niceSOFT/diffutils

doc/diff.texi
4 682 строки177 KB
Jim Meyering
maint: udpate GFDL license to 1.3
09 янв 2010, 13:06
09 янв 2010, 13:06cadd497
\input texinfo @c -*-texinfo-*-
@comment %**start of header
@setfilename diff.info
@include version.texi
@settitle Comparing and Merging Files
@syncodeindex vr cp
@setchapternewpage odd
@comment %**end of header
@copying
This manual is for GNU Diffutils
(version @value{VERSION}, @value{UPDATED}),
and documents the @acronym{GNU} @command{diff}, @command{diff3},
@command{sdiff}, and @command{cmp} commands for showing the
differences between files and the @acronym{GNU} @command{patch} command for
using their output to update files.

Copyright @copyright{} 1992-1994, 1998, 2001-2002, 2004, 2006, 2009-2010 Free
Software Foundation, Inc.

@quotation
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software Foundation; with no
Invariant Sections, with the Front-Cover texts being ``A GNU Manual,''
and with the Back-Cover Texts as in (a) below.  A copy of the
license is included in the section entitled ``GNU Free Documentation
License.''

(a) The FSF's Back-Cover Text is: ``You have the freedom to
copy and modify this GNU manual.  Buying copies from the FSF
supports it in developing GNU and promoting software freedom.''
@end quotation
@end copying

@c Debian install-info (up through at least version 1.9.20) uses only the
@c first dircategory.  Put this one first, as it is more useful in practice.
@dircategory Individual utilities
@direntry
* cmp: (diff)Invoking cmp.                      Compare 2 files byte by byte.
* diff: (diff)Invoking diff.                    Compare 2 files line by line.
* diff3: (diff)Invoking diff3.                  Compare 3 files line by line.
* patch: (diff)Invoking patch.                  Apply a patch to a file.
* sdiff: (diff)Invoking sdiff.                  Merge 2 files side-by-side.
@end direntry

@dircategory Text creation and manipulation
@direntry
* Diff: (diff).                 Comparing and merging files.
@end direntry

@titlepage
@title Comparing and Merging Files
@subtitle for Diffutils @value{VERSION} and @code{patch} 2.5.4
@subtitle @value{UPDATED}
@author David MacKenzie, Paul Eggert, and Richard Stallman
@page
@vskip 0pt plus 1filll
@insertcopying
@end titlepage

@shortcontents
@contents

@ifnottex
@node Top
@top Comparing and Merging Files

@insertcopying
@end ifnottex

@menu
* Overview::              Preliminary information.
* Comparison::            What file comparison means.

* Output Formats::        Formats for two-way difference reports.
* Incomplete Lines::      Lines that lack trailing newlines.
* Comparing Directories:: Comparing files and directories.
* Adjusting Output::      Making @command{diff} output prettier.
* diff Performance::      Making @command{diff} smarter or faster.

* Comparing Three Files:: Formats for three-way difference reports.
* diff3 Merging::         Merging from a common ancestor.

* Interactive Merging::   Interactive merging with @command{sdiff}.

* Merging with patch::    Using @command{patch} to change old files into new ones.
* Making Patches::        Tips for making and using patch distributions.

* Invoking cmp::          Compare two files byte by byte.
* Invoking diff::         Compare two files line by line.
* Invoking diff3::        Compare three files line by line.
* Invoking patch::        Apply a diff file to an original.
* Invoking sdiff::        Side-by-side merge of file differences.

* Standards conformance:: Conformance to the @acronym{POSIX} standard.
* Projects::              If you've found a bug or other shortcoming.

* Copying This Manual::   How to make copies of this manual.
* Translations::          Available translations of this manual.
* Index::                 Index.
@end menu

@node Overview
@unnumbered Overview
@cindex overview of @command{diff} and @command{patch}

Computer users often find occasion to ask how two files differ.  Perhaps
one file is a newer version of the other file.  Or maybe the two files
started out as identical copies but were changed by different people.

You can use the @command{diff} command to show differences between two
files, or each corresponding file in two directories.  @command{diff}
outputs differences between files line by line in any of several
formats, selectable by command line options.  This set of differences is
often called a @dfn{diff} or @dfn{patch}.  For files that are identical,
@command{diff} normally produces no output; for binary (non-text) files,
@command{diff} normally reports only that they are different.

You can use the @command{cmp} command to show the byte and line numbers
where two files differ.  @command{cmp} can also show all the bytes
that differ between the two files, side by side.  A way to compare
two files character by character is the Emacs command @kbd{M-x
compare-windows}.  @xref{Other Window, , Other Window, emacs, The @acronym{GNU}
Emacs Manual}, for more information on that command.

You can use the @command{diff3} command to show differences among three
files.  When two people have made independent changes to a common
original, @command{diff3} can report the differences between the original
and the two changed versions, and can produce a merged file that
contains both persons' changes together with warnings about conflicts.

You can use the @command{sdiff} command to merge two files interactively.

You can use the set of differences produced by @command{diff} to distribute
updates to text files (such as program source code) to other people.
This method is especially useful when the differences are small compared
to the complete files.  Given @command{diff} output, you can use the
@command{patch} program to update, or @dfn{patch}, a copy of the file.  If you
think of @command{diff} as subtracting one file from another to produce
their difference, you can think of @command{patch} as adding the difference
to one file to reproduce the other.

This manual first concentrates on making diffs, and later shows how to
use diffs to update files.

@acronym{GNU} @command{diff} was written by Paul Eggert, Mike Haertel,
David Hayes, Richard Stallman, and Len Tower.  Wayne Davison designed and
implemented the unified output format.  The basic algorithm is described
by Eugene W. Myers in ``An O(ND) Difference Algorithm and its Variations'',
@cite{Algorithmica} Vol.@: 1 No.@: 2, 1986, pp.@: 251--266; and in ``A File
Comparison Program'', Webb Miller and Eugene W. Myers,
@cite{Software---Practice and Experience} Vol.@: 15 No.@: 11, 1985,
pp.@: 1025--1040.
@c From: "Gene Myers" <gene@cs.arizona.edu>
@c They are about the same basic algorithm; the Algorithmica
@c paper gives a rigorous treatment and the sub-algorithm for
@c delivering scripts and should be the primary reference, but
@c both should be mentioned.
The algorithm was independently discovered as described by E. Ukkonen in
``Algorithms for Approximate String Matching'',
@cite{Information and Control} Vol.@: 64, 1985, pp.@: 100--118.
@c From: "Gene Myers" <gene@cs.arizona.edu>
@c Date: Wed, 29 Sep 1993 08:27:55 MST
@c Ukkonen should be given credit for also discovering the algorithm used
@c in GNU diff.
Unless the @option{--minimal} option is used, @command{diff} uses a
heuristic by Paul Eggert that limits the cost to @math{O(N^1.5 log N)}
at the price of producing suboptimal output for large inputs with many
differences.  Related algorithms are surveyed by Alfred V. Aho in
section 6.3 of ``Algorithms for Finding Patterns in Strings'',
@cite{Handbook of Theoretical Computer Science} (Jan Van Leeuwen,
ed.), Vol.@: A, @cite{Algorithms and Complexity}, Elsevier/MIT Press,
1990, pp.@: 255--300.

@acronym{GNU} @command{diff3} was written by Randy Smith.  @acronym{GNU}
@command{sdiff} was written by Thomas Lord.  @acronym{GNU} @command{cmp}
was written by Torbj@"orn Granlund and David MacKenzie.

@acronym{GNU} @command{patch} was written mainly by Larry Wall and Paul Eggert;
several @acronym{GNU} enhancements were contributed by Wayne Davison and
David MacKenzie.  Parts of this manual are adapted from a manual page
written by Larry Wall, with his permission.

@node Comparison
@chapter What Comparison Means
@cindex introduction

There are several ways to think about the differences between two files.
One way to think of the differences is as a series of lines that were
deleted from, inserted in, or changed in one file to produce the other
file.  @command{diff} compares two files line by line, finds groups of
lines that differ, and reports each group of differing lines.  It can
report the differing lines in several formats, which have different
purposes.

@acronym{GNU} @command{diff} can show whether files are different
without detailing the differences.  It also provides ways to suppress
certain kinds of differences that are not important to you.  Most
commonly, such differences are changes in the amount of white space
between words or lines.  @command{diff} also provides ways to suppress
differences in alphabetic case or in lines that match a regular
expression that you provide.  These options can accumulate; for
example, you can ignore changes in both white space and alphabetic
case.

Another way to think of the differences between two files is as a
sequence of pairs of bytes that can be either identical or
different.  @command{cmp} reports the differences between two files
byte by byte, instead of line by line.  As a result, it is often
more useful than @command{diff} for comparing binary files.  For text
files, @command{cmp} is useful mainly when you want to know only whether
two files are identical, or whether one file is a prefix of the other.

To illustrate the effect that considering changes byte by byte
can have compared with considering them line by line, think of what
happens if a single newline character is added to the beginning of a
file.  If that file is then compared with an otherwise identical file
that lacks the newline at the beginning, @command{diff} will report that a
blank line has been added to the file, while @command{cmp} will report that
almost every byte of the two files differs.

@command{diff3} normally compares three input files line by line, finds
groups of lines that differ, and reports each group of differing lines.
Its output is designed to make it easy to inspect two different sets of
changes to the same file.

@menu
* Hunks::             Groups of differing lines.
* White Space::       Suppressing differences in white space.
* Blank Lines::       Suppressing differences whose lines are all blank.
* Specified Lines::   Suppressing differences whose lines all match a pattern.
* Case Folding::      Suppressing differences in alphabetic case.
* Brief::             Summarizing which files are different.
* Binary::            Comparing binary files or forcing text comparisons.
@end menu

@node Hunks
@section Hunks
@cindex hunks

When comparing two files, @command{diff} finds sequences of lines common to
both files, interspersed with groups of differing lines called
@dfn{hunks}.  Comparing two identical files yields one sequence of
common lines and no hunks, because no lines differ.  Comparing two
entirely different files yields no common lines and one large hunk that
contains all lines of both files.  In general, there are many ways to
match up lines between two given files.  @command{diff} tries to minimize
the total hunk size by finding large sequences of common lines
interspersed with small hunks of differing lines.

For example, suppose the file @file{F} contains the three lines
@samp{a}, @samp{b}, @samp{c}, and the file @file{G} contains the same
three lines in reverse order @samp{c}, @samp{b}, @samp{a}.  If
@command{diff} finds the line @samp{c} as common, then the command
@samp{diff F G} produces this output:

@example
1,2d0
< a
< b
3a2,3
> b
> a
@end example

@noindent
But if @command{diff} notices the common line @samp{b} instead, it produces
this output:

@example
1c1
< a
---
> c
3c3
< c
---
> a
@end example

@noindent
It is also possible to find @samp{a} as the common line.  @command{diff}
does not always find an optimal matching between the files; it takes
shortcuts to run faster.  But its output is usually close to the
shortest possible.  You can adjust this tradeoff with the
@option{-d} or @option{--minimal} option (@pxref{diff Performance}).

@node White Space
@section Suppressing Differences in Blank and Tab Spacing
@cindex blank and tab difference suppression
@cindex tab and blank difference suppression

The @option{-E} or @option{--ignore-tab-expansion} option ignores the
distinction between tabs and spaces on input.  A tab is considered to be
equivalent to the number of spaces to the next tab stop (@pxref{Tabs}).

The @option{-b} or @option{--ignore-space-change} option is stronger.
It ignores white space at line end, and considers all other sequences of
one or more white space characters within a line to be equivalent.  With this
option, @command{diff} considers the following two lines to be equivalent,
where @samp{$} denotes the line end:

@example
Here lyeth  muche rychnesse  in lytell space.   -- John Heywood$
Here lyeth muche rychnesse in lytell space. -- John Heywood   $
@end example

The @option{-w} or @option{--ignore-all-space} option is stronger still.
It ignores differences even if one line has white space where
the other line has none.  @dfn{White space} characters include
tab, vertical tab, form feed, carriage return, and space;
some locales may define additional characters to be white space.
With this option, @command{diff} considers the
following two lines to be equivalent, where @samp{$} denotes the line
end and @samp{^M} denotes a carriage return:

@example
Here lyeth  muche  rychnesse in lytell space.--  John Heywood$
  He relyeth much erychnes  seinly tells pace.  --John Heywood   ^M$
@end example

For many other programs newline is also a white space character, but
@command{diff} is a line-oriented program and a newline character
always ends a line.  Hence the @option{-w} or
@option{--ignore-all-space} option does not ignore newline-related
changes; it ignores only other white space changes.

@node Blank Lines
@section Suppressing Differences Whose Lines Are All Blank
@cindex blank line difference suppression

The @option{-B} or @option{--ignore-blank-lines} option ignores changes
that consist entirely of blank lines.  With this option, for example, a
file containing
@example
1.  A point is that which has no part.

2.  A line is breadthless length.
-- Euclid, The Elements, I
@end example
@noindent
is considered identical to a file containing
@example
1.  A point is that which has no part.
2.  A line is breadthless length.


-- Euclid, The Elements, I
@end example

Normally this option affects only lines that are completely empty, but
if you also specify the @option{-b} or @option{--ignore-space-change}
option, or the @option{-w} or @option{--ignore-all-space} option,
lines are also affected if they look empty but contain white space.
In other words, @option{-B} is equivalent to @samp{-I '^$'} by
default, but it is equivalent to @option{-I '^[[:space:]]*$'} if
@option{-b} or @option{-w} is also specified.

@node Specified Lines
@section Suppressing Differences Whose Lines All Match a Regular Expression
@cindex regular expression suppression

To ignore insertions and deletions of lines that match a
@command{grep}-style regular expression, use the @option{-I
@var{regexp}} or @option{--ignore-matching-lines=@var{regexp}} option.
You should escape
regular expressions that contain shell metacharacters to prevent the
shell from expanding them.  For example, @samp{diff -I '^[[:digit:]]'} ignores
all changes to lines beginning with a digit.

However, @option{-I} only ignores the insertion or deletion of lines that
contain the regular expression if every changed line in the hunk---every
insertion and every deletion---matches the regular expression.  In other
words, for each nonignorable change, @command{diff} prints the complete set
of changes in its vicinity, including the ignorable ones.

You can specify more than one regular expression for lines to ignore by
using more than one @option{-I} option.  @command{diff} tries to match each
line against each regular expression.

@node Case Folding
@section Suppressing Case Differences
@cindex case difference suppression

@acronym{GNU} @command{diff} can treat lower case letters as
equivalent to their upper case counterparts, so that, for example, it
considers @samp{Funky Stuff}, @samp{funky STUFF}, and @samp{fUNKy
stuFf} to all be the same.  To request this, use the @option{-i} or
@option{--ignore-case} option.

@node Brief
@section Summarizing Which Files Differ
@cindex summarizing which files differ
@cindex brief difference reports

When you only want to find out whether files are different, and you
don't care what the differences are, you can use the summary output
format.  In this format, instead of showing the differences between the
files, @command{diff} simply reports whether files differ.  The @option{-q}
or @option{--brief} option selects this output format.

This format is especially useful when comparing the contents of two
directories.  It is also much faster than doing the normal line by line
comparisons, because @command{diff} can stop analyzing the files as soon as
it knows that there are any differences.

You can also get a brief indication of whether two files differ by using
@command{cmp}.  For files that are identical, @command{cmp} produces no
output.  When the files differ, by default, @command{cmp} outputs the byte
and line number where the first difference occurs, or reports that one
file is a prefix of the other.  You can use
the @option{-s}, @option{--quiet}, or @option{--silent} option to
suppress that information, so that @command{cmp}
produces no output and reports whether the files differ using only its
exit status (@pxref{Invoking cmp}).

@c Fix this.
Unlike @command{diff}, @command{cmp} cannot compare directories; it can only
compare two files.

@node Binary
@section Binary Files and Forcing Text Comparisons
@cindex binary file diff
@cindex text versus binary diff

If @command{diff} thinks that either of the two files it is comparing is
binary (a non-text file), it normally treats that pair of files much as
if the summary output format had been selected (@pxref{Brief}), and
reports only that the binary files are different.  This is because line
by line comparisons are usually not meaningful for binary files.

@command{diff} determines whether a file is text or binary by checking the
first few bytes in the file; the exact number of bytes is system
dependent, but it is typically several thousand.  If every byte in
that part of the file is non-null, @command{diff} considers the file to be
text; otherwise it considers the file to be binary.

Sometimes you might want to force @command{diff} to consider files to be
text.  For example, you might be comparing text files that contain
null characters; @command{diff} would erroneously decide that those are
non-text files.  Or you might be comparing documents that are in a
format used by a word processing system that uses null characters to
indicate special formatting.  You can force @command{diff} to consider all
files to be text files, and compare them line by line, by using the
@option{-a} or @option{--text} option.  If the files you compare using this
option do not in fact contain text, they will probably contain few
newline characters, and the @command{diff} output will consist of hunks
showing differences between long lines of whatever characters the files
contain.

You can also force @command{diff} to report only whether files differ
(but not how).  Use the @option{-q} or @option{--brief} option for
this.

Normally, differing binary files count as trouble because the
resulting @command{diff} output does not capture all the differences.
This trouble causes @command{diff} to exit with status 2.  However,
this trouble cannot occur with the @option{-a} or @option{--text}
option, or with the @option{-q} or @option{--brief} option, as these
options both cause @command{diff} to generate a form of output that
represents differences as requested.

In operating systems that distinguish between text and binary files,
@command{diff} normally reads and writes all data as text.  Use the
@option{--binary} option to force @command{diff} to read and write binary
data instead.  This option has no effect on a @acronym{POSIX}-compliant system
like @acronym{GNU} or traditional Unix.  However, many personal computer
operating systems represent the end of a line with a carriage return
followed by a newline.  On such systems, @command{diff} normally ignores
these carriage returns on input and generates them at the end of each
output line, but with the @option{--binary} option @command{diff} treats
each carriage return as just another input character, and does not
generate a carriage return at the end of each output line.  This can be
useful when dealing with non-text files that are meant to be
interchanged with @acronym{POSIX}-compliant systems.

The @option{--strip-trailing-cr} causes @command{diff} to treat input
lines that end in carriage return followed by newline as if they end
in plain newline.  This can be useful when comparing text that is
imperfectly imported from many personal computer operating systems.
This option affects how lines are read, which in turn affects how they
are compared and output.

If you want to compare two files byte by byte, you can use the
@command{cmp} program with the @option{-l} or @option{--verbose}
option to show the values of each differing byte in the two files.
With @acronym{GNU} @command{cmp}, you can also use the @option{-b} or
@option{--print-bytes} option to show the @acronym{ASCII} representation of
those bytes.  @xref{Invoking cmp}, for more information.

If @command{diff3} thinks that any of the files it is comparing is binary
(a non-text file), it normally reports an error, because such
comparisons are usually not useful.  @command{diff3} uses the same test as
@command{diff} to decide whether a file is binary.  As with @command{diff}, if
the input files contain a few non-text bytes but otherwise are like
text files, you can force @command{diff3} to consider all files to be text
files and compare them line by line by using the @option{-a} or
@option{--text} option.

@node Output Formats
@chapter @command{diff} Output Formats
@cindex output formats
@cindex format of @command{diff} output

@command{diff} has several mutually exclusive options for output format.
The following sections describe each format, illustrating how
@command{diff} reports the differences between two sample input files.

@menu
* Sample diff Input:: Sample @command{diff} input files for examples.
* Context::           Showing differences with the surrounding text.
* Side by Side::      Showing differences in two columns.
* Normal::            Showing differences without surrounding text.
* Scripts::           Generating scripts for other programs.
* If-then-else::      Merging files with if-then-else.
@end menu

@node Sample diff Input
@section Two Sample Input Files
@cindex @command{diff} sample input
@cindex sample input for @command{diff}

Here are two sample files that we will use in numerous examples to
illustrate the output of @command{diff} and how various options can change
it.

This is the file @file{lao}:

@example
The Way that can be told of is not the eternal Way;
The name that can be named is not the eternal name.
The Nameless is the origin of Heaven and Earth;
The Named is the mother of all things.
Therefore let there always be non-being,
  so we may see their subtlety,
And let there always be being,
  so we may see their outcome.
The two are the same,
But after they are produced,
  they have different names.
@end example

This is the file @file{tzu}:

@example
The Nameless is the origin of Heaven and Earth;
The named is the mother of all things.

Therefore let there always be non-being,
  so we may see their subtlety,
And let there always be being,
  so we may see their outcome.
The two are the same,
But after they are produced,
  they have different names.
They both may be called deep and profound.
Deeper and more profound,
The door of all subtleties!
@end example

In this example, the first hunk contains just the first two lines of
@file{lao}, the second hunk contains the fourth line of @file{lao}
opposing the second and third lines of @file{tzu}, and the last hunk
contains just the last three lines of @file{tzu}.

@node Context
@section Showing Differences in Their Context
@cindex context output format
@cindex @samp{!} output format

Usually, when you are looking at the differences between files, you will
also want to see the parts of the files near the lines that differ, to
help you understand exactly what has changed.  These nearby parts of the
files are called the @dfn{context}.

@acronym{GNU} @command{diff} provides two output formats that show context
around the differing lines: @dfn{context format} and @dfn{unified
format}.  It can optionally show in which function or section of the
file the differing lines are found.

If you are distributing new versions of files to other people in the
form of @command{diff} output, you should use one of the output formats
that show context so that they can apply the diffs even if they have
made small changes of their own to the files.  @command{patch} can apply
the diffs in this case by searching in the files for the lines of
context around the differing lines; if those lines are actually a few
lines away from where the diff says they are, @command{patch} can adjust
the line numbers accordingly and still apply the diff correctly.
@xref{Imperfect}, for more information on using @command{patch} to apply
imperfect diffs.

@menu
* Context Format::  An output format that shows surrounding lines.
* Unified Format::  A more compact output format that shows context.
* Sections::        Showing which sections of the files differences are in.
* Alternate Names:: Showing alternate file names in context headers.
@end menu

@node Context Format
@subsection Context Format

The context output format shows several lines of context around the
lines that differ.  It is the standard format for distributing updates
to source code.

To select this output format, use the @option{-C @var{lines}},
@option{--context@r{[}=@var{lines}@r{]}}, or @option{-c} option.  The
argument @var{lines} that some of these options take is the number of
lines of context to show.  If you do not specify @var{lines}, it
defaults to three.  For proper operation, @command{patch} typically needs
at least two lines of context.

@menu
* Example Context::  Sample output in context format.
* Less Context::     Another sample with less context.
* Detailed Context:: A detailed description of the context output format.
@end menu

@node Example Context
@subsubsection An Example of Context Format

Here is the output of @samp{diff -c lao tzu} (@pxref{Sample diff Input},
for the complete contents of the two files).  Notice that up to three
lines that are not different are shown around each line that is
different; they are the context lines.  Also notice that the first two
hunks have run together, because their contents overlap.

@example
*** lao	2002-02-21 23:30:39.942229878 -0800
--- tzu	2002-02-21 23:30:50.442260588 -0800
***************
*** 1,7 ****
- The Way that can be told of is not the eternal Way;
- The name that can be named is not the eternal name.
  The Nameless is the origin of Heaven and Earth;
! The Named is the mother of all things.
  Therefore let there always be non-being,
    so we may see their subtlety,
  And let there always be being,
--- 1,6 ----
  The Nameless is the origin of Heaven and Earth;
! The named is the mother of all things.
! @-
  Therefore let there always be non-being,
    so we may see their subtlety,
  And let there always be being,
***************
*** 9,11 ****
--- 8,13 ----
  The two are the same,
  But after they are produced,
    they have different names.
+ They both may be called deep and profound.
+ Deeper and more profound,
+ The door of all subtleties!
@end example

@node Less Context
@subsubsection An Example of Context Format with Less Context

Here is the output of @samp{diff -C 1 lao tzu} (@pxref{Sample diff
Input}, for the complete contents of the two files).  Notice that at
most one context line is reported here.

@example
*** lao	2002-02-21 23:30:39.942229878 -0800
--- tzu	2002-02-21 23:30:50.442260588 -0800
***************
*** 1,5 ****
- The Way that can be told of is not the eternal Way;
- The name that can be named is not the eternal name.
  The Nameless is the origin of Heaven and Earth;
! The Named is the mother of all things.
  Therefore let there always be non-being,
--- 1,4 ----
  The Nameless is the origin of Heaven and Earth;
! The named is the mother of all things.
! @-
  Therefore let there always be non-being,
***************
*** 11 ****
--- 10,13 ----
    they have different names.
+ They both may be called deep and profound.
+ Deeper and more profound,
+ The door of all subtleties!
@end example

@node Detailed Context
@subsubsection Detailed Description of Context Format

The context output format starts with a two-line header, which looks
like this:

@example
*** @var{from-file} @var{from-file-modification-time}
--- @var{to-file} @var{to-file-modification time}
@end example

@noindent
@vindex LC_TIME
@cindex time stamp format, context diffs
The time stamp normally looks like @samp{2002-02-21 23:30:39.942229878
-0800} to indicate the date, time with fractional seconds, and time
zone in @uref{ftp://ftp.isi.edu/in-notes/rfc2822.txt, Internet RFC
2822 format}.  (The fractional seconds are omitted on hosts that do
not support fractional time stamps.)  However, a traditional time
stamp like @samp{Thu Feb 21 23:30:39 2002} is used if the
@env{LC_TIME} locale category is either @samp{C} or @samp{POSIX}.

You can change the header's content with the
@option{--label=@var{label}} option; see @ref{Alternate Names}.

Next come one or more hunks of differences; each hunk shows one area
where the files differ.  Context format hunks look like this:

@example
***************
*** @var{from-file-line-numbers} ****
  @var{from-file-line}
  @var{from-file-line}@dots{}
--- @var{to-file-line-numbers} ----
  @var{to-file-line}
  @var{to-file-line}@dots{}
@end example

If a hunk contains two or more lines, its line numbers look like
@samp{@var{start},@var{end}}.  Otherwise only its end line number
appears.  An empty hunk is considered to end at the line that precedes
the hunk.

The lines of context around the lines that differ start with two space
characters.  The lines that differ between the two files start with one
of the following indicator characters, followed by a space character:

@table @samp
@item !
A line that is part of a group of one or more lines that changed between
the two files.  There is a corresponding group of lines marked with
@samp{!} in the part of this hunk for the other file.

@item +
An ``inserted'' line in the second file that corresponds to nothing in
the first file.

@item -
A ``deleted'' line in the first file that corresponds to nothing in the
second file.
@end table

If all of the changes in a hunk are insertions, the lines of
@var{from-file} are omitted.  If all of the changes are deletions, the
lines of @var{to-file} are omitted.

@node Unified Format
@subsection Unified Format
@cindex unified output format
@cindex @samp{+-} output format

The unified output format is a variation on the context format that is
more compact because it omits redundant context lines.  To select this
output format, use the @option{-U @var{lines}},
@option{--unified@r{[}=@var{lines}@r{]}}, or @option{-u}
option.  The argument @var{lines} is the number of lines of context to
show.  When it is not given, it defaults to three.

At present, only @acronym{GNU} @command{diff} can produce this format and
only @acronym{GNU} @command{patch} can automatically apply diffs in this
format.  For proper operation, @command{patch} typically needs at
least three lines of context.

@menu
* Example Unified::  Sample output in unified format.
* Detailed Unified:: A detailed description of unified format.
@end menu

@node Example Unified
@subsubsection An Example of Unified Format

Here is the output of the command @samp{diff -u lao tzu}
(@pxref{Sample diff Input}, for the complete contents of the two files):

@example
--- lao	2002-02-21 23:30:39.942229878 -0800
+++ tzu	2002-02-21 23:30:50.442260588 -0800
@@@@ -1,7 +1,6 @@@@
-The Way that can be told of is not the eternal Way;
-The name that can be named is not the eternal name.
 The Nameless is the origin of Heaven and Earth;
-The Named is the mother of all things.
+The named is the mother of all things.
+
 Therefore let there always be non-being,
   so we may see their subtlety,
 And let there always be being,
@@@@ -9,3 +8,6 @@@@
 The two are the same,
 But after they are produced,
   they have different names.
+They both may be called deep and profound.
+Deeper and more profound,
+The door of all subtleties!
@end example

@node Detailed Unified
@subsubsection Detailed Description of Unified Format

The unified output format starts with a two-line header, which looks
like this:

@example
--- @var{from-file} @var{from-file-modification-time}
+++ @var{to-file} @var{to-file-modification-time}
@end example

@noindent
@cindex time stamp format, unified diffs
The time stamp looks like @samp{2002-02-21 23:30:39.942229878 -0800}
to indicate the date, time with fractional seconds, and time zone.
The fractional seconds are omitted on hosts that do not support
fractional time stamps.

You can change the header's content with the
@option{--label=@var{label}} option; see @xref{Alternate Names}.

Next come one or more hunks of differences; each hunk shows one area
where the files differ.  Unified format hunks look like this:

@example
@@@@ @var{from-file-line-numbers} @var{to-file-line-numbers} @@@@
 @var{line-from-either-file}
 @var{line-from-either-file}@dots{}
@end example

If a hunk contains just one line, only its start line number appears.
Otherwise its line numbers look like @samp{@var{start},@var{count}}.
An empty hunk is considered to start at the line that follows the hunk.

If a hunk and its context contain two or more lines, its
line numbers look like @samp{@var{start},@var{count}}.  Otherwise only
its end line number appears.  An empty hunk is considered to end at
the line that precedes the hunk.

The lines common to both files begin with a space character.  The lines
that actually differ between the two files have one of the following
indicator characters in the left print column:

@table @samp
@item +
A line was added here to the first file.

@item -
A line was removed here from the first file.
@end table

@node Sections
@subsection Showing Which Sections Differences Are in
@cindex headings
@cindex section headings

Sometimes you might want to know which part of the files each change
falls in.  If the files are source code, this could mean which
function was changed.  If the files are documents, it could mean which
chapter or appendix was changed.  @acronym{GNU} @command{diff} can
show this by displaying the nearest section heading line that precedes
the differing lines.  Which lines are ``section headings'' is
determined by a regular expression.

@menu
* Specified Headings::  Showing headings that match regular expressions.
* C Function Headings:: Showing headings of C functions.
@end menu

@node Specified Headings
@subsubsection Showing Lines That Match Regular Expressions
@cindex specified headings
@cindex regular expression matching headings

To show in which sections differences occur for files that are not
source code for C or similar languages, use the @option{-F @var{regexp}}
or @option{--show-function-line=@var{regexp}} option.  @command{diff}
considers lines that match the @command{grep}-style regular expression
@var{regexp} to be the beginning
of a section of the file.  Here are suggested regular expressions for
some common languages:

@c Please add to this list, e.g. Fortran, Pascal, Perl, Python.
@table @samp
@item ^[[:alpha:]$_]
C, C++, Prolog
@item ^(
Lisp
@item ^@@node
Texinfo
@end table

This option does not automatically select an output format; in order to
use it, you must select the context format (@pxref{Context Format}) or
unified format (@pxref{Unified Format}).  In other output formats it
has no effect.

The @option{-F} or @option{--show-function-line} option finds the nearest
unchanged line that precedes each hunk of differences and matches the
given regular expression.  Then it adds that line to the end of the
line of asterisks in the context format, or to the @samp{@@@@} line in
unified format.  If no matching line exists, this option leaves the output for
that hunk unchanged.  If that line is more than 40 characters long, it
outputs only the first 40 characters.  You can specify more than one
regular expression for such lines; @command{diff} tries to match each line
against each regular expression, starting with the last one given.  This
means that you can use @option{-p} and @option{-F} together, if you wish.

@node C Function Headings
@subsubsection Showing C Function Headings
@cindex C function headings
@cindex function headings, C

To show in which functions differences occur for C and similar
languages, you can use the @option{-p} or @option{--show-c-function} option.
This option automatically defaults to the context output format
(@pxref{Context Format}), with the default number of lines of context.
You can override that number with @option{-C @var{lines}} elsewhere in the
command line.  You can override both the format and the number with
@option{-U @var{lines}} elsewhere in the command line.

The @option{-p} or @option{--show-c-function} option is equivalent to
@option{-F '^[[:alpha:]$_]'} if the unified format is specified, otherwise
@option{-c -F '^[[:alpha:]$_]'} (@pxref{Specified Headings}).  @acronym{GNU}
@command{diff} provides this option for the sake of convenience.

@node Alternate Names
@subsection Showing Alternate File Names
@cindex alternate file names
@cindex file name alternates

If you are comparing two files that have meaningless or uninformative
names, you might want @command{diff} to show alternate names in the header
of the context and unified output formats.  To do this, use the
@option{--label=@var{label}} option.  The first time
you give this option, its argument replaces the name and date of the
first file in the header; the second time, its argument replaces the
name and date of the second file.  If you give this option more than
twice, @command{diff} reports an error.  The @option{--label} option does not
affect the file names in the @command{pr} header when the @option{-l} or
@option{--paginate} option is used (@pxref{Pagination}).

Here are the first two lines of the output from @samp{diff -C 2
--label=original --label=modified lao tzu}:

@example
*** original
--- modified
@end example

@node Side by Side
@section Showing Differences Side by Side
@cindex side by side
@cindex two-column output
@cindex columnar output

@command{diff} can produce a side by side difference listing of two files.
The files are listed in two columns with a gutter between them.  The
gutter contains one of the following markers:

@table @asis
@item white space
The corresponding lines are in common.  That is, either the lines are
identical, or the difference is ignored because of one of the
@option{--ignore} options (@pxref{White Space}).

@item @samp{|}
The corresponding lines differ, and they are either both complete
or both incomplete.

@item @samp{<}
The files differ and only the first file contains the line.

@item @samp{>}
The files differ and only the second file contains the line.

@item @samp{(}
Only the first file contains the line, but the difference is ignored.

@item @samp{)}
Only the second file contains the line, but the difference is ignored.

@item @samp{\}
The corresponding lines differ, and only the first line is incomplete.

@item @samp{/}
The corresponding lines differ, and only the second line is incomplete.
@end table

Normally, an output line is incomplete if and only if the lines that it
contains are incomplete; @xref{Incomplete Lines}.  However, when an
output line represents two differing lines, one might be incomplete
while the other is not.  In this case, the output line is complete,
but its the gutter is marked @samp{\} if the first line is incomplete,
@samp{/} if the second line is.

Side by side format is sometimes easiest to read, but it has limitations.
It generates much wider output than usual, and truncates lines that are
too long to fit.  Also, it relies on lining up output more heavily than
usual, so its output looks particularly bad if you use varying
width fonts, nonstandard tab stops, or nonprinting characters.

You can use the @command{sdiff} command to interactively merge side by side
differences.  @xref{Interactive Merging}, for more information on merging files.

@menu
* Side by Side Format::  Controlling side by side output format.
* Example Side by Side:: Sample side by side output.
@end menu

@node Side by Side Format
@subsection Controlling Side by Side Format
@cindex side by side format

The @option{-y} or @option{--side-by-side} option selects side by side
format.  Because side by side output lines contain two input lines, the
output is wider than usual: normally 130 print columns, which can fit
onto a traditional printer line.  You can set the width of the output
with the @option{-W @var{columns}} or @option{--width=@var{columns}}
option.  The output is split into two halves of equal width, separated by a
small gutter to mark differences; the right half is aligned to a tab
stop so that tabs line up.  Input lines that are too long to fit in half
of an output line are truncated for output.

The @option{--left-column} option prints only the left column of two
common lines.  The @option{--suppress-common-lines} option suppresses
common lines entirely.

@node Example Side by Side
@subsection An Example of Side by Side Format

Here is the output of the command @samp{diff -y -W 72 lao tzu}
(@pxref{Sample diff Input}, for the complete contents of the two files).

@example
The Way that can be told of is n   <
The name that can be named is no   <
The Nameless is the origin of He        The Nameless is the origin of He
The Named is the mother of all t   |    The named is the mother of all t
                                   >
Therefore let there always be no        Therefore let there always be no
  so we may see their subtlety,           so we may see their subtlety,
And let there always be being,          And let there always be being,
  so we may see their outcome.            so we may see their outcome.
The two are the same,                   The two are the same,
But after they are produced,            But after they are produced,
  they have different names.              they have different names.
                                   >    They both may be called deep and
                                   >    Deeper and more profound,
                                   >    The door of all subtleties!
@end example

@node Normal
@section Showing Differences Without Context
@cindex normal output format
@cindex @samp{<} output format

The ``normal'' @command{diff} output format shows each hunk of differences
without any surrounding context.  Sometimes such output is the clearest
way to see how lines have changed, without the clutter of nearby
unchanged lines (although you can get similar results with the context
or unified formats by using 0 lines of context).  However, this format
is no longer widely used for sending out patches; for that purpose, the
context format (@pxref{Context Format}) and the unified format
(@pxref{Unified Format}) are superior.  Normal format is the default for
compatibility with older versions of @command{diff} and the @acronym{POSIX}
standard.  Use the @option{--normal} option to select this output
format explicitly.

@menu
* Example Normal::  Sample output in the normal format.
* Detailed Normal:: A detailed description of normal output format.
@end menu

@node Example Normal
@subsection An Example of Normal Format

Here is the output of the command @samp{diff lao tzu}
(@pxref{Sample diff Input}, for the complete contents of the two files).
Notice that it shows only the lines that are different between the two
files.

@example
1,2d0
< The Way that can be told of is not the eternal Way;
< The name that can be named is not the eternal name.
4c2,3
< The Named is the mother of all things.
---
> The named is the mother of all things.
> @-
11a11,13
> They both may be called deep and profound.
> Deeper and more profound,
> The door of all subtleties!
@end example

@node Detailed Normal
@subsection Detailed Description of Normal Format

The normal output format consists of one or more hunks of differences;
each hunk shows one area where the files differ.  Normal format hunks
look like this:

@example
@var{change-command}
< @var{from-file-line}
< @var{from-file-line}@dots{}
---
> @var{to-file-line}
> @var{to-file-line}@dots{}
@end example

There are three types of change commands.  Each consists of a line
number or comma-separated range of lines in the first file, a single
character indicating the kind of change to make, and a line number or
comma-separated range of lines in the second file.  All line numbers are
the original line numbers in each file.  The types of change commands
are:

@table @samp
@item @var{l}a@var{r}
Add the lines in range @var{r} of the second file after line @var{l} of
the first file.  For example, @samp{8a12,15} means append lines 12--15
of file 2 after line 8 of file 1; or, if changing file 2 into file 1,
delete lines 12--15 of file 2.

@item @var{f}c@var{t}
Replace the lines in range @var{f} of the first file with lines in range
@var{t} of the second file.  This is like a combined add and delete, but
more compact.  For example, @samp{5,7c8,10} means change lines 5--7 of
file 1 to read as lines 8--10 of file 2; or, if changing file 2 into
file 1, change lines 8--10 of file 2 to read as lines 5--7 of file 1.

@item @var{r}d@var{l}
Delete the lines in range @var{r} from the first file; line @var{l} is where
they would have appeared in the second file had they not been deleted.
For example, @samp{5,7d3} means delete lines 5--7 of file 1; or, if
changing file 2 into file 1, append lines 5--7 of file 1 after line 3 of
file 2.
@end table

@node Scripts
@section Making Edit Scripts
@cindex script output formats

Several output modes produce command scripts for editing @var{from-file}
to produce @var{to-file}.

@menu
* ed Scripts:: Using @command{diff} to produce commands for @command{ed}.
* Forward ed:: Making forward @command{ed} scripts.
* RCS::        A special @command{diff} output format used by @acronym{RCS}.
@end menu

@node ed Scripts
@subsection @command{ed} Scripts
@cindex @command{ed} script output format

@command{diff} can produce commands that direct the @command{ed} text editor
to change the first file into the second file.  Long ago, this was the
only output mode that was suitable for editing one file into another
automatically; today, with @command{patch}, it is almost obsolete.  Use the
@option{-e} or @option{--ed} option to select this output format.

Like the normal format (@pxref{Normal}), this output format does not
show any context; unlike the normal format, it does not include the
information necessary to apply the diff in reverse (to produce the first
file if all you have is the second file and the diff).

If the file @file{d} contains the output of @samp{diff -e old new}, then
the command @samp{(cat d && echo w) | ed - old} edits @file{old} to make
it a copy of @file{new}.  More generally, if @file{d1}, @file{d2},
@dots{}, @file{dN} contain the outputs of @samp{diff -e old new1},
@samp{diff -e new1 new2}, @dots{}, @samp{diff -e newN-1 newN},
respectively, then the command @samp{(cat d1 d2 @dots{} dN && echo w) |
ed - old} edits @file{old} to make it a copy of @file{newN}.

@menu
* Example ed::  A sample @command{ed} script.
* Detailed ed:: A detailed description of @command{ed} format.
@end menu

@node Example ed
@subsubsection Example @command{ed} Script

Here is the output of @samp{diff -e lao tzu} (@pxref{Sample
diff Input}, for the complete contents of the two files):

@example
11a
They both may be called deep and profound.
Deeper and more profound,
The door of all subtleties!
.
4c
The named is the mother of all things.

.
1,2d
@end example

@node Detailed ed
@subsubsection Detailed Description of @command{ed} Format

The @command{ed} output format consists of one or more hunks of
differences.  The changes closest to the ends of the files come first so
that commands that change the number of lines do not affect how
@command{ed} interprets line numbers in succeeding commands.  @command{ed}
format hunks look like this:

@example
@var{change-command}
@var{to-file-line}
@var{to-file-line}@dots{}
.
@end example

Because @command{ed} uses a single period on a line to indicate the
end of input, @acronym{GNU} @command{diff} protects lines of changes
that contain a single period on a line by writing two periods instead,
then writing a subsequent @command{ed} command to change the two
periods into one.  The @command{ed} format cannot represent an
incomplete line, so if the second file ends in a changed incomplete
line, @command{diff} reports an error and then pretends that a newline
was appended.

There are three types of change commands.  Each consists of a line
number or comma-separated range of lines in the first file and a single
character indicating the kind of change to make.  All line numbers are
the original line numbers in the file.  The types of change commands
are:

@table @samp
@item @var{l}a
Add text from the second file after line @var{l} in the first file.  For
example, @samp{8a} means to add the following lines after line 8 of file
1.

@item @var{r}c
Replace the lines in range @var{r} in the first file with the following
lines.  Like a combined add and delete, but more compact.  For example,
@samp{5,7c} means change lines 5--7 of file 1 to read as the text file
2.

@item @var{r}d
Delete the lines in range @var{r} from the first file.  For example,
@samp{5,7d} means delete lines 5--7 of file 1.
@end table

@node Forward ed
@subsection Forward @command{ed} Scripts
@cindex forward @command{ed} script output format

@command{diff} can produce output that is like an @command{ed} script, but
with hunks in forward (front to back) order.  The format of the commands
is also changed slightly: command characters precede the lines they
modify, spaces separate line numbers in ranges, and no attempt is made
to disambiguate hunk lines consisting of a single period.  Like
@command{ed} format, forward @command{ed} format cannot represent incomplete
lines.

Forward @command{ed} format is not very useful, because neither @command{ed}
nor @command{patch} can apply diffs in this format.  It exists mainly for
compatibility with older versions of @command{diff}.  Use the @option{-f} or
@option{--forward-ed} option to select it.

@node RCS
@subsection @acronym{RCS} Scripts
@cindex @acronym{RCS} script output format

The @acronym{RCS} output format is designed specifically for use by
the Revision Control System, which is a set of free programs used for
organizing different versions and systems of files.  Use the
@option{-n} or @option{--rcs} option to select this output format.  It
is like the forward @command{ed} format (@pxref{Forward ed}), but it
can represent arbitrary changes to the contents of a file because it
avoids the forward @command{ed} format's problems with lines
consisting of a single period and with incomplete lines.  Instead of
ending text sections with a line consisting of a single period, each
command specifies the number of lines it affects; a combination of the
@samp{a} and @samp{d} commands are used instead of @samp{c}.  Also, if
the second file ends in a changed incomplete line, then the output
also ends in an incomplete line.

Here is the output of @samp{diff -n lao tzu} (@pxref{Sample
diff Input}, for the complete contents of the two files):

@example
d1 2
d4 1
a4 2
The named is the mother of all things.

a11 3
They both may be called deep and profound.
Deeper and more profound,
The door of all subtleties!
@end example

@node If-then-else
@section Merging Files with If-then-else
@cindex merged output format
@cindex if-then-else output format
@cindex C if-then-else output format
@cindex @command{ifdef} output format

You can use @command{diff} to merge two files of C source code.  The output
of @command{diff} in this format contains all the lines of both files.
Lines common to both files are output just once; the differing parts are
separated by the C preprocessor directives @code{#ifdef @var{name}} or
@code{#ifndef @var{name}}, @code{#else}, and @code{#endif}.  When
compiling the output, you select which version to use by either defining
or leaving undefined the macro @var{name}.

To merge two files, use @command{diff} with the @option{-D @var{name}} or
@option{--ifdef=@var{name}} option.  The argument @var{name} is the C
preprocessor identifier to use in the @code{#ifdef} and @code{#ifndef}
directives.

For example, if you change an instance of @code{wait (&s)} to
@code{waitpid (-1, &s, 0)} and then merge the old and new files with
the @option{--ifdef=HAVE_WAITPID} option, then the affected part of your code
might look like this:

@example
    do @{
#ifndef HAVE_WAITPID
        if ((w = wait (&s)) < 0  &&  errno != EINTR)
#else /* HAVE_WAITPID */
        if ((w = waitpid (-1, &s, 0)) < 0  &&  errno != EINTR)
#endif /* HAVE_WAITPID */
            return w;
    @} while (w != child);
@end example

You can specify formats for languages other than C by using line group
formats and line formats, as described in the next sections.

@menu
* Line Group Formats::    Formats for general if-then-else line groups.
* Line Formats::          Formats for each line in a line group.
* Example If-then-else::  Sample if-then-else format output.
* Detailed If-then-else:: A detailed description of if-then-else format.
@end menu

@node Line Group Formats
@subsection Line Group Formats
@cindex line group formats
@cindex formats for if-then-else line groups

Line group formats let you specify formats suitable for many
applications that allow if-then-else input, including programming
languages and text formatting languages.  A line group format specifies
the output format for a contiguous group of similar lines.

For example, the following command compares the TeX files @file{old}
and @file{new}, and outputs a merged file in which old regions are
surrounded by @samp{\begin@{em@}}-@samp{\end@{em@}} lines, and new
regions are surrounded by @samp{\begin@{bf@}}-@samp{\end@{bf@}} lines.

@example
diff \
   --old-group-format='\begin@{em@}
%<\end@{em@}
' \
   --new-group-format='\begin@{bf@}
%>\end@{bf@}
' \
   old new
@end example

The following command is equivalent to the above example, but it is a
little more verbose, because it spells out the default line group formats.

@example
diff \
   --old-group-format='\begin@{em@}
%<\end@{em@}
' \
   --new-group-format='\begin@{bf@}
%>\end@{bf@}
' \
   --unchanged-group-format='%=' \
   --changed-group-format='\begin@{em@}
%<\end@{em@}
\begin@{bf@}
%>\end@{bf@}
' \
   old new
@end example

Here is a more advanced example, which outputs a diff listing with
headers containing line numbers in a ``plain English'' style.

@example
diff \
   --unchanged-group-format='' \
   --old-group-format='-------- %dn line%(n=1?:s) deleted at %df:
%<' \
   --new-group-format='-------- %dN line%(N=1?:s) added after %de:
%>' \
   --changed-group-format='-------- %dn line%(n=1?:s) changed at %df:
%<-------- to:
%>' \
   old new
@end example

To specify a line group format, use @command{diff} with one of the options
listed below.  You can specify up to four line group formats, one for
each kind of line group.  You should quote @var{format}, because it
typically contains shell metacharacters.

@table @option
@item --old-group-format=@var{format}
These line groups are hunks containing only lines from the first file.
The default old group format is the same as the changed group format if
it is specified; otherwise it is a format that outputs the line group as-is.

@item --new-group-format=@var{format}
These line groups are hunks containing only lines from the second
file.  The default new group format is same as the changed group
format if it is specified; otherwise it is a format that outputs the
line group as-is.

@item --changed-group-format=@var{format}
These line groups are hunks containing lines from both files.  The
default changed group format is the concatenation of the old and new
group formats.

@item --unchanged-group-format=@var{format}
These line groups contain lines common to both files.  The default
unchanged group format is a format that outputs the line group as-is.
@end table

In a line group format, ordinary characters represent themselves;
conversion specifications start with @samp{%} and have one of the
following forms.

@table @samp
@item %<
stands for the lines from the first file, including the trailing newline.
Each line is formatted according to the old line format (@pxref{Line Formats}).

@item %>
stands for the lines from the second file, including the trailing newline.
Each line is formatted according to the new line format.

@item %=
stands for the lines common to both files, including the trailing newline.
Each line is formatted according to the unchanged line format.

@item %%
stands for @samp{%}.

@item %c'@var{C}'
where @var{C} is a single character, stands for @var{C}.
@var{C} may not be a backslash or an apostrophe.
For example, @samp{%c':'} stands for a colon, even inside
the then-part of an if-then-else format, which a colon would
normally terminate.

@item %c'\@var{O}'
where @var{O} is a string of 1, 2, or 3 octal digits,
stands for the character with octal code @var{O}.
For example, @samp{%c'\0'} stands for a null character.

@item @var{F}@var{n}
where @var{F} is a @code{printf} conversion specification and @var{n} is one
of the following letters, stands for @var{n}'s value formatted with @var{F}.

@table @samp
@item e
The line number of the line just before the group in the old file.

@item f
The line number of the first line in the group in the old file;
equals @var{e} + 1.

@item l
The line number of the last line in the group in the old file.

@item m
The line number of the line just after the group in the old file;
equals @var{l} + 1.

@item n
The number of lines in the group in the old file; equals @var{l} - @var{f} + 1.

@item E, F, L, M, N
Likewise, for lines in the new file.

@end table

@vindex LC_NUMERIC
The @code{printf} conversion specification can be @samp{%d},
@samp{%o}, @samp{%x}, or @samp{%X}, specifying decimal, octal,
lower case hexadecimal, or upper case hexadecimal output
respectively.  After the @samp{%} the following options can appear in
sequence: a series of zero or more flags; an integer
specifying the minimum field width; and a period followed by an
optional integer specifying the minimum number of digits.
The flags are @samp{-} for left-justification, @samp{'} for separating
the digit into groups as specified by the @env{LC_NUMERIC} locale category,
and @samp{0} for padding with zeros instead of spaces.
For example, @samp{%5dN} prints the number of new lines in the group
in a field of width 5 characters, using the @code{printf} format @code{"%5d"}.

@item (@var{A}=@var{B}?@var{T}:@var{E})
If @var{A} equals @var{B} then @var{T} else @var{E}.
@var{A} and @var{B} are each either a decimal constant
or a single letter interpreted as above.
This format spec is equivalent to @var{T} if
@var{A}'s value equals @var{B}'s; otherwise it is equivalent to @var{E}.

For example, @samp{%(N=0?no:%dN) line%(N=1?:s)} is equivalent to
@samp{no lines} if @var{N} (the number of lines in the group in the
new file) is 0, to @samp{1 line} if @var{N} is 1, and to @samp{%dN lines}
otherwise.
@end table

@node Line Formats
@subsection Line Formats
@cindex line formats

Line formats control how each line taken from an input file is
output as part of a line group in if-then-else format.

For example, the following command outputs text with a one-character
change indicator to the left of the text.  The first character of output
is @samp{-} for deleted lines, @samp{|} for added lines, and a space for
unchanged lines.  The formats contain newline characters where newlines
are desired on output.

@example
diff \
   --old-line-format='-%l
' \
   --new-line-format='|%l
' \
   --unchanged-line-format=' %l
' \
   old new
@end example

To specify a line format, use one of the following options.  You should
quote @var{format}, since it often contains shell metacharacters.

@table @option
@item --old-line-format=@var{format}
formats lines just from the first file.

@item --new-line-format=@var{format}
formats lines just from the second file.

@item --unchanged-line-format=@var{format}
formats lines common to both files.

@item --line-format=@var{format}
formats all lines; in effect, it sets all three above options simultaneously.
@end table

In a line format, ordinary characters represent themselves;
conversion specifications start with @samp{%} and have one of the
following forms.

@table @samp
@item %l
stands for the contents of the line, not counting its trailing
newline (if any).  This format ignores whether the line is incomplete;
@xref{Incomplete Lines}.

@item %L
stands for the contents of the line, including its trailing newline
(if any).  If a line is incomplete, this format preserves its
incompleteness.

@item %%
stands for @samp{%}.

@item %c'@var{C}'
where @var{C} is a single character, stands for @var{C}.
@var{C} may not be a backslash or an apostrophe.
For example, @samp{%c':'} stands for a colon.

@item %c'\@var{O}'
where @var{O} is a string of 1, 2, or 3 octal digits,
stands for the character with octal code @var{O}.
For example, @samp{%c'\0'} stands for a null character.

@item @var{F}n
where @var{F} is a @code{printf} conversion specification,
stands for the line number formatted with @var{F}.
For example, @samp{%.5dn} prints the line number using the
@code{printf} format @code{"%.5d"}.  @xref{Line Group Formats}, for
more about printf conversion specifications.

@end table

The default line format is @samp{%l} followed by a newline character.

If the input contains tab characters and it is important that they line
up on output, you should ensure that @samp{%l} or @samp{%L} in a line
format is just after a tab stop (e.g.@: by preceding @samp{%l} or
@samp{%L} with a tab character), or you should use the @option{-t} or
@option{--expand-tabs} option.

Taken together, the line and line group formats let you specify many
different formats.  For example, the following command uses a format
similar to normal @command{diff} format.  You can tailor this command
to get fine control over @command{diff} output.

@example
diff \
   --old-line-format='< %l
' \
   --new-line-format='> %l
' \
   --old-group-format='%df%(f=l?:,%dl)d%dE
%<' \
   --new-group-format='%dea%dF%(F=L?:,%dL)
%>' \
   --changed-group-format='%df%(f=l?:,%dl)c%dF%(F=L?:,%dL)
%<---
%>' \
   --unchanged-group-format='' \
   old new
@end example

@node Example If-then-else
@subsection An Example of If-then-else Format

Here is the output of @samp{diff -DTWO lao tzu} (@pxref{Sample
diff Input}, for the complete contents of the two files):

@example
#ifndef TWO
The Way that can be told of is not the eternal Way;
The name that can be named is not the eternal name.
#endif /* ! TWO */
The Nameless is the origin of Heaven and Earth;
#ifndef TWO
The Named is the mother of all things.
#else /* TWO */
The named is the mother of all things.

#endif /* TWO */
Therefore let there always be non-being,
  so we may see their subtlety,
And let there always be being,
  so we may see their outcome.
The two are the same,
But after they are produced,
  they have different names.
#ifdef TWO
They both may be called deep and profound.
Deeper and more profound,
The door of all subtleties!
#endif /* TWO */
@end example

@node Detailed If-then-else
@subsection Detailed Description of If-then-else Format

For lines common to both files, @command{diff} uses the unchanged line
group format.  For each hunk of differences in the merged output
format, if the hunk contains only lines from the first file,
@command{diff} uses the old line group format; if the hunk contains only
lines from the second file, @command{diff} uses the new group format;
otherwise, @command{diff} uses the changed group format.

The old, new, and unchanged line formats specify the output format of
lines from the first file, lines from the second file, and lines common
to both files, respectively.

The option @option{--ifdef=@var{name}} is equivalent to
the following sequence of options using shell syntax:

@example
--old-group-format='#ifndef @var{name}
%<#endif /* ! @var{name} */
' \
--new-group-format='#ifdef @var{name}
%>#endif /* @var{name} */
' \
--unchanged-group-format='%=' \
--changed-group-format='#ifndef @var{name}
%<#else /* @var{name} */
%>#endif /* @var{name} */
'
@end example

You should carefully check the @command{diff} output for proper nesting.
For example, when using the @option{-D @var{name}} or
@option{--ifdef=@var{name}} option, you should check that if the
differing lines contain any of the C preprocessor directives
@samp{#ifdef}, @samp{#ifndef}, @samp{#else}, @samp{#elif}, or
@samp{#endif}, they are nested properly and match.  If they don't, you
must make corrections manually.  It is a good idea to carefully check
the resulting code anyway to make sure that it really does what you
want it to; depending on how the input files were produced, the output
might contain duplicate or otherwise incorrect code.

The @command{patch} @option{-D @var{name}} option behaves like
the @command{diff} @option{-D @var{name}} option, except it operates on
a file and a diff to produce a merged file; @xref{patch Options}.

@node Incomplete Lines
@chapter Incomplete Lines
@cindex incomplete lines
@cindex full lines
@cindex newline treatment by @command{diff}

When an input file ends in a non-newline character, its last line is
called an @dfn{incomplete line} because its last character is not a
newline.  All other lines are called @dfn{full lines} and end in a
newline character.  Incomplete lines do not match full lines unless
differences in white space are ignored (@pxref{White Space}).

An incomplete line is normally distinguished on output from a full
line by a following line that starts with @samp{\}.  However, the
@acronym{RCS} format (@pxref{RCS}) outputs the incomplete line as-is,
without any trailing newline or following line.  The side by side
format normally represents incomplete lines as-is, but in some cases
uses a @samp{\} or @samp{/} gutter marker; @xref{Side by Side}.  The
if-then-else line format preserves a line's incompleteness with
@samp{%L}, and discards the newline with @samp{%l}; @xref{Line
Formats}.  Finally, with the @command{ed} and forward @command{ed}
output formats (@pxref{Output Formats}) @command{diff} cannot
represent an incomplete line, so it pretends there was a newline and
reports an error.

For example, suppose @file{F} and @file{G} are one-byte files that
contain just @samp{f} and @samp{g}, respectively.  Then @samp{diff F G}
outputs

@example
1c1
< f
\ No newline at end of file
---
> g
\ No newline at end of file
@end example

@noindent
(The exact message may differ in non-English locales.)
@samp{diff -n F G} outputs the following without a trailing newline:

@example
d1 1
a1 1
g
@end example

@noindent
@samp{diff -e F G} reports two errors and outputs the following:

@example
1c
g
.
@end example

@node Comparing Directories
@chapter Comparing Directories

@vindex LC_COLLATE
You can use @command{diff} to compare some or all of the files in two
directory trees.  When both file name arguments to @command{diff} are
directories, it compares each file that is contained in both
directories, examining file names in alphabetical order as specified by
the @env{LC_COLLATE} locale category.  Normally
@command{diff} is silent about pairs of files that contain no differences,
but if you use the @option{-s} or @option{--report-identical-files} option,
it reports pairs of identical files.  Normally @command{diff} reports
subdirectories common to both directories without comparing
subdirectories' files, but if you use the @option{-r} or
@option{--recursive} option, it compares every corresponding pair of files
in the directory trees, as many levels deep as they go.

For file names that are in only one of the directories, @command{diff}
normally does not show the contents of the file that exists; it reports
only that the file exists in that directory and not in the other.  You
can make @command{diff} act as though the file existed but was empty in the
other directory, so that it outputs the entire contents of the file that
actually exists.  (It is output as either an insertion or a
deletion, depending on whether it is in the first or the second
directory given.)  To do this, use the @option{-N} or @option{--new-file}
option.

If the older directory contains one or more large files that are not in
the newer directory, you can make the patch smaller by using the
@option{--unidirectional-new-file} option instead of @option{-N}.
This option is like @option{-N} except that it only inserts the contents
of files that appear in the second directory but not the first (that is,
files that were added).  At the top of the patch, write instructions for
the user applying the patch to remove the files that were deleted before
applying the patch.  @xref{Making Patches}, for more discussion of
making patches for distribution.

To ignore some files while comparing directories, use the @option{-x
@var{pattern}} or @option{--exclude=@var{pattern}} option.  This option
ignores any files or subdirectories whose base names match the shell
pattern @var{pattern}.  Unlike in the shell, a period at the start of
the base of a file name matches a wildcard at the start of a pattern.
You should enclose @var{pattern} in quotes so that the shell does not
expand it.  For example, the option @option{-x '*.[ao]'} ignores any file
whose name ends with @samp{.a} or @samp{.o}.

This option accumulates if you specify it more than once.  For example,
using the options @option{-x 'RCS' -x '*,v'} ignores any file or
subdirectory whose base name is @samp{RCS} or ends with @samp{,v}.

If you need to give this option many times, you can instead put the
patterns in a file, one pattern per line, and use the @option{-X
@var{file}} or @option{--exclude-from=@var{file}} option.  Trailing
white space and empty lines are ignored in the pattern file.

If you have been comparing two directories and stopped partway through,
later you might want to continue where you left off.  You can do this by
using the @option{-S @var{file}} or @option{--starting-file=@var{file}}
option.  This compares only the file @var{file} and all alphabetically
later files in the topmost directory level.

If two directories differ only in that file names are lower case in
one directory and upper case in the upper, @command{diff} normally
reports many differences because it compares file names in a
case sensitive way.  With the @option{--ignore-file-name-case} option,
@command{diff} ignores case differences in file names, so that for example
the contents of the file @file{Tao} in one directory are compared to
the contents of the file @file{TAO} in the other.  The
@option{--no-ignore-file-name-case} option cancels the effect of the
@option{--ignore-file-name-case} option, reverting to the default
behavior.

If an @option{-x @var{pattern}} or @option{--exclude=@var{pattern}}
option, or an @option{-X @var{file}} or
@option{--exclude-from=@var{file}} option,
is specified while the @option{--ignore-file-name-case} option is in
effect, case is ignored when excluding file names matching the
specified patterns.

@node Adjusting Output
@chapter Making @command{diff} Output Prettier

@command{diff} provides several ways to adjust the appearance of its output.
These adjustments can be applied to any output format.

@menu
* Tabs::            Preserving the alignment of tab stops.
* Trailing Blanks:: Suppressing blanks before empty output lines.
* Pagination::      Page numbering and time-stamping @command{diff} output.
@end menu

@node Tabs
@section Preserving Tab Stop Alignment
@cindex tab stop alignment
@cindex aligning tab stops

The lines of text in some of the @command{diff} output formats are
preceded by one or two characters that indicate whether the text is
inserted, deleted, or changed.  The addition of those characters can
cause tabs to move to the next tab stop, throwing off the alignment of
columns in the line.  @acronym{GNU} @command{diff} provides two ways
to make tab-aligned columns line up correctly.

The first way is to have @command{diff} convert all tabs into the correct
number of spaces before outputting them; select this method with the
@option{-t} or @option{--expand-tabs} option.  To use this form of output with
@command{patch}, you must give @command{patch} the @option{-l} or
@option{--ignore-white-space} option (@pxref{Changed White Space}, for more
information).  @command{diff} normally assumes that tab stops are set
every 8 print columns, but this can be altered by the
@option{--tabsize=@var{columns}} option.

The other method for making tabs line up correctly is to add a tab
character instead of a space after the indicator character at the
beginning of the line.  This ensures that all following tab characters
are in the same position relative to tab stops that they were in the
original files, so that the output is aligned correctly.  Its
disadvantage is that it can make long lines too long to fit on one line
of the screen or the paper.  It also does not work with the unified
output format, which does not have a space character after the change
type indicator character.  Select this method with the @option{-T} or
@option{--initial-tab} option.

@node Trailing Blanks
@section Omitting trailing blanks
@cindex trailing blanks
When outputting lines in normal or context format, or outputting an
unchanged line in unified format, @command{diff} normally outputs a
blank just before each line.  If the line is empty, the output of
@command{diff} therefore contains trailing blanks even though the
input does not contain them.  For example, when outputting an
unchanged empty line in context format, @command{diff} normally
outputs a line with two leading spaces.

Some text editors and email agents routinely delete trailing blanks,
so it can be a problem to deal with diff output files that contain
them.  You can avoid this problem with the
@option{--suppress-blank-empty} option.  It causes @command{diff} to
omit trailing blanks at the end of output lines in normal, context,
and unified format, unless the trailing blanks were already present in
the input.  This changes the output format slightly, so that output
lines are guaranteed to never end in a blank unless an input line ends
in a blank.  This format is less likely to be munged by text editors
or by transmission via email.  It is accepted by @acronym{GNU}
@command{patch} as well.

@node Pagination
@section Paginating @command{diff} Output
@cindex paginating @command{diff} output

It can be convenient to have long output page-numbered and time-stamped.
The @option{-l} or @option{--paginate} option does this by sending the
@command{diff} output through the @command{pr} program.  Here is what the page
header might look like for @samp{diff -lc lao tzu}:

@example
2002-02-22 14:20                 diff -lc lao tzu                 Page 1
@end example

@node diff Performance
@chapter @command{diff} Performance Tradeoffs
@cindex performance of @command{diff}

@acronym{GNU} @command{diff} runs quite efficiently; however, in some
circumstances you can cause it to run faster or produce a more compact
set of changes.

One way to improve @command{diff} performance is to use hard or
symbolic links to files instead of copies.  This improves performance
because @command{diff} normally does not need to read two hard or
symbolic links to the same file, since their contents must be
identical.  For example, suppose you copy a large directory hierarchy,
make a few changes to the copy, and then often use @samp{diff -r} to
compare the original to the copy.  If the original files are
read-only, you can greatly improve performance by creating the copy
using hard or symbolic links (e.g., with @acronym{GNU} @samp{cp -lR} or
@samp{cp -sR}).  Before editing a file in the copy for the first time,
you should break the link and replace it with a regular copy.

You can also affect the performance of @acronym{GNU} @command{diff} by
giving it options that change the way it compares files.
Performance has more than one dimension.  These options improve one
aspect of performance at the cost of another, or they improve
performance in some cases while hurting it in others.

The way that @acronym{GNU} @command{diff} determines which lines have
changed always comes up with a near-minimal set of differences.
Usually it is good enough for practical purposes.  If the
@command{diff} output is large, you might want @command{diff} to use a
modified algorithm that sometimes produces a smaller set of
differences.  The @option{-d} or @option{--minimal} option does this;
however, it can also cause @command{diff} to run more slowly than
usual, so it is not the default behavior.

When the files you are comparing are large and have small groups of
changes scattered throughout them, you can use the
@option{--speed-large-files} option to make a different modification to
the algorithm that @command{diff} uses.  If the input files have a constant
small density of changes, this option speeds up the comparisons without
changing the output.  If not, @command{diff} might produce a larger set of
differences; however, the output will still be correct.

Normally @command{diff} discards the prefix and suffix that is common to
both files before it attempts to find a minimal set of differences.
This makes @command{diff} run faster, but occasionally it may produce
non-minimal output.  The @option{--horizon-lines=@var{lines}} option
prevents @command{diff} from discarding the last @var{lines} lines of the
prefix and the first @var{lines} lines of the suffix.  This gives
@command{diff} further opportunities to find a minimal output.

Suppose a run of changed lines includes a sequence of lines at one end
and there is an identical sequence of lines just outside the other end.
The @command{diff} command is free to choose which identical sequence is
included in the hunk.  In this case, @command{diff} normally shifts the
hunk's boundaries when this merges adjacent hunks, or shifts a hunk's
lines towards the end of the file.  Merging hunks can make the output
look nicer in some cases.

@node Comparing Three Files
@chapter Comparing Three Files
@cindex comparing three files
@cindex format of @command{diff3} output

Use the program @command{diff3} to compare three files and show any
differences among them.  (@command{diff3} can also merge files; see
@ref{diff3 Merging}).

The ``normal'' @command{diff3} output format shows each hunk of
differences without surrounding context.  Hunks are labeled depending
on whether they are two-way or three-way, and lines are annotated by
their location in the input files.

@xref{Invoking diff3}, for more information on how to run @command{diff3}.

@menu
* Sample diff3 Input::    Sample @command{diff3} input for examples.
* Example diff3 Normal::  Sample output in the normal format.
* diff3 Hunks::           The format of normal output format.
* Detailed diff3 Normal:: A detailed description of normal output format.
@end menu

@node Sample diff3 Input
@section A Third Sample Input File
@cindex @command{diff3} sample input
@cindex sample input for @command{diff3}

Here is a third sample file that will be used in examples to illustrate
the output of @command{diff3} and how various options can change it.  The
first two files are the same that we used for @command{diff} (@pxref{Sample
diff Input}).  This is the third sample file, called @file{tao}:

@example
The Way that can be told of is not the eternal Way;
The name that can be named is not the eternal name.
The Nameless is the origin of Heaven and Earth;
The named is the mother of all things.

Therefore let there always be non-being,
  so we may see their subtlety,
And let there always be being,
  so we may see their result.
The two are the same,
But after they are produced,
  they have different names.

  -- The Way of Lao-Tzu, tr. Wing-tsit Chan
@end example

@node Example diff3 Normal
@section An Example of @command{diff3} Normal Format

Here is the output of the command @samp{diff3 lao tzu tao}
(@pxref{Sample diff3 Input}, for the complete contents of the files).
Notice that it shows only the lines that are different among the three
files.

@example
====2
1:1,2c
3:1,2c
  The Way that can be told of is not the eternal Way;
  The name that can be named is not the eternal name.
2:0a
====1
1:4c
  The Named is the mother of all things.
2:2,3c
3:4,5c
  The named is the mother of all things.
  @-
====3
1:8c
2:7c
    so we may see their outcome.
3:9c
    so we may see their result.
====
1:11a
2:11,13c
  They both may be called deep and profound.
  Deeper and more profound,
  The door of all subtleties!
3:13,14c
  @-
    -- The Way of Lao-Tzu, tr. Wing-tsit Chan
@end example

@node Detailed diff3 Normal
@section Detailed Description of @command{diff3} Normal Format

Each hunk begins with a line marked @samp{====}.  Three-way hunks have
plain @samp{====} lines, and two-way hunks have @samp{1}, @samp{2}, or
@samp{3} appended to specify which of the three input files differ in
that hunk.  The hunks contain copies of two or three sets of input
lines each preceded by one or two commands identifying where the lines
came from.

Normally, two spaces precede each copy of an input line to distinguish
it from the commands.  But with the @option{-T} or @option{--initial-tab}
option, @command{diff3} uses a tab instead of two spaces; this lines up
tabs correctly.  @xref{Tabs}, for more information.

Commands take the following forms:

@table @samp
@item @var{file}:@var{l}a
This hunk appears after line @var{l} of file @var{file}, and
contains no lines in that file.  To edit this file to yield the other
files, one must append hunk lines taken from the other files.  For
example, @samp{1:11a} means that the hunk follows line 11 in the first
file and contains no lines from that file.

@item @var{file}:@var{r}c
This hunk contains the lines in the range @var{r} of file @var{file}.
The range @var{r} is a comma-separated pair of line numbers, or just one
number if there is only one line.  To edit this file to yield the
other files, one must change the specified lines to be the lines taken
from the other files.  For example, @samp{2:11,13c} means that the hunk
contains lines 11 through 13 from the second file.
@end table

If the last line in a set of input lines is incomplete
(@pxref{Incomplete Lines}), it is distinguished on output from a full
line by a following line that starts with @samp{\}.

@node diff3 Hunks
@section @command{diff3} Hunks
@cindex hunks for @command{diff3}
@cindex @command{diff3} hunks

Groups of lines that differ in two or three of the input files are
called @dfn{diff3 hunks}, by analogy with @command{diff} hunks
(@pxref{Hunks}).  If all three input files differ in a @command{diff3}
hunk, the hunk is called a @dfn{three-way hunk}; if just two input files
differ, it is a @dfn{two-way hunk}.

As with @command{diff}, several solutions are possible.  When comparing the
files @samp{A}, @samp{B}, and @samp{C}, @command{diff3} normally finds
@command{diff3} hunks by merging the two-way hunks output by the two
commands @samp{diff A B} and @samp{diff A C}.  This does not necessarily
minimize the size of the output, but exceptions should be rare.

For example, suppose @file{F} contains the three lines @samp{a},
@samp{b}, @samp{f}, @file{G} contains the lines @samp{g}, @samp{b},
@samp{g}, and @file{H} contains the lines @samp{a}, @samp{b},
@samp{h}.  @samp{diff3 F G H} might output the following:

@example
====2
1:1c
3:1c
  a
2:1c
  g
====
1:3c
  f
2:3c
  g
3:3c
  h
@end example

@noindent
because it found a two-way hunk containing @samp{a} in the first and
third files and @samp{g} in the second file, then the single line
@samp{b} common to all three files, then a three-way hunk containing
the last line of each file.

@node diff3 Merging
@chapter Merging From a Common Ancestor
@cindex merging from a common ancestor

When two people have made changes to copies of the same file,
@command{diff3} can produce a merged output that contains both sets of
changes together with warnings about conflicts.

One might imagine programs with names like @command{diff4} and @command{diff5}
to compare more than three files simultaneously, but in practice the
need rarely arises.  You can use @command{diff3} to merge three or more
sets of changes to a file by merging two change sets at a time.

@command{diff3} can incorporate changes from two modified versions into a
common preceding version.  This lets you merge the sets of changes
represented by the two newer files.  Specify the common ancestor version
as the second argument and the two newer versions as the first and third
arguments, like this:

@example
diff3 @var{mine} @var{older} @var{yours}
@end example

@noindent
You can remember the order of the arguments by noting that they are in
alphabetical order.

@cindex conflict
@cindex overlap
You can think of this as subtracting @var{older} from @var{yours} and
adding the result to @var{mine}, or as merging into @var{mine} the
changes that would turn @var{older} into @var{yours}.  This merging is
well-defined as long as @var{mine} and @var{older} match in the
neighborhood of each such change.  This fails to be true when all three
input files differ or when only @var{older} differs; we call this
a @dfn{conflict}.  When all three input files differ, we call the
conflict an @dfn{overlap}.

@command{diff3} gives you several ways to handle overlaps and conflicts.
You can omit overlaps or conflicts, or select only overlaps,
or mark conflicts with special @samp{<<<<<<<} and @samp{>>>>>>>} lines.

@command{diff3} can output the merge results as an @command{ed} script that
that can be applied to the first file to yield the merged output.
However, it is usually better to have @command{diff3} generate the merged
output directly; this bypasses some problems with @command{ed}.

@menu
* Which Changes::            Selecting changes to incorporate.
* Marking Conflicts::        Marking conflicts.
* Bypassing ed::             Generating merged output directly.
* Merging Incomplete Lines:: How @command{diff3} merges incomplete lines.
* Saving the Changed File::  Emulating System V behavior.
@end menu

@node Which Changes
@section Selecting Which Changes to Incorporate
@cindex overlapping change, selection of
@cindex unmerged change

You can select all unmerged changes from @var{older} to @var{yours} for merging
into @var{mine} with the @option{-e} or @option{--ed} option.  You can
select only the nonoverlapping unmerged changes with @option{-3} or
@option{--easy-only}, and you can select only the overlapping changes with
@option{-x} or @option{--overlap-only}.

The @option{-e}, @option{-3} and @option{-x} options select only
@dfn{unmerged changes}, i.e.@: changes where @var{mine} and @var{yours}
differ; they ignore changes from @var{older} to @var{yours} where
@var{mine} and @var{yours} are identical, because they assume that such
changes have already been merged.  If this assumption is not a safe
one, you can use the @option{-A} or @option{--show-all} option
(@pxref{Marking Conflicts}).

Here is the output of the command @command{diff3} with each of these three
options (@pxref{Sample diff3 Input}, for the complete contents of the files).
Notice that @option{-e} outputs the union of the disjoint sets of changes
output by @option{-3} and @option{-x}.

Output of @samp{diff3 -e lao tzu tao}:
@example
11a

  -- The Way of Lao-Tzu, tr. Wing-tsit Chan
.
8c
  so we may see their result.
.
@end example

Output of @samp{diff3 -3 lao tzu tao}:
@example
8c
  so we may see their result.
.
@end example

Output of @samp{diff3 -x lao tzu tao}:
@example
11a

  -- The Way of Lao-Tzu, tr. Wing-tsit Chan
.
@end example

@node Marking Conflicts
@section Marking Conflicts
@cindex conflict marking
@cindex @samp{<<<<<<<} for marking conflicts

@command{diff3} can mark conflicts in the merged output by
bracketing them with special marker lines.  A conflict
that comes from two files @var{A} and @var{B} is marked as follows:

@example
<<<<<<< @var{A}
@r{lines from @var{A}}
=======
@r{lines from @var{B}}
>>>>>>> @var{B}
@end example

A conflict that comes from three files @var{A}, @var{B} and @var{C} is
marked as follows:

@example
<<<<<<< @var{A}
@r{lines from @var{A}}
||||||| @var{B}
@r{lines from @var{B}}
=======
@r{lines from @var{C}}
>>>>>>> @var{C}
@end example

The @option{-A} or @option{--show-all} option acts like the @option{-e}
option, except that it brackets conflicts, and it outputs all changes
from @var{older} to @var{yours}, not just the unmerged changes.  Thus,
given the sample input files (@pxref{Sample diff3 Input}), @samp{diff3
-A lao tzu tao} puts brackets around the conflict where only @file{tzu}
differs:

@example
<<<<<<< tzu
=======
The Way that can be told of is not the eternal Way;
The name that can be named is not the eternal name.
>>>>>>> tao
@end example

And it outputs the three-way conflict as follows:

@example
<<<<<<< lao
||||||| tzu
They both may be called deep and profound.
Deeper and more profound,
The door of all subtleties!
=======

  -- The Way of Lao-Tzu, tr. Wing-tsit Chan
>>>>>>> tao
@end example

The @option{-E} or @option{--show-overlap} option outputs less information
than the @option{-A} or @option{--show-all} option, because it outputs only
unmerged changes, and it never outputs the contents of the second
file.  Thus the @option{-E} option acts like the @option{-e} option,
except that it brackets the first and third files from three-way
overlapping changes.  Similarly, @option{-X} acts like @option{-x}, except
it brackets all its (necessarily overlapping) changes.  For example,
for the three-way overlapping change above, the @option{-E} and @option{-X}
options output the following:

@example
<<<<<<< lao
=======

  -- The Way of Lao-Tzu, tr. Wing-tsit Chan
>>>>>>> tao
@end example

If you are comparing files that have meaningless or uninformative names,
you can use the @option{--label=@var{label}}
option to show alternate names in the @samp{<<<<<<<}, @samp{|||||||}
and @samp{>>>>>>>} brackets.  This option can be given up to three
times, once for each input file.  Thus @samp{diff3 -A --label X
--label Y --label Z A
B C} acts like @samp{diff3 -A A B C}, except that the output looks like
it came from files named @samp{X}, @samp{Y} and @samp{Z} rather than
from files named @samp{A}, @samp{B} and @samp{C}.

@node Bypassing ed
@section Generating the Merged Output Directly
@cindex merged @command{diff3} format

With the @option{-m} or @option{--merge} option, @command{diff3} outputs the
merged file directly.  This is more efficient than using @command{ed} to
generate it, and works even with non-text files that @command{ed} would
reject.  If you specify @option{-m} without an @command{ed} script option,
@option{-A} is assumed.

For example, the command @samp{diff3 -m lao tzu tao}
(@pxref{Sample diff3 Input} for a copy of the input files) would output
the following:

@example
<<<<<<< tzu
=======
The Way that can be told of is not the eternal Way;
The name that can be named is not the eternal name.
>>>>>>> tao
The Nameless is the origin of Heaven and Earth;
The Named is the mother of all things.
Therefore let there always be non-being,
  so we may see their subtlety,
And let there always be being,
  so we may see their result.
The two are the same,
But after they are produced,
  they have different names.
<<<<<<< lao
||||||| tzu
They both may be called deep and profound.
Deeper and more profound,
The door of all subtleties!
=======

  -- The Way of Lao-Tzu, tr. Wing-tsit Chan
>>>>>>> tao
@end example

@node Merging Incomplete Lines
@section How @command{diff3} Merges Incomplete Lines
@cindex incomplete line merging

With @option{-m}, incomplete lines (@pxref{Incomplete Lines}) are simply
copied to the output as they are found; if the merged output ends in an
conflict and one of the input files ends in an incomplete
line, succeeding @samp{|||||||}, @samp{=======} or @samp{>>>>>>>}
brackets appear somewhere other than the start of a line because
they are appended to the incomplete line.

Without @option{-m}, if an @command{ed} script option is specified and an
incomplete line is found, @command{diff3} generates a warning and acts as
if a newline had been present.

@node Saving the Changed File
@section Saving the Changed File
@cindex System V @command{diff3} compatibility

Traditional Unix @command{diff3} generates an @command{ed} script without the
trailing @samp{w} and @samp{q} commands that save the changes.
System V @command{diff3} generates these extra commands.  @acronym{GNU}
@command{diff3} normally behaves like traditional Unix
@command{diff3}, but with the @option{-i} option it behaves like
System V @command{diff3} and appends the @samp{w} and @samp{q}
commands.

The @option{-i} option requires one of the @command{ed} script options
@option{-AeExX3}, and is incompatible with the merged output option
@option{-m}.

@node Interactive Merging
@chapter Interactive Merging with @command{sdiff}
@cindex diff merging
@cindex interactive merging

With @command{sdiff}, you can merge two files interactively based on a
side-by-side @option{-y} format comparison (@pxref{Side by Side}).  Use
@option{-o @var{file}} or @option{--output=@var{file}} to specify where to
put the merged text.  @xref{Invoking sdiff}, for more details on the
options to @command{sdiff}.

Another way to merge files interactively is to use the Emacs Lisp
package @command{emerge}.  @xref{emerge, , emerge, emacs, The
@acronym{GNU} Emacs Manual}, for more information.

@menu
* sdiff Option Summary:: Summary of @command{sdiff} options.
* Merge Commands::       Merging two files interactively.
@end menu

@node sdiff Option Summary
@section Specifying @command{diff} Options to @command{sdiff}
@cindex @command{sdiff} output format

The following @command{sdiff} options have the same meaning as for
@command{diff}.  @xref{diff Options}, for the use of these options.

@example
-a -b -d -i -t -v
-B -E -I @var{regexp}

--expand-tabs
--ignore-blank-lines  --ignore-case
--ignore-matching-lines=@var{regexp}  --ignore-space-change
--ignore-tab-expansion
--left-column  --minimal  --speed-large-files
--strip-trailing-cr  --suppress-common-lines
--tabsize=@var{columns}  --text  --version  --width=@var{columns}
@end example

For historical reasons, @command{sdiff} has alternate names for some
options.  The @option{-l} option is equivalent to the
@option{--left-column} option, and similarly @option{-s} is equivalent
to @option{--suppress-common-lines}.  The meaning of the @command{sdiff}
@option{-w} and @option{-W} options is interchanged from that of
@command{diff}: with @command{sdiff}, @option{-w @var{columns}} is
equivalent to @option{--width=@var{columns}}, and @option{-W} is
equivalent to @option{--ignore-all-space}.  @command{sdiff} without the
@option{-o} option is equivalent to @command{diff} with the @option{-y}
or @option{--side-by-side} option (@pxref{Side by Side}).

@node Merge Commands
@section Merge Commands
@cindex merge commands
@cindex merging interactively

Groups of common lines, with a blank gutter, are copied from the first
file to the output.  After each group of differing lines, @command{sdiff}
prompts with @samp{%} and pauses, waiting for one of the following
commands.  Follow each command with @key{RET}.

@table @samp
@item e
Discard both versions.
Invoke a text editor on an empty temporary file,
then copy the resulting file to the output.

@item eb
Concatenate the two versions, edit the result in a temporary file,
then copy the edited result to the output.

@item ed
Like @samp{eb}, except precede each version with a header that
shows what file and lines the version came from.

@item el
@itemx e1
Edit a copy of the left version, then copy the result to the output.

@item er
@itemx e2
Edit a copy of the right version, then copy the result to the output.

@item l
@itemx 1
Copy the left version to the output.

@item q
Quit.

@item r
@itemx 2
Copy the right version to the output.

@item s
Silently copy common lines.

@item v
Verbosely copy common lines.  This is the default.
@end table

@vindex EDITOR
The text editor invoked is specified by the @env{EDITOR} environment
variable if it is set.  The default is system-dependent.

@node Merging with patch
@chapter Merging with @command{patch}

@command{patch} takes comparison output produced by @command{diff} and applies
the differences to a copy of the original file, producing a patched
version.  With @command{patch}, you can distribute just the changes to a
set of files instead of distributing the entire file set; your
correspondents can apply @command{patch} to update their copy of the files
with your changes.  @command{patch} automatically determines the diff
format, skips any leading or trailing headers, and uses the headers to
determine which file to patch.  This lets your correspondents feed a
mail message containing a difference listing directly to
@command{patch}.

@command{patch} detects and warns about common problems like forward
patches.  It saves any patches that it could not apply.  It can also maintain a
@code{patchlevel.h} file to ensure that your correspondents apply
diffs in the proper order.

@command{patch} accepts a series of diffs in its standard input, usually
separated by headers that specify which file to patch.  It applies
@command{diff} hunks (@pxref{Hunks}) one by one.  If a hunk does not
exactly match the original file, @command{patch} uses heuristics to try to
patch the file as well as it can.  If no approximate match can be found,
@command{patch} rejects the hunk and skips to the next hunk.  @command{patch}
normally replaces each file @var{f} with its new version, putting reject
hunks (if any) into @samp{@var{f}.rej}.

@xref{Invoking patch}, for detailed information on the options to
@command{patch}.

@menu
* patch Input::            Selecting the type of @command{patch} input.
* Revision Control::       Getting files from @acronym{RCS}, @acronym{SCCS}, etc.
* Imperfect::              Dealing with imperfect patches.
* Creating and Removing::  Creating and removing files with a patch.
* Patching Time Stamps::   Updating time stamps on patched files.
* Multiple Patches::       Handling multiple patches in a file.
* patch Directories::      Changing directory and stripping directories.
* Backups::                Whether backup files are made.
* Backup Names::           Backup file names.
* Reject Names::           Reject file names.
* patch Messages::         Messages and questions @command{patch} can produce.
* patch and POSIX::        Conformance to the @acronym{POSIX} standard.
* patch and Tradition::    @acronym{GNU} versus traditional @command{patch}.
@end menu

@node patch Input
@section Selecting the @command{patch} Input Format
@cindex @command{patch} input format

@command{patch} normally determines which @command{diff} format the patch
file uses by examining its contents.  For patch files that contain
particularly confusing leading text, you might need to use one of the
following options to force @command{patch} to interpret the patch file as a
certain format of diff.  The output formats listed here are the only
ones that @command{patch} can understand.

@table @option
@item -c
@itemx --context
context diff.

@item -e
@itemx --ed
@command{ed} script.

@item -n
@itemx --normal
normal diff.

@item -u
@itemx --unified
unified diff.
@end table

@node Revision Control
@section Revision Control
@cindex revision control
@cindex version control
@cindex @acronym{RCS}
@cindex ClearCase
@cindex @acronym{SCCS}

If a nonexistent input file is under a revision control system
supported by @command{patch}, @command{patch} normally asks the user
whether to get (or check out) the file from the revision control
system.  Patch currently supports @acronym{RCS}, ClearCase and
@acronym{SCCS}.  Under @acronym{RCS} and @acronym{SCCS},
@command{patch} also asks when the input file is read-only and matches
the default version in the revision control system.

@vindex PATCH_GET
The @option{-g @var{num}} or @option{--get=@var{num}} option affects access
to files under supported revision control systems.  If @var{num} is
positive, @command{patch} gets the file without asking the user; if
zero, @command{patch} neither asks the user nor gets the file; and if
negative, @command{patch} asks the user before getting the file.  The
default value of @var{num} is given by the value of the
@env{PATCH_GET} environment variable if it is set; if not, the default
value is zero if @command{patch} is conforming to @acronym{POSIX}, negative
otherwise.  @xref{patch and POSIX}.

@vindex VERSION_CONTROL
The choice of revision control system is unaffected by the
@env{VERSION_CONTROL} environment variable (@pxref{Backup Names}).

@node Imperfect
@section Applying Imperfect Patches
@cindex imperfect patch application

@command{patch} tries to skip any leading text in the patch file,
apply the diff, and then skip any trailing text.  Thus you can feed a
mail message directly to @command{patch}, and it should work.  If the
entire diff is indented by a constant amount of white space,
@command{patch} automatically ignores the indentation.  If a context
diff contains trailing carriage return on each line, @command{patch}
automatically ignores the carriage return.  If a context diff has been
encapsulated by prepending @w{@samp{- }} to lines beginning with @samp{-}
as per @uref{ftp://ftp.isi.edu/in-notes/rfc934.txt, Internet RFC 934},
@command{patch} automatically unencapsulates the input.

However, certain other types of imperfect input require user
intervention or testing.

@menu
* Changed White Space:: When tabs and spaces don't match exactly.
* Reversed Patches::    Applying reversed patches correctly.
* Inexact::             Helping @command{patch} find close matches.
* Dry Runs::            Predicting what @command{patch} will do.
@end menu

@node Changed White Space
@subsection Applying Patches with Changed White Space
@cindex white space in patches

Sometimes mailers, editors, or other programs change spaces into tabs,
or vice versa.  If this happens to a patch file or an input file, the
files might look the same, but @command{patch} will not be able to match
them properly.  If this problem occurs, use the @option{-l} or
@option{--ignore-white-space} option, which makes @command{patch} compare
blank characters (i.e.@: spaces and tabs) loosely so that any nonempty
sequence of blanks in the patch file matches any nonempty sequence of
blanks in the input files.  Non-blank
characters must still match exactly.  Each line of the context must
still match a line in the input file.

@node Reversed Patches
@subsection Applying Reversed Patches
@cindex reversed patches

Sometimes people run @command{diff} with the new file first instead of
second.  This creates a diff that is ``reversed''.  To apply such
patches, give @command{patch} the @option{-R} or @option{--reverse} option.
@command{patch} then attempts to swap each hunk around before applying it.
Rejects come out in the swapped format.

Often @command{patch} can guess that the patch is reversed.  If the first
hunk of a patch fails, @command{patch} reverses the hunk to see if it can
apply it that way.  If it can, @command{patch} asks you if you want to have
the @option{-R} option set; if it can't, @command{patch} continues to apply
the patch normally.  This method cannot detect a reversed patch if it is
a normal diff and the first command is an append (which should have been
a delete) since appends always succeed, because a null context matches
anywhere.  But most patches add or change lines rather than delete them,
so most reversed normal diffs begin with a delete, which fails, and
@command{patch} notices.

If you apply a patch that you have already applied, @command{patch} thinks
it is a reversed patch and offers to un-apply the patch.  This could be
construed as a feature.  If you did this inadvertently and you don't
want to un-apply the patch, just answer @samp{n} to this offer and to
the subsequent ``apply anyway'' question---or type @kbd{C-c} to kill the
@command{patch} process.

@node Inexact
@subsection Helping @command{patch} Find Inexact Matches
@cindex inexact patches
@cindex fuzz factor when patching

For context diffs, and to a lesser extent normal diffs, @command{patch} can
detect when the line numbers mentioned in the patch are incorrect, and
it attempts to find the correct place to apply each hunk of the patch.
As a first guess, it takes the line number mentioned in the hunk, plus
or minus any offset used in applying the previous hunk.  If that is not
the correct place, @command{patch} scans both forward and backward for a
set of lines matching the context given in the hunk.

First @command{patch} looks for a place where all lines of the context
match.  If it cannot find such a place, and it is reading a context or
unified diff, and the maximum fuzz factor is set to 1 or more, then
@command{patch} makes another scan, ignoring the first and last line of
context.  If that fails, and the maximum fuzz factor is set to 2 or
more, it makes another scan, ignoring the first two and last two lines
of context are ignored.  It continues similarly if the maximum fuzz
factor is larger.

The @option{-F @var{lines}} or @option{--fuzz=@var{lines}} option sets the
maximum fuzz factor to @var{lines}.  This option only applies to context
and unified diffs; it ignores up to @var{lines} lines while looking for
the place to install a hunk.  Note that a larger fuzz factor increases
the odds of making a faulty patch.  The default fuzz factor is 2; there
is no point to setting it to more than the number of lines of context
in the diff, ordinarily 3.

If @command{patch} cannot find a place to install a hunk of the patch, it
writes the hunk out to a reject file (@pxref{Reject Names}, for information
on how reject files are named).  It writes out rejected hunks in context
format no matter what form the input patch is in.  If the input is a
normal or @command{ed} diff, many of the contexts are simply null.  The
line numbers on the hunks in the reject file may be different from those
in the patch file: they show the approximate location where @command{patch}
thinks the failed hunks belong in the new file rather than in the old
one.

If the @option{--verbose} option is given, then
as it completes each hunk @command{patch} tells you whether the hunk
succeeded or failed, and if it failed, on which line (in the new file)
@command{patch} thinks the hunk should go.  If this is different from the
line number specified in the diff, it tells you the offset.  A single
large offset @emph{may} indicate that @command{patch} installed a hunk in
the wrong place.  @command{patch} also tells you if it used a fuzz factor
to make the match, in which case you should also be slightly suspicious.

@command{patch} cannot tell if the line numbers are off in an @command{ed}
script, and can only detect wrong line numbers in a normal diff when it
finds a change or delete command.  It may have the same problem with a
context diff using a fuzz factor equal to or greater than the number of
lines of context shown in the diff (typically 3).  In these cases, you
should probably look at a context diff between your original and patched
input files to see if the changes make sense.  Compiling without errors
is a pretty good indication that the patch worked, but not a guarantee.

A patch against an empty file applies to a nonexistent file, and vice
versa.  @xref{Creating and Removing}.

@command{patch} usually produces the correct results, even when it must
make many guesses.  However, the results are guaranteed only when
the patch is applied to an exact copy of the file that the patch was
generated from.

@node Dry Runs
@subsection Predicting what @command{patch} will do
@cindex testing @command{patch}
@cindex dry runs for @command{patch}

It may not be obvious in advance what @command{patch} will do with a
complicated or poorly formatted patch.  If you are concerned that the
input might cause @command{patch} to modify the wrong files, you can
use the @option{--dry-run} option, which causes @command{patch} to
print the results of applying patches without actually changing any
files.  You can then inspect the diagnostics generated by the dry run
to see whether @command{patch} will modify the files that you expect.
If the patch does not do what you want, you can modify the patch (or
the other options to @command{patch}) and try another dry run.  Once
you are satisfied with the proposed patch you can apply it by invoking
@command{patch} as before, but this time without the
@option{--dry-run} option.

@node Creating and Removing
@section Creating and Removing Files
@cindex creating files
@cindex empty files, removing
@cindex removing empty files

Sometimes when comparing two directories, a file may exist in one
directory but not the other.  If you give @command{diff} the
@option{-N} or @option{--new-file} option, or if you supply an old or
new file that is named @file{/dev/null} or is empty and is dated the
Epoch (1970-01-01 00:00:00 UTC), @command{diff} outputs a patch that
adds or deletes the contents of this file.  When given such a patch,
@command{patch} normally creates a new file or removes the old file.
However, when conforming to @acronym{POSIX} (@pxref{patch and POSIX}),
@command{patch} does not remove the old file, but leaves it empty.
The @option{-E} or @option{--remove-empty-files} option causes
@command{patch} to remove output files that are empty after applying a
patch, even if the patch does not appear to be one that removed the
file.

If the patch appears to create a file that already exists,
@command{patch} asks for confirmation before applying the patch.

@node Patching Time Stamps
@section Updating Time Stamps on Patched Files
@cindex time stamps on patched files

When @command{patch} updates a file, it normally sets the file's
last-modified time stamp to the current time of day.  If you are using
@command{patch} to track a software distribution, this can cause
@command{make} to incorrectly conclude that a patched file is out of
date.  For example, if @file{syntax.c} depends on @file{syntax.y}, and
@command{patch} updates @file{syntax.c} and then @file{syntax.y}, then
@file{syntax.c} will normally appear to be out of date with respect to
@file{syntax.y} even though its contents are actually up to date.

The @option{-Z} or @option{--set-utc} option causes @command{patch} to
set a patched file's modification and access times to the time stamps
given in context diff headers.  If the context diff headers do not
specify a time zone, they are assumed to use Coordinated Universal
Time (@acronym{UTC}, often known as @acronym{GMT}).

The @option{-T} or @option{--set-time} option acts like @option{-Z} or
@option{--set-utc}, except that it assumes that the context diff
headers' time stamps use local time instead of @acronym{UTC}.  This option
is not recommended, because patches using local time cannot easily be
used by people in other time zones, and because local time stamps are
ambiguous when local clocks move backwards during daylight-saving time
adjustments.  If the context diff headers specify a time zone, this
option is equivalent to @option{-Z} or @option{--set-utc}.

@command{patch} normally refrains from setting a file's time stamps if
the file's original last-modified time stamp does not match the time
given in the diff header, of if the file's contents do not exactly
match the patch.  However, if the @option{-f} or @option{--force}
option is given, the file's time stamps are set regardless.

Due to the limitations of the current @command{diff} format,
@command{patch} cannot update the times of files whose contents have
not changed.  Also, if you set file time stamps to values other than
the current time of day, you should also remove (e.g., with @samp{make
clean}) all files that depend on the patched files, so that later
invocations of @command{make} do not get confused by the patched
files' times.

@node Multiple Patches
@section Multiple Patches in a File
@cindex multiple patches
@cindex intuiting file names from patches

If the patch file contains more than one patch, and if you do not
specify an input file on the command line, @command{patch} tries to
apply each patch as if they came from separate patch files.  This
means that it determines the name of the file to patch for each patch,
and that it examines the leading text before each patch for file names
and prerequisite revision level (@pxref{Making Patches}, for more on
that topic).

@command{patch} uses the following rules to intuit a file name from
the leading text before a patch.  First, @command{patch} takes an
ordered list of candidate file names as follows:

@itemize @bullet
@item
If the header is that of a context diff, @command{patch} takes the old
and new file names in the header.  A name is ignored if it does not
have enough slashes to satisfy the @option{-p@var{num}} or
@option{--strip=@var{num}} option.  The name @file{/dev/null} is also
ignored.

@item
If there is an @samp{Index:} line in the leading garbage and if either
the old and new names are both absent or if @command{patch} is
conforming to @acronym{POSIX}, @command{patch} takes the name in the
@samp{Index:} line.

@item
For the purpose of the following rules, the candidate file names are
considered to be in the order (old, new, index), regardless of the
order that they appear in the header.
@end itemize

@noindent
Then @command{patch} selects a file name from the candidate list as
follows:

@itemize @bullet
@item
If some of the named files exist, @command{patch} selects the first
name if conforming to @acronym{POSIX}, and the best name otherwise.

@item
If @command{patch} is not ignoring @acronym{RCS}, ClearCase, and @acronym{SCCS}
(@pxref{Revision Control}), and no named files exist but an @acronym{RCS},
ClearCase, or @acronym{SCCS} master is found, @command{patch} selects the
first named file with an @acronym{RCS}, ClearCase, or @acronym{SCCS} master.

@item
If no named files exist, no @acronym{RCS}, ClearCase, or @acronym{SCCS} master
was found, some names are given, @command{patch} is not conforming to
@acronym{POSIX}, and the patch appears to create a file, @command{patch}
selects the best name requiring the creation of the fewest
directories.

@item
If no file name results from the above heuristics, you are asked for
the name of the file to patch, and @command{patch} selects that name.
@end itemize

To determine the @dfn{best} of a nonempty list of file names,
@command{patch} first takes all the names with the fewest path name
components; of those, it then takes all the names with the shortest
basename; of those, it then takes all the shortest names; finally, it
takes the first remaining name.

@xref{patch and POSIX}, to see whether @command{patch} is conforming
to @acronym{POSIX}.

@node patch Directories
@section Applying Patches in Other Directories
@cindex directories and patch
@cindex patching directories

The @option{-d @var{directory}} or @option{--directory=@var{directory}}
option to @command{patch} makes directory @var{directory} the current
directory for interpreting both file names in the patch file, and file
names given as arguments to other options (such as @option{-B} and
@option{-o}).  For example, while in a mail reading program, you can patch
a file in the @file{/usr/src/emacs} directory directly from a message
containing the patch like this:

@example
| patch -d /usr/src/emacs
@end example

Sometimes the file names given in a patch contain leading directories,
but you keep your files in a directory different from the one given in
the patch.  In those cases, you can use the
@option{-p@var{number}} or @option{--strip=@var{number}}
option to set the file name strip count to @var{number}.  The strip
count tells @command{patch} how many slashes, along with the directory
names between them, to strip from the front of file names.  A sequence
of one or more adjacent slashes is counted as a single slash.  By
default, @command{patch} strips off all leading directories, leaving
just the base file names.

For example, suppose the file name in the patch file is
@file{/gnu/src/emacs/etc/NEWS}.  Using @option{-p0} gives the
entire file name unmodified, @option{-p1} gives
@file{gnu/src/emacs/etc/NEWS} (no leading slash), @option{-p4} gives
@file{etc/NEWS}, and not specifying @option{-p} at all gives @file{NEWS}.

@command{patch} looks for each file (after any slashes have been stripped)
in the current directory, or if you used the @option{-d @var{directory}}
option, in that directory.

@node Backups
@section Backup Files
@cindex backup file strategy

Normally, @command{patch} creates a backup file if the patch does not
exactly match the original input file, because in that case the
original data might not be recovered if you undo the patch with
@samp{patch -R} (@pxref{Reversed Patches}).  However, when conforming
to @acronym{POSIX}, @command{patch} does not create backup files by
default.  @xref{patch and POSIX}.

The @option{-b} or @option{--backup} option causes @command{patch} to
make a backup file regardless of whether the patch matches the
original input.  The @option{--backup-if-mismatch} option causes
@command{patch} to create backup files for mismatches files; this is
the default when not conforming to @acronym{POSIX}.  The
@option{--no-backup-if-mismatch} option causes @command{patch} to not
create backup files, even for mismatched patches; this is the default
when conforming to @acronym{POSIX}.

When backing up a file that does not exist, an empty, unreadable
backup file is created as a placeholder to represent the nonexistent
file.

@node Backup Names
@section Backup File Names
@cindex backup file names

Normally, @command{patch} renames an original input file into a backup
file by appending to its name the extension @samp{.orig}, or @samp{~}
if using @samp{.orig} would make the backup file name too
long.@footnote{A coding error in @acronym{GNU} @command{patch} version
2.5.4 causes it to always use @samp{~}, but this should be fixed in
the next release.}  The @option{-z @var{backup-suffix}} or
@option{--suffix=@var{backup-suffix}} option causes @command{patch} to
use @var{backup-suffix} as the backup extension instead.

@vindex SIMPLE_BACKUP_SUFFIX
Alternately, you can specify the extension for backup files with the
@env{SIMPLE_BACKUP_SUFFIX} environment variable, which the options
override.

@command{patch} can also create numbered backup files the way
@acronym{GNU} Emacs does.  With this method, instead of having a
single backup of each file, @command{patch} makes a new backup file
name each time it patches a file.  For example, the backups of a file
named @file{sink} would be called, successively, @file{sink.~1~},
@file{sink.~2~}, @file{sink.~3~}, etc.

@vindex PATCH_VERSION_CONTROL
@vindex VERSION_CONTROL
The @option{-V @var{backup-style}} or
@option{--version-control=@var{backup-style}} option takes as an
argument a method for creating backup file names.  You can alternately
control the type of backups that @command{patch} makes with the
@env{PATCH_VERSION_CONTROL} environment variable, which the
@option{-V} option overrides.  If @env{PATCH_VERSION_CONTROL} is not
set, the @env{VERSION_CONTROL} environment variable is used instead.
Please note that these options and variables control backup file
names; they do not affect the choice of revision control system
(@pxref{Revision Control}).

The values of these environment variables and the argument to the
@option{-V} option are like the @acronym{GNU} Emacs @code{version-control}
variable (@pxref{Backup Names, , , emacs, The @acronym{GNU} Emacs Manual},
for more information on backup versions in Emacs).  They also
recognize synonyms that are more descriptive.  The valid values are
listed below; unique abbreviations are acceptable.

@table @option
@item t
@itemx numbered
Always make numbered backups.

@item nil
@itemx existing
Make numbered backups of files that already have them, simple backups of
the others.  This is the default.

@item never
@itemx simple
Always make simple backups.
@end table

You can also tell @command{patch} to prepend a prefix, such as a
directory name, to produce backup file names.  The @option{-B
@var{prefix}} or @option{--prefix=@var{prefix}} option makes backup
files by prepending @var{prefix} to them.  The @option{-Y
@var{prefix}} or @option{--basename-prefix=@var{prefix}} prepends
@var{prefix} to the last file name component of backup file names
instead; for example, @option{-Y ~} causes the backup name for
@file{dir/file.c} to be @file{dir/~file.c}.  If you use either of
these prefix options, the suffix-based options are ignored.

If you specify the output file with the @option{-o} option, that file is
the one that is backed up, not the input file.

Options that affect the names of backup files do not affect whether
backups are made.  For example, if you specify the
@option{--no-backup-if-mismatch} option, none of the options described
in this section have any affect, because no backups are made.

@node Reject Names
@section Reject File Names
@cindex reject file names

The names for reject files (files containing patches that
@command{patch} could not find a place to apply) are normally the name
of the output file with @samp{.rej} appended (or @samp{#} if using
@samp{.rej} would make the backup file name too long).

Alternatively, you can tell @command{patch} to place all of the rejected
patches in a single file.  The @option{-r @var{reject-file}} or
@option{--reject-file=@var{reject-file}} option uses @var{reject-file} as
the reject file name.

@node patch Messages
@section Messages and Questions from @command{patch}
@cindex @command{patch} messages and questions
@cindex diagnostics from @command{patch}
@cindex messages from @command{patch}

@command{patch} can produce a variety of messages, especially if it
has trouble decoding its input.  In a few situations where it's not
sure how to proceed, @command{patch} normally prompts you for more
information from the keyboard.  There are options to produce more or
fewer messages, to have it not ask for keyboard input, and to
affect the way that file names are quoted in messages.

@menu
* More or Fewer Messages::    Controlling the verbosity of @command{patch}.
* patch and Keyboard Input::  Inhibiting keyboard input.
* patch Quoting Style::       Quoting file names in diagnostics.
@end menu

@command{patch} exits with status 0 if all hunks are applied successfully,
1 if some hunks cannot be applied, and 2 if there is more serious trouble.
When applying a set of patches in a loop, you should check the
exit status, so you don't apply a later patch to a partially patched
file.

@node More or Fewer Messages
@subsection Controlling the Verbosity of @command{patch}
@cindex verbose messages from @command{patch}
@cindex inhibit messages from @command{patch}

You can cause @command{patch} to produce more messages by using the
@option{--verbose} option.  For example, when you give this option,
the message @samp{Hmm...} indicates that @command{patch} is reading text in
the patch file, attempting to determine whether there is a patch in that
text, and if so, what kind of patch it is.

You can inhibit all terminal output from @command{patch}, unless an error
occurs, by using the @option{-s}, @option{--quiet}, or @option{--silent}
option.

@node patch and Keyboard Input
@subsection Inhibiting Keyboard Input
@cindex keyboard input to @command{patch}

There are two ways you can prevent @command{patch} from asking you any
questions.  The @option{-f} or @option{--force} option assumes that you know
what you are doing.  It causes @command{patch} to do the following:

@itemize @bullet
@item
Skip patches that do not contain file names in their headers.

@item
Patch files even though they have the wrong version for the
@samp{Prereq:} line in the patch;

@item
Assume that patches are not reversed even if they look like they are.
@end itemize

@noindent
The @option{-t} or @option{--batch} option is similar to @option{-f}, in that
it suppresses questions, but it makes somewhat different assumptions:

@itemize @bullet
@item
Skip patches that do not contain file names in their headers
(the same as @option{-f}).

@item
Skip patches for which the file has the wrong version for the
@samp{Prereq:} line in the patch;

@item
Assume that patches are reversed if they look like they are.
@end itemize

@node patch Quoting Style
@subsection @command{patch} Quoting Style
@cindex quoting style

When @command{patch} outputs a file name in a diagnostic message, it
can format the name in any of several ways.  This can be useful to
output file names unambiguously, even if they contain punctuation or
special characters like newlines.  The
@option{--quoting-style=@var{word}} option controls how names are
output.  The @var{word} should be one of the following:

@table @samp
@item literal
Output names as-is.
@item shell
Quote names for the shell if they contain shell metacharacters or would
cause ambiguous output.
@item shell-always
Quote names for the shell, even if they would normally not require quoting.
@item c
Quote names as for a C language string.
@item escape
Quote as with @samp{c} except omit the surrounding double-quote
characters.
@c The following are not yet implemented in patch 2.5.4.
@c @item clocale
@c Quote as with @samp{c} except use quotation marks appropriate for the
@c locale.
@c @item locale
@c @c Use @t instead of @samp to avoid duplicate quoting in some output styles.
@c Like @samp{clocale}, but quote @t{`like this'} instead of @t{"like
@c this"} in the default C locale.  This looks nicer on many displays.
@end table

@vindex QUOTING_STYLE
You can specify the default value of the @option{--quoting-style}
option with the environment variable @env{QUOTING_STYLE}.  If that
environment variable is not set, the default value is @samp{shell},
but this default may change in a future version of @command{patch}.

@node patch and POSIX
@section @command{patch} and the @acronym{POSIX} Standard
@cindex @acronym{POSIX}

@vindex POSIXLY_CORRECT
If you specify the @option{--posix} option, or set the
@env{POSIXLY_CORRECT} environment variable, @command{patch} conforms
more strictly to the @acronym{POSIX} standard, as follows:

@itemize @bullet
@item
Take the first existing file from the list (old, new, index)
when intuiting file names from diff headers.  @xref{Multiple Patches}.

@item
Do not remove files that are removed by a diff.
@xref{Creating and Removing}.

@item
Do not ask whether to get files from @acronym{RCS}, ClearCase, or
@acronym{SCCS}.  @xref{Revision Control}.

@item
Require that all options precede the files in the command line.

@item
Do not backup files, even when there is a mismatch.  @xref{Backups}.

@end itemize

@node patch and Tradition
@section @acronym{GNU} @command{patch} and Traditional @command{patch}
@cindex traditional @command{patch}

The current version of @acronym{GNU} @command{patch} normally follows the
@acronym{POSIX} standard.  @xref{patch and POSIX}, for the few exceptions
to this general rule.

Unfortunately, @acronym{POSIX} redefined the behavior of @command{patch} in
several important ways.  You should be aware of the following
differences if you must interoperate with traditional @command{patch},
or with @acronym{GNU} @command{patch} version 2.1 and earlier.

@itemize @bullet
@item
In traditional @command{patch}, the @option{-p} option's operand was
optional, and a bare @option{-p} was equivalent to @option{-p0}.  The
@option{-p} option now requires an operand, and @option{-p@ 0} is now
equivalent to @option{-p0}.  For maximum compatibility, use options
like @option{-p0} and @option{-p1}.

Also, traditional @command{patch} simply counted slashes when
stripping path prefixes; @command{patch} now counts pathname
components.  That is, a sequence of one or more adjacent slashes now
counts as a single slash.  For maximum portability, avoid sending
patches containing @file{//} in file names.

@item
In traditional @command{patch}, backups were enabled by default.  This
behavior is now enabled with the @option{-b} or @option{--backup}
option.

Conversely, in @acronym{POSIX} @command{patch}, backups are never made,
even when there is a mismatch.  In @acronym{GNU} @command{patch}, this
behavior is enabled with the @option{--no-backup-if-mismatch} option,
or by conforming to @acronym{POSIX}.

The @option{-b@ @var{suffix}} option of traditional @command{patch} is
equivalent to the @samp{-b -z@ @var{suffix}} options of @acronym{GNU}
@command{patch}.

@item
Traditional @command{patch} used a complicated (and incompletely
documented) method to intuit the name of the file to be patched from
the patch header.  This method did not conform to @acronym{POSIX}, and had
a few gotchas.  Now @command{patch} uses a different, equally
complicated (but better documented) method that is optionally
@acronym{POSIX}-conforming; we hope it has fewer gotchas.  The two methods
are compatible if the file names in the context diff header and the
@samp{Index:} line are all identical after prefix-stripping.  Your
patch is normally compatible if each header's file names all contain
the same number of slashes.

@item
When traditional @command{patch} asked the user a question, it sent
the question to standard error and looked for an answer from the first
file in the following list that was a terminal: standard error,
standard output, @file{/dev/tty}, and standard input.  Now
@command{patch} sends questions to standard output and gets answers
from @file{/dev/tty}.  Defaults for some answers have been changed so
that @command{patch} never goes into an infinite loop when using
default answers.

@item
Traditional @command{patch} exited with a status value that counted
the number of bad hunks, or with status 1 if there was real trouble.
Now @command{patch} exits with status 1 if some hunks failed, or with
2 if there was real trouble.

@item
Limit yourself to the following options when sending instructions
meant to be executed by anyone running @acronym{GNU} @command{patch},
traditional @command{patch}, or a @command{patch} that conforms to
@acronym{POSIX}.  Spaces are significant in the following list, and
operands are required.

@example
@option{-c}
@option{-d @var{dir}}
@option{-D @var{define}}
@option{-e}
@option{-l}
@option{-n}
@option{-N}
@option{-o @var{outfile}}
@option{-p@var{num}}
@option{-R}
@option{-r @var{rejectfile}}
@end example

@end itemize

@node Making Patches
@chapter Tips for Making and Using Patches

Use some common sense when making and using patches.  For example,
when sending bug fixes to a program's maintainer, send several small
patches, one per independent subject, instead of one large,
harder-to-digest patch that covers all the subjects.

Here are some other things you should keep in mind if you are going to
distribute patches for updating a software package.

@menu
* Tips for Patch Producers::    Advice for making patches.
* Tips for Patch Consumers::    Advice for using patches.
* Avoiding Common Mistakes::    Avoiding common mistakes when using @command{patch}.
* Generating Smaller Patches::  How to generate smaller patches.
@end menu

@node Tips for Patch Producers
@section Tips for Patch Producers
@cindex patch producer tips

To create a patch that changes an older version of a package into a
newer version, first make a copy of the older and newer versions in
adjacent subdirectories.  It is common to do that by unpacking
@command{tar} archives of the two versions.

To generate the patch, use the command @samp{diff -Naur @var{old}
@var{new}} where @var{old} and @var{new} identify the old and new
directories.  The names @var{old} and @var{new} should not contain any
slashes.  The @option{-N} option lets the patch create and remove
files; @option{-a} lets the patch update non-text files; @option{-u}
generates useful time stamps and enough context; and @option{-r} lets
the patch update subdirectories.  Here is an example command, using
Bourne shell syntax:

@example
diff -Naur gcc-3.0.3 gcc-3.0.4
@end example

Tell your recipients how to apply the patches.  This should include
which working directory to use, and which @command{patch} options to
use; the option @samp{-p1} is recommended.  Test your procedure by
pretending to be a recipient and applying your patches to a copy of
the original files.

@xref{Avoiding Common Mistakes}, for how to avoid common mistakes when
generating a patch.

@node Tips for Patch Consumers
@section Tips for Patch Consumers
@cindex patch consumer tips

A patch producer should tell recipients how to apply the patches, so
the first rule of thumb for a patch consumer is to follow the
instructions supplied with the patch.

@acronym{GNU} @command{diff} can analyze files with arbitrarily long lines
and files that end in incomplete lines.  However, older versions of
@command{patch} cannot patch such files.  If you are having trouble
applying such patches, try upgrading to a recent version of @acronym{GNU}
@command{patch}.

@node Avoiding Common Mistakes
@section Avoiding Common Mistakes
@cindex common mistakes with patches
@cindex patch, common mistakes

When producing a patch for multiple files, apply @command{diff} to
directories whose names do not have slashes.  This reduces confusion
when the patch consumer specifies the @option{-p@var{number}} option,
since this option can have surprising results when the old and new
file names have different numbers of slashes.  For example, do not
send a patch with a header that looks like this:

@example
diff -Naur v2.0.29/prog/README prog/README
--- v2.0.29/prog/README	2002-03-10 23:30:39.942229878 -0800
+++ prog/README	2002-03-17 20:49:32.442260588 -0800
@end example

@noindent
because the two file names have different numbers of slashes, and
different versions of @command{patch} interpret the file names
differently.  To avoid confusion, send output that looks like this
instead:

@example
diff -Naur v2.0.29/prog/README v2.0.30/prog/README
--- v2.0.29/prog/README	2002-03-10 23:30:39.942229878 -0800
+++ v2.0.30/prog/README	2002-03-17 20:49:32.442260588 -0800
@end example

Make sure you have specified the file names correctly, either in a
context diff header or with an @samp{Index:} line.  Take care to not send out
reversed patches, since these make people wonder whether they have
already applied the patch.

Avoid sending patches that compare backup file names like
@file{README.orig} or @file{README~}, since this might confuse
@command{patch} into patching a backup file instead of the real file.
Instead, send patches that compare the same base file names in
different directories, e.g.@: @file{old/README} and @file{new/README}.

To save people from partially applying a patch before other patches that
should have gone before it, you can make the first patch in the patch
file update a file with a name like @file{patchlevel.h} or
@file{version.c}, which contains a patch level or version number.  If
the input file contains the wrong version number, @command{patch} will
complain immediately.

An even clearer way to prevent this problem is to put a @samp{Prereq:}
line before the patch.  If the leading text in the patch file contains a
line that starts with @samp{Prereq:}, @command{patch} takes the next word
from that line (normally a version number) and checks whether the next
input file contains that word, preceded and followed by either
white space or a newline.  If not, @command{patch} prompts you for
confirmation before proceeding.  This makes it difficult to accidentally
apply patches in the wrong order.

@node Generating Smaller Patches
@section Generating Smaller Patches
@cindex patches, shrinking

The simplest way to generate a patch is to use @samp{diff -Naur}
(@pxref{Tips for Patch Producers}), but you might be able to reduce
the size of the patch by renaming or removing some files before making
the patch.  If the older version of the package contains any files
that the newer version does not, or if any files have been renamed
between the two versions, make a list of @command{rm} and @command{mv}
commands for the user to execute in the old version directory before
applying the patch.  Then run those commands yourself in the scratch
directory.

If there are any files that you don't need to include in the patch
because they can easily be rebuilt from other files (for example,
@file{TAGS} and output from @command{yacc} and @command{makeinfo}),
exclude them from the patch by giving @command{diff} the @option{-x
@var{pattern}} option (@pxref{Comparing Directories}).  If you want
your patch to modify a derived file because your recipients lack tools
to build it, make sure that the patch for the derived file follows any
patches for files that it depends on, so that the recipients' time
stamps will not confuse @command{make}.

Now you can create the patch using @samp{diff -Naur}.  Make sure to
specify the scratch directory first and the newer directory second.

Add to the top of the patch a note telling the user any @command{rm} and
@command{mv} commands to run before applying the patch.  Then you can
remove the scratch directory.

You can also shrink the patch size by using fewer lines of context,
but bear in mind that @command{patch} typically needs at least two
lines for proper operation when patches do not exactly match the input
files.

@node Invoking cmp
@chapter Invoking @command{cmp}
@cindex invoking @command{cmp}
@cindex @command{cmp} invocation

The @command{cmp} command compares two files, and if they differ,
tells the first byte and line number where they differ or reports
that one file is a prefix of the other.  Bytes and
lines are numbered starting with 1.  The arguments of @command{cmp}
are as follows:

@example
cmp @var{options}@dots{} @var{from-file} @r{[}@var{to-file} @r{[}@var{from-skip} @r{[}@var{to-skip}@r{]}@r{]}@r{]}
@end example

The file name @file{-} is always the standard input.  @command{cmp}
also uses the standard input if one file name is omitted.  The
@var{from-skip} and @var{to-skip} operands specify how many bytes to
ignore at the start of each file; they are equivalent to the
@option{--ignore-initial=@var{from-skip}:@var{to-skip}} option.

By default, @command{cmp} outputs nothing if the two files have the
same contents.  If one file is a prefix of the other, @command{cmp}
prints to standard error a message of the following form:

@example
cmp: EOF on @var{shorter-file}
@end example

Otherwise, @command{cmp} prints to standard output a message of the
following form:

@example
@var{from-file} @var{to-file} differ: char @var{byte-number}, line @var{line-number}
@end example

The message formats can differ outside the @acronym{POSIX} locale.
Also, @acronym{POSIX} allows the @acronym{EOF} message to be followed
by a blank and some additional information.

An exit status of 0 means no differences were found, 1 means some
differences were found, and 2 means trouble.

@menu
* cmp Options:: Summary of options to @command{cmp}.
@end menu

@node cmp Options
@section Options to @command{cmp}
@cindex @command{cmp} options
@cindex options for @command{cmp}

Below is a summary of all of the options that @acronym{GNU}
@command{cmp} accepts.  Most options have two equivalent names, one of
which is a single letter preceded by @samp{-}, and the other of which
is a long name preceded by @samp{--}.  Multiple single letter options
(unless they take an argument) can be combined into a single command
line word: @option{-bl} is equivalent to @option{-b -l}.

@table @option
@item -b
@itemx --print-bytes
Print the differing bytes.  Display control bytes as a
@samp{^} followed by a letter of the alphabet and precede bytes
that have the high bit set with @samp{M-} (which stands for ``meta'').

@item --help
Output a summary of usage and then exit.

@item -i @var{skip}
@itemx --ignore-initial=@var{skip}
Ignore any differences in the first @var{skip} bytes of the input
files.  Treat files with fewer than @var{skip} bytes as if they are
empty.  If @var{skip} is of the form
@option{@var{from-skip}:@var{to-skip}}, skip the first @var{from-skip}
bytes of the first input file and the first @var{to-skip} bytes of the
second.

@item -l
@itemx --verbose
Output the (decimal) byte numbers and (octal) values of all differing bytes,
instead of the default standard output.
Also, output the @acronym{EOF} message if one file is shorter than the other.

@item -n @var{count}
@itemx --bytes=@var{count}
Compare at most @var{count} input bytes.

@item -s
@itemx --quiet
@itemx --silent
Do not print anything; only return an exit status indicating whether
the files differ.

@item -v
@itemx --version
Output version information and then exit.
@end table

In the above table, operands that are byte counts are normally
decimal, but may be preceded by @samp{0} for octal and @samp{0x} for
hexadecimal.

A byte count can be followed by a suffix to specify a multiple of that
count; in this case an omitted integer is understood to be 1.  A bare
size letter, or one followed by @samp{iB}, specifies a multiple using
powers of 1024.  A size letter followed by @samp{B} specifies powers
of 1000 instead.  For example, @option{-n 4M} and @option{-n 4MiB} are
equivalent to @option{-n 4194304}, whereas @option{-n 4MB} is
equivalent to @option{-n 4000000}.  This notation is upward compatible
with the @uref{http://www.bipm.fr/enus/3_SI/si-prefixes.html, SI
prefixes} for decimal multiples and with the
@uref{http://physics.nist.gov/cuu/Units/binary.html, IEC 60027-2
prefixes for binary multiples}.

The following suffixes are defined.  Large sizes like @code{1Y} may be
rejected by your computer due to limitations of its arithmetic.

@table @samp
@item kB
@cindex kilobyte, definition of
kilobyte: @math{10^3 = 1000}.
@item k
@itemx K
@itemx KiB
@cindex kibibyte, definition of
kibibyte: @math{2^10 = 1024}.  @samp{K} is special: the SI prefix is
@samp{k} and the IEC 60027-2 prefix is @samp{Ki}, but tradition and
@acronym{POSIX} use @samp{k} to mean @samp{KiB}.
@item MB
@cindex megabyte, definition of
megabyte: @math{10^6 = 1,000,000}.
@item M
@itemx MiB
@cindex mebibyte, definition of
mebibyte: @math{2^20 = 1,048,576}.
@item GB
@cindex gigabyte, definition of
gigabyte: @math{10^9 = 1,000,000,000}.
@item G
@itemx GiB
@cindex gibibyte, definition of
gibibyte: @math{2^30 = 1,073,741,824}.
@item TB
@cindex terabyte, definition of
terabyte:  @math{10^12 = 1,000,000,000,000}.
@item T
@itemx TiB
@cindex tebibyte, definition of
tebibyte: @math{2^40 = 1,099,511,627,776}.
@item PB
@cindex petabyte, definition of
petabyte: @math{10^15 = 1,000,000,000,000,000}.
@item P
@itemx PiB
@cindex pebibyte, definition of
pebibyte: @math{2^50 = 1,125,899,906,842,624}.
@item EB
@cindex exabyte, definition of
exabyte: @math{10^18 = 1,000,000,000,000,000,000}.
@item E
@itemx EiB
@cindex exbibyte, definition of
exbibyte: @math{2^60 = 1,152,921,504,606,846,976}.
@item ZB
@cindex zettabyte, definition of
zettabyte: @math{10^21 = 1,000,000,000,000,000,000,000}
@item Z
@itemx ZiB
@math{2^70 = 1,180,591,620,717,411,303,424}.
(@samp{Zi} is a GNU extension to IEC 60027-2.)
@item YB
@cindex yottabyte, definition of
yottabyte: @math{10^24 = 1,000,000,000,000,000,000,000,000}.
@item Y
@itemx YiB
@math{2^80 = 1,208,925,819,614,629,174,706,176}.
(@samp{Yi} is a GNU extension to IEC 60027-2.)
@end table

@node Invoking diff
@chapter Invoking @command{diff}
@cindex invoking @command{diff}
@cindex @command{diff} invocation

The format for running the @command{diff} command is:

@example
diff @var{options}@dots{} @var{files}@dots{}
@end example

In the simplest case, two file names @var{from-file} and
@var{to-file} are given, and @command{diff} compares the contents of
@var{from-file} and @var{to-file}.  A file name of @file{-} stands for
text read from the standard input.  As a special case, @samp{diff - -}
compares a copy of standard input to itself.

If one file is a directory and the other is not, @command{diff} compares
the file in the directory whose name is that of the non-directory.
The non-directory file must not be @file{-}.

If two file names are given and both are directories,
@command{diff} compares corresponding files in both directories, in
alphabetical order; this comparison is not recursive unless the
@option{-r} or @option{--recursive} option is given.  @command{diff} never
compares the actual contents of a directory as if it were a file.  The
file that is fully specified may not be standard input, because standard
input is nameless and the notion of ``file with the same name'' does not
apply.

If the @option{--from-file=@var{file}} option is given, the number of
file names is arbitrary, and @var{file} is compared to each named file.
Similarly, if the @option{--to-file=@var{file}} option is given, each
named file is compared to @var{file}.

@command{diff} options begin with @samp{-}, so normally file names
may not begin with @samp{-}.  However, @option{--} as an
argument by itself treats the remaining arguments as file names even if
they begin with @samp{-}.

An exit status of 0 means no differences were found, 1 means some
differences were found, and 2 means trouble.  Normally, differing
binary files count as trouble, but this can be altered by using the
@option{-a} or @option{--text} option, or the @option{-q} or
@option{--brief} option.

@menu
* diff Options:: Summary of options to @command{diff}.
@end menu

@node diff Options
@section Options to @command{diff}
@cindex @command{diff} options
@cindex options for @command{diff}

Below is a summary of all of the options that @acronym{GNU}
@command{diff} accepts.  Most options have two equivalent names, one
of which is a single letter preceded by @samp{-}, and the other of
which is a long name preceded by @samp{--}.  Multiple single letter
options (unless they take an argument) can be combined into a single
command line word: @option{-ac} is equivalent to @option{-a -c}.  Long
named options can be abbreviated to any unique prefix of their name.
Brackets ([ and ]) indicate that an option takes an optional argument.

@table @option
@item -a
@itemx --text
Treat all files as text and compare them line-by-line, even if they
do not seem to be text.  @xref{Binary}.

@item -b
@itemx --ignore-space-change
Ignore changes in amount of white space.  @xref{White Space}.

@item -B
@itemx --ignore-blank-lines
Ignore changes that just insert or delete blank lines.  @xref{Blank
Lines}.

@item --binary
Read and write data in binary mode.  @xref{Binary}.

@item -c
Use the context output format, showing three lines of context.
@xref{Context Format}.

@item -C @var{lines}
@itemx --context@r{[}=@var{lines}@r{]}
Use the context output format, showing @var{lines} (an integer) lines of
context, or three if @var{lines} is not given.  @xref{Context Format}.
For proper operation, @command{patch} typically needs at least two lines of
context.

For compatibility @command{diff} also supports an obsolete option
syntax @option{-@var{lines}} that has effect when combined with
@option{-c}, @option{-p}, or @option{-u}.  New scripts should use
@option{-C @var{lines}} or @option{-U @var{lines}} instead.

@item --changed-group-format=@var{format}
Use @var{format} to output a line group containing differing lines from
both files in if-then-else format.  @xref{Line Group Formats}.

@item -d
@itemx --minimal
Change the algorithm perhaps find a smaller set of changes.  This makes
@command{diff} slower (sometimes much slower).  @xref{diff Performance}.

@item -D @var{name}
@itemx --ifdef=@var{name}
Make merged @samp{#ifdef} format output, conditional on the preprocessor
macro @var{name}.  @xref{If-then-else}.

@item -e
@itemx --ed
Make output that is a valid @command{ed} script.  @xref{ed Scripts}.

@item -E
@itemx --ignore-tab-expansion
Ignore changes due to tab expansion.
@xref{White Space}.

@item -f
@itemx --forward-ed
Make output that looks vaguely like an @command{ed} script but has changes
in the order they appear in the file.  @xref{Forward ed}.

@item -F @var{regexp}
@itemx --show-function-line=@var{regexp}
In context and unified format, for each hunk of differences, show some
of the last preceding line that matches @var{regexp}.  @xref{Specified
Headings}.

@item --from-file=@var{file}
Compare @var{file} to each operand; @var{file} may be a directory.

@item --help
Output a summary of usage and then exit.

@item --horizon-lines=@var{lines}
Do not discard the last @var{lines} lines of the common prefix
and the first @var{lines} lines of the common suffix.
@xref{diff Performance}.

@item -i
@itemx --ignore-case
Ignore changes in case; consider upper- and lower-case letters
equivalent.  @xref{Case Folding}.

@item -I @var{regexp}
@itemx --ignore-matching-lines=@var{regexp}
Ignore changes that just insert or delete lines that match @var{regexp}.
@xref{Specified Lines}.

@item --ignore-file-name-case
Ignore case when comparing file names during recursive comparison.
@xref{Comparing Directories}.

@item -l
@itemx --paginate
Pass the output through @command{pr} to paginate it.  @xref{Pagination}.

@item --label=@var{label}
Use @var{label} instead of the file name in the context format
(@pxref{Context Format}) and unified format (@pxref{Unified Format})
headers.  @xref{RCS}.

@item --left-column
Print only the left column of two common lines in side by side format.
@xref{Side by Side Format}.

@item --line-format=@var{format}
Use @var{format} to output all input lines in if-then-else format.
@xref{Line Formats}.

@item -n
@itemx --rcs
Output @acronym{RCS}-format diffs; like @option{-f} except that each command
specifies the number of lines affected.  @xref{RCS}.

@item -N
@itemx --new-file
In directory comparison, if a file is found in only one directory,
treat it as present but empty in the other directory.  @xref{Comparing
Directories}.

@item --new-group-format=@var{format}
Use @var{format} to output a group of lines taken from just the second
file in if-then-else format.  @xref{Line Group Formats}.

@item --new-line-format=@var{format}
Use @var{format} to output a line taken from just the second file in
if-then-else format.  @xref{Line Formats}.

@item --old-group-format=@var{format}
Use @var{format} to output a group of lines taken from just the first
file in if-then-else format.  @xref{Line Group Formats}.

@item --old-line-format=@var{format}
Use @var{format} to output a line taken from just the first file in
if-then-else format.  @xref{Line Formats}.

@item -p
@itemx --show-c-function
Show which C function each change is in.  @xref{C Function Headings}.

@item -q
@itemx --brief
Report only whether the files differ, not the details of the
differences.  @xref{Brief}.

@item -r
@itemx --recursive
When comparing directories, recursively compare any subdirectories
found.  @xref{Comparing Directories}.

@item -s
@itemx --report-identical-files
Report when two files are the same.  @xref{Comparing Directories}.

@item -S @var{file}
@itemx --starting-file=@var{file}
When comparing directories, start with the file @var{file}.  This is
used for resuming an aborted comparison.  @xref{Comparing Directories}.

@item --speed-large-files
Use heuristics to speed handling of large files that have numerous
scattered small changes.  @xref{diff Performance}.

@item --strip-trailing-cr
Strip any trailing carriage return at the end of an input line.
@xref{Binary}.

@item --suppress-common-lines
Do not print common lines in side by side format.
@xref{Side by Side Format}.

@item -t
@itemx --expand-tabs
Expand tabs to spaces in the output, to preserve the alignment of tabs
in the input files.  @xref{Tabs}.

@item -T
@itemx --initial-tab
Output a tab rather than a space before the text of a line in normal or
context format.  This causes the alignment of tabs in the line to look
normal.  @xref{Tabs}.

@item --tabsize=@var{columns}
Assume that tab stops are set every @var{columns} (default 8) print
columns.  @xref{Tabs}.

@item --suppress-blank-empty
Suppress any blanks before newlines when printing the representation
of an empty line, when outputting normal, context, or unified format.
@xref{Trailing Blanks}.

@item --to-file=@var{file}
Compare each operand to @var{file}; @var{file} may be a directory.

@item -u
Use the unified output format, showing three lines of context.
@xref{Unified Format}.

@item --unchanged-group-format=@var{format}
Use @var{format} to output a group of common lines taken from both files
in if-then-else format.  @xref{Line Group Formats}.

@item --unchanged-line-format=@var{format}
Use @var{format} to output a line common to both files in if-then-else
format.  @xref{Line Formats}.

@item --unidirectional-new-file
When comparing directories, if a file appears only in the second
directory of the two, treat it as present but empty in the other.
@xref{Comparing Directories}.

@item -U @var{lines}
@itemx --unified@r{[}=@var{lines}@r{]}
Use the unified output format, showing @var{lines} (an integer) lines of
context, or three if @var{lines} is not given.  @xref{Unified Format}.
For proper operation, @command{patch} typically needs at least two lines of
context.

On older systems, @command{diff} supports an obsolete option
@option{-@var{lines}} that has effect when combined with @option{-u}.
@acronym{POSIX} 1003.1-2001 (@pxref{Standards conformance}) does not allow
this; use @option{-U @var{lines}} instead.

@item -v
@itemx --version
Output version information and then exit.

@item -w
@itemx --ignore-all-space
Ignore white space when comparing lines.  @xref{White Space}.

@item -W @var{columns}
@itemx --width=@var{columns}
Output at most @var{columns} (default 130) print columns per line in
side by side format.  @xref{Side by Side Format}.

@item -x @var{pattern}
@itemx --exclude=@var{pattern}
When comparing directories, ignore files and subdirectories whose basenames
match @var{pattern}.  @xref{Comparing Directories}.

@item -X @var{file}
@itemx --exclude-from=@var{file}
When comparing directories, ignore files and subdirectories whose basenames
match any pattern contained in @var{file}.  @xref{Comparing Directories}.

@item -y
@itemx --side-by-side
Use the side by side output format.  @xref{Side by Side Format}.
@end table

@node Invoking diff3
@chapter Invoking @command{diff3}
@cindex invoking @command{diff3}
@cindex @command{diff3} invocation

The @command{diff3} command compares three files and outputs descriptions
of their differences.  Its arguments are as follows:

@example
diff3 @var{options}@dots{} @var{mine} @var{older} @var{yours}
@end example

The files to compare are @var{mine}, @var{older}, and @var{yours}.
At most one of these three file names may be @file{-},
which tells @command{diff3} to read the standard input for that file.

An exit status of 0 means @command{diff3} was successful, 1 means some
conflicts were found, and 2 means trouble.

@menu
* diff3 Options:: Summary of options to @command{diff3}.
@end menu

@node diff3 Options
@section Options to @command{diff3}
@cindex @command{diff3} options
@cindex options for @command{diff3}

Below is a summary of all of the options that @acronym{GNU} @command{diff3}
accepts.  Multiple single letter options (unless they take an argument)
can be combined into a single command line argument.

@table @option
@item -a
@itemx --text
Treat all files as text and compare them line-by-line, even if they
do not appear to be text.  @xref{Binary}.

@item -A
@itemx --show-all
Incorporate all unmerged changes from @var{older} to @var{yours} into
@var{mine}, surrounding conflicts with bracket lines.
@xref{Marking Conflicts}.

@item --diff-program=@var{program}
Use the compatible comparison program @var{program} to compare files
instead of @command{diff}.

@item -e
@itemx --ed
Generate an @command{ed} script that incorporates all the changes from
@var{older} to @var{yours} into @var{mine}.  @xref{Which Changes}.

@item -E
@itemx --show-overlap
Like @option{-e}, except bracket lines from overlapping changes' first
and third files.
@xref{Marking Conflicts}.
With @option{-E}, an overlapping change looks like this:

@example
<<<<<<< @var{mine}
@r{lines from @var{mine}}
=======
@r{lines from @var{yours}}
>>>>>>> @var{yours}
@end example

@item --help
Output a summary of usage and then exit.

@item -i
Generate @samp{w} and @samp{q} commands at the end of the @command{ed}
script for System V compatibility.  This option must be combined with
one of the @option{-AeExX3} options, and may not be combined with @option{-m}.
@xref{Saving the Changed File}.

@item --label=@var{label}
Use the label @var{label} for the brackets output by the @option{-A},
@option{-E} and @option{-X} options.  This option may be given up to three
times, one for each input file.  The default labels are the names of
the input files.  Thus @samp{diff3 --label X --label Y --label Z -m A
B C} acts like
@samp{diff3 -m A B C}, except that the output looks like it came from
files named @samp{X}, @samp{Y} and @samp{Z} rather than from files
named @samp{A}, @samp{B} and @samp{C}.  @xref{Marking Conflicts}.

@item -m
@itemx --merge
Apply the edit script to the first file and send the result to standard
output.  Unlike piping the output from @command{diff3} to @command{ed}, this
works even for binary files and incomplete lines.  @option{-A} is assumed
if no edit script option is specified.  @xref{Bypassing ed}.

@item --strip-trailing-cr
Strip any trailing carriage return at the end of an input line.
@xref{Binary}.

@item -T
@itemx --initial-tab
Output a tab rather than two spaces before the text of a line in normal format.
This causes the alignment of tabs in the line to look normal.  @xref{Tabs}.

@item -v
@itemx --version
Output version information and then exit.

@item -x
@itemx --overlap-only
Like @option{-e}, except output only the overlapping changes.
@xref{Which Changes}.

@item -X
Like @option{-E}, except output only the overlapping changes.
In other words, like @option{-x}, except bracket changes as in @option{-E}.
@xref{Marking Conflicts}.

@item -3
@itemx --easy-only
Like @option{-e}, except output only the nonoverlapping changes.
@xref{Which Changes}.
@end table

@node Invoking patch
@chapter Invoking @command{patch}
@cindex invoking @command{patch}
@cindex @command{patch} invocation

Normally @command{patch} is invoked like this:

@example
patch <@var{patchfile}
@end example

The full format for invoking @command{patch} is:

@example
patch @var{options}@dots{} @r{[}@var{origfile} @r{[}@var{patchfile}@r{]}@r{]}
@end example

You can also specify where to read the patch from with the @option{-i
@var{patchfile}} or @option{--input=@var{patchfile}} option.
If you do not specify @var{patchfile}, or if @var{patchfile} is
@file{-}, @command{patch} reads the patch (that is, the @command{diff} output)
from the standard input.

If you do not specify an input file on the command line, @command{patch}
tries to intuit from the @dfn{leading text} (any text in the patch
that comes before the @command{diff} output) which file to edit.
@xref{Multiple Patches}.

By default, @command{patch} replaces the original input file with the
patched version, possibly after renaming the original file into a
backup file (@pxref{Backup Names}, for a description of how
@command{patch} names backup files).  You can also specify where to
put the output with the @option{-o @var{file}} or
@option{--output=@var{file}} option; however, do not use this option
if @var{file} is one of the input files.

@menu
* patch Options::     Summary table of options to @command{patch}.
@end menu

@node patch Options
@section Options to @command{patch}
@cindex @command{patch} options
@cindex options for @command{patch}

Here is a summary of all of the options that @acronym{GNU} @command{patch}
accepts.  @xref{patch and Tradition}, for which of these options are
safe to use in older versions of @command{patch}.

Multiple single-letter options that do not take an argument can be
combined into a single command line argument with only one dash.

@table @option
@item -b
@itemx --backup
Back up the original contents of each file, even if backups would
normally not be made.  @xref{Backups}.

@item -B @var{prefix}
@itemx --prefix=@var{prefix}
Prepend @var{prefix} to backup file names.  @xref{Backup Names}.

@item --backup-if-mismatch
Back up the original contents of each file if the patch does not
exactly match the file.  This is the default behavior when not
conforming to @acronym{POSIX}.  @xref{Backups}.

@item --binary
Read and write all files in binary mode, except for standard output
and @file{/dev/tty}.  This option has no effect on
@acronym{POSIX}-conforming systems like @acronym{GNU}/Linux.  On systems where
this option makes a difference, the patch should be generated by
@samp{diff -a --binary}.  @xref{Binary}.

@item -c
@itemx --context
Interpret the patch file as a context diff.  @xref{patch Input}.

@item -d @var{directory}
@itemx --directory=@var{directory}
Make directory @var{directory} the current directory for interpreting
both file names in the patch file, and file names given as arguments to
other options.  @xref{patch Directories}.

@item -D @var{name}
@itemx --ifdef=@var{name}
Make merged if-then-else output using @var{name}.  @xref{If-then-else}.

@item --dry-run
Print the results of applying the patches without actually changing
any files.  @xref{Dry Runs}.

@item -e
@itemx --ed
Interpret the patch file as an @command{ed} script.  @xref{patch Input}.

@item -E
@itemx --remove-empty-files
Remove output files that are empty after the patches have been applied.
@xref{Creating and Removing}.

@item -f
@itemx --force
Assume that the user knows exactly what he or she is doing, and do not
ask any questions.  @xref{patch Messages}.

@item -F @var{lines}
@itemx --fuzz=@var{lines}
Set the maximum fuzz factor to @var{lines}.  @xref{Inexact}.

@item -g @var{num}
@itemx --get=@var{num}
If @var{num} is positive, get input files from a revision control
system as necessary; if zero, do not get the files; if negative, ask
the user whether to get the files.  @xref{Revision Control}.

@item --help
Output a summary of usage and then exit.

@item -i @var{patchfile}
@itemx --input=@var{patchfile}
Read the patch from @var{patchfile} rather than from standard input.
@xref{patch Options}.

@item -l
@itemx --ignore-white-space
Let any sequence of blanks (spaces or tabs) in the patch file match
any sequence of blanks in the input file.  @xref{Changed White Space}.

@item -n
@itemx --normal
Interpret the patch file as a normal diff.  @xref{patch Input}.

@item -N
@itemx --forward
Ignore patches that @command{patch} thinks are reversed or already applied.
See also @option{-R}.  @xref{Reversed Patches}.

@item --no-backup-if-mismatch
Do not back up the original contents of files.  This is the default
behavior when conforming to @acronym{POSIX}.  @xref{Backups}.

@item -o @var{file}
@itemx --output=@var{file}
Use @var{file} as the output file name.  @xref{patch Options}.

@item -p@var{number}
@itemx --strip=@var{number}
Set the file name strip count to @var{number}.  @xref{patch Directories}.

@item --posix
Conform to @acronym{POSIX}, as if the @env{POSIXLY_CORRECT} environment
variable had been set.  @xref{patch and POSIX}.

@item --quoting-style=@var{word}
Use style @var{word} to quote names in diagnostics, as if the
@env{QUOTING_STYLE} environment variable had been set to @var{word}.
@xref{patch Quoting Style}.

@item -r @var{reject-file}
@itemx --reject-file=@var{reject-file}
Use @var{reject-file} as the reject file name.  @xref{Reject Names}.

@item -R
@itemx --reverse
Assume that this patch was created with the old and new files swapped.
@xref{Reversed Patches}.

@item -s
@itemx --quiet
@itemx --silent
Work silently unless an error occurs.  @xref{patch Messages}.

@item -t
@itemx --batch
Do not ask any questions.  @xref{patch Messages}.

@item -T
@itemx --set-time
Set the modification and access times of patched files from time
stamps given in context diff headers, assuming that the context diff
headers use local time.  @xref{Patching Time Stamps}.

@item -u
@itemx --unified
Interpret the patch file as a unified diff.  @xref{patch Input}.

@item -v
@itemx --version
Output version information and then exit.

@item -V @var{backup-style}
@itemx --version=control=@var{backup-style}
Select the naming convention for backup file names.  @xref{Backup Names}.

@item --verbose
Print more diagnostics than usual.  @xref{patch Messages}.

@item -x @var{number}
@itemx --debug=@var{number}
Set internal debugging flags.  Of interest only to @command{patch}
patchers.

@item -Y @var{prefix}
@itemx --basename-prefix=@var{prefix}
Prepend @var{prefix} to base names of backup files.  @xref{Backup Names}.

@item -z @var{suffix}
@itemx --suffix=@var{suffix}
Use @var{suffix} as the backup extension instead of @samp{.orig} or
@samp{~}.  @xref{Backup Names}.

@item -Z
@itemx --set-utc
Set the modification and access times of patched files from time
stamps given in context diff headers, assuming that the context diff
headers use @acronym{UTC}.  @xref{Patching Time Stamps}.

@end table

@node Invoking sdiff
@chapter Invoking @command{sdiff}
@cindex invoking @command{sdiff}
@cindex @command{sdiff} invocation

The @command{sdiff} command merges two files and interactively outputs the
results.  Its arguments are as follows:

@example
sdiff -o @var{outfile} @var{options}@dots{} @var{from-file} @var{to-file}
@end example

This merges @var{from-file} with @var{to-file}, with output to @var{outfile}.
If @var{from-file} is a directory and @var{to-file} is not, @command{sdiff}
compares the file in @var{from-file} whose file name is that of @var{to-file},
and vice versa.  @var{from-file} and @var{to-file} may not both be
directories.

@command{sdiff} options begin with @samp{-}, so normally @var{from-file}
and @var{to-file} may not begin with @samp{-}.  However, @option{--} as an
argument by itself treats the remaining arguments as file names even if
they begin with @samp{-}.  You may not use @file{-} as an input file.

@command{sdiff} without @option{-o} (or @option{--output}) produces a
side-by-side difference.  This usage is obsolete; use the @option{-y}
or @option{--side-by-side} option of @command{diff} instead.

An exit status of 0 means no differences were found, 1 means some
differences were found, and 2 means trouble.

@menu
* sdiff Options:: Summary of options to @command{diff}.
@end menu

@node sdiff Options
@section Options to @command{sdiff}
@cindex @command{sdiff} options
@cindex options for @command{sdiff}

Below is a summary of all of the options that @acronym{GNU}
@command{sdiff} accepts.  Each option has two equivalent names, one of
which is a single letter preceded by @samp{-}, and the other of which
is a long name preceded by @samp{--}.  Multiple single letter options
(unless they take an argument) can be combined into a single command
line argument.  Long named options can be abbreviated to any unique
prefix of their name.

@table @option
@item -a
@itemx --text
Treat all files as text and compare them line-by-line, even if they
do not appear to be text.  @xref{Binary}.

@item -b
@itemx --ignore-space-change
Ignore changes in amount of white space.  @xref{White Space}.

@item -B
@itemx --ignore-blank-lines
Ignore changes that just insert or delete blank lines.  @xref{Blank
Lines}.

@item -d
@itemx --minimal
Change the algorithm to perhaps find a smaller set of changes.  This
makes @command{sdiff} slower (sometimes much slower).  @xref{diff
Performance}.

@item --diff-program=@var{program}
Use the compatible comparison program @var{program} to compare files
instead of @command{diff}.

@item -E
@itemx --ignore-tab-expansion
Ignore changes due to tab expansion.
@xref{White Space}.

@item --help
Output a summary of usage and then exit.

@item -i
@itemx --ignore-case
Ignore changes in case; consider upper- and lower-case to be the same.
@xref{Case Folding}.

@item -I @var{regexp}
@itemx --ignore-matching-lines=@var{regexp}
Ignore changes that just insert or delete lines that match @var{regexp}.
@xref{Specified Lines}.

@item -l
@itemx --left-column
Print only the left column of two common lines.
@xref{Side by Side Format}.

@item -o @var{file}
@itemx --output=@var{file}
Put merged output into @var{file}.  This option is required for merging.

@item -s
@itemx --suppress-common-lines
Do not print common lines.  @xref{Side by Side Format}.

@item --speed-large-files
Use heuristics to speed handling of large files that have numerous
scattered small changes.  @xref{diff Performance}.

@item --strip-trailing-cr
Strip any trailing carriage return at the end of an input line.
@xref{Binary}.

@item -t
@itemx --expand-tabs
Expand tabs to spaces in the output, to preserve the alignment of tabs
in the input files.  @xref{Tabs}.

@item --tabsize=@var{columns}
Assume that tab stops are set every @var{columns} (default 8) print
columns.  @xref{Tabs}.

@item -v
@itemx --version
Output version information and then exit.

@item -w @var{columns}
@itemx --width=@var{columns}
Output at most @var{columns} (default 130) print columns per line.
@xref{Side by Side Format}.  Note that for historical reasons, this
option is @option{-W} in @command{diff}, @option{-w} in @command{sdiff}.

@item -W
@itemx --ignore-all-space
Ignore white space when comparing lines.  @xref{White Space}.
Note that for historical reasons, this option is @option{-w} in @command{diff},
@option{-W} in @command{sdiff}.
@end table

@node Standards conformance
@chapter Standards conformance
@cindex @acronym{POSIX}

@vindex POSIXLY_CORRECT
In a few cases, the @acronym{GNU} utilities' default behavior is
incompatible with the @acronym{POSIX} standard.  To suppress these
incompatibilities, define the @env{POSIXLY_CORRECT} environment
variable.  Unless you are checking for @acronym{POSIX} conformance, you
probably do not need to define @env{POSIXLY_CORRECT}.

Normally options and operands can appear in any order, and programs act
as if all the options appear before any operands.  For example,
@samp{diff lao tzu -C 2} acts like @samp{diff -C 2 lao tzu}, since
@samp{2} is an option-argument of @option{-C}.  However, if the
@env{POSIXLY_CORRECT} environment variable is set, options must appear
before operands, unless otherwise specified for a particular command.

Newer versions of @acronym{POSIX} are occasionally incompatible with older
versions.  For example, older versions of @acronym{POSIX} allowed the
command @samp{diff -c -10} to have the same meaning as @samp{diff -C
10}, but @acronym{POSIX} 1003.1-2001 @samp{diff} no longer allows
digit-string options like @option{-10}.

@vindex _POSIX2_VERSION
The @acronym{GNU} utilities normally conform to the version of @acronym{POSIX}
that is standard for your system.  To cause them to conform to a
different version of @acronym{POSIX}, define the @env{_POSIX2_VERSION}
environment variable to a value of the form @var{yyyymm} specifying
the year and month the standard was adopted.  Two values are currently
supported for @env{_POSIX2_VERSION}: @samp{199209} stands for
@acronym{POSIX} 1003.2-1992, and @samp{200112} stands for @acronym{POSIX}
1003.1-2001.  For example, if you are running older software that
assumes an older version of @acronym{POSIX} and uses @samp{diff -c -10},
you can work around the compatibility problems by setting
@samp{_POSIX2_VERSION=199209} in your environment.

@node Projects
@chapter Future Projects

Here are some ideas for improving @acronym{GNU} @command{diff} and
@command{patch}.  The @acronym{GNU} project has identified some
improvements as potential programming projects for volunteers.  You
can also help by reporting any bugs that you find.

If you are a programmer and would like to contribute something to the
@acronym{GNU} project, please consider volunteering for one of these
projects.  If you are seriously contemplating work, please write to
@email{gvc@@gnu.org} to coordinate with other volunteers.

@menu
* Shortcomings:: Suggested projects for improvements.
* Bugs::         Reporting bugs.
@end menu

@node Shortcomings
@section Suggested Projects for Improving @acronym{GNU} @command{diff} and @command{patch}
@cindex projects for directories

One should be able to use @acronym{GNU} @command{diff} to generate a
patch from any pair of directory trees, and given the patch and a copy
of one such tree, use @command{patch} to generate a faithful copy of
the other.  Unfortunately, some changes to directory trees cannot be
expressed using current patch formats; also, @command{patch} does not
handle some of the existing formats.  These shortcomings motivate the
following suggested projects.

@menu
* Internationalization:: Handling multibyte and varying-width characters.
* Changing Structure::   Handling changes to the directory structure.
* Special Files::        Handling symbolic links, device special files, etc.
* Unusual File Names::   Handling file names that contain unusual characters.
* Time Stamp Order::     Outputting diffs in time stamp order.
* Ignoring Changes::     Ignoring certain changes while showing others.
* Speedups::             Improving performance.
@end menu

@node Internationalization
@subsection Handling Multibyte and Varying-Width Characters
@cindex multibyte characters
@cindex varying-width characters

@command{diff}, @command{diff3} and @command{sdiff} treat each line of
input as a string of unibyte characters.  This can mishandle multibyte
characters in some cases.  For example, when asked to ignore spaces,
@command{diff} does not properly ignore a multibyte space character.

Also, @command{diff} currently assumes that each byte is one column
wide, and this assumption is incorrect in some locales, e.g., locales
that use UTF-8 encoding.  This causes problems with the @option{-y} or
@option{--side-by-side} option of @command{diff}.

These problems need to be fixed without unduly affecting the
performance of the utilities in unibyte environments.

The IBM GNU/Linux Technology Center Internationalization Team has
proposed
@uref{http://oss.software.ibm.com/developer/opensource/linux/patches/i18n/diffutils-2.7.2-i18n-0.1.patch.gz,patches
to support internationalized @command{diff}}.
Unfortunately, these patches are incomplete and are to an older
version of @command{diff}, so more work needs to be done in this area.

@node Changing Structure
@subsection Handling Changes to the Directory Structure
@cindex directory structure changes

@command{diff} and @command{patch} do not handle some changes to directory
structure.  For example, suppose one directory tree contains a directory
named @samp{D} with some subsidiary files, and another contains a file
with the same name @samp{D}.  @samp{diff -r} does not output enough
information for @command{patch} to transform the directory subtree into
the file.

There should be a way to specify that a file has been removed without
having to include its entire contents in the patch file.  There should
also be a way to tell @command{patch} that a file was renamed, even if
there is no way for @command{diff} to generate such information.
There should be a way to tell @command{patch} that a file's time stamp
has changed, even if its contents have not changed.

These problems can be fixed by extending the @command{diff} output format
to represent changes in directory structure, and extending @command{patch}
to understand these extensions.

@node Special Files
@subsection Files that are Neither Directories Nor Regular Files
@cindex special files

Some files are neither directories nor regular files: they are unusual
files like symbolic links, device special files, named pipes, and
sockets.  Currently, @command{diff} treats symbolic links as if they
were the pointed-to files, except that a recursive @command{diff}
reports an error if it detects infinite loops of symbolic links (e.g.,
symbolic links to @file{..}).  @command{diff} treats other special
files like regular files if they are specified at the top level, but
simply reports their presence when comparing directories.  This means
that @command{patch} cannot represent changes to such files.  For
example, if you change which file a symbolic link points to,
@command{diff} outputs the difference between the two files, instead
of the change to the symbolic link.

@c This might not be a good idea; is it wise for root to install devices
@c this way?
@command{diff} should optionally report changes to special files specially,
and @command{patch} should be extended to understand these extensions.

@node Unusual File Names
@subsection File Names that Contain Unusual Characters
@cindex file names with unusual characters

When a file name contains an unusual character like a newline or
white space, @samp{diff -r} generates a patch that @command{patch} cannot
parse.  The problem is with format of @command{diff} output, not just with
@command{patch}, because with odd enough file names one can cause
@command{diff} to generate a patch that is syntactically correct but
patches the wrong files.  The format of @command{diff} output should be
extended to handle all possible file names.

@node Time Stamp Order
@subsection Outputting Diffs in Time Stamp Order

Applying @command{patch} to a multiple-file diff can result in files
whose time stamps are out of order.  @acronym{GNU} @command{patch} has
options to restore the time stamps of the updated files
(@pxref{Patching Time Stamps}), but sometimes it is useful to generate
a patch that works even if the recipient does not have @acronym{GNU} patch,
or does not use these options.  One way to do this would be to
implement a @command{diff} option to output diffs in time stamp order.

@node Ignoring Changes
@subsection Ignoring Certain Changes

It would be nice to have a feature for specifying two strings, one in
@var{from-file} and one in @var{to-file}, which should be considered to
match.  Thus, if the two strings are @samp{foo} and @samp{bar}, then if
two lines differ only in that @samp{foo} in file 1 corresponds to
@samp{bar} in file 2, the lines are treated as identical.

It is not clear how general this feature can or should be, or
what syntax should be used for it.

A partial substitute is to filter one or both files before comparing,
e.g.:

@example
sed 's/foo/bar/g' file1 | diff - file2
@end example

However, this outputs the filtered text, not the original.

@node Speedups
@subsection Improving Performance

When comparing two large directory structures, one of which was
originally copied from the other with time stamps preserved (e.g.,
with @samp{cp -pR}), it would greatly improve performance if an option
told @command{diff} to assume that two files with the same size and
time stamps have the same content.  @xref{diff Performance}.

@node Bugs
@section Reporting Bugs
@cindex bug reports
@cindex reporting bugs

If you think you have found a bug in @acronym{GNU} @command{cmp},
@command{diff}, @command{diff3}, or @command{sdiff}, please report it
by electronic mail to the
@uref{http://mail.gnu.org/mailman/listinfo/bug-gnu-utils,GNU utilities
bug report mailing list} @email{bug-gnu-utils@@gnu.org}.  Please send
bug reports for @acronym{GNU} @command{patch} to
@email{bug-patch@@gnu.org}.  Send as precise a description of the
problem as you can, including the output of the @option{--version}
option and sample input files that produce the bug, if applicable.  If
you have a nontrivial fix for the bug, please send it as well.  If you
have a patch, please send it too.  It may simplify the maintainer's
job if the patch is relative to a recent test release, which you can
find in the directory @uref{ftp://alpha.gnu.org/gnu/diffutils/}.

@node Copying This Manual
@appendix Copying This Manual
@include fdl.texi

@node Translations
@appendix Translations of This Manual

Nishio Futoshi of the GNUjdoc project has prepared a Japanese
translation of this manual.  Its most recent version can be found at
@uref{http://openlab.ring.gr.jp/gnujdoc/cvsweb/cvsweb.cgi/gnujdoc/}.

@node Index
@appendix Index

@printindex cp

@bye