NAME

diff — differential file and directory comparator

SYNOPSIS

DESCRIPTION

The diff utility compares the contents of file1 and file2 and writes to the standard output the list of changes necessary to convert one file into the other. No output is produced if the files are identical.

Output options (mutually exclusive):

-C number --context number

Like -c but produces a diff with number lines of context.

-c

Produces a diff with 3 lines of context. With -c the output format is modified slightly: the output begins with identification of the files involved and their creation dates and then each change is separated by a line with fifteen *'s. The lines removed from file1 are marked with ‘- ’; those added to file2 are marked ‘+ ’. Lines which are changed from one file to the other are marked in both files with ‘! ’. Changes which lie within 3 lines of each other are grouped together on output.

-D string --ifdef string

Creates a merged version of file1 and file2 on the standard output, with C preprocessor controls included so that a compilation of the result without defining string is equivalent to compiling file1, while defining string will yield file2.

-e --ed

Produces output in a form suitable as input for the editor utility, ed(1), which can then be used to convert file1 into file2.

Extra commands are added to the output when comparing directories with -e, so that the result is a sh(1) script for converting text files which are common to the two directories from their state in dir1 to their state in dir2. Note that when comparing directories with -e, the resulting file may no longer be interpreted as an ed(1) script. Output is added to indicate which file each set of ed(1) commands applies to. These hunks can be manually extracted to produce an ed(1) script, which can also be applied with patch(1).

-f --forward-ed

Identical output to that of the -e flag, but in reverse order. It cannot be digested by ed(1).

--help

This option prints a summary to stdout and exits with status 0.

-n

Produces a script similar to that of -e, but in the opposite order and with a count of changed lines on each insert or delete command. This is the form used by rcsdiff.

-q --brief

Just print a line when the files differ. Does not output a list of changes.

-U number --unified number

Like -u but produces a diff with number lines of context.

-u

Produces a unified diff with 3 lines of context. A unified diff is similar to the context diff produced by the -c option. However, unlike with -c, all lines to be changed (added and/or removed) are present in a single section.

--version

This option prints a version string to stdout and exits with status 0.

-y --side-by-side

Output in two columns with a marker between them. The marker can be one of the following:

space

Corresponding lines are identical.

'|'

Corresponding lines are different.

'<'

Files differ and only the first file contains the line.

'>'

Files differ and only the second file contains the line.

Comparison options:

-A algo, --algorithm algo

Configure the algorithm used when comparing files. diff supports 3 algorithms:

myers

The Myers diff algorithm finds the shortest edit which transforms one input into the other. It generally runs in O(N+D²) time, requiring O(N) space, where N is the sum of the lengths of the inputs and D is the length of the difference between them, with a theoretical O(N·D) worst case. If it encounters worst-case input, the implementation used by diff falls back to a less optimal but faster algorithm.

patience

The Patience variant of the Myers algorithm attempts to create more aesthetically pleasing diff output by logically grouping lines.

stone

The Stone algorithm (commonly known as Hunt-McIlroy or Hunt-Szymanski) looks for the longest common subsequence between compared files. Stone encounters worst case performance when there are long common subsequences. In large files this can lead to a significant performance impact. The Stone algorithm is maintained for compatibility.

The diff utility defaults to the Myers algorithm, but will fall back to the Stone algorithm if the input or output options are not supported by the Myers implementation.

-a --text

Treat all files as ASCII text. Normally diff will simply print “Binary files ... differ” if files contain binary characters. Use of this option forces diff to produce a diff.

-B --ignore-blank-lines

Causes chunks that include only blank lines to be ignored.

-b --ignore-space-change

Causes trailing blanks (spaces and tabs) to be ignored, and other strings of blanks to compare equal.

--color=[when]

Color the additions green, and removals red, or the value in the DIFFCOLORS environment variable. The possible values of when are “never”, “always” and “auto”. auto will use color if the output is a tty and the COLORTERM environment variable is set to a non-empty string.

-d --minimal

Try very hard to produce a diff as small as possible. This may consume a lot of processing power and memory when processing large files with many changes.

-F pattern, --show-function-line pattern

Like -p, but display the last line that matches provided pattern.

-I pattern --ignore-matching-lines pattern

Ignores changes, insertions, and deletions whose lines match the extended regular expression pattern. Multiple -I patterns may be specified. All lines in the change must match some pattern for the change to be ignored. See re_format(7) for more information on regular expression patterns.

-i --ignore-case

Ignores the case of letters. E.g., “A” will compare equal to “a”.

-l --paginate

Pass the output through pr(1) to paginate it.

-L label --label label

Print label instead of the first (and second, if this option is specified twice) file name and time in the context or unified diff header.

-p --show-c-function

With unified and context diffs, show with each change the first 40 characters of the last line before the context beginning with a letter, an underscore or a dollar sign. For C and Objective-C source code following standard layout conventions, this will show the prototype of the function the change applies to.

-T --initial-tab

Print a tab rather than a space before the rest of the line for the normal, context or unified output formats. This makes the alignment of tabs in the line consistent.

-t --expand-tabs

Will expand tabs in output lines. Normal or -c output adds character(s) to the front of each line which may screw up the indentation of the original source lines and make the output listing difficult to interpret. This option will preserve the original source's indentation.

-w --ignore-all-space

Is similar to -b --ignore-space-change but causes whitespace (blanks and tabs) to be totally ignored. E.g., “if ( a == b )” will compare equal to “if(a==b)”.

-W number --width number

Output at most number columns when using side by side format. The default value is 130. Note that unless

-t

was specified, diff will always align the second column to a tab stop, so values of --width smaller than approximately five times the value of --tabsize may yield surprising results.

--changed-group-format GFMT

Format input groups in the provided

the format is a string with special keywords:

%<

lines from FILE1

%<

lines from FILE2

--ignore-file-name-case

ignore case when comparing file names

--no-dereference

do not follow symbolic links

--no-ignore-file-name-case

do not ignore case when comparing file names (default)

--normal

default diff output

--speed-large-files

stub option for compatibility with GNU diff

--strip-trailing-cr

strip carriage return on input files

--suppress-common-lines

Do not output common lines when using the side by side format

--tabsize number

Number of spaces representing a tab (default 8)

Directory comparison options:

-N --new-file

If a file is found in only one directory, act as if it was found in the other directory too but was of zero size.

-P --unidirectional-new-file

If a file is found only in dir2, act as if it was found in dir1 too but was of zero size.

-r --recursive

Causes application of diff recursively to common subdirectories encountered.

-S name --starting-file name

Re-starts a directory diff in the middle, beginning with file name.

-s --report-identical-files

Causes diff to report files which are the same, which are otherwise not mentioned.

-X file --exclude-from file

Exclude files and subdirectories from comparison whose basenames match lines in file. Multiple -X options may be specified.

-x pattern --exclude pattern

Exclude files and subdirectories from comparison whose basenames match pattern. Patterns are matched using shell-style globbing via fnmatch(3). Multiple -x options may be specified.

If both arguments are directories, diff sorts the contents of the directories by name, and then runs the regular file diff algorithm, producing a change list, on text files which are different. Binary files which differ, common subdirectories, and files which appear in only one directory are described as such. In directory mode only regular files and directories are compared. If a non-regular file such as a device special file or FIFO is encountered, a diagnostic message is printed.

If only one of file1 and file2 is a directory, diff is applied to the non-directory file and the file contained in the directory file with a filename that is the same as the last component of the non-directory file.

If either file1 or file2 is ‘-’, the standard input is used in its place.

ENVIRONMENT

DIFFCOLORS

The value of this variable is the form add:rm, where add is the ASCII escape sequence for additions and rm is the ASCII escape sequence for deletions. If this is unset, diff uses green for additions and red for removals.

FILES

/tmp/diff.XXXXXXXX

Temporary file used when comparing a device or the standard input. Note that the temporary file is unlinked as soon as it is created so it will not show up in a directory listing.

EXIT STATUS

The diff utility exits with one of the following values:

The --help and --version options exit with a status of 0.

EXAMPLES

Compare old_dir and new_dir recursively generating an unified diff and treating files found only in one of those directories as new files:

$ diff -ruN /path/to/old_dir /path/to/new_dir

Same as above but excluding files matching the expressions “*.h” and “*.c”:

$ diff -ruN -x '*.h' -x '*.c' /path/to/old_dir /path/to/new_dir

Show a single line indicating if the files differ:

$ diff -q /boot/loader.conf /boot/defaults/loader.conf
Files /boot/loader.conf and /boot/defaults/loader.conf differ

Assuming a file named example.txt with the following contents:

FreeBSD is an operating system
Linux is a kernel
OpenBSD is an operating system

Compare stdin with example.txt excluding from the comparison those lines containing either "Linux" or "Open":

$ echo "FreeBSD is an operating system" | diff -q -I 'Linux|Open' example.txt -

SEE ALSO

cmp(1), comm(1), diff3(1), ed(1), patch(1), pr(1), sdiff(1)

James W. Hunt and M. Douglas McIlroy, An Algorithm for Differential File Comparison, Computing Science Technical Report, Bell Laboratories 41, June 1976.

STANDARDS

The diff utility is compliant with the IEEE Std 1003.1-2008 (“POSIX.1”) specification.

The flags [-AaDdIiLlNnPpqSsTtwXxy] are extensions to that specification.

HISTORY

A diff command appeared in Version 6 AT&T UNIX. The diff implementation used in FreeBSD was GNU diff until FreeBSD 11.4. This was replaced in FreeBSD 12.0 by a BSD-licensed implementation written by Todd Miller. Some GNUisms were lost in the process.

libdiff was imported from the Game of Trees version control system and default algorithm was changed to Myers for FreeBSD 15.