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.
26 #include "archive_platform.h"
28 __FBSDID("$FreeBSD$");
48 #include "archive_private.h"
49 #include "archive_read_private.h"
51 #if defined(HAVE_BZLIB_H) && defined(BZ_CONFIG_ERROR)
55 size_t out_block_size;
56 char valid; /* True = decompressor is initialized */
57 char eof; /* True = found end of compressed data. */
61 static ssize_t bzip2_filter_read(struct archive_read_filter *, const void **);
62 static int bzip2_filter_close(struct archive_read_filter *);
66 * Note that we can detect bzip2 archives even if we can't decompress
67 * them. (In fact, we like detecting them because we can give better
68 * error messages.) So the bid framework here gets compiled even
69 * if bzlib is unavailable.
71 static int bzip2_reader_bid(struct archive_read_filter_bidder *, struct archive_read_filter *);
72 static int bzip2_reader_init(struct archive_read_filter *);
73 static int bzip2_reader_free(struct archive_read_filter_bidder *);
76 archive_read_support_compression_bzip2(struct archive *_a)
78 struct archive_read *a = (struct archive_read *)_a;
79 struct archive_read_filter_bidder *reader = __archive_read_get_bidder(a);
82 return (ARCHIVE_FATAL);
85 reader->bid = bzip2_reader_bid;
86 reader->init = bzip2_reader_init;
87 reader->options = NULL;
88 reader->free = bzip2_reader_free;
89 #if defined(HAVE_BZLIB_H) && defined(BZ_CONFIG_ERROR)
92 archive_set_error(_a, ARCHIVE_ERRNO_MISC,
93 "Using external bunzip2 program");
94 return (ARCHIVE_WARN);
99 bzip2_reader_free(struct archive_read_filter_bidder *self){
100 (void)self; /* UNUSED */
105 * Test whether we can handle this data.
107 * This logic returns zero if any part of the signature fails. It
108 * also tries to Do The Right Thing if a very short buffer prevents us
109 * from verifying as much as we would like.
112 bzip2_reader_bid(struct archive_read_filter_bidder *self, struct archive_read_filter *filter)
114 const unsigned char *buffer;
118 (void)self; /* UNUSED */
120 /* Minimal bzip2 archive is 14 bytes. */
121 buffer = __archive_read_filter_ahead(filter, 14, &avail);
125 /* First three bytes must be "BZh" */
127 if (buffer[0] != 'B' || buffer[1] != 'Z' || buffer[2] != 'h')
131 /* Next follows a compression flag which must be an ASCII digit. */
132 if (buffer[3] < '1' || buffer[3] > '9')
136 /* After BZh[1-9], there must be either a data block
137 * which begins with 0x314159265359 or an end-of-data
138 * marker of 0x177245385090. */
139 if (memcmp(buffer + 4, "\x31\x41\x59\x26\x53\x59", 6) == 0)
141 else if (memcmp(buffer + 4, "\x17\x72\x45\x38\x50\x90", 6) == 0)
146 return (bits_checked);
149 #if !defined(HAVE_BZLIB_H) || !defined(BZ_CONFIG_ERROR)
152 * If we don't have the library on this system, we can't actually do the
153 * decompression. We can, however, still detect compressed archives
154 * and emit a useful message.
157 bzip2_reader_init(struct archive_read_filter *self)
161 r = __archive_read_program(self, "bunzip2");
162 /* Note: We set the format here even if __archive_read_program()
163 * above fails. We do, after all, know what the format is
164 * even if we weren't able to read it. */
165 self->code = ARCHIVE_COMPRESSION_BZIP2;
166 self->name = "bzip2";
174 * Setup the callbacks.
177 bzip2_reader_init(struct archive_read_filter *self)
179 static const size_t out_block_size = 64 * 1024;
181 struct private_data *state;
183 self->code = ARCHIVE_COMPRESSION_BZIP2;
184 self->name = "bzip2";
186 state = (struct private_data *)calloc(sizeof(*state), 1);
187 out_block = (unsigned char *)malloc(out_block_size);
188 if (self == NULL || state == NULL || out_block == NULL) {
189 archive_set_error(&self->archive->archive, ENOMEM,
190 "Can't allocate data for bzip2 decompression");
193 return (ARCHIVE_FATAL);
197 state->out_block_size = out_block_size;
198 state->out_block = out_block;
199 self->read = bzip2_filter_read;
200 self->skip = NULL; /* not supported */
201 self->close = bzip2_filter_close;
207 * Return the next block of decompressed data.
210 bzip2_filter_read(struct archive_read_filter *self, const void **p)
212 struct private_data *state;
214 const char *read_buf;
217 state = (struct private_data *)self->data;
224 /* Empty our output buffer. */
225 state->stream.next_out = state->out_block;
226 state->stream.avail_out = state->out_block_size;
228 /* Try to fill the output buffer. */
231 if (bzip2_reader_bid(self->bidder, self->upstream) == 0) {
233 *p = state->out_block;
234 decompressed = state->stream.next_out
236 return (decompressed);
238 /* Initialize compression library. */
239 ret = BZ2_bzDecompressInit(&(state->stream),
240 0 /* library verbosity */,
241 0 /* don't use low-mem algorithm */);
243 /* If init fails, try low-memory algorithm instead. */
244 if (ret == BZ_MEM_ERROR)
245 ret = BZ2_bzDecompressInit(&(state->stream),
246 0 /* library verbosity */,
247 1 /* do use low-mem algo */);
250 const char *detail = NULL;
251 int err = ARCHIVE_ERRNO_MISC;
254 detail = "invalid setup parameter";
258 detail = "out of memory";
260 case BZ_CONFIG_ERROR:
261 detail = "mis-compiled library";
264 archive_set_error(&self->archive->archive, err,
265 "Internal error initializing decompressor%s%s",
266 detail == NULL ? "" : ": ",
268 return (ARCHIVE_FATAL);
273 /* stream.next_in is really const, but bzlib
274 * doesn't declare it so. <sigh> */
276 __archive_read_filter_ahead(self->upstream, 1, &ret);
277 if (read_buf == NULL)
278 return (ARCHIVE_FATAL);
279 state->stream.next_in = (char *)(uintptr_t)read_buf;
280 state->stream.avail_in = ret;
281 /* There is no more data, return whatever we have. */
284 *p = state->out_block;
285 decompressed = state->stream.next_out
287 return (decompressed);
290 /* Decompress as much as we can in one pass. */
291 ret = BZ2_bzDecompress(&(state->stream));
292 __archive_read_filter_consume(self->upstream,
293 state->stream.next_in - read_buf);
296 case BZ_STREAM_END: /* Found end of stream. */
297 switch (BZ2_bzDecompressEnd(&(state->stream))) {
301 archive_set_error(&(self->archive->archive),
303 "Failed to clean up decompressor");
304 return (ARCHIVE_FATAL);
308 case BZ_OK: /* Decompressor made some progress. */
309 /* If we filled our buffer, update stats and return. */
310 if (state->stream.avail_out == 0) {
311 *p = state->out_block;
312 decompressed = state->stream.next_out
314 return (decompressed);
317 default: /* Return an error. */
318 archive_set_error(&self->archive->archive,
319 ARCHIVE_ERRNO_MISC, "bzip decompression failed");
320 return (ARCHIVE_FATAL);
326 * Clean up the decompressor.
329 bzip2_filter_close(struct archive_read_filter *self)
331 struct private_data *state;
332 int ret = ARCHIVE_OK;
334 state = (struct private_data *)self->data;
337 switch (BZ2_bzDecompressEnd(&state->stream)) {
341 archive_set_error(&self->archive->archive,
343 "Failed to clean up decompressor");
348 free(state->out_block);
353 #endif /* HAVE_BZLIB_H */