1 .\" Copyright (c) 2003-2007 Tim Kientzle
2 .\" All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
13 .\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
14 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
15 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
16 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
17 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
18 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
19 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
20 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
21 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
22 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
25 .\" $FreeBSD: head/lib/libarchive/archive_read.3 191595 2009-04-27 20:13:13Z kientzle $
31 .Nm archive_read_new ,
32 .Nm archive_read_set_filter_options ,
33 .Nm archive_read_set_format_options ,
34 .Nm archive_read_set_options ,
35 .Nm archive_read_support_compression_all ,
36 .Nm archive_read_support_compression_bzip2 ,
37 .Nm archive_read_support_compression_compress ,
38 .Nm archive_read_support_compression_gzip ,
39 .Nm archive_read_support_compression_lzma ,
40 .Nm archive_read_support_compression_none ,
41 .Nm archive_read_support_compression_xz ,
42 .Nm archive_read_support_compression_program ,
43 .Nm archive_read_support_compression_program_signature ,
44 .Nm archive_read_support_format_all ,
45 .Nm archive_read_support_format_ar ,
46 .Nm archive_read_support_format_cpio ,
47 .Nm archive_read_support_format_empty ,
48 .Nm archive_read_support_format_iso9660 ,
49 .Nm archive_read_support_format_mtree,
50 .Nm archive_read_support_format_raw,
51 .Nm archive_read_support_format_tar ,
52 .Nm archive_read_support_format_zip ,
53 .Nm archive_read_open ,
54 .Nm archive_read_open2 ,
55 .Nm archive_read_open_fd ,
56 .Nm archive_read_open_FILE ,
57 .Nm archive_read_open_filename ,
58 .Nm archive_read_open_memory ,
59 .Nm archive_read_next_header ,
60 .Nm archive_read_next_header2 ,
61 .Nm archive_read_data ,
62 .Nm archive_read_data_block ,
63 .Nm archive_read_data_skip ,
64 .\" #if ARCHIVE_API_VERSION < 3
65 .Nm archive_read_data_into_buffer ,
67 .Nm archive_read_data_into_fd ,
68 .Nm archive_read_extract ,
69 .Nm archive_read_extract2 ,
70 .Nm archive_read_extract_set_progress_callback ,
71 .Nm archive_read_close ,
72 .Nm archive_read_finish
73 .Nd functions for reading streaming archives
77 .Fn archive_read_new "void"
79 .Fn archive_read_support_compression_all "struct archive *"
81 .Fn archive_read_support_compression_bzip2 "struct archive *"
83 .Fn archive_read_support_compression_compress "struct archive *"
85 .Fn archive_read_support_compression_gzip "struct archive *"
87 .Fn archive_read_support_compression_lzma "struct archive *"
89 .Fn archive_read_support_compression_none "struct archive *"
91 .Fn archive_read_support_compression_xz "struct archive *"
93 .Fo archive_read_support_compression_program
94 .Fa "struct archive *"
98 .Fo archive_read_support_compression_program_signature
99 .Fa "struct archive *"
100 .Fa "const char *cmd"
101 .Fa "const void *signature"
102 .Fa "size_t signature_length"
105 .Fn archive_read_support_format_all "struct archive *"
107 .Fn archive_read_support_format_ar "struct archive *"
109 .Fn archive_read_support_format_cpio "struct archive *"
111 .Fn archive_read_support_format_empty "struct archive *"
113 .Fn archive_read_support_format_iso9660 "struct archive *"
115 .Fn archive_read_support_format_mtree "struct archive *"
117 .Fn archive_read_support_format_raw "struct archive *"
119 .Fn archive_read_support_format_tar "struct archive *"
121 .Fn archive_read_support_format_zip "struct archive *"
123 .Fn archive_read_set_filter_options "struct archive *" "const char *"
125 .Fn archive_read_set_format_options "struct archive *" "const char *"
127 .Fn archive_read_set_options "struct archive *" "const char *"
129 .Fo archive_read_open
130 .Fa "struct archive *"
131 .Fa "void *client_data"
132 .Fa "archive_open_callback *"
133 .Fa "archive_read_callback *"
134 .Fa "archive_close_callback *"
137 .Fo archive_read_open2
138 .Fa "struct archive *"
139 .Fa "void *client_data"
140 .Fa "archive_open_callback *"
141 .Fa "archive_read_callback *"
142 .Fa "archive_skip_callback *"
143 .Fa "archive_close_callback *"
146 .Fn archive_read_open_FILE "struct archive *" "FILE *file"
148 .Fn archive_read_open_fd "struct archive *" "int fd" "size_t block_size"
150 .Fo archive_read_open_filename
151 .Fa "struct archive *"
152 .Fa "const char *filename"
153 .Fa "size_t block_size"
156 .Fn archive_read_open_memory "struct archive *" "void *buff" "size_t size"
158 .Fn archive_read_next_header "struct archive *" "struct archive_entry **"
160 .Fn archive_read_next_header2 "struct archive *" "struct archive_entry *"
162 .Fn archive_read_data "struct archive *" "void *buff" "size_t len"
164 .Fo archive_read_data_block
165 .Fa "struct archive *"
166 .Fa "const void **buff"
171 .Fn archive_read_data_skip "struct archive *"
172 .\" #if ARCHIVE_API_VERSION < 3
174 .Fn archive_read_data_into_buffer "struct archive *" "void *" "ssize_t len"
177 .Fn archive_read_data_into_fd "struct archive *" "int fd"
179 .Fo archive_read_extract
180 .Fa "struct archive *"
181 .Fa "struct archive_entry *"
185 .Fo archive_read_extract2
186 .Fa "struct archive *src"
187 .Fa "struct archive_entry *"
188 .Fa "struct archive *dest"
191 .Fo archive_read_extract_set_progress_callback
192 .Fa "struct archive *"
193 .Fa "void (*func)(void *)"
194 .Fa "void *user_data"
197 .Fn archive_read_close "struct archive *"
199 .Fn archive_read_finish "struct archive *"
201 These functions provide a complete API for reading streaming archives.
202 The general process is to first create the
204 object, set options, initialize the reader, iterate over the archive
205 headers and associated data, then close the archive and release all
207 The following summary describes the functions in approximately the
208 order they would be used:
209 .Bl -tag -compact -width indent
210 .It Fn archive_read_new
211 Allocates and initializes a
213 object suitable for reading from an archive.
215 .Fn archive_read_support_compression_bzip2 ,
216 .Fn archive_read_support_compression_compress ,
217 .Fn archive_read_support_compression_gzip ,
218 .Fn archive_read_support_compression_lzma ,
219 .Fn archive_read_support_compression_none ,
220 .Fn archive_read_support_compression_xz
222 Enables auto-detection code and decompression support for the
223 specified compression.
226 if the compression is fully supported, or
228 if the compression is supported only through an external program.
229 Note that decompression using an external program is usually slower than
230 decompression through built-in libraries.
233 is always enabled by default.
234 .It Fn archive_read_support_compression_all
235 Enables all available decompression filters.
236 .It Fn archive_read_support_compression_program
237 Data is fed through the specified external program before being dearchived.
238 Note that this disables automatic detection of the compression format,
239 so it makes no sense to specify this in conjunction with any other
240 decompression option.
241 .It Fn archive_read_support_compression_program_signature
242 This feeds data through the specified external program
243 but only if the initial bytes of the data match the specified
246 .Fn archive_read_support_format_all ,
247 .Fn archive_read_support_format_ar ,
248 .Fn archive_read_support_format_cpio ,
249 .Fn archive_read_support_format_empty ,
250 .Fn archive_read_support_format_iso9660 ,
251 .Fn archive_read_support_format_mtree ,
252 .Fn archive_read_support_format_tar ,
253 .Fn archive_read_support_format_zip
255 Enables support---including auto-detection code---for the
256 specified archive format.
258 .Fn archive_read_support_format_tar
259 enables support for a variety of standard tar formats, old-style tar,
260 ustar, pax interchange format, and many common variants.
262 .Fn archive_read_support_format_all
263 enables support for all available formats.
264 Only empty archives are supported by default.
265 .It Fn archive_read_support_format_raw
268 format handler allows libarchive to be used to read arbitrary data.
269 It treats any data stream as an archive with a single entry.
270 The pathname of this entry is
272 all other entry fields are unset.
273 This is not enabled by
274 .Fn archive_read_support_format_all
275 in order to avoid erroneous handling of damaged archives.
277 .Fn archive_read_set_filter_options ,
278 .Fn archive_read_set_format_options ,
279 .Fn archive_read_set_options
281 Specifies options that will be passed to currently-registered
282 filters (including decompression filters) and/or format readers.
283 The argument is a comma-separated list of individual options.
284 Individual options have one of the following forms:
285 .Bl -tag -compact -width indent
287 The option/value pair will be provided to every module.
288 Modules that do not accept an option with this name will ignore it.
290 The option will be provided to every module with a value of
293 The option will be provided to every module with a NULL value.
294 .It Ar module:option=value , Ar module:option , Ar module:!option
295 As above, but the corresponding option and value will be provided
296 only to modules whose name matches
299 The return value will be
301 if any module accepts the option, or
303 if no module accepted the option, or
305 if there was a fatal error while attempting to process the option.
307 The currently supported options are:
308 .Bl -tag -compact -width indent
310 .Bl -tag -compact -width indent
312 Support Joliet extensions.
313 Defaults to enabled, use
318 .It Fn archive_read_open
320 .Fn archive_read_open2 ,
321 except that the skip callback is assumed to be
323 .It Fn archive_read_open2
324 Freeze the settings, open the archive, and prepare for reading entries.
325 This is the most generic version of this call, which accepts
326 four callback functions.
327 Most clients will want to use
328 .Fn archive_read_open_filename ,
329 .Fn archive_read_open_FILE ,
330 .Fn archive_read_open_fd ,
332 .Fn archive_read_open_memory
334 The library invokes the client-provided functions to obtain
335 raw bytes from the archive.
336 .It Fn archive_read_open_FILE
338 .Fn archive_read_open ,
339 except that it accepts a
342 This function should not be used with tape drives or other devices
343 that require strict I/O blocking.
344 .It Fn archive_read_open_fd
346 .Fn archive_read_open ,
347 except that it accepts a file descriptor and block size rather than
348 a set of function pointers.
349 Note that the file descriptor will not be automatically closed at
351 This function is safe for use with tape drives or other blocked devices.
352 .It Fn archive_read_open_file
353 This is a deprecated synonym for
354 .Fn archive_read_open_filename .
355 .It Fn archive_read_open_filename
357 .Fn archive_read_open ,
358 except that it accepts a simple filename and a block size.
359 A NULL filename represents standard input.
360 This function is safe for use with tape drives or other blocked devices.
361 .It Fn archive_read_open_memory
363 .Fn archive_read_open ,
364 except that it accepts a pointer and size of a block of
365 memory containing the archive data.
366 .It Fn archive_read_next_header
367 Read the header for the next entry and return a pointer to
369 .Tn struct archive_entry .
370 This is a convenience wrapper around
371 .Fn archive_read_next_header2
372 that reuses an internal
373 .Tn struct archive_entry
374 object for each request.
375 .It Fn archive_read_next_header2
376 Read the header for the next entry and populate the provided
377 .Tn struct archive_entry .
378 .It Fn archive_read_data
379 Read data associated with the header just read.
380 Internally, this is a convenience function that calls
381 .Fn archive_read_data_block
382 and fills any gaps with nulls so that callers see a single
383 continuous stream of data.
384 .It Fn archive_read_data_block
385 Return the next available block of data for this entry.
387 .Fn archive_read_data ,
389 .Fn archive_read_data_block
390 function avoids copying data and allows you to correctly handle
391 sparse files, as supported by some archive formats.
392 The library guarantees that offsets will increase and that blocks
394 Note that the blocks returned from this function can be much larger
395 than the block size read from disk, due to compression
396 and internal buffer optimizations.
397 .It Fn archive_read_data_skip
398 A convenience function that repeatedly calls
399 .Fn archive_read_data_block
400 to skip all of the data for this archive entry.
401 .\" #if ARCHIVE_API_VERSION < 3
402 .It Fn archive_read_data_into_buffer
403 This function is deprecated and will be removed.
405 .Fn archive_read_data
408 .It Fn archive_read_data_into_fd
409 A convenience function that repeatedly calls
410 .Fn archive_read_data_block
411 to copy the entire entry to the provided file descriptor.
412 .It Fn archive_read_extract , Fn archive_read_extract_set_skip_file
413 A convenience function that wraps the corresponding
414 .Xr archive_write_disk 3
417 .Fn archive_read_extract
418 creates a restore object using
419 .Xr archive_write_disk_new 3
421 .Xr archive_write_disk_set_standard_lookup 3 ,
422 then transparently invokes
423 .Xr archive_write_disk_set_options 3 ,
424 .Xr archive_write_header 3 ,
425 .Xr archive_write_data 3 ,
427 .Xr archive_write_finish_entry 3
428 to create the entry on disk and copy data into it.
431 argument is passed unmodified to
432 .Xr archive_write_disk_set_options 3 .
433 .It Fn archive_read_extract2
434 This is another version of
435 .Fn archive_read_extract
436 that allows you to provide your own restore object.
437 In particular, this allows you to override the standard lookup functions
439 .Xr archive_write_disk_set_group_lookup 3 ,
441 .Xr archive_write_disk_set_user_lookup 3 .
443 .Fn archive_read_extract2
446 argument; you should use
447 .Fn archive_write_disk_set_options
448 to set the restore options yourself.
449 .It Fn archive_read_extract_set_progress_callback
450 Sets a pointer to a user-defined callback that can be used
451 for updating progress displays during extraction.
452 The progress function will be invoked during the extraction of large
454 The progress function will be invoked with the pointer provided to this call.
455 Generally, the data pointed to should include a reference to the archive
456 object and the archive_entry object so that various statistics
457 can be retrieved for the progress display.
458 .It Fn archive_read_close
459 Complete the archive and invoke the close callback.
460 .It Fn archive_read_finish
462 .Fn archive_read_close
463 if it was not invoked manually, then release all resources.
464 Note: In libarchive 1.x, this function was declared to return
466 which made it impossible to detect certain errors when
467 .Fn archive_read_close
468 was invoked implicitly from this function.
469 The declaration is corrected beginning with libarchive 2.0.
472 Note that the library determines most of the relevant information about
473 the archive by inspection.
474 In particular, it automatically detects
478 compression and transparently performs the appropriate decompression.
479 It also automatically detects the archive format.
481 A complete description of the
484 .Tn struct archive_entry
485 objects can be found in the overview manual page for
488 The callback functions must match the following prototypes:
489 .Bl -item -offset indent
492 .Fo archive_read_callback
493 .Fa "struct archive *"
494 .Fa "void *client_data"
495 .Fa "const void **buffer"
498 .\" #if ARCHIVE_API_VERSION < 2
500 .Fo archive_skip_callback
501 .Fa "struct archive *"
502 .Fa "void *client_data"
506 .\" .Ft typedef off_t
507 .\" .Fo archive_skip_callback
508 .\" .Fa "struct archive *"
509 .\" .Fa "void *client_data"
510 .\" .Fa "off_t request"
515 .Fn archive_open_callback "struct archive *" "void *client_data"
518 .Fn archive_close_callback "struct archive *" "void *client_data"
521 The open callback is invoked by
525 if the underlying file or data source is successfully
527 If the open fails, it should call
528 .Fn archive_set_error
529 to register an error code and message and return
532 The read callback is invoked whenever the library
533 requires raw bytes from the archive.
534 The read callback should read data into a buffer,
536 .Li const void **buffer
537 argument to point to the available data, and
538 return a count of the number of bytes available.
539 The library will invoke the read callback again
540 only after it has consumed this data.
541 The library imposes no constraints on the size
542 of the data blocks returned.
543 On end-of-file, the read callback should
545 On error, the read callback should invoke
546 .Fn archive_set_error
547 to register an error code and message and
550 The skip callback is invoked when the
551 library wants to ignore a block of data.
552 The return value is the number of bytes actually
553 skipped, which may differ from the request.
554 If the callback cannot skip data, it should return
556 If the skip callback is not provided (the
559 the library will invoke the read function
560 instead and simply discard the result.
561 A skip callback can provide significant
562 performance gains when reading uncompressed
563 archives from slow disk drives or other media
564 that can skip quickly.
566 The close callback is invoked by archive_close when
567 the archive processing is complete.
568 The callback should return
571 On failure, the callback should invoke
572 .Fn archive_set_error
573 to register an error code and message and
577 The following illustrates basic usage of the library.
579 the callback functions are simply wrappers around the standard
585 .Bd -literal -offset indent
587 list_archive(const char *name)
589 struct mydata *mydata;
591 struct archive_entry *entry;
593 mydata = malloc(sizeof(struct mydata));
594 a = archive_read_new();
596 archive_read_support_compression_all(a);
597 archive_read_support_format_all(a);
598 archive_read_open(a, mydata, myopen, myread, myclose);
599 while (archive_read_next_header(a, &entry) == ARCHIVE_OK) {
600 printf("%s\en",archive_entry_pathname(entry));
601 archive_read_data_skip(a);
603 archive_read_finish(a);
608 myread(struct archive *a, void *client_data, const void **buff)
610 struct mydata *mydata = client_data;
612 *buff = mydata->buff;
613 return (read(mydata->fd, mydata->buff, 10240));
617 myopen(struct archive *a, void *client_data)
619 struct mydata *mydata = client_data;
621 mydata->fd = open(mydata->name, O_RDONLY);
622 return (mydata->fd >= 0 ? ARCHIVE_OK : ARCHIVE_FATAL);
626 myclose(struct archive *a, void *client_data)
628 struct mydata *mydata = client_data;
636 Most functions return zero on success, non-zero on error.
637 The possible return codes include:
639 (the operation succeeded),
641 (the operation succeeded but a non-critical error was encountered),
643 (end-of-archive was encountered),
645 (the operation failed but can be retried),
648 (there was a fatal error; the archive should be closed immediately).
649 Detailed error codes and textual descriptions are available from the
652 .Fn archive_error_string
656 returns a pointer to a freshly allocated
663 .Fn archive_read_data
664 returns a count of bytes actually read or zero at the end of the entry.
670 is returned and an error code and textual description can be retrieved from the
673 .Fn archive_error_string
676 The library expects the client callbacks to behave similarly.
677 If there is an error, you can use
678 .Fn archive_set_error
679 to set an appropriate error code and description,
680 then return one of the non-zero values above.
681 (Note that the value eventually returned to the client may
682 not be the same; many errors that are not critical at the level
683 of basic I/O can prevent the archive from being properly read,
684 thus most I/O errors eventually cause
696 library first appeared in
702 library was written by
703 .An Tim Kientzle Aq kientzle@acm.org .
705 Many traditional archiver programs treat
706 empty files as valid empty archives.
707 For example, many implementations of
709 allow you to append entries to an empty file.
710 Of course, it is impossible to determine the format of an empty file
711 by inspecting the contents, so this library treats empty files as