diff options
Diffstat (limited to 'src/Common/zlib/trees.c')
| -rw-r--r-- | src/Common/zlib/trees.c | 674 |
1 files changed, 294 insertions, 380 deletions
diff --git a/src/Common/zlib/trees.c b/src/Common/zlib/trees.c index 50cf4b45..6a523ef3 100644 --- a/src/Common/zlib/trees.c +++ b/src/Common/zlib/trees.c @@ -1,6 +1,6 @@ /* trees.c -- output deflated data using Huffman coding - * Copyright (C) 1995-2017 Jean-loup Gailly + * Copyright (C) 1995-2024 Jean-loup Gailly * detect_data_type() function provided freely by Cosmin Truta, 2006 * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -121,41 +121,118 @@ struct static_tree_desc_s { int elems; /* max number of elements in the tree */ int max_length; /* max bit length for the codes */ }; -local const static_tree_desc static_l_desc = +#ifdef NO_INIT_GLOBAL_POINTERS +# define TCONST +#else +# define TCONST const +#endif + +local TCONST static_tree_desc static_l_desc = {static_ltree, extra_lbits, LITERALS+1, L_CODES, MAX_BITS}; -local const static_tree_desc static_d_desc = +local TCONST static_tree_desc static_d_desc = {static_dtree, extra_dbits, 0, D_CODES, MAX_BITS}; -local const static_tree_desc static_bl_desc = +local TCONST static_tree_desc static_bl_desc = {(const ct_data *)0, extra_blbits, 0, BL_CODES, MAX_BL_BITS}; /* =========================================================================== - * Local (static) routines in this file. + * Output a short LSB first on the stream. + * IN assertion: there is enough room in pendingBuf. + */ +#define put_short(s, w) { \ + put_byte(s, (uch)((w) & 0xff)); \ + put_byte(s, (uch)((ush)(w) >> 8)); \ +} + +/* =========================================================================== + * Reverse the first len bits of a code, using straightforward code (a faster + * method would use a table) + * IN assertion: 1 <= len <= 15 */ +local unsigned bi_reverse(unsigned code, int len) { + register unsigned res = 0; + do { + res |= code & 1; + code >>= 1, res <<= 1; + } while (--len > 0); + return res >> 1; +} -local void tr_static_init OF((void)); -local void init_block OF((deflate_state *s)); -local void pqdownheap OF((deflate_state *s, ct_data *tree, int k)); -local void gen_bitlen OF((deflate_state *s, tree_desc *desc)); -local void gen_codes OF((ct_data *tree, int max_code, ushf *bl_count)); -local void build_tree OF((deflate_state *s, tree_desc *desc)); -local void scan_tree OF((deflate_state *s, ct_data *tree, int max_code)); -local void send_tree OF((deflate_state *s, ct_data *tree, int max_code)); -local int build_bl_tree OF((deflate_state *s)); -local void send_all_trees OF((deflate_state *s, int lcodes, int dcodes, - int blcodes)); -local void compress_block OF((deflate_state *s, const ct_data *ltree, - const ct_data *dtree)); -local int detect_data_type OF((deflate_state *s)); -local unsigned bi_reverse OF((unsigned value, int length)); -local void bi_windup OF((deflate_state *s)); -local void bi_flush OF((deflate_state *s)); +/* =========================================================================== + * Flush the bit buffer, keeping at most 7 bits in it. + */ +local void bi_flush(deflate_state *s) { + if (s->bi_valid == 16) { + put_short(s, s->bi_buf); + s->bi_buf = 0; + s->bi_valid = 0; + } else if (s->bi_valid >= 8) { + put_byte(s, (Byte)s->bi_buf); + s->bi_buf >>= 8; + s->bi_valid -= 8; + } +} + +/* =========================================================================== + * Flush the bit buffer and align the output on a byte boundary + */ +local void bi_windup(deflate_state *s) { + if (s->bi_valid > 8) { + put_short(s, s->bi_buf); + } else if (s->bi_valid > 0) { + put_byte(s, (Byte)s->bi_buf); + } + s->bi_buf = 0; + s->bi_valid = 0; +#ifdef ZLIB_DEBUG + s->bits_sent = (s->bits_sent + 7) & ~7; +#endif +} + +/* =========================================================================== + * Generate the codes for a given tree and bit counts (which need not be + * optimal). + * IN assertion: the array bl_count contains the bit length statistics for + * the given tree and the field len is set for all tree elements. + * OUT assertion: the field code is set for all tree elements of non + * zero code length. + */ +local void gen_codes(ct_data *tree, int max_code, ushf *bl_count) { + ush next_code[MAX_BITS+1]; /* next code value for each bit length */ + unsigned code = 0; /* running code value */ + int bits; /* bit index */ + int n; /* code index */ + + /* The distribution counts are first used to generate the code values + * without bit reversal. + */ + for (bits = 1; bits <= MAX_BITS; bits++) { + code = (code + bl_count[bits - 1]) << 1; + next_code[bits] = (ush)code; + } + /* Check that the bit counts in bl_count are consistent. The last code + * must be all ones. + */ + Assert (code + bl_count[MAX_BITS] - 1 == (1 << MAX_BITS) - 1, + "inconsistent bit counts"); + Tracev((stderr,"\ngen_codes: max_code %d ", max_code)); + + for (n = 0; n <= max_code; n++) { + int len = tree[n].Len; + if (len == 0) continue; + /* Now reverse the bits */ + tree[n].Code = (ush)bi_reverse(next_code[len]++, len); + + Tracecv(tree != static_ltree, (stderr,"\nn %3d %c l %2d c %4x (%x) ", + n, (isgraph(n) ? n : ' '), len, tree[n].Code, next_code[len] - 1)); + } +} #ifdef GEN_TREES_H -local void gen_trees_header OF((void)); +local void gen_trees_header(void); #endif #ifndef ZLIB_DEBUG # define send_code(s, c, tree) send_bits(s, tree[c].Code, tree[c].Len) @@ -167,34 +244,19 @@ local void gen_trees_header OF((void)); send_bits(s, tree[c].Code, tree[c].Len); } #endif /* =========================================================================== - * Output a short LSB first on the stream. - * IN assertion: there is enough room in pendingBuf. - */ -#define put_short(s, w) { \ - put_byte(s, (uch)((w) & 0xff)); \ - put_byte(s, (uch)((ush)(w) >> 8)); \ -} - -/* =========================================================================== * Send a value on a given number of bits. * IN assertion: length <= 16 and value fits in length bits. */ #ifdef ZLIB_DEBUG -local void send_bits OF((deflate_state *s, int value, int length)); - -local void send_bits(s, value, length) - deflate_state *s; - int value; /* value to send */ - int length; /* number of bits */ -{ +local void send_bits(deflate_state *s, int value, int length) { Tracevv((stderr," l %2d v %4x ", length, value)); Assert(length > 0 && length <= 15, "invalid length"); s->bits_sent += (ulg)length; /* If not enough room in bi_buf, use (valid) bits from bi_buf and - * (16 - bi_valid) bits from value, leaving (width - (16-bi_valid)) + * (16 - bi_valid) bits from value, leaving (width - (16 - bi_valid)) * unused bits in value. */ if (s->bi_valid > (int)Buf_size - length) { s->bi_buf |= (ush)value << s->bi_valid; @@ -228,10 +290,9 @@ local void send_bits(s, value, length) /* =========================================================================== * Initialize the various 'constant' tables. */ -local void tr_static_init() -{ +local void tr_static_init(void) { #if defined(GEN_TREES_H) || !defined(STDC) static int static_init_done = 0; int n; /* iterates over tree elements */ int bits; /* bit counter */ @@ -255,36 +316,36 @@ local void tr_static_init() /* Initialize the mapping length (0..255) -> length code (0..28) */ length = 0; for (code = 0; code < LENGTH_CODES-1; code++) { base_length[code] = length; - for (n = 0; n < (1<<extra_lbits[code]); n++) { + for (n = 0; n < (1 << extra_lbits[code]); n++) { _length_code[length++] = (uch)code; } } Assert (length == 256, "tr_static_init: length != 256"); /* Note that the length 255 (match length 258) can be represented * in two different ways: code 284 + 5 bits or code 285, so we * overwrite length_code[255] to use the best encoding: */ - _length_code[length-1] = (uch)code; + _length_code[length - 1] = (uch)code; /* Initialize the mapping dist (0..32K) -> dist code (0..29) */ dist = 0; for (code = 0 ; code < 16; code++) { base_dist[code] = dist; - for (n = 0; n < (1<<extra_dbits[code]); n++) { + for (n = 0; n < (1 << extra_dbits[code]); n++) { _dist_code[dist++] = (uch)code; } } Assert (dist == 256, "tr_static_init: dist != 256"); dist >>= 7; /* from now on, all distances are divided by 128 */ for ( ; code < D_CODES; code++) { base_dist[code] = dist << 7; - for (n = 0; n < (1<<(extra_dbits[code]-7)); n++) { + for (n = 0; n < (1 << (extra_dbits[code] - 7)); n++) { _dist_code[256 + dist++] = (uch)code; } } - Assert (dist == 256, "tr_static_init: 256+dist != 512"); + Assert (dist == 256, "tr_static_init: 256 + dist != 512"); /* Construct the codes of the static literal tree */ for (bits = 0; bits <= MAX_BITS; bits++) bl_count[bits] = 0; n = 0; @@ -311,21 +372,20 @@ local void tr_static_init() #endif /* defined(GEN_TREES_H) || !defined(STDC) */ } /* =========================================================================== - * Genererate the file trees.h describing the static trees. + * Generate the file trees.h describing the static trees. */ #ifdef GEN_TREES_H # ifndef ZLIB_DEBUG # include <stdio.h> # endif # define SEPARATOR(i, last, width) \ ((i) == (last)? "\n};\n\n" : \ - ((i) % (width) == (width)-1 ? ",\n" : ", ")) + ((i) % (width) == (width) - 1 ? ",\n" : ", ")) -void gen_trees_header() -{ +void gen_trees_header(void) { FILE *header = fopen("trees.h", "w"); int i; Assert (header != NULL, "Can't open trees.h"); @@ -373,13 +433,27 @@ void gen_trees_header() } #endif /* GEN_TREES_H */ /* =========================================================================== + * Initialize a new block. + */ +local void init_block(deflate_state *s) { + int n; /* iterates over tree elements */ + + /* Initialize the trees. */ + for (n = 0; n < L_CODES; n++) s->dyn_ltree[n].Freq = 0; + for (n = 0; n < D_CODES; n++) s->dyn_dtree[n].Freq = 0; + for (n = 0; n < BL_CODES; n++) s->bl_tree[n].Freq = 0; + + s->dyn_ltree[END_BLOCK].Freq = 1; + s->opt_len = s->static_len = 0L; + s->sym_next = s->matches = 0; +} + +/* =========================================================================== * Initialize the tree data structures for a new zlib stream. */ -void ZLIB_INTERNAL _tr_init(s) - deflate_state *s; -{ +void ZLIB_INTERNAL _tr_init(deflate_state *s) { tr_static_init(); s->l_desc.dyn_tree = s->dyn_ltree; s->l_desc.stat_desc = &static_l_desc; @@ -400,26 +474,8 @@ void ZLIB_INTERNAL _tr_init(s) /* Initialize the first block of the first file: */ init_block(s); } -/* =========================================================================== - * Initialize a new block. - */ -local void init_block(s) - deflate_state *s; -{ - int n; /* iterates over tree elements */ - - /* Initialize the trees. */ - for (n = 0; n < L_CODES; n++) s->dyn_ltree[n].Freq = 0; - for (n = 0; n < D_CODES; n++) s->dyn_dtree[n].Freq = 0; - for (n = 0; n < BL_CODES; n++) s->bl_tree[n].Freq = 0; - - s->dyn_ltree[END_BLOCK].Freq = 1; - s->opt_len = s->static_len = 0L; - s->last_lit = s->matches = 0; -} - #define SMALLEST 1 /* Index within the heap array of least frequent node in the Huffman tree */ @@ -447,19 +503,15 @@ local void init_block(s) * exchanging a node with the smallest of its two sons if necessary, stopping * when the heap property is re-established (each father smaller than its * two sons). */ -local void pqdownheap(s, tree, k) - deflate_state *s; - ct_data *tree; /* the tree to restore */ - int k; /* node to move down */ -{ +local void pqdownheap(deflate_state *s, ct_data *tree, int k) { int v = s->heap[k]; int j = k << 1; /* left son of k */ while (j <= s->heap_len) { /* Set j to the smallest of the two sons: */ if (j < s->heap_len && - smaller(tree, s->heap[j+1], s->heap[j], s->depth)) { + smaller(tree, s->heap[j + 1], s->heap[j], s->depth)) { j++; } /* Exit if v is smaller than both sons */ if (smaller(tree, v, s->heap[j], s->depth)) break; @@ -482,12 +534,9 @@ local void pqdownheap(s, tree, k) * array bl_count contains the frequencies for each bit length. * The length opt_len is updated; static_len is also updated if stree is * not null. */ -local void gen_bitlen(s, desc) - deflate_state *s; - tree_desc *desc; /* the tree descriptor */ -{ +local void gen_bitlen(deflate_state *s, tree_desc *desc) { ct_data *tree = desc->dyn_tree; int max_code = desc->max_code; const ct_data *stree = desc->stat_desc->static_tree; const intf *extra = desc->stat_desc->extra_bits; @@ -506,9 +555,9 @@ local void gen_bitlen(s, desc) * overflow in the case of the bit length tree). */ tree[s->heap[s->heap_max]].Len = 0; /* root of the heap */ - for (h = s->heap_max+1; h < HEAP_SIZE; h++) { + for (h = s->heap_max + 1; h < HEAP_SIZE; h++) { n = s->heap[h]; bits = tree[tree[n].Dad].Len + 1; if (bits > max_length) bits = max_length, overflow++; tree[n].Len = (ush)bits; @@ -517,9 +566,9 @@ local void gen_bitlen(s, desc) if (n > max_code) continue; /* not a leaf node */ s->bl_count[bits]++; xbits = 0; - if (n >= base) xbits = extra[n-base]; + if (n >= base) xbits = extra[n - base]; f = tree[n].Freq; s->opt_len += (ulg)f * (unsigned)(bits + xbits); if (stree) s->static_len += (ulg)f * (unsigned)(stree[n].Len + xbits); } @@ -529,12 +578,12 @@ local void gen_bitlen(s, desc) /* This happens for example on obj2 and pic of the Calgary corpus */ /* Find the first bit length which could increase: */ do { - bits = max_length-1; + bits = max_length - 1; while (s->bl_count[bits] == 0) bits--; - s->bl_count[bits]--; /* move one leaf down the tree */ - s->bl_count[bits+1] += 2; /* move one overflow item as its brother */ + s->bl_count[bits]--; /* move one leaf down the tree */ + s->bl_count[bits + 1] += 2; /* move one overflow item as its brother */ s->bl_count[max_length]--; /* The brother of the overflow item also moves one step up, * but this does not affect bl_count[max_length] */ @@ -560,50 +609,11 @@ local void gen_bitlen(s, desc) } } } -/* =========================================================================== - * Generate the codes for a given tree and bit counts (which need not be - * optimal). - * IN assertion: the array bl_count contains the bit length statistics for - * the given tree and the field len is set for all tree elements. - * OUT assertion: the field code is set for all tree elements of non - * zero code length. - */ -local void gen_codes (tree, max_code, bl_count) - ct_data *tree; /* the tree to decorate */ - int max_code; /* largest code with non zero frequency */ - ushf *bl_count; /* number of codes at each bit length */ -{ - ush next_code[MAX_BITS+1]; /* next code value for each bit length */ - unsigned code = 0; /* running code value */ - int bits; /* bit index */ - int n; /* code index */ - - /* The distribution counts are first used to generate the code values - * without bit reversal. - */ - for (bits = 1; bits <= MAX_BITS; bits++) { - code = (code + bl_count[bits-1]) << 1; - next_code[bits] = (ush)code; - } - /* Check that the bit counts in bl_count are consistent. The last code - * must be all ones. - */ - Assert (code + bl_count[MAX_BITS]-1 == (1<<MAX_BITS)-1, - "inconsistent bit counts"); - Tracev((stderr,"\ngen_codes: max_code %d ", max_code)); - - for (n = 0; n <= max_code; n++) { - int len = tree[n].Len; - if (len == 0) continue; - /* Now reverse the bits */ - tree[n].Code = (ush)bi_reverse(next_code[len]++, len); - - Tracecv(tree != static_ltree, (stderr,"\nn %3d %c l %2d c %4x (%x) ", - n, (isgraph(n) ? n : ' '), len, tree[n].Code, next_code[len]-1)); - } -} +#ifdef DUMP_BL_TREE +# include <stdio.h> +#endif /* =========================================================================== * Construct one Huffman tree and assigns the code bit strings and lengths. * Update the total bit length for the current block. @@ -611,21 +621,18 @@ local void gen_codes (tree, max_code, bl_count) * OUT assertions: the fields len and code are set to the optimal bit length * and corresponding code. The length opt_len is updated; static_len is * also updated if stree is not null. The field max_code is set. */ -local void build_tree(s, desc) - deflate_state *s; - tree_desc *desc; /* the tree descriptor */ -{ +local void build_tree(deflate_state *s, tree_desc *desc) { ct_data *tree = desc->dyn_tree; const ct_data *stree = desc->stat_desc->static_tree; int elems = desc->stat_desc->elems; int n, m; /* iterate over heap elements */ int max_code = -1; /* largest code with non zero frequency */ int node; /* new node being created */ /* Construct the initial heap, with least frequent element in - * heap[SMALLEST]. The sons of heap[n] are heap[2*n] and heap[2*n+1]. + * heap[SMALLEST]. The sons of heap[n] are heap[2*n] and heap[2*n + 1]. * heap[0] is not used. */ s->heap_len = 0, s->heap_max = HEAP_SIZE; @@ -651,9 +658,9 @@ local void build_tree(s, desc) /* node is 0 or 1 so it does not have extra bits */ } desc->max_code = max_code; - /* The elements heap[heap_len/2+1 .. heap_len] are leaves of the tree, + /* The elements heap[heap_len/2 + 1 .. heap_len] are leaves of the tree, * establish sub-heaps of increasing lengths: */ for (n = s->heap_len/2; n >= 1; n--) pqdownheap(s, tree, n); @@ -699,13 +706,9 @@ local void build_tree(s, desc) /* =========================================================================== * Scan a literal or distance tree to determine the frequencies of the codes * in the bit length tree. */ -local void scan_tree (s, tree, max_code) - deflate_state *s; - ct_data *tree; /* the tree to be scanned */ - int max_code; /* and its largest code of non zero frequency */ -{ +local void scan_tree(deflate_state *s, ct_data *tree, int max_code) { int n; /* iterates over all tree elements */ int prevlen = -1; /* last emitted length */ int curlen; /* length of current code */ int nextlen = tree[0].Len; /* length of next code */ @@ -713,12 +716,12 @@ local void scan_tree (s, tree, max_code) int max_count = 7; /* max repeat count */ int min_count = 4; /* min repeat count */ if (nextlen == 0) max_count = 138, min_count = 3; - tree[max_code+1].Len = (ush)0xffff; /* guard */ + tree[max_code + 1].Len = (ush)0xffff; /* guard */ for (n = 0; n <= max_code; n++) { - curlen = nextlen; nextlen = tree[n+1].Len; + curlen = nextlen; nextlen = tree[n + 1].Len; if (++count < max_count && curlen == nextlen) { continue; } else if (count < min_count) { s->bl_tree[curlen].Freq += count; @@ -744,26 +747,22 @@ local void scan_tree (s, tree, max_code) /* =========================================================================== * Send a literal or distance tree in compressed form, using the codes in * bl_tree. */ -local void send_tree (s, tree, max_code) - deflate_state *s; - ct_data *tree; /* the tree to be scanned */ - int max_code; /* and its largest code of non zero frequency */ -{ +local void send_tree(deflate_state *s, ct_data *tree, int max_code) { int n; /* iterates over all tree elements */ int prevlen = -1; /* last emitted length */ int curlen; /* length of current code */ int nextlen = tree[0].Len; /* length of next code */ int count = 0; /* repeat count of the current code */ int max_count = 7; /* max repeat count */ int min_count = 4; /* min repeat count */ - /* tree[max_code+1].Len = -1; */ /* guard already set */ + /* tree[max_code + 1].Len = -1; */ /* guard already set */ if (nextlen == 0) max_count = 138, min_count = 3; for (n = 0; n <= max_code; n++) { - curlen = nextlen; nextlen = tree[n+1].Len; + curlen = nextlen; nextlen = tree[n + 1].Len; if (++count < max_count && curlen == nextlen) { continue; } else if (count < min_count) { do { send_code(s, curlen, s->bl_tree); } while (--count != 0); @@ -772,15 +771,15 @@ local void send_tree (s, tree, max_code) if (curlen != prevlen) { send_code(s, curlen, s->bl_tree); count--; } Assert(count >= 3 && count <= 6, " 3_6?"); - send_code(s, REP_3_6, s->bl_tree); send_bits(s, count-3, 2); + send_code(s, REP_3_6, s->bl_tree); send_bits(s, count - 3, 2); } else if (count <= 10) { - send_code(s, REPZ_3_10, s->bl_tree); send_bits(s, count-3, 3); + send_code(s, REPZ_3_10, s->bl_tree); send_bits(s, count - 3, 3); } else { - send_code(s, REPZ_11_138, s->bl_tree); send_bits(s, count-11, 7); + send_code(s, REPZ_11_138, s->bl_tree); send_bits(s, count - 11, 7); } count = 0; prevlen = curlen; if (nextlen == 0) { max_count = 138, min_count = 3; @@ -795,21 +794,19 @@ local void send_tree (s, tree, max_code) /* =========================================================================== * Construct the Huffman tree for the bit lengths and return the index in * bl_order of the last bit length code to send. */ -local int build_bl_tree(s) - deflate_state *s; -{ +local int build_bl_tree(deflate_state *s) { int max_blindex; /* index of last bit length code of non zero freq */ /* Determine the bit length frequencies for literal and distance trees */ scan_tree(s, (ct_data *)s->dyn_ltree, s->l_desc.max_code); scan_tree(s, (ct_data *)s->dyn_dtree, s->d_desc.max_code); /* Build the bit length tree: */ build_tree(s, (tree_desc *)(&(s->bl_desc))); - /* opt_len now includes the length of the tree representations, except - * the lengths of the bit lengths codes and the 5+5+4 bits for the counts. + /* opt_len now includes the length of the tree representations, except the + * lengths of the bit lengths codes and the 5 + 5 + 4 bits for the counts. */ /* Determine the number of bit length codes to send. The pkzip format * requires that at least 4 bit length codes be sent. (appnote.txt says @@ -818,9 +815,9 @@ local int build_bl_tree(s) for (max_blindex = BL_CODES-1; max_blindex >= 3; max_blindex--) { if (s->bl_tree[bl_order[max_blindex]].Len != 0) break; } /* Update opt_len to include the bit length tree and counts */ - s->opt_len += 3*((ulg)max_blindex+1) + 5+5+4; + s->opt_len += 3*((ulg)max_blindex + 1) + 5 + 5 + 4; Tracev((stderr, "\ndyn trees: dyn %ld, stat %ld", s->opt_len, s->static_len)); return max_blindex; @@ -830,73 +827,64 @@ local int build_bl_tree(s) * Send the header for a block using dynamic Huffman trees: the counts, the * lengths of the bit length codes, the literal tree and the distance tree. * IN assertion: lcodes >= 257, dcodes >= 1, blcodes >= 4. */ -local void send_all_trees(s, lcodes, dcodes, blcodes) - deflate_state *s; - int lcodes, dcodes, blcodes; /* number of codes for each tree */ -{ +local void send_all_trees(deflate_state *s, int lcodes, int dcodes, + int blcodes) { int rank; /* index in bl_order */ Assert (lcodes >= 257 && dcodes >= 1 && blcodes >= 4, "not enough codes"); Assert (lcodes <= L_CODES && dcodes <= D_CODES && blcodes <= BL_CODES, "too many codes"); Tracev((stderr, "\nbl counts: ")); - send_bits(s, lcodes-257, 5); /* not +255 as stated in appnote.txt */ - send_bits(s, dcodes-1, 5); - send_bits(s, blcodes-4, 4); /* not -3 as stated in appnote.txt */ + send_bits(s, lcodes - 257, 5); /* not +255 as stated in appnote.txt */ + send_bits(s, dcodes - 1, 5); + send_bits(s, blcodes - 4, 4); /* not -3 as stated in appnote.txt */ for (rank = 0; rank < blcodes; rank++) { Tracev((stderr, "\nbl code %2d ", bl_order[rank])); send_bits(s, s->bl_tree[bl_order[rank]].Len, 3); } Tracev((stderr, "\nbl tree: sent %ld", s->bits_sent)); - send_tree(s, (ct_data *)s->dyn_ltree, lcodes-1); /* literal tree */ + send_tree(s, (ct_data *)s->dyn_ltree, lcodes - 1); /* literal tree */ Tracev((stderr, "\nlit tree: sent %ld", s->bits_sent)); - send_tree(s, (ct_data *)s->dyn_dtree, dcodes-1); /* distance tree */ + send_tree(s, (ct_data *)s->dyn_dtree, dcodes - 1); /* distance tree */ Tracev((stderr, "\ndist tree: sent %ld", s->bits_sent)); } /* =========================================================================== * Send a stored block */ -void ZLIB_INTERNAL _tr_stored_block(s, buf, stored_len, last) - deflate_state *s; - charf *buf; /* input block */ - ulg stored_len; /* length of input block */ - int last; /* one if this is the last block for a file */ -{ - send_bits(s, (STORED_BLOCK<<1)+last, 3); /* send block type */ +void ZLIB_INTERNAL _tr_stored_block(deflate_state *s, charf *buf, + ulg stored_len, int last) { + send_bits(s, (STORED_BLOCK<<1) + last, 3); /* send block type */ bi_windup(s); /* align on byte boundary */ put_short(s, (ush)stored_len); put_short(s, (ush)~stored_len); - zmemcpy(s->pending_buf + s->pending, (Bytef *)buf, stored_len); + if (stored_len) + zmemcpy(s->pending_buf + s->pending, (Bytef *)buf, stored_len); s->pending += stored_len; #ifdef ZLIB_DEBUG s->compressed_len = (s->compressed_len + 3 + 7) & (ulg)~7L; s->compressed_len += (stored_len + 4) << 3; s->bits_sent += 2*16; - s->bits_sent += stored_len<<3; + s->bits_sent += stored_len << 3; #endif } /* =========================================================================== * Flush the bits in the bit buffer to pending output (leaves at most 7 bits) */ -void ZLIB_INTERNAL _tr_flush_bits(s) - deflate_state *s; -{ +void ZLIB_INTERNAL _tr_flush_bits(deflate_state *s) { bi_flush(s); } /* =========================================================================== * Send one empty static block to give enough lookahead for inflate. * This takes 10 bits, of which 7 may remain in the bit buffer. */ -void ZLIB_INTERNAL _tr_align(s) - deflate_state *s; -{ +void ZLIB_INTERNAL _tr_align(deflate_state *s) { send_bits(s, STATIC_TREES<<1, 3); send_code(s, END_BLOCK, static_ltree); #ifdef ZLIB_DEBUG s->compressed_len += 10L; /* 3 for block type, 7 for EOB */ @@ -904,17 +892,109 @@ void ZLIB_INTERNAL _tr_align(s) bi_flush(s); } /* =========================================================================== + * Send the block data compressed using the given Huffman trees + */ +local void compress_block(deflate_state *s, const ct_data *ltree, + const ct_data *dtree) { + unsigned dist; /* distance of matched string */ + int lc; /* match length or unmatched char (if dist == 0) */ + unsigned sx = 0; /* running index in symbol buffers */ + unsigned code; /* the code to send */ + int extra; /* number of extra bits to send */ + + if (s->sym_next != 0) do { +#ifdef LIT_MEM + dist = s->d_buf[sx]; + lc = s->l_buf[sx++]; +#else + dist = s->sym_buf[sx++] & 0xff; + dist += (unsigned)(s->sym_buf[sx++] & 0xff) << 8; + lc = s->sym_buf[sx++]; +#endif + if (dist == 0) { + send_code(s, lc, ltree); /* send a literal byte */ + Tracecv(isgraph(lc), (stderr," '%c' ", lc)); + } else { + /* Here, lc is the match length - MIN_MATCH */ + code = _length_code[lc]; + send_code(s, code + LITERALS + 1, ltree); /* send length code */ + extra = extra_lbits[code]; + if (extra != 0) { + lc -= base_length[code]; + send_bits(s, lc, extra); /* send the extra length bits */ + } + dist--; /* dist is now the match distance - 1 */ + code = d_code(dist); + Assert (code < D_CODES, "bad d_code"); + + send_code(s, code, dtree); /* send the distance code */ + extra = extra_dbits[code]; + if (extra != 0) { + dist -= (unsigned)base_dist[code]; + send_bits(s, dist, extra); /* send the extra distance bits */ + } + } /* literal or match pair ? */ + + /* Check for no overlay of pending_buf on needed symbols */ +#ifdef LIT_MEM + Assert(s->pending < 2 * (s->lit_bufsize + sx), "pendingBuf overflow"); +#else + Assert(s->pending < s->lit_bufsize + sx, "pendingBuf overflow"); +#endif + + } while (sx < s->sym_next); + + send_code(s, END_BLOCK, ltree); +} + +/* =========================================================================== + * Check if the data type is TEXT or BINARY, using the following algorithm: + * - TEXT if the two conditions below are satisfied: + * a) There are no non-portable control characters belonging to the + * "block list" (0..6, 14..25, 28..31). + * b) There is at least one printable character belonging to the + * "allow list" (9 {TAB}, 10 {LF}, 13 {CR}, 32..255). + * - BINARY otherwise. + * - The following partially-portable control characters form a + * "gray list" that is ignored in this detection algorithm: + * (7 {BEL}, 8 {BS}, 11 {VT}, 12 {FF}, 26 {SUB}, 27 {ESC}). + * IN assertion: the fields Freq of dyn_ltree are set. + */ +local int detect_data_type(deflate_state *s) { + /* block_mask is the bit mask of block-listed bytes + * set bits 0..6, 14..25, and 28..31 + * 0xf3ffc07f = binary 11110011111111111100000001111111 + */ + unsigned long block_mask = 0xf3ffc07fUL; + int n; + + /* Check for non-textual ("block-listed") bytes. */ + for (n = 0; n <= 31; n++, block_mask >>= 1) + if ((block_mask & 1) && (s->dyn_ltree[n].Freq != 0)) + return Z_BINARY; + + /* Check for textual ("allow-listed") bytes. */ + if (s->dyn_ltree[9].Freq != 0 || s->dyn_ltree[10].Freq != 0 + || s->dyn_ltree[13].Freq != 0) + return Z_TEXT; + for (n = 32; n < LITERALS; n++) + if (s->dyn_ltree[n].Freq != 0) + return Z_TEXT; + + /* There are no "block-listed" or "allow-listed" bytes: + * this stream either is empty or has tolerated ("gray-listed") bytes only. + */ + return Z_BINARY; +} + +/* =========================================================================== * Determine the best encoding for the current block: dynamic trees, static * trees or store, and write out the encoded block. */ -void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last) - deflate_state *s; - charf *buf; /* input block, or NULL if too old */ - ulg stored_len; /* length of input block */ - int last; /* one if this is the last block for a file */ -{ +void ZLIB_INTERNAL _tr_flush_block(deflate_state *s, charf *buf, + ulg stored_len, int last) { ulg opt_lenb, static_lenb; /* opt_len and static_len in bytes */ int max_blindex = 0; /* index of last bit length code of non zero freq */ /* Build the Huffman trees unless a stored block is forced */ @@ -941,16 +1021,19 @@ void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last) */ max_blindex = build_bl_tree(s); /* Determine the best encoding. Compute the block lengths in bytes. */ - opt_lenb = (s->opt_len+3+7)>>3; - static_lenb = (s->static_len+3+7)>>3; + opt_lenb = (s->opt_len + 3 + 7) >> 3; + static_lenb = (s->static_len + 3 + 7) >> 3; Tracev((stderr, "\nopt %lu(%lu) stat %lu(%lu) stored %lu lit %u ", opt_lenb, s->opt_len, static_lenb, s->static_len, stored_len, - s->last_lit)); + s->sym_next / 3)); - if (static_lenb <= opt_lenb) opt_lenb = static_lenb; +#ifndef FORCE_STATIC + if (static_lenb <= opt_lenb || s->strategy == Z_FIXED) +#endif + opt_lenb = static_lenb; } else { Assert(buf != (char*)0, "lost buf"); opt_lenb = static_lenb = stored_len + 5; /* force a stored block */ @@ -958,9 +1041,9 @@ void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last) #ifdef FORCE_STORED if (buf != (char*)0) { /* force stored block */ #else - if (stored_len+4 <= opt_lenb && buf != (char*)0) { + if (stored_len + 4 <= opt_lenb && buf != (char*)0) { /* 4: two words for the lengths */ #endif /* The test buf != NULL is only necessary if LIT_BUFSIZE > WSIZE. * Otherwise we can't have processed more than WSIZE input bytes since @@ -969,23 +1052,19 @@ void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last) * transform a block into a stored block. */ _tr_stored_block(s, buf, stored_len, last); -#ifdef FORCE_STATIC - } else if (static_lenb >= 0) { /* force static trees */ -#else - } else if (s->strategy == Z_FIXED || static_lenb == opt_lenb) { -#endif - send_bits(s, (STATIC_TREES<<1)+last, 3); + } else if (static_lenb == opt_lenb) { + send_bits(s, (STATIC_TREES<<1) + last, 3); compress_block(s, (const ct_data *)static_ltree, (const ct_data *)static_dtree); #ifdef ZLIB_DEBUG s->compressed_len += 3 + s->static_len; #endif } else { - send_bits(s, (DYN_TREES<<1)+last, 3); - send_all_trees(s, s->l_desc.max_code+1, s->d_desc.max_code+1, - max_blindex+1); + send_bits(s, (DYN_TREES<<1) + last, 3); + send_all_trees(s, s->l_desc.max_code + 1, s->d_desc.max_code + 1, + max_blindex + 1); compress_block(s, (const ct_data *)s->dyn_ltree, (const ct_data *)s->dyn_dtree); #ifdef ZLIB_DEBUG s->compressed_len += 3 + s->opt_len; @@ -1002,23 +1081,25 @@ void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last) #ifdef ZLIB_DEBUG s->compressed_len += 7; /* align on byte boundary */ #endif } - Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len>>3, - s->compressed_len-7*last)); + Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len >> 3, + s->compressed_len - 7*last)); } /* =========================================================================== * Save the match info and tally the frequency counts. Return true if * the current block must be flushed. */ -int ZLIB_INTERNAL _tr_tally (s, dist, lc) - deflate_state *s; - unsigned dist; /* distance of matched string */ - unsigned lc; /* match length-MIN_MATCH or unmatched char (if dist==0) */ -{ - s->d_buf[s->last_lit] = (ush)dist; - s->l_buf[s->last_lit++] = (uch)lc; +int ZLIB_INTERNAL _tr_tally(deflate_state *s, unsigned dist, unsigned lc) { +#ifdef LIT_MEM + s->d_buf[s->sym_next] = (ush)dist; + s->l_buf[s->sym_next++] = (uch)lc; +#else + s->sym_buf[s->sym_next++] = (uch)dist; + s->sym_buf[s->sym_next++] = (uch)(dist >> 8); + s->sym_buf[s->sym_next++] = (uch)lc; +#endif if (dist == 0) { /* lc is the unmatched char */ s->dyn_ltree[lc].Freq++; } else { @@ -1028,176 +1109,9 @@ int ZLIB_INTERNAL _tr_tally (s, dist, lc) Assert((ush)dist < (ush)MAX_DIST(s) && (ush)lc <= (ush)(MAX_MATCH-MIN_MATCH) && (ush)d_code(dist) < (ush)D_CODES, "_tr_tally: bad match"); - s->dyn_ltree[_length_code[lc]+LITERALS+1].Freq++; + s->dyn_ltree[_length_code[lc] + LITERALS + 1].Freq++; s->dyn_dtree[d_code(dist)].Freq++; } - -#ifdef TRUNCATE_BLOCK - /* Try to guess if it is profitable to stop the current block here */ - if ((s->last_lit & 0x1fff) == 0 && s->level > 2) { - /* Compute an upper bound for the compressed length */ - ulg out_length = (ulg)s->last_lit*8L; - ulg in_length = (ulg)((long)s->strstart - s->block_start); - int dcode; - for (dcode = 0; dcode < D_CODES; dcode++) { - out_length += (ulg)s->dyn_dtree[dcode].Freq * - (5L+extra_dbits[dcode]); - } - out_length >>= 3; - Tracev((stderr,"\nlast_lit %u, in %ld, out ~%ld(%ld%%) ", - s->last_lit, in_length, out_length, - 100L - out_length*100L/in_length)); - if (s->matches < s->last_lit/2 && out_length < in_length/2) return 1; - } -#endif - return (s->last_lit == s->lit_bufsize-1); - /* We avoid equality with lit_bufsize because of wraparound at 64K - * on 16 bit machines and because stored blocks are restricted to - * 64K-1 bytes. - */ -} - -/* =========================================================================== - * Send the block data compressed using the given Huffman trees - */ -local void compress_block(s, ltree, dtree) - deflate_state *s; - const ct_data *ltree; /* literal tree */ - const ct_data *dtree; /* distance tree */ -{ - unsigned dist; /* distance of matched string */ - int lc; /* match length or unmatched char (if dist == 0) */ - unsigned lx = 0; /* running index in l_buf */ - unsigned code; /* the code to send */ - int extra; /* number of extra bits to send */ - - if (s->last_lit != 0) do { - dist = s->d_buf[lx]; - lc = s->l_buf[lx++]; - if (dist == 0) { - send_code(s, lc, ltree); /* send a literal byte */ - Tracecv(isgraph(lc), (stderr," '%c' ", lc)); - } else { - /* Here, lc is the match length - MIN_MATCH */ - code = _length_code[lc]; - send_code(s, code+LITERALS+1, ltree); /* send the length code */ - extra = extra_lbits[code]; - if (extra != 0) { - lc -= base_length[code]; - send_bits(s, lc, extra); /* send the extra length bits */ - } - dist--; /* dist is now the match distance - 1 */ - code = d_code(dist); - Assert (code < D_CODES, "bad d_code"); - - send_code(s, code, dtree); /* send the distance code */ - extra = extra_dbits[code]; - if (extra != 0) { - dist -= (unsigned)base_dist[code]; - send_bits(s, dist, extra); /* send the extra distance bits */ - } - } /* literal or match pair ? */ - - /* Check that the overlay between pending_buf and d_buf+l_buf is ok: */ - Assert((uInt)(s->pending) < s->lit_bufsize + 2*lx, - "pendingBuf overflow"); - - } while (lx < s->last_lit); - - send_code(s, END_BLOCK, ltree); -} - -/* =========================================================================== - * Check if the data type is TEXT or BINARY, using the following algorithm: - * - TEXT if the two conditions below are satisfied: - * a) There are no non-portable control characters belonging to the - * "black list" (0..6, 14..25, 28..31). - * b) There is at least one printable character belonging to the - * "white list" (9 {TAB}, 10 {LF}, 13 {CR}, 32..255). - * - BINARY otherwise. - * - The following partially-portable control characters form a - * "gray list" that is ignored in this detection algorithm: - * (7 {BEL}, 8 {BS}, 11 {VT}, 12 {FF}, 26 {SUB}, 27 {ESC}). - * IN assertion: the fields Freq of dyn_ltree are set. - */ -local int detect_data_type(s) - deflate_state *s; -{ - /* black_mask is the bit mask of black-listed bytes - * set bits 0..6, 14..25, and 28..31 - * 0xf3ffc07f = binary 11110011111111111100000001111111 - */ - unsigned long black_mask = 0xf3ffc07fUL; - int n; - - /* Check for non-textual ("black-listed") bytes. */ - for (n = 0; n <= 31; n++, black_mask >>= 1) - if ((black_mask & 1) && (s->dyn_ltree[n].Freq != 0)) - return Z_BINARY; - - /* Check for textual ("white-listed") bytes. */ - if (s->dyn_ltree[9].Freq != 0 || s->dyn_ltree[10].Freq != 0 - || s->dyn_ltree[13].Freq != 0) - return Z_TEXT; - for (n = 32; n < LITERALS; n++) - if (s->dyn_ltree[n].Freq != 0) - return Z_TEXT; - - /* There are no "black-listed" or "white-listed" bytes: - * this stream either is empty or has tolerated ("gray-listed") bytes only. - */ - return Z_BINARY; -} - -/* =========================================================================== - * Reverse the first len bits of a code, using straightforward code (a faster - * method would use a table) - * IN assertion: 1 <= len <= 15 - */ -local unsigned bi_reverse(code, len) - unsigned code; /* the value to invert */ - int len; /* its bit length */ -{ - register unsigned res = 0; - do { - res |= code & 1; - code >>= 1, res <<= 1; - } while (--len > 0); - return res >> 1; -} - -/* =========================================================================== - * Flush the bit buffer, keeping at most 7 bits in it. - */ -local void bi_flush(s) - deflate_state *s; -{ - if (s->bi_valid == 16) { - put_short(s, s->bi_buf); - s->bi_buf = 0; - s->bi_valid = 0; - } else if (s->bi_valid >= 8) { - put_byte(s, (Byte)s->bi_buf); - s->bi_buf >>= 8; - s->bi_valid -= 8; - } -} - -/* =========================================================================== - * Flush the bit buffer and align the output on a byte boundary - */ -local void bi_windup(s) - deflate_state *s; -{ - if (s->bi_valid > 8) { - put_short(s, s->bi_buf); - } else if (s->bi_valid > 0) { - put_byte(s, (Byte)s->bi_buf); - } - s->bi_buf = 0; - s->bi_valid = 0; -#ifdef ZLIB_DEBUG - s->bits_sent = (s->bits_sent+7) & ~7; -#endif + return (s->sym_next == s->sym_end); } |