compress 7.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246
  1. #------------------------------------------------------------------------------
  2. # $File: compress,v 1.54 2014/03/14 18:47:29 christos Exp $
  3. # compress: file(1) magic for pure-compression formats (no archives)
  4. #
  5. # compress, gzip, pack, compact, huf, squeeze, crunch, freeze, yabba, etc.
  6. #
  7. # Formats for various forms of compressed data
  8. # Formats for "compress" proper have been moved into "compress.c",
  9. # because it tries to uncompress it to figure out what's inside.
  10. # standard unix compress
  11. 0 string \037\235 compress'd data
  12. !:mime application/x-compress
  13. !:apple LZIVZIVU
  14. >2 byte&0x80 >0 block compressed
  15. >2 byte&0x1f x %d bits
  16. # gzip (GNU zip, not to be confused with Info-ZIP or PKWARE zip archiver)
  17. # Edited by Chris Chittleborough <cchittleborough@yahoo.com.au>, March 2002
  18. # * Original filename is only at offset 10 if "extra field" absent
  19. # * Produce shorter output - notably, only report compression methods
  20. # other than 8 ("deflate", the only method defined in RFC 1952).
  21. 0 string \037\213 gzip compressed data
  22. !:mime application/x-gzip
  23. !:strength * 2
  24. >2 byte <8 \b, reserved method
  25. >2 byte >8 \b, unknown method
  26. >3 byte &0x01 \b, ASCII
  27. >3 byte &0x02 \b, has CRC
  28. >3 byte &0x04 \b, extra field
  29. >3 byte&0xC =0x08
  30. >>10 string x \b, was "%s"
  31. >3 byte &0x10 \b, has comment
  32. >3 byte &0x20 \b, encrypted
  33. >4 ledate >0 \b, last modified: %s
  34. >8 byte 2 \b, max compression
  35. >8 byte 4 \b, max speed
  36. >9 byte =0x00 \b, from FAT filesystem (MS-DOS, OS/2, NT)
  37. >9 byte =0x01 \b, from Amiga
  38. >9 byte =0x02 \b, from VMS
  39. >9 byte =0x03 \b, from Unix
  40. >9 byte =0x04 \b, from VM/CMS
  41. >9 byte =0x05 \b, from Atari
  42. >9 byte =0x06 \b, from HPFS filesystem (OS/2, NT)
  43. >9 byte =0x07 \b, from MacOS
  44. >9 byte =0x08 \b, from Z-System
  45. >9 byte =0x09 \b, from CP/M
  46. >9 byte =0x0A \b, from TOPS/20
  47. >9 byte =0x0B \b, from NTFS filesystem (NT)
  48. >9 byte =0x0C \b, from QDOS
  49. >9 byte =0x0D \b, from Acorn RISCOS
  50. # packed data, Huffman (minimum redundancy) codes on a byte-by-byte basis
  51. 0 string \037\036 packed data
  52. !:mime application/octet-stream
  53. >2 belong >1 \b, %d characters originally
  54. >2 belong =1 \b, %d character originally
  55. #
  56. # This magic number is byte-order-independent.
  57. 0 short 0x1f1f old packed data
  58. !:mime application/octet-stream
  59. # XXX - why *two* entries for "compacted data", one of which is
  60. # byte-order independent, and one of which is byte-order dependent?
  61. #
  62. 0 short 0x1fff compacted data
  63. !:mime application/octet-stream
  64. # This string is valid for SunOS (BE) and a matching "short" is listed
  65. # in the Ultrix (LE) magic file.
  66. 0 string \377\037 compacted data
  67. !:mime application/octet-stream
  68. 0 short 0145405 huf output
  69. !:mime application/octet-stream
  70. # bzip2
  71. 0 string BZh bzip2 compressed data
  72. !:mime application/x-bzip2
  73. >3 byte >47 \b, block size = %c00k
  74. # lzip
  75. 0 string LZIP lzip compressed data
  76. !:mime application/x-lzip
  77. >4 byte x \b, version: %d
  78. # squeeze and crunch
  79. # Michael Haardt <michael@cantor.informatik.rwth-aachen.de>
  80. 0 beshort 0x76FF squeezed data,
  81. >4 string x original name %s
  82. 0 beshort 0x76FE crunched data,
  83. >2 string x original name %s
  84. 0 beshort 0x76FD LZH compressed data,
  85. >2 string x original name %s
  86. # Freeze
  87. 0 string \037\237 frozen file 2.1
  88. 0 string \037\236 frozen file 1.0 (or gzip 0.5)
  89. # SCO compress -H (LZH)
  90. 0 string \037\240 SCO compress -H (LZH) data
  91. # European GSM 06.10 is a provisional standard for full-rate speech
  92. # transcoding, prI-ETS 300 036, which uses RPE/LTP (residual pulse
  93. # excitation/long term prediction) coding at 13 kbit/s.
  94. #
  95. # There's only a magic nibble (4 bits); that nibble repeats every 33
  96. # bytes. This isn't suited for use, but maybe we can use it someday.
  97. #
  98. # This will cause very short GSM files to be declared as data and
  99. # mismatches to be declared as data too!
  100. #0 byte&0xF0 0xd0 data
  101. #>33 byte&0xF0 0xd0
  102. #>66 byte&0xF0 0xd0
  103. #>99 byte&0xF0 0xd0
  104. #>132 byte&0xF0 0xd0 GSM 06.10 compressed audio
  105. # bzip a block-sorting file compressor
  106. # by Julian Seward <sewardj@cs.man.ac.uk> and others
  107. #
  108. #0 string BZ bzip compressed data
  109. #>2 byte x \b, version: %c
  110. #>3 string =1 \b, compression block size 100k
  111. #>3 string =2 \b, compression block size 200k
  112. #>3 string =3 \b, compression block size 300k
  113. #>3 string =4 \b, compression block size 400k
  114. #>3 string =5 \b, compression block size 500k
  115. #>3 string =6 \b, compression block size 600k
  116. #>3 string =7 \b, compression block size 700k
  117. #>3 string =8 \b, compression block size 800k
  118. #>3 string =9 \b, compression block size 900k
  119. # lzop from <markus.oberhumer@jk.uni-linz.ac.at>
  120. 0 string \x89\x4c\x5a\x4f\x00\x0d\x0a\x1a\x0a lzop compressed data
  121. >9 beshort <0x0940
  122. >>9 byte&0xf0 =0x00 - version 0.
  123. >>9 beshort&0x0fff x \b%03x,
  124. >>13 byte 1 LZO1X-1,
  125. >>13 byte 2 LZO1X-1(15),
  126. >>13 byte 3 LZO1X-999,
  127. ## >>22 bedate >0 last modified: %s,
  128. >>14 byte =0x00 os: MS-DOS
  129. >>14 byte =0x01 os: Amiga
  130. >>14 byte =0x02 os: VMS
  131. >>14 byte =0x03 os: Unix
  132. >>14 byte =0x05 os: Atari
  133. >>14 byte =0x06 os: OS/2
  134. >>14 byte =0x07 os: MacOS
  135. >>14 byte =0x0A os: Tops/20
  136. >>14 byte =0x0B os: WinNT
  137. >>14 byte =0x0E os: Win32
  138. >9 beshort >0x0939
  139. >>9 byte&0xf0 =0x00 - version 0.
  140. >>9 byte&0xf0 =0x10 - version 1.
  141. >>9 byte&0xf0 =0x20 - version 2.
  142. >>9 beshort&0x0fff x \b%03x,
  143. >>15 byte 1 LZO1X-1,
  144. >>15 byte 2 LZO1X-1(15),
  145. >>15 byte 3 LZO1X-999,
  146. ## >>25 bedate >0 last modified: %s,
  147. >>17 byte =0x00 os: MS-DOS
  148. >>17 byte =0x01 os: Amiga
  149. >>17 byte =0x02 os: VMS
  150. >>17 byte =0x03 os: Unix
  151. >>17 byte =0x05 os: Atari
  152. >>17 byte =0x06 os: OS/2
  153. >>17 byte =0x07 os: MacOS
  154. >>17 byte =0x0A os: Tops/20
  155. >>17 byte =0x0B os: WinNT
  156. >>17 byte =0x0E os: Win32
  157. # 4.3BSD-Quasijarus Strong Compression
  158. # http://minnie.tuhs.org/Quasijarus/compress.html
  159. 0 string \037\241 Quasijarus strong compressed data
  160. # From: Cory Dikkers <cdikkers@swbell.net>
  161. 0 string XPKF Amiga xpkf.library compressed data
  162. 0 string PP11 Power Packer 1.1 compressed data
  163. 0 string PP20 Power Packer 2.0 compressed data,
  164. >4 belong 0x09090909 fast compression
  165. >4 belong 0x090A0A0A mediocre compression
  166. >4 belong 0x090A0B0B good compression
  167. >4 belong 0x090A0C0C very good compression
  168. >4 belong 0x090A0C0D best compression
  169. # 7-zip archiver, from Thomas Klausner (wiz@danbala.tuwien.ac.at)
  170. # http://www.7-zip.org or DOC/7zFormat.txt
  171. #
  172. 0 string 7z\274\257\047\034 7-zip archive data,
  173. >6 byte x version %d
  174. >7 byte x \b.%d
  175. !:mime application/x-7z-compressed
  176. # Type: LZMA
  177. 0 lelong&0xffffff =0x5d
  178. >12 leshort 0xff LZMA compressed data,
  179. !:mime application/x-lzma
  180. >>5 lequad =0xffffffffffffffff streamed
  181. >>5 lequad !0xffffffffffffffff non-streamed, size %lld
  182. >12 leshort 0 LZMA compressed data,
  183. >>5 lequad =0xffffffffffffffff streamed
  184. >>5 lequad !0xffffffffffffffff non-streamed, size %lld
  185. # http://tukaani.org/xz/xz-file-format.txt
  186. 0 ustring \xFD7zXZ\x00 XZ compressed data
  187. !:mime application/x-xz
  188. # https://github.com/ckolivas/lrzip/blob/master/doc/magic.header.txt
  189. 0 string LRZI LRZIP compressed data
  190. >4 byte x - version %d
  191. >5 byte x \b.%d
  192. !:mime application/x-lrzip
  193. # http://fastcompression.blogspot.fi/2013/04/lz4-streaming-format-final.html
  194. 0 lelong 0x184d2204 LZ4 compressed data
  195. !:mime application/x-lz4
  196. 0 lelong 0x184c2102 LZ4 compressed data, legacy format
  197. !:mime application/x-lz4
  198. # AFX compressed files (Wolfram Kleff)
  199. 2 string -afx- AFX compressed file data
  200. # Supplementary magic data for the file(1) command to support
  201. # rzip(1). The format is described in magic(5).
  202. #
  203. # Copyright (C) 2003 by Andrew Tridgell. You may do whatever you want with
  204. # this file.
  205. #
  206. 0 string RZIP rzip compressed data
  207. >4 byte x - version %d
  208. >5 byte x \b.%d
  209. >6 belong x (%d bytes)
  210. 0 string ArC\x01 FreeArc archive <http://freearc.org>
  211. # Type: DACT compressed files
  212. 0 long 0x444354C3 DACT compressed data
  213. >4 byte >-1 (version %i.
  214. >5 byte >-1 %i.
  215. >6 byte >-1 %i)
  216. >7 long >0 , original size: %i bytes
  217. >15 long >30 , block size: %i bytes
  218. # Valve Pack (VPK) files
  219. 0 lelong 0x55aa1234 Valve Pak file
  220. >0x4 lelong x \b, version %u
  221. >0x8 lelong x \b, %u entries