush)); /* Compression function. Returns the block state after the call. */ local int deflateStateCheck OF((z_streamp strm)); local void slide_hash OF((deflate_state *s)); local void fill_window OF((deflate_state *s)); local block_state deflate_stored OF((deflate_state *s, int flush)); local block_state deflate_fast OF((deflate_state *s, int flush)); #ifndef FASTEST local block_state deflate_slow OF((deflate_state *s, int flush)); #endif local block_state deflate_rle OF((deflate_state *s, int flush)); local block_state deflate_huff OF((deflate_state *s, int flush)); local void lm_init OF((deflate_state *s)); local void putShortMSB OF((deflate_state *s, uInt b)); local void flush_pending OF((z_streamp strm)); local unsigned read_buf OF((z_streamp strm, Bytef *buf, unsigned size)); #ifdef ASMV @@ -158,219 +158,259 @@ local const config configuration_table[10] = { * Update a hash value with the given input byte * IN assertion: all calls to UPDATE_HASH are made with consecutive input * characters, so that a running hash key can be computed from the previous * key instead of complete recalculation each time. */ #define UPDATE_HASH(s,h,c) (h = (((h)<<s->hash_shift) ^ (c)) & s->hash_mask) /* =========================================================================== * Insert string str in the dictionary and set match_head to the previous head * of the hash chain (the most recent string with same hash key). Return * the previous length of the hash chain. * If this file is compiled with -DFASTEST, the compression level is forced * to 1, and no hash chains are maintained. * IN assertion: all calls to INSERT_STRING are made with consecutive input * characters and the first MIN_MATCH bytes of str are valid (except for * the last MIN_MATCH-1 bytes of the input file). */ #ifdef FASTEST #define INSERT_STRING(s, str, match_head) \ (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \ match_head = s->head[s->ins_h], \ s->head[s->ins_h] = (Pos)(str)) #else #define INSERT_STRING(s, str, match_head) \ (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \ match_head = s->prev[(str) & s->w_mask] = s->head[s->ins_h], \ s->head[s->ins_h] = (Pos)(str)) #endif /* =========================================================================== * Initialize the hash table (avoiding 64K overflow for 16 bit systems). * prev[] will be initialized on the fly. */ #define CLEAR_HASH(s) \ - s->head[s->hash_size-1] = NIL; \ - zmemzero((Bytef *)s->head, (unsigned)(s->hash_size-1)*sizeof(*s->head)); + do { \ + s->head[s->hash_size-1] = NIL; \ + zmemzero((Bytef *)s->head, \ + (unsigned)(s->hash_size-1)*sizeof(*s->head)); \ + } while (0) /* =========================================================================== * Slide the hash table when sliding the window down (could be avoided with 32 * bit values at the expense of memory usage). We slide even when level == 0 to * keep the hash table consistent if we switch back to level > 0 later. */ local void slide_hash(s) deflate_state *s; { unsigned n, m; Posf *p; uInt wsize = s->w_size; n = s->hash_size; p = &s->head[n]; do { m = *--p; *p = (Pos)(m >= wsize ? m - wsize : NIL); } while (--n); n = wsize; #ifndef FASTEST p = &s->prev[n]; do { m = *--p; *p = (Pos)(m >= wsize ? m - wsize : NIL); /* If n is not on any hash chain, prev[n] is garbage but * its value will never be used. */ } while (--n); #endif } /* ========================================================================= */ int ZEXPORT deflateInit_(strm, level, version, stream_size) z_streamp strm; int level; const char *version; int stream_size; { return deflateInit2_(strm, level, Z_DEFLATED, MAX_WBITS, DEF_MEM_LEVEL, Z_DEFAULT_STRATEGY, version, stream_size); /* To do: ignore strm->next_in if we use it as window */ } /* ========================================================================= */ int ZEXPORT deflateInit2_(strm, level, method, windowBits, memLevel, strategy, version, stream_size) z_streamp strm; int level; int method; int windowBits; int memLevel; int strategy; const char *version; int stream_size; { deflate_state *s; int wrap = 1; static const char my_version[] = ZLIB_VERSION; - ushf *overlay; - /* We overlay pending_buf and d_buf+l_buf. This works since the average - * output size for (length,distance) codes is <= 24 bits. - */ - if (version == Z_NULL || version[0] != my_version[0] || stream_size != sizeof(z_stream)) { return Z_VERSION_ERROR; } if (strm == Z_NULL) return Z_STREAM_ERROR; strm->msg = Z_NULL; if (strm->zalloc == (alloc_func)0) { #ifdef Z_SOLO return Z_STREAM_ERROR; #else strm->zalloc = zcalloc; strm->opaque = (voidpf)0; #endif } if (strm->zfree == (free_func)0) #ifdef Z_SOLO return Z_STREAM_ERROR; #else strm->zfree = zcfree; #endif #ifdef FASTEST if (level != 0) level = 1; #else if (level == Z_DEFAULT_COMPRESSION) level = 6; #endif if (windowBits < 0) { /* suppress zlib wrapper */ wrap = 0; windowBits = -windowBits; } #ifdef GZIP else if (windowBits > 15) { wrap = 2; /* write gzip wrapper instead */ windowBits -= 16; } #endif if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || method != Z_DEFLATED || windowBits < 8 || windowBits > 15 || level < 0 || level > 9 || strategy < 0 || strategy > Z_FIXED || (windowBits == 8 && wrap != 1)) { return Z_STREAM_ERROR; } if (windowBits == 8) windowBits = 9; /* until 256-byte window bug fixed */ s = (deflate_state *) ZALLOC(strm, 1, sizeof(deflate_state)); if (s == Z_NULL) return Z_MEM_ERROR; strm->state = (struct internal_state FAR *)s; s->strm = strm; s->status = INIT_STATE; /* to pass state test in deflateReset() */ s->wrap = wrap; s->gzhead = Z_NULL; s->w_bits = (uInt)windowBits; s->w_size = 1 << s->w_bits; s->w_mask = s->w_size - 1; s->hash_bits = (uInt)memLevel + 7; s->hash_size = 1 << s->hash_bits; s->hash_mask = s->hash_size - 1; s->hash_shift = ((s->hash_bits+MIN_MATCH-1)/MIN_MATCH); s->window = (Bytef *) ZALLOC(strm, s->w_size, 2*sizeof(Byte)); s->prev = (Posf *) ZALLOC(strm, s->w_size, sizeof(Pos)); s->head = (Posf *) ZALLOC(strm, s->hash_size, sizeof(Pos)); s->high_water = 0; /* nothing written to s->window yet */ s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */ - overlay = (ushf *) ZALLOC(strm, s->lit_bufsize, sizeof(ush)+2); - s->pending_buf = (uchf *) overlay; - s->pending_buf_size = (ulg)s->lit_bufsize * (sizeof(ush)+2L); + /* We overlay pending_buf and sym_buf. This works since the average size + * for length/distance pairs over any compressed block is assured to be 31 + * bits or less. + * + * Analysis: The longest fixed codes are a length code of 8 bits plus 5 + * extra bits, for lengths 131 to 257. The longest fixed distance codes are + * 5 bits plus 13 extra bits, for distances 16385 to 32768. The longest + * possible fixed-codes length/distance pair is then 31 bits total. + * + * sym_buf starts one-fourth of the way into pending_buf. So there are + * three bytes in sym_buf for every four bytes in pending_buf. Each symbol + * in sym_buf is three bytes -- two for the distance and one for the + * literal/length. As each symbol is consumed, the pointer to the next + * sym_buf value to read moves forward three bytes. From that symbol, up to + * 31 bits are written to pending_buf. The closest the written pending_buf + * bits gets to the next sym_buf symbol to read is just before the last + * code is written. At that time, 31*(n-2) bits have been written, just + * after 24*(n-2) bits have been consumed from sym_buf. sym_buf starts at + * 8*n bits into pending_buf. (Note that the symbol buffer fills when n-1 + * symbols are written.) The closest the writing gets to what is unread is + * then n+14 bits. Here n is lit_bufsize, which is 16384 by default, and + * can range from 128 to 32768. + * + * Therefore, at a minimum, there are 142 bits of space between what is + * written and what is read in the overlain buffers, so the symbols cannot + * be overwritten by the compressed data. That space is actually 139 bits, + * due to the three-bit fixed-code block header. + * + * That covers the case where either Z_FIXED is specified, forcing fixed + * codes, or when the use of fixed codes is chosen, because that choice + * results in a smaller compressed block than dynamic codes. That latter + * condition then assures that the above analysis also covers all dynamic + * blocks. A dynamic-code block will only be chosen to be emitted if it has + * fewer bits than a fixed-code block would for the same set of symbols. + * Therefore its average symbol length is assured to be less than 31. So + * the compressed data for a dynamic block also cannot overwrite the + * symbols from which it is being constructed. + */ + + s->pending_buf = (uchf *) ZALLOC(strm, s->lit_bufsize, 4); + s->pending_buf_size = (ulg)s->lit_bufsize * 4; if (s->window == Z_NULL || s->prev == Z_NULL || s->head == Z_NULL || s->pending_buf == Z_NULL) { s->status = FINISH_STATE; strm->msg = ERR_MSG(Z_MEM_ERROR); deflateEnd (strm); return Z_MEM_ERROR; } - s->d_buf = overlay + s->lit_bufsize/sizeof(ush); - s->l_buf = s->pending_buf + (1+sizeof(ush))*s->lit_bufsize; + s->sym_buf = s->pending_buf + s->lit_bufsize; + s->sym_end = (s->lit_bufsize - 1) * 3; + /* We avoid equality with lit_bufsize*3 because of wraparound at 64K + * on 16 bit machines and because stored blocks are restricted to + * 64K-1 bytes. + */ s->level = level; s->strategy = strategy; s->method = (Byte)method; return deflateReset(strm); } /* ========================================================================= * Check for a valid deflate stream state. Return 0 if ok, 1 if not. */ local int deflateStateCheck (strm) z_streamp strm; { deflate_state *s; if (strm == Z_NULL || strm->zalloc == (alloc_func)0 || strm->zfree == (free_func)0) return 1; s = strm->state; if (s == Z_NULL || s->strm != strm || (s->status != INIT_STATE && #ifdef GZIP s->status != GZIP_STATE && #endif s->status != EXTRA_STATE && s->status != NAME_STATE && s->status != COMMENT_STATE && s->status != HCRC_STATE && s->status != BUSY_STATE && s->status != FINISH_STATE)) return 1; return 0; } /* ========================================================================= */ int ZEXPORT deflateSetDictionary (strm, dictionary, dictLength) @@ -456,175 +496,176 @@ int ZEXPORT deflateGetDictionary (strm, dictionary, dictLength) len = s->strstart + s->lookahead; if (len > s->w_size) len = s->w_size; if (dictionary != Z_NULL && len) zmemcpy(dictionary, s->window + s->strstart + s->lookahead - len, len); if (dictLength != Z_NULL) *dictLength = len; return Z_OK; } /* ========================================================================= */ int ZEXPORT deflateResetKeep (strm) z_streamp strm; { deflate_state *s; if (deflateStateCheck(strm)) { return Z_STREAM_ERROR; } strm->total_in = strm->total_out = 0; strm->msg = Z_NULL; /* use zfree if we ever allocate msg dynamically */ strm->data_type = Z_UNKNOWN; s = (deflate_state *)strm->state; s->pending = 0; s->pending_out = s->pending_buf; if (s->wrap < 0) { s->wrap = -s->wrap; /* was made negative by deflate(..., Z_FINISH); */ } s->status = #ifdef GZIP s->wrap == 2 ? GZIP_STATE : #endif - s->wrap ? INIT_STATE : BUSY_STATE; + INIT_STATE; strm->adler = #ifdef GZIP s->wrap == 2 ? crc32(0L, Z_NULL, 0) : #endif adler32(0L, Z_NULL, 0); - s->last_flush = Z_NO_FLUSH; + s->last_flush = -2; _tr_init(s); return Z_OK; } /* ========================================================================= */ int ZEXPORT deflateReset (strm) z_streamp strm; { int ret; ret = deflateResetKeep(strm); if (ret == Z_OK) lm_init(strm->state); return ret; } /* ========================================================================= */ int ZEXPORT deflateSetHeader (strm, head) z_streamp strm; gz_headerp head; { if (deflateStateCheck(strm) || strm->state->wrap != 2) return Z_STREAM_ERROR; strm->state->gzhead = head; return Z_OK; } /* ========================================================================= */ int ZEXPORT deflatePending (strm, pending, bits) unsigned *pending; int *bits; z_streamp strm; { if (deflateStateCheck(strm)) return Z_STREAM_ERROR; if (pending != Z_NULL) *pending = strm->state->pending; if (bits != Z_NULL) *bits = strm->state->bi_valid; return Z_OK; } /* ========================================================================= */ int ZEXPORT deflatePrime (strm, bits, value) z_streamp strm; int bits; int value; { deflate_state *s; int put; if (deflateStateCheck(strm)) return Z_STREAM_ERROR; s = strm->state; - if ((Bytef *)(s->d_buf) < s->pending_out + ((Buf_size + 7) >> 3)) + if (bits < 0 || bits > 16 || + s->sym_buf < s->pending_out + ((Buf_size + 7) >> 3)) return Z_BUF_ERROR; do { put = Buf_size - s->bi_valid; if (put > bits) put = bits; s->bi_buf |= (ush)((value & ((1 << put) - 1)) << s->bi_valid); s->bi_valid += put; _tr_flush_bits(s); value >>= put; bits -= put; } while (bits); return Z_OK; } /* ========================================================================= */ int ZEXPORT deflateParams(strm, level, strategy) z_streamp strm; int level; int strategy; { deflate_state *s; compress_func func; if (deflateStateCheck(strm)) return Z_STREAM_ERROR; s = strm->state; #ifdef FASTEST if (level != 0) level = 1; #else if (level == Z_DEFAULT_COMPRESSION) level = 6; #endif if (level < 0 || level > 9 || strategy < 0 || strategy > Z_FIXED) { return Z_STREAM_ERROR; } func = configuration_table[s->level].func; if ((strategy != s->strategy || func != configuration_table[level].func) && - s->high_water) { + s->last_flush != -2) { /* Flush the last buffer: */ int err = deflate(strm, Z_BLOCK); if (err == Z_STREAM_ERROR) return err; - if (strm->avail_out == 0) + if (strm->avail_in || (s->strstart - s->block_start) + s->lookahead) return Z_BUF_ERROR; } if (s->level != level) { if (s->level == 0 && s->matches != 0) { if (s->matches == 1) slide_hash(s); else CLEAR_HASH(s); s->matches = 0; } s->level = level; s->max_lazy_match = configuration_table[level].max_lazy; s->good_match = configuration_table[level].good_length; s->nice_match = configuration_table[level].nice_length; s->max_chain_length = configuration_table[level].max_chain; } s->strategy = strategy; return Z_OK; } /* ========================================================================= */ int ZEXPORT deflateTune(strm, good_length, max_lazy, nice_length, max_chain) z_streamp strm; int good_length; int max_lazy; int nice_length; int max_chain; { deflate_state *s; if (deflateStateCheck(strm)) return Z_STREAM_ERROR; s = strm->state; s->good_match = (uInt)good_length; s->max_lazy_match = (uInt)max_lazy; s->nice_match = nice_length; @@ -779,70 +820,72 @@ int ZEXPORT deflate (strm, flush) } if (strm->avail_out == 0) ERR_RETURN(strm, Z_BUF_ERROR); old_flush = s->last_flush; s->last_flush = flush; /* Flush as much pending output as possible */ if (s->pending != 0) { flush_pending(strm); if (strm->avail_out == 0) { /* Since avail_out is 0, deflate will be called again with * more output space, but possibly with both pending and * avail_in equal to zero. There won't be anything to do, * but this is not an error situation so make sure we * return OK instead of BUF_ERROR at next call of deflate: */ s->last_flush = -1; return Z_OK; } /* Make sure there is something to do and avoid duplicate consecutive * flushes. For repeated and useless calls with Z_FINISH, we keep * returning Z_STREAM_END instead of Z_BUF_ERROR. */ } else if (strm->avail_in == 0 && RANK(flush) <= RANK(old_flush) && flush != Z_FINISH) { ERR_RETURN(strm, Z_BUF_ERROR); } /* User must not provide more input after the first FINISH: */ if (s->status == FINISH_STATE && strm->avail_in != 0) { ERR_RETURN(strm, Z_BUF_ERROR); } /* Write the header */ + if (s->status == INIT_STATE && s->wrap == 0) + s->status = BUSY_STATE; if (s->status == INIT_STATE) { /* zlib header */ uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8; uInt level_flags; if (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2) level_flags = 0; else if (s->level < 6) level_flags = 1; else if (s->level == 6) level_flags = 2; else level_flags = 3; header |= (level_flags << 6); if (s->strstart != 0) header |= PRESET_DICT; header += 31 - (header % 31); putShortMSB(s, header); /* Save the adler32 of the preset dictionary: */ if (s->strstart != 0) { putShortMSB(s, (uInt)(strm->adler >> 16)); putShortMSB(s, (uInt)(strm->adler & 0xffff)); } strm->adler = adler32(0L, Z_NULL, 0); s->status = BUSY_STATE; /* Compression must start with an empty pending buffer */ flush_pending(strm); if (s->pending != 0) { s->last_flush = -1; return Z_OK; } } #ifdef GZIP @@ -1076,107 +1119,104 @@ int ZEXPORT deflate (strm, flush) int ZEXPORT deflateEnd (strm) z_streamp strm; { int status; if (deflateStateCheck(strm)) return Z_STREAM_ERROR; status = strm->state->status; /* Deallocate in reverse order of allocations: */ TRY_FREE(strm, strm->state->pending_buf); TRY_FREE(strm, strm->state->head); TRY_FREE(strm, strm->state->prev); TRY_FREE(strm, strm->state->window); ZFREE(strm, strm->state); strm->state = Z_NULL; return status == BUSY_STATE ? Z_DATA_ERROR : Z_OK; } /* ========================================================================= * Copy the source state to the destination state. * To simplify the source, this is not supported for 16-bit MSDOS (which * doesn't have enough memory anyway to duplicate compression states). */ int ZEXPORT deflateCopy (dest, source) z_streamp dest; z_streamp source; { #ifdef MAXSEG_64K return Z_STREAM_ERROR; #else deflate_state *ds; deflate_state *ss; - ushf *overlay; if (deflateStateCheck(source) || dest == Z_NULL) { return Z_STREAM_ERROR; } ss = source->state; zmemcpy((voidpf)dest, (voidpf)source, sizeof(z_stream)); ds = (deflate_state *) ZALLOC(dest, 1, sizeof(deflate_state)); if (ds == Z_NULL) return Z_MEM_ERROR; dest->state = (struct internal_state FAR *) ds; zmemcpy((voidpf)ds, (voidpf)ss, sizeof(deflate_state)); ds->strm = dest; ds->window = (Bytef *) ZALLOC(dest, ds->w_size, 2*sizeof(Byte)); ds->prev = (Posf *) ZALLOC(dest, ds->w_size, sizeof(Pos)); ds->head = (Posf *) ZALLOC(dest, ds->hash_size, sizeof(Pos)); - overlay = (ushf *) ZALLOC(dest, ds->lit_bufsize, sizeof(ush)+2); - ds->pending_buf = (uchf *) overlay; + ds->pending_buf = (uchf *) ZALLOC(dest, ds->lit_bufsize, 4); if (ds->window == Z_NULL || ds->prev == Z_NULL || ds->head == Z_NULL || ds->pending_buf == Z_NULL) { deflateEnd (dest); return Z_MEM_ERROR; } /* following zmemcpy do not work for 16-bit MSDOS */ zmemcpy(ds->window, ss->window, ds->w_size * 2 * sizeof(Byte)); zmemcpy((voidpf)ds->prev, (voidpf)ss->prev, ds->w_size * sizeof(Pos)); zmemcpy((voidpf)ds->head, (voidpf)ss->head, ds->hash_size * sizeof(Pos)); zmemcpy(ds->pending_buf, ss->pending_buf, (uInt)ds->pending_buf_size); ds->pending_out = ds->pending_buf + (ss->pending_out - ss->pending_buf); - ds->d_buf = overlay + ds->lit_bufsize/sizeof(ush); - ds->l_buf = ds->pending_buf + (1+sizeof(ush))*ds->lit_bufsize; + ds->sym_buf = ds->pending_buf + ds->lit_bufsize; ds->l_desc.dyn_tree = ds->dyn_ltree; ds->d_desc.dyn_tree = ds->dyn_dtree; ds->bl_desc.dyn_tree = ds->bl_tree; return Z_OK; #endif /* MAXSEG_64K */ } /* =========================================================================== * Read a new buffer from the current input stream, update the adler32 * and total number of bytes read. All deflate() input goes through * this function so some applications may wish to modify it to avoid * allocating a large strm->next_in buffer and copying from it. * (See also flush_pending()). */ local unsigned read_buf(strm, buf, size) z_streamp strm; Bytef *buf; unsigned size; { unsigned len = strm->avail_in; if (len > size) len = size; if (len == 0) return 0; strm->avail_in -= len; zmemcpy(buf, strm->next_in, len); if (strm->state->wrap == 1) { strm->adler = adler32(strm->adler, buf, len); } #ifdef GZIP else if (strm->state->wrap == 2) { strm->adler = crc32(strm->adler, buf, len); @@ -1481,70 +1521,72 @@ local void check_match(s, start, match, length) */ local void fill_window(s) deflate_state *s; { unsigned n; unsigned more; /* Amount of free space at the end of the window. */ uInt wsize = s->w_size; Assert(s->lookahead < MIN_LOOKAHEAD, "already enough lookahead"); do { more = (unsigned)(s->window_size -(ulg)s->lookahead -(ulg)s->strstart); /* Deal with !@#$% 64K limit: */ if (sizeof(int) <= 2) { if (more == 0 && s->strstart == 0 && s->lookahead == 0) { more = wsize; } else if (more == (unsigned)(-1)) { /* Very unlikely, but possible on 16 bit machine if * strstart == 0 && lookahead == 1 (input done a byte at time) */ more--; } } /* If the window is almost full and there is insufficient lookahead, * move the upper half to the lower one to make room in the upper half. */ if (s->strstart >= wsize+MAX_DIST(s)) { zmemcpy(s->window, s->window+wsize, (unsigned)wsize - more); s->match_start -= wsize; s->strstart -= wsize; /* we now have strstart >= MAX_DIST */ s->block_start -= (long) wsize; + if (s->insert > s->strstart) + s->insert = s->strstart; slide_hash(s); more += wsize; } if (s->strm->avail_in == 0) break; /* If there was no sliding: * strstart <= WSIZE+MAX_DIST-1 && lookahead <= MIN_LOOKAHEAD - 1 && * more == window_size - lookahead - strstart * => more >= window_size - (MIN_LOOKAHEAD-1 + WSIZE + MAX_DIST-1) * => more >= window_size - 2*WSIZE + 2 * In the BIG_MEM or MMAP case (not yet supported), * window_size == input_size + MIN_LOOKAHEAD && * strstart + s->lookahead <= input_size => more >= MIN_LOOKAHEAD. * Otherwise, window_size == 2*WSIZE so more >= 2. * If there was sliding, more >= WSIZE. So in all cases, more >= 2. */ Assert(more >= 2, "more < 2"); n = read_buf(s->strm, s->window + s->strstart + s->lookahead, more); s->lookahead += n; /* Initialize the hash value now that we have some input: */ if (s->lookahead + s->insert >= MIN_MATCH) { uInt str = s->strstart - s->insert; s->ins_h = s->window[str]; UPDATE_HASH(s, s->ins_h, s->window[str + 1]); #if MIN_MATCH != 3 Call UPDATE_HASH() MIN_MATCH-3 more times #endif while (s->insert) { UPDATE_HASH(s, s->ins_h, s->window[str + MIN_MATCH-1]); #ifndef FASTEST s->prev[str & s->w_mask] = s->head[s->ins_h]; #endif s->head[s->ins_h] = (Pos)str; @@ -1710,113 +1752,119 @@ local block_state deflate_stored(s, flush) left = len; zmemcpy(s->strm->next_out, s->window + s->block_start, left); s->strm->next_out += left; s->strm->avail_out -= left; s->strm->total_out += left; s->block_start += left; len -= left; } /* Copy uncompressed bytes directly from next_in to next_out, updating * the check value. */ if (len) { read_buf(s->strm, s->strm->next_out, len); s->strm->next_out += len; s->strm->avail_out -= len; s->strm->total_out += len; } } while (last == 0); /* Update the sliding window with the last s->w_size bytes of the copied * data, or append all of the copied data to the existing window if less * than s->w_size bytes were copied. Also update the number of bytes to * insert in the hash tables, in the event that deflateParams() switches to * a non-zero compression level. */ used -= s->strm->avail_in; /* number of input bytes directly copied */ if (used) { /* If any input was used, then no unused input remains in the window, * therefore s->block_start == s->strstart. */ if (used >= s->w_size) { /* supplant the previous history */ s->matches = 2; /* clear hash */ zmemcpy(s->window, s->strm->next_in - s->w_size, s->w_size); s->strstart = s->w_size; + s->insert = s->strstart; } else { if (s->window_size - s->strstart <= used) { /* Slide the window down. */ s->strstart -= s->w_size; zmemcpy(s->window, s->window + s->w_size, s->strstart); if (s->matches < 2) s->matches++; /* add a pending slide_hash() */ + if (s->insert > s->strstart) + s->insert = s->strstart; } zmemcpy(s->window + s->strstart, s->strm->next_in - used, used); s->strstart += used; + s->insert += MIN(used, s->w_size - s->insert); } s->block_start = s->strstart; - s->insert += MIN(used, s->w_size - s->insert); } if (s->high_water < s->strstart) s->high_water = s->strstart; /* If the last block was written to next_out, then done. */ if (last) return finish_done; /* If flushing and all input has been consumed, then done. */ if (flush != Z_NO_FLUSH && flush != Z_FINISH && s->strm->avail_in == 0 && (long)s->strstart == s->block_start) return block_done; /* Fill the window with any remaining input. */ - have = s->window_size - s->strstart - 1; + have = s->window_size - s->strstart; if (s->strm->avail_in > have && s->block_start >= (long)s->w_size) { /* Slide the window down. */ s->block_start -= s->w_size; s->strstart -= s->w_size; zmemcpy(s->window, s->window + s->w_size, s->strstart); if (s->matches < 2) s->matches++; /* add a pending slide_hash() */ have += s->w_size; /* more space now */ + if (s->insert > s->strstart) + s->insert = s->strstart; } if (have > s->strm->avail_in) have = s->strm->avail_in; if (have) { read_buf(s->strm, s->window + s->strstart, have); s->strstart += have; + s->insert += MIN(have, s->w_size - s->insert); } if (s->high_water < s->strstart) s->high_water = s->strstart; /* There was not enough avail_out to write a complete worthy or flushed * stored block to next_out. Write a stored block to pending instead, if we * have enough input for a worthy block, or if flushing and there is enough * room for the remaining input as a stored block in the pending buffer. */ have = (s->bi_valid + 42) >> 3; /* number of header bytes */ /* maximum stored block length that will fit in pending: */ have = MIN(s->pending_buf_size - have, MAX_STORED); min_block = MIN(have, s->w_size); left = s->strstart - s->block_start; if (left >= min_block || ((left || flush == Z_FINISH) && flush != Z_NO_FLUSH && s->strm->avail_in == 0 && left <= have)) { len = MIN(left, have); last = flush == Z_FINISH && s->strm->avail_in == 0 && len == left ? 1 : 0; _tr_stored_block(s, (charf *)s->window + s->block_start, len, last); s->block_start += len; flush_pending(s->strm); } /* We've done all we can with the available input and output. */ return last ? finish_started : need_more; } /* =========================================================================== * Compress as much as possible from the input stream, return the current * block state. * This function does not perform lazy evaluation of matches and inserts * new strings in the dictionary only for unmatched strings or for short * matches. It is used only for the fast compression options. @@ -1880,71 +1928,71 @@ local block_state deflate_fast(s, flush) s->strstart++; INSERT_STRING(s, s->strstart, hash_head); /* strstart never exceeds WSIZE-MAX_MATCH, so there are * always MIN_MATCH bytes ahead. */ } while (--s->match_length != 0); s->strstart++; } else #endif { s->strstart += s->match_length; s->match_length = 0; s->ins_h = s->window[s->strstart]; UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]); #if MIN_MATCH != 3 Call UPDATE_HASH() MIN_MATCH-3 more times #endif /* If lookahead < MIN_MATCH, ins_h is garbage, but it does not * matter since it will be recomputed at next deflate call. */ } } else { /* No match, output a literal byte */ Tracevv((stderr,"%c", s->window[s->strstart])); _tr_tally_lit (s, s->window[s->strstart], bflush); s->lookahead--; s->strstart++; } if (bflush) FLUSH_BLOCK(s, 0); } s->insert = s->strstart < MIN_MATCH-1 ? s->strstart : MIN_MATCH-1; if (flush == Z_FINISH) { FLUSH_BLOCK(s, 1); return finish_done; } - if (s->last_lit) + if (s->sym_next) FLUSH_BLOCK(s, 0); return block_done; } #ifndef FASTEST /* =========================================================================== * Same as above, but achieves better compression. We use a lazy * evaluation for matches: a match is finally adopted only if there is * no better match at the next window position. */ local block_state deflate_slow(s, flush) deflate_state *s; int flush; { IPos hash_head; /* head of hash chain */ int bflush; /* set if current block must be flushed */ /* Process the input block. */ for (;;) { /* Make sure that we always have enough lookahead, except * at the end of the input file. We need MAX_MATCH bytes * for the next match, plus MIN_MATCH bytes to insert the * string following the next match. */ if (s->lookahead < MIN_LOOKAHEAD) { fill_window(s); if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) { return need_more; } if (s->lookahead == 0) break; /* flush the current block */ } /* Insert the string window[strstart .. strstart+2] in the * dictionary, and set hash_head to the head of the hash chain: */ @@ -2011,71 +2059,71 @@ local block_state deflate_slow(s, flush) if (bflush) FLUSH_BLOCK(s, 0); } else if (s->match_available) { /* If there was no match at the previous position, output a * single literal. If there was a match but the current match * is longer, truncate the previous match to a single literal. */ Tracevv((stderr,"%c", s->window[s->strstart-1])); _tr_tally_lit(s, s->window[s->strstart-1], bflush); if (bflush) { FLUSH_BLOCK_ONLY(s, 0); } s->strstart++; s->lookahead--; if (s->strm->avail_out == 0) return need_more; } else { /* There is no previous match to compare with, wait for * the next step to decide. */ s->match_available = 1; s->strstart++; s->lookahead--; } } Assert (flush != Z_NO_FLUSH, "no flush?"); if (s->match_available) { Tracevv((stderr,"%c", s->window[s->strstart-1])); _tr_tally_lit(s, s->window[s->strstart-1], bflush); s->match_available = 0; } s->insert = s->strstart < MIN_MATCH-1 ? s->strstart : MIN_MATCH-1; if (flush == Z_FINISH) { FLUSH_BLOCK(s, 1); return finish_done; } - if (s->last_lit) + if (s->sym_next) FLUSH_BLOCK(s, 0); return block_done; } #endif /* FASTEST */ /* =========================================================================== * For Z_RLE, simply look for runs of bytes, generate matches only of distance * one. Do not maintain a hash table. (It will be regenerated if this run of * deflate switches away from Z_RLE.) */ local block_state deflate_rle(s, flush) deflate_state *s; int flush; { int bflush; /* set if current block must be flushed */ uInt prev; /* byte at distance one to match */ Bytef *scan, *strend; /* scan goes up to strend for length of run */ for (;;) { /* Make sure that we always have enough lookahead, except * at the end of the input file. We need MAX_MATCH bytes * for the longest run, plus one for the unrolled loop. */ if (s->lookahead <= MAX_MATCH) { fill_window(s); if (s->lookahead <= MAX_MATCH && flush == Z_NO_FLUSH) { return need_more; } if (s->lookahead == 0) break; /* flush the current block */ } /* See how many times the previous byte repeats */ s->match_length = 0; if (s->lookahead >= MIN_MATCH && s->strstart > 0) { scan = s->window + s->strstart - 1; @@ -2086,78 +2134,78 @@ local block_state deflate_rle(s, flush) } while (prev == *++scan && prev == *++scan && prev == *++scan && prev == *++scan && prev == *++scan && prev == *++scan && prev == *++scan && prev == *++scan && scan < strend); s->match_length = MAX_MATCH - (uInt)(strend - scan); if (s->match_length > s->lookahead) s->match_length = s->lookahead; } Assert(scan <= s->window+(uInt)(s->window_size-1), "wild scan"); } /* Emit match if have run of MIN_MATCH or longer, else emit literal */ if (s->match_length >= MIN_MATCH) { check_match(s, s->strstart, s->strstart - 1, s->match_length); _tr_tally_dist(s, 1, s->match_length - MIN_MATCH, bflush); s->lookahead -= s->match_length; s->strstart += s->match_length; s->match_length = 0; } else { /* No match, output a literal byte */ Tracevv((stderr,"%c", s->window[s->strstart])); _tr_tally_lit (s, s->window[s->strstart], bflush); s->lookahead--; s->strstart++; } if (bflush) FLUSH_BLOCK(s, 0); } s->insert = 0; if (flush == Z_FINISH) { FLUSH_BLOCK(s, 1); return finish_done; } - if (s->last_lit) + if (s->sym_next) FLUSH_BLOCK(s, 0); return block_done; } /* =========================================================================== * For Z_HUFFMAN_ONLY, do not look for matches. Do not maintain a hash table. * (It will be regenerated if this run of deflate switches away from Huffman.) */ local block_state deflate_huff(s, flush) deflate_state *s; int flush; { int bflush; /* set if current block must be flushed */ for (;;) { /* Make sure that we have a literal to write. */ if (s->lookahead == 0) { fill_window(s); if (s->lookahead == 0) { if (flush == Z_NO_FLUSH) return need_more; break; /* flush the current block */ } } /* Output a literal byte */ s->match_length = 0; Tracevv((stderr,"%c", s->window[s->strstart])); _tr_tally_lit (s, s->window[s->strstart], bflush); s->lookahead--; s->strstart++; if (bflush) FLUSH_BLOCK(s, 0); } s->insert = 0; if (flush == Z_FINISH) { FLUSH_BLOCK(s, 1); return finish_done; } - if (s->last_lit) + if (s->sym_next) FLUSH_BLOCK(s, 0); return block_done; } |