This manual page documents version __VERSION__ of the command. tests each
argument in an attempt to classify it. There are three sets of tests,
performed in this order: filesystem tests, magic tests, and language tests.
The test that succeeds causes the file type to be printed. The type
printed will usually contain one of the words (the file contains only
printing characters and a few common control characters and is probably
safe to read on an terminal), (the file contains the result of compiling a
program in a form understandable to some kernel or another), or meaning
anything else (data is usually or non-printable). Exceptions are well-
known file formats (core files, tar archives) that are known to contain
binary data. When modifying magic files or the program itself, make sure
to Users depend on knowing that all the readable files in a directory have
the word printed. Don't do as Berkeley did and change to The filesystem
tests are based on examining the return from a system call. The program
checks to see if the file is empty, or if it's some sort of special file.
Any known file types appropriate to the system you are running on (sockets,
symbolic links, or named pipes (FIFOs) on those systems that implement
them) are intuited if they are defined in the system header file The magic
tests are used to check for files with data in particular fixed formats.
The canonical example of this is a binary executable (compiled program)
file, whose format is defined in and possibly in the standard include
directory. These files have a stored in a particular place near the
beginning of the file that tells the operating system that the file is a
binary executable, and which of several types thereof. The concept of a
has been applied by extension to data files. Any file with some invariant
identifier at a small fixed offset into the file can usually be described
in this way. The information identifying these files is read from the
compiled magic file or the files in the directory if the compiled file does
not exist. In addition, if or exists, it will be used in preference to the
system magic files. If a file does not match any of the entries in the
magic file, it is examined to see if it seems to be a text file. ASCII,
ISO-8859-x, non-ISO 8-bit extended-ASCII character sets (such as those used
on Macintosh and IBM PC systems), UTF-8-encoded Unicode, UTF-16-encoded
Unicode, and EBCDIC character sets can be distinguished by the different
ranges and sequences of bytes that constitute printable text in each set.
If a file passes any of these tests, its character set is reported. ASCII,
ISO-8859-x, UTF-8, and extended-ASCII files are identified as because they
will be mostly readable on nearly any terminal; UTF-16 and EBCDIC are only
because, while they contain text, it is text that will require translation
before it can be read. In addition, will attempt to determine other
characteristics of text-type files. If the lines of a file are terminated
by CR, CRLF, or NEL, instead of the Unix-standard LF, this will be
reported. Files that contain embedded escape sequences or overstriking
will also be identified. Once has determined the character set used in a
text-type file, it will attempt to determine in what language the file is
written. The language tests look for particular strings (cf. that can
appear anywhere in the first few blocks of a file. For example, the
keyword indicates that the file is most likely a input file, just as the
keyword indicates a C program. These tests are less reliable than the
previous two groups, so they are performed last. The language test
routines also test for some miscellany (such as archives, JSON files). Any
file that cannot be identified as having been written in any of the
character sets listed above is simply said to be Causes the command to
output the file type and creator code as used by older MacOS versions. The
code consists of eight letters, the first describing the file type, the
latter the creator. This option works properly only for file formats that
have the apple-style output defined. Do not prepend filenames to output
lines (brief mode). Write a output file that contains a pre-parsed version
of the magic file or directory. Cause a checking printout of the parsed
form of the magic file. This is usually used in conjunction with the
option to debug a new magic file before installing it. Prints internal
debugging information to stderr. On filesystem errors (file not found
etc), instead of handling the error as regular output as POSIX mandates and
keep going, issue an error message and exit. Exclude the test named in
from the list of tests made to determine the file type. Valid test names
are: application type (only on EMX). Various types of text files (this
test will try to guess the text encoding, irrespective of the setting of
the option). Different text encodings for soft magic tests. Ignored for
backwards compatibility. Prints details of Compound Document Files.
Checks for, and looks inside, compressed files. Checks Comma Separated
Value files. Prints ELF file details, provided soft magic tests are
enabled and the elf magic is found. Examines JSON (RFC-7159) files by
parsing them for compliance. Consults magic files. Examines SIMH tape
files. Examines tar files by verifying the checksum of the 512 byte tar
header. Excluding this test can provide more detailed content description
by using the soft magic method. A synonym for Like but ignore tests that
does not know about. This is intended for compatibility with older
versions of Print a slash-separated list of valid extensions for the file
type found. Use the specified string as the separator between the filename
and the file result returned. Defaults to Read the names of the files to
be examined from (one per line) before the argument list. Either or at
least one filename argument must be present; to test the standard input,
use as a filename argument. Please note that is unwrapped and the enclosed
filenames are processed when this option is encountered and before any
further options processing is done. This allows one to process multiple
lists of files with different command line arguments on the same
invocation. Thus if you want to set the delimiter, you need to do it
before you specify the list of files, like: instead of: This option causes
symlinks not to be followed (on systems that support symbolic links). This
is the default if the environment variable is not defined. Causes the
command to output mime type strings rather than the more traditional human
readable ones. Thus it may say rather than Like but print only the
specified element(s). Don't stop at the first match, keep going.
Subsequent matches will be have the string prepended. (If you want a
newline, see the option.) The magic pattern with the highest strength (see
the option) comes first. Shows a list of patterns and their strength
sorted descending by strength which is used for the matching (see also the
option). This option causes symlinks to be followed, as the like-named
option in (on systems that support symbolic links). This is the default if
the environment variable is defined. Specify an alternate list of files
and directories containing magic. This can be a single item, or a colon-
separated list. If a compiled magic file is found alongside a file or
directory, it will be used instead. Don't pad filenames so that they align
in the output. Force stdout to be flushed after checking each file. This
is only useful if checking a list of files. It is intended to be used by
programs that want filetype output from a pipe. On systems that support or
attempt to preserve the access time of files analyzed, to pretend that
never read them. Set various parameter limits. Don't translate
unprintable characters to \ooo. Normally translates unprintable characters
to their octal representation. Normally, only attempts to read and
determine the type of argument files which reports are ordinary files.
This prevents problems, because reading special files may have peculiar
consequences. Specifying the option causes to also read argument files
which are block or character special files. This is useful for determining
the filesystem types of the data in raw disk partitions, which are block
special files. This option also causes to disregard the file size as
reported by since on some systems it reports a zero size for raw disk
partitions. On systems where libseccomp is available, the option disables
sandboxing which is enabled by default. This option is needed for to
execute external decompressing programs, i.e. when the option is specified
and the built-in decompressors are not available. On systems where
sandboxing is not available, this option has no effect. Print the version
of the program and exit. Try to look inside compressed files. Try to look
inside compressed files, but report information about the contents only not
the compression. Output a null character after the end of the filename.
Nice to the output. This does not affect the separator, which is still
printed. If this option is repeated more than once, then prints just the
filename followed by a NUL followed by the description (or ERROR: text)
followed by a second NUL for each entry. Print a help message and exit.
The environment variable can be used to set the default magic file name.
If that variable is set, then will not attempt to open adds to the value of
this variable as appropriate. The environment variable controls (on
systems that support symbolic links), whether will attempt to follow
symlinks or not. If set, then follows symlink, otherwise it does not.
This is also controlled by the and options. Default compiled list of
magic. Directory containing default magic files. will exit with if the
operation was successful or if an error was encountered. The following
errors cause diagnostic messages, but don't affect the program exit code
(as POSIX requires), unless is specified: A file cannot be found There is
no permission to read a file The file type cannot be determined $ file
file.c file /dev/{wd0a,hda} file.c: C program text file: ELF 32-bit
LSB executable, Intel 80386, version 1 (SYSV), dynamically linked
(uses shared libs), stripped /dev/wd0a: block special (0/0) /dev/hda: block
special (3/0)
$ file -s /dev/wd0{b,d} /dev/wd0b: data /dev/wd0d: x86 boot sector
$ file -s /dev/hda{,1,2,3,4,5,6,7,8,9,10} /dev/hda: x86 boot sector
/dev/hda1: Linux/i386 ext2 filesystem /dev/hda2: x86 boot sector
/dev/hda3: x86 boot sector, extended partition table /dev/hda4:
Linux/i386 ext2 filesystem /dev/hda5: Linux/i386 swap file /dev/hda6:
Linux/i386 swap file /dev/hda7: Linux/i386 swap file /dev/hda8:
Linux/i386 swap file /dev/hda9: empty /dev/hda10: empty
$ file -i file.c file /dev/{wd0a,hda} file.c: text/x-c file:
application/x-executable /dev/hda: application/x-not-regular-file
/dev/wd0a: application/x-not-regular-file
This program is believed to exceed the System V Interface Definition of
FILE(CMD), as near as one can determine from the vague language contained
therein. Its behavior is mostly compatible with the System V program of
the same name. This version knows more magic, however, so it will produce
different (albeit more accurate) output in many cases. The one significant
difference between this version and System V is that this version treats
any white space as a delimiter, so that spaces in pattern strings must be
escaped. For example, Gt]10 string language impress(imPRESS data) in an
existing magic file would have to be changed to Gt]10 string language\
impress (imPRESS data) In addition, in this version, if a pattern
string contains a backslash, it must be escaped. For example
0 string \begindata Andrew Toolkit document in an
existing magic file would have to be changed to
0 string \\begindata Andrew Toolkit document SunOS
releases 3.2 and later from Sun Microsystems include a command derived from
the System V one, but with some extensions. This version differs from
Sun's only in minor ways. It includes the extension of the operator, used
as, for example, Gt]16 longAm]0x7fffffff Gt]0 not
stripped On systems where libseccomp is available, is enforces limiting
system calls to only the ones necessary for the operation of the program.
This enforcement does not provide any security benefit when is asked to
decompress input files running external programs with the option. To
enable execution of external decompressors, one needs to disable sandboxing
using the option. The magic file entries have been collected from various
sources, mainly USENET, and contributed by various authors. Christos
Zoulas (address below) will collect additional or corrected magic file
entries. A consolidation of magic file entries will be distributed
periodically. The order of entries in the magic file is significant.
Depending on what system you are using, the order that they are put
together may be incorrect. If your old command uses a magic file, keep the
old magic file around for comparison purposes (rename it to There has been
a command in every (man page dated November, 1973). The System V version
introduced one significant major change: the external list of magic types.
This slowed the program down slightly but made it a lot more flexible.
This program, based on the System V version, was written by Ian Darwin
without looking at anybody else's source code. John Gilmore revised the
code extensively, making it better than the first version. Geoff Collyer
found several inadequacies and provided some magic file entries.
Contributions of the operator by Rob McMahon, 1989. Guy Harris, made many
changes from 1993 to the present. Primary development and maintenance from
1990 to the present by Christos Zoulas Altered by Chris Lowth 2000: handle
the option to output mime type strings, using an alternative magic file and
internal logic. Altered by Eric Fischer July, 2000, to identify character
codes and attempt to identify the languages of non-ASCII files. Altered by
Reuben Thomas 2007-2011, to improve MIME support, merge MIME and non-MIME
magic, support directories as well as files of magic, apply many bug fixes,
update and fix a lot of magic, improve the build system, improve the
documentation, and rewrite the Python bindings in pure Python. The list of
contributors to the directory (magic files) is too long to include here.
You know who you are; thank you. Many contributors are listed in the
source files. Copyright (c) Ian F. Darwin, Toronto, Canada, 1986-1999.
Covered by the standard Berkeley Software Distribution copyright; see the
file COPYING in the source distribution. The files and were written by
John Gilmore from his public-domain program, and are not covered by the
above license. Please report bugs and send patches to the bug tracker at
or the mailing list at (visit first to subscribe). Fix output so that
tests for MIME and APPLE flags are not needed all over the place, and
actual output is only done in one place. This needs a design. Suggestion:
push possible outputs on to a list, then pick the last-pushed (most
specific, one hopes) value at the end, or use a default if the list is
empty. This should not slow down evaluation. The handling of and printing
\012- between entries is clumsy and complicated; refactor and centralize.
Some of the encoding logic is hard-coded in encoding.c and can be moved to
the magic files if we had a !:charset annotation. Continue to squash all
magic bugs. See Debian BTS for a good source. Store arbitrarily long
strings, for example for %s patterns, so that they can be printed out.
Fixes Debian bug #271672. This can be done by allocating strings in a
string pool, storing the string pool at the end of the magic file and
converting all the string pointers to relative offsets from the string
pool. Add syntax for relative offsets after current level (Debian bug
#466037). Make file -ki work, i.e. give multiple MIME types. Add a zip
library so we can peek inside Office2007 documents to print more details
about their contents. Add an option to print URLs for the sources of the
file descriptions. Combine script searches and add a way to map executable
names to MIME types (e.g. have a magic value for !:mime which causes the
resulting string to be looked up in a table). This would avoid adding the
same magic repeatedly for each new hash-bang interpreter. When a file
descriptor is available, we can skip and adjust the buffer instead of the
hacky buffer management we do now. Fix and to check for consistency at
compile time (duplicate pointing to undefined ). Make / more efficient by
keeping a sorted list of names. Special-case ^ to flip endianness in the
parser so that it does not have to be escaped, and document it. If the
offsets specified internally in the file exceed the buffer size ( variable
in file.h), then we don't seek to that offset, but we give up. It would be
better if buffer managements was done when the file descriptor is available
so we can seek around the file. One must be careful though because this
has performance and thus security considerations, because one can slow down
things by repeatedly seeking. There is support now for keeping separate
buffers and having offsets from the end of the file, but the internal
buffer management still needs an overhaul. You can obtain the original
author's latest version by anonymous FTP on in the directory
