2 .\" $Id: file.man,v 1.44 2003/02/27 20:47:46 christos Exp $
4 .Dt FILE 1 "Copyright but distributable"
8 .Nd determine file type
14 .Op Fl m Ar magicfiles
20 This manual page documents version 3.41 of the
22 utility which tests each argument in an attempt to classify it.
23 There are three sets of tests, performed in this order:
24 file system tests, magic number tests, and language tests.
27 test that succeeds causes the file type to be printed.
29 The type printed will usually contain one of the words
31 (the file contains only
32 printing characters and a few common control
33 characters and is probably safe to read on an
37 (the file contains the result of compiling a program
38 in a form understandable to some
43 meaning anything else (data is usually
46 Exceptions are well-known file formats (core files, tar archives)
47 that are known to contain binary data.
48 When modifying the file
49 .Pa /usr/share/misc/magic
50 or the program itself,
51 .Em "preserve these keywords" .
52 People depend on knowing that all the readable files in a directory
56 Don't do as Berkeley did and change
57 .Dq Li "shell commands text"
59 .Dq Li "shell script" .
61 .Pa /usr/share/misc/magic
62 is built mechanically from a large number of small files in
65 in the source distribution of this program.
67 The file system tests are based on examining the return from a
70 The program checks to see if the file is empty,
71 or if it's some sort of special file.
72 Any known file types appropriate to the system you are running on
73 (sockets, symbolic links, or named pipes (FIFOs) on those systems that
75 are intuited if they are defined in
76 the system header file
79 The magic number tests are used to check for files with data in
80 particular fixed formats.
81 The canonical example of this is a binary executable (compiled program)
83 file, whose format is defined in
87 in the standard include directory.
90 stored in a particular place
91 near the beginning of the file that tells the
94 that the file is a binary executable, and which of several types thereof.
97 has been applied by extension to data files.
98 Any file with some invariant identifier at a small fixed
99 offset into the file can usually be described in this way.
100 The information identifying these files is read from the compiled
102 .Pa /usr/share/misc/magic.mgc ,
104 .Pa /usr/share/misc/magic
105 if the compile file does not exist.
107 If a file does not match any of the entries in the magic file,
108 it is examined to see if it seems to be a text file.
109 ASCII, ISO-8859-x, non-ISO 8-bit extended-ASCII character sets
110 (such as those used on Macintosh and IBM PC systems),
111 UTF-8-encoded Unicode, UTF-16-encoded Unicode, and EBCDIC
112 character sets can be distinguished by the different
113 ranges and sequences of bytes that constitute printable text
115 If a file passes any of these tests, its character set is reported.
116 ASCII, ISO-8859-x, UTF-8, and extended-ASCII files are identified
119 because they will be mostly readable on nearly any terminal;
120 UTF-16 and EBCDIC are only
121 .Dq Li "character data"
123 they contain text, it is text that will require translation
124 before it can be read.
127 will attempt to determine other characteristics of text-type files.
128 If the lines of a file are terminated by CR, CRLF, or NEL, instead
131 LF, this will be reported.
132 Files that contain embedded escape sequences or overstriking
133 will also be identified.
137 has determined the character set used in a text-type file,
139 attempt to determine in what language the file is written.
140 The language tests look for particular strings (cf
142 that can appear anywhere in the first few blocks of a file.
143 For example, the keyword
145 indicates that the file is most likely a
147 input file, just as the keyword
149 indicates a C program.
150 These tests are less reliable than the previous
151 two groups, so they are performed last.
152 The language test routines also test for some miscellany
157 Any file that cannot be identified as having been written
158 in any of the character sets listed above is simply said to be
161 .Bl -tag -width indent
163 Do not prepend filenames to output lines (brief mode).
165 Cause a checking printout of the parsed form of the magic file.
166 This is usually used in conjunction with
168 to debug a new magic file before installing it.
172 output file that contains a pre-parsed version of
175 Read the names of the files to be examined from
178 before the argument list.
181 or at least one filename argument must be present;
182 to test the standard input, use
184 as a filename argument.
185 .It Fl F Ar separator
186 Use the specified separator character instead of
189 Causes the file command to output mime type strings rather than the more
190 traditional human readable ones.
192 .Dq Li "text/plain; charset=us-ascii"
194 .Dq Li "ASCII text" .
195 In order for this option to work, file changes the way
196 it handles files recognised by the command itself (such as many of the
197 text file types, directories etc), and makes use of an alternative
204 Don't stop at the first match, keep going.
206 Specify an alternate list of files containing magic numbers.
207 This can be a single file, or a colon-separated list of files.
209 Force stdout to be flushed after checking each file.
210 This is only useful if checking a list of files.
211 It is intended to be used by programs that want
212 filetype output from a pipe.
214 Don't pad output to align filenames nicely.
216 Print the version of the program and exit.
218 Try to look inside compressed files.
220 option causes symlinks to be followed, as the like-named option in
222 (on systems that support symbolic links).
226 only attempts to read and determine the type of argument files which
228 reports are ordinary files.
229 This prevents problems, because reading special files may have peculiar
235 to also read argument files which are block or character special files.
236 This is useful for determining the file system types of the data in raw
237 disk partitions, which are block special files.
238 This option also causes
240 to disregard the file size as reported by
242 since on some systems it reports a zero size for raw disk partitions.
245 .Bl -tag -width ".Pa /usr/share/misc/magic.mime" -compact
246 .It Pa /usr/share/misc/magic.mgc
247 default compiled list of magic numbers
248 .It Pa /usr/share/misc/magic
249 default list of magic numbers
250 .It Pa /usr/share/misc/magic.mime
251 default list of magic numbers, used to output mime types when the
256 The environment variable
258 can be used to set the default magic number files.
264 .Sh STANDARDS CONFORMANCE
265 This program is believed to exceed the
267 of FILE(CMD), as near as one can determine from the vague language
269 Its behaviour is mostly compatible with the System V program of the same name.
270 This version knows more magic, however, so it will produce
271 different (albeit more accurate) output in many cases.
273 The one significant difference
274 between this version and System V
275 is that this version treats any white space
276 as a delimiter, so that spaces in pattern strings must be escaped.
279 .Dl ">10 string language impress\ (imPRESS data)"
281 in an existing magic file would have to be changed to
283 .Dl ">10 string language\e impress (imPRESS data)"
285 In addition, in this version, if a pattern string contains a backslash,
289 .Dl "0 string \ebegindata Andrew Toolkit document"
291 in an existing magic file would have to be changed to
293 .Dl "0 string \e\ebegindata Andrew Toolkit document"
295 SunOS releases 3.2 and later from Sun Microsystems include a
297 command derived from the System V one, but with some extensions.
298 My version differs from Sun's only in minor ways.
299 It includes the extension of the
304 .Dl ">16 long&0x7fffffff >0 not stripped"
306 The magic file entries have been collected from various sources,
307 mainly USENET, and contributed by various authors.
309 (address below) will collect additional
310 or corrected magic file entries.
311 A consolidation of magic file entries
312 will be distributed periodically.
314 The order of entries in the magic file is significant.
315 Depending on what system you are using, the order that
316 they are put together may be incorrect.
319 command uses a magic file,
320 keep the old magic file around for comparison purposes
322 .Pa /usr/share/misc/magic.orig ) .
325 $ file file.c file /dev/{wd0a,hda}
326 file.c: C program text
327 file: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV),
328 dynamically linked (uses shared libs), stripped
329 /dev/wd0a: block special (0/0)
330 /dev/hda: block special (3/0)
331 $ file -s /dev/wd0{b,d}
333 /dev/wd0d: x86 boot sector
334 $ file -s /dev/hda{,1,2,3,4,5,6,7,8,9,10} # Linux
335 /dev/hda: x86 boot sector
336 /dev/hda1: Linux/i386 ext2 filesystem
337 /dev/hda2: x86 boot sector
338 /dev/hda3: x86 boot sector, extended partition table
339 /dev/hda4: Linux/i386 ext2 filesystem
340 /dev/hda5: Linux/i386 swap file
341 /dev/hda6: Linux/i386 swap file
342 /dev/hda7: Linux/i386 swap file
343 /dev/hda8: Linux/i386 swap file
347 $ file -s /dev/rwd0e # BSD
349 Unix Fast File system (little-endian),
350 last mounted on /usr,
351 last written at Mon Feb 10 13:22:40 2003,
353 number of blocks 28754208,
354 number of data blocks 27812712,
355 number of cylinder groups 3566,
358 minimum percentage of free blocks 5,
359 rotational delay 0ms,
360 disk rotational speed 60rps,
363 $ file -i file.c file /dev/{wd0a,hda}
365 file: application/x-executable, dynamically linked (uses shared libs),
367 /dev/hda: application/x-not-regular-file
368 /dev/wd0a: application/x-not-regular-file
375 since at least Research Version 4
376 (man page dated November, 1973).
377 The System V version introduced one significant major change:
378 the external list of magic number types.
379 This slowed the program down slightly but made it a lot more flexible.
381 This program, based on the System V version,
383 .An Ian Darwin Aq ian@darwinsys.com
384 without looking at anybody else's source code.
387 revised the code extensively, making it better than
390 found several inadequacies
391 and provided some magic file entries.
395 .An Rob McMahon Aq cudcv@warwick.ac.uk ,
398 .An Guy Harris Aq guy@netapp.com ,
399 made many changes from 1993 to the present.
401 Primary development and maintenance from 1990 to the present by
402 .An Christos Zoulas Aq christos@astron.com .
405 .An Chris Lowth Aq chris@lowth.com ,
409 option to output mime type strings and using an alternative
410 magic file and internal logic.
413 .An Eric Fischer Aq enf@pobox.com ,
415 to identify character codes and attempt to identify the languages
420 The list of contributors to the
422 directory (source for the
423 .Pa /usr/share/misc/magic
424 file) is too long to include here.
425 You know who you are; thank you.
427 Copyright (c) Ian F. Darwin, Toronto, Canada, 1986-1999.
428 Covered by the standard Berkeley Software Distribution copyright; see the file
430 in the source distribution.
438 from his public-domain
440 program, and are not covered by the above license.
442 There must be a better way to automate the construction of the
444 file from all the glop in
447 Better yet, the magic file should be compiled into binary (say,
449 or, better yet, fixed-length
451 strings for use in heterogenous network environments) for faster startup.
452 Then the program would run as fast as the Version 7 program of the same name,
453 with the flexibility of the System V version.
457 utility uses several algorithms that favor speed over accuracy,
458 thus it can be misled about the contents of
464 files (primarily for programming languages)
465 is simplistic, inefficient and requires recompilation to update.
469 clause to follow a series of continuation lines.
471 The magic file and keywords should have regular expression support.
474 as a field delimiter is ugly and makes
475 it hard to edit the files, but is entrenched.
477 It might be advisable to allow upper-case letters in keywords
480 commands vs man page macros.
481 Regular expression support would make this easy.
483 The program doesn't grok
485 It should be able to figure
487 by seeing some keywords which
488 appear indented at the start of line.
489 Regular expression support would make this easy.
491 The list of keywords in
493 probably belongs in the
496 This could be done by using some keyword like
498 for the offset value.
500 Another optimisation would be to sort
501 the magic file so that we can just run down all the
502 tests for the first byte, first word, first long, etc, once we
504 Complain about conflicts in the magic file entries.
505 Make a rule that the magic entries sort based on file offset rather
506 than position within the magic file?
508 The program should provide a way to give an estimate
512 We end up removing guesses (e.g.\&
514 as first 5 chars of file) because
515 they are not as good as other guesses (e.g.\&
518 .Dq Li "Return-Path:" ) .
519 Still, if the others don't pan out, it should be
520 possible to use the first guess.
522 This program is slower than some vendors' file commands.
523 The new support for multiple character codes makes it even slower.
525 This manual page, and particularly this section, is too long.
527 You can obtain the original author's latest version by anonymous FTP
531 .Pa /pub/file/file-X.YZ.tar.gz