pod2man -c pristine-bz2 pristine-bz2 > pristine-bz2.1
pod2man -c zgz zgz.pod > zgz.1
-zgz: zgz.c
- gcc -Wall -O2 -lz -o zgz zgz.c
+ZGZ_SOURCES = zgz.c bits.c deflate.c gzip.c trees.c util.c
+zgz: $(ZGZ_SOURCES) gzip.h
+ gcc -Wall -O2 -lz -o $@ $(ZGZ_SOURCES)
install:
install -d $(DESTDIR)/usr/bin
--- /dev/null
+/* bits.c -- output variable-length bit strings
+
+ Copyright (C) 1999 Free Software Foundation, Inc.
+ Copyright (C) 1992-1993 Jean-loup Gailly
+ Copyright (C) 2008 Josh Triplett
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+
+/*
+ * PURPOSE
+ *
+ * Output variable-length bit strings. Compression can be done
+ * to a file or to memory. (The latter is not supported in this version.)
+ *
+ * DISCUSSION
+ *
+ * The PKZIP "deflate" file format interprets compressed file data
+ * as a sequence of bits. Multi-bit strings in the file may cross
+ * byte boundaries without restriction.
+ *
+ * The first bit of each byte is the low-order bit.
+ *
+ * The routines in this file allow a variable-length bit value to
+ * be output right-to-left (useful for literal values). For
+ * left-to-right output (useful for code strings from the tree routines),
+ * the bits must have been reversed first with bi_reverse().
+ *
+ * For in-memory compression, the compressed bit stream goes directly
+ * into the requested output buffer. The input data is read in blocks
+ * by the mem_read() function. The buffer is limited to 64K on 16 bit
+ * machines.
+ *
+ * INTERFACE
+ *
+ * void bi_init (FILE *zipfile)
+ * Initialize the bit string routines.
+ *
+ * void send_bits (int value, int length)
+ * Write out a bit string, taking the source bits right to
+ * left.
+ *
+ * int bi_reverse (int value, int length)
+ * Reverse the bits of a bit string, taking the source bits left to
+ * right and emitting them right to left.
+ *
+ * void bi_windup (void)
+ * Write out any remaining bits in an incomplete byte.
+ *
+ * void copy_block(char *buf, unsigned len, int header)
+ * Copy a stored block to the zip file, storing first the length and
+ * its one's complement if requested.
+ *
+ */
+
+#include "gzip.h"
+
+/* ===========================================================================
+ * Local data used by the "bit string" routines.
+ */
+
+static int zfile; /* output gzip file */
+
+static unsigned short bi_buf;
+/* Output buffer. bits are inserted starting at the bottom (least significant
+ * bits).
+ */
+
+#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.)
+ */
+
+static int bi_valid;
+/* Number of valid bits in bi_buf. All bits above the last valid bit
+ * are always zero.
+ */
+
+int (*read_buf)(char *buf, unsigned size);
+/* Current input function. Set to mem_read for in-memory compression */
+
+/* ===========================================================================
+ * Initialize the bit string routines.
+ */
+void bi_init (int zipfile) /* output zip file, NO_FILE for in-memory compression */
+{
+ zfile = zipfile;
+ bi_buf = 0;
+ bi_valid = 0;
+
+ /* Set the defaults for file compression. They are set by memcompress
+ * for in-memory compression.
+ */
+ if (zfile != NO_FILE) {
+ read_buf = file_read;
+ }
+}
+
+/* ===========================================================================
+ * Send a value on a given number of bits.
+ * IN assertion: length <= 16 and value fits in length bits.
+ */
+void send_bits(int value, /* value to send */
+ int length) /* number of bits */
+{
+ /* 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))
+ * unused bits in value.
+ */
+ if (bi_valid > (int)Buf_size - length) {
+ bi_buf |= (value << bi_valid);
+ put_short(bi_buf);
+ bi_buf = (ush)value >> (Buf_size - bi_valid);
+ bi_valid += length - Buf_size;
+ } else {
+ bi_buf |= value << bi_valid;
+ bi_valid += length;
+ }
+}
+
+/* ===========================================================================
+ * Reverse the first len bits of a code, using straightforward code (a faster
+ * method would use a table)
+ * IN assertion: 1 <= len <= 15
+ */
+unsigned bi_reverse(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;
+}
+
+/* ===========================================================================
+ * Write out any remaining bits in an incomplete byte.
+ */
+void bi_windup(void)
+{
+ if (bi_valid > 8) {
+ put_short(bi_buf);
+ } else if (bi_valid > 0) {
+ put_byte(bi_buf);
+ }
+ bi_buf = 0;
+ bi_valid = 0;
+}
+
+/* ===========================================================================
+ * Copy a stored block to the zip file, storing first the length and its
+ * one's complement if requested.
+ */
+void copy_block(char *buf, /* the input data */
+ unsigned len, /* its length */
+ int header) /* true if block header must be written */
+{
+ bi_windup(); /* align on byte boundary */
+
+ if (header) {
+ put_short((ush)len);
+ put_short((ush)~len);
+ }
+ while (len--) {
+ put_byte(*buf++);
+ }
+}
* Clarify bit in man page about what file formats are supported.
Closes: #507322
- -- Josh Triplett <josh@freedesktop.org> Thu, 27 Nov 2008 19:54:29 -0800
+ [ Josh Triplett ]
+ * zgz: Incorporate a cut-down version of GNU gzip to support the --gnu
+ option. The cut-down gzip includes support for explicitly setting the
+ timestamp and original filename. Closes: #506627
+ * pristine-gz: Now that zgz supports explicitly setting the timestamp and
+ original filename with both --zlib and --gnu, use that support to do all
+ compression using -c with standard input and standard output. This
+ removes the need to create a temporary file with the original filename and
+ set its timestamp, as well as the need to continuously recreate the file
+ when compressed and removed.
+
+ -- Josh Triplett <josh@freedesktop.org> Thu, 11 Dec 2008 14:22:56 -0800
pristine-tar (0.18) unstable; urgency=low
Copyright: © 2007-2008 Joey Hess <joeyh@debian.org>
© 2007 Faidon Liambotis
© 2008 Cyril Brulebois <cyril.brulebois@enst-bretagne.fr>
+ © 2008 Josh Triplett <josh@joshtriplett.org>
License: GPL-2+
On Debian systems, the complete text of the GPL can be found in
/usr/share/common-licenses/GPL.
License: BSD-3
On Debian systems, the complete text of the 3 clause BSD license can be
found in /usr/share/common-licenses/BSD.
+
+Files: gzip.h bits.c deflate.c gzip.c trees.c util.c
+Copyright: © 1997, 1998, 1999, 2001, 2002, 2006, 2007 Free Software Foundation, Inc.
+ © 1992-1993 Jean-loup Gailly
+ © 2008 Josh Triplett
+License: GPL-2+
+ On Debian systems, the complete text of the GPL can be found in
+ /usr/share/common-licenses/GPL.
--- /dev/null
+/* deflate.c -- compress data using the deflation algorithm
+
+ Copyright (C) 1999, 2006 Free Software Foundation, Inc.
+ Copyright (C) 1992-1993 Jean-loup Gailly
+ Copyright (C) 2008 Josh Triplett
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+/*
+ * PURPOSE
+ *
+ * Identify new text as repetitions of old text within a fixed-
+ * length sliding window trailing behind the new text.
+ *
+ * DISCUSSION
+ *
+ * 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 info-zippers for bug reports and testing.
+ *
+ * REFERENCES
+ *
+ * APPNOTE.TXT documentation file in PKZIP 1.93a distribution.
+ *
+ * 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
+ *
+ * INTERFACE
+ *
+ * void lm_init (int pack_level, ush *flags)
+ * Initialize the "longest match" routines for a new file
+ *
+ * void deflate (void)
+ * Processes a new input file. Sets the compressed length, crc,
+ * deflate flags and internal file attributes.
+ */
+
+#include <stdio.h>
+
+#include "gzip.h"
+
+/* ===========================================================================
+ * Configuration parameters
+ */
+
+#define HASH_BITS 15
+
+#define HASH_SIZE (unsigned)(1<<HASH_BITS)
+#define HASH_MASK (HASH_SIZE-1)
+#define WMASK (WSIZE-1)
+/* HASH_SIZE and WSIZE must be powers of two */
+
+#define NIL 0
+/* Tail of hash chains */
+
+#define FAST 4
+#define SLOW 2
+/* speed options for the general purpose bit flag */
+
+#ifndef TOO_FAR
+# define TOO_FAR 4096
+#endif
+/* Matches of length 3 are discarded if their distance exceeds TOO_FAR */
+
+#ifndef RSYNC_WIN
+# define RSYNC_WIN 4096
+#endif
+/* Size of rsync window, must be < MAX_DIST */
+
+#define RSYNC_SUM_MATCH(sum) ((sum) % RSYNC_WIN == 0)
+/* Whether window sum matches magic value */
+
+/* ===========================================================================
+ * Local data used by the "longest match" routines.
+ */
+
+typedef ush Pos;
+typedef unsigned IPos;
+/* A Pos is an index in the character window. We use short instead of int to
+ * save space in the various tables. IPos is used only for parameter passing.
+ */
+
+static uch window[2L*WSIZE];
+/* Sliding window. Input bytes are read into the second half of the window,
+ * and move to the first half later to keep a dictionary of at least WSIZE
+ * bytes. With this organization, matches are limited to a distance of
+ * WSIZE-MAX_MATCH bytes, but this ensures that IO is always
+ * performed with a length multiple of the block size. Also, it limits
+ * the window size to 64K, which is quite useful on MSDOS.
+ */
+
+Pos prev[WSIZE];
+/* Link to older string with same hash index. To limit the size of this
+ * array to 64K, this link is maintained only for the last 32K strings.
+ * An index in this array is thus a window index modulo 32K.
+ */
+
+Pos head[1<<HASH_BITS];
+/* Heads of the hash chains or NIL. */
+
+ulg window_size = (ulg)2*WSIZE;
+/* window size, 2*WSIZE except for MMAP or BIG_MEM, where it is the
+ * input file length plus MIN_LOOKAHEAD.
+ */
+
+long block_start;
+/* window position at the beginning of the current output block. Gets
+ * negative when the window is moved backwards.
+ */
+
+static unsigned ins_h; /* hash index of string to be inserted */
+
+#define H_SHIFT ((HASH_BITS+MIN_MATCH-1)/MIN_MATCH)
+/* Number of bits by which ins_h and del_h must be shifted at each
+ * input step. It must be such that after MIN_MATCH steps, the oldest
+ * byte no longer takes part in the hash key, that is:
+ * H_SHIFT * MIN_MATCH >= HASH_BITS
+ */
+
+unsigned int prev_length;
+/* Length of the best match at previous step. Matches not greater than this
+ * are discarded. This is used in the lazy match evaluation.
+ */
+
+unsigned strstart; /* start of string to insert */
+unsigned match_start; /* start of matching string */
+static int eofile; /* flag set at end of input file */
+static unsigned lookahead; /* number of valid bytes ahead in window */
+
+unsigned max_chain_length;
+/* To speed up deflation, hash chains are never searched beyond this length.
+ * A higher limit improves compression ratio but degrades the speed.
+ */
+
+static unsigned int max_lazy_match;
+/* Attempt to find a better match only when the current match is strictly
+ * smaller than this value. This mechanism is used only for compression
+ * levels >= 4.
+ */
+#define max_insert_length max_lazy_match
+/* Insert new strings in the hash table only if the match length
+ * is not greater than this length. This saves time but degrades compression.
+ * max_insert_length is used only for compression levels <= 3.
+ */
+
+static unsigned good_match;
+/* Use a faster search when the previous match is longer than this */
+
+static ulg rsync_sum; /* rolling sum of rsync window */
+static ulg rsync_chunk_end; /* next rsync sequence point */
+
+/* Values for max_lazy_match, good_match and max_chain_length, depending on
+ * the desired pack level (0..9). The values given below have been tuned to
+ * exclude worst case performance for pathological files. Better values may be
+ * found for specific files.
+ */
+
+typedef struct config {
+ ush good_length; /* reduce lazy search above this match length */
+ ush max_lazy; /* do not perform lazy search above this match length */
+ ush nice_length; /* quit search above this match length */
+ ush max_chain;
+} config;
+
+#ifdef FULL_SEARCH
+# define nice_match MAX_MATCH
+#else
+ int nice_match; /* Stop searching when current match exceeds this */
+#endif
+
+static config configuration_table[10] = {
+/* good lazy nice chain */
+/* 0 */ {0, 0, 0, 0}, /* store only */
+/* 1 */ {4, 4, 8, 4}, /* maximum speed, no lazy matches */
+/* 2 */ {4, 5, 16, 8},
+/* 3 */ {4, 6, 32, 32},
+
+/* 4 */ {4, 4, 16, 16}, /* lazy matches */
+/* 5 */ {8, 16, 32, 32},
+/* 6 */ {8, 16, 128, 128},
+/* 7 */ {8, 32, 128, 256},
+/* 8 */ {32, 128, 258, 1024},
+/* 9 */ {32, 258, 258, 4096}}; /* maximum compression */
+
+/* Note: the deflate() code requires max_lazy >= MIN_MATCH and max_chain >= 4
+ * For deflate_fast() (levels <= 3) good is ignored and lazy has a different
+ * meaning.
+ */
+
+#define EQUAL 0
+/* result of memcmp for equal strings */
+
+/* ===========================================================================
+ * Prototypes for local functions.
+ */
+static void fill_window(void);
+
+ int longest_match(IPos cur_match);
+
+/* ===========================================================================
+ * Update a hash value with the given input byte
+ * IN assertion: all calls to 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(h,c) (h = (((h)<<H_SHIFT) ^ (c)) & HASH_MASK)
+
+/* ===========================================================================
+ * Insert string s 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.
+ * IN assertion: all calls to to INSERT_STRING are made with consecutive
+ * input characters and the first MIN_MATCH bytes of s are valid
+ * (except for the last MIN_MATCH-1 bytes of the input file).
+ */
+#define INSERT_STRING(s, match_head) \
+ (UPDATE_HASH(ins_h, window[(s) + MIN_MATCH-1]), \
+ prev[(s) & WMASK] = match_head = head[ins_h], \
+ head[ins_h] = (s))
+
+/* ===========================================================================
+ * Initialize the "longest match" routines for a new file
+ */
+void lm_init (int pack_level, /* 1: best speed, 9: best compression */
+ ush *flags) /* general purpose bit flag */
+{
+ register unsigned j;
+
+ if (pack_level < 1 || pack_level > 9) gzip_error ("bad pack level");
+
+ /* Initialize the hash table. */
+ memzero((char*)head, HASH_SIZE*sizeof(*head));
+ /* prev will be initialized on the fly */
+
+ /* rsync params */
+ rsync_chunk_end = 0xFFFFFFFFUL;
+ rsync_sum = 0;
+
+ /* Set the default configuration parameters:
+ */
+ max_lazy_match = configuration_table[pack_level].max_lazy;
+ good_match = configuration_table[pack_level].good_length;
+#ifndef FULL_SEARCH
+ nice_match = configuration_table[pack_level].nice_length;
+#endif
+ max_chain_length = configuration_table[pack_level].max_chain;
+ if (pack_level == 1) {
+ *flags |= FAST;
+ } else if (pack_level == 9) {
+ *flags |= SLOW;
+ }
+ /* ??? reduce max_chain_length for binary files */
+
+ strstart = 0;
+ block_start = 0L;
+
+ lookahead = read_buf((char*)window,
+ sizeof(int) <= 2 ? (unsigned)WSIZE : 2*WSIZE);
+
+ if (lookahead == 0 || lookahead == (unsigned)EOF) {
+ eofile = 1, lookahead = 0;
+ return;
+ }
+ eofile = 0;
+ /* Make sure that we always have enough lookahead. This is important
+ * if input comes from a device such as a tty.
+ */
+ while (lookahead < MIN_LOOKAHEAD && !eofile) fill_window();
+
+ ins_h = 0;
+ for (j=0; j<MIN_MATCH-1; j++) UPDATE_HASH(ins_h, window[j]);
+ /* If lookahead < MIN_MATCH, ins_h is garbage, but this is
+ * not important since only literal bytes will be emitted.
+ */
+}
+
+/* ===========================================================================
+ * Set match_start to the longest match starting at the given string and
+ * return its length. Matches shorter or equal to prev_length are discarded,
+ * in which case the result is equal to prev_length and match_start is
+ * garbage.
+ * IN assertions: cur_match is the head of the hash chain for the current
+ * string (strstart) and its distance is <= MAX_DIST, and prev_length >= 1
+ */
+int longest_match(IPos cur_match)
+{
+ unsigned chain_length = max_chain_length; /* max hash chain length */
+ register uch *scan = window + strstart; /* current string */
+ register uch *match; /* matched string */
+ register int len; /* length of current match */
+ int best_len = prev_length; /* best match length so far */
+ IPos limit = strstart > (IPos)MAX_DIST ? strstart - (IPos)MAX_DIST : NIL;
+ /* Stop when cur_match becomes <= limit. To simplify the code,
+ * we prevent matches with the string of window index 0.
+ */
+
+/* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16.
+ * It is easy to get rid of this optimization if necessary.
+ */
+#if HASH_BITS < 8 || MAX_MATCH != 258
+ error: Code too clever
+#endif
+
+#ifdef UNALIGNED_OK
+ /* Compare two bytes at a time. Note: this is not always beneficial.
+ * Try with and without -DUNALIGNED_OK to check.
+ */
+ register uch *strend = window + strstart + MAX_MATCH - 1;
+ register ush scan_start = *(ush*)scan;
+ register ush scan_end = *(ush*)(scan+best_len-1);
+#else
+ register uch *strend = window + strstart + MAX_MATCH;
+ register uch scan_end1 = scan[best_len-1];
+ register uch scan_end = scan[best_len];
+#endif
+
+ /* Do not waste too much time if we already have a good match: */
+ if (prev_length >= good_match) {
+ chain_length >>= 2;
+ }
+ Assert(strstart <= window_size-MIN_LOOKAHEAD, "insufficient lookahead");
+
+ do {
+ Assert(cur_match < strstart, "no future");
+ match = window + cur_match;
+
+ /* Skip to next match if the match length cannot increase
+ * or if the match length is less than 2:
+ */
+#if (defined(UNALIGNED_OK) && MAX_MATCH == 258)
+ /* This code assumes sizeof(unsigned short) == 2. Do not use
+ * UNALIGNED_OK if your compiler uses a different size.
+ */
+ if (*(ush*)(match+best_len-1) != scan_end ||
+ *(ush*)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
+ * 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
+ * necessary to put more guard bytes at the end of the window, or
+ * to check more often for insufficient lookahead.
+ */
+ scan++, match++;
+ do {
+ } while (*(ush*)(scan+=2) == *(ush*)(match+=2) &&
+ *(ush*)(scan+=2) == *(ush*)(match+=2) &&
+ *(ush*)(scan+=2) == *(ush*)(match+=2) &&
+ *(ush*)(scan+=2) == *(ush*)(match+=2) &&
+ scan < strend);
+ /* The funny "do {}" generates better code on most compilers */
+
+ /* Here, scan <= window+strstart+257 */
+ Assert(scan <= window+(unsigned)(window_size-1), "wild scan");
+ if (*scan == *match) 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;
+
+ /* 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
+ * the hash keys are equal and that HASH_BITS >= 8.
+ */
+ scan += 2, match++;
+
+ /* We check for insufficient lookahead only every 8th comparison;
+ * the 256th check will be made at strstart+258.
+ */
+ do {
+ } while (*++scan == *++match && *++scan == *++match &&
+ *++scan == *++match && *++scan == *++match &&
+ *++scan == *++match && *++scan == *++match &&
+ *++scan == *++match && *++scan == *++match &&
+ scan < strend);
+
+ len = MAX_MATCH - (int)(strend - scan);
+ scan = strend - MAX_MATCH;
+
+#endif /* UNALIGNED_OK */
+
+ if (len > best_len) {
+ match_start = cur_match;
+ best_len = len;
+ if (len >= nice_match) break;
+#ifdef UNALIGNED_OK
+ scan_end = *(ush*)(scan+best_len-1);
+#else
+ scan_end1 = scan[best_len-1];
+ scan_end = scan[best_len];
+#endif
+ }
+ } while ((cur_match = prev[cur_match & WMASK]) > limit
+ && --chain_length != 0);
+
+ return best_len;
+}
+
+/* ===========================================================================
+ * Fill the window when the lookahead becomes insufficient.
+ * Updates strstart and lookahead, and sets eofile if end of input file.
+ * IN assertion: lookahead < MIN_LOOKAHEAD && strstart + lookahead > 0
+ * OUT assertions: at least one byte has been read, or eofile is set;
+ * file reads are performed for at least two bytes (required for the
+ * translate_eol option).
+ */
+static void fill_window(void)
+{
+ register unsigned n, m;
+ unsigned more = (unsigned)(window_size - (ulg)lookahead - (ulg)strstart);
+ /* Amount of free space at the end of the window. */
+
+ /* 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 (more == (unsigned)EOF) {
+ /* Very unlikely, but possible on 16 bit machine if strstart == 0
+ * and lookahead == 1 (input done one byte at time)
+ */
+ more--;
+ } else if (strstart >= WSIZE+MAX_DIST) {
+ /* By the IN assertion, the window is not empty so we can't confuse
+ * more == 0 with more == 64K on a 16 bit machine.
+ */
+ Assert(window_size == (ulg)2*WSIZE, "no sliding with BIG_MEM");
+
+ memcpy((char*)window, (char*)window+WSIZE, (unsigned)WSIZE);
+ match_start -= WSIZE;
+ strstart -= WSIZE; /* we now have strstart >= MAX_DIST: */
+ if (rsync_chunk_end != 0xFFFFFFFFUL)
+ rsync_chunk_end -= WSIZE;
+
+ block_start -= (long) WSIZE;
+
+ for (n = 0; n < HASH_SIZE; n++) {
+ m = head[n];
+ head[n] = (Pos)(m >= WSIZE ? m-WSIZE : NIL);
+ }
+ for (n = 0; n < WSIZE; n++) {
+ m = prev[n];
+ prev[n] = (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.
+ */
+ }
+ more += WSIZE;
+ }
+ /* At this point, more >= 2 */
+ if (!eofile) {
+ n = read_buf((char*)window+strstart+lookahead, more);
+ if (n == 0 || n == (unsigned)EOF) {
+ eofile = 1;
+ } else {
+ lookahead += n;
+ }
+ }
+}
+
+static void rsync_roll(unsigned start, unsigned num)
+{
+ unsigned i;
+
+ if (start < RSYNC_WIN) {
+ /* before window fills. */
+ for (i = start; i < RSYNC_WIN; i++) {
+ if (i == start + num) return;
+ rsync_sum += (ulg)window[i];
+ }
+ num -= (RSYNC_WIN - start);
+ start = RSYNC_WIN;
+ }
+
+ /* buffer after window full */
+ for (i = start; i < start+num; i++) {
+ /* New character in */
+ rsync_sum += (ulg)window[i];
+ /* Old character out */
+ rsync_sum -= (ulg)window[i - RSYNC_WIN];
+ if (rsync_chunk_end == 0xFFFFFFFFUL && RSYNC_SUM_MATCH(rsync_sum))
+ rsync_chunk_end = i;
+ }
+}
+
+/* ===========================================================================
+ * Set rsync_chunk_end if window sum matches magic value.
+ */
+#define RSYNC_ROLL(s, n) \
+ do { if (rsync) rsync_roll((s), (n)); } while(0)
+
+/* ===========================================================================
+ * Flush the current block, with given end-of-file flag.
+ * IN assertion: strstart is set to the end of the current match.
+ */
+#define FLUSH_BLOCK(eof) \
+ flush_block(block_start >= 0L ? (char*)&window[(unsigned)block_start] : \
+ (char*)NULL, (long)strstart - block_start, flush-1, (eof))
+
+/* ===========================================================================
+ * Processes a new input file and return its compressed length. This
+ * function does not perform lazy evaluationof 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.
+ */
+static void deflate_fast(int pack_level, int rsync)
+{
+ IPos hash_head; /* head of the hash chain */
+ int flush; /* set if current block must be flushed, 2=>and padded */
+ unsigned match_length = 0; /* length of best match */
+
+ prev_length = MIN_MATCH-1;
+ while (lookahead != 0) {
+ /* Insert the string window[strstart .. strstart+2] in the
+ * dictionary, and set hash_head to the head of the hash chain:
+ */
+ INSERT_STRING(strstart, hash_head);
+
+ /* Find the longest match, discarding those <= prev_length.
+ * At this point we have always match_length < MIN_MATCH
+ */
+ if (hash_head != NIL && strstart - hash_head <= MAX_DIST
+ && strstart <= window_size - MIN_LOOKAHEAD) {
+ /* To simplify the code, we prevent matches with the string
+ * of window index 0 (in particular we have to avoid a match
+ * of the string with itself at the start of the input file).
+ */
+ match_length = longest_match (hash_head);
+ /* longest_match() sets match_start */
+ if (match_length > lookahead) match_length = lookahead;
+ }
+ if (match_length >= MIN_MATCH) {
+ flush = ct_tally(pack_level, strstart-match_start, match_length - MIN_MATCH);
+
+ lookahead -= match_length;
+
+ RSYNC_ROLL(strstart, match_length);
+ /* Insert new strings in the hash table only if the match length
+ * is not too large. This saves time but degrades compression.
+ */
+ if (match_length <= max_insert_length) {
+ match_length--; /* string at strstart already in hash table */
+ do {
+ strstart++;
+ INSERT_STRING(strstart, hash_head);
+ /* strstart never exceeds WSIZE-MAX_MATCH, so there are
+ * always MIN_MATCH bytes ahead. If lookahead < MIN_MATCH
+ * these bytes are garbage, but it does not matter since
+ * the next lookahead bytes will be emitted as literals.
+ */
+ } while (--match_length != 0);
+ strstart++;
+ } else {
+ strstart += match_length;
+ match_length = 0;
+ ins_h = window[strstart];
+ UPDATE_HASH(ins_h, window[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",window[strstart]));
+ flush = ct_tally (pack_level, 0, window[strstart]);
+ RSYNC_ROLL(strstart, 1);
+ lookahead--;
+ strstart++;
+ }
+ if (rsync && strstart > rsync_chunk_end) {
+ rsync_chunk_end = 0xFFFFFFFFUL;
+ flush = 2;
+ }
+ if (flush) FLUSH_BLOCK(0), block_start = strstart;
+
+ /* 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.
+ */
+ while (lookahead < MIN_LOOKAHEAD && !eofile) fill_window();
+
+ }
+ FLUSH_BLOCK(1); /* eof */
+}
+
+/* ===========================================================================
+ * 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.
+ */
+void gnu_deflate(int pack_level, int rsync)
+{
+ IPos hash_head; /* head of hash chain */
+ IPos prev_match; /* previous match */
+ int flush = 0; /* set if current block must be flushed */
+ int match_available = 0; /* set if previous match exists */
+ register unsigned match_length = MIN_MATCH-1; /* length of best match */
+
+ if (pack_level <= 3) {
+ deflate_fast(pack_level, rsync); /* optimized for speed */
+ return;
+ }
+
+ /* Process the input block. */
+ while (lookahead != 0) {
+ /* Insert the string window[strstart .. strstart+2] in the
+ * dictionary, and set hash_head to the head of the hash chain:
+ */
+ INSERT_STRING(strstart, hash_head);
+
+ /* Find the longest match, discarding those <= prev_length.
+ */
+ prev_length = match_length, prev_match = match_start;
+ match_length = MIN_MATCH-1;
+
+ if (hash_head != NIL && prev_length < max_lazy_match &&
+ strstart - hash_head <= MAX_DIST &&
+ strstart <= window_size - MIN_LOOKAHEAD) {
+ /* To simplify the code, we prevent matches with the string
+ * of window index 0 (in particular we have to avoid a match
+ * of the string with itself at the start of the input file).
+ */
+ match_length = longest_match (hash_head);
+ /* longest_match() sets match_start */
+ if (match_length > lookahead) match_length = lookahead;
+
+ /* Ignore a length 3 match if it is too distant: */
+ if (match_length == MIN_MATCH && strstart-match_start > TOO_FAR){
+ /* If prev_match is also MIN_MATCH, match_start is garbage
+ * but we will ignore the current match anyway.
+ */
+ match_length--;
+ }
+ }
+ /* If there was a match at the previous step and the current
+ * match is not better, output the previous match:
+ */
+ if (prev_length >= MIN_MATCH && match_length <= prev_length) {
+ flush = ct_tally(pack_level, strstart-1-prev_match, prev_length - MIN_MATCH);
+
+ /* Insert in hash table all strings up to the end of the match.
+ * strstart-1 and strstart are already inserted.
+ */
+ lookahead -= prev_length-1;
+ prev_length -= 2;
+ RSYNC_ROLL(strstart, prev_length+1);
+ do {
+ strstart++;
+ INSERT_STRING(strstart, hash_head);
+ /* strstart never exceeds WSIZE-MAX_MATCH, so there are
+ * always MIN_MATCH bytes ahead. If lookahead < MIN_MATCH
+ * these bytes are garbage, but it does not matter since the
+ * next lookahead bytes will always be emitted as literals.
+ */
+ } while (--prev_length != 0);
+ match_available = 0;
+ match_length = MIN_MATCH-1;
+ strstart++;
+
+ if (rsync && strstart > rsync_chunk_end) {
+ rsync_chunk_end = 0xFFFFFFFFUL;
+ flush = 2;
+ }
+ if (flush) FLUSH_BLOCK(0), block_start = strstart;
+ } else if (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",window[strstart-1]));
+ flush = ct_tally (pack_level, 0, window[strstart-1]);
+ if (rsync && strstart > rsync_chunk_end) {
+ rsync_chunk_end = 0xFFFFFFFFUL;
+ flush = 2;
+ }
+ if (flush) FLUSH_BLOCK(0), block_start = strstart;
+ RSYNC_ROLL(strstart, 1);
+ strstart++;
+ lookahead--;
+ } else {
+ /* There is no previous match to compare with, wait for
+ * the next step to decide.
+ */
+ if (rsync && strstart > rsync_chunk_end) {
+ /* Reset huffman tree */
+ rsync_chunk_end = 0xFFFFFFFFUL;
+ flush = 2;
+ FLUSH_BLOCK(0), block_start = strstart;
+ }
+ match_available = 1;
+ RSYNC_ROLL(strstart, 1);
+ strstart++;
+ lookahead--;
+ }
+ /* Assert (strstart <= bytes_in && lookahead <= bytes_in, "a bit too far"); */
+
+ /* 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.
+ */
+ while (lookahead < MIN_LOOKAHEAD && !eofile) fill_window();
+ }
+ if (match_available) ct_tally (pack_level, 0, window[strstart-1]);
+
+ FLUSH_BLOCK(1); /* eof */
+}
--- /dev/null
+/* zip.c -- compress files to the gzip or pkzip format
+
+ Copyright (C) 1997, 1998, 1999, 2006, 2007 Free Software Foundation, Inc.
+ Copyright (C) 1992-1993 Jean-loup Gailly
+ Copyright (C) 2008 Josh Triplett
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+#include <ctype.h>
+#include <fcntl.h>
+#include <unistd.h>
+
+#include "gzip.h"
+
+uch outbuf[OUTBUFSIZ];
+unsigned outcnt; /* bytes in output buffer */
+
+static ulg crc; /* crc on uncompressed file data */
+
+static int ifd; /* input file descriptor */
+static int ofd; /* output file descriptor */
+
+static off_t bytes_in; /* number of input bytes */
+
+/* ===========================================================================
+ * Deflate in to out.
+ * IN assertions: the input and output buffers are cleared.
+ */
+void gnuzip(int in, int out, char *origname, ulg timestamp, int level, int osflag, int rsync)
+{
+ uch flags = 0; /* general purpose bit flags */
+ ush deflate_flags = 0; /* pkzip -es, -en or -ex equivalent */
+
+ ifd = in;
+ ofd = out;
+ outcnt = 0;
+ bytes_in = 0L;
+
+ /* Write the header to the gzip file. */
+
+ put_byte(GZIP_MAGIC[0]); /* magic header */
+ put_byte(GZIP_MAGIC[1]);
+ put_byte(DEFLATED); /* compression method */
+
+ if (origname)
+ flags |= ORIG_NAME;
+ put_byte(flags); /* general flags */
+ put_long(timestamp);
+
+ /* Write deflated file to zip file */
+ crc = updcrc(0, 0);
+
+ bi_init(out);
+ ct_init();
+ lm_init(level, &deflate_flags);
+
+ put_byte((uch)deflate_flags); /* extra flags */
+ put_byte(osflag); /* OS identifier */
+
+ if (origname) {
+ char *p = origname;
+ do {
+ put_byte(*p);
+ } while (*p++);
+ }
+
+ gnu_deflate(level, rsync);
+
+ /* Write the crc and uncompressed size */
+ put_long(crc);
+ put_long((ulg)bytes_in);
+
+ flush_outbuf();
+}
+
+
+/* ===========================================================================
+ * Read a new buffer from the current input file, perform end-of-line
+ * translation, and update the crc and input file size.
+ * IN assertion: size >= 2 (for end-of-line translation)
+ */
+int file_read(char *buf, unsigned size)
+{
+ unsigned len;
+
+ len = read_buffer (ifd, buf, size);
+ if (len == 0) return (int)len;
+ if (len == (unsigned)-1) {
+ read_error();
+ return EOF;
+ }
+
+ crc = updcrc((uch*)buf, len);
+ bytes_in += (off_t)len;
+ return (int)len;
+}
+
+/* ===========================================================================
+ * Write the output buffer outbuf[0..outcnt-1].
+ * (used for the compressed data only)
+ */
+void flush_outbuf(void)
+{
+ if (outcnt == 0) return;
+
+ write_buf(ofd, (char *)outbuf, outcnt);
+ outcnt = 0;
+}
--- /dev/null
+/* gzip.h -- common declarations for all gzip modules
+
+ Copyright (C) 1997, 1998, 1999, 2001, 2006, 2007 Free Software
+ Foundation, Inc.
+ Copyright (C) 1992-1993 Jean-loup Gailly.
+ Copyright (C) 2008 Josh Triplett
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+#define BITS 16
+
+#ifndef __attribute__
+# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 8) || __STRICT_ANSI__
+# define __attribute__(x)
+# endif
+#endif
+
+#define ATTRIBUTE_NORETURN __attribute__ ((__noreturn__))
+
+/* I don't like nested includes, but the following headers are used
+ * too often
+ */
+#include <stdio.h>
+#include <sys/types.h> /* for off_t */
+#include <time.h>
+#include <string.h>
+#define memzero(s, n) memset ((s), 0, (n))
+
+typedef unsigned char uch;
+typedef unsigned short ush;
+typedef unsigned long ulg;
+
+/* Return codes from gzip */
+#define OK 0
+#define ERROR 1
+
+/* Compression methods */
+#define DEFLATED 8
+
+#define INBUFSIZ 0x8000 /* input buffer size */
+#define OUTBUFSIZ 16384 /* output buffer size */
+#define DIST_BUFSIZE 0x8000 /* buffer for distances, see trees.c */
+
+extern uch outbuf[]; /* output buffer */
+
+extern unsigned outcnt; /* bytes in output buffer */
+
+#define NO_FILE (-1) /* in memory compression */
+
+
+#define GZIP_MAGIC "\037\213" /* Magic header for gzip files, 1F 8B */
+#define OLD_GZIP_MAGIC "\037\236" /* Magic header for gzip 0.5 = freeze 1.x */
+#define PKZIP_MAGIC "\120\113\003\004" /* Magic header for pkzip files */
+
+/* gzip flag byte */
+#define ASCII_FLAG 0x01 /* bit 0 set: file probably ascii text */
+#define CONTINUATION 0x02 /* bit 1 set: continuation of multi-part gzip file */
+#define EXTRA_FIELD 0x04 /* bit 2 set: extra field present */
+#define ORIG_NAME 0x08 /* bit 3 set: original file name present */
+#define COMMENT 0x10 /* bit 4 set: file comment present */
+#define ENCRYPTED 0x20 /* bit 5 set: file is encrypted */
+#define RESERVED 0xC0 /* bit 6,7: reserved */
+
+/* internal file attribute */
+#define UNKNOWN 0xffff
+#define BINARY 0
+#define ASCII 1
+
+#define WSIZE 0x8000 /* window size--must be a power of two, and */
+ /* at least 32K for zip's deflate method */
+
+#define MIN_MATCH 3
+#define MAX_MATCH 258
+/* The minimum and maximum match lengths */
+
+#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1)
+/* Minimum amount of lookahead, except at the end of the input file.
+ * See deflate.c for comments about the MIN_MATCH+1.
+ */
+
+#define MAX_DIST (WSIZE-MIN_LOOKAHEAD)
+/* In order to simplify the code, particularly on 16 bit machines, match
+ * distances are limited to MAX_DIST instead of WSIZE.
+ */
+
+/* put_byte is used for the compressed output. */
+#define put_byte(c) {outbuf[outcnt++]=(uch)(c); if (outcnt==OUTBUFSIZ)\
+ flush_outbuf();}
+
+/* Output a 16 bit value, lsb first */
+#define put_short(w) \
+{ if (outcnt < OUTBUFSIZ-2) { \
+ outbuf[outcnt++] = (uch) ((w) & 0xff); \
+ outbuf[outcnt++] = (uch) ((ush)(w) >> 8); \
+ } else { \
+ put_byte((uch)((w) & 0xff)); \
+ put_byte((uch)((ush)(w) >> 8)); \
+ } \
+}
+
+/* Output a 32 bit value to the bit stream, lsb first */
+#define put_long(n) { \
+ put_short((n) & 0xffff); \
+ put_short(((ulg)(n)) >> 16); \
+}
+
+/* Diagnostic functions */
+#define Assert(cond,msg)
+#define Trace(x)
+#define Tracev(x)
+#define Tracevv(x)
+#define Tracec(c,x)
+#define Tracecv(c,x)
+
+ /* in zip.c: */
+extern int file_read(char *buf, unsigned size);
+
+ /* in deflate.c */
+void lm_init(int pack_level, ush *flags);
+void gnu_deflate(int pack_level, int rsync);
+
+ /* in trees.c */
+void ct_init(void);
+int ct_tally(int pack_level, int dist, int lc);
+void flush_block(char *buf, ulg stored_len, int pad, int eof);
+
+ /* in bits.c */
+void bi_init(int zipfile);
+void send_bits(int value, int length);
+unsigned bi_reverse(unsigned value, int length);
+void bi_windup(void);
+void copy_block(char *buf, unsigned len, int header);
+extern int (*read_buf)(char *buf, unsigned size);
+
+ /* in util.c: */
+extern ulg updcrc(uch *s, unsigned n);
+extern void flush_outbuf(void);
+extern void write_buf(int fd, void *buf, unsigned cnt);
+extern int read_buffer(int fd, void *buf, unsigned int cnt);
+extern void gzip_error(char *m);
+extern void read_error(void);
+extern void write_error(void);
Joey Hess <joeyh@debian.org>,
Faidon Liambotis <paravoid@debian.org>
+Josh Triplett <josh@joshtriplett.org>
Licensed under the GPL, version 2.
use File::Temp;
use Getopt::Long;
use File::Basename qw/basename/;
+use IPC::Open2;
use constant GZIP_DEBUG => 1;
}
}
+sub doit_redir {
+ no warnings 'once';
+ my ($in, $out, @args) = @_;
+ vprint(@args, "<", $in, ">", $out);
+ open INFILE, "<", $in or die("Could not open '$in' for reading: $!\n");
+ open OUTFILE, ">", $out or die("Could not open '$out' for reading: $!\n");
+ my $pid = open2(">&OUTFILE", "<&INFILE", @args);
+ waitpid $pid, 0;
+}
+
sub tempdir {
return File::Temp::tempdir("pristine-gz.XXXXXXXXXX",
TMPDIR => 1, CLEANUP => !$keep);
unless ($flags[GZIP_FLAG_FNAME]) {
push @args, '-n';
push @args, '-M' if $timestamp;
- } else {
- push @args, '-m'
- unless($timestamp);
}
-
+
if ($level == GZIP_COMPRESSION_BEST) {
push @args, '-9'
} elsif ($level == GZIP_COMPRESSION_FAST) {
}
sub testvariant {
- my ($old, $new, $origname, @args) = @_;
-
- # We need an uncompressed file to test the compressor with.
- # See if one already exists.
- my $uncompressed = $new;
- $uncompressed =~ s/\.gz$//;
- if ($origname && -e $origname) {
- $uncompressed = $origname;
- }
- elsif (! -e $uncompressed) {
- # Generate the uncompressed file.
- doit('gunzip', '-N', $new);
- if (-e $new) {
- die "gunzip failed, aborting";
- }
- elsif ($origname && -e $origname) {
- $uncompressed = $origname;
- }
- elsif (! -e $uncompressed) {
- die("gunzip succeded but I can't find the uncompressed file");
- }
- }
-
- # Cache a copy of the uncompressed file to avoid needing to
- # uncompress it again. (By hard linking, we avoid an
- # expensive copy, but need to pass -f when gzipping.)
- doit("ln", "-f", $uncompressed, "$uncompressed.cached");
-
- # try gzipping with the arguments passed
- doit('zgz', @args, "-f", $uncompressed);
- $new = "$uncompressed.gz";
- unless (-e $new) {
- die("zgz failed, aborting");
- }
- # move cached uncompressed file into place for later use
- doit("mv", "-f", "$uncompressed.cached", $uncompressed);
+ my ($orig, $tempin, $tempout, @args) = @_;
+ # Use the variant of gzip defined by @args to compress $tempin to
+ # $tempout and compare the result to $orig
- # and compare the generated with the original
- return !comparefiles($old, $new);
+ doit_redir($tempin, $tempout, 'zgz', @args, '-c');
+ return !comparefiles($orig, $tempout);
}
sub reproducegz {
- my ($wd, $orig, $new) = (shift, shift, shift);
+ my ($orig, $tempin, $tempout) = (shift, shift, shift);
+ doit_redir($orig, $tempin, "gzip", "-dc");
# read fields from gzip headers
- my ($flags, $timestamp, $level, $os, $name) = readgzip($new);
+ my ($flags, $timestamp, $level, $os, $name) = readgzip($orig);
debug("flags: [".join(", ", @$flags).
"] timestamp: $timestamp level: $level os: $os name: $name");
# try to guess the gzip arguments that are needed by the header
# information
my @args = predictgzipargs($flags, $timestamp, $level);
-
- if ($name !~ /\//) {
- if ($name) {
- $name = "$wd/$name";
- doit("mv", "-f", $new, "$name.gz");
- $new = "$name.gz";
- }
- if ($os == GZIP_OS_UNIX) {
- # for 98% of the cases the simple heuristic above works
- testvariant($orig, $new, $name, '--gnu', @args)
- && return $name, $timestamp, '--gnu', @args;
-
- # some .gz are created using --rsyncable
- # it is a Debian-specific option not very popular
- testvariant($orig, $new, $name, '--gnu', @args, '--rsyncable')
- && return $name, $timestamp, '--gnu', @args, '--rsyncable';
- }
+ my @extraargs = ("-F", $name, "-T", $timestamp);
+
+ if ($os == GZIP_OS_UNIX) {
+ # for 98% of the cases the simple heuristic above works
+ testvariant($orig, $tempin, $tempout, '--gnu', @args, @extraargs)
+ && return $name, $timestamp, '--gnu', @args;
+ testvariant($orig, $tempin, $tempout, '--gnu', @args, @extraargs, '--rsyncable')
+ && return $name, $timestamp, '--gnu', @args, '--rsyncable';
}
- else {
+
+ if ($name =~ /\//) {
push @args, "--original-name", $name;
+ @extraargs = ("-T", $timestamp);
$name = basename($name);
- $name = "$wd/$name";
- doit("mv", "-f", $new, "$name.gz");
- $new = "$name.gz";
}
# set the Operating System flag to the one found in the original
# gzip which is using the zlib library; try with our version of
# bsd-gzip with added support for the undocumented GNU gzip options
# -m and -M
- testvariant($orig, $new, $name, @args)
+ testvariant($orig, $tempin, $tempout, @args, @extraargs)
&& return $name, $timestamp, @args;
-
+
# apparently, there is an old version of bsd-gzip (or a similar tool
# based on zlib) that creates gz using maximum compression (-9) but
# does not indicate so in the headers. surprisingly, there are many
# .gz out there.
- testvariant($orig, $new, $name, @args, '--quirk', 'buggy-bsd')
+ testvariant($orig, $tempin, $tempout, @args, @extraargs, '--quirk', 'buggy-bsd')
&& return $name. $timestamp, @args, '--quirk', 'buggy-bsd';
# Windows' NTFS gzip implementation; quirk is really really evil
# it should be the last test: it can result in a corrupted archive!
if ($os == GZIP_OS_NTFS) {
pop @args; pop @args; # ntfs quirk implies NTFS osflag
- testvariant($orig, $new, $name, @args, '--quirk', 'ntfs')
+ testvariant($orig, $tempin, $tempout, @args, @extraargs, '--quirk', 'ntfs')
&& return $name, $timestamp, @args, '--quirk', 'ntfs';
}
chomp $timestamp;
close IN;
- doit("cp", $file, "$tempdir/$filename");
- utime($timestamp, $timestamp, "$tempdir/$filename") ||
- die "utime: $!";
- doit("zgz", @params, "-f", "$tempdir/$filename");
- doit("mv", "-f", "$tempdir/$filename.gz", "$file.gz");
- doit("rm", "-f", $file);
+ doit_redir("$file", "$file.gz", "zgz", @params, "-T", $timestamp, "-F", "$filename", "-c");
}
sub gendelta {
my @files=qw(version type params filename timestamp);
- doit("cp", $gzfile, "$tempdir/test.gz");
my ($filename, $timestamp, @params)=
- reproducegz($tempdir, $gzfile, "$tempdir/test.gz");
+ reproducegz($gzfile, "$tempdir/test", "$tempdir/test.gz");
open(OUT, ">", "$tempdir/version") || die "$!";
print OUT "2.0\n";
--- /dev/null
+/* trees.c -- output deflated data using Huffman coding
+
+ Copyright (C) 1997, 1998, 1999 Free Software Foundation, Inc.
+ Copyright (C) 1992-1993 Jean-loup Gailly
+ Copyright (C) 2008 Josh Triplett
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+/*
+ * PURPOSE
+ *
+ * Encode various sets of source values using variable-length
+ * binary code trees.
+ *
+ * DISCUSSION
+ *
+ * The PKZIP "deflation" process uses several Huffman trees. The more
+ * common source values are represented by shorter bit sequences.
+ *
+ * Each code tree is stored in the ZIP file in a compressed form
+ * which is itself a Huffman encoding of the lengths of
+ * all the code strings (in ascending order by source values).
+ * The actual code strings are reconstructed from the lengths in
+ * the UNZIP process, as described in the "application note"
+ * (APPNOTE.TXT) distributed as part of PKWARE's PKZIP program.
+ *
+ * REFERENCES
+ *
+ * Lynch, Thomas J.
+ * Data Compression: Techniques and Applications, pp. 53-55.
+ * Lifetime Learning Publications, 1985. ISBN 0-534-03418-7.
+ *
+ * Storer, James A.
+ * Data Compression: Methods and Theory, pp. 49-50.
+ * Computer Science Press, 1988. ISBN 0-7167-8156-5.
+ *
+ * Sedgewick, R.
+ * Algorithms, p290.
+ * Addison-Wesley, 1983. ISBN 0-201-06672-6.
+ *
+ * INTERFACE
+ *
+ * void ct_init (void)
+ * Allocate the match buffer and initialize the various tables
+ *
+ * void ct_tally (int pack_level, int dist, int lc);
+ * Save the match info and tally the frequency counts.
+ *
+ * void flush_block (char *buf, ulg stored_len, int pad, int eof)
+ * Determine the best encoding for the current block: dynamic trees,
+ * static trees or store, and output the encoded block to the zip
+ * file. If pad is set, pads the block to the next byte.
+ * */
+
+#include <ctype.h>
+
+#include "gzip.h"
+
+/* ===========================================================================
+ * Constants
+ */
+
+#define MAX_BITS 15
+/* All codes must not exceed MAX_BITS bits */
+
+#define MAX_BL_BITS 7
+/* Bit length codes must not exceed MAX_BL_BITS bits */
+
+#define LENGTH_CODES 29
+/* number of length codes, not counting the special END_BLOCK code */
+
+#define LITERALS 256
+/* number of literal bytes 0..255 */
+
+#define END_BLOCK 256
+/* end of block literal code */
+
+#define L_CODES (LITERALS+1+LENGTH_CODES)
+/* number of Literal or Length codes, including the END_BLOCK code */
+
+#define D_CODES 30
+/* number of distance codes */
+
+#define BL_CODES 19
+/* number of codes used to transfer the bit lengths */
+
+
+static int extra_lbits[LENGTH_CODES] /* extra bits for each length code */
+ = {0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0};
+
+static int extra_dbits[D_CODES] /* extra bits for each distance code */
+ = {0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13};
+
+static int extra_blbits[BL_CODES]/* extra bits for each bit length code */
+ = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,3,7};
+
+#define STORED_BLOCK 0
+#define STATIC_TREES 1
+#define DYN_TREES 2
+/* The three kinds of block type */
+
+#define LIT_BUFSIZE 0x8000
+/* Sizes of match buffers for literals/lengths and distances. There are
+ * 4 reasons for limiting LIT_BUFSIZE to 64K:
+ * - frequencies can be kept in 16 bit counters
+ * - if compression is not successful for the first block, all input data is
+ * still in the window so we can still emit a stored block even when input
+ * comes from standard input. (This can also be done for all blocks if
+ * LIT_BUFSIZE is not greater than 32K.)
+ * - if compression is not successful for a file smaller than 64K, we can
+ * even emit a stored file instead of a stored block (saving 5 bytes).
+ * - creating new Huffman trees less frequently may not provide fast
+ * adaptation to changes in the input data statistics. (Take for
+ * example a binary file with poorly compressible code followed by
+ * a highly compressible string table.) Smaller buffer sizes give
+ * fast adaptation but have of course the overhead of transmitting trees
+ * more frequently.
+ * - I can't count above 4
+ * The current code is general and allows DIST_BUFSIZE < LIT_BUFSIZE (to save
+ * memory at the expense of compression). Some optimizations would be possible
+ * if we rely on DIST_BUFSIZE == LIT_BUFSIZE.
+ */
+
+#define REP_3_6 16
+/* repeat previous bit length 3-6 times (2 bits of repeat count) */
+
+#define REPZ_3_10 17
+/* repeat a zero length 3-10 times (3 bits of repeat count) */
+
+#define REPZ_11_138 18
+/* repeat a zero length 11-138 times (7 bits of repeat count) */
+
+/* ===========================================================================
+ * Local data
+ */
+
+/* Data structure describing a single value and its code string. */
+typedef struct ct_data {
+ union {
+ ush freq; /* frequency count */
+ ush code; /* bit string */
+ } fc;
+ union {
+ ush dad; /* father node in Huffman tree */
+ ush len; /* length of bit string */
+ } dl;
+} ct_data;
+
+#define Freq fc.freq
+#define Code fc.code
+#define Dad dl.dad
+#define Len dl.len
+
+#define HEAP_SIZE (2*L_CODES+1)
+/* maximum heap size */
+
+static ct_data dyn_ltree[HEAP_SIZE]; /* literal and length tree */
+static ct_data dyn_dtree[2*D_CODES+1]; /* distance tree */
+
+static 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
+ * The codes 286 and 287 are needed to build a canonical tree (see ct_init
+ * below).
+ */
+
+static ct_data static_dtree[D_CODES];
+/* The static distance tree. (Actually a trivial tree since all codes use
+ * 5 bits.)
+ */
+
+static ct_data bl_tree[2*BL_CODES+1];
+/* Huffman tree for the bit lengths */
+
+typedef struct tree_desc {
+ ct_data *dyn_tree; /* the dynamic tree */
+ ct_data *static_tree; /* corresponding static tree or NULL */
+ int *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 */
+ int max_length; /* max bit length for the codes */
+ int max_code; /* largest code with non zero frequency */
+} tree_desc;
+
+static tree_desc l_desc =
+{dyn_ltree, static_ltree, extra_lbits, LITERALS+1, L_CODES, MAX_BITS, 0};
+
+static tree_desc d_desc =
+{dyn_dtree, static_dtree, extra_dbits, 0, D_CODES, MAX_BITS, 0};
+
+static tree_desc bl_desc =
+{bl_tree, (ct_data *)0, extra_blbits, 0, BL_CODES, MAX_BL_BITS, 0};
+
+
+static ush bl_count[MAX_BITS+1];
+/* number of codes at each bit length for an optimal tree */
+
+static uch bl_order[BL_CODES]
+ = {16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15};
+/* The lengths of the bit length codes are sent in order of decreasing
+ * probability, to avoid transmitting the lengths for unused bit length codes.
+ */
+
+static int heap[2*L_CODES+1]; /* heap used to build the Huffman trees */
+static int heap_len; /* number of elements in the heap */
+static int heap_max; /* element of largest frequency */
+/* The sons of heap[n] are heap[2*n] and heap[2*n+1]. heap[0] is not used.
+ * The same heap array is used to build all trees.
+ */
+
+static uch depth[2*L_CODES+1];
+/* Depth of each subtree used as tie breaker for trees of equal frequency */
+
+static uch length_code[MAX_MATCH-MIN_MATCH+1];
+/* length code for each normalized match length (0 == MIN_MATCH) */
+
+static uch dist_code[512];
+/* 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.
+ */
+
+static int base_length[LENGTH_CODES];
+/* First normalized length for each code (0 = MIN_MATCH) */
+
+static int base_dist[D_CODES];
+/* First normalized distance for each code (0 = distance of 1) */
+
+static uch l_buf[INBUFSIZ]; /* buffer for literals or lengths */
+static ush d_buf[DIST_BUFSIZE]; /* buffer for distances */
+
+static uch flag_buf[(LIT_BUFSIZE/8)];
+/* flag_buf is a bit array distinguishing literals from lengths in
+ * l_buf, thus indicating the presence or absence of a distance.
+ */
+
+static unsigned last_lit; /* running index in l_buf */
+static unsigned last_dist; /* running index in d_buf */
+static unsigned last_flags; /* running index in flag_buf */
+static uch flags; /* current flags not yet saved in flag_buf */
+static uch flag_bit; /* current bit used in flags */
+/* bits are filled in flags starting at bit 0 (least significant).
+ * Note: these flags are overkill in the current code since we don't
+ * take advantage of DIST_BUFSIZE == LIT_BUFSIZE.
+ */
+
+static ulg opt_len; /* bit length of current block with optimal trees */
+static ulg static_len; /* bit length of current block with static trees */
+
+static off_t compressed_len; /* total bit length of compressed file */
+
+extern long block_start; /* window offset of current block */
+extern unsigned strstart; /* window offset of current string */
+
+/* ===========================================================================
+ * Local (static) routines in this file.
+ */
+
+static void init_block(void);
+static void pqdownheap(ct_data *tree, int k);
+static void gen_bitlen(tree_desc *desc);
+static void gen_codes(ct_data *tree, int max_code);
+static void build_tree(tree_desc *desc);
+static void scan_tree(ct_data *tree, int max_code);
+static void send_tree(ct_data *tree, int max_code);
+static int build_bl_tree(void);
+static void send_all_trees(int lcodes, int dcodes, int blcodes);
+static void compress_block(ct_data *ltree, ct_data *dtree);
+
+#define send_code(c, tree) send_bits(tree[c].Code, tree[c].Len)
+/* Send a code of the given tree. c and tree must not have side effects */
+
+#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.
+ */
+
+#define MAX(a,b) (a >= b ? a : b)
+/* the arguments must not have side effects */
+
+/* ===========================================================================
+ * Allocate the match buffer and initialize the various tables
+ */
+void ct_init(void)
+{
+ int n; /* iterates over tree elements */
+ int bits; /* bit counter */
+ int length; /* length value */
+ int code; /* code value */
+ int dist; /* distance index */
+
+ compressed_len = 0L;
+
+ if (static_dtree[0].Len != 0) return; /* ct_init already called */
+
+ /* 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;
+ }
+ }
+ Assert (length == 256, "ct_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;
+
+ /* 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;
+ }
+ }
+ Assert (dist == 256, "ct_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++) {
+ dist_code[256 + dist++] = (uch)code;
+ }
+ }
+ Assert (dist == 256, "ct_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;
+ while (n <= 143) static_ltree[n++].Len = 8, bl_count[8]++;
+ while (n <= 255) static_ltree[n++].Len = 9, bl_count[9]++;
+ while (n <= 279) static_ltree[n++].Len = 7, bl_count[7]++;
+ while (n <= 287) static_ltree[n++].Len = 8, bl_count[8]++;
+ /* Codes 286 and 287 do not exist, but we must include them in the
+ * tree construction to get a canonical Huffman tree (longest code
+ * all ones)
+ */
+ gen_codes((ct_data *)static_ltree, L_CODES+1);
+
+ /* The static distance tree is trivial: */
+ for (n = 0; n < D_CODES; n++) {
+ static_dtree[n].Len = 5;
+ static_dtree[n].Code = bi_reverse(n, 5);
+ }
+
+ /* Initialize the first block of the first file: */
+ init_block();
+}
+
+/* ===========================================================================
+ * Initialize a new block.
+ */
+static void init_block(void)
+{
+ int n; /* iterates over tree elements */
+
+ /* Initialize the trees. */
+ for (n = 0; n < L_CODES; n++) dyn_ltree[n].Freq = 0;
+ for (n = 0; n < D_CODES; n++) dyn_dtree[n].Freq = 0;
+ for (n = 0; n < BL_CODES; n++) bl_tree[n].Freq = 0;
+
+ dyn_ltree[END_BLOCK].Freq = 1;
+ opt_len = static_len = 0L;
+ last_lit = last_dist = last_flags = 0;
+ flags = 0; flag_bit = 1;
+}
+
+#define SMALLEST 1
+/* Index within the heap array of least frequent node in the Huffman tree */
+
+
+/* ===========================================================================
+ * Remove the smallest element from the heap and recreate the heap with
+ * one less element. Updates heap and heap_len.
+ */
+#define pqremove(tree, top) \
+{\
+ top = heap[SMALLEST]; \
+ heap[SMALLEST] = heap[heap_len--]; \
+ pqdownheap(tree, SMALLEST); \
+}
+
+/* ===========================================================================
+ * Compares to subtrees, using the tree depth as tie breaker when
+ * the subtrees have equal frequency. This minimizes the worst case length.
+ */
+#define smaller(tree, n, m) \
+ (tree[n].Freq < tree[m].Freq || \
+ (tree[n].Freq == tree[m].Freq && depth[n] <= depth[m]))
+
+/* ===========================================================================
+ * Restore the heap property by moving down the tree starting at node k,
+ * 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).
+ */
+static void pqdownheap(ct_data *tree, /* the tree to restore */
+ int k) /* node to move down */
+{
+ int v = heap[k];
+ int j = k << 1; /* left son of k */
+ while (j <= heap_len) {
+ /* Set j to the smallest of the two sons: */
+ if (j < heap_len && smaller(tree, heap[j+1], heap[j])) j++;
+
+ /* Exit if v is smaller than both sons */
+ if (smaller(tree, v, heap[j])) break;
+
+ /* Exchange v with the smallest son */
+ heap[k] = heap[j]; k = j;
+
+ /* And continue down the tree, setting j to the left son of k */
+ j <<= 1;
+ }
+ heap[k] = v;
+}
+
+/* ===========================================================================
+ * Compute the optimal bit lengths for a tree and update the total bit length
+ * for the current block.
+ * IN assertion: the fields freq and dad are set, heap[heap_max] and
+ * above are the tree nodes sorted by increasing frequency.
+ * OUT assertions: the field len is set to the optimal bit length, the
+ * 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.
+ */
+static void gen_bitlen(tree_desc *desc)
+{
+ ct_data *tree = desc->dyn_tree;
+ int *extra = desc->extra_bits;
+ int base = desc->extra_base;
+ int max_code = desc->max_code;
+ int max_length = desc->max_length;
+ ct_data *stree = desc->static_tree;
+ int h; /* heap index */
+ int n, m; /* iterate over the tree elements */
+ int bits; /* bit length */
+ int xbits; /* extra bits */
+ ush f; /* frequency */
+ int overflow = 0; /* number of elements with bit length too large */
+
+ for (bits = 0; bits <= MAX_BITS; bits++) bl_count[bits] = 0;
+
+ /* In a first pass, compute the optimal bit lengths (which may
+ * overflow in the case of the bit length tree).
+ */
+ tree[heap[heap_max]].Len = 0; /* root of the heap */
+
+ for (h = heap_max+1; h < HEAP_SIZE; h++) {
+ n = heap[h];
+ bits = tree[tree[n].Dad].Len + 1;
+ if (bits > max_length) bits = max_length, overflow++;
+ tree[n].Len = (ush)bits;
+ /* We overwrite tree[n].Dad which is no longer needed */
+
+ if (n > max_code) continue; /* not a leaf node */
+
+ bl_count[bits]++;
+ xbits = 0;
+ if (n >= base) xbits = extra[n-base];
+ f = tree[n].Freq;
+ opt_len += (ulg)f * (bits + xbits);
+ if (stree) static_len += (ulg)f * (stree[n].Len + xbits);
+ }
+ if (overflow == 0) return;
+
+ Trace((stderr,"\nbit length overflow\n"));
+ /* 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;
+ while (bl_count[bits] == 0) bits--;
+ bl_count[bits]--; /* move one leaf down the tree */
+ bl_count[bits+1] += 2; /* move one overflow item as its brother */
+ bl_count[max_length]--;
+ /* The brother of the overflow item also moves one step up,
+ * but this does not affect bl_count[max_length]
+ */
+ overflow -= 2;
+ } while (overflow > 0);
+
+ /* Now recompute all bit lengths, scanning in increasing frequency.
+ * h is still equal to HEAP_SIZE. (It is simpler to reconstruct all
+ * lengths instead of fixing only the wrong ones. This idea is taken
+ * from 'ar' written by Haruhiko Okumura.)
+ */
+ for (bits = max_length; bits != 0; bits--) {
+ n = bl_count[bits];
+ while (n != 0) {
+ m = heap[--h];
+ if (m > max_code) continue;
+ if (tree[m].Len != (unsigned) bits) {
+ Trace((stderr,"code %d bits %d->%d\n", m, tree[m].Len, bits));
+ opt_len += ((long)bits-(long)tree[m].Len)*(long)tree[m].Freq;
+ tree[m].Len = (ush)bits;
+ }
+ n--;
+ }
+ }
+}
+
+/* ===========================================================================
+ * 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.
+ */
+static void gen_codes (ct_data *tree, /* the tree to decorate */
+ int max_code) /* largest code with non zero frequency */
+{
+ ush next_code[MAX_BITS+1]; /* next code value for each bit length */
+ ush 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++) {
+ next_code[bits] = code = (code + bl_count[bits-1]) << 1;
+ }
+ /* 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 = bi_reverse(next_code[len]++, len);
+
+ Tracec(tree != static_ltree, (stderr,"\nn %3d %c l %2d c %4x (%x) ",
+ n, (isgraph(n) ? n : ' '), len, tree[n].Code, next_code[len]-1));
+ }
+}
+
+/* ===========================================================================
+ * Construct one Huffman tree and assigns the code bit strings and lengths.
+ * Update the total bit length for the current block.
+ * IN assertion: the field freq is set for all tree elements.
+ * 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.
+ */
+static void build_tree(tree_desc *desc)
+{
+ ct_data *tree = desc->dyn_tree;
+ ct_data *stree = desc->static_tree;
+ int elems = desc->elems;
+ int n, m; /* iterate over heap elements */
+ int max_code = -1; /* largest code with non zero frequency */
+ int node = elems; /* next internal node of the tree */
+
+ /* 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[0] is not used.
+ */
+ heap_len = 0, heap_max = HEAP_SIZE;
+
+ for (n = 0; n < elems; n++) {
+ if (tree[n].Freq != 0) {
+ heap[++heap_len] = max_code = n;
+ depth[n] = 0;
+ } else {
+ tree[n].Len = 0;
+ }
+ }
+
+ /* The pkzip format requires that at least one distance code exists,
+ * and that at least one bit should be sent even if there is only one
+ * possible code. So to avoid special checks later on we force at least
+ * two codes of non zero frequency.
+ */
+ while (heap_len < 2) {
+ int new = heap[++heap_len] = (max_code < 2 ? ++max_code : 0);
+ tree[new].Freq = 1;
+ depth[new] = 0;
+ opt_len--; if (stree) static_len -= stree[new].Len;
+ /* new 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,
+ * establish sub-heaps of increasing lengths:
+ */
+ for (n = heap_len/2; n >= 1; n--) pqdownheap(tree, n);
+
+ /* Construct the Huffman tree by repeatedly combining the least two
+ * frequent nodes.
+ */
+ do {
+ pqremove(tree, n); /* n = node of least frequency */
+ m = heap[SMALLEST]; /* m = node of next least frequency */
+
+ heap[--heap_max] = n; /* keep the nodes sorted by frequency */
+ heap[--heap_max] = m;
+
+ /* Create a new node father of n and m */
+ tree[node].Freq = tree[n].Freq + tree[m].Freq;
+ depth[node] = (uch) (MAX(depth[n], depth[m]) + 1);
+ tree[n].Dad = tree[m].Dad = (ush)node;
+#ifdef DUMP_BL_TREE
+ if (tree == bl_tree) {
+ fprintf(stderr,"\nnode %d(%d), sons %d(%d) %d(%d)",
+ node, tree[node].Freq, n, tree[n].Freq, m, tree[m].Freq);
+ }
+#endif
+ /* and insert the new node in the heap */
+ heap[SMALLEST] = node++;
+ pqdownheap(tree, SMALLEST);
+
+ } while (heap_len >= 2);
+
+ heap[--heap_max] = heap[SMALLEST];
+
+ /* At this point, the fields freq and dad are set. We can now
+ * generate the bit lengths.
+ */
+ gen_bitlen((tree_desc *)desc);
+
+ /* The field len is now set, we can generate the bit codes */
+ gen_codes ((ct_data *)tree, max_code);
+}
+
+/* ===========================================================================
+ * Scan a literal or distance tree to determine the frequencies of the codes
+ * in the bit length tree. Updates opt_len to take into account the repeat
+ * counts. (The contribution of the bit length codes will be added later
+ * during the construction of bl_tree.)
+ */
+static void scan_tree (ct_data *tree, /* the tree to be scanned */
+ int max_code) /* and its largest code of non zero frequency */
+{
+ 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 */
+
+ if (nextlen == 0) max_count = 138, min_count = 3;
+ tree[max_code+1].Len = (ush)0xffff; /* guard */
+
+ for (n = 0; n <= max_code; n++) {
+ curlen = nextlen; nextlen = tree[n+1].Len;
+ if (++count < max_count && curlen == nextlen) {
+ continue;
+ } else if (count < min_count) {
+ bl_tree[curlen].Freq += count;
+ } else if (curlen != 0) {
+ if (curlen != prevlen) bl_tree[curlen].Freq++;
+ bl_tree[REP_3_6].Freq++;
+ } else if (count <= 10) {
+ bl_tree[REPZ_3_10].Freq++;
+ } else {
+ bl_tree[REPZ_11_138].Freq++;
+ }
+ count = 0; prevlen = curlen;
+ if (nextlen == 0) {
+ max_count = 138, min_count = 3;
+ } else if (curlen == nextlen) {
+ max_count = 6, min_count = 3;
+ } else {
+ max_count = 7, min_count = 4;
+ }
+ }
+}
+
+/* ===========================================================================
+ * Send a literal or distance tree in compressed form, using the codes in
+ * bl_tree.
+ */
+static void send_tree (ct_data *tree, /* the tree to be scanned */
+ int max_code) /* and its largest code of non zero frequency */
+{
+ 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 */
+ if (nextlen == 0) max_count = 138, min_count = 3;
+
+ for (n = 0; n <= max_code; n++) {
+ curlen = nextlen; nextlen = tree[n+1].Len;
+ if (++count < max_count && curlen == nextlen) {
+ continue;
+ } else if (count < min_count) {
+ do { send_code(curlen, bl_tree); } while (--count != 0);
+
+ } else if (curlen != 0) {
+ if (curlen != prevlen) {
+ send_code(curlen, bl_tree); count--;
+ }
+ Assert(count >= 3 && count <= 6, " 3_6?");
+ send_code(REP_3_6, bl_tree); send_bits(count-3, 2);
+
+ } else if (count <= 10) {
+ send_code(REPZ_3_10, bl_tree); send_bits(count-3, 3);
+
+ } else {
+ send_code(REPZ_11_138, bl_tree); send_bits(count-11, 7);
+ }
+ count = 0; prevlen = curlen;
+ if (nextlen == 0) {
+ max_count = 138, min_count = 3;
+ } else if (curlen == nextlen) {
+ max_count = 6, min_count = 3;
+ } else {
+ max_count = 7, min_count = 4;
+ }
+ }
+}
+
+/* ===========================================================================
+ * Construct the Huffman tree for the bit lengths and return the index in
+ * bl_order of the last bit length code to send.
+ */
+static int build_bl_tree(void)
+{
+ 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((ct_data *)dyn_ltree, l_desc.max_code);
+ scan_tree((ct_data *)dyn_dtree, d_desc.max_code);
+
+ /* Build the bit length tree: */
+ build_tree((tree_desc *)(&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.
+ */
+
+ /* 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
+ * 3 but the actual value used is 4.)
+ */
+ for (max_blindex = BL_CODES-1; max_blindex >= 3; max_blindex--) {
+ if (bl_tree[bl_order[max_blindex]].Len != 0) break;
+ }
+ /* Update opt_len to include the bit length tree and counts */
+ opt_len += 3*(max_blindex+1) + 5+5+4;
+ Tracev((stderr, "\ndyn trees: dyn %lu, stat %lu", opt_len, static_len));
+
+ return max_blindex;
+}
+
+/* ===========================================================================
+ * 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.
+ */
+static void send_all_trees(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(lcodes-257, 5); /* not +255 as stated in appnote.txt */
+ send_bits(dcodes-1, 5);
+ send_bits(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(bl_tree[bl_order[rank]].Len, 3);
+ }
+
+ send_tree((ct_data *)dyn_ltree, lcodes-1); /* send the literal tree */
+
+ send_tree((ct_data *)dyn_dtree, dcodes-1); /* send the distance tree */
+}
+
+/* ===========================================================================
+ * 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.
+ */
+void flush_block(char *buf, /* input block, or NULL if too old */
+ ulg stored_len, /* length of input block */
+ int pad, /* pad output to byte boundary */
+ int eof) /* true if this is the last block for a file */
+{
+ ulg opt_lenb, static_lenb; /* opt_len and static_len in bytes */
+ int max_blindex; /* index of last bit length code of non zero freq */
+
+ flag_buf[last_flags] = flags; /* Save the flags for the last 8 items */
+
+ /* Construct the literal and distance trees */
+ build_tree((tree_desc *)(&l_desc));
+ Tracev((stderr, "\nlit data: dyn %lu, stat %lu", opt_len, static_len));
+
+ build_tree((tree_desc *)(&d_desc));
+ Tracev((stderr, "\ndist data: dyn %lu, stat %lu", opt_len, 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();
+
+ /* Determine the best encoding. Compute first the block length in bytes */
+ opt_lenb = (opt_len+3+7)>>3;
+ static_lenb = (static_len+3+7)>>3;
+
+ Trace((stderr, "\nopt %lu(%lu) stat %lu(%lu) stored %lu lit %u dist %u ",
+ opt_lenb, opt_len, static_lenb, static_len, stored_len,
+ last_lit, last_dist));
+
+ if (static_lenb <= opt_lenb) opt_lenb = static_lenb;
+
+ if (stored_len+4 <= opt_lenb && buf != (char*)0) {
+ /* 4: two words for the lengths */
+ /* The test buf != NULL is only necessary if LIT_BUFSIZE > WSIZE.
+ * Otherwise we can't have processed more than WSIZE input bytes since
+ * the last block flush, because compression would have been
+ * successful. If LIT_BUFSIZE <= WSIZE, it is never too late to
+ * transform a block into a stored block.
+ */
+ send_bits((STORED_BLOCK<<1)+eof, 3); /* send block type */
+ compressed_len = (compressed_len + 3 + 7) & ~7L;
+ compressed_len += (stored_len + 4) << 3;
+
+ copy_block(buf, (unsigned)stored_len, 1); /* with header */
+
+ } else if (static_lenb == opt_lenb) {
+ send_bits((STATIC_TREES<<1)+eof, 3);
+ compress_block((ct_data *)static_ltree, (ct_data *)static_dtree);
+ compressed_len += 3 + static_len;
+ } else {
+ send_bits((DYN_TREES<<1)+eof, 3);
+ send_all_trees(l_desc.max_code+1, d_desc.max_code+1, max_blindex+1);
+ compress_block((ct_data *)dyn_ltree, (ct_data *)dyn_dtree);
+ compressed_len += 3 + opt_len;
+ }
+ Assert (compressed_len == bits_sent, "bad compressed size");
+ init_block();
+
+ if (eof) {
+ /* Assert (input_len == bytes_in, "bad input size"); */
+ bi_windup();
+ compressed_len += 7; /* align on byte boundary */
+ } else if (pad && (compressed_len % 8) != 0) {
+ send_bits((STORED_BLOCK<<1)+eof, 3); /* send block type */
+ compressed_len = (compressed_len + 3 + 7) & ~7L;
+ copy_block(buf, 0, 1); /* with header */
+ }
+}
+
+/* ===========================================================================
+ * Save the match info and tally the frequency counts. Return true if
+ * the current block must be flushed.
+ */
+int ct_tally (int pack_level, /* Compression level, 1 to 9 */
+ int dist, /* distance of matched string */
+ int lc) /* match length-MIN_MATCH or unmatched char (if dist==0) */
+{
+ l_buf[last_lit++] = (uch)lc;
+ if (dist == 0) {
+ /* lc is the unmatched char */
+ dyn_ltree[lc].Freq++;
+ } else {
+ /* Here, lc is the match length - MIN_MATCH */
+ dist--; /* dist = match distance - 1 */
+ Assert((ush)dist < (ush)MAX_DIST &&
+ (ush)lc <= (ush)(MAX_MATCH-MIN_MATCH) &&
+ (ush)d_code(dist) < (ush)D_CODES, "ct_tally: bad match");
+
+ dyn_ltree[length_code[lc]+LITERALS+1].Freq++;
+ dyn_dtree[d_code(dist)].Freq++;
+
+ d_buf[last_dist++] = (ush)dist;
+ flags |= flag_bit;
+ }
+ flag_bit <<= 1;
+
+ /* Output the flags if they fill a byte: */
+ if ((last_lit & 7) == 0) {
+ flag_buf[last_flags++] = flags;
+ flags = 0, flag_bit = 1;
+ }
+ /* Try to guess if it is profitable to stop the current block here */
+ if (pack_level > 2 && (last_lit & 0xfff) == 0) {
+ /* Compute an upper bound for the compressed length */
+ ulg out_length = (ulg)last_lit*8L;
+ ulg in_length = (ulg)strstart-block_start;
+ int dcode;
+ for (dcode = 0; dcode < D_CODES; dcode++) {
+ out_length += (ulg)dyn_dtree[dcode].Freq*(5L+extra_dbits[dcode]);
+ }
+ out_length >>= 3;
+ Trace((stderr,"\nlast_lit %u, last_dist %u, in %ld, out ~%ld(%ld%%) ",
+ last_lit, last_dist, in_length, out_length,
+ 100L - out_length*100L/in_length));
+ if (last_dist < last_lit/2 && out_length < in_length/2) return 1;
+ }
+ return (last_lit == LIT_BUFSIZE-1 || last_dist == DIST_BUFSIZE);
+ /* 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
+ */
+static void compress_block(ct_data *ltree, /* literal tree */
+ 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 dx = 0; /* running index in d_buf */
+ unsigned fx = 0; /* running index in flag_buf */
+ uch flag = 0; /* current flags */
+ unsigned code; /* the code to send */
+ int extra; /* number of extra bits to send */
+
+ if (last_lit != 0) do {
+ if ((lx & 7) == 0) flag = flag_buf[fx++];
+ lc = l_buf[lx++];
+ if ((flag & 1) == 0) {
+ send_code(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(code+LITERALS+1, ltree); /* send the length code */
+ extra = extra_lbits[code];
+ if (extra != 0) {
+ lc -= base_length[code];
+ send_bits(lc, extra); /* send the extra length bits */
+ }
+ dist = d_buf[dx++];
+ /* Here, dist is the match distance - 1 */
+ code = d_code(dist);
+ Assert (code < D_CODES, "bad d_code");
+
+ send_code(code, dtree); /* send the distance code */
+ extra = extra_dbits[code];
+ if (extra != 0) {
+ dist -= base_dist[code];
+ send_bits(dist, extra); /* send the extra distance bits */
+ }
+ } /* literal or match pair ? */
+ flag >>= 1;
+ } while (lx < last_lit);
+
+ send_code(END_BLOCK, ltree);
+}
--- /dev/null
+/* util.c -- utility functions for gzip support
+
+ Copyright (C) 1997, 1998, 1999, 2001, 2002, 2006 Free Software
+ Foundation, Inc.
+ Copyright (C) 1992-1993 Jean-loup Gailly
+ Copyright (C) 2008 Josh Triplett
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+#include <ctype.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <limits.h>
+#include <stdlib.h>
+#include <unistd.h>
+
+#include "gzip.h"
+
+static int write_buffer(int, void *, unsigned int);
+
+extern ulg crc_32_tab[]; /* crc table, defined below */
+
+/* ===========================================================================
+ * Run a set of bytes through the crc shift register. If s is a NULL
+ * pointer, then initialize the crc shift register contents instead.
+ * Return the current crc in either case.
+ */
+ulg updcrc(uch *s, unsigned n)
+{
+ register ulg c; /* temporary variable */
+
+ static ulg crc = (ulg)0xffffffffL; /* shift register contents */
+
+ if (s == NULL) {
+ c = 0xffffffffL;
+ } else {
+ c = crc;
+ if (n) do {
+ c = crc_32_tab[((int)c ^ (*s++)) & 0xff] ^ (c >> 8);
+ } while (--n);
+ }
+ crc = c;
+ return c ^ 0xffffffffL; /* (instead of ~c for 64-bit machines) */
+}
+
+/* Like the standard read function, except do not attempt to read more
+ than SSIZE_MAX bytes at a time. */
+int
+read_buffer (int fd, void *buf, unsigned int cnt)
+{
+#ifdef SSIZE_MAX
+ if (SSIZE_MAX < cnt)
+ cnt = (unsigned int)SSIZE_MAX;
+#endif
+ return read (fd, buf, cnt);
+}
+
+/* Likewise for 'write'. */
+static int
+write_buffer (int fd, void *buf, unsigned int cnt)
+{
+#ifdef SSIZE_MAX
+ if (SSIZE_MAX < cnt)
+ cnt = (unsigned int)SSIZE_MAX;
+#endif
+ return write (fd, buf, cnt);
+}
+
+/* ===========================================================================
+ * Does the same as write(), but also handles partial pipe writes and checks
+ * for error return.
+ */
+void write_buf(int fd, void *buf, unsigned cnt)
+{
+ unsigned n;
+
+ while ((n = write_buffer (fd, buf, cnt)) != cnt) {
+ if (n == (unsigned)(-1)) {
+ write_error();
+ }
+ cnt -= n;
+ buf = ((char*)buf)+n;
+ }
+}
+
+/* ========================================================================
+ * Error handlers.
+ */
+void
+gzip_error (char *m)
+{
+ fprintf (stderr, "\nzgz: stdin: %s\n", m);
+ exit(ERROR);
+}
+
+void read_error(void)
+{
+ if (errno != 0)
+ perror("\nzgz: stdin");
+ else
+ fprintf(stderr, "\nzgz: stdin: unexpected end of file\n");
+ exit(ERROR);
+}
+
+void write_error(void)
+{
+ perror("\nzgz: stdout");
+ exit(ERROR);
+}
+
+/* ========================================================================
+ * Table of CRC-32's of all single-byte values (made by makecrc.c)
+ */
+ulg crc_32_tab[] = {
+ 0x00000000L, 0x77073096L, 0xee0e612cL, 0x990951baL, 0x076dc419L,
+ 0x706af48fL, 0xe963a535L, 0x9e6495a3L, 0x0edb8832L, 0x79dcb8a4L,
+ 0xe0d5e91eL, 0x97d2d988L, 0x09b64c2bL, 0x7eb17cbdL, 0xe7b82d07L,
+ 0x90bf1d91L, 0x1db71064L, 0x6ab020f2L, 0xf3b97148L, 0x84be41deL,
+ 0x1adad47dL, 0x6ddde4ebL, 0xf4d4b551L, 0x83d385c7L, 0x136c9856L,
+ 0x646ba8c0L, 0xfd62f97aL, 0x8a65c9ecL, 0x14015c4fL, 0x63066cd9L,
+ 0xfa0f3d63L, 0x8d080df5L, 0x3b6e20c8L, 0x4c69105eL, 0xd56041e4L,
+ 0xa2677172L, 0x3c03e4d1L, 0x4b04d447L, 0xd20d85fdL, 0xa50ab56bL,
+ 0x35b5a8faL, 0x42b2986cL, 0xdbbbc9d6L, 0xacbcf940L, 0x32d86ce3L,
+ 0x45df5c75L, 0xdcd60dcfL, 0xabd13d59L, 0x26d930acL, 0x51de003aL,
+ 0xc8d75180L, 0xbfd06116L, 0x21b4f4b5L, 0x56b3c423L, 0xcfba9599L,
+ 0xb8bda50fL, 0x2802b89eL, 0x5f058808L, 0xc60cd9b2L, 0xb10be924L,
+ 0x2f6f7c87L, 0x58684c11L, 0xc1611dabL, 0xb6662d3dL, 0x76dc4190L,
+ 0x01db7106L, 0x98d220bcL, 0xefd5102aL, 0x71b18589L, 0x06b6b51fL,
+ 0x9fbfe4a5L, 0xe8b8d433L, 0x7807c9a2L, 0x0f00f934L, 0x9609a88eL,
+ 0xe10e9818L, 0x7f6a0dbbL, 0x086d3d2dL, 0x91646c97L, 0xe6635c01L,
+ 0x6b6b51f4L, 0x1c6c6162L, 0x856530d8L, 0xf262004eL, 0x6c0695edL,
+ 0x1b01a57bL, 0x8208f4c1L, 0xf50fc457L, 0x65b0d9c6L, 0x12b7e950L,
+ 0x8bbeb8eaL, 0xfcb9887cL, 0x62dd1ddfL, 0x15da2d49L, 0x8cd37cf3L,
+ 0xfbd44c65L, 0x4db26158L, 0x3ab551ceL, 0xa3bc0074L, 0xd4bb30e2L,
+ 0x4adfa541L, 0x3dd895d7L, 0xa4d1c46dL, 0xd3d6f4fbL, 0x4369e96aL,
+ 0x346ed9fcL, 0xad678846L, 0xda60b8d0L, 0x44042d73L, 0x33031de5L,
+ 0xaa0a4c5fL, 0xdd0d7cc9L, 0x5005713cL, 0x270241aaL, 0xbe0b1010L,
+ 0xc90c2086L, 0x5768b525L, 0x206f85b3L, 0xb966d409L, 0xce61e49fL,
+ 0x5edef90eL, 0x29d9c998L, 0xb0d09822L, 0xc7d7a8b4L, 0x59b33d17L,
+ 0x2eb40d81L, 0xb7bd5c3bL, 0xc0ba6cadL, 0xedb88320L, 0x9abfb3b6L,
+ 0x03b6e20cL, 0x74b1d29aL, 0xead54739L, 0x9dd277afL, 0x04db2615L,
+ 0x73dc1683L, 0xe3630b12L, 0x94643b84L, 0x0d6d6a3eL, 0x7a6a5aa8L,
+ 0xe40ecf0bL, 0x9309ff9dL, 0x0a00ae27L, 0x7d079eb1L, 0xf00f9344L,
+ 0x8708a3d2L, 0x1e01f268L, 0x6906c2feL, 0xf762575dL, 0x806567cbL,
+ 0x196c3671L, 0x6e6b06e7L, 0xfed41b76L, 0x89d32be0L, 0x10da7a5aL,
+ 0x67dd4accL, 0xf9b9df6fL, 0x8ebeeff9L, 0x17b7be43L, 0x60b08ed5L,
+ 0xd6d6a3e8L, 0xa1d1937eL, 0x38d8c2c4L, 0x4fdff252L, 0xd1bb67f1L,
+ 0xa6bc5767L, 0x3fb506ddL, 0x48b2364bL, 0xd80d2bdaL, 0xaf0a1b4cL,
+ 0x36034af6L, 0x41047a60L, 0xdf60efc3L, 0xa867df55L, 0x316e8eefL,
+ 0x4669be79L, 0xcb61b38cL, 0xbc66831aL, 0x256fd2a0L, 0x5268e236L,
+ 0xcc0c7795L, 0xbb0b4703L, 0x220216b9L, 0x5505262fL, 0xc5ba3bbeL,
+ 0xb2bd0b28L, 0x2bb45a92L, 0x5cb36a04L, 0xc2d7ffa7L, 0xb5d0cf31L,
+ 0x2cd99e8bL, 0x5bdeae1dL, 0x9b64c2b0L, 0xec63f226L, 0x756aa39cL,
+ 0x026d930aL, 0x9c0906a9L, 0xeb0e363fL, 0x72076785L, 0x05005713L,
+ 0x95bf4a82L, 0xe2b87a14L, 0x7bb12baeL, 0x0cb61b38L, 0x92d28e9bL,
+ 0xe5d5be0dL, 0x7cdcefb7L, 0x0bdbdf21L, 0x86d3d2d4L, 0xf1d4e242L,
+ 0x68ddb3f8L, 0x1fda836eL, 0x81be16cdL, 0xf6b9265bL, 0x6fb077e1L,
+ 0x18b74777L, 0x88085ae6L, 0xff0f6a70L, 0x66063bcaL, 0x11010b5cL,
+ 0x8f659effL, 0xf862ae69L, 0x616bffd3L, 0x166ccf45L, 0xa00ae278L,
+ 0xd70dd2eeL, 0x4e048354L, 0x3903b3c2L, 0xa7672661L, 0xd06016f7L,
+ 0x4969474dL, 0x3e6e77dbL, 0xaed16a4aL, 0xd9d65adcL, 0x40df0b66L,
+ 0x37d83bf0L, 0xa9bcae53L, 0xdebb9ec5L, 0x47b2cf7fL, 0x30b5ffe9L,
+ 0xbdbdf21cL, 0xcabac28aL, 0x53b39330L, 0x24b4a3a6L, 0xbad03605L,
+ 0xcdd70693L, 0x54de5729L, 0x23d967bfL, 0xb3667a2eL, 0xc4614ab8L,
+ 0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL,
+ 0x2d02ef8dL
+};
#include <getopt.h>
#include <time.h>
+extern void gnuzip(int in, int out, char *origname, unsigned long timestamp, int level, int osflag, int rsync);
+
#define BUFLEN (64 * 1024)
#define GZIP_MAGIC0 0x1F
};
#define NUM_SUFFIXES (sizeof suffixes / sizeof suffixes[0])
-static const char gzip_version[] = "zgz 20071002 based on NetBSD gzip 20060927";
+static const char gzip_version[] = "zgz 20081211 based on NetBSD gzip 20060927 and GNU gzip 1.3.12";
static const char gzip_copyright[] = \
" Authors: Faidon Liambotis <paravoid@debian.org>\n"
static int mflag; /* undocumented: don't save timestamp */
static int qflag; /* quiet mode */
static int Tflag; /* force timestamp */
-static uint32_t timestamp; /* forced timestamp */
+static uint32_t timestamp = 0; /* forced timestamp */
static int xflag = -1; /* don't set Extra Flags (i.e. compression level)
binary compatibility with an older, buggy version */
static int osflag = GZIP_OS_UNIX; /* Unix or otherwise */
{ "best", no_argument, 0, '9' },
{ "ascii", no_argument, 0, 'a' },
/* new options */
+ { "gnu", no_argument, 0, 'G' },
+ { "zlib", no_argument, 0, 'Z' },
+ { "rsyncable", no_argument, 0, 'R' },
{ "no-timestamp", no_argument, 0, 'm' },
{ "force-timestamp", no_argument, 0, 'M' },
{ "timestamp", required_argument, 0, 'T' },
{ "osflag", required_argument, 0, 's' },
{ "original-name", required_argument, 0, 'o' },
+ { "filename", required_argument, 0, 'F' },
{ "quirk", required_argument, 0, 'k' },
/* end */
{ "version", no_argument, 0, 'V' },
main(int argc, char **argv)
{
const char *progname = argv[0];
+ int gnu = 0;
char *origname = NULL;
+ int rsync = 0;
int ch;
if (strcmp(progname, "gunzip") == 0 ||
usage();
}
- if (argc > 1 && strcmp(argv[1], "--gnu") == 0) {
- /* omit first argument, i.e. --gnu */
- argv++;
- /* works because "--gnu" is bigger than "gzip" */
- strcpy(argv[0], "gzip");
- execv("/bin/gzip", argv);
-
- /* NOT REACHED */
- fprintf(stderr, "Failed to spawn /bin/gzip\n");
- exit(1);
- } else if (argc > 1 && strcmp(argv[1], "--zlib") == 0) {
- /* skip --zlib argument if existent */
- argc--;
- argv++;
- }
-
-#define OPT_LIST "123456789acdfhLNnMmqrT:Vo:k:s:"
+#define OPT_LIST "123456789acdfhF:GLNnMmqRrT:Vo:k:s:Z"
while ((ch = getopt_long(argc, argv, OPT_LIST, longopts, NULL)) != -1) {
switch (ch) {
+ case 'G':
+ gnu = 1;
+ break;
+ case 'Z':
+ if (gnu)
+ fprintf(stderr, "%s: Both --gnu and --zlib given; using --gnu\n", progname);
+ break;
case '1': case '2': case '3':
case '4': case '5': case '6':
case '7': case '8': case '9':
case 's':
osflag = atoi(optarg);
break;
+ case 'F':
case 'o':
- if (nflag)
- fprintf(stderr, "%s: ignoring original-name because no-name was passed\n", progname);
origname = optarg;
break;
case 'k':
timestamp = atoi(optarg);
Tflag = 1;
break;
+ case 'R':
+ rsync = 1;
+ break;
case 'r':
fprintf(stderr, "%s: recursive is not supported on this version\n", progname);
usage();
argv += optind;
argc -= optind;
- if (argc == 0) {
- handle_stdout(origname);
+ if (gnu) {
+ if (ntfs_quirk || xflag >= 0) {
+ fprintf(stderr, "%s: quirks not supported with --gnu\n", progname);
+ return 1;
+ }
+ if (argc != 0) {
+ fprintf(stderr, "%s: filenames not supported with --gnu\n", progname);
+ return 1;
+ }
+ if (nflag)
+ origname = NULL;
+ if (mflag)
+ timestamp = 0;
+ gnuzip(STDIN_FILENO, STDOUT_FILENO, origname, timestamp, numflag, osflag, rsync);
} else {
- do {
- handle_pathname(argv[0], origname);
- } while (*++argv);
+ if (rsync) {
+ fprintf(stderr, "%s: --rsyncable not supported with --zlib\n", progname);
+ return 1;
+ }
+ if (argc == 0) {
+ handle_stdout(origname);
+ } else {
+ do {
+ handle_pathname(argv[0], origname);
+ } while (*++argv);
+ }
}
exit(exit_value);
}
fprintf(stderr, "%s\n", gzip_version);
fprintf(stderr,
- "usage: zgz [--gnu | --zlib] [-" OPT_LIST "] [<file> [<file> ...]]\n"
- " --gnu use GNU gzip (/bin/gzip)\n"
- " --zlib use zlib's implementation (default)\n"
+ "usage: zgz [-" OPT_LIST "] [<file> [<file> ...]]\n"
+ " -G --gnu use GNU gzip implementation\n"
+ " -Z --zlib use zlib's implementation (default)\n"
" -1 --fast fastest (worst) compression\n"
" -2 .. -8 set compression level\n"
" -9 --best best (slowest) compression\n"
" -q --quiet output no warnings\n"
" -V --version display program version\n"
" -h --help display this help\n"
- " \nzlib-specific options:\n"
" -o NAME\n"
" --original-name NAME use NAME as the original file name\n"
- " -k --quirk QUIRK enable a format quirk (buggy-bsd, ntfs)\n"
+ " -F NAME --filename NAME same as --original-name\n"
" -s --osflag set the OS flag to something different than 03 (Unix)\n"
- " -T --timestamp SECONDS set the timestamp to the specified number of seconds\n");
+ " -T --timestamp SECONDS set the timestamp to the specified number of seconds\n"
+ " \ngnu-specific options:\n"
+ " -R --rsyncable make rsync-friendly archive\n"
+ " \nzlib-specific options:\n"
+ " -k --quirk QUIRK enable a format quirk (buggy-bsd, ntfs)\n");
exit(0);
}
However, this is a modified version that removes support for some options, most
notably decompression (-d) and recursive (-r).
+zgz can also build GNU gzip files with the --gnu option. zgz incorporates parts
+of GNU gzip to support this option.
+
This version supports a variety of special options which aid pristine-gz(1) in
reproducing builds of gz files that are bit-identical to the originals.
=head1 AUTHOR
Matthew R. Green,
+Jean-loup Gailly,
Faidon Liambotis,
Josh Triplett