#include "deflate.h"
const char deflate_copyright[] =
- " deflate 1.2.12 Copyright 1995-2022 Jean-loup Gailly and Mark Adler ";
+ " deflate 1.2.13 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
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
-# pragma message("Assembler code may have bugs -- use at your own risk")
- void match_init OF((void)); /* asm code initialization */
- uInt longest_match OF((deflate_state *s, IPos cur_match));
-#else
local uInt longest_match OF((deflate_state *s, IPos cur_match));
-#endif
#ifdef ZLIB_DEBUG
local void check_match OF((deflate_state *s, IPos start, IPos match,
* 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)
+#define UPDATE_HASH(s,h,c) (h = (((h) << s->hash_shift) ^ (c)) & s->hash_mask)
/* ===========================================================================
*/
#define CLEAR_HASH(s) \
do { \
- s->head[s->hash_size-1] = NIL; \
+ s->head[s->hash_size - 1] = NIL; \
zmemzero((Bytef *)s->head, \
- (unsigned)(s->hash_size-1)*sizeof(*s->head)); \
+ (unsigned)(s->hash_size - 1)*sizeof(*s->head)); \
} while (0)
/* ===========================================================================
if (windowBits < 0) { /* suppress zlib wrapper */
wrap = 0;
+ if (windowBits < -15)
+ return Z_STREAM_ERROR;
windowBits = -windowBits;
}
#ifdef GZIP
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->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));
* 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
+ * 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
+ * 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
}
/* =========================================================================
- * For the default windowBits of 15 and memLevel of 8, this function returns
- * a close to exact, as well as small, upper bound on the compressed size.
- * They are coded as constants here for a reason--if the #define's are
- * changed, then this function needs to be changed as well. The return
- * value for 15 and 8 only works for those exact settings.
+ * For the default windowBits of 15 and memLevel of 8, this function returns a
+ * close to exact, as well as small, upper bound on the compressed size. This
+ * is an expansion of ~0.03%, plus a small constant.
+ *
+ * For any setting other than those defaults for windowBits and memLevel, one
+ * of two worst case bounds is returned. This is at most an expansion of ~4% or
+ * ~13%, plus a small constant.
*
- * For any setting other than those defaults for windowBits and memLevel,
- * the value returned is a conservative worst case for the maximum expansion
- * resulting from using fixed blocks instead of stored blocks, which deflate
- * can emit on compressed data for some combinations of the parameters.
+ * Both the 0.03% and 4% derive from the overhead of stored blocks. The first
+ * one is for stored blocks of 16383 bytes (memLevel == 8), whereas the second
+ * is for stored blocks of 127 bytes (the worst case memLevel == 1). The
+ * expansion results from five bytes of header for each stored block.
*
- * This function could be more sophisticated to provide closer upper bounds for
- * every combination of windowBits and memLevel. But even the conservative
- * upper bound of about 14% expansion does not seem onerous for output buffer
- * allocation.
+ * The larger expansion of 13% results from a window size less than or equal to
+ * the symbols buffer size (windowBits <= memLevel + 7). In that case some of
+ * the data being compressed may have slid out of the sliding window, impeding
+ * a stored block from being emitted. Then the only choice is a fixed or
+ * dynamic block, where a fixed block limits the maximum expansion to 9 bits
+ * per 8-bit byte, plus 10 bits for every block. The smallest block size for
+ * which this can occur is 255 (memLevel == 2).
+ *
+ * Shifts are used to approximate divisions, for speed.
*/
uLong ZEXPORT deflateBound(
z_streamp strm,
uLong sourceLen)
{
deflate_state *s;
- uLong complen, wraplen;
+ uLong fixedlen, storelen, wraplen;
+
+ /* upper bound for fixed blocks with 9-bit literals and length 255
+ (memLevel == 2, which is the lowest that may not use stored blocks) --
+ ~13% overhead plus a small constant */
+ fixedlen = sourceLen + (sourceLen >> 3) + (sourceLen >> 8) +
+ (sourceLen >> 9) + 4;
- /* conservative upper bound for compressed data */
- complen = sourceLen +
- ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 5;
+ /* upper bound for stored blocks with length 127 (memLevel == 1) --
+ ~4% overhead plus a small constant */
+ storelen = sourceLen + (sourceLen >> 5) + (sourceLen >> 7) +
+ (sourceLen >> 11) + 7;
- /* if can't get parameters, return conservative bound plus zlib wrapper */
+ /* if can't get parameters, return larger bound plus a zlib wrapper */
if (deflateStateCheck(strm))
- return complen + 6;
+ return (fixedlen > storelen ? fixedlen : storelen) + 6;
/* compute wrapper length */
s = strm->state;
wraplen = 6;
}
- /* if not default parameters, return conservative bound */
+ /* if not default parameters, return one of the conservative bounds */
if (s->w_bits != 15 || s->hash_bits != 8 + 7)
- return complen + wraplen;
+ return (s->w_bits <= s->hash_bits ? fixedlen : storelen) + wraplen;
- /* default settings: return tight bound for that case */
+ /* default settings: return tight bound for that case -- ~0.03% overhead
+ plus a small constant */
return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) +
(sourceLen >> 25) + 13 - 6 + wraplen;
}
s->status = BUSY_STATE;
if (s->status == INIT_STATE) {
/* zlib header */
- uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8;
+ uInt header = (Z_DEFLATED + ((s->w_bits - 8) << 4)) << 8;
uInt level_flags;
if (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2)
s->match_length = s->prev_length = MIN_MATCH-1;
s->match_available = 0;
s->ins_h = 0;
-#ifndef FASTEST
-#ifdef ASMV
- match_init(); /* initialize the asm code */
-#endif
-#endif
}
#ifndef FASTEST
* string (strstart) and its distance is <= MAX_DIST, and prev_length >= 1
* OUT assertion: the match length is not greater than s->lookahead.
*/
-#ifndef ASMV
-/* For 80x86 and 680x0, an optimized version will be provided in match.asm or
- * match.S. The code will be functionally equivalent.
- */
local uInt longest_match(
deflate_state *s,
IPos cur_match)
*/
register Bytef *strend = s->window + s->strstart + MAX_MATCH - 1;
register ush scan_start = *(ushf*)scan;
- register ush scan_end = *(ushf*)(scan+best_len-1);
+ register ush scan_end = *(ushf*)(scan + best_len - 1);
#else
register Bytef *strend = s->window + s->strstart + MAX_MATCH;
- register Byte scan_end1 = scan[best_len-1];
+ register Byte scan_end1 = scan[best_len - 1];
register Byte scan_end = scan[best_len];
#endif
*/
if ((uInt)nice_match > s->lookahead) nice_match = (int)s->lookahead;
- Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead");
+ Assert((ulg)s->strstart <= s->window_size - MIN_LOOKAHEAD,
+ "need lookahead");
do {
Assert(cur_match < s->strstart, "no future");
/* This code assumes sizeof(unsigned short) == 2. Do not use
* UNALIGNED_OK if your compiler uses a different size.
*/
- if (*(ushf*)(match+best_len-1) != scan_end ||
+ if (*(ushf*)(match + best_len - 1) != scan_end ||
*(ushf*)match != scan_start) continue;
/* It is not necessary to compare scan[2] and match[2] since they are
* always equal when the other bytes match, given that the hash keys
* are equal and that HASH_BITS >= 8. Compare 2 bytes at a time at
- * strstart+3, +5, ... up to strstart+257. We check for insufficient
+ * strstart + 3, + 5, up to strstart + 257. We check for insufficient
* lookahead only every 4th comparison; the 128th check will be made
- * at strstart+257. If MAX_MATCH-2 is not a multiple of 8, it is
+ * at strstart + 257. If MAX_MATCH-2 is not a multiple of 8, it is
* necessary to put more guard bytes at the end of the window, or
* to check more often for insufficient lookahead.
*/
Assert(scan[2] == match[2], "scan[2]?");
scan++, match++;
do {
- } while (*(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
- *(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
- *(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
- *(ushf*)(scan+=2) == *(ushf*)(match+=2) &&
+ } while (*(ushf*)(scan += 2) == *(ushf*)(match += 2) &&
+ *(ushf*)(scan += 2) == *(ushf*)(match += 2) &&
+ *(ushf*)(scan += 2) == *(ushf*)(match += 2) &&
+ *(ushf*)(scan += 2) == *(ushf*)(match += 2) &&
scan < strend);
/* The funny "do {}" generates better code on most compilers */
- /* Here, scan <= window+strstart+257 */
- Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan");
+ /* Here, scan <= window + strstart + 257 */
+ Assert(scan <= s->window + (unsigned)(s->window_size - 1),
+ "wild scan");
if (*scan == *match) scan++;
- len = (MAX_MATCH - 1) - (int)(strend-scan);
+ len = (MAX_MATCH - 1) - (int)(strend - scan);
scan = strend - (MAX_MATCH-1);
#else /* UNALIGNED_OK */
- if (match[best_len] != scan_end ||
- match[best_len-1] != scan_end1 ||
- *match != *scan ||
- *++match != scan[1]) continue;
+ if (match[best_len] != scan_end ||
+ match[best_len - 1] != scan_end1 ||
+ *match != *scan ||
+ *++match != scan[1]) continue;
- /* The check at best_len-1 can be removed because it will be made
+ /* The check at best_len - 1 can be removed because it will be made
* again later. (This heuristic is not always a win.)
* It is not necessary to compare scan[2] and match[2] since they
* are always equal when the other bytes match, given that
Assert(*scan == *match, "match[2]?");
/* We check for insufficient lookahead only every 8th comparison;
- * the 256th check will be made at strstart+258.
+ * the 256th check will be made at strstart + 258.
*/
do {
} while (*++scan == *++match && *++scan == *++match &&
*++scan == *++match && *++scan == *++match &&
scan < strend);
- Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan");
+ Assert(scan <= s->window + (unsigned)(s->window_size - 1),
+ "wild scan");
len = MAX_MATCH - (int)(strend - scan);
scan = strend - MAX_MATCH;
best_len = len;
if (len >= nice_match) break;
#ifdef UNALIGNED_OK
- scan_end = *(ushf*)(scan+best_len-1);
+ scan_end = *(ushf*)(scan + best_len - 1);
#else
- scan_end1 = scan[best_len-1];
+ scan_end1 = scan[best_len - 1];
scan_end = scan[best_len];
#endif
}
if ((uInt)best_len <= s->lookahead) return (uInt)best_len;
return s->lookahead;
}
-#endif /* ASMV */
#else /* FASTEST */
*/
Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever");
- Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead");
+ Assert((ulg)s->strstart <= s->window_size - MIN_LOOKAHEAD,
+ "need lookahead");
Assert(cur_match < s->strstart, "no future");
*/
if (match[0] != scan[0] || match[1] != scan[1]) return MIN_MATCH-1;
- /* The check at best_len-1 can be removed because it will be made
+ /* The check at best_len - 1 can be removed because it will be made
* again later. (This heuristic is not always a win.)
* It is not necessary to compare scan[2] and match[2] since they
* are always equal when the other bytes match, given that
Assert(*scan == *match, "match[2]?");
/* We check for insufficient lookahead only every 8th comparison;
- * the 256th check will be made at strstart+258.
+ * the 256th check will be made at strstart + 258.
*/
do {
} while (*++scan == *++match && *++scan == *++match &&
*++scan == *++match && *++scan == *++match &&
scan < strend);
- Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan");
+ Assert(scan <= s->window + (unsigned)(s->window_size - 1), "wild scan");
len = MAX_MATCH - (int)(strend - scan);
z_error("invalid match");
}
if (z_verbose > 1) {
- fprintf(stderr,"\\[%d,%d]", start-match, length);
+ fprintf(stderr,"\\[%d,%d]", start - match, length);
do { putc(s->window[start++], stderr); } while (--length != 0);
}
}
/* 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)) {
+ if (s->strstart >= wsize + MAX_DIST(s)) {
- zmemcpy(s->window, s->window+wsize, (unsigned)wsize - more);
+ 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;
*
* deflate_stored() is written to minimize the number of times an input byte is
* copied. It is most efficient with large input and output buffers, which
- * maximizes the opportunites to have a single copy from next_in to next_out.
+ * maximizes the opportunities to have a single copy from next_in to next_out.
*/
local block_state deflate_stored(
deflate_state *s,
if (s->lookahead == 0) break; /* flush the current block */
}
- /* Insert the string window[strstart .. strstart+2] in the
+ /* Insert the string window[strstart .. strstart + 2] in the
* dictionary, and set hash_head to the head of the hash chain:
*/
hash_head = NIL;
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]);
+ UPDATE_HASH(s, s->ins_h, s->window[s->strstart + 1]);
#if MIN_MATCH != 3
Call UPDATE_HASH() MIN_MATCH-3 more times
#endif
} else {
/* No match, output a literal byte */
Tracevv((stderr,"%c", s->window[s->strstart]));
- _tr_tally_lit (s, s->window[s->strstart], bflush);
+ _tr_tally_lit(s, s->window[s->strstart], bflush);
s->lookahead--;
s->strstart++;
}
if (s->lookahead == 0) break; /* flush the current block */
}
- /* Insert the string window[strstart .. strstart+2] in the
+ /* Insert the string window[strstart .. strstart + 2] in the
* dictionary, and set hash_head to the head of the hash chain:
*/
hash_head = NIL;
uInt max_insert = s->strstart + s->lookahead - MIN_MATCH;
/* Do not insert strings in hash table beyond this. */
- check_match(s, s->strstart-1, s->prev_match, s->prev_length);
+ check_match(s, s->strstart - 1, s->prev_match, s->prev_length);
- _tr_tally_dist(s, s->strstart -1 - s->prev_match,
+ _tr_tally_dist(s, s->strstart - 1 - s->prev_match,
s->prev_length - MIN_MATCH, bflush);
/* Insert in hash table all strings up to the end of the match.
- * strstart-1 and strstart are already inserted. If there is not
+ * strstart - 1 and strstart are already inserted. If there is not
* enough lookahead, the last two strings are not inserted in
* the hash table.
*/
- s->lookahead -= s->prev_length-1;
+ s->lookahead -= s->prev_length - 1;
s->prev_length -= 2;
do {
if (++s->strstart <= max_insert) {
* 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);
+ 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);
}
}
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);
+ 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 (s->match_length > s->lookahead)
s->match_length = s->lookahead;
}
- Assert(scan <= s->window+(uInt)(s->window_size-1), "wild scan");
+ Assert(scan <= s->window + (uInt)(s->window_size - 1),
+ "wild scan");
}
/* Emit match if have run of MIN_MATCH or longer, else emit literal */
} else {
/* No match, output a literal byte */
Tracevv((stderr,"%c", s->window[s->strstart]));
- _tr_tally_lit (s, s->window[s->strstart], bflush);
+ _tr_tally_lit(s, s->window[s->strstart], bflush);
s->lookahead--;
s->strstart++;
}
/* Output a literal byte */
s->match_length = 0;
Tracevv((stderr,"%c", s->window[s->strstart]));
- _tr_tally_lit (s, s->window[s->strstart], bflush);
+ _tr_tally_lit(s, s->window[s->strstart], bflush);
s->lookahead--;
s->strstart++;
if (bflush) FLUSH_BLOCK(s, 0);
https://perl5.git.perl.org / perl5.git / blobdiff