1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
|
$FreeBSD$
libarchive: a library for reading and writing streaming archives
This is all under a BSD license. Use, enjoy, but don't blame me if it breaks!
As of February, 2004, the library proper is fairly complete and compiles
cleanly on FreeBSD 5-CURRENT. The API should be stable now.
Documentation:
* libarchive(3) gives an overview of the library as a whole
* archive_read(3) and archive_write(3) provide detailed calling
sequences for the read and write APIs
* archive_entry(3) details the "struct archive_entry" utility class
* tar(5) documents the "tar" file formats supported by the library
You should also read the copious comments in "archive.h" and the source
code for the sample "bsdtar" program for more details. Please let me know
about any errors or omissions you find. (In particular, I no doubt missed
a few things when researching the tar(5) page.)
Notes:
* This is a heavily stream-oriented system. There is no direct
support for in-place modification or random access and no intention
of ever adding such support. Adding such support would require
sacrificing a lot of other features, so don't bother asking.
* The library is designed to be extended with new compression and
archive formats. The only requirement is that the format be
readable or writable as a stream and that each archive entry be
independent. For example, zip archives can't be written as a
stream because they require the compressed size of the data as part
of the file header. Similarly, some file attributes for zip
archives can't be extracted when streaming because those attributes
are only stored in the end-of-archive central directory and thus
aren't available when the corresponding entry is actually
extracted.
* Under certain circumstances, you can append entries to an archive
by opening the file for reading, skimming to the end of the archive,
noting the file location, then opening it for write with a custom write
callback that seeks to the appropriate position before writing. Be
sure to not enable any compression support if you do this!
* Compression and blocking are handled implicitly and, as far as
possible, transparently. All archive I/O is correctly blocked, even if
it's compressed. On read, the compression format is detected
automatically and the appropriate decompressor is invoked.
* It should be easy to implement a system that reads one
archive and writes entries to another archive, omitting
or adding entries as appropriate along the way. This permits
"re-writing" of archive streams in lieu of in-place modification.
bsdtar has some code to demonstrate this.
* The archive itself is read/written using callback functions.
You can read an archive directly from an in-memory buffer or
write it to a socket, if you wish. There are some utility
functions to provide easy-to-use "open file," etc, capabilities.
* The read/write APIs are designed to allow individual entries
to be read or written to any data source: You can create
a block of data in memory and add it to a tar archive without
first writing a temporary file. You can also read an entry from
an archive and write the data directly to a socket. If you want
to read/write entries to disk, there are convenience functions to
make this especially easy.
* Read supports most common tar formats, including GNU tar,
POSIX-compliant "ustar interchange format", and the
shiny-and-improved POSIX "pax extended interchange format." The
pax format, in particular, eliminates most of the traditional tar
limitations in a standard way that is increasingly well supported.
(GNU tar notably does not support "pax interchange format"; the
GPL-licensed 'star' archiver does, however.) GNU format is only
incompletely supported at this time; if you really need GNU-format
sparse file support, volume headers, or GNU-format split archives,
let me know.
There's also support for a grab-bag of non-tar formats, including
POSIX cpio and shar.
* When writing tar formats, consider using "pax restricted" format
by default. This avoids the pax extensions whenever it can, enabling
them only on entries that cannot be correctly archived with ustar
format. Thus, you get the broad compatibility of ustar with the
safety of pax's support for very long filenames, etc.
* Note: "pax interchange format" is really an extended tar format,
despite what the name says.
|
