1 This is diff.info, produced by makeinfo version 4.13 from diff.texi.
3 This manual is for GNU Diffutils (version 3.0, 15 April 2010), and
4 documents the GNU `diff', `diff3', `sdiff', and `cmp' commands for
5 showing the differences between files and the GNU `patch' command for
6 using their output to update files.
8 Copyright (C) 1992-1994, 1998, 2001-2002, 2004, 2006, 2009-2010 Free
9 Software Foundation, Inc.
11 Permission is granted to copy, distribute and/or modify this
12 document under the terms of the GNU Free Documentation License,
13 Version 1.3 or any later version published by the Free Software
14 Foundation; with no Invariant Sections, with the Front-Cover texts
15 being "A GNU Manual," and with the Back-Cover Texts as in (a)
16 below. A copy of the license is included in the section entitled
17 "GNU Free Documentation License."
19 (a) The FSF's Back-Cover Text is: "You have the freedom to copy
20 and modify this GNU manual. Buying copies from the FSF supports
21 it in developing GNU and promoting software freedom."
23 INFO-DIR-SECTION Individual utilities
25 * cmp: (diff)Invoking cmp. Compare 2 files byte by byte.
26 * diff: (diff)Invoking diff. Compare 2 files line by line.
27 * diff3: (diff)Invoking diff3. Compare 3 files line by line.
28 * patch: (diff)Invoking patch. Apply a patch to a file.
29 * sdiff: (diff)Invoking sdiff. Merge 2 files side-by-side.
32 INFO-DIR-SECTION Text creation and manipulation
34 * Diff: (diff). Comparing and merging files.
38 File: diff.info, Node: Top, Next: Overview, Up: (dir)
40 Comparing and Merging Files
41 ***************************
43 This manual is for GNU Diffutils (version 3.0, 15 April 2010), and
44 documents the GNU `diff', `diff3', `sdiff', and `cmp' commands for
45 showing the differences between files and the GNU `patch' command for
46 using their output to update files.
48 Copyright (C) 1992-1994, 1998, 2001-2002, 2004, 2006, 2009-2010 Free
49 Software Foundation, Inc.
51 Permission is granted to copy, distribute and/or modify this
52 document under the terms of the GNU Free Documentation License,
53 Version 1.3 or any later version published by the Free Software
54 Foundation; with no Invariant Sections, with the Front-Cover texts
55 being "A GNU Manual," and with the Back-Cover Texts as in (a)
56 below. A copy of the license is included in the section entitled
57 "GNU Free Documentation License."
59 (a) The FSF's Back-Cover Text is: "You have the freedom to copy
60 and modify this GNU manual. Buying copies from the FSF supports
61 it in developing GNU and promoting software freedom."
65 * Overview:: Preliminary information.
66 * Comparison:: What file comparison means.
68 * Output Formats:: Formats for two-way difference reports.
69 * Incomplete Lines:: Lines that lack trailing newlines.
70 * Comparing Directories:: Comparing files and directories.
71 * Adjusting Output:: Making `diff' output prettier.
72 * diff Performance:: Making `diff' smarter or faster.
74 * Comparing Three Files:: Formats for three-way difference reports.
75 * diff3 Merging:: Merging from a common ancestor.
77 * Interactive Merging:: Interactive merging with `sdiff'.
79 * Merging with patch:: Using `patch' to change old files into new ones.
80 * Making Patches:: Tips for making and using patch distributions.
82 * Invoking cmp:: Compare two files byte by byte.
83 * Invoking diff:: Compare two files line by line.
84 * Invoking diff3:: Compare three files line by line.
85 * Invoking patch:: Apply a diff file to an original.
86 * Invoking sdiff:: Side-by-side merge of file differences.
88 * Standards conformance:: Conformance to the POSIX standard.
89 * Projects:: If you've found a bug or other shortcoming.
91 * Copying This Manual:: How to make copies of this manual.
92 * Translations:: Available translations of this manual.
96 File: diff.info, Node: Overview, Next: Comparison, Prev: Top, Up: Top
101 Computer users often find occasion to ask how two files differ. Perhaps
102 one file is a newer version of the other file. Or maybe the two files
103 started out as identical copies but were changed by different people.
105 You can use the `diff' command to show differences between two
106 files, or each corresponding file in two directories. `diff' outputs
107 differences between files line by line in any of several formats,
108 selectable by command line options. This set of differences is often
109 called a "diff" or "patch". For files that are identical, `diff'
110 normally produces no output; for binary (non-text) files, `diff'
111 normally reports only that they are different.
113 You can use the `cmp' command to show the byte and line numbers
114 where two files differ. `cmp' can also show all the bytes that differ
115 between the two files, side by side. A way to compare two files
116 character by character is the Emacs command `M-x compare-windows'.
117 *Note Other Window: (emacs)Other Window, for more information on that
120 You can use the `diff3' command to show differences among three
121 files. When two people have made independent changes to a common
122 original, `diff3' can report the differences between the original and
123 the two changed versions, and can produce a merged file that contains
124 both persons' changes together with warnings about conflicts.
126 You can use the `sdiff' command to merge two files interactively.
128 You can use the set of differences produced by `diff' to distribute
129 updates to text files (such as program source code) to other people.
130 This method is especially useful when the differences are small compared
131 to the complete files. Given `diff' output, you can use the `patch'
132 program to update, or "patch", a copy of the file. If you think of
133 `diff' as subtracting one file from another to produce their
134 difference, you can think of `patch' as adding the difference to one
135 file to reproduce the other.
137 This manual first concentrates on making diffs, and later shows how
138 to use diffs to update files.
140 GNU `diff' was written by Paul Eggert, Mike Haertel, David Hayes,
141 Richard Stallman, and Len Tower. Wayne Davison designed and
142 implemented the unified output format. The basic algorithm is described
143 by Eugene W. Myers in "An O(ND) Difference Algorithm and its
144 Variations", `Algorithmica' Vol. 1 No. 2, 1986, pp. 251-266; and in "A
145 File Comparison Program", Webb Miller and Eugene W. Myers,
146 `Software--Practice and Experience' Vol. 15 No. 11, 1985, pp. 1025-1040.
147 The algorithm was independently discovered as described by E. Ukkonen in
148 "Algorithms for Approximate String Matching", `Information and Control'
149 Vol. 64, 1985, pp. 100-118. Unless the `--minimal' option is used,
150 `diff' uses a heuristic by Paul Eggert that limits the cost to O(N^1.5
151 log N) at the price of producing suboptimal output for large inputs
152 with many differences. Related algorithms are surveyed by Alfred V.
153 Aho in section 6.3 of "Algorithms for Finding Patterns in Strings",
154 `Handbook of Theoretical Computer Science' (Jan Van Leeuwen, ed.), Vol.
155 A, `Algorithms and Complexity', Elsevier/MIT Press, 1990, pp. 255-300.
157 GNU `diff3' was written by Randy Smith. GNU `sdiff' was written by
158 Thomas Lord. GNU `cmp' was written by Torbjo"rn Granlund and David
161 GNU `patch' was written mainly by Larry Wall and Paul Eggert;
162 several GNU enhancements were contributed by Wayne Davison and David
163 MacKenzie. Parts of this manual are adapted from a manual page written
164 by Larry Wall, with his permission.
167 File: diff.info, Node: Comparison, Next: Output Formats, Prev: Overview, Up: Top
169 1 What Comparison Means
170 ***********************
172 There are several ways to think about the differences between two files.
173 One way to think of the differences is as a series of lines that were
174 deleted from, inserted in, or changed in one file to produce the other
175 file. `diff' compares two files line by line, finds groups of lines
176 that differ, and reports each group of differing lines. It can report
177 the differing lines in several formats, which have different purposes.
179 GNU `diff' can show whether files are different without detailing
180 the differences. It also provides ways to suppress certain kinds of
181 differences that are not important to you. Most commonly, such
182 differences are changes in the amount of white space between words or
183 lines. `diff' also provides ways to suppress differences in alphabetic
184 case or in lines that match a regular expression that you provide.
185 These options can accumulate; for example, you can ignore changes in
186 both white space and alphabetic case.
188 Another way to think of the differences between two files is as a
189 sequence of pairs of bytes that can be either identical or different.
190 `cmp' reports the differences between two files byte by byte, instead
191 of line by line. As a result, it is often more useful than `diff' for
192 comparing binary files. For text files, `cmp' is useful mainly when
193 you want to know only whether two files are identical, or whether one
194 file is a prefix of the other.
196 To illustrate the effect that considering changes byte by byte can
197 have compared with considering them line by line, think of what happens
198 if a single newline character is added to the beginning of a file. If
199 that file is then compared with an otherwise identical file that lacks
200 the newline at the beginning, `diff' will report that a blank line has
201 been added to the file, while `cmp' will report that almost every byte
202 of the two files differs.
204 `diff3' normally compares three input files line by line, finds
205 groups of lines that differ, and reports each group of differing lines.
206 Its output is designed to make it easy to inspect two different sets of
207 changes to the same file.
211 * Hunks:: Groups of differing lines.
212 * White Space:: Suppressing differences in white space.
213 * Blank Lines:: Suppressing differences whose lines are all blank.
214 * Specified Lines:: Suppressing differences whose lines all match a pattern.
215 * Case Folding:: Suppressing differences in alphabetic case.
216 * Brief:: Summarizing which files are different.
217 * Binary:: Comparing binary files or forcing text comparisons.
220 File: diff.info, Node: Hunks, Next: White Space, Up: Comparison
225 When comparing two files, `diff' finds sequences of lines common to
226 both files, interspersed with groups of differing lines called "hunks".
227 Comparing two identical files yields one sequence of common lines and
228 no hunks, because no lines differ. Comparing two entirely different
229 files yields no common lines and one large hunk that contains all lines
230 of both files. In general, there are many ways to match up lines
231 between two given files. `diff' tries to minimize the total hunk size
232 by finding large sequences of common lines interspersed with small
233 hunks of differing lines.
235 For example, suppose the file `F' contains the three lines `a', `b',
236 `c', and the file `G' contains the same three lines in reverse order
237 `c', `b', `a'. If `diff' finds the line `c' as common, then the command
238 `diff F G' produces this output:
247 But if `diff' notices the common line `b' instead, it produces this
259 It is also possible to find `a' as the common line. `diff' does not
260 always find an optimal matching between the files; it takes shortcuts
261 to run faster. But its output is usually close to the shortest
262 possible. You can adjust this tradeoff with the `-d' or `--minimal'
263 option (*note diff Performance::).
266 File: diff.info, Node: White Space, Next: Blank Lines, Prev: Hunks, Up: Comparison
268 1.2 Suppressing Differences in Blank and Tab Spacing
269 ====================================================
271 The `-E' or `--ignore-tab-expansion' option ignores the distinction
272 between tabs and spaces on input. A tab is considered to be equivalent
273 to the number of spaces to the next tab stop (*note Tabs::).
275 The `-b' or `--ignore-space-change' option is stronger. It ignores
276 white space at line end, and considers all other sequences of one or
277 more white space characters within a line to be equivalent. With this
278 option, `diff' considers the following two lines to be equivalent,
279 where `$' denotes the line end:
281 Here lyeth muche rychnesse in lytell space. -- John Heywood$
282 Here lyeth muche rychnesse in lytell space. -- John Heywood $
284 The `-w' or `--ignore-all-space' option is stronger still. It
285 ignores differences even if one line has white space where the other
286 line has none. "White space" characters include tab, vertical tab,
287 form feed, carriage return, and space; some locales may define
288 additional characters to be white space. With this option, `diff'
289 considers the following two lines to be equivalent, where `$' denotes
290 the line end and `^M' denotes a carriage return:
292 Here lyeth muche rychnesse in lytell space.-- John Heywood$
293 He relyeth much erychnes seinly tells pace. --John Heywood ^M$
295 For many other programs newline is also a white space character, but
296 `diff' is a line-oriented program and a newline character always ends a
297 line. Hence the `-w' or `--ignore-all-space' option does not ignore
298 newline-related changes; it ignores only other white space changes.
301 File: diff.info, Node: Blank Lines, Next: Specified Lines, Prev: White Space, Up: Comparison
303 1.3 Suppressing Differences Whose Lines Are All Blank
304 =====================================================
306 The `-B' or `--ignore-blank-lines' option ignores changes that consist
307 entirely of blank lines. With this option, for example, a file
309 1. A point is that which has no part.
311 2. A line is breadthless length.
312 -- Euclid, The Elements, I
313 is considered identical to a file containing
314 1. A point is that which has no part.
315 2. A line is breadthless length.
318 -- Euclid, The Elements, I
320 Normally this option affects only lines that are completely empty,
321 but if you also specify the `-b' or `--ignore-space-change' option, or
322 the `-w' or `--ignore-all-space' option, lines are also affected if
323 they look empty but contain white space. In other words, `-B' is
324 equivalent to `-I '^$'' by default, but it is equivalent to `-I
325 '^[[:space:]]*$'' if `-b' or `-w' is also specified.
328 File: diff.info, Node: Specified Lines, Next: Case Folding, Prev: Blank Lines, Up: Comparison
330 1.4 Suppressing Differences Whose Lines All Match a Regular Expression
331 ======================================================================
333 To ignore insertions and deletions of lines that match a `grep'-style
334 regular expression, use the `-I REGEXP' or
335 `--ignore-matching-lines=REGEXP' option. You should escape regular
336 expressions that contain shell metacharacters to prevent the shell from
337 expanding them. For example, `diff -I '^[[:digit:]]'' ignores all
338 changes to lines beginning with a digit.
340 However, `-I' only ignores the insertion or deletion of lines that
341 contain the regular expression if every changed line in the hunk--every
342 insertion and every deletion--matches the regular expression. In other
343 words, for each nonignorable change, `diff' prints the complete set of
344 changes in its vicinity, including the ignorable ones.
346 You can specify more than one regular expression for lines to ignore
347 by using more than one `-I' option. `diff' tries to match each line
348 against each regular expression.
351 File: diff.info, Node: Case Folding, Next: Brief, Prev: Specified Lines, Up: Comparison
353 1.5 Suppressing Case Differences
354 ================================
356 GNU `diff' can treat lower case letters as equivalent to their upper
357 case counterparts, so that, for example, it considers `Funky Stuff',
358 `funky STUFF', and `fUNKy stuFf' to all be the same. To request this,
359 use the `-i' or `--ignore-case' option.
362 File: diff.info, Node: Brief, Next: Binary, Prev: Case Folding, Up: Comparison
364 1.6 Summarizing Which Files Differ
365 ==================================
367 When you only want to find out whether files are different, and you
368 don't care what the differences are, you can use the summary output
369 format. In this format, instead of showing the differences between the
370 files, `diff' simply reports whether files differ. The `-q' or
371 `--brief' option selects this output format.
373 This format is especially useful when comparing the contents of two
374 directories. It is also much faster than doing the normal line by line
375 comparisons, because `diff' can stop analyzing the files as soon as it
376 knows that there are any differences.
378 You can also get a brief indication of whether two files differ by
379 using `cmp'. For files that are identical, `cmp' produces no output.
380 When the files differ, by default, `cmp' outputs the byte and line
381 number where the first difference occurs, or reports that one file is a
382 prefix of the other. You can use the `-s', `--quiet', or `--silent'
383 option to suppress that information, so that `cmp' produces no output
384 and reports whether the files differ using only its exit status (*note
387 Unlike `diff', `cmp' cannot compare directories; it can only compare
391 File: diff.info, Node: Binary, Prev: Brief, Up: Comparison
393 1.7 Binary Files and Forcing Text Comparisons
394 =============================================
396 If `diff' thinks that either of the two files it is comparing is binary
397 (a non-text file), it normally treats that pair of files much as if the
398 summary output format had been selected (*note Brief::), and reports
399 only that the binary files are different. This is because line by line
400 comparisons are usually not meaningful for binary files.
402 `diff' determines whether a file is text or binary by checking the
403 first few bytes in the file; the exact number of bytes is system
404 dependent, but it is typically several thousand. If every byte in that
405 part of the file is non-null, `diff' considers the file to be text;
406 otherwise it considers the file to be binary.
408 Sometimes you might want to force `diff' to consider files to be
409 text. For example, you might be comparing text files that contain null
410 characters; `diff' would erroneously decide that those are non-text
411 files. Or you might be comparing documents that are in a format used
412 by a word processing system that uses null characters to indicate
413 special formatting. You can force `diff' to consider all files to be
414 text files, and compare them line by line, by using the `-a' or
415 `--text' option. If the files you compare using this option do not in
416 fact contain text, they will probably contain few newline characters,
417 and the `diff' output will consist of hunks showing differences between
418 long lines of whatever characters the files contain.
420 You can also force `diff' to report only whether files differ (but
421 not how). Use the `-q' or `--brief' option for this.
423 Normally, differing binary files count as trouble because the
424 resulting `diff' output does not capture all the differences. This
425 trouble causes `diff' to exit with status 2. However, this trouble
426 cannot occur with the `-a' or `--text' option, or with the `-q' or
427 `--brief' option, as these options both cause `diff' to generate a form
428 of output that represents differences as requested.
430 In operating systems that distinguish between text and binary files,
431 `diff' normally reads and writes all data as text. Use the `--binary'
432 option to force `diff' to read and write binary data instead. This
433 option has no effect on a POSIX-compliant system like GNU or
434 traditional Unix. However, many personal computer operating systems
435 represent the end of a line with a carriage return followed by a
436 newline. On such systems, `diff' normally ignores these carriage
437 returns on input and generates them at the end of each output line, but
438 with the `--binary' option `diff' treats each carriage return as just
439 another input character, and does not generate a carriage return at the
440 end of each output line. This can be useful when dealing with non-text
441 files that are meant to be interchanged with POSIX-compliant systems.
443 The `--strip-trailing-cr' causes `diff' to treat input lines that
444 end in carriage return followed by newline as if they end in plain
445 newline. This can be useful when comparing text that is imperfectly
446 imported from many personal computer operating systems. This option
447 affects how lines are read, which in turn affects how they are compared
450 If you want to compare two files byte by byte, you can use the `cmp'
451 program with the `-l' or `--verbose' option to show the values of each
452 differing byte in the two files. With GNU `cmp', you can also use the
453 `-b' or `--print-bytes' option to show the ASCII representation of
454 those bytes. *Note Invoking cmp::, for more information.
456 If `diff3' thinks that any of the files it is comparing is binary (a
457 non-text file), it normally reports an error, because such comparisons
458 are usually not useful. `diff3' uses the same test as `diff' to decide
459 whether a file is binary. As with `diff', if the input files contain a
460 few non-text bytes but otherwise are like text files, you can force
461 `diff3' to consider all files to be text files and compare them line by
462 line by using the `-a' or `--text' option.
465 File: diff.info, Node: Output Formats, Next: Incomplete Lines, Prev: Comparison, Up: Top
467 2 `diff' Output Formats
468 ***********************
470 `diff' has several mutually exclusive options for output format. The
471 following sections describe each format, illustrating how `diff'
472 reports the differences between two sample input files.
476 * Sample diff Input:: Sample `diff' input files for examples.
477 * Context:: Showing differences with the surrounding text.
478 * Side by Side:: Showing differences in two columns.
479 * Normal:: Showing differences without surrounding text.
480 * Scripts:: Generating scripts for other programs.
481 * If-then-else:: Merging files with if-then-else.
484 File: diff.info, Node: Sample diff Input, Next: Context, Up: Output Formats
486 2.1 Two Sample Input Files
487 ==========================
489 Here are two sample files that we will use in numerous examples to
490 illustrate the output of `diff' and how various options can change it.
492 This is the file `lao':
494 The Way that can be told of is not the eternal Way;
495 The name that can be named is not the eternal name.
496 The Nameless is the origin of Heaven and Earth;
497 The Named is the mother of all things.
498 Therefore let there always be non-being,
499 so we may see their subtlety,
500 And let there always be being,
501 so we may see their outcome.
502 The two are the same,
503 But after they are produced,
504 they have different names.
506 This is the file `tzu':
508 The Nameless is the origin of Heaven and Earth;
509 The named is the mother of all things.
511 Therefore let there always be non-being,
512 so we may see their subtlety,
513 And let there always be being,
514 so we may see their outcome.
515 The two are the same,
516 But after they are produced,
517 they have different names.
518 They both may be called deep and profound.
519 Deeper and more profound,
520 The door of all subtleties!
522 In this example, the first hunk contains just the first two lines of
523 `lao', the second hunk contains the fourth line of `lao' opposing the
524 second and third lines of `tzu', and the last hunk contains just the
525 last three lines of `tzu'.
528 File: diff.info, Node: Context, Next: Side by Side, Prev: Sample diff Input, Up: Output Formats
530 2.2 Showing Differences in Their Context
531 ========================================
533 Usually, when you are looking at the differences between files, you will
534 also want to see the parts of the files near the lines that differ, to
535 help you understand exactly what has changed. These nearby parts of the
536 files are called the "context".
538 GNU `diff' provides two output formats that show context around the
539 differing lines: "context format" and "unified format". It can
540 optionally show in which function or section of the file the differing
543 If you are distributing new versions of files to other people in the
544 form of `diff' output, you should use one of the output formats that
545 show context so that they can apply the diffs even if they have made
546 small changes of their own to the files. `patch' can apply the diffs
547 in this case by searching in the files for the lines of context around
548 the differing lines; if those lines are actually a few lines away from
549 where the diff says they are, `patch' can adjust the line numbers
550 accordingly and still apply the diff correctly. *Note Imperfect::, for
551 more information on using `patch' to apply imperfect diffs.
555 * Context Format:: An output format that shows surrounding lines.
556 * Unified Format:: A more compact output format that shows context.
557 * Sections:: Showing which sections of the files differences are in.
558 * Alternate Names:: Showing alternate file names in context headers.
561 File: diff.info, Node: Context Format, Next: Unified Format, Up: Context
566 The context output format shows several lines of context around the
567 lines that differ. It is the standard format for distributing updates
570 To select this output format, use the `-C LINES',
571 `--context[=LINES]', or `-c' option. The argument LINES that some of
572 these options take is the number of lines of context to show. If you
573 do not specify LINES, it defaults to three. For proper operation,
574 `patch' typically needs at least two lines of context.
578 * Example Context:: Sample output in context format.
579 * Less Context:: Another sample with less context.
580 * Detailed Context:: A detailed description of the context output format.
583 File: diff.info, Node: Example Context, Next: Less Context, Up: Context Format
585 2.2.1.1 An Example of Context Format
586 ....................................
588 Here is the output of `diff -c lao tzu' (*note Sample diff Input::, for
589 the complete contents of the two files). Notice that up to three lines
590 that are not different are shown around each line that is different;
591 they are the context lines. Also notice that the first two hunks have
592 run together, because their contents overlap.
594 *** lao 2002-02-21 23:30:39.942229878 -0800
595 --- tzu 2002-02-21 23:30:50.442260588 -0800
598 - The Way that can be told of is not the eternal Way;
599 - The name that can be named is not the eternal name.
600 The Nameless is the origin of Heaven and Earth;
601 ! The Named is the mother of all things.
602 Therefore let there always be non-being,
603 so we may see their subtlety,
604 And let there always be being,
606 The Nameless is the origin of Heaven and Earth;
607 ! The named is the mother of all things.
609 Therefore let there always be non-being,
610 so we may see their subtlety,
611 And let there always be being,
615 The two are the same,
616 But after they are produced,
617 they have different names.
618 + They both may be called deep and profound.
619 + Deeper and more profound,
620 + The door of all subtleties!
623 File: diff.info, Node: Less Context, Next: Detailed Context, Prev: Example Context, Up: Context Format
625 2.2.1.2 An Example of Context Format with Less Context
626 ......................................................
628 Here is the output of `diff -C 1 lao tzu' (*note Sample diff Input::,
629 for the complete contents of the two files). Notice that at most one
630 context line is reported here.
632 *** lao 2002-02-21 23:30:39.942229878 -0800
633 --- tzu 2002-02-21 23:30:50.442260588 -0800
636 - The Way that can be told of is not the eternal Way;
637 - The name that can be named is not the eternal name.
638 The Nameless is the origin of Heaven and Earth;
639 ! The Named is the mother of all things.
640 Therefore let there always be non-being,
642 The Nameless is the origin of Heaven and Earth;
643 ! The named is the mother of all things.
645 Therefore let there always be non-being,
649 they have different names.
650 + They both may be called deep and profound.
651 + Deeper and more profound,
652 + The door of all subtleties!
655 File: diff.info, Node: Detailed Context, Prev: Less Context, Up: Context Format
657 2.2.1.3 Detailed Description of Context Format
658 ..............................................
660 The context output format starts with a two-line header, which looks
663 *** FROM-FILE FROM-FILE-MODIFICATION-TIME
664 --- TO-FILE TO-FILE-MODIFICATION TIME
666 The time stamp normally looks like `2002-02-21 23:30:39.942229878
667 -0800' to indicate the date, time with fractional seconds, and time
668 zone in Internet RFC 2822 format
669 (ftp://ftp.isi.edu/in-notes/rfc2822.txt). (The fractional seconds are
670 omitted on hosts that do not support fractional time stamps.) However,
671 a traditional time stamp like `Thu Feb 21 23:30:39 2002' is used if the
672 `LC_TIME' locale category is either `C' or `POSIX'.
674 You can change the header's content with the `--label=LABEL' option;
675 see *note Alternate Names::.
677 Next come one or more hunks of differences; each hunk shows one area
678 where the files differ. Context format hunks look like this:
681 *** FROM-FILE-LINE-NUMBERS ****
684 --- TO-FILE-LINE-NUMBERS ----
688 If a hunk contains two or more lines, its line numbers look like
689 `START,END'. Otherwise only its end line number appears. An empty
690 hunk is considered to end at the line that precedes the hunk.
692 The lines of context around the lines that differ start with two
693 space characters. The lines that differ between the two files start
694 with one of the following indicator characters, followed by a space
698 A line that is part of a group of one or more lines that changed
699 between the two files. There is a corresponding group of lines
700 marked with `!' in the part of this hunk for the other file.
703 An "inserted" line in the second file that corresponds to nothing
707 A "deleted" line in the first file that corresponds to nothing in
710 If all of the changes in a hunk are insertions, the lines of
711 FROM-FILE are omitted. If all of the changes are deletions, the lines
712 of TO-FILE are omitted.
715 File: diff.info, Node: Unified Format, Next: Sections, Prev: Context Format, Up: Context
720 The unified output format is a variation on the context format that is
721 more compact because it omits redundant context lines. To select this
722 output format, use the `-U LINES', `--unified[=LINES]', or `-u' option.
723 The argument LINES is the number of lines of context to show. When it
724 is not given, it defaults to three.
726 At present, only GNU `diff' can produce this format and only GNU
727 `patch' can automatically apply diffs in this format. For proper
728 operation, `patch' typically needs at least three lines of context.
732 * Example Unified:: Sample output in unified format.
733 * Detailed Unified:: A detailed description of unified format.
736 File: diff.info, Node: Example Unified, Next: Detailed Unified, Up: Unified Format
738 2.2.2.1 An Example of Unified Format
739 ....................................
741 Here is the output of the command `diff -u lao tzu' (*note Sample diff
742 Input::, for the complete contents of the two files):
744 --- lao 2002-02-21 23:30:39.942229878 -0800
745 +++ tzu 2002-02-21 23:30:50.442260588 -0800
747 -The Way that can be told of is not the eternal Way;
748 -The name that can be named is not the eternal name.
749 The Nameless is the origin of Heaven and Earth;
750 -The Named is the mother of all things.
751 +The named is the mother of all things.
753 Therefore let there always be non-being,
754 so we may see their subtlety,
755 And let there always be being,
757 The two are the same,
758 But after they are produced,
759 they have different names.
760 +They both may be called deep and profound.
761 +Deeper and more profound,
762 +The door of all subtleties!
765 File: diff.info, Node: Detailed Unified, Prev: Example Unified, Up: Unified Format
767 2.2.2.2 Detailed Description of Unified Format
768 ..............................................
770 The unified output format starts with a two-line header, which looks
773 --- FROM-FILE FROM-FILE-MODIFICATION-TIME
774 +++ TO-FILE TO-FILE-MODIFICATION-TIME
776 The time stamp looks like `2002-02-21 23:30:39.942229878 -0800' to
777 indicate the date, time with fractional seconds, and time zone. The
778 fractional seconds are omitted on hosts that do not support fractional
781 You can change the header's content with the `--label=LABEL' option;
782 see *Note Alternate Names::.
784 Next come one or more hunks of differences; each hunk shows one area
785 where the files differ. Unified format hunks look like this:
787 @@ FROM-FILE-LINE-NUMBERS TO-FILE-LINE-NUMBERS @@
788 LINE-FROM-EITHER-FILE
789 LINE-FROM-EITHER-FILE...
791 If a hunk contains just one line, only its start line number appears.
792 Otherwise its line numbers look like `START,COUNT'. An empty hunk is
793 considered to start at the line that follows the hunk.
795 If a hunk and its context contain two or more lines, its line
796 numbers look like `START,COUNT'. Otherwise only its end line number
797 appears. An empty hunk is considered to end at the line that precedes
800 The lines common to both files begin with a space character. The
801 lines that actually differ between the two files have one of the
802 following indicator characters in the left print column:
805 A line was added here to the first file.
808 A line was removed here from the first file.
811 File: diff.info, Node: Sections, Next: Alternate Names, Prev: Unified Format, Up: Context
813 2.2.3 Showing Which Sections Differences Are in
814 -----------------------------------------------
816 Sometimes you might want to know which part of the files each change
817 falls in. If the files are source code, this could mean which function
818 was changed. If the files are documents, it could mean which chapter
819 or appendix was changed. GNU `diff' can show this by displaying the
820 nearest section heading line that precedes the differing lines. Which
821 lines are "section headings" is determined by a regular expression.
825 * Specified Headings:: Showing headings that match regular expressions.
826 * C Function Headings:: Showing headings of C functions.
829 File: diff.info, Node: Specified Headings, Next: C Function Headings, Up: Sections
831 2.2.3.1 Showing Lines That Match Regular Expressions
832 ....................................................
834 To show in which sections differences occur for files that are not
835 source code for C or similar languages, use the `-F REGEXP' or
836 `--show-function-line=REGEXP' option. `diff' considers lines that
837 match the `grep'-style regular expression REGEXP to be the beginning of
838 a section of the file. Here are suggested regular expressions for some
850 This option does not automatically select an output format; in order
851 to use it, you must select the context format (*note Context Format::)
852 or unified format (*note Unified Format::). In other output formats it
855 The `-F' or `--show-function-line' option finds the nearest
856 unchanged line that precedes each hunk of differences and matches the
857 given regular expression. Then it adds that line to the end of the
858 line of asterisks in the context format, or to the `@@' line in unified
859 format. If no matching line exists, this option leaves the output for
860 that hunk unchanged. If that line is more than 40 characters long, it
861 outputs only the first 40 characters. You can specify more than one
862 regular expression for such lines; `diff' tries to match each line
863 against each regular expression, starting with the last one given. This
864 means that you can use `-p' and `-F' together, if you wish.
867 File: diff.info, Node: C Function Headings, Prev: Specified Headings, Up: Sections
869 2.2.3.2 Showing C Function Headings
870 ...................................
872 To show in which functions differences occur for C and similar
873 languages, you can use the `-p' or `--show-c-function' option. This
874 option automatically defaults to the context output format (*note
875 Context Format::), with the default number of lines of context. You
876 can override that number with `-C LINES' elsewhere in the command line.
877 You can override both the format and the number with `-U LINES'
878 elsewhere in the command line.
880 The `-p' or `--show-c-function' option is equivalent to `-F
881 '^[[:alpha:]$_]'' if the unified format is specified, otherwise `-c -F
882 '^[[:alpha:]$_]'' (*note Specified Headings::). GNU `diff' provides
883 this option for the sake of convenience.
886 File: diff.info, Node: Alternate Names, Prev: Sections, Up: Context
888 2.2.4 Showing Alternate File Names
889 ----------------------------------
891 If you are comparing two files that have meaningless or uninformative
892 names, you might want `diff' to show alternate names in the header of
893 the context and unified output formats. To do this, use the
894 `--label=LABEL' option. The first time you give this option, its
895 argument replaces the name and date of the first file in the header;
896 the second time, its argument replaces the name and date of the second
897 file. If you give this option more than twice, `diff' reports an
898 error. The `--label' option does not affect the file names in the `pr'
899 header when the `-l' or `--paginate' option is used (*note
902 Here are the first two lines of the output from `diff -C 2
903 --label=original --label=modified lao tzu':
909 File: diff.info, Node: Side by Side, Next: Normal, Prev: Context, Up: Output Formats
911 2.3 Showing Differences Side by Side
912 ====================================
914 `diff' can produce a side by side difference listing of two files. The
915 files are listed in two columns with a gutter between them. The gutter
916 contains one of the following markers:
919 The corresponding lines are in common. That is, either the lines
920 are identical, or the difference is ignored because of one of the
921 `--ignore' options (*note White Space::).
924 The corresponding lines differ, and they are either both complete
928 The files differ and only the first file contains the line.
931 The files differ and only the second file contains the line.
934 Only the first file contains the line, but the difference is
938 Only the second file contains the line, but the difference is
942 The corresponding lines differ, and only the first line is
946 The corresponding lines differ, and only the second line is
949 Normally, an output line is incomplete if and only if the lines that
950 it contains are incomplete; *Note Incomplete Lines::. However, when an
951 output line represents two differing lines, one might be incomplete
952 while the other is not. In this case, the output line is complete, but
953 its the gutter is marked `\' if the first line is incomplete, `/' if
956 Side by side format is sometimes easiest to read, but it has
957 limitations. It generates much wider output than usual, and truncates
958 lines that are too long to fit. Also, it relies on lining up output
959 more heavily than usual, so its output looks particularly bad if you
960 use varying width fonts, nonstandard tab stops, or nonprinting
963 You can use the `sdiff' command to interactively merge side by side
964 differences. *Note Interactive Merging::, for more information on
969 * Side by Side Format:: Controlling side by side output format.
970 * Example Side by Side:: Sample side by side output.
973 File: diff.info, Node: Side by Side Format, Next: Example Side by Side, Up: Side by Side
975 2.3.1 Controlling Side by Side Format
976 -------------------------------------
978 The `-y' or `--side-by-side' option selects side by side format.
979 Because side by side output lines contain two input lines, the output
980 is wider than usual: normally 130 print columns, which can fit onto a
981 traditional printer line. You can set the width of the output with the
982 `-W COLUMNS' or `--width=COLUMNS' option. The output is split into two
983 halves of equal width, separated by a small gutter to mark differences;
984 the right half is aligned to a tab stop so that tabs line up. Input
985 lines that are too long to fit in half of an output line are truncated
988 The `--left-column' option prints only the left column of two common
989 lines. The `--suppress-common-lines' option suppresses common lines
993 File: diff.info, Node: Example Side by Side, Prev: Side by Side Format, Up: Side by Side
995 2.3.2 An Example of Side by Side Format
996 ---------------------------------------
998 Here is the output of the command `diff -y -W 72 lao tzu' (*note Sample
999 diff Input::, for the complete contents of the two files).
1001 The Way that can be told of is n <
1002 The name that can be named is no <
1003 The Nameless is the origin of He The Nameless is the origin of He
1004 The Named is the mother of all t | The named is the mother of all t
1006 Therefore let there always be no Therefore let there always be no
1007 so we may see their subtlety, so we may see their subtlety,
1008 And let there always be being, And let there always be being,
1009 so we may see their outcome. so we may see their outcome.
1010 The two are the same, The two are the same,
1011 But after they are produced, But after they are produced,
1012 they have different names. they have different names.
1013 > They both may be called deep and
1014 > Deeper and more profound,
1015 > The door of all subtleties!
1018 File: diff.info, Node: Normal, Next: Scripts, Prev: Side by Side, Up: Output Formats
1020 2.4 Showing Differences Without Context
1021 =======================================
1023 The "normal" `diff' output format shows each hunk of differences
1024 without any surrounding context. Sometimes such output is the clearest
1025 way to see how lines have changed, without the clutter of nearby
1026 unchanged lines (although you can get similar results with the context
1027 or unified formats by using 0 lines of context). However, this format
1028 is no longer widely used for sending out patches; for that purpose, the
1029 context format (*note Context Format::) and the unified format (*note
1030 Unified Format::) are superior. Normal format is the default for
1031 compatibility with older versions of `diff' and the POSIX standard.
1032 Use the `--normal' option to select this output format explicitly.
1036 * Example Normal:: Sample output in the normal format.
1037 * Detailed Normal:: A detailed description of normal output format.
1040 File: diff.info, Node: Example Normal, Next: Detailed Normal, Up: Normal
1042 2.4.1 An Example of Normal Format
1043 ---------------------------------
1045 Here is the output of the command `diff lao tzu' (*note Sample diff
1046 Input::, for the complete contents of the two files). Notice that it
1047 shows only the lines that are different between the two files.
1050 < The Way that can be told of is not the eternal Way;
1051 < The name that can be named is not the eternal name.
1053 < The Named is the mother of all things.
1055 > The named is the mother of all things.
1058 > They both may be called deep and profound.
1059 > Deeper and more profound,
1060 > The door of all subtleties!
1063 File: diff.info, Node: Detailed Normal, Prev: Example Normal, Up: Normal
1065 2.4.2 Detailed Description of Normal Format
1066 -------------------------------------------
1068 The normal output format consists of one or more hunks of differences;
1069 each hunk shows one area where the files differ. Normal format hunks
1079 There are three types of change commands. Each consists of a line
1080 number or comma-separated range of lines in the first file, a single
1081 character indicating the kind of change to make, and a line number or
1082 comma-separated range of lines in the second file. All line numbers are
1083 the original line numbers in each file. The types of change commands
1087 Add the lines in range R of the second file after line L of the
1088 first file. For example, `8a12,15' means append lines 12-15 of
1089 file 2 after line 8 of file 1; or, if changing file 2 into file 1,
1090 delete lines 12-15 of file 2.
1093 Replace the lines in range F of the first file with lines in range
1094 T of the second file. This is like a combined add and delete, but
1095 more compact. For example, `5,7c8,10' means change lines 5-7 of
1096 file 1 to read as lines 8-10 of file 2; or, if changing file 2 into
1097 file 1, change lines 8-10 of file 2 to read as lines 5-7 of file 1.
1100 Delete the lines in range R from the first file; line L is where
1101 they would have appeared in the second file had they not been
1102 deleted. For example, `5,7d3' means delete lines 5-7 of file 1;
1103 or, if changing file 2 into file 1, append lines 5-7 of file 1
1104 after line 3 of file 2.
1107 File: diff.info, Node: Scripts, Next: If-then-else, Prev: Normal, Up: Output Formats
1109 2.5 Making Edit Scripts
1110 =======================
1112 Several output modes produce command scripts for editing FROM-FILE to
1117 * ed Scripts:: Using `diff' to produce commands for `ed'.
1118 * Forward ed:: Making forward `ed' scripts.
1119 * RCS:: A special `diff' output format used by RCS.
1122 File: diff.info, Node: ed Scripts, Next: Forward ed, Up: Scripts
1127 `diff' can produce commands that direct the `ed' text editor to change
1128 the first file into the second file. Long ago, this was the only
1129 output mode that was suitable for editing one file into another
1130 automatically; today, with `patch', it is almost obsolete. Use the
1131 `-e' or `--ed' option to select this output format.
1133 Like the normal format (*note Normal::), this output format does not
1134 show any context; unlike the normal format, it does not include the
1135 information necessary to apply the diff in reverse (to produce the first
1136 file if all you have is the second file and the diff).
1138 If the file `d' contains the output of `diff -e old new', then the
1139 command `(cat d && echo w) | ed - old' edits `old' to make it a copy of
1140 `new'. More generally, if `d1', `d2', ..., `dN' contain the outputs of
1141 `diff -e old new1', `diff -e new1 new2', ..., `diff -e newN-1 newN',
1142 respectively, then the command `(cat d1 d2 ... dN && echo w) | ed -
1143 old' edits `old' to make it a copy of `newN'.
1147 * Example ed:: A sample `ed' script.
1148 * Detailed ed:: A detailed description of `ed' format.
1151 File: diff.info, Node: Example ed, Next: Detailed ed, Up: ed Scripts
1153 2.5.1.1 Example `ed' Script
1154 ...........................
1156 Here is the output of `diff -e lao tzu' (*note Sample diff Input::, for
1157 the complete contents of the two files):
1160 They both may be called deep and profound.
1161 Deeper and more profound,
1162 The door of all subtleties!
1165 The named is the mother of all things.
1171 File: diff.info, Node: Detailed ed, Prev: Example ed, Up: ed Scripts
1173 2.5.1.2 Detailed Description of `ed' Format
1174 ...........................................
1176 The `ed' output format consists of one or more hunks of differences.
1177 The changes closest to the ends of the files come first so that
1178 commands that change the number of lines do not affect how `ed'
1179 interprets line numbers in succeeding commands. `ed' format hunks look
1187 Because `ed' uses a single period on a line to indicate the end of
1188 input, GNU `diff' protects lines of changes that contain a single
1189 period on a line by writing two periods instead, then writing a
1190 subsequent `ed' command to change the two periods into one. The `ed'
1191 format cannot represent an incomplete line, so if the second file ends
1192 in a changed incomplete line, `diff' reports an error and then pretends
1193 that a newline was appended.
1195 There are three types of change commands. Each consists of a line
1196 number or comma-separated range of lines in the first file and a single
1197 character indicating the kind of change to make. All line numbers are
1198 the original line numbers in the file. The types of change commands
1202 Add text from the second file after line L in the first file. For
1203 example, `8a' means to add the following lines after line 8 of file
1207 Replace the lines in range R in the first file with the following
1208 lines. Like a combined add and delete, but more compact. For
1209 example, `5,7c' means change lines 5-7 of file 1 to read as the
1213 Delete the lines in range R from the first file. For example,
1214 `5,7d' means delete lines 5-7 of file 1.
1217 File: diff.info, Node: Forward ed, Next: RCS, Prev: ed Scripts, Up: Scripts
1219 2.5.2 Forward `ed' Scripts
1220 --------------------------
1222 `diff' can produce output that is like an `ed' script, but with hunks
1223 in forward (front to back) order. The format of the commands is also
1224 changed slightly: command characters precede the lines they modify,
1225 spaces separate line numbers in ranges, and no attempt is made to
1226 disambiguate hunk lines consisting of a single period. Like `ed'
1227 format, forward `ed' format cannot represent incomplete lines.
1229 Forward `ed' format is not very useful, because neither `ed' nor
1230 `patch' can apply diffs in this format. It exists mainly for
1231 compatibility with older versions of `diff'. Use the `-f' or
1232 `--forward-ed' option to select it.
1235 File: diff.info, Node: RCS, Prev: Forward ed, Up: Scripts
1240 The RCS output format is designed specifically for use by the Revision
1241 Control System, which is a set of free programs used for organizing
1242 different versions and systems of files. Use the `-n' or `--rcs'
1243 option to select this output format. It is like the forward `ed'
1244 format (*note Forward ed::), but it can represent arbitrary changes to
1245 the contents of a file because it avoids the forward `ed' format's
1246 problems with lines consisting of a single period and with incomplete
1247 lines. Instead of ending text sections with a line consisting of a
1248 single period, each command specifies the number of lines it affects; a
1249 combination of the `a' and `d' commands are used instead of `c'. Also,
1250 if the second file ends in a changed incomplete line, then the output
1251 also ends in an incomplete line.
1253 Here is the output of `diff -n lao tzu' (*note Sample diff Input::,
1254 for the complete contents of the two files):
1259 The named is the mother of all things.
1262 They both may be called deep and profound.
1263 Deeper and more profound,
1264 The door of all subtleties!
1267 File: diff.info, Node: If-then-else, Prev: Scripts, Up: Output Formats
1269 2.6 Merging Files with If-then-else
1270 ===================================
1272 You can use `diff' to merge two files of C source code. The output of
1273 `diff' in this format contains all the lines of both files. Lines
1274 common to both files are output just once; the differing parts are
1275 separated by the C preprocessor directives `#ifdef NAME' or `#ifndef
1276 NAME', `#else', and `#endif'. When compiling the output, you select
1277 which version to use by either defining or leaving undefined the macro
1280 To merge two files, use `diff' with the `-D NAME' or `--ifdef=NAME'
1281 option. The argument NAME is the C preprocessor identifier to use in
1282 the `#ifdef' and `#ifndef' directives.
1284 For example, if you change an instance of `wait (&s)' to `waitpid
1285 (-1, &s, 0)' and then merge the old and new files with the
1286 `--ifdef=HAVE_WAITPID' option, then the affected part of your code
1287 might look like this:
1290 #ifndef HAVE_WAITPID
1291 if ((w = wait (&s)) < 0 && errno != EINTR)
1292 #else /* HAVE_WAITPID */
1293 if ((w = waitpid (-1, &s, 0)) < 0 && errno != EINTR)
1294 #endif /* HAVE_WAITPID */
1296 } while (w != child);
1298 You can specify formats for languages other than C by using line
1299 group formats and line formats, as described in the next sections.
1303 * Line Group Formats:: Formats for general if-then-else line groups.
1304 * Line Formats:: Formats for each line in a line group.
1305 * Example If-then-else:: Sample if-then-else format output.
1306 * Detailed If-then-else:: A detailed description of if-then-else format.
1309 File: diff.info, Node: Line Group Formats, Next: Line Formats, Up: If-then-else
1311 2.6.1 Line Group Formats
1312 ------------------------
1314 Line group formats let you specify formats suitable for many
1315 applications that allow if-then-else input, including programming
1316 languages and text formatting languages. A line group format specifies
1317 the output format for a contiguous group of similar lines.
1319 For example, the following command compares the TeX files `old' and
1320 `new', and outputs a merged file in which old regions are surrounded by
1321 `\begin{em}'-`\end{em}' lines, and new regions are surrounded by
1322 `\begin{bf}'-`\end{bf}' lines.
1325 --old-group-format='\begin{em}
1328 --new-group-format='\begin{bf}
1333 The following command is equivalent to the above example, but it is a
1334 little more verbose, because it spells out the default line group
1338 --old-group-format='\begin{em}
1341 --new-group-format='\begin{bf}
1344 --unchanged-group-format='%=' \
1345 --changed-group-format='\begin{em}
1352 Here is a more advanced example, which outputs a diff listing with
1353 headers containing line numbers in a "plain English" style.
1356 --unchanged-group-format='' \
1357 --old-group-format='-------- %dn line%(n=1?:s) deleted at %df:
1359 --new-group-format='-------- %dN line%(N=1?:s) added after %de:
1361 --changed-group-format='-------- %dn line%(n=1?:s) changed at %df:
1366 To specify a line group format, use `diff' with one of the options
1367 listed below. You can specify up to four line group formats, one for
1368 each kind of line group. You should quote FORMAT, because it typically
1369 contains shell metacharacters.
1371 `--old-group-format=FORMAT'
1372 These line groups are hunks containing only lines from the first
1373 file. The default old group format is the same as the changed
1374 group format if it is specified; otherwise it is a format that
1375 outputs the line group as-is.
1377 `--new-group-format=FORMAT'
1378 These line groups are hunks containing only lines from the second
1379 file. The default new group format is same as the changed group
1380 format if it is specified; otherwise it is a format that outputs
1381 the line group as-is.
1383 `--changed-group-format=FORMAT'
1384 These line groups are hunks containing lines from both files. The
1385 default changed group format is the concatenation of the old and
1388 `--unchanged-group-format=FORMAT'
1389 These line groups contain lines common to both files. The default
1390 unchanged group format is a format that outputs the line group
1393 In a line group format, ordinary characters represent themselves;
1394 conversion specifications start with `%' and have one of the following
1398 stands for the lines from the first file, including the trailing
1399 newline. Each line is formatted according to the old line format
1400 (*note Line Formats::).
1403 stands for the lines from the second file, including the trailing
1404 newline. Each line is formatted according to the new line format.
1407 stands for the lines common to both files, including the trailing
1408 newline. Each line is formatted according to the unchanged line
1415 where C is a single character, stands for C. C may not be a
1416 backslash or an apostrophe. For example, `%c':'' stands for a
1417 colon, even inside the then-part of an if-then-else format, which
1418 a colon would normally terminate.
1421 where O is a string of 1, 2, or 3 octal digits, stands for the
1422 character with octal code O. For example, `%c'\0'' stands for a
1426 where F is a `printf' conversion specification and N is one of the
1427 following letters, stands for N's value formatted with F.
1430 The line number of the line just before the group in the old
1434 The line number of the first line in the group in the old
1438 The line number of the last line in the group in the old file.
1441 The line number of the line just after the group in the old
1445 The number of lines in the group in the old file; equals L -
1449 Likewise, for lines in the new file.
1452 The `printf' conversion specification can be `%d', `%o', `%x', or
1453 `%X', specifying decimal, octal, lower case hexadecimal, or upper
1454 case hexadecimal output respectively. After the `%' the following
1455 options can appear in sequence: a series of zero or more flags; an
1456 integer specifying the minimum field width; and a period followed
1457 by an optional integer specifying the minimum number of digits.
1458 The flags are `-' for left-justification, `'' for separating the
1459 digit into groups as specified by the `LC_NUMERIC' locale category,
1460 and `0' for padding with zeros instead of spaces. For example,
1461 `%5dN' prints the number of new lines in the group in a field of
1462 width 5 characters, using the `printf' format `"%5d"'.
1465 If A equals B then T else E. A and B are each either a decimal
1466 constant or a single letter interpreted as above. This format
1467 spec is equivalent to T if A's value equals B's; otherwise it is
1470 For example, `%(N=0?no:%dN) line%(N=1?:s)' is equivalent to `no
1471 lines' if N (the number of lines in the group in the new file) is
1472 0, to `1 line' if N is 1, and to `%dN lines' otherwise.
1475 File: diff.info, Node: Line Formats, Next: Example If-then-else, Prev: Line Group Formats, Up: If-then-else
1480 Line formats control how each line taken from an input file is output
1481 as part of a line group in if-then-else format.
1483 For example, the following command outputs text with a one-character
1484 change indicator to the left of the text. The first character of output
1485 is `-' for deleted lines, `|' for added lines, and a space for
1486 unchanged lines. The formats contain newline characters where newlines
1487 are desired on output.
1490 --old-line-format='-%l
1492 --new-line-format='|%l
1494 --unchanged-line-format=' %l
1498 To specify a line format, use one of the following options. You
1499 should quote FORMAT, since it often contains shell metacharacters.
1501 `--old-line-format=FORMAT'
1502 formats lines just from the first file.
1504 `--new-line-format=FORMAT'
1505 formats lines just from the second file.
1507 `--unchanged-line-format=FORMAT'
1508 formats lines common to both files.
1510 `--line-format=FORMAT'
1511 formats all lines; in effect, it sets all three above options
1514 In a line format, ordinary characters represent themselves;
1515 conversion specifications start with `%' and have one of the following
1519 stands for the contents of the line, not counting its trailing
1520 newline (if any). This format ignores whether the line is
1521 incomplete; *Note Incomplete Lines::.
1524 stands for the contents of the line, including its trailing newline
1525 (if any). If a line is incomplete, this format preserves its
1532 where C is a single character, stands for C. C may not be a
1533 backslash or an apostrophe. For example, `%c':'' stands for a
1537 where O is a string of 1, 2, or 3 octal digits, stands for the
1538 character with octal code O. For example, `%c'\0'' stands for a
1542 where F is a `printf' conversion specification, stands for the
1543 line number formatted with F. For example, `%.5dn' prints the
1544 line number using the `printf' format `"%.5d"'. *Note Line Group
1545 Formats::, for more about printf conversion specifications.
1548 The default line format is `%l' followed by a newline character.
1550 If the input contains tab characters and it is important that they
1551 line up on output, you should ensure that `%l' or `%L' in a line format
1552 is just after a tab stop (e.g. by preceding `%l' or `%L' with a tab
1553 character), or you should use the `-t' or `--expand-tabs' option.
1555 Taken together, the line and line group formats let you specify many
1556 different formats. For example, the following command uses a format
1557 similar to normal `diff' format. You can tailor this command to get
1558 fine control over `diff' output.
1561 --old-line-format='< %l
1563 --new-line-format='> %l
1565 --old-group-format='%df%(f=l?:,%dl)d%dE
1567 --new-group-format='%dea%dF%(F=L?:,%dL)
1569 --changed-group-format='%df%(f=l?:,%dl)c%dF%(F=L?:,%dL)
1572 --unchanged-group-format='' \
1576 File: diff.info, Node: Example If-then-else, Next: Detailed If-then-else, Prev: Line Formats, Up: If-then-else
1578 2.6.3 An Example of If-then-else Format
1579 ---------------------------------------
1581 Here is the output of `diff -DTWO lao tzu' (*note Sample diff Input::,
1582 for the complete contents of the two files):
1585 The Way that can be told of is not the eternal Way;
1586 The name that can be named is not the eternal name.
1588 The Nameless is the origin of Heaven and Earth;
1590 The Named is the mother of all things.
1592 The named is the mother of all things.
1595 Therefore let there always be non-being,
1596 so we may see their subtlety,
1597 And let there always be being,
1598 so we may see their outcome.
1599 The two are the same,
1600 But after they are produced,
1601 they have different names.
1603 They both may be called deep and profound.
1604 Deeper and more profound,
1605 The door of all subtleties!
1609 File: diff.info, Node: Detailed If-then-else, Prev: Example If-then-else, Up: If-then-else
1611 2.6.4 Detailed Description of If-then-else Format
1612 -------------------------------------------------
1614 For lines common to both files, `diff' uses the unchanged line group
1615 format. For each hunk of differences in the merged output format, if
1616 the hunk contains only lines from the first file, `diff' uses the old
1617 line group format; if the hunk contains only lines from the second
1618 file, `diff' uses the new group format; otherwise, `diff' uses the
1619 changed group format.
1621 The old, new, and unchanged line formats specify the output format of
1622 lines from the first file, lines from the second file, and lines common
1623 to both files, respectively.
1625 The option `--ifdef=NAME' is equivalent to the following sequence of
1626 options using shell syntax:
1628 --old-group-format='#ifndef NAME
1629 %<#endif /* ! NAME */
1631 --new-group-format='#ifdef NAME
1634 --unchanged-group-format='%=' \
1635 --changed-group-format='#ifndef NAME
1640 You should carefully check the `diff' output for proper nesting.
1641 For example, when using the `-D NAME' or `--ifdef=NAME' option, you
1642 should check that if the differing lines contain any of the C
1643 preprocessor directives `#ifdef', `#ifndef', `#else', `#elif', or
1644 `#endif', they are nested properly and match. If they don't, you must
1645 make corrections manually. It is a good idea to carefully check the
1646 resulting code anyway to make sure that it really does what you want it
1647 to; depending on how the input files were produced, the output might
1648 contain duplicate or otherwise incorrect code.
1650 The `patch' `-D NAME' option behaves like the `diff' `-D NAME'
1651 option, except it operates on a file and a diff to produce a merged
1652 file; *Note patch Options::.
1655 File: diff.info, Node: Incomplete Lines, Next: Comparing Directories, Prev: Output Formats, Up: Top
1660 When an input file ends in a non-newline character, its last line is
1661 called an "incomplete line" because its last character is not a
1662 newline. All other lines are called "full lines" and end in a newline
1663 character. Incomplete lines do not match full lines unless differences
1664 in white space are ignored (*note White Space::).
1666 An incomplete line is normally distinguished on output from a full
1667 line by a following line that starts with `\'. However, the RCS format
1668 (*note RCS::) outputs the incomplete line as-is, without any trailing
1669 newline or following line. The side by side format normally represents
1670 incomplete lines as-is, but in some cases uses a `\' or `/' gutter
1671 marker; *Note Side by Side::. The if-then-else line format preserves a
1672 line's incompleteness with `%L', and discards the newline with `%l';
1673 *Note Line Formats::. Finally, with the `ed' and forward `ed' output
1674 formats (*note Output Formats::) `diff' cannot represent an incomplete
1675 line, so it pretends there was a newline and reports an error.
1677 For example, suppose `F' and `G' are one-byte files that contain
1678 just `f' and `g', respectively. Then `diff F G' outputs
1682 \ No newline at end of file
1685 \ No newline at end of file
1687 (The exact message may differ in non-English locales.) `diff -n F G'
1688 outputs the following without a trailing newline:
1694 `diff -e F G' reports two errors and outputs the following:
1701 File: diff.info, Node: Comparing Directories, Next: Adjusting Output, Prev: Incomplete Lines, Up: Top
1703 4 Comparing Directories
1704 ***********************
1706 You can use `diff' to compare some or all of the files in two directory
1707 trees. When both file name arguments to `diff' are directories, it
1708 compares each file that is contained in both directories, examining
1709 file names in alphabetical order as specified by the `LC_COLLATE'
1710 locale category. Normally `diff' is silent about pairs of files that
1711 contain no differences, but if you use the `-s' or
1712 `--report-identical-files' option, it reports pairs of identical files.
1713 Normally `diff' reports subdirectories common to both directories
1714 without comparing subdirectories' files, but if you use the `-r' or
1715 `--recursive' option, it compares every corresponding pair of files in
1716 the directory trees, as many levels deep as they go.
1718 For file names that are in only one of the directories, `diff'
1719 normally does not show the contents of the file that exists; it reports
1720 only that the file exists in that directory and not in the other. You
1721 can make `diff' act as though the file existed but was empty in the
1722 other directory, so that it outputs the entire contents of the file that
1723 actually exists. (It is output as either an insertion or a deletion,
1724 depending on whether it is in the first or the second directory given.)
1725 To do this, use the `-N' or `--new-file' option.
1727 If the older directory contains one or more large files that are not
1728 in the newer directory, you can make the patch smaller by using the
1729 `--unidirectional-new-file' option instead of `-N'. This option is
1730 like `-N' except that it only inserts the contents of files that appear
1731 in the second directory but not the first (that is, files that were
1732 added). At the top of the patch, write instructions for the user
1733 applying the patch to remove the files that were deleted before
1734 applying the patch. *Note Making Patches::, for more discussion of
1735 making patches for distribution.
1737 To ignore some files while comparing directories, use the `-x
1738 PATTERN' or `--exclude=PATTERN' option. This option ignores any files
1739 or subdirectories whose base names match the shell pattern PATTERN.
1740 Unlike in the shell, a period at the start of the base of a file name
1741 matches a wildcard at the start of a pattern. You should enclose
1742 PATTERN in quotes so that the shell does not expand it. For example,
1743 the option `-x '*.[ao]'' ignores any file whose name ends with `.a' or
1746 This option accumulates if you specify it more than once. For
1747 example, using the options `-x 'RCS' -x '*,v'' ignores any file or
1748 subdirectory whose base name is `RCS' or ends with `,v'.
1750 If you need to give this option many times, you can instead put the
1751 patterns in a file, one pattern per line, and use the `-X FILE' or
1752 `--exclude-from=FILE' option. Trailing white space and empty lines are
1753 ignored in the pattern file.
1755 If you have been comparing two directories and stopped partway
1756 through, later you might want to continue where you left off. You can
1757 do this by using the `-S FILE' or `--starting-file=FILE' option. This
1758 compares only the file FILE and all alphabetically later files in the
1759 topmost directory level.
1761 If two directories differ only in that file names are lower case in
1762 one directory and upper case in the upper, `diff' normally reports many
1763 differences because it compares file names in a case sensitive way.
1764 With the `--ignore-file-name-case' option, `diff' ignores case
1765 differences in file names, so that for example the contents of the file
1766 `Tao' in one directory are compared to the contents of the file `TAO'
1767 in the other. The `--no-ignore-file-name-case' option cancels the
1768 effect of the `--ignore-file-name-case' option, reverting to the default
1771 If an `-x PATTERN' or `--exclude=PATTERN' option, or an `-X FILE' or
1772 `--exclude-from=FILE' option, is specified while the
1773 `--ignore-file-name-case' option is in effect, case is ignored when
1774 excluding file names matching the specified patterns.
1777 File: diff.info, Node: Adjusting Output, Next: diff Performance, Prev: Comparing Directories, Up: Top
1779 5 Making `diff' Output Prettier
1780 *******************************
1782 `diff' provides several ways to adjust the appearance of its output.
1783 These adjustments can be applied to any output format.
1787 * Tabs:: Preserving the alignment of tab stops.
1788 * Trailing Blanks:: Suppressing blanks before empty output lines.
1789 * Pagination:: Page numbering and time-stamping `diff' output.
1792 File: diff.info, Node: Tabs, Next: Trailing Blanks, Up: Adjusting Output
1794 5.1 Preserving Tab Stop Alignment
1795 =================================
1797 The lines of text in some of the `diff' output formats are preceded by
1798 one or two characters that indicate whether the text is inserted,
1799 deleted, or changed. The addition of those characters can cause tabs
1800 to move to the next tab stop, throwing off the alignment of columns in
1801 the line. GNU `diff' provides two ways to make tab-aligned columns
1804 The first way is to have `diff' convert all tabs into the correct
1805 number of spaces before outputting them; select this method with the
1806 `-t' or `--expand-tabs' option. To use this form of output with
1807 `patch', you must give `patch' the `-l' or `--ignore-white-space'
1808 option (*note Changed White Space::, for more information). `diff'
1809 normally assumes that tab stops are set every 8 print columns, but this
1810 can be altered by the `--tabsize=COLUMNS' option.
1812 The other method for making tabs line up correctly is to add a tab
1813 character instead of a space after the indicator character at the
1814 beginning of the line. This ensures that all following tab characters
1815 are in the same position relative to tab stops that they were in the
1816 original files, so that the output is aligned correctly. Its
1817 disadvantage is that it can make long lines too long to fit on one line
1818 of the screen or the paper. It also does not work with the unified
1819 output format, which does not have a space character after the change
1820 type indicator character. Select this method with the `-T' or
1821 `--initial-tab' option.
1824 File: diff.info, Node: Trailing Blanks, Next: Pagination, Prev: Tabs, Up: Adjusting Output
1826 5.2 Omitting trailing blanks
1827 ============================
1829 When outputting lines in normal or context format, or outputting an
1830 unchanged line in unified format, `diff' normally outputs a blank just
1831 before each line. If the line is empty, the output of `diff' therefore
1832 contains trailing blanks even though the input does not contain them.
1833 For example, when outputting an unchanged empty line in context format,
1834 `diff' normally outputs a line with two leading spaces.
1836 Some text editors and email agents routinely delete trailing blanks,
1837 so it can be a problem to deal with diff output files that contain
1838 them. You can avoid this problem with the `--suppress-blank-empty'
1839 option. It causes `diff' to omit trailing blanks at the end of output
1840 lines in normal, context, and unified format, unless the trailing
1841 blanks were already present in the input. This changes the output
1842 format slightly, so that output lines are guaranteed to never end in a
1843 blank unless an input line ends in a blank. This format is less likely
1844 to be munged by text editors or by transmission via email. It is
1845 accepted by GNU `patch' as well.
1848 File: diff.info, Node: Pagination, Prev: Trailing Blanks, Up: Adjusting Output
1850 5.3 Paginating `diff' Output
1851 ============================
1853 It can be convenient to have long output page-numbered and time-stamped.
1854 The `-l' or `--paginate' option does this by sending the `diff' output
1855 through the `pr' program. Here is what the page header might look like
1856 for `diff -lc lao tzu':
1858 2002-02-22 14:20 diff -lc lao tzu Page 1
1861 File: diff.info, Node: diff Performance, Next: Comparing Three Files, Prev: Adjusting Output, Up: Top
1863 6 `diff' Performance Tradeoffs
1864 ******************************
1866 GNU `diff' runs quite efficiently; however, in some circumstances you
1867 can cause it to run faster or produce a more compact set of changes.
1869 One way to improve `diff' performance is to use hard or symbolic
1870 links to files instead of copies. This improves performance because
1871 `diff' normally does not need to read two hard or symbolic links to the
1872 same file, since their contents must be identical. For example,
1873 suppose you copy a large directory hierarchy, make a few changes to the
1874 copy, and then often use `diff -r' to compare the original to the copy.
1875 If the original files are read-only, you can greatly improve
1876 performance by creating the copy using hard or symbolic links (e.g.,
1877 with GNU `cp -lR' or `cp -sR'). Before editing a file in the copy for
1878 the first time, you should break the link and replace it with a regular
1881 You can also affect the performance of GNU `diff' by giving it
1882 options that change the way it compares files. Performance has more
1883 than one dimension. These options improve one aspect of performance at
1884 the cost of another, or they improve performance in some cases while
1885 hurting it in others.
1887 The way that GNU `diff' determines which lines have changed always
1888 comes up with a near-minimal set of differences. Usually it is good
1889 enough for practical purposes. If the `diff' output is large, you
1890 might want `diff' to use a modified algorithm that sometimes produces a
1891 smaller set of differences. The `-d' or `--minimal' option does this;
1892 however, it can also cause `diff' to run more slowly than usual, so it
1893 is not the default behavior.
1895 When the files you are comparing are large and have small groups of
1896 changes scattered throughout them, you can use the
1897 `--speed-large-files' option to make a different modification to the
1898 algorithm that `diff' uses. If the input files have a constant small
1899 density of changes, this option speeds up the comparisons without
1900 changing the output. If not, `diff' might produce a larger set of
1901 differences; however, the output will still be correct.
1903 Normally `diff' discards the prefix and suffix that is common to
1904 both files before it attempts to find a minimal set of differences.
1905 This makes `diff' run faster, but occasionally it may produce
1906 non-minimal output. The `--horizon-lines=LINES' option prevents `diff'
1907 from discarding the last LINES lines of the prefix and the first LINES
1908 lines of the suffix. This gives `diff' further opportunities to find a
1911 Suppose a run of changed lines includes a sequence of lines at one
1912 end and there is an identical sequence of lines just outside the other
1913 end. The `diff' command is free to choose which identical sequence is
1914 included in the hunk. In this case, `diff' normally shifts the hunk's
1915 boundaries when this merges adjacent hunks, or shifts a hunk's lines
1916 towards the end of the file. Merging hunks can make the output look
1917 nicer in some cases.
1920 File: diff.info, Node: Comparing Three Files, Next: diff3 Merging, Prev: diff Performance, Up: Top
1922 7 Comparing Three Files
1923 ***********************
1925 Use the program `diff3' to compare three files and show any differences
1926 among them. (`diff3' can also merge files; see *note diff3 Merging::).
1928 The "normal" `diff3' output format shows each hunk of differences
1929 without surrounding context. Hunks are labeled depending on whether
1930 they are two-way or three-way, and lines are annotated by their
1931 location in the input files.
1933 *Note Invoking diff3::, for more information on how to run `diff3'.
1937 * Sample diff3 Input:: Sample `diff3' input for examples.
1938 * Example diff3 Normal:: Sample output in the normal format.
1939 * diff3 Hunks:: The format of normal output format.
1940 * Detailed diff3 Normal:: A detailed description of normal output format.
1943 File: diff.info, Node: Sample diff3 Input, Next: Example diff3 Normal, Up: Comparing Three Files
1945 7.1 A Third Sample Input File
1946 =============================
1948 Here is a third sample file that will be used in examples to illustrate
1949 the output of `diff3' and how various options can change it. The first
1950 two files are the same that we used for `diff' (*note Sample diff
1951 Input::). This is the third sample file, called `tao':
1953 The Way that can be told of is not the eternal Way;
1954 The name that can be named is not the eternal name.
1955 The Nameless is the origin of Heaven and Earth;
1956 The named is the mother of all things.
1958 Therefore let there always be non-being,
1959 so we may see their subtlety,
1960 And let there always be being,
1961 so we may see their result.
1962 The two are the same,
1963 But after they are produced,
1964 they have different names.
1966 -- The Way of Lao-Tzu, tr. Wing-tsit Chan
1969 File: diff.info, Node: Example diff3 Normal, Next: diff3 Hunks, Prev: Sample diff3 Input, Up: Comparing Three Files
1971 7.2 An Example of `diff3' Normal Format
1972 =======================================
1974 Here is the output of the command `diff3 lao tzu tao' (*note Sample
1975 diff3 Input::, for the complete contents of the files). Notice that it
1976 shows only the lines that are different among the three files.
1981 The Way that can be told of is not the eternal Way;
1982 The name that can be named is not the eternal name.
1986 The Named is the mother of all things.
1989 The named is the mother of all things.
1994 so we may see their outcome.
1996 so we may see their result.
2000 They both may be called deep and profound.
2001 Deeper and more profound,
2002 The door of all subtleties!
2005 -- The Way of Lao-Tzu, tr. Wing-tsit Chan
2008 File: diff.info, Node: Detailed diff3 Normal, Prev: diff3 Hunks, Up: Comparing Three Files
2010 7.3 Detailed Description of `diff3' Normal Format
2011 =================================================
2013 Each hunk begins with a line marked `===='. Three-way hunks have plain
2014 `====' lines, and two-way hunks have `1', `2', or `3' appended to
2015 specify which of the three input files differ in that hunk. The hunks
2016 contain copies of two or three sets of input lines each preceded by one
2017 or two commands identifying where the lines came from.
2019 Normally, two spaces precede each copy of an input line to
2020 distinguish it from the commands. But with the `-T' or `--initial-tab'
2021 option, `diff3' uses a tab instead of two spaces; this lines up tabs
2022 correctly. *Note Tabs::, for more information.
2024 Commands take the following forms:
2027 This hunk appears after line L of file FILE, and contains no lines
2028 in that file. To edit this file to yield the other files, one
2029 must append hunk lines taken from the other files. For example,
2030 `1:11a' means that the hunk follows line 11 in the first file and
2031 contains no lines from that file.
2034 This hunk contains the lines in the range R of file FILE. The
2035 range R is a comma-separated pair of line numbers, or just one
2036 number if there is only one line. To edit this file to yield the
2037 other files, one must change the specified lines to be the lines
2038 taken from the other files. For example, `2:11,13c' means that
2039 the hunk contains lines 11 through 13 from the second file.
2041 If the last line in a set of input lines is incomplete (*note
2042 Incomplete Lines::), it is distinguished on output from a full line by
2043 a following line that starts with `\'.
2046 File: diff.info, Node: diff3 Hunks, Next: Detailed diff3 Normal, Prev: Example diff3 Normal, Up: Comparing Three Files
2051 Groups of lines that differ in two or three of the input files are
2052 called "diff3 hunks", by analogy with `diff' hunks (*note Hunks::). If
2053 all three input files differ in a `diff3' hunk, the hunk is called a
2054 "three-way hunk"; if just two input files differ, it is a "two-way
2057 As with `diff', several solutions are possible. When comparing the
2058 files `A', `B', and `C', `diff3' normally finds `diff3' hunks by
2059 merging the two-way hunks output by the two commands `diff A B' and
2060 `diff A C'. This does not necessarily minimize the size of the output,
2061 but exceptions should be rare.
2063 For example, suppose `F' contains the three lines `a', `b', `f', `G'
2064 contains the lines `g', `b', `g', and `H' contains the lines `a', `b',
2065 `h'. `diff3 F G H' might output the following:
2081 because it found a two-way hunk containing `a' in the first and third
2082 files and `g' in the second file, then the single line `b' common to
2083 all three files, then a three-way hunk containing the last line of each
2087 File: diff.info, Node: diff3 Merging, Next: Interactive Merging, Prev: Comparing Three Files, Up: Top
2089 8 Merging From a Common Ancestor
2090 ********************************
2092 When two people have made changes to copies of the same file, `diff3'
2093 can produce a merged output that contains both sets of changes together
2094 with warnings about conflicts.
2096 One might imagine programs with names like `diff4' and `diff5' to
2097 compare more than three files simultaneously, but in practice the need
2098 rarely arises. You can use `diff3' to merge three or more sets of
2099 changes to a file by merging two change sets at a time.
2101 `diff3' can incorporate changes from two modified versions into a
2102 common preceding version. This lets you merge the sets of changes
2103 represented by the two newer files. Specify the common ancestor version
2104 as the second argument and the two newer versions as the first and third
2105 arguments, like this:
2107 diff3 MINE OLDER YOURS
2109 You can remember the order of the arguments by noting that they are in
2112 You can think of this as subtracting OLDER from YOURS and adding the
2113 result to MINE, or as merging into MINE the changes that would turn
2114 OLDER into YOURS. This merging is well-defined as long as MINE and
2115 OLDER match in the neighborhood of each such change. This fails to be
2116 true when all three input files differ or when only OLDER differs; we
2117 call this a "conflict". When all three input files differ, we call the
2118 conflict an "overlap".
2120 `diff3' gives you several ways to handle overlaps and conflicts.
2121 You can omit overlaps or conflicts, or select only overlaps, or mark
2122 conflicts with special `<<<<<<<' and `>>>>>>>' lines.
2124 `diff3' can output the merge results as an `ed' script that that can
2125 be applied to the first file to yield the merged output. However, it
2126 is usually better to have `diff3' generate the merged output directly;
2127 this bypasses some problems with `ed'.
2131 * Which Changes:: Selecting changes to incorporate.
2132 * Marking Conflicts:: Marking conflicts.
2133 * Bypassing ed:: Generating merged output directly.
2134 * Merging Incomplete Lines:: How `diff3' merges incomplete lines.
2135 * Saving the Changed File:: Emulating System V behavior.
2138 File: diff.info, Node: Which Changes, Next: Marking Conflicts, Up: diff3 Merging
2140 8.1 Selecting Which Changes to Incorporate
2141 ==========================================
2143 You can select all unmerged changes from OLDER to YOURS for merging
2144 into MINE with the `-e' or `--ed' option. You can select only the
2145 nonoverlapping unmerged changes with `-3' or `--easy-only', and you can
2146 select only the overlapping changes with `-x' or `--overlap-only'.
2148 The `-e', `-3' and `-x' options select only "unmerged changes", i.e.
2149 changes where MINE and YOURS differ; they ignore changes from OLDER to
2150 YOURS where MINE and YOURS are identical, because they assume that such
2151 changes have already been merged. If this assumption is not a safe
2152 one, you can use the `-A' or `--show-all' option (*note Marking
2155 Here is the output of the command `diff3' with each of these three
2156 options (*note Sample diff3 Input::, for the complete contents of the
2157 files). Notice that `-e' outputs the union of the disjoint sets of
2158 changes output by `-3' and `-x'.
2160 Output of `diff3 -e lao tzu tao':
2163 -- The Way of Lao-Tzu, tr. Wing-tsit Chan
2166 so we may see their result.
2169 Output of `diff3 -3 lao tzu tao':
2171 so we may see their result.
2174 Output of `diff3 -x lao tzu tao':
2177 -- The Way of Lao-Tzu, tr. Wing-tsit Chan
2181 File: diff.info, Node: Marking Conflicts, Next: Bypassing ed, Prev: Which Changes, Up: diff3 Merging
2183 8.2 Marking Conflicts
2184 =====================
2186 `diff3' can mark conflicts in the merged output by bracketing them with
2187 special marker lines. A conflict that comes from two files A and B is
2196 A conflict that comes from three files A, B and C is marked as
2207 The `-A' or `--show-all' option acts like the `-e' option, except
2208 that it brackets conflicts, and it outputs all changes from OLDER to
2209 YOURS, not just the unmerged changes. Thus, given the sample input
2210 files (*note Sample diff3 Input::), `diff3 -A lao tzu tao' puts
2211 brackets around the conflict where only `tzu' differs:
2215 The Way that can be told of is not the eternal Way;
2216 The name that can be named is not the eternal name.
2219 And it outputs the three-way conflict as follows:
2223 They both may be called deep and profound.
2224 Deeper and more profound,
2225 The door of all subtleties!
2228 -- The Way of Lao-Tzu, tr. Wing-tsit Chan
2231 The `-E' or `--show-overlap' option outputs less information than
2232 the `-A' or `--show-all' option, because it outputs only unmerged
2233 changes, and it never outputs the contents of the second file. Thus
2234 the `-E' option acts like the `-e' option, except that it brackets the
2235 first and third files from three-way overlapping changes. Similarly,
2236 `-X' acts like `-x', except it brackets all its (necessarily
2237 overlapping) changes. For example, for the three-way overlapping
2238 change above, the `-E' and `-X' options output the following:
2243 -- The Way of Lao-Tzu, tr. Wing-tsit Chan
2246 If you are comparing files that have meaningless or uninformative
2247 names, you can use the `--label=LABEL' option to show alternate names
2248 in the `<<<<<<<', `|||||||' and `>>>>>>>' brackets. This option can be
2249 given up to three times, once for each input file. Thus `diff3 -A
2250 --label X --label Y --label Z A B C' acts like `diff3 -A A B C', except
2251 that the output looks like it came from files named `X', `Y' and `Z'
2252 rather than from files named `A', `B' and `C'.
2255 File: diff.info, Node: Bypassing ed, Next: Merging Incomplete Lines, Prev: Marking Conflicts, Up: diff3 Merging
2257 8.3 Generating the Merged Output Directly
2258 =========================================
2260 With the `-m' or `--merge' option, `diff3' outputs the merged file
2261 directly. This is more efficient than using `ed' to generate it, and
2262 works even with non-text files that `ed' would reject. If you specify
2263 `-m' without an `ed' script option, `-A' is assumed.
2265 For example, the command `diff3 -m lao tzu tao' (*note Sample diff3
2266 Input:: for a copy of the input files) would output the following:
2270 The Way that can be told of is not the eternal Way;
2271 The name that can be named is not the eternal name.
2273 The Nameless is the origin of Heaven and Earth;
2274 The Named is the mother of all things.
2275 Therefore let there always be non-being,
2276 so we may see their subtlety,
2277 And let there always be being,
2278 so we may see their result.
2279 The two are the same,
2280 But after they are produced,
2281 they have different names.
2284 They both may be called deep and profound.
2285 Deeper and more profound,
2286 The door of all subtleties!
2289 -- The Way of Lao-Tzu, tr. Wing-tsit Chan
2293 File: diff.info, Node: Merging Incomplete Lines, Next: Saving the Changed File, Prev: Bypassing ed, Up: diff3 Merging
2295 8.4 How `diff3' Merges Incomplete Lines
2296 =======================================
2298 With `-m', incomplete lines (*note Incomplete Lines::) are simply
2299 copied to the output as they are found; if the merged output ends in an
2300 conflict and one of the input files ends in an incomplete line,
2301 succeeding `|||||||', `=======' or `>>>>>>>' brackets appear somewhere
2302 other than the start of a line because they are appended to the
2305 Without `-m', if an `ed' script option is specified and an
2306 incomplete line is found, `diff3' generates a warning and acts as if a
2307 newline had been present.
2310 File: diff.info, Node: Saving the Changed File, Prev: Merging Incomplete Lines, Up: diff3 Merging
2312 8.5 Saving the Changed File
2313 ===========================
2315 Traditional Unix `diff3' generates an `ed' script without the trailing
2316 `w' and `q' commands that save the changes. System V `diff3' generates
2317 these extra commands. GNU `diff3' normally behaves like traditional
2318 Unix `diff3', but with the `-i' option it behaves like System V `diff3'
2319 and appends the `w' and `q' commands.
2321 The `-i' option requires one of the `ed' script options `-AeExX3',
2322 and is incompatible with the merged output option `-m'.
2325 File: diff.info, Node: Interactive Merging, Next: Merging with patch, Prev: diff3 Merging, Up: Top
2327 9 Interactive Merging with `sdiff'
2328 **********************************
2330 With `sdiff', you can merge two files interactively based on a
2331 side-by-side `-y' format comparison (*note Side by Side::). Use `-o
2332 FILE' or `--output=FILE' to specify where to put the merged text.
2333 *Note Invoking sdiff::, for more details on the options to `sdiff'.
2335 Another way to merge files interactively is to use the Emacs Lisp
2336 package `emerge'. *Note emerge: (emacs)emerge, for more information.
2340 * sdiff Option Summary:: Summary of `sdiff' options.
2341 * Merge Commands:: Merging two files interactively.
2344 File: diff.info, Node: sdiff Option Summary, Next: Merge Commands, Up: Interactive Merging
2346 9.1 Specifying `diff' Options to `sdiff'
2347 ========================================
2349 The following `sdiff' options have the same meaning as for `diff'.
2350 *Note diff Options::, for the use of these options.
2356 --ignore-blank-lines --ignore-case
2357 --ignore-matching-lines=REGEXP --ignore-space-change
2358 --ignore-tab-expansion
2359 --left-column --minimal --speed-large-files
2360 --strip-trailing-cr --suppress-common-lines
2361 --tabsize=COLUMNS --text --version --width=COLUMNS
2363 For historical reasons, `sdiff' has alternate names for some
2364 options. The `-l' option is equivalent to the `--left-column' option,
2365 and similarly `-s' is equivalent to `--suppress-common-lines'. The
2366 meaning of the `sdiff' `-w' and `-W' options is interchanged from that
2367 of `diff': with `sdiff', `-w COLUMNS' is equivalent to
2368 `--width=COLUMNS', and `-W' is equivalent to `--ignore-all-space'.
2369 `sdiff' without the `-o' option is equivalent to `diff' with the `-y'
2370 or `--side-by-side' option (*note Side by Side::).
2373 File: diff.info, Node: Merge Commands, Prev: sdiff Option Summary, Up: Interactive Merging
2378 Groups of common lines, with a blank gutter, are copied from the first
2379 file to the output. After each group of differing lines, `sdiff'
2380 prompts with `%' and pauses, waiting for one of the following commands.
2381 Follow each command with <RET>.
2384 Discard both versions. Invoke a text editor on an empty temporary
2385 file, then copy the resulting file to the output.
2388 Concatenate the two versions, edit the result in a temporary file,
2389 then copy the edited result to the output.
2392 Like `eb', except precede each version with a header that shows
2393 what file and lines the version came from.
2397 Edit a copy of the left version, then copy the result to the
2402 Edit a copy of the right version, then copy the result to the
2407 Copy the left version to the output.
2414 Copy the right version to the output.
2417 Silently copy common lines.
2420 Verbosely copy common lines. This is the default.
2422 The text editor invoked is specified by the `EDITOR' environment
2423 variable if it is set. The default is system-dependent.
2426 File: diff.info, Node: Merging with patch, Next: Making Patches, Prev: Interactive Merging, Up: Top
2428 10 Merging with `patch'
2429 ***********************
2431 `patch' takes comparison output produced by `diff' and applies the
2432 differences to a copy of the original file, producing a patched
2433 version. With `patch', you can distribute just the changes to a set of
2434 files instead of distributing the entire file set; your correspondents
2435 can apply `patch' to update their copy of the files with your changes.
2436 `patch' automatically determines the diff format, skips any leading or
2437 trailing headers, and uses the headers to determine which file to
2438 patch. This lets your correspondents feed a mail message containing a
2439 difference listing directly to `patch'.
2441 `patch' detects and warns about common problems like forward
2442 patches. It saves any patches that it could not apply. It can also
2443 maintain a `patchlevel.h' file to ensure that your correspondents apply
2444 diffs in the proper order.
2446 `patch' accepts a series of diffs in its standard input, usually
2447 separated by headers that specify which file to patch. It applies
2448 `diff' hunks (*note Hunks::) one by one. If a hunk does not exactly
2449 match the original file, `patch' uses heuristics to try to patch the
2450 file as well as it can. If no approximate match can be found, `patch'
2451 rejects the hunk and skips to the next hunk. `patch' normally replaces
2452 each file F with its new version, putting reject hunks (if any) into
2455 *Note Invoking patch::, for detailed information on the options to
2460 * patch Input:: Selecting the type of `patch' input.
2461 * Revision Control:: Getting files from RCS, SCCS, etc.
2462 * Imperfect:: Dealing with imperfect patches.
2463 * Creating and Removing:: Creating and removing files with a patch.
2464 * Patching Time Stamps:: Updating time stamps on patched files.
2465 * Multiple Patches:: Handling multiple patches in a file.
2466 * patch Directories:: Changing directory and stripping directories.
2467 * Backups:: Whether backup files are made.
2468 * Backup Names:: Backup file names.
2469 * Reject Names:: Reject file names.
2470 * patch Messages:: Messages and questions `patch' can produce.
2471 * patch and POSIX:: Conformance to the POSIX standard.
2472 * patch and Tradition:: GNU versus traditional `patch'.
2475 File: diff.info, Node: patch Input, Next: Revision Control, Up: Merging with patch
2477 10.1 Selecting the `patch' Input Format
2478 =======================================
2480 `patch' normally determines which `diff' format the patch file uses by
2481 examining its contents. For patch files that contain particularly
2482 confusing leading text, you might need to use one of the following
2483 options to force `patch' to interpret the patch file as a certain
2484 format of diff. The output formats listed here are the only ones that
2485 `patch' can understand.
2504 File: diff.info, Node: Revision Control, Next: Imperfect, Prev: patch Input, Up: Merging with patch
2506 10.2 Revision Control
2507 =====================
2509 If a nonexistent input file is under a revision control system
2510 supported by `patch', `patch' normally asks the user whether to get (or
2511 check out) the file from the revision control system. Patch currently
2512 supports RCS, ClearCase and SCCS. Under RCS and SCCS, `patch' also
2513 asks when the input file is read-only and matches the default version
2514 in the revision control system.
2516 The `-g NUM' or `--get=NUM' option affects access to files under
2517 supported revision control systems. If NUM is positive, `patch' gets
2518 the file without asking the user; if zero, `patch' neither asks the
2519 user nor gets the file; and if negative, `patch' asks the user before
2520 getting the file. The default value of NUM is given by the value of the
2521 `PATCH_GET' environment variable if it is set; if not, the default
2522 value is zero if `patch' is conforming to POSIX, negative otherwise.
2523 *Note patch and POSIX::.
2525 The choice of revision control system is unaffected by the
2526 `VERSION_CONTROL' environment variable (*note Backup Names::).
2529 File: diff.info, Node: Imperfect, Next: Creating and Removing, Prev: Revision Control, Up: Merging with patch
2531 10.3 Applying Imperfect Patches
2532 ===============================
2534 `patch' tries to skip any leading text in the patch file, apply the
2535 diff, and then skip any trailing text. Thus you can feed a mail
2536 message directly to `patch', and it should work. If the entire diff is
2537 indented by a constant amount of white space, `patch' automatically
2538 ignores the indentation. If a context diff contains trailing carriage
2539 return on each line, `patch' automatically ignores the carriage return.
2540 If a context diff has been encapsulated by prepending `- ' to lines
2541 beginning with `-' as per Internet RFC 934
2542 (ftp://ftp.isi.edu/in-notes/rfc934.txt), `patch' automatically
2543 unencapsulates the input.
2545 However, certain other types of imperfect input require user
2546 intervention or testing.
2550 * Changed White Space:: When tabs and spaces don't match exactly.
2551 * Reversed Patches:: Applying reversed patches correctly.
2552 * Inexact:: Helping `patch' find close matches.
2553 * Dry Runs:: Predicting what `patch' will do.
2556 File: diff.info, Node: Changed White Space, Next: Reversed Patches, Up: Imperfect
2558 10.3.1 Applying Patches with Changed White Space
2559 ------------------------------------------------
2561 Sometimes mailers, editors, or other programs change spaces into tabs,
2562 or vice versa. If this happens to a patch file or an input file, the
2563 files might look the same, but `patch' will not be able to match them
2564 properly. If this problem occurs, use the `-l' or
2565 `--ignore-white-space' option, which makes `patch' compare blank
2566 characters (i.e. spaces and tabs) loosely so that any nonempty sequence
2567 of blanks in the patch file matches any nonempty sequence of blanks in
2568 the input files. Non-blank characters must still match exactly. Each
2569 line of the context must still match a line in the input file.
2572 File: diff.info, Node: Reversed Patches, Next: Inexact, Prev: Changed White Space, Up: Imperfect
2574 10.3.2 Applying Reversed Patches
2575 --------------------------------
2577 Sometimes people run `diff' with the new file first instead of second.
2578 This creates a diff that is "reversed". To apply such patches, give
2579 `patch' the `-R' or `--reverse' option. `patch' then attempts to swap
2580 each hunk around before applying it. Rejects come out in the swapped
2583 Often `patch' can guess that the patch is reversed. If the first
2584 hunk of a patch fails, `patch' reverses the hunk to see if it can apply
2585 it that way. If it can, `patch' asks you if you want to have the `-R'
2586 option set; if it can't, `patch' continues to apply the patch normally.
2587 This method cannot detect a reversed patch if it is a normal diff and
2588 the first command is an append (which should have been a delete) since
2589 appends always succeed, because a null context matches anywhere. But
2590 most patches add or change lines rather than delete them, so most
2591 reversed normal diffs begin with a delete, which fails, and `patch'
2594 If you apply a patch that you have already applied, `patch' thinks
2595 it is a reversed patch and offers to un-apply the patch. This could be
2596 construed as a feature. If you did this inadvertently and you don't
2597 want to un-apply the patch, just answer `n' to this offer and to the
2598 subsequent "apply anyway" question--or type `C-c' to kill the `patch'
2602 File: diff.info, Node: Inexact, Next: Dry Runs, Prev: Reversed Patches, Up: Imperfect
2604 10.3.3 Helping `patch' Find Inexact Matches
2605 -------------------------------------------
2607 For context diffs, and to a lesser extent normal diffs, `patch' can
2608 detect when the line numbers mentioned in the patch are incorrect, and
2609 it attempts to find the correct place to apply each hunk of the patch.
2610 As a first guess, it takes the line number mentioned in the hunk, plus
2611 or minus any offset used in applying the previous hunk. If that is not
2612 the correct place, `patch' scans both forward and backward for a set of
2613 lines matching the context given in the hunk.
2615 First `patch' looks for a place where all lines of the context
2616 match. If it cannot find such a place, and it is reading a context or
2617 unified diff, and the maximum fuzz factor is set to 1 or more, then
2618 `patch' makes another scan, ignoring the first and last line of
2619 context. If that fails, and the maximum fuzz factor is set to 2 or
2620 more, it makes another scan, ignoring the first two and last two lines
2621 of context are ignored. It continues similarly if the maximum fuzz
2624 The `-F LINES' or `--fuzz=LINES' option sets the maximum fuzz factor
2625 to LINES. This option only applies to context and unified diffs; it
2626 ignores up to LINES lines while looking for the place to install a
2627 hunk. Note that a larger fuzz factor increases the odds of making a
2628 faulty patch. The default fuzz factor is 2; there is no point to
2629 setting it to more than the number of lines of context in the diff,
2632 If `patch' cannot find a place to install a hunk of the patch, it
2633 writes the hunk out to a reject file (*note Reject Names::, for
2634 information on how reject files are named). It writes out rejected
2635 hunks in context format no matter what form the input patch is in. If
2636 the input is a normal or `ed' diff, many of the contexts are simply
2637 null. The line numbers on the hunks in the reject file may be
2638 different from those in the patch file: they show the approximate
2639 location where `patch' thinks the failed hunks belong in the new file
2640 rather than in the old one.
2642 If the `--verbose' option is given, then as it completes each hunk
2643 `patch' tells you whether the hunk succeeded or failed, and if it
2644 failed, on which line (in the new file) `patch' thinks the hunk should
2645 go. If this is different from the line number specified in the diff,
2646 it tells you the offset. A single large offset _may_ indicate that
2647 `patch' installed a hunk in the wrong place. `patch' also tells you if
2648 it used a fuzz factor to make the match, in which case you should also
2649 be slightly suspicious.
2651 `patch' cannot tell if the line numbers are off in an `ed' script,
2652 and can only detect wrong line numbers in a normal diff when it finds a
2653 change or delete command. It may have the same problem with a context
2654 diff using a fuzz factor equal to or greater than the number of lines
2655 of context shown in the diff (typically 3). In these cases, you should
2656 probably look at a context diff between your original and patched input
2657 files to see if the changes make sense. Compiling without errors is a
2658 pretty good indication that the patch worked, but not a guarantee.
2660 A patch against an empty file applies to a nonexistent file, and vice
2661 versa. *Note Creating and Removing::.
2663 `patch' usually produces the correct results, even when it must make
2664 many guesses. However, the results are guaranteed only when the patch
2665 is applied to an exact copy of the file that the patch was generated
2669 File: diff.info, Node: Dry Runs, Prev: Inexact, Up: Imperfect
2671 10.3.4 Predicting what `patch' will do
2672 --------------------------------------
2674 It may not be obvious in advance what `patch' will do with a
2675 complicated or poorly formatted patch. If you are concerned that the
2676 input might cause `patch' to modify the wrong files, you can use the
2677 `--dry-run' option, which causes `patch' to print the results of
2678 applying patches without actually changing any files. You can then
2679 inspect the diagnostics generated by the dry run to see whether `patch'
2680 will modify the files that you expect. If the patch does not do what
2681 you want, you can modify the patch (or the other options to `patch')
2682 and try another dry run. Once you are satisfied with the proposed
2683 patch you can apply it by invoking `patch' as before, but this time
2684 without the `--dry-run' option.
2687 File: diff.info, Node: Creating and Removing, Next: Patching Time Stamps, Prev: Imperfect, Up: Merging with patch
2689 10.4 Creating and Removing Files
2690 ================================
2692 Sometimes when comparing two directories, a file may exist in one
2693 directory but not the other. If you give `diff' the `-N' or
2694 `--new-file' option, or if you supply an old or new file that is named
2695 `/dev/null' or is empty and is dated the Epoch (1970-01-01 00:00:00
2696 UTC), `diff' outputs a patch that adds or deletes the contents of this
2697 file. When given such a patch, `patch' normally creates a new file or
2698 removes the old file. However, when conforming to POSIX (*note patch
2699 and POSIX::), `patch' does not remove the old file, but leaves it empty.
2700 The `-E' or `--remove-empty-files' option causes `patch' to remove
2701 output files that are empty after applying a patch, even if the patch
2702 does not appear to be one that removed the file.
2704 If the patch appears to create a file that already exists, `patch'
2705 asks for confirmation before applying the patch.
2708 File: diff.info, Node: Patching Time Stamps, Next: Multiple Patches, Prev: Creating and Removing, Up: Merging with patch
2710 10.5 Updating Time Stamps on Patched Files
2711 ==========================================
2713 When `patch' updates a file, it normally sets the file's last-modified
2714 time stamp to the current time of day. If you are using `patch' to
2715 track a software distribution, this can cause `make' to incorrectly
2716 conclude that a patched file is out of date. For example, if
2717 `syntax.c' depends on `syntax.y', and `patch' updates `syntax.c' and
2718 then `syntax.y', then `syntax.c' will normally appear to be out of date
2719 with respect to `syntax.y' even though its contents are actually up to
2722 The `-Z' or `--set-utc' option causes `patch' to set a patched
2723 file's modification and access times to the time stamps given in
2724 context diff headers. If the context diff headers do not specify a
2725 time zone, they are assumed to use Coordinated Universal Time (UTC,
2726 often known as GMT).
2728 The `-T' or `--set-time' option acts like `-Z' or `--set-utc',
2729 except that it assumes that the context diff headers' time stamps use
2730 local time instead of UTC. This option is not recommended, because
2731 patches using local time cannot easily be used by people in other time
2732 zones, and because local time stamps are ambiguous when local clocks
2733 move backwards during daylight-saving time adjustments. If the context
2734 diff headers specify a time zone, this option is equivalent to `-Z' or
2737 `patch' normally refrains from setting a file's time stamps if the
2738 file's original last-modified time stamp does not match the time given
2739 in the diff header, of if the file's contents do not exactly match the
2740 patch. However, if the `-f' or `--force' option is given, the file's
2741 time stamps are set regardless.
2743 Due to the limitations of the current `diff' format, `patch' cannot
2744 update the times of files whose contents have not changed. Also, if
2745 you set file time stamps to values other than the current time of day,
2746 you should also remove (e.g., with `make clean') all files that depend
2747 on the patched files, so that later invocations of `make' do not get
2748 confused by the patched files' times.
2751 File: diff.info, Node: Multiple Patches, Next: patch Directories, Prev: Patching Time Stamps, Up: Merging with patch
2753 10.6 Multiple Patches in a File
2754 ===============================
2756 If the patch file contains more than one patch, and if you do not
2757 specify an input file on the command line, `patch' tries to apply each
2758 patch as if they came from separate patch files. This means that it
2759 determines the name of the file to patch for each patch, and that it
2760 examines the leading text before each patch for file names and
2761 prerequisite revision level (*note Making Patches::, for more on that
2764 `patch' uses the following rules to intuit a file name from the
2765 leading text before a patch. First, `patch' takes an ordered list of
2766 candidate file names as follows:
2768 * If the header is that of a context diff, `patch' takes the old and
2769 new file names in the header. A name is ignored if it does not
2770 have enough slashes to satisfy the `-pNUM' or `--strip=NUM'
2771 option. The name `/dev/null' is also ignored.
2773 * If there is an `Index:' line in the leading garbage and if either
2774 the old and new names are both absent or if `patch' is conforming
2775 to POSIX, `patch' takes the name in the `Index:' line.
2777 * For the purpose of the following rules, the candidate file names
2778 are considered to be in the order (old, new, index), regardless of
2779 the order that they appear in the header.
2781 Then `patch' selects a file name from the candidate list as follows:
2783 * If some of the named files exist, `patch' selects the first name
2784 if conforming to POSIX, and the best name otherwise.
2786 * If `patch' is not ignoring RCS, ClearCase, and SCCS (*note
2787 Revision Control::), and no named files exist but an RCS,
2788 ClearCase, or SCCS master is found, `patch' selects the first
2789 named file with an RCS, ClearCase, or SCCS master.
2791 * If no named files exist, no RCS, ClearCase, or SCCS master was
2792 found, some names are given, `patch' is not conforming to POSIX,
2793 and the patch appears to create a file, `patch' selects the best
2794 name requiring the creation of the fewest directories.
2796 * If no file name results from the above heuristics, you are asked
2797 for the name of the file to patch, and `patch' selects that name.
2799 To determine the "best" of a nonempty list of file names, `patch'
2800 first takes all the names with the fewest path name components; of
2801 those, it then takes all the names with the shortest basename; of
2802 those, it then takes all the shortest names; finally, it takes the
2803 first remaining name.
2805 *Note patch and POSIX::, to see whether `patch' is conforming to
2809 File: diff.info, Node: patch Directories, Next: Backups, Prev: Multiple Patches, Up: Merging with patch
2811 10.7 Applying Patches in Other Directories
2812 ==========================================
2814 The `-d DIRECTORY' or `--directory=DIRECTORY' option to `patch' makes
2815 directory DIRECTORY the current directory for interpreting both file
2816 names in the patch file, and file names given as arguments to other
2817 options (such as `-B' and `-o'). For example, while in a mail reading
2818 program, you can patch a file in the `/usr/src/emacs' directory
2819 directly from a message containing the patch like this:
2821 | patch -d /usr/src/emacs
2823 Sometimes the file names given in a patch contain leading
2824 directories, but you keep your files in a directory different from the
2825 one given in the patch. In those cases, you can use the `-pNUMBER' or
2826 `--strip=NUMBER' option to set the file name strip count to NUMBER.
2827 The strip count tells `patch' how many slashes, along with the directory
2828 names between them, to strip from the front of file names. A sequence
2829 of one or more adjacent slashes is counted as a single slash. By
2830 default, `patch' strips off all leading directories, leaving just the
2833 For example, suppose the file name in the patch file is
2834 `/gnu/src/emacs/etc/NEWS'. Using `-p0' gives the entire file name
2835 unmodified, `-p1' gives `gnu/src/emacs/etc/NEWS' (no leading slash),
2836 `-p4' gives `etc/NEWS', and not specifying `-p' at all gives `NEWS'.
2838 `patch' looks for each file (after any slashes have been stripped)
2839 in the current directory, or if you used the `-d DIRECTORY' option, in
2843 File: diff.info, Node: Backups, Next: Backup Names, Prev: patch Directories, Up: Merging with patch
2848 Normally, `patch' creates a backup file if the patch does not exactly
2849 match the original input file, because in that case the original data
2850 might not be recovered if you undo the patch with `patch -R' (*note
2851 Reversed Patches::). However, when conforming to POSIX, `patch' does
2852 not create backup files by default. *Note patch and POSIX::.
2854 The `-b' or `--backup' option causes `patch' to make a backup file
2855 regardless of whether the patch matches the original input. The
2856 `--backup-if-mismatch' option causes `patch' to create backup files for
2857 mismatches files; this is the default when not conforming to POSIX. The
2858 `--no-backup-if-mismatch' option causes `patch' to not create backup
2859 files, even for mismatched patches; this is the default when conforming
2862 When backing up a file that does not exist, an empty, unreadable
2863 backup file is created as a placeholder to represent the nonexistent
2867 File: diff.info, Node: Backup Names, Next: Reject Names, Prev: Backups, Up: Merging with patch
2869 10.9 Backup File Names
2870 ======================
2872 Normally, `patch' renames an original input file into a backup file by
2873 appending to its name the extension `.orig', or `~' if using `.orig'
2874 would make the backup file name too long.(1) The `-z BACKUP-SUFFIX' or
2875 `--suffix=BACKUP-SUFFIX' option causes `patch' to use BACKUP-SUFFIX as
2876 the backup extension instead.
2878 Alternately, you can specify the extension for backup files with the
2879 `SIMPLE_BACKUP_SUFFIX' environment variable, which the options override.
2881 `patch' can also create numbered backup files the way GNU Emacs
2882 does. With this method, instead of having a single backup of each
2883 file, `patch' makes a new backup file name each time it patches a file.
2884 For example, the backups of a file named `sink' would be called,
2885 successively, `sink.~1~', `sink.~2~', `sink.~3~', etc.
2887 The `-V BACKUP-STYLE' or `--version-control=BACKUP-STYLE' option
2888 takes as an argument a method for creating backup file names. You can
2889 alternately control the type of backups that `patch' makes with the
2890 `PATCH_VERSION_CONTROL' environment variable, which the `-V' option
2891 overrides. If `PATCH_VERSION_CONTROL' is not set, the
2892 `VERSION_CONTROL' environment variable is used instead. Please note
2893 that these options and variables control backup file names; they do not
2894 affect the choice of revision control system (*note Revision Control::).
2896 The values of these environment variables and the argument to the
2897 `-V' option are like the GNU Emacs `version-control' variable (*note
2898 Backup Names: (emacs)Backup Names, for more information on backup
2899 versions in Emacs). They also recognize synonyms that are more
2900 descriptive. The valid values are listed below; unique abbreviations
2905 Always make numbered backups.
2909 Make numbered backups of files that already have them, simple
2910 backups of the others. This is the default.
2914 Always make simple backups.
2916 You can also tell `patch' to prepend a prefix, such as a directory
2917 name, to produce backup file names. The `-B PREFIX' or
2918 `--prefix=PREFIX' option makes backup files by prepending PREFIX to
2919 them. The `-Y PREFIX' or `--basename-prefix=PREFIX' prepends PREFIX to
2920 the last file name component of backup file names instead; for example,
2921 `-Y ~' causes the backup name for `dir/file.c' to be `dir/~file.c'. If
2922 you use either of these prefix options, the suffix-based options are
2925 If you specify the output file with the `-o' option, that file is
2926 the one that is backed up, not the input file.
2928 Options that affect the names of backup files do not affect whether
2929 backups are made. For example, if you specify the
2930 `--no-backup-if-mismatch' option, none of the options described in this
2931 section have any affect, because no backups are made.
2933 ---------- Footnotes ----------
2935 (1) A coding error in GNU `patch' version 2.5.4 causes it to always
2936 use `~', but this should be fixed in the next release.
2939 File: diff.info, Node: Reject Names, Next: patch Messages, Prev: Backup Names, Up: Merging with patch
2941 10.10 Reject File Names
2942 =======================
2944 The names for reject files (files containing patches that `patch' could
2945 not find a place to apply) are normally the name of the output file
2946 with `.rej' appended (or `#' if using `.rej' would make the backup file
2949 Alternatively, you can tell `patch' to place all of the rejected
2950 patches in a single file. The `-r REJECT-FILE' or
2951 `--reject-file=REJECT-FILE' option uses REJECT-FILE as the reject file
2955 File: diff.info, Node: patch Messages, Next: patch and POSIX, Prev: Reject Names, Up: Merging with patch
2957 10.11 Messages and Questions from `patch'
2958 =========================================
2960 `patch' can produce a variety of messages, especially if it has trouble
2961 decoding its input. In a few situations where it's not sure how to
2962 proceed, `patch' normally prompts you for more information from the
2963 keyboard. There are options to produce more or fewer messages, to have
2964 it not ask for keyboard input, and to affect the way that file names
2965 are quoted in messages.
2969 * More or Fewer Messages:: Controlling the verbosity of `patch'.
2970 * patch and Keyboard Input:: Inhibiting keyboard input.
2971 * patch Quoting Style:: Quoting file names in diagnostics.
2973 `patch' exits with status 0 if all hunks are applied successfully, 1
2974 if some hunks cannot be applied, and 2 if there is more serious trouble.
2975 When applying a set of patches in a loop, you should check the exit
2976 status, so you don't apply a later patch to a partially patched file.
2979 File: diff.info, Node: More or Fewer Messages, Next: patch and Keyboard Input, Up: patch Messages
2981 10.11.1 Controlling the Verbosity of `patch'
2982 --------------------------------------------
2984 You can cause `patch' to produce more messages by using the `--verbose'
2985 option. For example, when you give this option, the message `Hmm...'
2986 indicates that `patch' is reading text in the patch file, attempting to
2987 determine whether there is a patch in that text, and if so, what kind
2990 You can inhibit all terminal output from `patch', unless an error
2991 occurs, by using the `-s', `--quiet', or `--silent' option.
2994 File: diff.info, Node: patch and Keyboard Input, Next: patch Quoting Style, Prev: More or Fewer Messages, Up: patch Messages
2996 10.11.2 Inhibiting Keyboard Input
2997 ---------------------------------
2999 There are two ways you can prevent `patch' from asking you any
3000 questions. The `-f' or `--force' option assumes that you know what you
3001 are doing. It causes `patch' to do the following:
3003 * Skip patches that do not contain file names in their headers.
3005 * Patch files even though they have the wrong version for the
3006 `Prereq:' line in the patch;
3008 * Assume that patches are not reversed even if they look like they
3011 The `-t' or `--batch' option is similar to `-f', in that it suppresses
3012 questions, but it makes somewhat different assumptions:
3014 * Skip patches that do not contain file names in their headers (the
3017 * Skip patches for which the file has the wrong version for the
3018 `Prereq:' line in the patch;
3020 * Assume that patches are reversed if they look like they are.
3023 File: diff.info, Node: patch Quoting Style, Prev: patch and Keyboard Input, Up: patch Messages
3025 10.11.3 `patch' Quoting Style
3026 -----------------------------
3028 When `patch' outputs a file name in a diagnostic message, it can format
3029 the name in any of several ways. This can be useful to output file
3030 names unambiguously, even if they contain punctuation or special
3031 characters like newlines. The `--quoting-style=WORD' option controls
3032 how names are output. The WORD should be one of the following:
3038 Quote names for the shell if they contain shell metacharacters or
3039 would cause ambiguous output.
3042 Quote names for the shell, even if they would normally not require
3046 Quote names as for a C language string.
3049 Quote as with `c' except omit the surrounding double-quote
3052 You can specify the default value of the `--quoting-style' option
3053 with the environment variable `QUOTING_STYLE'. If that environment
3054 variable is not set, the default value is `shell', but this default may
3055 change in a future version of `patch'.
3058 File: diff.info, Node: patch and POSIX, Next: patch and Tradition, Prev: patch Messages, Up: Merging with patch
3060 10.12 `patch' and the POSIX Standard
3061 ====================================
3063 If you specify the `--posix' option, or set the `POSIXLY_CORRECT'
3064 environment variable, `patch' conforms more strictly to the POSIX
3065 standard, as follows:
3067 * Take the first existing file from the list (old, new, index) when
3068 intuiting file names from diff headers. *Note Multiple Patches::.
3070 * Do not remove files that are removed by a diff. *Note Creating
3073 * Do not ask whether to get files from RCS, ClearCase, or SCCS.
3074 *Note Revision Control::.
3076 * Require that all options precede the files in the command line.
3078 * Do not backup files, even when there is a mismatch. *Note
3083 File: diff.info, Node: patch and Tradition, Prev: patch and POSIX, Up: Merging with patch
3085 10.13 GNU `patch' and Traditional `patch'
3086 =========================================
3088 The current version of GNU `patch' normally follows the POSIX standard.
3089 *Note patch and POSIX::, for the few exceptions to this general rule.
3091 Unfortunately, POSIX redefined the behavior of `patch' in several
3092 important ways. You should be aware of the following differences if
3093 you must interoperate with traditional `patch', or with GNU `patch'
3094 version 2.1 and earlier.
3096 * In traditional `patch', the `-p' option's operand was optional,
3097 and a bare `-p' was equivalent to `-p0'. The `-p' option now
3098 requires an operand, and `-p 0' is now equivalent to `-p0'. For
3099 maximum compatibility, use options like `-p0' and `-p1'.
3101 Also, traditional `patch' simply counted slashes when stripping
3102 path prefixes; `patch' now counts pathname components. That is, a
3103 sequence of one or more adjacent slashes now counts as a single
3104 slash. For maximum portability, avoid sending patches containing
3107 * In traditional `patch', backups were enabled by default. This
3108 behavior is now enabled with the `-b' or `--backup' option.
3110 Conversely, in POSIX `patch', backups are never made, even when
3111 there is a mismatch. In GNU `patch', this behavior is enabled
3112 with the `--no-backup-if-mismatch' option, or by conforming to
3115 The `-b SUFFIX' option of traditional `patch' is equivalent to the
3116 `-b -z SUFFIX' options of GNU `patch'.
3118 * Traditional `patch' used a complicated (and incompletely
3119 documented) method to intuit the name of the file to be patched
3120 from the patch header. This method did not conform to POSIX, and
3121 had a few gotchas. Now `patch' uses a different, equally
3122 complicated (but better documented) method that is optionally
3123 POSIX-conforming; we hope it has fewer gotchas. The two methods
3124 are compatible if the file names in the context diff header and the
3125 `Index:' line are all identical after prefix-stripping. Your
3126 patch is normally compatible if each header's file names all
3127 contain the same number of slashes.
3129 * When traditional `patch' asked the user a question, it sent the
3130 question to standard error and looked for an answer from the first
3131 file in the following list that was a terminal: standard error,
3132 standard output, `/dev/tty', and standard input. Now `patch'
3133 sends questions to standard output and gets answers from
3134 `/dev/tty'. Defaults for some answers have been changed so that
3135 `patch' never goes into an infinite loop when using default
3138 * Traditional `patch' exited with a status value that counted the
3139 number of bad hunks, or with status 1 if there was real trouble.
3140 Now `patch' exits with status 1 if some hunks failed, or with 2 if
3141 there was real trouble.
3143 * Limit yourself to the following options when sending instructions
3144 meant to be executed by anyone running GNU `patch', traditional
3145 `patch', or a `patch' that conforms to POSIX. Spaces are
3146 significant in the following list, and operands are required.
3162 File: diff.info, Node: Making Patches, Next: Invoking cmp, Prev: Merging with patch, Up: Top
3164 11 Tips for Making and Using Patches
3165 ************************************
3167 Use some common sense when making and using patches. For example, when
3168 sending bug fixes to a program's maintainer, send several small
3169 patches, one per independent subject, instead of one large,
3170 harder-to-digest patch that covers all the subjects.
3172 Here are some other things you should keep in mind if you are going
3173 to distribute patches for updating a software package.
3177 * Tips for Patch Producers:: Advice for making patches.
3178 * Tips for Patch Consumers:: Advice for using patches.
3179 * Avoiding Common Mistakes:: Avoiding common mistakes when using `patch'.
3180 * Generating Smaller Patches:: How to generate smaller patches.
3183 File: diff.info, Node: Tips for Patch Producers, Next: Tips for Patch Consumers, Up: Making Patches
3185 11.1 Tips for Patch Producers
3186 =============================
3188 To create a patch that changes an older version of a package into a
3189 newer version, first make a copy of the older and newer versions in
3190 adjacent subdirectories. It is common to do that by unpacking `tar'
3191 archives of the two versions.
3193 To generate the patch, use the command `diff -Naur OLD NEW' where
3194 OLD and NEW identify the old and new directories. The names OLD and
3195 NEW should not contain any slashes. The `-N' option lets the patch
3196 create and remove files; `-a' lets the patch update non-text files; `-u'
3197 generates useful time stamps and enough context; and `-r' lets the
3198 patch update subdirectories. Here is an example command, using Bourne
3201 diff -Naur gcc-3.0.3 gcc-3.0.4
3203 Tell your recipients how to apply the patches. This should include
3204 which working directory to use, and which `patch' options to use; the
3205 option `-p1' is recommended. Test your procedure by pretending to be a
3206 recipient and applying your patches to a copy of the original files.
3208 *Note Avoiding Common Mistakes::, for how to avoid common mistakes
3209 when generating a patch.
3212 File: diff.info, Node: Tips for Patch Consumers, Next: Avoiding Common Mistakes, Prev: Tips for Patch Producers, Up: Making Patches
3214 11.2 Tips for Patch Consumers
3215 =============================
3217 A patch producer should tell recipients how to apply the patches, so
3218 the first rule of thumb for a patch consumer is to follow the
3219 instructions supplied with the patch.
3221 GNU `diff' can analyze files with arbitrarily long lines and files
3222 that end in incomplete lines. However, older versions of `patch'
3223 cannot patch such files. If you are having trouble applying such
3224 patches, try upgrading to a recent version of GNU `patch'.
3227 File: diff.info, Node: Avoiding Common Mistakes, Next: Generating Smaller Patches, Prev: Tips for Patch Consumers, Up: Making Patches
3229 11.3 Avoiding Common Mistakes
3230 =============================
3232 When producing a patch for multiple files, apply `diff' to directories
3233 whose names do not have slashes. This reduces confusion when the patch
3234 consumer specifies the `-pNUMBER' option, since this option can have
3235 surprising results when the old and new file names have different
3236 numbers of slashes. For example, do not send a patch with a header
3237 that looks like this:
3239 diff -Naur v2.0.29/prog/README prog/README
3240 --- v2.0.29/prog/README 2002-03-10 23:30:39.942229878 -0800
3241 +++ prog/README 2002-03-17 20:49:32.442260588 -0800
3243 because the two file names have different numbers of slashes, and
3244 different versions of `patch' interpret the file names differently. To
3245 avoid confusion, send output that looks like this instead:
3247 diff -Naur v2.0.29/prog/README v2.0.30/prog/README
3248 --- v2.0.29/prog/README 2002-03-10 23:30:39.942229878 -0800
3249 +++ v2.0.30/prog/README 2002-03-17 20:49:32.442260588 -0800
3251 Make sure you have specified the file names correctly, either in a
3252 context diff header or with an `Index:' line. Take care to not send out
3253 reversed patches, since these make people wonder whether they have
3254 already applied the patch.
3256 Avoid sending patches that compare backup file names like
3257 `README.orig' or `README~', since this might confuse `patch' into
3258 patching a backup file instead of the real file. Instead, send patches
3259 that compare the same base file names in different directories, e.g.
3260 `old/README' and `new/README'.
3262 To save people from partially applying a patch before other patches
3263 that should have gone before it, you can make the first patch in the
3264 patch file update a file with a name like `patchlevel.h' or
3265 `version.c', which contains a patch level or version number. If the
3266 input file contains the wrong version number, `patch' will complain
3269 An even clearer way to prevent this problem is to put a `Prereq:'
3270 line before the patch. If the leading text in the patch file contains a
3271 line that starts with `Prereq:', `patch' takes the next word from that
3272 line (normally a version number) and checks whether the next input file
3273 contains that word, preceded and followed by either white space or a
3274 newline. If not, `patch' prompts you for confirmation before
3275 proceeding. This makes it difficult to accidentally apply patches in
3279 File: diff.info, Node: Generating Smaller Patches, Prev: Avoiding Common Mistakes, Up: Making Patches
3281 11.4 Generating Smaller Patches
3282 ===============================
3284 The simplest way to generate a patch is to use `diff -Naur' (*note Tips
3285 for Patch Producers::), but you might be able to reduce the size of the
3286 patch by renaming or removing some files before making the patch. If
3287 the older version of the package contains any files that the newer
3288 version does not, or if any files have been renamed between the two
3289 versions, make a list of `rm' and `mv' commands for the user to execute
3290 in the old version directory before applying the patch. Then run those
3291 commands yourself in the scratch directory.
3293 If there are any files that you don't need to include in the patch
3294 because they can easily be rebuilt from other files (for example,
3295 `TAGS' and output from `yacc' and `makeinfo'), exclude them from the
3296 patch by giving `diff' the `-x PATTERN' option (*note Comparing
3297 Directories::). If you want your patch to modify a derived file
3298 because your recipients lack tools to build it, make sure that the
3299 patch for the derived file follows any patches for files that it
3300 depends on, so that the recipients' time stamps will not confuse `make'.
3302 Now you can create the patch using `diff -Naur'. Make sure to
3303 specify the scratch directory first and the newer directory second.
3305 Add to the top of the patch a note telling the user any `rm' and
3306 `mv' commands to run before applying the patch. Then you can remove
3307 the scratch directory.
3309 You can also shrink the patch size by using fewer lines of context,
3310 but bear in mind that `patch' typically needs at least two lines for
3311 proper operation when patches do not exactly match the input files.
3314 File: diff.info, Node: Invoking cmp, Next: Invoking diff, Prev: Making Patches, Up: Top
3319 The `cmp' command compares two files, and if they differ, tells the
3320 first byte and line number where they differ or reports that one file
3321 is a prefix of the other. Bytes and lines are numbered starting with
3322 1. The arguments of `cmp' are as follows:
3324 cmp OPTIONS... FROM-FILE [TO-FILE [FROM-SKIP [TO-SKIP]]]
3326 The file name `-' is always the standard input. `cmp' also uses the
3327 standard input if one file name is omitted. The FROM-SKIP and TO-SKIP
3328 operands specify how many bytes to ignore at the start of each file;
3329 they are equivalent to the `--ignore-initial=FROM-SKIP:TO-SKIP' option.
3331 By default, `cmp' outputs nothing if the two files have the same
3332 contents. If one file is a prefix of the other, `cmp' prints to
3333 standard error a message of the following form:
3335 cmp: EOF on SHORTER-FILE
3337 Otherwise, `cmp' prints to standard output a message of the
3340 FROM-FILE TO-FILE differ: char BYTE-NUMBER, line LINE-NUMBER
3342 The message formats can differ outside the POSIX locale. Also,
3343 POSIX allows the EOF message to be followed by a blank and some
3344 additional information.
3346 An exit status of 0 means no differences were found, 1 means some
3347 differences were found, and 2 means trouble.
3351 * cmp Options:: Summary of options to `cmp'.
3354 File: diff.info, Node: cmp Options, Up: Invoking cmp
3356 12.1 Options to `cmp'
3357 =====================
3359 Below is a summary of all of the options that GNU `cmp' accepts. Most
3360 options have two equivalent names, one of which is a single letter
3361 preceded by `-', and the other of which is a long name preceded by
3362 `--'. Multiple single letter options (unless they take an argument)
3363 can be combined into a single command line word: `-bl' is equivalent to
3368 Print the differing bytes. Display control bytes as a `^'
3369 followed by a letter of the alphabet and precede bytes that have
3370 the high bit set with `M-' (which stands for "meta").
3373 Output a summary of usage and then exit.
3376 `--ignore-initial=SKIP'
3377 Ignore any differences in the first SKIP bytes of the input files.
3378 Treat files with fewer than SKIP bytes as if they are empty. If
3379 SKIP is of the form `FROM-SKIP:TO-SKIP', skip the first FROM-SKIP
3380 bytes of the first input file and the first TO-SKIP bytes of the
3385 Output the (decimal) byte numbers and (octal) values of all
3386 differing bytes, instead of the default standard output. Also,
3387 output the EOF message if one file is shorter than the other.
3391 Compare at most COUNT input bytes.
3396 Do not print anything; only return an exit status indicating
3397 whether the files differ.
3401 Output version information and then exit.
3403 In the above table, operands that are byte counts are normally
3404 decimal, but may be preceded by `0' for octal and `0x' for hexadecimal.
3406 A byte count can be followed by a suffix to specify a multiple of
3407 that count; in this case an omitted integer is understood to be 1. A
3408 bare size letter, or one followed by `iB', specifies a multiple using
3409 powers of 1024. A size letter followed by `B' specifies powers of 1000
3410 instead. For example, `-n 4M' and `-n 4MiB' are equivalent to `-n
3411 4194304', whereas `-n 4MB' is equivalent to `-n 4000000'. This
3412 notation is upward compatible with the SI prefixes
3413 (http://www.bipm.fr/enus/3_SI/si-prefixes.html) for decimal multiples
3414 and with the IEC 60027-2 prefixes for binary multiples
3415 (http://physics.nist.gov/cuu/Units/binary.html).
3417 The following suffixes are defined. Large sizes like `1Y' may be
3418 rejected by your computer due to limitations of its arithmetic.
3421 kilobyte: 10^3 = 1000.
3426 kibibyte: 2^10 = 1024. `K' is special: the SI prefix is `k' and
3427 the IEC 60027-2 prefix is `Ki', but tradition and POSIX use `k' to
3431 megabyte: 10^6 = 1,000,000.
3435 mebibyte: 2^20 = 1,048,576.
3438 gigabyte: 10^9 = 1,000,000,000.
3442 gibibyte: 2^30 = 1,073,741,824.
3445 terabyte: 10^12 = 1,000,000,000,000.
3449 tebibyte: 2^40 = 1,099,511,627,776.
3452 petabyte: 10^15 = 1,000,000,000,000,000.
3456 pebibyte: 2^50 = 1,125,899,906,842,624.
3459 exabyte: 10^18 = 1,000,000,000,000,000,000.
3463 exbibyte: 2^60 = 1,152,921,504,606,846,976.
3466 zettabyte: 10^21 = 1,000,000,000,000,000,000,000
3470 2^70 = 1,180,591,620,717,411,303,424. (`Zi' is a GNU extension to
3474 yottabyte: 10^24 = 1,000,000,000,000,000,000,000,000.
3478 2^80 = 1,208,925,819,614,629,174,706,176. (`Yi' is a GNU
3479 extension to IEC 60027-2.)
3482 File: diff.info, Node: Invoking diff, Next: Invoking diff3, Prev: Invoking cmp, Up: Top
3487 The format for running the `diff' command is:
3489 diff OPTIONS... FILES...
3491 In the simplest case, two file names FROM-FILE and TO-FILE are
3492 given, and `diff' compares the contents of FROM-FILE and TO-FILE. A
3493 file name of `-' stands for text read from the standard input. As a
3494 special case, `diff - -' compares a copy of standard input to itself.
3496 If one file is a directory and the other is not, `diff' compares the
3497 file in the directory whose name is that of the non-directory. The
3498 non-directory file must not be `-'.
3500 If two file names are given and both are directories, `diff'
3501 compares corresponding files in both directories, in alphabetical
3502 order; this comparison is not recursive unless the `-r' or
3503 `--recursive' option is given. `diff' never compares the actual
3504 contents of a directory as if it were a file. The file that is fully
3505 specified may not be standard input, because standard input is nameless
3506 and the notion of "file with the same name" does not apply.
3508 If the `--from-file=FILE' option is given, the number of file names
3509 is arbitrary, and FILE is compared to each named file. Similarly, if
3510 the `--to-file=FILE' option is given, each named file is compared to
3513 `diff' options begin with `-', so normally file names may not begin
3514 with `-'. However, `--' as an argument by itself treats the remaining
3515 arguments as file names even if they begin with `-'.
3517 An exit status of 0 means no differences were found, 1 means some
3518 differences were found, and 2 means trouble. Normally, differing
3519 binary files count as trouble, but this can be altered by using the
3520 `-a' or `--text' option, or the `-q' or `--brief' option.
3524 * diff Options:: Summary of options to `diff'.
3527 File: diff.info, Node: diff Options, Up: Invoking diff
3529 13.1 Options to `diff'
3530 ======================
3532 Below is a summary of all of the options that GNU `diff' accepts. Most
3533 options have two equivalent names, one of which is a single letter
3534 preceded by `-', and the other of which is a long name preceded by
3535 `--'. Multiple single letter options (unless they take an argument)
3536 can be combined into a single command line word: `-ac' is equivalent to
3537 `-a -c'. Long named options can be abbreviated to any unique prefix of
3538 their name. Brackets ([ and ]) indicate that an option takes an
3543 Treat all files as text and compare them line-by-line, even if they
3544 do not seem to be text. *Note Binary::.
3547 `--ignore-space-change'
3548 Ignore changes in amount of white space. *Note White Space::.
3551 `--ignore-blank-lines'
3552 Ignore changes that just insert or delete blank lines. *Note
3556 Read and write data in binary mode. *Note Binary::.
3559 Use the context output format, showing three lines of context.
3560 *Note Context Format::.
3564 Use the context output format, showing LINES (an integer) lines of
3565 context, or three if LINES is not given. *Note Context Format::.
3566 For proper operation, `patch' typically needs at least two lines of
3569 For compatibility `diff' also supports an obsolete option syntax
3570 `-LINES' that has effect when combined with `-c', `-p', or `-u'.
3571 New scripts should use `-C LINES' or `-U LINES' instead.
3573 `--changed-group-format=FORMAT'
3574 Use FORMAT to output a line group containing differing lines from
3575 both files in if-then-else format. *Note Line Group Formats::.
3579 Change the algorithm perhaps find a smaller set of changes. This
3580 makes `diff' slower (sometimes much slower). *Note diff
3585 Make merged `#ifdef' format output, conditional on the preprocessor
3586 macro NAME. *Note If-then-else::.
3590 Make output that is a valid `ed' script. *Note ed Scripts::.
3593 `--ignore-tab-expansion'
3594 Ignore changes due to tab expansion. *Note White Space::.
3598 Make output that looks vaguely like an `ed' script but has changes
3599 in the order they appear in the file. *Note Forward ed::.
3602 `--show-function-line=REGEXP'
3603 In context and unified format, for each hunk of differences, show
3604 some of the last preceding line that matches REGEXP. *Note
3605 Specified Headings::.
3608 Compare FILE to each operand; FILE may be a directory.
3611 Output a summary of usage and then exit.
3613 `--horizon-lines=LINES'
3614 Do not discard the last LINES lines of the common prefix and the
3615 first LINES lines of the common suffix. *Note diff Performance::.
3619 Ignore changes in case; consider upper- and lower-case letters
3620 equivalent. *Note Case Folding::.
3623 `--ignore-matching-lines=REGEXP'
3624 Ignore changes that just insert or delete lines that match REGEXP.
3625 *Note Specified Lines::.
3627 `--ignore-file-name-case'
3628 Ignore case when comparing file names during recursive comparison.
3629 *Note Comparing Directories::.
3633 Pass the output through `pr' to paginate it. *Note Pagination::.
3636 Use LABEL instead of the file name in the context format (*note
3637 Context Format::) and unified format (*note Unified Format::)
3638 headers. *Note RCS::.
3641 Print only the left column of two common lines in side by side
3642 format. *Note Side by Side Format::.
3644 `--line-format=FORMAT'
3645 Use FORMAT to output all input lines in if-then-else format.
3646 *Note Line Formats::.
3650 Output RCS-format diffs; like `-f' except that each command
3651 specifies the number of lines affected. *Note RCS::.
3655 In directory comparison, if a file is found in only one directory,
3656 treat it as present but empty in the other directory. *Note
3657 Comparing Directories::.
3659 `--new-group-format=FORMAT'
3660 Use FORMAT to output a group of lines taken from just the second
3661 file in if-then-else format. *Note Line Group Formats::.
3663 `--new-line-format=FORMAT'
3664 Use FORMAT to output a line taken from just the second file in
3665 if-then-else format. *Note Line Formats::.
3667 `--old-group-format=FORMAT'
3668 Use FORMAT to output a group of lines taken from just the first
3669 file in if-then-else format. *Note Line Group Formats::.
3671 `--old-line-format=FORMAT'
3672 Use FORMAT to output a line taken from just the first file in
3673 if-then-else format. *Note Line Formats::.
3677 Show which C function each change is in. *Note C Function
3682 Report only whether the files differ, not the details of the
3683 differences. *Note Brief::.
3687 When comparing directories, recursively compare any subdirectories
3688 found. *Note Comparing Directories::.
3691 `--report-identical-files'
3692 Report when two files are the same. *Note Comparing Directories::.
3695 `--starting-file=FILE'
3696 When comparing directories, start with the file FILE. This is
3697 used for resuming an aborted comparison. *Note Comparing
3700 `--speed-large-files'
3701 Use heuristics to speed handling of large files that have numerous
3702 scattered small changes. *Note diff Performance::.
3704 `--strip-trailing-cr'
3705 Strip any trailing carriage return at the end of an input line.
3708 `--suppress-common-lines'
3709 Do not print common lines in side by side format. *Note Side by
3714 Expand tabs to spaces in the output, to preserve the alignment of
3715 tabs in the input files. *Note Tabs::.
3719 Output a tab rather than a space before the text of a line in
3720 normal or context format. This causes the alignment of tabs in
3721 the line to look normal. *Note Tabs::.
3724 Assume that tab stops are set every COLUMNS (default 8) print
3725 columns. *Note Tabs::.
3727 `--suppress-blank-empty'
3728 Suppress any blanks before newlines when printing the
3729 representation of an empty line, when outputting normal, context,
3730 or unified format. *Note Trailing Blanks::.
3733 Compare each operand to FILE; FILE may be a directory.
3736 Use the unified output format, showing three lines of context.
3737 *Note Unified Format::.
3739 `--unchanged-group-format=FORMAT'
3740 Use FORMAT to output a group of common lines taken from both files
3741 in if-then-else format. *Note Line Group Formats::.
3743 `--unchanged-line-format=FORMAT'
3744 Use FORMAT to output a line common to both files in if-then-else
3745 format. *Note Line Formats::.
3747 `--unidirectional-new-file'
3748 When comparing directories, if a file appears only in the second
3749 directory of the two, treat it as present but empty in the other.
3750 *Note Comparing Directories::.
3754 Use the unified output format, showing LINES (an integer) lines of
3755 context, or three if LINES is not given. *Note Unified Format::.
3756 For proper operation, `patch' typically needs at least two lines of
3759 On older systems, `diff' supports an obsolete option `-LINES' that
3760 has effect when combined with `-u'. POSIX 1003.1-2001 (*note
3761 Standards conformance::) does not allow this; use `-U LINES'
3766 Output version information and then exit.
3769 `--ignore-all-space'
3770 Ignore white space when comparing lines. *Note White Space::.
3774 Output at most COLUMNS (default 130) print columns per line in
3775 side by side format. *Note Side by Side Format::.
3779 When comparing directories, ignore files and subdirectories whose
3780 basenames match PATTERN. *Note Comparing Directories::.
3783 `--exclude-from=FILE'
3784 When comparing directories, ignore files and subdirectories whose
3785 basenames match any pattern contained in FILE. *Note Comparing
3790 Use the side by side output format. *Note Side by Side Format::.
3793 File: diff.info, Node: Invoking diff3, Next: Invoking patch, Prev: Invoking diff, Up: Top
3798 The `diff3' command compares three files and outputs descriptions of
3799 their differences. Its arguments are as follows:
3801 diff3 OPTIONS... MINE OLDER YOURS
3803 The files to compare are MINE, OLDER, and YOURS. At most one of
3804 these three file names may be `-', which tells `diff3' to read the
3805 standard input for that file.
3807 An exit status of 0 means `diff3' was successful, 1 means some
3808 conflicts were found, and 2 means trouble.
3812 * diff3 Options:: Summary of options to `diff3'.
3815 File: diff.info, Node: diff3 Options, Up: Invoking diff3
3817 14.1 Options to `diff3'
3818 =======================
3820 Below is a summary of all of the options that GNU `diff3' accepts.
3821 Multiple single letter options (unless they take an argument) can be
3822 combined into a single command line argument.
3826 Treat all files as text and compare them line-by-line, even if they
3827 do not appear to be text. *Note Binary::.
3831 Incorporate all unmerged changes from OLDER to YOURS into MINE,
3832 surrounding conflicts with bracket lines. *Note Marking
3835 `--diff-program=PROGRAM'
3836 Use the compatible comparison program PROGRAM to compare files
3841 Generate an `ed' script that incorporates all the changes from
3842 OLDER to YOURS into MINE. *Note Which Changes::.
3846 Like `-e', except bracket lines from overlapping changes' first
3847 and third files. *Note Marking Conflicts::. With `-E', an
3848 overlapping change looks like this:
3857 Output a summary of usage and then exit.
3860 Generate `w' and `q' commands at the end of the `ed' script for
3861 System V compatibility. This option must be combined with one of
3862 the `-AeExX3' options, and may not be combined with `-m'. *Note
3863 Saving the Changed File::.
3866 Use the label LABEL for the brackets output by the `-A', `-E' and
3867 `-X' options. This option may be given up to three times, one for
3868 each input file. The default labels are the names of the input
3869 files. Thus `diff3 --label X --label Y --label Z -m A B C' acts
3870 like `diff3 -m A B C', except that the output looks like it came
3871 from files named `X', `Y' and `Z' rather than from files named
3872 `A', `B' and `C'. *Note Marking Conflicts::.
3876 Apply the edit script to the first file and send the result to
3877 standard output. Unlike piping the output from `diff3' to `ed',
3878 this works even for binary files and incomplete lines. `-A' is
3879 assumed if no edit script option is specified. *Note Bypassing
3882 `--strip-trailing-cr'
3883 Strip any trailing carriage return at the end of an input line.
3888 Output a tab rather than two spaces before the text of a line in
3889 normal format. This causes the alignment of tabs in the line to
3890 look normal. *Note Tabs::.
3894 Output version information and then exit.
3898 Like `-e', except output only the overlapping changes. *Note
3902 Like `-E', except output only the overlapping changes. In other
3903 words, like `-x', except bracket changes as in `-E'. *Note
3904 Marking Conflicts::.
3908 Like `-e', except output only the nonoverlapping changes. *Note
3912 File: diff.info, Node: Invoking patch, Next: Invoking sdiff, Prev: Invoking diff3, Up: Top
3917 Normally `patch' is invoked like this:
3921 The full format for invoking `patch' is:
3923 patch OPTIONS... [ORIGFILE [PATCHFILE]]
3925 You can also specify where to read the patch from with the `-i
3926 PATCHFILE' or `--input=PATCHFILE' option. If you do not specify
3927 PATCHFILE, or if PATCHFILE is `-', `patch' reads the patch (that is,
3928 the `diff' output) from the standard input.
3930 If you do not specify an input file on the command line, `patch'
3931 tries to intuit from the "leading text" (any text in the patch that
3932 comes before the `diff' output) which file to edit. *Note Multiple
3935 By default, `patch' replaces the original input file with the
3936 patched version, possibly after renaming the original file into a
3937 backup file (*note Backup Names::, for a description of how `patch'
3938 names backup files). You can also specify where to put the output with
3939 the `-o FILE' or `--output=FILE' option; however, do not use this option
3940 if FILE is one of the input files.
3944 * patch Options:: Summary table of options to `patch'.
3947 File: diff.info, Node: patch Options, Up: Invoking patch
3949 15.1 Options to `patch'
3950 =======================
3952 Here is a summary of all of the options that GNU `patch' accepts.
3953 *Note patch and Tradition::, for which of these options are safe to use
3954 in older versions of `patch'.
3956 Multiple single-letter options that do not take an argument can be
3957 combined into a single command line argument with only one dash.
3961 Back up the original contents of each file, even if backups would
3962 normally not be made. *Note Backups::.
3966 Prepend PREFIX to backup file names. *Note Backup Names::.
3968 `--backup-if-mismatch'
3969 Back up the original contents of each file if the patch does not
3970 exactly match the file. This is the default behavior when not
3971 conforming to POSIX. *Note Backups::.
3974 Read and write all files in binary mode, except for standard output
3975 and `/dev/tty'. This option has no effect on POSIX-conforming
3976 systems like GNU/Linux. On systems where this option makes a
3977 difference, the patch should be generated by `diff -a --binary'.
3982 Interpret the patch file as a context diff. *Note patch Input::.
3985 `--directory=DIRECTORY'
3986 Make directory DIRECTORY the current directory for interpreting
3987 both file names in the patch file, and file names given as
3988 arguments to other options. *Note patch Directories::.
3992 Make merged if-then-else output using NAME. *Note If-then-else::.
3995 Print the results of applying the patches without actually changing
3996 any files. *Note Dry Runs::.
4000 Interpret the patch file as an `ed' script. *Note patch Input::.
4003 `--remove-empty-files'
4004 Remove output files that are empty after the patches have been
4005 applied. *Note Creating and Removing::.
4009 Assume that the user knows exactly what he or she is doing, and do
4010 not ask any questions. *Note patch Messages::.
4014 Set the maximum fuzz factor to LINES. *Note Inexact::.
4018 If NUM is positive, get input files from a revision control system
4019 as necessary; if zero, do not get the files; if negative, ask the
4020 user whether to get the files. *Note Revision Control::.
4023 Output a summary of usage and then exit.
4027 Read the patch from PATCHFILE rather than from standard input.
4028 *Note patch Options::.
4031 `--ignore-white-space'
4032 Let any sequence of blanks (spaces or tabs) in the patch file match
4033 any sequence of blanks in the input file. *Note Changed White
4038 Interpret the patch file as a normal diff. *Note patch Input::.
4042 Ignore patches that `patch' thinks are reversed or already applied.
4043 See also `-R'. *Note Reversed Patches::.
4045 `--no-backup-if-mismatch'
4046 Do not back up the original contents of files. This is the default
4047 behavior when conforming to POSIX. *Note Backups::.
4051 Use FILE as the output file name. *Note patch Options::.
4055 Set the file name strip count to NUMBER. *Note patch
4059 Conform to POSIX, as if the `POSIXLY_CORRECT' environment variable
4060 had been set. *Note patch and POSIX::.
4062 `--quoting-style=WORD'
4063 Use style WORD to quote names in diagnostics, as if the
4064 `QUOTING_STYLE' environment variable had been set to WORD. *Note
4065 patch Quoting Style::.
4068 `--reject-file=REJECT-FILE'
4069 Use REJECT-FILE as the reject file name. *Note Reject Names::.
4073 Assume that this patch was created with the old and new files
4074 swapped. *Note Reversed Patches::.
4079 Work silently unless an error occurs. *Note patch Messages::.
4083 Do not ask any questions. *Note patch Messages::.
4087 Set the modification and access times of patched files from time
4088 stamps given in context diff headers, assuming that the context
4089 diff headers use local time. *Note Patching Time Stamps::.
4093 Interpret the patch file as a unified diff. *Note patch Input::.
4097 Output version information and then exit.
4100 `--version=control=BACKUP-STYLE'
4101 Select the naming convention for backup file names. *Note Backup
4105 Print more diagnostics than usual. *Note patch Messages::.
4109 Set internal debugging flags. Of interest only to `patch'
4113 `--basename-prefix=PREFIX'
4114 Prepend PREFIX to base names of backup files. *Note Backup
4119 Use SUFFIX as the backup extension instead of `.orig' or `~'.
4120 *Note Backup Names::.
4124 Set the modification and access times of patched files from time
4125 stamps given in context diff headers, assuming that the context
4126 diff headers use UTC. *Note Patching Time Stamps::.
4130 File: diff.info, Node: Invoking sdiff, Next: Standards conformance, Prev: Invoking patch, Up: Top
4135 The `sdiff' command merges two files and interactively outputs the
4136 results. Its arguments are as follows:
4138 sdiff -o OUTFILE OPTIONS... FROM-FILE TO-FILE
4140 This merges FROM-FILE with TO-FILE, with output to OUTFILE. If
4141 FROM-FILE is a directory and TO-FILE is not, `sdiff' compares the file
4142 in FROM-FILE whose file name is that of TO-FILE, and vice versa.
4143 FROM-FILE and TO-FILE may not both be directories.
4145 `sdiff' options begin with `-', so normally FROM-FILE and TO-FILE
4146 may not begin with `-'. However, `--' as an argument by itself treats
4147 the remaining arguments as file names even if they begin with `-'. You
4148 may not use `-' as an input file.
4150 `sdiff' without `-o' (or `--output') produces a side-by-side
4151 difference. This usage is obsolete; use the `-y' or `--side-by-side'
4152 option of `diff' instead.
4154 An exit status of 0 means no differences were found, 1 means some
4155 differences were found, and 2 means trouble.
4159 * sdiff Options:: Summary of options to `diff'.
4162 File: diff.info, Node: sdiff Options, Up: Invoking sdiff
4164 16.1 Options to `sdiff'
4165 =======================
4167 Below is a summary of all of the options that GNU `sdiff' accepts.
4168 Each option has two equivalent names, one of which is a single letter
4169 preceded by `-', and the other of which is a long name preceded by
4170 `--'. Multiple single letter options (unless they take an argument)
4171 can be combined into a single command line argument. Long named
4172 options can be abbreviated to any unique prefix of their name.
4176 Treat all files as text and compare them line-by-line, even if they
4177 do not appear to be text. *Note Binary::.
4180 `--ignore-space-change'
4181 Ignore changes in amount of white space. *Note White Space::.
4184 `--ignore-blank-lines'
4185 Ignore changes that just insert or delete blank lines. *Note
4190 Change the algorithm to perhaps find a smaller set of changes.
4191 This makes `sdiff' slower (sometimes much slower). *Note diff
4194 `--diff-program=PROGRAM'
4195 Use the compatible comparison program PROGRAM to compare files
4199 `--ignore-tab-expansion'
4200 Ignore changes due to tab expansion. *Note White Space::.
4203 Output a summary of usage and then exit.
4207 Ignore changes in case; consider upper- and lower-case to be the
4208 same. *Note Case Folding::.
4211 `--ignore-matching-lines=REGEXP'
4212 Ignore changes that just insert or delete lines that match REGEXP.
4213 *Note Specified Lines::.
4217 Print only the left column of two common lines. *Note Side by
4222 Put merged output into FILE. This option is required for merging.
4225 `--suppress-common-lines'
4226 Do not print common lines. *Note Side by Side Format::.
4228 `--speed-large-files'
4229 Use heuristics to speed handling of large files that have numerous
4230 scattered small changes. *Note diff Performance::.
4232 `--strip-trailing-cr'
4233 Strip any trailing carriage return at the end of an input line.
4238 Expand tabs to spaces in the output, to preserve the alignment of
4239 tabs in the input files. *Note Tabs::.
4242 Assume that tab stops are set every COLUMNS (default 8) print
4243 columns. *Note Tabs::.
4247 Output version information and then exit.
4251 Output at most COLUMNS (default 130) print columns per line.
4252 *Note Side by Side Format::. Note that for historical reasons,
4253 this option is `-W' in `diff', `-w' in `sdiff'.
4256 `--ignore-all-space'
4257 Ignore white space when comparing lines. *Note White Space::.
4258 Note that for historical reasons, this option is `-w' in `diff',
4262 File: diff.info, Node: Standards conformance, Next: Projects, Prev: Invoking sdiff, Up: Top
4264 17 Standards conformance
4265 ************************
4267 In a few cases, the GNU utilities' default behavior is incompatible
4268 with the POSIX standard. To suppress these incompatibilities, define
4269 the `POSIXLY_CORRECT' environment variable. Unless you are checking
4270 for POSIX conformance, you probably do not need to define
4273 Normally options and operands can appear in any order, and programs
4274 act as if all the options appear before any operands. For example,
4275 `diff lao tzu -C 2' acts like `diff -C 2 lao tzu', since `2' is an
4276 option-argument of `-C'. However, if the `POSIXLY_CORRECT' environment
4277 variable is set, options must appear before operands, unless otherwise
4278 specified for a particular command.
4280 Newer versions of POSIX are occasionally incompatible with older
4281 versions. For example, older versions of POSIX allowed the command
4282 `diff -c -10' to have the same meaning as `diff -C 10', but POSIX
4283 1003.1-2001 `diff' no longer allows digit-string options like `-10'.
4285 The GNU utilities normally conform to the version of POSIX that is
4286 standard for your system. To cause them to conform to a different
4287 version of POSIX, define the `_POSIX2_VERSION' environment variable to
4288 a value of the form YYYYMM specifying the year and month the standard
4289 was adopted. Two values are currently supported for `_POSIX2_VERSION':
4290 `199209' stands for POSIX 1003.2-1992, and `200112' stands for POSIX
4291 1003.1-2001. For example, if you are running older software that
4292 assumes an older version of POSIX and uses `diff -c -10', you can work
4293 around the compatibility problems by setting `_POSIX2_VERSION=199209'
4294 in your environment.
4297 File: diff.info, Node: Projects, Next: Copying This Manual, Prev: Standards conformance, Up: Top
4302 Here are some ideas for improving GNU `diff' and `patch'. The GNU
4303 project has identified some improvements as potential programming
4304 projects for volunteers. You can also help by reporting any bugs that
4307 If you are a programmer and would like to contribute something to the
4308 GNU project, please consider volunteering for one of these projects.
4309 If you are seriously contemplating work, please write to <gvc@gnu.org>
4310 to coordinate with other volunteers.
4314 * Shortcomings:: Suggested projects for improvements.
4315 * Bugs:: Reporting bugs.
4318 File: diff.info, Node: Shortcomings, Next: Bugs, Up: Projects
4320 18.1 Suggested Projects for Improving GNU `diff' and `patch'
4321 ============================================================
4323 One should be able to use GNU `diff' to generate a patch from any pair
4324 of directory trees, and given the patch and a copy of one such tree,
4325 use `patch' to generate a faithful copy of the other. Unfortunately,
4326 some changes to directory trees cannot be expressed using current patch
4327 formats; also, `patch' does not handle some of the existing formats.
4328 These shortcomings motivate the following suggested projects.
4332 * Internationalization:: Handling multibyte and varying-width characters.
4333 * Changing Structure:: Handling changes to the directory structure.
4334 * Special Files:: Handling symbolic links, device special files, etc.
4335 * Unusual File Names:: Handling file names that contain unusual characters.
4336 * Time Stamp Order:: Outputting diffs in time stamp order.
4337 * Ignoring Changes:: Ignoring certain changes while showing others.
4338 * Speedups:: Improving performance.
4341 File: diff.info, Node: Internationalization, Next: Changing Structure, Up: Shortcomings
4343 18.1.1 Handling Multibyte and Varying-Width Characters
4344 ------------------------------------------------------
4346 `diff', `diff3' and `sdiff' treat each line of input as a string of
4347 unibyte characters. This can mishandle multibyte characters in some
4348 cases. For example, when asked to ignore spaces, `diff' does not
4349 properly ignore a multibyte space character.
4351 Also, `diff' currently assumes that each byte is one column wide,
4352 and this assumption is incorrect in some locales, e.g., locales that
4353 use UTF-8 encoding. This causes problems with the `-y' or
4354 `--side-by-side' option of `diff'.
4356 These problems need to be fixed without unduly affecting the
4357 performance of the utilities in unibyte environments.
4359 The IBM GNU/Linux Technology Center Internationalization Team has
4360 proposed patches to support internationalized `diff'
4361 (http://oss.software.ibm.com/developer/opensource/linux/patches/i18n/diffutils-2.7.2-i18n-0.1.patch.gz).
4362 Unfortunately, these patches are incomplete and are to an older version
4363 of `diff', so more work needs to be done in this area.
4366 File: diff.info, Node: Changing Structure, Next: Special Files, Prev: Internationalization, Up: Shortcomings
4368 18.1.2 Handling Changes to the Directory Structure
4369 --------------------------------------------------
4371 `diff' and `patch' do not handle some changes to directory structure.
4372 For example, suppose one directory tree contains a directory named `D'
4373 with some subsidiary files, and another contains a file with the same
4374 name `D'. `diff -r' does not output enough information for `patch' to
4375 transform the directory subtree into the file.
4377 There should be a way to specify that a file has been removed without
4378 having to include its entire contents in the patch file. There should
4379 also be a way to tell `patch' that a file was renamed, even if there is
4380 no way for `diff' to generate such information. There should be a way
4381 to tell `patch' that a file's time stamp has changed, even if its
4382 contents have not changed.
4384 These problems can be fixed by extending the `diff' output format to
4385 represent changes in directory structure, and extending `patch' to
4386 understand these extensions.
4389 File: diff.info, Node: Special Files, Next: Unusual File Names, Prev: Changing Structure, Up: Shortcomings
4391 18.1.3 Files that are Neither Directories Nor Regular Files
4392 -----------------------------------------------------------
4394 Some files are neither directories nor regular files: they are unusual
4395 files like symbolic links, device special files, named pipes, and
4396 sockets. Currently, `diff' treats symbolic links as if they were the
4397 pointed-to files, except that a recursive `diff' reports an error if it
4398 detects infinite loops of symbolic links (e.g., symbolic links to
4399 `..'). `diff' treats other special files like regular files if they
4400 are specified at the top level, but simply reports their presence when
4401 comparing directories. This means that `patch' cannot represent
4402 changes to such files. For example, if you change which file a
4403 symbolic link points to, `diff' outputs the difference between the two
4404 files, instead of the change to the symbolic link.
4406 `diff' should optionally report changes to special files specially,
4407 and `patch' should be extended to understand these extensions.
4410 File: diff.info, Node: Unusual File Names, Next: Time Stamp Order, Prev: Special Files, Up: Shortcomings
4412 18.1.4 File Names that Contain Unusual Characters
4413 -------------------------------------------------
4415 When a file name contains an unusual character like a newline or white
4416 space, `diff -r' generates a patch that `patch' cannot parse. The
4417 problem is with format of `diff' output, not just with `patch', because
4418 with odd enough file names one can cause `diff' to generate a patch
4419 that is syntactically correct but patches the wrong files. The format
4420 of `diff' output should be extended to handle all possible file names.
4423 File: diff.info, Node: Time Stamp Order, Next: Ignoring Changes, Prev: Unusual File Names, Up: Shortcomings
4425 18.1.5 Outputting Diffs in Time Stamp Order
4426 -------------------------------------------
4428 Applying `patch' to a multiple-file diff can result in files whose time
4429 stamps are out of order. GNU `patch' has options to restore the time
4430 stamps of the updated files (*note Patching Time Stamps::), but
4431 sometimes it is useful to generate a patch that works even if the
4432 recipient does not have GNU patch, or does not use these options. One
4433 way to do this would be to implement a `diff' option to output diffs in
4437 File: diff.info, Node: Ignoring Changes, Next: Speedups, Prev: Time Stamp Order, Up: Shortcomings
4439 18.1.6 Ignoring Certain Changes
4440 -------------------------------
4442 It would be nice to have a feature for specifying two strings, one in
4443 FROM-FILE and one in TO-FILE, which should be considered to match.
4444 Thus, if the two strings are `foo' and `bar', then if two lines differ
4445 only in that `foo' in file 1 corresponds to `bar' in file 2, the lines
4446 are treated as identical.
4448 It is not clear how general this feature can or should be, or what
4449 syntax should be used for it.
4451 A partial substitute is to filter one or both files before comparing,
4454 sed 's/foo/bar/g' file1 | diff - file2
4456 However, this outputs the filtered text, not the original.
4459 File: diff.info, Node: Speedups, Prev: Ignoring Changes, Up: Shortcomings
4461 18.1.7 Improving Performance
4462 ----------------------------
4464 When comparing two large directory structures, one of which was
4465 originally copied from the other with time stamps preserved (e.g., with
4466 `cp -pR'), it would greatly improve performance if an option told
4467 `diff' to assume that two files with the same size and time stamps have
4468 the same content. *Note diff Performance::.
4471 File: diff.info, Node: Bugs, Prev: Shortcomings, Up: Projects
4476 If you think you have found a bug in GNU `cmp', `diff', `diff3', or
4477 `sdiff', please report it by electronic mail to the GNU utilities bug
4479 (http://mail.gnu.org/mailman/listinfo/bug-gnu-utils)
4480 <bug-gnu-utils@gnu.org>. Please send bug reports for GNU `patch' to
4481 <bug-patch@gnu.org>. Send as precise a description of the problem as
4482 you can, including the output of the `--version' option and sample
4483 input files that produce the bug, if applicable. If you have a
4484 nontrivial fix for the bug, please send it as well. If you have a
4485 patch, please send it too. It may simplify the maintainer's job if the
4486 patch is relative to a recent test release, which you can find in the
4487 directory `ftp://alpha.gnu.org/gnu/diffutils/'.
4490 File: diff.info, Node: Copying This Manual, Next: Translations, Prev: Projects, Up: Top
4492 Appendix A Copying This Manual
4493 ******************************
4495 Version 1.3, 3 November 2008
4497 Copyright (C) 2000, 2001, 2002, 2007, 2008 Free Software Foundation, Inc.
4500 Everyone is permitted to copy and distribute verbatim copies
4501 of this license document, but changing it is not allowed.
4505 The purpose of this License is to make a manual, textbook, or other
4506 functional and useful document "free" in the sense of freedom: to
4507 assure everyone the effective freedom to copy and redistribute it,
4508 with or without modifying it, either commercially or
4509 noncommercially. Secondarily, this License preserves for the
4510 author and publisher a way to get credit for their work, while not
4511 being considered responsible for modifications made by others.
4513 This License is a kind of "copyleft", which means that derivative
4514 works of the document must themselves be free in the same sense.
4515 It complements the GNU General Public License, which is a copyleft
4516 license designed for free software.
4518 We have designed this License in order to use it for manuals for
4519 free software, because free software needs free documentation: a
4520 free program should come with manuals providing the same freedoms
4521 that the software does. But this License is not limited to
4522 software manuals; it can be used for any textual work, regardless
4523 of subject matter or whether it is published as a printed book.
4524 We recommend this License principally for works whose purpose is
4525 instruction or reference.
4527 1. APPLICABILITY AND DEFINITIONS
4529 This License applies to any manual or other work, in any medium,
4530 that contains a notice placed by the copyright holder saying it
4531 can be distributed under the terms of this License. Such a notice
4532 grants a world-wide, royalty-free license, unlimited in duration,
4533 to use that work under the conditions stated herein. The
4534 "Document", below, refers to any such manual or work. Any member
4535 of the public is a licensee, and is addressed as "you". You
4536 accept the license if you copy, modify or distribute the work in a
4537 way requiring permission under copyright law.
4539 A "Modified Version" of the Document means any work containing the
4540 Document or a portion of it, either copied verbatim, or with
4541 modifications and/or translated into another language.
4543 A "Secondary Section" is a named appendix or a front-matter section
4544 of the Document that deals exclusively with the relationship of the
4545 publishers or authors of the Document to the Document's overall
4546 subject (or to related matters) and contains nothing that could
4547 fall directly within that overall subject. (Thus, if the Document
4548 is in part a textbook of mathematics, a Secondary Section may not
4549 explain any mathematics.) The relationship could be a matter of
4550 historical connection with the subject or with related matters, or
4551 of legal, commercial, philosophical, ethical or political position
4554 The "Invariant Sections" are certain Secondary Sections whose
4555 titles are designated, as being those of Invariant Sections, in
4556 the notice that says that the Document is released under this
4557 License. If a section does not fit the above definition of
4558 Secondary then it is not allowed to be designated as Invariant.
4559 The Document may contain zero Invariant Sections. If the Document
4560 does not identify any Invariant Sections then there are none.
4562 The "Cover Texts" are certain short passages of text that are
4563 listed, as Front-Cover Texts or Back-Cover Texts, in the notice
4564 that says that the Document is released under this License. A
4565 Front-Cover Text may be at most 5 words, and a Back-Cover Text may
4566 be at most 25 words.
4568 A "Transparent" copy of the Document means a machine-readable copy,
4569 represented in a format whose specification is available to the
4570 general public, that is suitable for revising the document
4571 straightforwardly with generic text editors or (for images
4572 composed of pixels) generic paint programs or (for drawings) some
4573 widely available drawing editor, and that is suitable for input to
4574 text formatters or for automatic translation to a variety of
4575 formats suitable for input to text formatters. A copy made in an
4576 otherwise Transparent file format whose markup, or absence of
4577 markup, has been arranged to thwart or discourage subsequent
4578 modification by readers is not Transparent. An image format is
4579 not Transparent if used for any substantial amount of text. A
4580 copy that is not "Transparent" is called "Opaque".
4582 Examples of suitable formats for Transparent copies include plain
4583 ASCII without markup, Texinfo input format, LaTeX input format,
4584 SGML or XML using a publicly available DTD, and
4585 standard-conforming simple HTML, PostScript or PDF designed for
4586 human modification. Examples of transparent image formats include
4587 PNG, XCF and JPG. Opaque formats include proprietary formats that
4588 can be read and edited only by proprietary word processors, SGML or
4589 XML for which the DTD and/or processing tools are not generally
4590 available, and the machine-generated HTML, PostScript or PDF
4591 produced by some word processors for output purposes only.
4593 The "Title Page" means, for a printed book, the title page itself,
4594 plus such following pages as are needed to hold, legibly, the
4595 material this License requires to appear in the title page. For
4596 works in formats which do not have any title page as such, "Title
4597 Page" means the text near the most prominent appearance of the
4598 work's title, preceding the beginning of the body of the text.
4600 The "publisher" means any person or entity that distributes copies
4601 of the Document to the public.
4603 A section "Entitled XYZ" means a named subunit of the Document
4604 whose title either is precisely XYZ or contains XYZ in parentheses
4605 following text that translates XYZ in another language. (Here XYZ
4606 stands for a specific section name mentioned below, such as
4607 "Acknowledgements", "Dedications", "Endorsements", or "History".)
4608 To "Preserve the Title" of such a section when you modify the
4609 Document means that it remains a section "Entitled XYZ" according
4612 The Document may include Warranty Disclaimers next to the notice
4613 which states that this License applies to the Document. These
4614 Warranty Disclaimers are considered to be included by reference in
4615 this License, but only as regards disclaiming warranties: any other
4616 implication that these Warranty Disclaimers may have is void and
4617 has no effect on the meaning of this License.
4621 You may copy and distribute the Document in any medium, either
4622 commercially or noncommercially, provided that this License, the
4623 copyright notices, and the license notice saying this License
4624 applies to the Document are reproduced in all copies, and that you
4625 add no other conditions whatsoever to those of this License. You
4626 may not use technical measures to obstruct or control the reading
4627 or further copying of the copies you make or distribute. However,
4628 you may accept compensation in exchange for copies. If you
4629 distribute a large enough number of copies you must also follow
4630 the conditions in section 3.
4632 You may also lend copies, under the same conditions stated above,
4633 and you may publicly display copies.
4635 3. COPYING IN QUANTITY
4637 If you publish printed copies (or copies in media that commonly
4638 have printed covers) of the Document, numbering more than 100, and
4639 the Document's license notice requires Cover Texts, you must
4640 enclose the copies in covers that carry, clearly and legibly, all
4641 these Cover Texts: Front-Cover Texts on the front cover, and
4642 Back-Cover Texts on the back cover. Both covers must also clearly
4643 and legibly identify you as the publisher of these copies. The
4644 front cover must present the full title with all words of the
4645 title equally prominent and visible. You may add other material
4646 on the covers in addition. Copying with changes limited to the
4647 covers, as long as they preserve the title of the Document and
4648 satisfy these conditions, can be treated as verbatim copying in
4651 If the required texts for either cover are too voluminous to fit
4652 legibly, you should put the first ones listed (as many as fit
4653 reasonably) on the actual cover, and continue the rest onto
4656 If you publish or distribute Opaque copies of the Document
4657 numbering more than 100, you must either include a
4658 machine-readable Transparent copy along with each Opaque copy, or
4659 state in or with each Opaque copy a computer-network location from
4660 which the general network-using public has access to download
4661 using public-standard network protocols a complete Transparent
4662 copy of the Document, free of added material. If you use the
4663 latter option, you must take reasonably prudent steps, when you
4664 begin distribution of Opaque copies in quantity, to ensure that
4665 this Transparent copy will remain thus accessible at the stated
4666 location until at least one year after the last time you
4667 distribute an Opaque copy (directly or through your agents or
4668 retailers) of that edition to the public.
4670 It is requested, but not required, that you contact the authors of
4671 the Document well before redistributing any large number of
4672 copies, to give them a chance to provide you with an updated
4673 version of the Document.
4677 You may copy and distribute a Modified Version of the Document
4678 under the conditions of sections 2 and 3 above, provided that you
4679 release the Modified Version under precisely this License, with
4680 the Modified Version filling the role of the Document, thus
4681 licensing distribution and modification of the Modified Version to
4682 whoever possesses a copy of it. In addition, you must do these
4683 things in the Modified Version:
4685 A. Use in the Title Page (and on the covers, if any) a title
4686 distinct from that of the Document, and from those of
4687 previous versions (which should, if there were any, be listed
4688 in the History section of the Document). You may use the
4689 same title as a previous version if the original publisher of
4690 that version gives permission.
4692 B. List on the Title Page, as authors, one or more persons or
4693 entities responsible for authorship of the modifications in
4694 the Modified Version, together with at least five of the
4695 principal authors of the Document (all of its principal
4696 authors, if it has fewer than five), unless they release you
4697 from this requirement.
4699 C. State on the Title page the name of the publisher of the
4700 Modified Version, as the publisher.
4702 D. Preserve all the copyright notices of the Document.
4704 E. Add an appropriate copyright notice for your modifications
4705 adjacent to the other copyright notices.
4707 F. Include, immediately after the copyright notices, a license
4708 notice giving the public permission to use the Modified
4709 Version under the terms of this License, in the form shown in
4712 G. Preserve in that license notice the full lists of Invariant
4713 Sections and required Cover Texts given in the Document's
4716 H. Include an unaltered copy of this License.
4718 I. Preserve the section Entitled "History", Preserve its Title,
4719 and add to it an item stating at least the title, year, new
4720 authors, and publisher of the Modified Version as given on
4721 the Title Page. If there is no section Entitled "History" in
4722 the Document, create one stating the title, year, authors,
4723 and publisher of the Document as given on its Title Page,
4724 then add an item describing the Modified Version as stated in
4725 the previous sentence.
4727 J. Preserve the network location, if any, given in the Document
4728 for public access to a Transparent copy of the Document, and
4729 likewise the network locations given in the Document for
4730 previous versions it was based on. These may be placed in
4731 the "History" section. You may omit a network location for a
4732 work that was published at least four years before the
4733 Document itself, or if the original publisher of the version
4734 it refers to gives permission.
4736 K. For any section Entitled "Acknowledgements" or "Dedications",
4737 Preserve the Title of the section, and preserve in the
4738 section all the substance and tone of each of the contributor
4739 acknowledgements and/or dedications given therein.
4741 L. Preserve all the Invariant Sections of the Document,
4742 unaltered in their text and in their titles. Section numbers
4743 or the equivalent are not considered part of the section
4746 M. Delete any section Entitled "Endorsements". Such a section
4747 may not be included in the Modified Version.
4749 N. Do not retitle any existing section to be Entitled
4750 "Endorsements" or to conflict in title with any Invariant
4753 O. Preserve any Warranty Disclaimers.
4755 If the Modified Version includes new front-matter sections or
4756 appendices that qualify as Secondary Sections and contain no
4757 material copied from the Document, you may at your option
4758 designate some or all of these sections as invariant. To do this,
4759 add their titles to the list of Invariant Sections in the Modified
4760 Version's license notice. These titles must be distinct from any
4761 other section titles.
4763 You may add a section Entitled "Endorsements", provided it contains
4764 nothing but endorsements of your Modified Version by various
4765 parties--for example, statements of peer review or that the text
4766 has been approved by an organization as the authoritative
4767 definition of a standard.
4769 You may add a passage of up to five words as a Front-Cover Text,
4770 and a passage of up to 25 words as a Back-Cover Text, to the end
4771 of the list of Cover Texts in the Modified Version. Only one
4772 passage of Front-Cover Text and one of Back-Cover Text may be
4773 added by (or through arrangements made by) any one entity. If the
4774 Document already includes a cover text for the same cover,
4775 previously added by you or by arrangement made by the same entity
4776 you are acting on behalf of, you may not add another; but you may
4777 replace the old one, on explicit permission from the previous
4778 publisher that added the old one.
4780 The author(s) and publisher(s) of the Document do not by this
4781 License give permission to use their names for publicity for or to
4782 assert or imply endorsement of any Modified Version.
4784 5. COMBINING DOCUMENTS
4786 You may combine the Document with other documents released under
4787 this License, under the terms defined in section 4 above for
4788 modified versions, provided that you include in the combination
4789 all of the Invariant Sections of all of the original documents,
4790 unmodified, and list them all as Invariant Sections of your
4791 combined work in its license notice, and that you preserve all
4792 their Warranty Disclaimers.
4794 The combined work need only contain one copy of this License, and
4795 multiple identical Invariant Sections may be replaced with a single
4796 copy. If there are multiple Invariant Sections with the same name
4797 but different contents, make the title of each such section unique
4798 by adding at the end of it, in parentheses, the name of the
4799 original author or publisher of that section if known, or else a
4800 unique number. Make the same adjustment to the section titles in
4801 the list of Invariant Sections in the license notice of the
4804 In the combination, you must combine any sections Entitled
4805 "History" in the various original documents, forming one section
4806 Entitled "History"; likewise combine any sections Entitled
4807 "Acknowledgements", and any sections Entitled "Dedications". You
4808 must delete all sections Entitled "Endorsements."
4810 6. COLLECTIONS OF DOCUMENTS
4812 You may make a collection consisting of the Document and other
4813 documents released under this License, and replace the individual
4814 copies of this License in the various documents with a single copy
4815 that is included in the collection, provided that you follow the
4816 rules of this License for verbatim copying of each of the
4817 documents in all other respects.
4819 You may extract a single document from such a collection, and
4820 distribute it individually under this License, provided you insert
4821 a copy of this License into the extracted document, and follow
4822 this License in all other respects regarding verbatim copying of
4825 7. AGGREGATION WITH INDEPENDENT WORKS
4827 A compilation of the Document or its derivatives with other
4828 separate and independent documents or works, in or on a volume of
4829 a storage or distribution medium, is called an "aggregate" if the
4830 copyright resulting from the compilation is not used to limit the
4831 legal rights of the compilation's users beyond what the individual
4832 works permit. When the Document is included in an aggregate, this
4833 License does not apply to the other works in the aggregate which
4834 are not themselves derivative works of the Document.
4836 If the Cover Text requirement of section 3 is applicable to these
4837 copies of the Document, then if the Document is less than one half
4838 of the entire aggregate, the Document's Cover Texts may be placed
4839 on covers that bracket the Document within the aggregate, or the
4840 electronic equivalent of covers if the Document is in electronic
4841 form. Otherwise they must appear on printed covers that bracket
4842 the whole aggregate.
4846 Translation is considered a kind of modification, so you may
4847 distribute translations of the Document under the terms of section
4848 4. Replacing Invariant Sections with translations requires special
4849 permission from their copyright holders, but you may include
4850 translations of some or all Invariant Sections in addition to the
4851 original versions of these Invariant Sections. You may include a
4852 translation of this License, and all the license notices in the
4853 Document, and any Warranty Disclaimers, provided that you also
4854 include the original English version of this License and the
4855 original versions of those notices and disclaimers. In case of a
4856 disagreement between the translation and the original version of
4857 this License or a notice or disclaimer, the original version will
4860 If a section in the Document is Entitled "Acknowledgements",
4861 "Dedications", or "History", the requirement (section 4) to
4862 Preserve its Title (section 1) will typically require changing the
4867 You may not copy, modify, sublicense, or distribute the Document
4868 except as expressly provided under this License. Any attempt
4869 otherwise to copy, modify, sublicense, or distribute it is void,
4870 and will automatically terminate your rights under this License.
4872 However, if you cease all violation of this License, then your
4873 license from a particular copyright holder is reinstated (a)
4874 provisionally, unless and until the copyright holder explicitly
4875 and finally terminates your license, and (b) permanently, if the
4876 copyright holder fails to notify you of the violation by some
4877 reasonable means prior to 60 days after the cessation.
4879 Moreover, your license from a particular copyright holder is
4880 reinstated permanently if the copyright holder notifies you of the
4881 violation by some reasonable means, this is the first time you have
4882 received notice of violation of this License (for any work) from
4883 that copyright holder, and you cure the violation prior to 30 days
4884 after your receipt of the notice.
4886 Termination of your rights under this section does not terminate
4887 the licenses of parties who have received copies or rights from
4888 you under this License. If your rights have been terminated and
4889 not permanently reinstated, receipt of a copy of some or all of
4890 the same material does not give you any rights to use it.
4892 10. FUTURE REVISIONS OF THIS LICENSE
4894 The Free Software Foundation may publish new, revised versions of
4895 the GNU Free Documentation License from time to time. Such new
4896 versions will be similar in spirit to the present version, but may
4897 differ in detail to address new problems or concerns. See
4898 `http://www.gnu.org/copyleft/'.
4900 Each version of the License is given a distinguishing version
4901 number. If the Document specifies that a particular numbered
4902 version of this License "or any later version" applies to it, you
4903 have the option of following the terms and conditions either of
4904 that specified version or of any later version that has been
4905 published (not as a draft) by the Free Software Foundation. If
4906 the Document does not specify a version number of this License,
4907 you may choose any version ever published (not as a draft) by the
4908 Free Software Foundation. If the Document specifies that a proxy
4909 can decide which future versions of this License can be used, that
4910 proxy's public statement of acceptance of a version permanently
4911 authorizes you to choose that version for the Document.
4915 "Massive Multiauthor Collaboration Site" (or "MMC Site") means any
4916 World Wide Web server that publishes copyrightable works and also
4917 provides prominent facilities for anybody to edit those works. A
4918 public wiki that anybody can edit is an example of such a server.
4919 A "Massive Multiauthor Collaboration" (or "MMC") contained in the
4920 site means any set of copyrightable works thus published on the MMC
4923 "CC-BY-SA" means the Creative Commons Attribution-Share Alike 3.0
4924 license published by Creative Commons Corporation, a not-for-profit
4925 corporation with a principal place of business in San Francisco,
4926 California, as well as future copyleft versions of that license
4927 published by that same organization.
4929 "Incorporate" means to publish or republish a Document, in whole or
4930 in part, as part of another Document.
4932 An MMC is "eligible for relicensing" if it is licensed under this
4933 License, and if all works that were first published under this
4934 License somewhere other than this MMC, and subsequently
4935 incorporated in whole or in part into the MMC, (1) had no cover
4936 texts or invariant sections, and (2) were thus incorporated prior
4937 to November 1, 2008.
4939 The operator of an MMC Site may republish an MMC contained in the
4940 site under CC-BY-SA on the same site at any time before August 1,
4941 2009, provided the MMC is eligible for relicensing.
4944 ADDENDUM: How to use this License for your documents
4945 ====================================================
4947 To use this License in a document you have written, include a copy of
4948 the License in the document and put the following copyright and license
4949 notices just after the title page:
4951 Copyright (C) YEAR YOUR NAME.
4952 Permission is granted to copy, distribute and/or modify this document
4953 under the terms of the GNU Free Documentation License, Version 1.3
4954 or any later version published by the Free Software Foundation;
4955 with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
4956 Texts. A copy of the license is included in the section entitled ``GNU
4957 Free Documentation License''.
4959 If you have Invariant Sections, Front-Cover Texts and Back-Cover
4960 Texts, replace the "with...Texts." line with this:
4962 with the Invariant Sections being LIST THEIR TITLES, with
4963 the Front-Cover Texts being LIST, and with the Back-Cover Texts
4966 If you have Invariant Sections without Cover Texts, or some other
4967 combination of the three, merge those two alternatives to suit the
4970 If your document contains nontrivial examples of program code, we
4971 recommend releasing these examples in parallel under your choice of
4972 free software license, such as the GNU General Public License, to
4973 permit their use in free software.
4976 File: diff.info, Node: Translations, Next: Index, Prev: Copying This Manual, Up: Top
4978 Appendix B Translations of This Manual
4979 **************************************
4981 Nishio Futoshi of the GNUjdoc project has prepared a Japanese
4982 translation of this manual. Its most recent version can be found at
4983 `http://openlab.ring.gr.jp/gnujdoc/cvsweb/cvsweb.cgi/gnujdoc/'.
4986 File: diff.info, Node: Index, Prev: Translations, Up: Top
4994 * ! output format: Context. (line 6)
4995 * +- output format: Unified Format. (line 6)
4996 * < output format: Normal. (line 6)
4997 * <<<<<<< for marking conflicts: Marking Conflicts. (line 6)
4998 * _POSIX2_VERSION: Standards conformance.
5000 * aligning tab stops: Tabs. (line 6)
5001 * alternate file names: Alternate Names. (line 6)
5002 * backup file names: Backup Names. (line 6)
5003 * backup file strategy: Backups. (line 6)
5004 * binary file diff: Binary. (line 6)
5005 * blank and tab difference suppression: White Space. (line 6)
5006 * blank line difference suppression: Blank Lines. (line 6)
5007 * brief difference reports: Brief. (line 6)
5008 * bug reports: Bugs. (line 6)
5009 * C function headings: C Function Headings. (line 6)
5010 * C if-then-else output format: If-then-else. (line 6)
5011 * case difference suppression: Case Folding. (line 6)
5012 * ClearCase: Revision Control. (line 6)
5013 * cmp invocation: Invoking cmp. (line 6)
5014 * cmp options: cmp Options. (line 6)
5015 * columnar output: Side by Side. (line 6)
5016 * common mistakes with patches: Avoiding Common Mistakes.
5018 * comparing three files: Comparing Three Files.
5020 * conflict: diff3 Merging. (line 26)
5021 * conflict marking: Marking Conflicts. (line 6)
5022 * context output format: Context. (line 6)
5023 * creating files: Creating and Removing.
5025 * diagnostics from patch: patch Messages. (line 6)
5026 * diff invocation: Invoking diff. (line 6)
5027 * diff merging: Interactive Merging. (line 6)
5028 * diff options: diff Options. (line 6)
5029 * diff sample input: Sample diff Input. (line 6)
5030 * diff3 hunks: diff3 Hunks. (line 6)
5031 * diff3 invocation: Invoking diff3. (line 6)
5032 * diff3 options: diff3 Options. (line 6)
5033 * diff3 sample input: Sample diff3 Input. (line 6)
5034 * directories and patch: patch Directories. (line 6)
5035 * directory structure changes: Changing Structure. (line 6)
5036 * dry runs for patch: Dry Runs. (line 6)
5037 * ed script output format: ed Scripts. (line 6)
5038 * EDITOR: Merge Commands. (line 50)
5039 * empty files, removing: Creating and Removing.
5041 * exabyte, definition of: cmp Options. (line 106)
5042 * exbibyte, definition of: cmp Options. (line 110)
5043 * file name alternates: Alternate Names. (line 6)
5044 * file names with unusual characters: Unusual File Names. (line 6)
5045 * format of diff output: Output Formats. (line 6)
5046 * format of diff3 output: Comparing Three Files.
5048 * formats for if-then-else line groups: Line Group Formats. (line 6)
5049 * forward ed script output format: Forward ed. (line 6)
5050 * full lines: Incomplete Lines. (line 6)
5051 * function headings, C: C Function Headings. (line 6)
5052 * fuzz factor when patching: Inexact. (line 6)
5053 * gibibyte, definition of: cmp Options. (line 89)
5054 * gigabyte, definition of: cmp Options. (line 85)
5055 * headings: Sections. (line 6)
5056 * hunks: Hunks. (line 6)
5057 * hunks for diff3: diff3 Hunks. (line 6)
5058 * if-then-else output format: If-then-else. (line 6)
5059 * ifdef output format: If-then-else. (line 6)
5060 * imperfect patch application: Imperfect. (line 6)
5061 * incomplete line merging: Merging Incomplete Lines.
5063 * incomplete lines: Incomplete Lines. (line 6)
5064 * inexact patches: Inexact. (line 6)
5065 * inhibit messages from patch: More or Fewer Messages.
5067 * interactive merging: Interactive Merging. (line 6)
5068 * introduction: Comparison. (line 6)
5069 * intuiting file names from patches: Multiple Patches. (line 6)
5070 * invoking cmp: Invoking cmp. (line 6)
5071 * invoking diff: Invoking diff. (line 6)
5072 * invoking diff3: Invoking diff3. (line 6)
5073 * invoking patch: Invoking patch. (line 6)
5074 * invoking sdiff: Invoking sdiff. (line 6)
5075 * keyboard input to patch: patch and Keyboard Input.
5077 * kibibyte, definition of: cmp Options. (line 73)
5078 * kilobyte, definition of: cmp Options. (line 68)
5079 * LC_COLLATE: Comparing Directories.
5081 * LC_NUMERIC: Line Group Formats. (line 144)
5082 * LC_TIME: Detailed Context. (line 12)
5083 * line formats: Line Formats. (line 6)
5084 * line group formats: Line Group Formats. (line 6)
5085 * mebibyte, definition of: cmp Options. (line 82)
5086 * megabyte, definition of: cmp Options. (line 78)
5087 * merge commands: Merge Commands. (line 6)
5088 * merged diff3 format: Bypassing ed. (line 6)
5089 * merged output format: If-then-else. (line 6)
5090 * merging from a common ancestor: diff3 Merging. (line 6)
5091 * merging interactively: Merge Commands. (line 6)
5092 * messages from patch: patch Messages. (line 6)
5093 * multibyte characters: Internationalization.
5095 * multiple patches: Multiple Patches. (line 6)
5096 * newline treatment by diff: Incomplete Lines. (line 6)
5097 * normal output format: Normal. (line 6)
5098 * options for cmp: cmp Options. (line 6)
5099 * options for diff: diff Options. (line 6)
5100 * options for diff3: diff3 Options. (line 6)
5101 * options for patch: patch Options. (line 6)
5102 * options for sdiff: sdiff Options. (line 6)
5103 * output formats: Output Formats. (line 6)
5104 * overlap: diff3 Merging. (line 26)
5105 * overlapping change, selection of: Which Changes. (line 6)
5106 * overview of diff and patch: Overview. (line 6)
5107 * paginating diff output: Pagination. (line 6)
5108 * patch consumer tips: Tips for Patch Consumers.
5110 * patch input format: patch Input. (line 6)
5111 * patch invocation: Invoking patch. (line 6)
5112 * patch messages and questions: patch Messages. (line 6)
5113 * patch options: patch Options. (line 6)
5114 * patch producer tips: Tips for Patch Producers.
5116 * patch, common mistakes: Avoiding Common Mistakes.
5118 * PATCH_GET: Revision Control. (line 13)
5119 * PATCH_VERSION_CONTROL: Backup Names. (line 21)
5120 * patches, shrinking: Generating Smaller Patches.
5122 * patching directories: patch Directories. (line 6)
5123 * pebibyte, definition of: cmp Options. (line 103)
5124 * performance of diff: diff Performance. (line 6)
5125 * petabyte, definition of: cmp Options. (line 99)
5126 * POSIX <1>: Standards conformance.
5128 * POSIX: patch and POSIX. (line 6)
5129 * POSIXLY_CORRECT <1>: Standards conformance.
5131 * POSIXLY_CORRECT: patch and POSIX. (line 6)
5132 * projects for directories: Shortcomings. (line 6)
5133 * quoting style: patch Quoting Style. (line 6)
5134 * QUOTING_STYLE: patch Quoting Style. (line 30)
5135 * RCS: Revision Control. (line 6)
5136 * RCS script output format: RCS. (line 6)
5137 * regular expression matching headings: Specified Headings. (line 6)
5138 * regular expression suppression: Specified Lines. (line 6)
5139 * reject file names: Reject Names. (line 6)
5140 * removing empty files: Creating and Removing.
5142 * reporting bugs: Bugs. (line 6)
5143 * reversed patches: Reversed Patches. (line 6)
5144 * revision control: Revision Control. (line 6)
5145 * sample input for diff: Sample diff Input. (line 6)
5146 * sample input for diff3: Sample diff3 Input. (line 6)
5147 * SCCS: Revision Control. (line 6)
5148 * script output formats: Scripts. (line 6)
5149 * sdiff invocation: Invoking sdiff. (line 6)
5150 * sdiff options: sdiff Options. (line 6)
5151 * sdiff output format: sdiff Option Summary.
5153 * section headings: Sections. (line 6)
5154 * side by side: Side by Side. (line 6)
5155 * side by side format: Side by Side Format. (line 6)
5156 * SIMPLE_BACKUP_SUFFIX: Backup Names. (line 12)
5157 * special files: Special Files. (line 6)
5158 * specified headings: Specified Headings. (line 6)
5159 * summarizing which files differ: Brief. (line 6)
5160 * System V diff3 compatibility: Saving the Changed File.
5162 * tab and blank difference suppression: White Space. (line 6)
5163 * tab stop alignment: Tabs. (line 6)
5164 * tebibyte, definition of: cmp Options. (line 96)
5165 * terabyte, definition of: cmp Options. (line 92)
5166 * testing patch: Dry Runs. (line 6)
5167 * text versus binary diff: Binary. (line 6)
5168 * time stamp format, context diffs: Detailed Context. (line 12)
5169 * time stamp format, unified diffs: Detailed Unified. (line 12)
5170 * time stamps on patched files: Patching Time Stamps.
5172 * traditional patch: patch and Tradition. (line 6)
5173 * trailing blanks: Trailing Blanks. (line 6)
5174 * two-column output: Side by Side. (line 6)
5175 * unified output format: Unified Format. (line 6)
5176 * unmerged change: Which Changes. (line 6)
5177 * varying-width characters: Internationalization.
5179 * verbose messages from patch: More or Fewer Messages.
5181 * version control: Revision Control. (line 6)
5182 * VERSION_CONTROL <1>: Backup Names. (line 21)
5183 * VERSION_CONTROL: Revision Control. (line 22)
5184 * white space in patches: Changed White Space. (line 6)
5185 * yottabyte, definition of: cmp Options. (line 121)
5186 * zettabyte, definition of: cmp Options. (line 113)
5192 Node: Overview
\7f4135
5193 Node: Comparison
\7f7759
5195 Node: White Space
\7f11895
5196 Node: Blank Lines
\7f13623
5197 Node: Specified Lines
\7f14653
5198 Node: Case Folding
\7f15775
5200 Node: Binary
\7f17506
5201 Node: Output Formats
\7f21591
5202 Node: Sample diff Input
\7f22311
5203 Node: Context
\7f23805
5204 Node: Context Format
\7f25377
5205 Node: Example Context
\7f26164
5206 Node: Less Context
\7f27666
5207 Node: Detailed Context
\7f28850
5208 Node: Unified Format
\7f31043
5209 Node: Example Unified
\7f31834
5210 Node: Detailed Unified
\7f32867
5211 Node: Sections
\7f34509
5212 Node: Specified Headings
\7f35263
5213 Node: C Function Headings
\7f36812
5214 Node: Alternate Names
\7f37654
5215 Node: Side by Side
\7f38563
5216 Node: Side by Side Format
\7f40707
5217 Node: Example Side by Side
\7f41605
5218 Node: Normal
\7f42940
5219 Node: Example Normal
\7f43935
5220 Node: Detailed Normal
\7f44666
5221 Node: Scripts
\7f46400
5222 Node: ed Scripts
\7f46800
5223 Node: Example ed
\7f48002
5224 Node: Detailed ed
\7f48447
5225 Node: Forward ed
\7f50201
5227 Node: If-then-else
\7f52184
5228 Node: Line Group Formats
\7f53857
5229 Node: Line Formats
\7f59728
5230 Node: Example If-then-else
\7f62993
5231 Node: Detailed If-then-else
\7f64067
5232 Node: Incomplete Lines
\7f65945
5233 Node: Comparing Directories
\7f67577
5234 Node: Adjusting Output
\7f71621
5236 Node: Trailing Blanks
\7f73733
5237 Node: Pagination
\7f74954
5238 Node: diff Performance
\7f75418
5239 Node: Comparing Three Files
\7f78501
5240 Node: Sample diff3 Input
\7f79374
5241 Node: Example diff3 Normal
\7f80317
5242 Node: Detailed diff3 Normal
\7f81362
5243 Node: diff3 Hunks
\7f83117
5244 Node: diff3 Merging
\7f84408
5245 Node: Which Changes
\7f86647
5246 Node: Marking Conflicts
\7f88046
5247 Node: Bypassing ed
\7f90500
5248 Node: Merging Incomplete Lines
\7f91839
5249 Node: Saving the Changed File
\7f92560
5250 Node: Interactive Merging
\7f93171
5251 Node: sdiff Option Summary
\7f93876
5252 Node: Merge Commands
\7f95047
5253 Node: Merging with patch
\7f96331
5254 Node: patch Input
\7f98698
5255 Node: Revision Control
\7f99375
5256 Node: Imperfect
\7f100541
5257 Node: Changed White Space
\7f101684
5258 Node: Reversed Patches
\7f102476
5259 Node: Inexact
\7f103936
5260 Node: Dry Runs
\7f107490
5261 Node: Creating and Removing
\7f108349
5262 Node: Patching Time Stamps
\7f109395
5263 Node: Multiple Patches
\7f111593
5264 Node: patch Directories
\7f114251
5265 Node: Backups
\7f115872
5266 Node: Backup Names
\7f116933
5267 Ref: Backup Names-Footnote-1
\7f119897
5268 Node: Reject Names
\7f120024
5269 Node: patch Messages
\7f120608
5270 Node: More or Fewer Messages
\7f121663
5271 Node: patch and Keyboard Input
\7f122289
5272 Node: patch Quoting Style
\7f123315
5273 Node: patch and POSIX
\7f124460
5274 Node: patch and Tradition
\7f125295
5275 Node: Making Patches
\7f128747
5276 Node: Tips for Patch Producers
\7f129568
5277 Node: Tips for Patch Consumers
\7f130819
5278 Node: Avoiding Common Mistakes
\7f131451
5279 Node: Generating Smaller Patches
\7f133971
5280 Node: Invoking cmp
\7f135727
5281 Node: cmp Options
\7f137145
5282 Node: Invoking diff
\7f140632
5283 Node: diff Options
\7f142485
5284 Node: Invoking diff3
\7f150912
5285 Node: diff3 Options
\7f151545
5286 Node: Invoking patch
\7f154573
5287 Node: patch Options
\7f155776
5288 Node: Invoking sdiff
\7f160998
5289 Node: sdiff Options
\7f162140
5290 Node: Standards conformance
\7f165015
5291 Node: Projects
\7f166756
5292 Node: Shortcomings
\7f167462
5293 Node: Internationalization
\7f168559
5294 Node: Changing Structure
\7f169720
5295 Node: Special Files
\7f170819
5296 Node: Unusual File Names
\7f171926
5297 Node: Time Stamp Order
\7f172560
5298 Node: Ignoring Changes
\7f173198
5299 Node: Speedups
\7f173963
5301 Node: Copying This Manual
\7f175270
5302 Node: Translations
\7f200404
5303 Node: Index
\7f200771