2 #------------------------------------------------------------------------------
3 # archive: file(1) magic for archive formats (see also "msdos" for self-
4 # extracting compressed archives)
6 # cpio, ar, arc, arj, hpack, lha/lharc, rar, squish, uc2, zip, zoo, etc.
7 # pre-POSIX "tar" archives are handled in the C code.
10 257 string ustar\0 POSIX tar archive
11 257 string ustar\040\040\0 GNU tar archive
15 # Yes, the top two "cpio archive" formats *are* supposed to just be "short".
16 # The idea is to indicate archives produced on machines with the same
17 # byte order as the machine running "file" with "cpio archive", and
18 # to indicate archives produced on machines with the opposite byte order
19 # from the machine running "file" with "byte-swapped cpio archive".
21 # The SVR4 "cpio(4)" hints that there are additional formats, but they
22 # are defined as "short"s; I think all the new formats are
23 # character-header formats and thus are strings, not numbers.
24 0 short 070707 cpio archive
25 0 short 0143561 byte-swapped cpio archive
26 0 string 070707 ASCII cpio archive (pre-SVR4 or odc)
27 0 string 070701 ASCII cpio archive (SVR4 with no CRC)
28 0 string 070702 ASCII cpio archive (SVR4 with CRC)
30 # Debian package (needs to go before regular portable archives)
32 0 string !<arch>\ndebian
33 >8 string debian-split part of multipart Debian package
34 >8 string debian-binary Debian binary package
35 >68 string >\n (format %s)
36 >136 ledate x created: %s
39 0 long 0177555 very old archive
40 0 short 0177555 very old PDP-11 archive
41 0 long 0177545 old archive
42 0 short 0177545 old PDP-11 archive
43 0 long 0100554 apl workspace
44 0 string =<ar> archive
46 # MIPS archive (needs to go before regular portable archives)
48 0 string !<arch>\n__________E MIPS archive
49 >20 string U with MIPS Ucode members
50 >21 string L with MIPSEL members
51 >21 string B with MIPSEB members
52 >19 string L and an EL hash table
53 >19 string B and an EB hash table
54 >22 string X -- out of date
56 0 string -h- Software Tools format archive text
59 # XXX - why are there multiple <ar> thingies? Note that 0x213c6172 is
60 # "!<ar", so, for new-style (4.xBSD/SVR2andup) archives, we have:
62 # 0 string !<arch> current ar archive
63 # 0 long 0x213c6172 archive file
65 # and for SVR1 archives, we have:
67 # 0 string \<ar> System V Release 1 ar archive
68 # 0 string =<ar> archive
70 # XXX - did Aegis really store shared libraries, breakpointed modules,
71 # and absolute code program modules in the same format as new-style
74 0 string !<arch> current ar archive
75 >8 string __.SYMDEF random library
76 >0 belong =65538 - pre SR9.5
77 >0 belong =65539 - post SR9.5
78 >0 beshort 2 - object archive
79 >0 beshort 3 - shared library module
80 >0 beshort 4 - debug break-pointed module
81 >0 beshort 5 - absolute code program module
82 0 string \<ar> System V Release 1 ar archive
83 0 string =<ar> archive
85 # XXX - from "vax", which appears to collect a bunch of byte-swapped
86 # thingies, to help you recognize VAX files on big-endian machines;
87 # with "leshort", "lelong", and "string", that's no longer necessary....
89 0 belong 0x65ff0000 VAX 3.0 archive
90 0 belong 0x3c61723e VAX 5.0 archive
92 0 long 0x213c6172 archive file
93 0 lelong 0177555 very old VAX archive
94 0 leshort 0177555 very old PDP-11 archive
96 # XXX - "pdp" claims that 0177545 can have an __.SYMDEF member and thus
97 # be a random library (it said 0xff65 rather than 0177545).
99 0 lelong 0177545 old VAX archive
100 >8 string __.SYMDEF random library
101 0 leshort 0177545 old PDP-11 archive
102 >8 string __.SYMDEF random library
104 # From "pdp" (but why a 4-byte quantity?)
106 0 lelong 0x39bed PDP-11 old archive
107 0 lelong 0x39bee PDP-11 4.0 archive
109 # ARC archiver, from Daniel Quinlan (quinlan@yggdrasil.com)
111 # The first byte is the magic (0x1a), byte 2 is the compression type for
112 # the first file (0x01 through 0x09), and bytes 3 to 15 are the MS-DOS
113 # filename of the first file (null terminated). Since some types collide
114 # we only test some types on basis of frequency: 0x08 (83%), 0x09 (5%),
115 # 0x02 (5%), 0x03 (3%), 0x04 (2%), 0x06 (2%). 0x01 collides with terminfo.
116 0 lelong&0x8080ffff 0x0000081a ARC archive data, dynamic LZW
117 0 lelong&0x8080ffff 0x0000091a ARC archive data, squashed
118 0 lelong&0x8080ffff 0x0000021a ARC archive data, uncompressed
119 0 lelong&0x8080ffff 0x0000031a ARC archive data, packed
120 0 lelong&0x8080ffff 0x0000041a ARC archive data, squeezed
121 0 lelong&0x8080ffff 0x0000061a ARC archive data, crunched
123 # Acorn archive formats (Disaster prone simpleton, m91dps@ecs.ox.ac.uk)
124 # I can't create either SPARK or ArcFS archives so I have not tested this stuff
125 # [GRR: the original entries collide with ARC, above; replaced with combined
126 # version (not tested)]
127 #0 byte 0x1a RISC OS archive
128 #>1 string archive (ArcFS format)
129 0 string \032archive RISC OS archive (ArcFS format)
131 # ARJ archiver (jason@jarthur.Claremont.EDU)
132 0 leshort 0xea60 ARJ archive data
134 >8 byte &0x04 multi-volume,
135 >8 byte &0x10 slash-switched,
136 >8 byte &0x20 backup,
137 >34 string x original name: %s,
142 >7 byte 4 os: Macintosh
144 >7 byte 6 os: Apple ][ GS
145 >7 byte 7 os: Atari ST
147 >7 byte 9 os: VAX/VMS
150 # HA archiver (Greg Roelofs, newt@uchicago.edu)
151 # This is a really bad format. A file containing HAWAII will match this...
152 #0 string HA HA archive data,
153 #>2 leshort =1 1 file,
154 #>2 leshort >1 %u files,
155 #>4 byte&0x0f =0 first is type CPY
156 #>4 byte&0x0f =1 first is type ASC
157 #>4 byte&0x0f =2 first is type HSC
158 #>4 byte&0x0f =0x0e first is type DIR
159 #>4 byte&0x0f =0x0f first is type SPECIAL
161 # HPACK archiver (Peter Gutmann, pgut1@cs.aukuni.ac.nz)
162 0 string HPAK HPACK archive data
164 # JAM Archive volume format, by Dmitry.Kohmanyuk@UA.net
165 0 string \351,\001JAM\ JAM archive,
166 >7 string >\0 version %.4s
168 >>0x2b string >\0 label %.11s,
169 >>0x27 lelong x serial %08x,
170 >>0x36 string >\0 fstype %.8s
172 # LHARC/LHA archiver (Greg Roelofs, newt@uchicago.edu)
173 2 string -lh0- LHarc 1.x archive data [lh0]
174 2 string -lh1- LHarc 1.x archive data [lh1]
175 2 string -lz4- LHarc 1.x archive data [lz4]
176 2 string -lz5- LHarc 1.x archive data [lz5]
177 # [never seen any but the last; -lh4- reported in comp.compression:]
178 2 string -lzs- LHa 2.x? archive data [lzs]
179 2 string -lh\40- LHa 2.x? archive data [lh ]
180 2 string -lhd- LHa 2.x? archive data [lhd]
181 2 string -lh2- LHa 2.x? archive data [lh2]
182 2 string -lh3- LHa 2.x? archive data [lh3]
183 2 string -lh4- LHa (2.x) archive data [lh4]
184 2 string -lh5- LHa (2.x) archive data [lh5]
185 2 string -lh6- LHa (2.x) archive data [lh6]
186 2 string -lh7- LHa (2.x) archive data [lh7]
187 >20 byte x - header level %d
189 # RAR archiver (Greg Roelofs, newt@uchicago.edu)
190 0 string Rar! RAR archive data
192 # SQUISH archiver (Greg Roelofs, newt@uchicago.edu)
193 0 string SQSH squished archive data (Acorn RISCOS)
195 # UC2 archiver (Greg Roelofs, newt@uchicago.edu)
196 # I can't figure out the self-extracting form of these buggers...
197 0 string UC2\x1a UC2 archive data
199 # ZIP archives (Greg Roelofs, c/o zip-bugs@wkuvx1.wku.edu)
200 0 string PK\003\004 Zip archive data
201 >4 byte 0x09 \b, at least v0.9 to extract
202 >4 byte 0x0a \b, at least v1.0 to extract
203 >4 byte 0x0b \b, at least v1.1 to extract
204 >4 byte 0x14 \b, at least v2.0 to extract
207 20 lelong 0xfdc4a7dc Zoo archive data
211 >32 byte >0 \b, modify: v%d
213 >42 lelong 0xfdc4a7dc \b,
214 >>70 byte >0 extract: v%d
218 10 string #\ This\ is\ a\ shell\ archive shell archive text
221 # LBR. NB: May conflict with the questionable
222 # "binary Computer Graphics Metafile" format.
224 0 string \0\ \ \ \ \ \ \ \ \ \ \ \0\0 LBR archive data
226 # PMA (CP/M derivative of LHA)
228 2 string -pm0- PMarc archive data [pm0]
229 2 string -pm1- PMarc archive data [pm1]
230 2 string -pm2- PMarc archive data [pm2]
231 2 string -pms- PMarc SFX archive (CP/M, DOS)
232 5 string -pc1- PopCom compressed executable (CP/M)
234 # From Rafael Laboissiere <rafael@laboissiere.net>
235 # The Project Revision Control System (see
236 # http://prcs.sourceforge.net) generates a packaged project
237 # file which is recognized by the following entry:
238 0 leshort 0xeb81 PRCS packaged project
241 # by David Necas (Yeti) <yeti@physics.muni.cz>
242 0 string MSCF\0\0\0\0 Microsoft cabinet file data,
247 # by David Necas (Yeti) <yeti@physics.muni.cz>
248 4 string gtktalog\ GTKtalog catalog data,
249 >13 string 3 version 3
250 >>14 beshort 0x677a (gzipped)
251 >>14 beshort !0x677a (not gzipped)
252 >13 string >3 version %s
254 ############################################################################
255 # Parity archive reconstruction file, the 'par' file format now used on Usenet.
256 0 string PAR\0 PARity archive data
257 >48 leshort =0 - Index file
258 >48 leshort >0 - file number %d