/* trees.c -- output deflated data using Huffman coding
- * Copyright (C) 1995-1998 Jean-loup Gailly
- * For conditions of distribution and use, see copyright notice in zlib.h
+ * Copyright (C) 1995-2012 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
*/
/*
/* @(#) $Id$ */
+/* #define GEN_TREES_H */
+
#include "deflate.h"
#ifdef DEBUG
* probability, to avoid transmitting the lengths for unused bit length codes.
*/
-#define Buf_size (8 * 2*sizeof(char))
-/* Number of bits used within bi_buf. (bi_buf might be implemented on
- * more than 16 bits on some systems.)
- */
-
/* ===========================================================================
* Local data. These are initialized only once.
*/
+#define DIST_CODE_LEN 512 /* see definition of array dist_code below */
+
+#if defined(GEN_TREES_H) || !defined(STDC)
+/* non ANSI compilers may not accept trees.h */
+
local ct_data static_ltree[L_CODES+2];
/* The static literal tree. Since the bit lengths are imposed, there is no
* need for the L_CODES extra codes used during heap construction. However
* 5 bits.)
*/
-local uch dist_code[512];
-/* distance codes. The first 256 values correspond to the distances
+uch _dist_code[DIST_CODE_LEN];
+/* Distance codes. The first 256 values correspond to the distances
* 3 .. 258, the last 256 values correspond to the top 8 bits of
* the 15 bit distances.
*/
-local uch length_code[MAX_MATCH-MIN_MATCH+1];
+uch _length_code[MAX_MATCH-MIN_MATCH+1];
/* length code for each normalized match length (0 == MIN_MATCH) */
local int base_length[LENGTH_CODES];
local int base_dist[D_CODES];
/* First normalized distance for each code (0 = distance of 1) */
+#else
+# include "trees.h"
+#endif /* GEN_TREES_H */
+
struct static_tree_desc_s {
- ct_data *static_tree; /* static tree or NULL */
+ const ct_data *static_tree; /* static tree or NULL */
const intf *extra_bits; /* extra bits for each code or NULL */
int extra_base; /* base index for extra_bits */
int elems; /* max number of elements in the tree */
{static_dtree, extra_dbits, 0, D_CODES, MAX_BITS};
local static_tree_desc static_bl_desc =
-{(ct_data *)0, extra_blbits, 0, BL_CODES, MAX_BL_BITS};
+{(const ct_data *)0, extra_blbits, 0, BL_CODES, MAX_BL_BITS};
/* ===========================================================================
* Local (static) routines in this file.
int blcodes));
local void compress_block OF((deflate_state *s, ct_data *ltree,
ct_data *dtree));
-local void set_data_type OF((deflate_state *s));
+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));
local void copy_block OF((deflate_state *s, charf *buf, unsigned len,
int header));
+#ifdef GEN_TREES_H
+local void gen_trees_header OF((void));
+#endif
+
#ifndef DEBUG
# define send_code(s, c, tree) send_bits(s, tree[c].Code, tree[c].Len)
/* Send a code of the given tree. c and tree must not have side effects */
send_bits(s, tree[c].Code, tree[c].Len); }
#endif
-#define d_code(dist) \
- ((dist) < 256 ? dist_code[dist] : dist_code[256+((dist)>>7)])
-/* Mapping from a distance to a distance code. dist is the distance - 1 and
- * must not have side effects. dist_code[256] and dist_code[257] are never
- * used.
- */
-
/* ===========================================================================
* Output a short LSB first on the stream.
* IN assertion: there is enough room in pendingBuf.
* unused bits in value.
*/
if (s->bi_valid > (int)Buf_size - length) {
- s->bi_buf |= (value << s->bi_valid);
+ s->bi_buf |= (ush)value << s->bi_valid;
put_short(s, s->bi_buf);
s->bi_buf = (ush)value >> (Buf_size - s->bi_valid);
s->bi_valid += length - Buf_size;
} else {
- s->bi_buf |= value << s->bi_valid;
+ s->bi_buf |= (ush)value << s->bi_valid;
s->bi_valid += length;
}
}
{ int len = length;\
if (s->bi_valid > (int)Buf_size - len) {\
int val = value;\
- s->bi_buf |= (val << s->bi_valid);\
+ s->bi_buf |= (ush)val << s->bi_valid;\
put_short(s, s->bi_buf);\
s->bi_buf = (ush)val >> (Buf_size - s->bi_valid);\
s->bi_valid += len - Buf_size;\
} else {\
- s->bi_buf |= (value) << s->bi_valid;\
+ s->bi_buf |= (ush)(value) << s->bi_valid;\
s->bi_valid += len;\
}\
}
#endif /* DEBUG */
-#define MAX(a,b) (a >= b ? a : b)
/* the arguments must not have side effects */
/* ===========================================================================
- * Initialize the various 'constant' tables. In a multi-threaded environment,
- * this function may be called by two threads concurrently, but this is
- * harmless since both invocations do exactly the same thing.
+ * Initialize the various 'constant' tables.
*/
local void tr_static_init()
{
+#if defined(GEN_TREES_H) || !defined(STDC)
static int static_init_done = 0;
int n; /* iterates over tree elements */
int bits; /* bit counter */
if (static_init_done) return;
+ /* For some embedded targets, global variables are not initialized: */
+#ifdef NO_INIT_GLOBAL_POINTERS
+ static_l_desc.static_tree = static_ltree;
+ static_l_desc.extra_bits = extra_lbits;
+ static_d_desc.static_tree = static_dtree;
+ static_d_desc.extra_bits = extra_dbits;
+ static_bl_desc.extra_bits = extra_blbits;
+#endif
+
/* 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++) {
- length_code[length++] = (uch)code;
+ _length_code[length++] = (uch)code;
}
}
Assert (length == 256, "tr_static_init: length != 256");
* 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++) {
- dist_code[dist++] = (uch)code;
+ _dist_code[dist++] = (uch)code;
}
}
Assert (dist == 256, "tr_static_init: dist != 256");
for ( ; code < D_CODES; code++) {
base_dist[code] = dist << 7;
for (n = 0; n < (1<<(extra_dbits[code]-7)); n++) {
- dist_code[256 + dist++] = (uch)code;
+ _dist_code[256 + dist++] = (uch)code;
}
}
Assert (dist == 256, "tr_static_init: 256+dist != 512");
static_dtree[n].Code = bi_reverse((unsigned)n, 5);
}
static_init_done = 1;
+
+# ifdef GEN_TREES_H
+ gen_trees_header();
+# endif
+#endif /* defined(GEN_TREES_H) || !defined(STDC) */
}
+/* ===========================================================================
+ * Genererate the file trees.h describing the static trees.
+ */
+#ifdef GEN_TREES_H
+# ifndef DEBUG
+# include <stdio.h>
+# endif
+
+# define SEPARATOR(i, last, width) \
+ ((i) == (last)? "\n};\n\n" : \
+ ((i) % (width) == (width)-1 ? ",\n" : ", "))
+
+void gen_trees_header()
+{
+ FILE *header = fopen("trees.h", "w");
+ int i;
+
+ Assert (header != NULL, "Can't open trees.h");
+ fprintf(header,
+ "/* header created automatically with -DGEN_TREES_H */\n\n");
+
+ fprintf(header, "local const ct_data static_ltree[L_CODES+2] = {\n");
+ for (i = 0; i < L_CODES+2; i++) {
+ fprintf(header, "{{%3u},{%3u}}%s", static_ltree[i].Code,
+ static_ltree[i].Len, SEPARATOR(i, L_CODES+1, 5));
+ }
+
+ fprintf(header, "local const ct_data static_dtree[D_CODES] = {\n");
+ for (i = 0; i < D_CODES; i++) {
+ fprintf(header, "{{%2u},{%2u}}%s", static_dtree[i].Code,
+ static_dtree[i].Len, SEPARATOR(i, D_CODES-1, 5));
+ }
+
+ fprintf(header, "const uch ZLIB_INTERNAL _dist_code[DIST_CODE_LEN] = {\n");
+ for (i = 0; i < DIST_CODE_LEN; i++) {
+ fprintf(header, "%2u%s", _dist_code[i],
+ SEPARATOR(i, DIST_CODE_LEN-1, 20));
+ }
+
+ fprintf(header,
+ "const uch ZLIB_INTERNAL _length_code[MAX_MATCH-MIN_MATCH+1]= {\n");
+ for (i = 0; i < MAX_MATCH-MIN_MATCH+1; i++) {
+ fprintf(header, "%2u%s", _length_code[i],
+ SEPARATOR(i, MAX_MATCH-MIN_MATCH, 20));
+ }
+
+ fprintf(header, "local const int base_length[LENGTH_CODES] = {\n");
+ for (i = 0; i < LENGTH_CODES; i++) {
+ fprintf(header, "%1u%s", base_length[i],
+ SEPARATOR(i, LENGTH_CODES-1, 20));
+ }
+
+ fprintf(header, "local const int base_dist[D_CODES] = {\n");
+ for (i = 0; i < D_CODES; i++) {
+ fprintf(header, "%5u%s", base_dist[i],
+ SEPARATOR(i, D_CODES-1, 10));
+ }
+
+ fclose(header);
+}
+#endif /* GEN_TREES_H */
+
/* ===========================================================================
* Initialize the tree data structures for a new zlib stream.
*/
-void _tr_init(s)
+void ZLIB_INTERNAL _tr_init(s)
deflate_state *s;
{
tr_static_init();
- s->compressed_len = 0L;
-
s->l_desc.dyn_tree = s->dyn_ltree;
s->l_desc.stat_desc = &static_l_desc;
s->bi_buf = 0;
s->bi_valid = 0;
- s->last_eob_len = 8; /* enough lookahead for inflate */
#ifdef DEBUG
+ s->compressed_len = 0L;
s->bits_sent = 0L;
#endif
deflate_state *s;
tree_desc *desc; /* the tree descriptor */
{
- ct_data *tree = desc->dyn_tree;
- int max_code = desc->max_code;
- ct_data *stree = desc->stat_desc->static_tree;
- const intf *extra = desc->stat_desc->extra_bits;
- int base = desc->stat_desc->extra_base;
- int max_length = desc->stat_desc->max_length;
+ 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;
+ int base = desc->stat_desc->extra_base;
+ int max_length = desc->stat_desc->max_length;
int h; /* heap index */
int n, m; /* iterate over the tree elements */
int bits; /* bit length */
while (n != 0) {
m = s->heap[--h];
if (m > max_code) continue;
- if (tree[m].Len != (unsigned) bits) {
+ if ((unsigned) tree[m].Len != (unsigned) bits) {
Trace((stderr,"code %d bits %d->%d\n", m, tree[m].Len, bits));
s->opt_len += ((long)bits - (long)tree[m].Len)
*(long)tree[m].Freq;
deflate_state *s;
tree_desc *desc; /* the tree descriptor */
{
- ct_data *tree = desc->dyn_tree;
- ct_data *stree = desc->stat_desc->static_tree;
- int elems = desc->stat_desc->elems;
+ 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 */
/* Create a new node father of n and m */
tree[node].Freq = tree[n].Freq + tree[m].Freq;
- s->depth[node] = (uch) (MAX(s->depth[n], s->depth[m]) + 1);
+ s->depth[node] = (uch)((s->depth[n] >= s->depth[m] ?
+ s->depth[n] : s->depth[m]) + 1);
tree[n].Dad = tree[m].Dad = (ush)node;
#ifdef DUMP_BL_TREE
if (tree == s->bl_tree) {
/* ===========================================================================
* Send a stored block
*/
-void _tr_stored_block(s, buf, stored_len, eof)
+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 eof; /* true if this is the last block for a file */
+ int last; /* one if this is the last block for a file */
{
- send_bits(s, (STORED_BLOCK<<1)+eof, 3); /* send block type */
+ send_bits(s, (STORED_BLOCK<<1)+last, 3); /* send block type */
+#ifdef DEBUG
s->compressed_len = (s->compressed_len + 3 + 7) & (ulg)~7L;
s->compressed_len += (stored_len + 4) << 3;
-
+#endif
copy_block(s, buf, (unsigned)stored_len, 1); /* with header */
}
+/* ===========================================================================
+ * 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;
+{
+ 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.
- * The current inflate code requires 9 bits of lookahead. If the
- * last two codes for the previous block (real code plus EOB) were coded
- * on 5 bits or less, inflate may have only 5+3 bits of lookahead to decode
- * the last real code. In this case we send two empty static blocks instead
- * of one. (There are no problems if the previous block is stored or fixed.)
- * To simplify the code, we assume the worst case of last real code encoded
- * on one bit only.
*/
-void _tr_align(s)
+void ZLIB_INTERNAL _tr_align(s)
deflate_state *s;
{
send_bits(s, STATIC_TREES<<1, 3);
send_code(s, END_BLOCK, static_ltree);
+#ifdef DEBUG
s->compressed_len += 10L; /* 3 for block type, 7 for EOB */
+#endif
bi_flush(s);
- /* Of the 10 bits for the empty block, we have already sent
- * (10 - bi_valid) bits. The lookahead for the last real code (before
- * the EOB of the previous block) was thus at least one plus the length
- * of the EOB plus what we have just sent of the empty static block.
- */
- if (1 + s->last_eob_len + 10 - s->bi_valid < 9) {
- send_bits(s, STATIC_TREES<<1, 3);
- send_code(s, END_BLOCK, static_ltree);
- s->compressed_len += 10L;
- bi_flush(s);
- }
- s->last_eob_len = 7;
}
/* ===========================================================================
* Determine the best encoding for the current block: dynamic trees, static
- * trees or store, and output the encoded block to the zip file. This function
- * returns the total compressed length for the file so far.
+ * trees or store, and output the encoded block to the zip file.
*/
-ulg _tr_flush_block(s, buf, stored_len, eof)
+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 eof; /* true if this is the last block for a file */
+ int last; /* one if this is the last block for a file */
{
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 */
if (s->level > 0) {
- /* Check if the file is ascii or binary */
- if (s->data_type == Z_UNKNOWN) set_data_type(s);
+ /* Check if the file is binary or text */
+ if (s->strm->data_type == Z_UNKNOWN)
+ s->strm->data_type = detect_data_type(s);
- /* Construct the literal and distance trees */
- build_tree(s, (tree_desc *)(&(s->l_desc)));
- Tracev((stderr, "\nlit data: dyn %ld, stat %ld", s->opt_len,
- s->static_len));
+ /* Construct the literal and distance trees */
+ build_tree(s, (tree_desc *)(&(s->l_desc)));
+ Tracev((stderr, "\nlit data: dyn %ld, stat %ld", s->opt_len,
+ s->static_len));
- build_tree(s, (tree_desc *)(&(s->d_desc)));
- Tracev((stderr, "\ndist data: dyn %ld, stat %ld", s->opt_len,
- s->static_len));
- /* At this point, opt_len and static_len are the total bit lengths of
- * the compressed block data, excluding the tree representations.
- */
+ build_tree(s, (tree_desc *)(&(s->d_desc)));
+ Tracev((stderr, "\ndist data: dyn %ld, stat %ld", s->opt_len,
+ s->static_len));
+ /* At this point, opt_len and static_len are the total bit lengths of
+ * the compressed block data, excluding the tree representations.
+ */
- /* Build the bit length tree for the above two trees, and get the index
- * in bl_order of the last bit length code to send.
- */
- max_blindex = build_bl_tree(s);
+ /* Build the bit length tree for the above two trees, and get the index
+ * in bl_order of the last bit length code to send.
+ */
+ max_blindex = build_bl_tree(s);
- /* Determine the best encoding. Compute first the block length in bytes*/
- opt_lenb = (s->opt_len+3+7)>>3;
- static_lenb = (s->static_len+3+7)>>3;
+ /* 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;
- 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));
+ 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));
- if (static_lenb <= opt_lenb) opt_lenb = static_lenb;
+ if (static_lenb <= opt_lenb) opt_lenb = static_lenb;
} else {
Assert(buf != (char*)0, "lost buf");
- opt_lenb = static_lenb = stored_len + 5; /* force a stored block */
+ opt_lenb = static_lenb = stored_len + 5; /* force a stored block */
}
- /* If compression failed and this is the first and last block,
- * and if the .zip file can be seeked (to rewrite the local header),
- * the whole file is transformed into a stored file:
- */
-#ifdef STORED_FILE_OK
-# ifdef FORCE_STORED_FILE
- if (eof && s->compressed_len == 0L) { /* force stored file */
-# else
- if (stored_len <= opt_lenb && eof && s->compressed_len==0L && seekable()) {
-# endif
- /* Since LIT_BUFSIZE <= 2*WSIZE, the input data must be there: */
- if (buf == (charf*)0) error ("block vanished");
-
- copy_block(buf, (unsigned)stored_len, 0); /* without header */
- s->compressed_len = stored_len << 3;
- s->method = STORED;
- } else
-#endif /* STORED_FILE_OK */
-
#ifdef FORCE_STORED
if (buf != (char*)0) { /* force stored block */
#else
* successful. If LIT_BUFSIZE <= WSIZE, it is never too late to
* transform a block into a stored block.
*/
- _tr_stored_block(s, buf, stored_len, eof);
+ _tr_stored_block(s, buf, stored_len, last);
#ifdef FORCE_STATIC
} else if (static_lenb >= 0) { /* force static trees */
#else
- } else if (static_lenb == opt_lenb) {
+ } else if (s->strategy == Z_FIXED || static_lenb == opt_lenb) {
#endif
- send_bits(s, (STATIC_TREES<<1)+eof, 3);
+ send_bits(s, (STATIC_TREES<<1)+last, 3);
compress_block(s, (ct_data *)static_ltree, (ct_data *)static_dtree);
+#ifdef DEBUG
s->compressed_len += 3 + s->static_len;
+#endif
} else {
- send_bits(s, (DYN_TREES<<1)+eof, 3);
+ 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, (ct_data *)s->dyn_ltree, (ct_data *)s->dyn_dtree);
+#ifdef DEBUG
s->compressed_len += 3 + s->opt_len;
+#endif
}
Assert (s->compressed_len == s->bits_sent, "bad compressed size");
+ /* The above check is made mod 2^32, for files larger than 512 MB
+ * and uLong implemented on 32 bits.
+ */
init_block(s);
- if (eof) {
+ if (last) {
bi_windup(s);
+#ifdef DEBUG
s->compressed_len += 7; /* align on byte boundary */
+#endif
}
Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len>>3,
- s->compressed_len-7*eof));
-
- return 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 _tr_tally (s, dist, lc)
+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) */
(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->level > 2 && (s->last_lit & 0xfff) == 0) {
+ 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);
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
Tracecv(isgraph(lc), (stderr," '%c' ", lc));
} else {
/* Here, lc is the match length - MIN_MATCH */
- code = length_code[lc];
+ code = _length_code[lc];
send_code(s, code+LITERALS+1, ltree); /* send the length code */
extra = extra_lbits[code];
if (extra != 0) {
} /* literal or match pair ? */
/* Check that the overlay between pending_buf and d_buf+l_buf is ok: */
- Assert(s->pending < s->lit_bufsize + 2*lx, "pendingBuf overflow");
+ Assert((uInt)(s->pending) < s->lit_bufsize + 2*lx,
+ "pendingBuf overflow");
} while (lx < s->last_lit);
send_code(s, END_BLOCK, ltree);
- s->last_eob_len = ltree[END_BLOCK].Len;
}
/* ===========================================================================
- * Set the data type to ASCII or BINARY, using a crude approximation:
- * binary if more than 20% of the bytes are <= 6 or >= 128, ascii otherwise.
- * IN assertion: the fields freq of dyn_ltree are set and the total of all
- * frequencies does not exceed 64K (to fit in an int on 16 bit machines).
+ * 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 void set_data_type(s)
+local int detect_data_type(s)
deflate_state *s;
{
- int n = 0;
- unsigned ascii_freq = 0;
- unsigned bin_freq = 0;
- while (n < 7) bin_freq += s->dyn_ltree[n++].Freq;
- while (n < 128) ascii_freq += s->dyn_ltree[n++].Freq;
- while (n < LITERALS) bin_freq += s->dyn_ltree[n++].Freq;
- s->data_type = (Byte)(bin_freq > (ascii_freq >> 2) ? Z_BINARY : Z_ASCII);
+ /* 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;
}
/* ===========================================================================
int header; /* true if block header must be written */
{
bi_windup(s); /* align on byte boundary */
- s->last_eob_len = 8; /* enough lookahead for inflate */
if (header) {
- put_short(s, (ush)len);
+ put_short(s, (ush)len);
put_short(s, (ush)~len);
#ifdef DEBUG
s->bits_sent += 2*16;