2 * Copyright (c) 2003-2007 Tim Kientzle
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
8 * 1. Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 * notice, this list of conditions and the following disclaimer in the
12 * documentation and/or other materials provided with the distribution.
14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
15 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
16 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
17 * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
18 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
19 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
20 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
21 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
22 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
23 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
27 * This code borrows heavily from "compress" source code, which is
28 * protected by the following copyright. (Clause 3 dropped by request
33 * Copyright (c) 1985, 1986, 1992, 1993
34 * The Regents of the University of California. All rights reserved.
36 * This code is derived from software contributed to Berkeley by
37 * Diomidis Spinellis and James A. Woods, derived from original
38 * work by Spencer Thomas and Joseph Orost.
40 * Redistribution and use in source and binary forms, with or without
41 * modification, are permitted provided that the following conditions
43 * 1. Redistributions of source code must retain the above copyright
44 * notice, this list of conditions and the following disclaimer.
45 * 2. Redistributions in binary form must reproduce the above copyright
46 * notice, this list of conditions and the following disclaimer in the
47 * documentation and/or other materials provided with the distribution.
48 * 4. Neither the name of the University nor the names of its contributors
49 * may be used to endorse or promote products derived from this software
50 * without specific prior written permission.
52 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
53 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
54 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
55 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
56 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
57 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
58 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
59 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
60 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
61 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
66 #include "archive_platform.h"
67 __FBSDID("$FreeBSD$");
83 #include "archive_private.h"
84 #include "archive_read_private.h"
87 * Because LZW decompression is pretty simple, I've just implemented
88 * the whole decompressor here (cribbing from "compress" source code,
89 * of course), rather than relying on an external library. I have
90 * made an effort to clarify and simplify the algorithm, so the
91 * names and structure here don't exactly match those used by compress.
95 /* Input variables. */
96 const unsigned char *next_in;
100 size_t bytes_in_section;
102 /* Output variables. */
103 size_t out_block_size;
106 /* Decompression status variables. */
108 int end_of_stream; /* EOF status. */
109 int maxcode; /* Largest code. */
110 int maxcode_bits; /* Length of largest code. */
111 int section_end_code; /* When to increase bits. */
112 int bits; /* Current code length. */
113 int oldcode; /* Previous code. */
114 int finbyte; /* Last byte of prev code. */
117 int free_ent; /* Next dictionary entry. */
118 unsigned char suffix[65536];
119 uint16_t prefix[65536];
122 * Scratch area for expanding dictionary entries. Note:
123 * "worst" case here comes from compressing /dev/zero: the
124 * last code in the dictionary will code a sequence of
125 * 65536-256 zero bytes. Thus, we need stack space to expand
126 * a 65280-byte dictionary entry. (Of course, 32640:1
127 * compression could also be considered the "best" case. ;-)
129 unsigned char *stackp;
130 unsigned char stack[65300];
133 static int compress_bidder_bid(struct archive_read_filter_bidder *, struct archive_read_filter *);
134 static int compress_bidder_init(struct archive_read_filter *);
135 static int compress_bidder_free(struct archive_read_filter_bidder *);
137 static ssize_t compress_filter_read(struct archive_read_filter *, const void **);
138 static int compress_filter_close(struct archive_read_filter *);
140 static int getbits(struct archive_read_filter *, int n);
141 static int next_code(struct archive_read_filter *);
144 archive_read_support_compression_compress(struct archive *_a)
146 struct archive_read *a = (struct archive_read *)_a;
147 struct archive_read_filter_bidder *bidder = __archive_read_get_bidder(a);
150 return (ARCHIVE_FATAL);
153 bidder->bid = compress_bidder_bid;
154 bidder->init = compress_bidder_init;
155 bidder->options = NULL;
156 bidder->free = compress_bidder_free;
161 * Test whether we can handle this data.
163 * This logic returns zero if any part of the signature fails. It
164 * also tries to Do The Right Thing if a very short buffer prevents us
165 * from verifying as much as we would like.
168 compress_bidder_bid(struct archive_read_filter_bidder *self,
169 struct archive_read_filter *filter)
171 const unsigned char *buffer;
175 (void)self; /* UNUSED */
177 buffer = __archive_read_filter_ahead(filter, 2, &avail);
183 if (buffer[0] != 037) /* Verify first ID byte. */
187 if (buffer[1] != 0235) /* Verify second ID byte. */
195 return (bits_checked);
199 * Setup the callbacks.
202 compress_bidder_init(struct archive_read_filter *self)
204 struct private_data *state;
205 static const size_t out_block_size = 64 * 1024;
209 self->code = ARCHIVE_COMPRESSION_COMPRESS;
210 self->name = "compress (.Z)";
212 state = (struct private_data *)calloc(sizeof(*state), 1);
213 out_block = malloc(out_block_size);
214 if (state == NULL || out_block == NULL) {
217 archive_set_error(&self->archive->archive, ENOMEM,
218 "Can't allocate data for %s decompression",
220 return (ARCHIVE_FATAL);
224 state->out_block_size = out_block_size;
225 state->out_block = out_block;
226 self->read = compress_filter_read;
227 self->skip = NULL; /* not supported */
228 self->close = compress_filter_close;
230 /* XXX MOVE THE FOLLOWING OUT OF INIT() XXX */
232 code = getbits(self, 8); /* Skip first signature byte. */
233 code = getbits(self, 8); /* Skip second signature byte. */
235 code = getbits(self, 8);
236 state->maxcode_bits = code & 0x1f;
237 state->maxcode = (1 << state->maxcode_bits);
238 state->use_reset_code = code & 0x80;
240 /* Initialize decompressor. */
241 state->free_ent = 256;
242 state->stackp = state->stack;
243 if (state->use_reset_code)
246 state->section_end_code = (1<<state->bits) - 1;
248 for (code = 255; code >= 0; code--) {
249 state->prefix[code] = 0;
250 state->suffix[code] = code;
258 * Return a block of data from the decompression buffer. Decompress more
262 compress_filter_read(struct archive_read_filter *self, const void **pblock)
264 struct private_data *state;
265 unsigned char *p, *start, *end;
268 state = (struct private_data *)self->data;
269 if (state->end_of_stream) {
273 p = start = (unsigned char *)state->out_block;
274 end = start + state->out_block_size;
276 while (p < end && !state->end_of_stream) {
277 if (state->stackp > state->stack) {
278 *p++ = *--state->stackp;
280 ret = next_code(self);
281 if (ret == ARCHIVE_EOF)
282 state->end_of_stream = ret;
283 else if (ret != ARCHIVE_OK)
293 * Clean up the reader.
296 compress_bidder_free(struct archive_read_filter_bidder *self)
303 * Close and release the filter.
306 compress_filter_close(struct archive_read_filter *self)
308 struct private_data *state = (struct private_data *)self->data;
310 free(state->out_block);
316 * Process the next code and fill the stack with the expansion
317 * of the code. Returns ARCHIVE_FATAL if there is a fatal I/O or
318 * format error, ARCHIVE_EOF if we hit end of data, ARCHIVE_OK otherwise.
321 next_code(struct archive_read_filter *self)
323 struct private_data *state = (struct private_data *)self->data;
326 static int debug_buff[1024];
327 static unsigned debug_index;
329 code = newcode = getbits(self, state->bits);
333 debug_buff[debug_index++] = code;
334 if (debug_index >= sizeof(debug_buff)/sizeof(debug_buff[0]))
337 /* If it's a reset code, reset the dictionary. */
338 if ((code == 256) && state->use_reset_code) {
340 * The original 'compress' implementation blocked its
341 * I/O in a manner that resulted in junk bytes being
342 * inserted after every reset. The next section skips
343 * this junk. (Yes, the number of *bytes* to skip is
344 * a function of the current *bit* length.)
346 int skip_bytes = state->bits -
347 (state->bytes_in_section % state->bits);
348 skip_bytes %= state->bits;
349 state->bits_avail = 0; /* Discard rest of this byte. */
350 while (skip_bytes-- > 0) {
351 code = getbits(self, 8);
355 /* Now, actually do the reset. */
356 state->bytes_in_section = 0;
358 state->section_end_code = (1 << state->bits) - 1;
359 state->free_ent = 257;
361 return (next_code(self));
364 if (code > state->free_ent) {
365 /* An invalid code is a fatal error. */
366 archive_set_error(&(self->archive->archive), -1,
367 "Invalid compressed data");
368 return (ARCHIVE_FATAL);
371 /* Special case for KwKwK string. */
372 if (code >= state->free_ent) {
373 *state->stackp++ = state->finbyte;
374 code = state->oldcode;
377 /* Generate output characters in reverse order. */
378 while (code >= 256) {
379 *state->stackp++ = state->suffix[code];
380 code = state->prefix[code];
382 *state->stackp++ = state->finbyte = code;
384 /* Generate the new entry. */
385 code = state->free_ent;
386 if (code < state->maxcode && state->oldcode >= 0) {
387 state->prefix[code] = state->oldcode;
388 state->suffix[code] = state->finbyte;
391 if (state->free_ent > state->section_end_code) {
393 state->bytes_in_section = 0;
394 if (state->bits == state->maxcode_bits)
395 state->section_end_code = state->maxcode;
397 state->section_end_code = (1 << state->bits) - 1;
400 /* Remember previous code. */
401 state->oldcode = newcode;
406 * Return next 'n' bits from stream.
408 * -1 indicates end of available data.
411 getbits(struct archive_read_filter *self, int n)
413 struct private_data *state = (struct private_data *)self->data;
416 static const int mask[] = {
417 0x00, 0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff,
418 0x1ff, 0x3ff, 0x7ff, 0xfff, 0x1fff, 0x3fff, 0x7fff, 0xffff
421 while (state->bits_avail < n) {
422 if (state->avail_in <= 0) {
424 = __archive_read_filter_ahead(self->upstream,
427 return (ARCHIVE_EOF);
428 if (ret < 0 || state->next_in == NULL)
429 return (ARCHIVE_FATAL);
430 state->avail_in = ret;
431 __archive_read_filter_consume(self->upstream, ret);
433 state->bit_buffer |= *state->next_in++ << state->bits_avail;
435 state->bits_avail += 8;
436 state->bytes_in_section++;
439 code = state->bit_buffer;
440 state->bit_buffer >>= n;
441 state->bits_avail -= n;
443 return (code & mask[n]);