1/*-
2 * Copyright (c) 2003-2007 Tim Kientzle
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
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.
13 *
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.
24 */
25
26#include "archive_platform.h"
27
28#ifdef HAVE_ERRNO_H
29#include <errno.h>
30#endif
31#include <stdio.h>
32#ifdef HAVE_STDLIB_H
33#include <stdlib.h>
34#endif
35#ifdef HAVE_STRING_H
36#include <string.h>
37#endif
38#ifdef HAVE_UNISTD_H
39#include <unistd.h>
40#endif
41#ifdef HAVE_BZLIB_H
42#include <bzlib.h>
43#endif
44
45#include "archive.h"
46#include "archive_private.h"
47#include "archive_read_private.h"
48
49#if defined(HAVE_BZLIB_H) && defined(BZ_CONFIG_ERROR)
50struct private_data {
51	bz_stream	 stream;
52	char		*out_block;
53	size_t		 out_block_size;
54	char		 valid; /* True = decompressor is initialized */
55	char		 eof; /* True = found end of compressed data. */
56};
57
58/* Bzip2 filter */
59static ssize_t	bzip2_filter_read(struct archive_read_filter *, const void **);
60static int	bzip2_filter_close(struct archive_read_filter *);
61#endif
62
63/*
64 * Note that we can detect bzip2 archives even if we can't decompress
65 * them.  (In fact, we like detecting them because we can give better
66 * error messages.)  So the bid framework here gets compiled even
67 * if bzlib is unavailable.
68 */
69static int	bzip2_reader_bid(struct archive_read_filter_bidder *, struct archive_read_filter *);
70static int	bzip2_reader_init(struct archive_read_filter *);
71
72#if ARCHIVE_VERSION_NUMBER < 4000000
73/* Deprecated; remove in libarchive 4.0 */
74int
75archive_read_support_compression_bzip2(struct archive *a)
76{
77	return archive_read_support_filter_bzip2(a);
78}
79#endif
80
81static const struct archive_read_filter_bidder_vtable
82bzip2_bidder_vtable = {
83	.bid = bzip2_reader_bid,
84	.init = bzip2_reader_init,
85};
86
87int
88archive_read_support_filter_bzip2(struct archive *_a)
89{
90	struct archive_read *a = (struct archive_read *)_a;
91
92	if (__archive_read_register_bidder(a, NULL, "bzip2",
93				&bzip2_bidder_vtable) != ARCHIVE_OK)
94		return (ARCHIVE_FATAL);
95
96#if defined(HAVE_BZLIB_H) && defined(BZ_CONFIG_ERROR)
97	return (ARCHIVE_OK);
98#else
99	archive_set_error(_a, ARCHIVE_ERRNO_MISC,
100	    "Using external bzip2 program");
101	return (ARCHIVE_WARN);
102#endif
103}
104
105/*
106 * Test whether we can handle this data.
107 *
108 * This logic returns zero if any part of the signature fails.  It
109 * also tries to Do The Right Thing if a very short buffer prevents us
110 * from verifying as much as we would like.
111 */
112static int
113bzip2_reader_bid(struct archive_read_filter_bidder *self, struct archive_read_filter *filter)
114{
115	const unsigned char *buffer;
116	ssize_t avail;
117	int bits_checked;
118
119	(void)self; /* UNUSED */
120
121	/* Minimal bzip2 archive is 14 bytes. */
122	buffer = __archive_read_filter_ahead(filter, 14, &avail);
123	if (buffer == NULL)
124		return (0);
125
126	/* First three bytes must be "BZh" */
127	bits_checked = 0;
128	if (memcmp(buffer, "BZh", 3) != 0)
129		return (0);
130	bits_checked += 24;
131
132	/* Next follows a compression flag which must be an ASCII digit. */
133	if (buffer[3] < '1' || buffer[3] > '9')
134		return (0);
135	bits_checked += 5;
136
137	/* After BZh[1-9], there must be either a data block
138	 * which begins with 0x314159265359 or an end-of-data
139	 * marker of 0x177245385090. */
140	if (memcmp(buffer + 4, "\x31\x41\x59\x26\x53\x59", 6) == 0)
141		bits_checked += 48;
142	else if (memcmp(buffer + 4, "\x17\x72\x45\x38\x50\x90", 6) == 0)
143		bits_checked += 48;
144	else
145		return (0);
146
147	return (bits_checked);
148}
149
150#if !defined(HAVE_BZLIB_H) || !defined(BZ_CONFIG_ERROR)
151
152/*
153 * If we don't have the library on this system, we can't actually do the
154 * decompression.  We can, however, still detect compressed archives
155 * and emit a useful message.
156 */
157static int
158bzip2_reader_init(struct archive_read_filter *self)
159{
160	int r;
161
162	r = __archive_read_program(self, "bzip2 -d");
163	/* Note: We set the format here even if __archive_read_program()
164	 * above fails.  We do, after all, know what the format is
165	 * even if we weren't able to read it. */
166	self->code = ARCHIVE_FILTER_BZIP2;
167	self->name = "bzip2";
168	return (r);
169}
170
171
172#else
173
174static const struct archive_read_filter_vtable
175bzip2_reader_vtable = {
176	.read = bzip2_filter_read,
177	.close = bzip2_filter_close,
178};
179
180/*
181 * Setup the callbacks.
182 */
183static int
184bzip2_reader_init(struct archive_read_filter *self)
185{
186	static const size_t out_block_size = 64 * 1024;
187	void *out_block;
188	struct private_data *state;
189
190	self->code = ARCHIVE_FILTER_BZIP2;
191	self->name = "bzip2";
192
193	state = (struct private_data *)calloc(1, sizeof(*state));
194	out_block = (unsigned char *)malloc(out_block_size);
195	if (state == NULL || out_block == NULL) {
196		archive_set_error(&self->archive->archive, ENOMEM,
197		    "Can't allocate data for bzip2 decompression");
198		free(out_block);
199		free(state);
200		return (ARCHIVE_FATAL);
201	}
202
203	self->data = state;
204	state->out_block_size = out_block_size;
205	state->out_block = out_block;
206	self->vtable = &bzip2_reader_vtable;
207
208	return (ARCHIVE_OK);
209}
210
211/*
212 * Return the next block of decompressed data.
213 */
214static ssize_t
215bzip2_filter_read(struct archive_read_filter *self, const void **p)
216{
217	struct private_data *state;
218	size_t decompressed;
219	const char *read_buf;
220	ssize_t ret;
221
222	state = (struct private_data *)self->data;
223
224	if (state->eof) {
225		*p = NULL;
226		return (0);
227	}
228
229	/* Empty our output buffer. */
230	state->stream.next_out = state->out_block;
231	state->stream.avail_out = (uint32_t)state->out_block_size;
232
233	/* Try to fill the output buffer. */
234	for (;;) {
235		if (!state->valid) {
236			if (bzip2_reader_bid(self->bidder, self->upstream) == 0) {
237				state->eof = 1;
238				*p = state->out_block;
239				decompressed = state->stream.next_out
240				    - state->out_block;
241				return (decompressed);
242			}
243			/* Initialize compression library. */
244			ret = BZ2_bzDecompressInit(&(state->stream),
245					   0 /* library verbosity */,
246					   0 /* don't use low-mem algorithm */);
247
248			/* If init fails, try low-memory algorithm instead. */
249			if (ret == BZ_MEM_ERROR)
250				ret = BZ2_bzDecompressInit(&(state->stream),
251					   0 /* library verbosity */,
252					   1 /* do use low-mem algo */);
253
254			if (ret != BZ_OK) {
255				const char *detail = NULL;
256				int err = ARCHIVE_ERRNO_MISC;
257				switch (ret) {
258				case BZ_PARAM_ERROR:
259					detail = "invalid setup parameter";
260					break;
261				case BZ_MEM_ERROR:
262					err = ENOMEM;
263					detail = "out of memory";
264					break;
265				case BZ_CONFIG_ERROR:
266					detail = "mis-compiled library";
267					break;
268				}
269				archive_set_error(&self->archive->archive, err,
270				    "Internal error initializing decompressor%s%s",
271				    detail == NULL ? "" : ": ",
272				    detail);
273				return (ARCHIVE_FATAL);
274			}
275			state->valid = 1;
276		}
277
278		/* stream.next_in is really const, but bzlib
279		 * doesn't declare it so. <sigh> */
280		read_buf =
281		    __archive_read_filter_ahead(self->upstream, 1, &ret);
282		if (read_buf == NULL) {
283			archive_set_error(&self->archive->archive,
284			    ARCHIVE_ERRNO_MISC,
285			    "truncated bzip2 input");
286			return (ARCHIVE_FATAL);
287		}
288		state->stream.next_in = (char *)(uintptr_t)read_buf;
289		state->stream.avail_in = (uint32_t)ret;
290		/* There is no more data, return whatever we have. */
291		if (ret == 0) {
292			state->eof = 1;
293			*p = state->out_block;
294			decompressed = state->stream.next_out
295			    - state->out_block;
296			return (decompressed);
297		}
298
299		/* Decompress as much as we can in one pass. */
300		ret = BZ2_bzDecompress(&(state->stream));
301		__archive_read_filter_consume(self->upstream,
302		    state->stream.next_in - read_buf);
303
304		switch (ret) {
305		case BZ_STREAM_END: /* Found end of stream. */
306			switch (BZ2_bzDecompressEnd(&(state->stream))) {
307			case BZ_OK:
308				break;
309			default:
310				archive_set_error(&(self->archive->archive),
311					  ARCHIVE_ERRNO_MISC,
312					  "Failed to clean up decompressor");
313				return (ARCHIVE_FATAL);
314			}
315			state->valid = 0;
316			/* FALLTHROUGH */
317		case BZ_OK: /* Decompressor made some progress. */
318			/* If we filled our buffer, update stats and return. */
319			if (state->stream.avail_out == 0) {
320				*p = state->out_block;
321				decompressed = state->stream.next_out
322				    - state->out_block;
323				return (decompressed);
324			}
325			break;
326		default: /* Return an error. */
327			archive_set_error(&self->archive->archive,
328			    ARCHIVE_ERRNO_MISC, "bzip decompression failed");
329			return (ARCHIVE_FATAL);
330		}
331	}
332}
333
334/*
335 * Clean up the decompressor.
336 */
337static int
338bzip2_filter_close(struct archive_read_filter *self)
339{
340	struct private_data *state;
341	int ret = ARCHIVE_OK;
342
343	state = (struct private_data *)self->data;
344
345	if (state->valid) {
346		switch (BZ2_bzDecompressEnd(&state->stream)) {
347		case BZ_OK:
348			break;
349		default:
350			archive_set_error(&self->archive->archive,
351					  ARCHIVE_ERRNO_MISC,
352					  "Failed to clean up decompressor");
353			ret = ARCHIVE_FATAL;
354		}
355		state->valid = 0;
356	}
357
358	free(state->out_block);
359	free(state);
360	return (ret);
361}
362
363#endif /* HAVE_BZLIB_H && BZ_CONFIG_ERROR */
364