1GITATTRIBUTES(5) Git Manual GITATTRIBUTES(5)
2
6 gitattributes - Defining attributes per path
7
9 $GIT_DIR/info/attributes, .gitattributes
10
12 A gitattributes file is a simple text file that gives attributes to
13 pathnames.
14 Each line in gitattributes file is of form:
16 pattern attr1 attr2 ...
18 That is, a pattern followed by an attributes list, separated by
20 whitespaces. Leading and trailing whitespaces are ignored. Lines that
21 begin with # are ignored. Patterns that begin with a double quote are
22 quoted in C style. When the pattern matches the path in question, the
23 attributes listed on the line are given to the path.
24 Each attribute can be in one of these states for a given path:
26 Set
28 The path has the attribute with special value "true"; this is
29 specified by listing only the name of the attribute in the
30 attribute list.
31 Unset
33 The path has the attribute with special value "false"; this is
34 specified by listing the name of the attribute prefixed with a dash
35 - in the attribute list.
36 Set to a value
38 The path has the attribute with specified string value; this is
39 specified by listing the name of the attribute followed by an equal
40 sign = and its value in the attribute list.
41 Unspecified
43 No pattern matches the path, and nothing says if the path has or
44 does not have the attribute, the attribute for the path is said to
45 be Unspecified.
46 When more than one pattern matches the path, a later line overrides an
48 earlier line. This overriding is done per attribute.
49 The rules by which the pattern matches paths are the same as in
51 .gitignore files (see gitignore(5)), with a few exceptions:
52 · negative patterns are forbidden
54 · patterns that match a directory do not recursively match paths
56 inside that directory (so using the trailing-slash path/ syntax is
57 pointless in an attributes file; use path/** instead)
58 When deciding what attributes are assigned to a path, Git consults
60 $GIT_DIR/info/attributes file (which has the highest precedence),
61 .gitattributes file in the same directory as the path in question, and
62 its parent directories up to the toplevel of the work tree (the further
63 the directory that contains .gitattributes is from the path in
64 question, the lower its precedence). Finally global and system-wide
65 files are considered (they have the lowest precedence).
66 When the .gitattributes file is missing from the work tree, the path in
68 the index is used as a fall-back. During checkout process,
69 .gitattributes in the index is used and then the file in the working
70 tree is used as a fall-back.
71 If you wish to affect only a single repository (i.e., to assign
73 attributes to files that are particular to one user’s workflow for that
74 repository), then attributes should be placed in the
75 $GIT_DIR/info/attributes file. Attributes which should be
76 version-controlled and distributed to other repositories (i.e.,
77 attributes of interest to all users) should go into .gitattributes
78 files. Attributes that should affect all repositories for a single user
79 should be placed in a file specified by the core.attributesFile
80 configuration option (see git-config(1)). Its default value is
81 $XDG_CONFIG_HOME/git/attributes. If $XDG_CONFIG_HOME is either not set
82 or empty, $HOME/.config/git/attributes is used instead. Attributes for
83 all users on a system should be placed in the
84 $(prefix)/etc/gitattributes file.
85 Sometimes you would need to override a setting of an attribute for a
87 path to Unspecified state. This can be done by listing the name of the
88 attribute prefixed with an exclamation point !.
89
91 Certain operations by Git can be influenced by assigning particular
92 attributes to a path. Currently, the following operations are
93 attributes-aware.
94 Checking-out and checking-in
96 These attributes affect how the contents stored in the repository are
97 copied to the working tree files when commands such as git switch, git
98 checkout and git merge run. They also affect how Git stores the
99 contents you prepare in the working tree in the repository upon git add
100 and git commit.
101 text
103 This attribute enables and controls end-of-line normalization. When
104 a text file is normalized, its line endings are converted to LF in
105 the repository. To control what line ending style is used in the
106 working directory, use the eol attribute for a single file and the
107 core.eol configuration variable for all text files. Note that
108 setting core.autocrlf to true or input overrides core.eol (see the
109 definitions of those options in git-config(1)).
110 Set
112 Setting the text attribute on a path enables end-of-line
113 normalization and marks the path as a text file. End-of-line
114 conversion takes place without guessing the content type.
115 Unset
117 Unsetting the text attribute on a path tells Git not to attempt
118 any end-of-line conversion upon checkin or checkout.
119 Set to string value "auto"
121 When text is set to "auto", the path is marked for automatic
122 end-of-line conversion. If Git decides that the content is
123 text, its line endings are converted to LF on checkin. When the
124 file has been committed with CRLF, no conversion is done.
125 Unspecified
127 If the text attribute is unspecified, Git uses the
128 core.autocrlf configuration variable to determine if the file
129 should be converted.
130 Any other value causes Git to act as if text has been left
132 unspecified.
133 eol
135 This attribute sets a specific line-ending style to be used in the
136 working directory. It enables end-of-line conversion without any
137 content checks, effectively setting the text attribute. Note that
138 setting this attribute on paths which are in the index with CRLF
139 line endings may make the paths to be considered dirty. Adding the
140 path to the index again will normalize the line endings in the
141 index.
142 Set to string value "crlf"
144 This setting forces Git to normalize line endings for this file
145 on checkin and convert them to CRLF when the file is checked
146 out.
147 Set to string value "lf"
149 This setting forces Git to normalize line endings to LF on
150 checkin and prevents conversion to CRLF when the file is
151 checked out.
152 Backwards compatibility with crlf attribute
154 For backwards compatibility, the crlf attribute is interpreted as
155 follows:
156 crlf text
158 -crlf -text
159 crlf=input eol=lf
160 End-of-line conversion
163 While Git normally leaves file contents alone, it can be configured
164 to normalize line endings to LF in the repository and, optionally,
165 to convert them to CRLF when files are checked out.
166 If you simply want to have CRLF line endings in your working
168 directory regardless of the repository you are working with, you
169 can set the config variable "core.autocrlf" without using any
170 attributes.
171 [core]
173 autocrlf = true
174 This does not force normalization of text files, but does ensure
177 that text files that you introduce to the repository have their
178 line endings normalized to LF when they are added, and that files
179 that are already normalized in the repository stay normalized.
180 If you want to ensure that text files that any contributor
182 introduces to the repository have their line endings normalized,
183 you can set the text attribute to "auto" for all files.
184 * text=auto
186 The attributes allow a fine-grained control, how the line endings
189 are converted. Here is an example that will make Git normalize
190 .txt, .vcproj and .sh files, ensure that .vcproj files have CRLF
191 and .sh files have LF in the working directory, and prevent .jpg
192 files from being normalized regardless of their content.
193 * text=auto
195 *.txt text
196 *.vcproj text eol=crlf
197 *.sh text eol=lf
198 *.jpg -text
199 Note
202 When text=auto conversion is enabled in a cross-platform
203 project using push and pull to a central repository the text
204 files containing CRLFs should be normalized.
205 From a clean working directory:
207 $ echo "* text=auto" >.gitattributes
209 $ git add --renormalize .
210 $ git status # Show files that will be normalized
211 $ git commit -m "Introduce end-of-line normalization"
212 If any files that should not be normalized show up in git status,
215 unset their text attribute before running git add -u.
216 manual.pdf -text
218 Conversely, text files that Git does not detect can have
221 normalization enabled manually.
222 weirdchars.txt text
224 If core.safecrlf is set to "true" or "warn", Git verifies if the
227 conversion is reversible for the current setting of core.autocrlf.
228 For "true", Git rejects irreversible conversions; for "warn", Git
229 only prints a warning but accepts an irreversible conversion. The
230 safety triggers to prevent such a conversion done to the files in
231 the work tree, but there are a few exceptions. Even though...
232 · git add itself does not touch the files in the work tree, the
234 next checkout would, so the safety triggers;
235 · git apply to update a text file with a patch does touch the
237 files in the work tree, but the operation is about text files
238 and CRLF conversion is about fixing the line ending
239 inconsistencies, so the safety does not trigger;
240 · git diff itself does not touch the files in the work tree, it
242 is often run to inspect the changes you intend to next git add.
243 To catch potential problems early, safety triggers.
244 working-tree-encoding
246 Git recognizes files encoded in ASCII or one of its supersets (e.g.
247 UTF-8, ISO-8859-1, ...) as text files. Files encoded in certain
248 other encodings (e.g. UTF-16) are interpreted as binary and
249 consequently built-in Git text processing tools (e.g. git diff) as
250 well as most Git web front ends do not visualize the contents of
251 these files by default.
252 In these cases you can tell Git the encoding of a file in the
254 working directory with the working-tree-encoding attribute. If a
255 file with this attribute is added to Git, then Git reencodes the
256 content from the specified encoding to UTF-8. Finally, Git stores
257 the UTF-8 encoded content in its internal data structure (called
258 "the index"). On checkout the content is reencoded back to the
259 specified encoding.
260 Please note that using the working-tree-encoding attribute may have
262 a number of pitfalls:
263 · Alternative Git implementations (e.g. JGit or libgit2) and
265 older Git versions (as of March 2018) do not support the
266 working-tree-encoding attribute. If you decide to use the
267 working-tree-encoding attribute in your repository, then it is
268 strongly recommended to ensure that all clients working with
269 the repository support it.
270 For example, Microsoft Visual Studio resources files (*.rc) or
272 PowerShell script files (*.ps1) are sometimes encoded in
273 UTF-16. If you declare *.ps1 as files as UTF-16 and you add
274 foo.ps1 with a working-tree-encoding enabled Git client, then
275 foo.ps1 will be stored as UTF-8 internally. A client without
276 working-tree-encoding support will checkout foo.ps1 as UTF-8
277 encoded file. This will typically cause trouble for the users
278 of this file.
279 If a Git client that does not support the working-tree-encoding
281 attribute adds a new file bar.ps1, then bar.ps1 will be stored
282 "as-is" internally (in this example probably as UTF-16). A
283 client with working-tree-encoding support will interpret the
284 internal contents as UTF-8 and try to convert it to UTF-16 on
285 checkout. That operation will fail and cause an error.
286 · Reencoding content to non-UTF encodings can cause errors as the
288 conversion might not be UTF-8 round trip safe. If you suspect
289 your encoding to not be round trip safe, then add it to
290 core.checkRoundtripEncoding to make Git check the round trip
291 encoding (see git-config(1)). SHIFT-JIS (Japanese character
292 set) is known to have round trip issues with UTF-8 and is
293 checked by default.
294 · Reencoding content requires resources that might slow down
296 certain Git operations (e.g git checkout or git add).
297 Use the working-tree-encoding attribute only if you cannot store a
299 file in UTF-8 encoding and if you want Git to be able to process
300 the content as text.
301 As an example, use the following attributes if your *.ps1 files are
303 UTF-16 encoded with byte order mark (BOM) and you want Git to
304 perform automatic line ending conversion based on your platform.
305 *.ps1 text working-tree-encoding=UTF-16
307 Use the following attributes if your *.ps1 files are UTF-16 little
310 endian encoded without BOM and you want Git to use Windows line
311 endings in the working directory (use UTF-16LE-BOM instead of
312 UTF-16LE if you want UTF-16 little endian with BOM). Please note,
313 it is highly recommended to explicitly define the line endings with
314 eol if the working-tree-encoding attribute is used to avoid
315 ambiguity.
316 *.ps1 text working-tree-encoding=UTF-16LE eol=CRLF
318 You can get a list of all available encodings on your platform with
321 the following command:
322 iconv --list
324 If you do not know the encoding of a file, then you can use the
327 file command to guess the encoding:
328 file foo.ps1
330 ident
333 When the attribute ident is set for a path, Git replaces $Id$ in
334 the blob object with $Id:, followed by the 40-character hexadecimal
335 blob object name, followed by a dollar sign $ upon checkout. Any
336 byte sequence that begins with $Id: and ends with $ in the worktree
337 file is replaced with $Id$ upon check-in.
338 filter
340 A filter attribute can be set to a string value that names a filter
341 driver specified in the configuration.
342 A filter driver consists of a clean command and a smudge command,
344 either of which can be left unspecified. Upon checkout, when the
345 smudge command is specified, the command is fed the blob object
346 from its standard input, and its standard output is used to update
347 the worktree file. Similarly, the clean command is used to convert
348 the contents of worktree file upon checkin. By default these
349 commands process only a single blob and terminate. If a long
350 running process filter is used in place of clean and/or smudge
351 filters, then Git can process all blobs with a single filter
352 command invocation for the entire life of a single Git command, for
353 example git add --all. If a long running process filter is
354 configured then it always takes precedence over a configured single
355 blob filter. See section below for the description of the protocol
356 used to communicate with a process filter.
357 One use of the content filtering is to massage the content into a
359 shape that is more convenient for the platform, filesystem, and the
360 user to use. For this mode of operation, the key phrase here is
361 "more convenient" and not "turning something unusable into usable".
362 In other words, the intent is that if someone unsets the filter
363 driver definition, or does not have the appropriate filter program,
364 the project should still be usable.
365 Another use of the content filtering is to store the content that
367 cannot be directly used in the repository (e.g. a UUID that refers
368 to the true content stored outside Git, or an encrypted content)
369 and turn it into a usable form upon checkout (e.g. download the
370 external content, or decrypt the encrypted content).
371 These two filters behave differently, and by default, a filter is
373 taken as the former, massaging the contents into more convenient
374 shape. A missing filter driver definition in the config, or a
375 filter driver that exits with a non-zero status, is not an error
376 but makes the filter a no-op passthru.
377 You can declare that a filter turns a content that by itself is
379 unusable into a usable content by setting the
380 filter.<driver>.required configuration variable to true.
381 Note: Whenever the clean filter is changed, the repo should be
383 renormalized: $ git add --renormalize .
384 For example, in .gitattributes, you would assign the filter
386 attribute for paths.
387 *.c filter=indent
389 Then you would define a "filter.indent.clean" and
392 "filter.indent.smudge" configuration in your .git/config to specify
393 a pair of commands to modify the contents of C programs when the
394 source files are checked in ("clean" is run) and checked out (no
395 change is made because the command is "cat").
396 [filter "indent"]
398 clean = indent
399 smudge = cat
400 For best results, clean should not alter its output further if it
403 is run twice ("clean→clean" should be equivalent to "clean"), and
404 multiple smudge commands should not alter clean's output
405 ("smudge→smudge→clean" should be equivalent to "clean"). See the
406 section on merging below.
407 The "indent" filter is well-behaved in this regard: it will not
409 modify input that is already correctly indented. In this case, the
410 lack of a smudge filter means that the clean filter must accept its
411 own output without modifying it.
412 If a filter must succeed in order to make the stored contents
414 usable, you can declare that the filter is required, in the
415 configuration:
416 [filter "crypt"]
418 clean = openssl enc ...
419 smudge = openssl enc -d ...
420 required
421 Sequence "%f" on the filter command line is replaced with the name
424 of the file the filter is working on. A filter might use this in
425 keyword substitution. For example:
426 [filter "p4"]
428 clean = git-p4-filter --clean %f
429 smudge = git-p4-filter --smudge %f
430 Note that "%f" is the name of the path that is being worked on.
433 Depending on the version that is being filtered, the corresponding
434 file on disk may not exist, or may have different contents. So,
435 smudge and clean commands should not try to access the file on
436 disk, but only act as filters on the content provided to them on
437 standard input.
438 Long Running Filter Process
440 If the filter command (a string value) is defined via
441 filter.<driver>.process then Git can process all blobs with a
442 single filter invocation for the entire life of a single Git
443 command. This is achieved by using the long-running process
444 protocol (described in
445 technical/long-running-process-protocol.txt).
446 When Git encounters the first file that needs to be cleaned or
448 smudged, it starts the filter and performs the handshake. In the
449 handshake, the welcome message sent by Git is "git-filter-client",
450 only version 2 is suppported, and the supported capabilities are
451 "clean", "smudge", and "delay".
452 Afterwards Git sends a list of "key=value" pairs terminated with a
454 flush packet. The list will contain at least the filter command
455 (based on the supported capabilities) and the pathname of the file
456 to filter relative to the repository root. Right after the flush
457 packet Git sends the content split in zero or more pkt-line packets
458 and a flush packet to terminate content. Please note, that the
459 filter must not send any response before it received the content
460 and the final flush packet. Also note that the "value" of a
461 "key=value" pair can contain the "=" character whereas the key
462 would never contain that character.
463 packet: git> command=smudge
465 packet: git> pathname=path/testfile.dat
466 packet: git> 0000
467 packet: git> CONTENT
468 packet: git> 0000
469 The filter is expected to respond with a list of "key=value" pairs
472 terminated with a flush packet. If the filter does not experience
473 problems then the list must contain a "success" status. Right after
474 these packets the filter is expected to send the content in zero or
475 more pkt-line packets and a flush packet at the end. Finally, a
476 second list of "key=value" pairs terminated with a flush packet is
477 expected. The filter can change the status in the second list or
478 keep the status as is with an empty list. Please note that the
479 empty list must be terminated with a flush packet regardless.
480 packet: git< status=success
482 packet: git< 0000
483 packet: git< SMUDGED_CONTENT
484 packet: git< 0000
485 packet: git< 0000 # empty list, keep "status=success" unchanged!
486 If the result content is empty then the filter is expected to
489 respond with a "success" status and a flush packet to signal the
490 empty content.
491 packet: git< status=success
493 packet: git< 0000
494 packet: git< 0000 # empty content!
495 packet: git< 0000 # empty list, keep "status=success" unchanged!
496 In case the filter cannot or does not want to process the content,
499 it is expected to respond with an "error" status.
500 packet: git< status=error
502 packet: git< 0000
503 If the filter experiences an error during processing, then it can
506 send the status "error" after the content was (partially or
507 completely) sent.
508 packet: git< status=success
510 packet: git< 0000
511 packet: git< HALF_WRITTEN_ERRONEOUS_CONTENT
512 packet: git< 0000
513 packet: git< status=error
514 packet: git< 0000
515 In case the filter cannot or does not want to process the content
518 as well as any future content for the lifetime of the Git process,
519 then it is expected to respond with an "abort" status at any point
520 in the protocol.
521 packet: git< status=abort
523 packet: git< 0000
524 Git neither stops nor restarts the filter process in case the
527 "error"/"abort" status is set. However, Git sets its exit code
528 according to the filter.<driver>.required flag, mimicking the
529 behavior of the filter.<driver>.clean / filter.<driver>.smudge
530 mechanism.
531 If the filter dies during the communication or does not adhere to
533 the protocol then Git will stop the filter process and restart it
534 with the next file that needs to be processed. Depending on the
535 filter.<driver>.required flag Git will interpret that as error.
536 Delay
538 If the filter supports the "delay" capability, then Git can send
539 the flag "can-delay" after the filter command and pathname. This
540 flag denotes that the filter can delay filtering the current blob
541 (e.g. to compensate network latencies) by responding with no
542 content but with the status "delayed" and a flush packet.
543 packet: git> command=smudge
545 packet: git> pathname=path/testfile.dat
546 packet: git> can-delay=1
547 packet: git> 0000
548 packet: git> CONTENT
549 packet: git> 0000
550 packet: git< status=delayed
551 packet: git< 0000
552 If the filter supports the "delay" capability then it must support
555 the "list_available_blobs" command. If Git sends this command, then
556 the filter is expected to return a list of pathnames representing
557 blobs that have been delayed earlier and are now available. The
558 list must be terminated with a flush packet followed by a "success"
559 status that is also terminated with a flush packet. If no blobs for
560 the delayed paths are available, yet, then the filter is expected
561 to block the response until at least one blob becomes available.
562 The filter can tell Git that it has no more delayed blobs by
563 sending an empty list. As soon as the filter responds with an empty
564 list, Git stops asking. All blobs that Git has not received at this
565 point are considered missing and will result in an error.
566 packet: git> command=list_available_blobs
568 packet: git> 0000
569 packet: git< pathname=path/testfile.dat
570 packet: git< pathname=path/otherfile.dat
571 packet: git< 0000
572 packet: git< status=success
573 packet: git< 0000
574 After Git received the pathnames, it will request the corresponding
577 blobs again. These requests contain a pathname and an empty content
578 section. The filter is expected to respond with the smudged content
579 in the usual way as explained above.
580 packet: git> command=smudge
582 packet: git> pathname=path/testfile.dat
583 packet: git> 0000
584 packet: git> 0000 # empty content!
585 packet: git< status=success
586 packet: git< 0000
587 packet: git< SMUDGED_CONTENT
588 packet: git< 0000
589 packet: git< 0000 # empty list, keep "status=success" unchanged!
590 Example
593 A long running filter demo implementation can be found in
594 contrib/long-running-filter/example.pl located in the Git core
595 repository. If you develop your own long running filter process
596 then the GIT_TRACE_PACKET environment variables can be very helpful
597 for debugging (see git(1)).
598 Please note that you cannot use an existing filter.<driver>.clean
600 or filter.<driver>.smudge command with filter.<driver>.process
601 because the former two use a different inter process communication
602 protocol than the latter one.
603 Interaction between checkin/checkout attributes
605 In the check-in codepath, the worktree file is first converted with
606 filter driver (if specified and corresponding driver defined), then
607 the result is processed with ident (if specified), and then finally
608 with text (again, if specified and applicable).
609 In the check-out codepath, the blob content is first converted with
611 text, and then ident and fed to filter.
612 Merging branches with differing checkin/checkout attributes
614 If you have added attributes to a file that cause the canonical
615 repository format for that file to change, such as adding a
616 clean/smudge filter or text/eol/ident attributes, merging anything
617 where the attribute is not in place would normally cause merge
618 conflicts.
619 To prevent these unnecessary merge conflicts, Git can be told to
621 run a virtual check-out and check-in of all three stages of a file
622 when resolving a three-way merge by setting the merge.renormalize
623 configuration variable. This prevents changes caused by check-in
624 conversion from causing spurious merge conflicts when a converted
625 file is merged with an unconverted file.
626 As long as a "smudge→clean" results in the same output as a "clean"
628 even on files that are already smudged, this strategy will
629 automatically resolve all filter-related conflicts. Filters that do
630 not act in this way may cause additional merge conflicts that must
631 be resolved manually.
632 Generating diff text
634 diff
635 The attribute diff affects how Git generates diffs for particular
636 files. It can tell Git whether to generate a textual patch for the
637 path or to treat the path as a binary file. It can also affect what
638 line is shown on the hunk header @@ -k,l +n,m @@ line, tell Git to
639 use an external command to generate the diff, or ask Git to convert
640 binary files to a text format before generating the diff.
641 Set
643 A path to which the diff attribute is set is treated as text,
644 even when they contain byte values that normally never appear
645 in text files, such as NUL.
646 Unset
648 A path to which the diff attribute is unset will generate
649 Binary files differ (or a binary patch, if binary patches are
650 enabled).
651 Unspecified
653 A path to which the diff attribute is unspecified first gets
654 its contents inspected, and if it looks like text and is
655 smaller than core.bigFileThreshold, it is treated as text.
656 Otherwise it would generate Binary files differ.
657 String
659 Diff is shown using the specified diff driver. Each driver may
660 specify one or more options, as described in the following
661 section. The options for the diff driver "foo" are defined by
662 the configuration variables in the "diff.foo" section of the
663 Git config file.
664 Defining an external diff driver
666 The definition of a diff driver is done in gitconfig, not
667 gitattributes file, so strictly speaking this manual page is a
668 wrong place to talk about it. However...
669 To define an external diff driver jcdiff, add a section to your
671 $GIT_DIR/config file (or $HOME/.gitconfig file) like this:
672 [diff "jcdiff"]
674 command = j-c-diff
675 When Git needs to show you a diff for the path with diff attribute
678 set to jcdiff, it calls the command you specified with the above
679 configuration, i.e. j-c-diff, with 7 parameters, just like
680 GIT_EXTERNAL_DIFF program is called. See git(1) for details.
681 Defining a custom hunk-header
683 Each group of changes (called a "hunk") in the textual diff output
684 is prefixed with a line of the form:
685 @@ -k,l +n,m @@ TEXT
687 This is called a hunk header. The "TEXT" portion is by default a
689 line that begins with an alphabet, an underscore or a dollar sign;
690 this matches what GNU diff -p output uses. This default selection
691 however is not suited for some contents, and you can use a
692 customized pattern to make a selection.
693 First, in .gitattributes, you would assign the diff attribute for
695 paths.
696 *.tex diff=tex
698 Then, you would define a "diff.tex.xfuncname" configuration to
701 specify a regular expression that matches a line that you would
702 want to appear as the hunk header "TEXT". Add a section to your
703 $GIT_DIR/config file (or $HOME/.gitconfig file) like this:
704 [diff "tex"]
706 xfuncname = "^(\\\\(sub)*section\\{.*)$"
707 Note. A single level of backslashes are eaten by the configuration
710 file parser, so you would need to double the backslashes; the
711 pattern above picks a line that begins with a backslash, and zero
712 or more occurrences of sub followed by section followed by open
713 brace, to the end of line.
714 There are a few built-in patterns to make this easier, and tex is
716 one of them, so you do not have to write the above in your
717 configuration file (you still need to enable this with the
718 attribute mechanism, via .gitattributes). The following built in
719 patterns are available:
720 · ada suitable for source code in the Ada language.
722 · bibtex suitable for files with BibTeX coded references.
724 · cpp suitable for source code in the C and C++ languages.
726 · csharp suitable for source code in the C# language.
728 · css suitable for cascading style sheets.
730 · dts suitable for devicetree (DTS) files.
732 · fortran suitable for source code in the Fortran language.
734 · fountain suitable for Fountain documents.
736 · golang suitable for source code in the Go language.
738 · html suitable for HTML/XHTML documents.
740 · java suitable for source code in the Java language.
742 · matlab suitable for source code in the MATLAB and Octave
744 languages.
745 · objc suitable for source code in the Objective-C language.
747 · pascal suitable for source code in the Pascal/Delphi language.
749 · perl suitable for source code in the Perl language.
751 · php suitable for source code in the PHP language.
753 · python suitable for source code in the Python language.
755 · ruby suitable for source code in the Ruby language.
757 · rust suitable for source code in the Rust language.
759 · tex suitable for source code for LaTeX documents.
761 Customizing word diff
763 You can customize the rules that git diff --word-diff uses to split
764 words in a line, by specifying an appropriate regular expression in
765 the "diff.*.wordRegex" configuration variable. For example, in TeX
766 a backslash followed by a sequence of letters forms a command, but
767 several such commands can be run together without intervening
768 whitespace. To separate them, use a regular expression in your
769 $GIT_DIR/config file (or $HOME/.gitconfig file) like this:
770 [diff "tex"]
772 wordRegex = "\\\\[a-zA-Z]+|[{}]|\\\\.|[^\\{}[:space:]]+"
773 A built-in pattern is provided for all languages listed in the
776 previous section.
777 Performing text diffs of binary files
779 Sometimes it is desirable to see the diff of a text-converted
780 version of some binary files. For example, a word processor
781 document can be converted to an ASCII text representation, and the
782 diff of the text shown. Even though this conversion loses some
783 information, the resulting diff is useful for human viewing (but
784 cannot be applied directly).
785 The textconv config option is used to define a program for
787 performing such a conversion. The program should take a single
788 argument, the name of a file to convert, and produce the resulting
789 text on stdout.
790 For example, to show the diff of the exif information of a file
792 instead of the binary information (assuming you have the exif tool
793 installed), add the following section to your $GIT_DIR/config file
794 (or $HOME/.gitconfig file):
795 [diff "jpg"]
797 textconv = exif
798 Note
801 The text conversion is generally a one-way conversion; in this
802 example, we lose the actual image contents and focus just on
803 the text data. This means that diffs generated by textconv are
804 not suitable for applying. For this reason, only git diff and
805 the git log family of commands (i.e., log, whatchanged, show)
806 will perform text conversion. git format-patch will never
807 generate this output. If you want to send somebody a
808 text-converted diff of a binary file (e.g., because it quickly
809 conveys the changes you have made), you should generate it
810 separately and send it as a comment in addition to the usual
811 binary diff that you might send.
812 Because text conversion can be slow, especially when doing a large
814 number of them with git log -p, Git provides a mechanism to cache
815 the output and use it in future diffs. To enable caching, set the
816 "cachetextconv" variable in your diff driver’s config. For example:
817 [diff "jpg"]
819 textconv = exif
820 cachetextconv = true
821 This will cache the result of running "exif" on each blob
824 indefinitely. If you change the textconv config variable for a diff
825 driver, Git will automatically invalidate the cache entries and
826 re-run the textconv filter. If you want to invalidate the cache
827 manually (e.g., because your version of "exif" was updated and now
828 produces better output), you can remove the cache manually with git
829 update-ref -d refs/notes/textconv/jpg (where "jpg" is the name of
830 the diff driver, as in the example above).
831 Choosing textconv versus external diff
833 If you want to show differences between binary or
834 specially-formatted blobs in your repository, you can choose to use
835 either an external diff command, or to use textconv to convert them
836 to a diff-able text format. Which method you choose depends on your
837 exact situation.
838 The advantage of using an external diff command is flexibility. You
840 are not bound to find line-oriented changes, nor is it necessary
841 for the output to resemble unified diff. You are free to locate and
842 report changes in the most appropriate way for your data format.
843 A textconv, by comparison, is much more limiting. You provide a
845 transformation of the data into a line-oriented text format, and
846 Git uses its regular diff tools to generate the output. There are
847 several advantages to choosing this method:
848 1. Ease of use. It is often much simpler to write a binary to text
850 transformation than it is to perform your own diff. In many
851 cases, existing programs can be used as textconv filters (e.g.,
852 exif, odt2txt).
853 2. Git diff features. By performing only the transformation step
855 yourself, you can still utilize many of Git’s diff features,
856 including colorization, word-diff, and combined diffs for
857 merges.
858 3. Caching. Textconv caching can speed up repeated diffs, such as
860 those you might trigger by running git log -p.
861 Marking files as binary
863 Git usually guesses correctly whether a blob contains text or
864 binary data by examining the beginning of the contents. However,
865 sometimes you may want to override its decision, either because a
866 blob contains binary data later in the file, or because the
867 content, while technically composed of text characters, is opaque
868 to a human reader. For example, many postscript files contain only
869 ASCII characters, but produce noisy and meaningless diffs.
870 The simplest way to mark a file as binary is to unset the diff
872 attribute in the .gitattributes file:
873 *.ps -diff
875 This will cause Git to generate Binary files differ (or a binary
878 patch, if binary patches are enabled) instead of a regular diff.
879 However, one may also want to specify other diff driver attributes.
881 For example, you might want to use textconv to convert postscript
882 files to an ASCII representation for human viewing, but otherwise
883 treat them as binary files. You cannot specify both -diff and
884 diff=ps attributes. The solution is to use the diff.*.binary config
885 option:
886 [diff "ps"]
888 textconv = ps2ascii
889 binary = true
890 Performing a three-way merge
893 merge
894 The attribute merge affects how three versions of a file are merged
895 when a file-level merge is necessary during git merge, and other
896 commands such as git revert and git cherry-pick.
897 Set
899 Built-in 3-way merge driver is used to merge the contents in a
900 way similar to merge command of RCS suite. This is suitable for
901 ordinary text files.
902 Unset
904 Take the version from the current branch as the tentative merge
905 result, and declare that the merge has conflicts. This is
906 suitable for binary files that do not have a well-defined merge
907 semantics.
908 Unspecified
910 By default, this uses the same built-in 3-way merge driver as
911 is the case when the merge attribute is set. However, the
912 merge.default configuration variable can name different merge
913 driver to be used with paths for which the merge attribute is
914 unspecified.
915 String
917 3-way merge is performed using the specified custom merge
918 driver. The built-in 3-way merge driver can be explicitly
919 specified by asking for "text" driver; the built-in "take the
920 current branch" driver can be requested with "binary".
921 Built-in merge drivers
923 There are a few built-in low-level merge drivers defined that can
924 be asked for via the merge attribute.
925 text
927 Usual 3-way file level merge for text files. Conflicted regions
928 are marked with conflict markers <<<<<<<, ======= and >>>>>>>.
929 The version from your branch appears before the ======= marker,
930 and the version from the merged branch appears after the
931 ======= marker.
932 binary
934 Keep the version from your branch in the work tree, but leave
935 the path in the conflicted state for the user to sort out.
936 union
938 Run 3-way file level merge for text files, but take lines from
939 both versions, instead of leaving conflict markers. This tends
940 to leave the added lines in the resulting file in random order
941 and the user should verify the result. Do not use this if you
942 do not understand the implications.
943 Defining a custom merge driver
945 The definition of a merge driver is done in the .git/config file,
946 not in the gitattributes file, so strictly speaking this manual
947 page is a wrong place to talk about it. However...
948 To define a custom merge driver filfre, add a section to your
950 $GIT_DIR/config file (or $HOME/.gitconfig file) like this:
951 [merge "filfre"]
953 name = feel-free merge driver
954 driver = filfre %O %A %B %L %P
955 recursive = binary
956 The merge.*.name variable gives the driver a human-readable name.
959 The ‘merge.*.driver` variable’s value is used to construct a
961 command to run to merge ancestor’s version (%O), current version
962 (%A) and the other branches’ version (%B). These three tokens are
963 replaced with the names of temporary files that hold the contents
964 of these versions when the command line is built. Additionally, %L
965 will be replaced with the conflict marker size (see below).
966 The merge driver is expected to leave the result of the merge in
968 the file named with %A by overwriting it, and exit with zero status
969 if it managed to merge them cleanly, or non-zero if there were
970 conflicts.
971 The merge.*.recursive variable specifies what other merge driver to
973 use when the merge driver is called for an internal merge between
974 common ancestors, when there are more than one. When left
975 unspecified, the driver itself is used for both internal merge and
976 the final merge.
977 The merge driver can learn the pathname in which the merged result
979 will be stored via placeholder %P.
980 conflict-marker-size
982 This attribute controls the length of conflict markers left in the
983 work tree file during a conflicted merge. Only setting to the value
984 to a positive integer has any meaningful effect.
985 For example, this line in .gitattributes can be used to tell the
987 merge machinery to leave much longer (instead of the usual
988 7-character-long) conflict markers when merging the file
989 Documentation/git-merge.txt results in a conflict.
990 Documentation/git-merge.txt conflict-marker-size=32
992 Checking whitespace errors
995 whitespace
996 The core.whitespace configuration variable allows you to define
997 what diff and apply should consider whitespace errors for all paths
998 in the project (See git-config(1)). This attribute gives you finer
999 control per path.
1000 Set
1002 Notice all types of potential whitespace errors known to Git.
1003 The tab width is taken from the value of the core.whitespace
1004 configuration variable.
1005 Unset
1007 Do not notice anything as error.
1008 Unspecified
1010 Use the value of the core.whitespace configuration variable to
1011 decide what to notice as error.
1012 String
1014 Specify a comma separate list of common whitespace problems to
1015 notice in the same format as the core.whitespace configuration
1016 variable.
1017 Creating an archive
1019 export-ignore
1020 Files and directories with the attribute export-ignore won’t be
1021 added to archive files.
1022 export-subst
1024 If the attribute export-subst is set for a file then Git will
1025 expand several placeholders when adding this file to an archive.
1026 The expansion depends on the availability of a commit ID, i.e., if
1027 git-archive(1) has been given a tree instead of a commit or a tag
1028 then no replacement will be done. The placeholders are the same as
1029 those for the option --pretty=format: of git-log(1), except that
1030 they need to be wrapped like this: $Format:PLACEHOLDERS$ in the
1031 file. E.g. the string $Format:%H$ will be replaced by the commit
1032 hash.
1033 Packing objects
1035 delta
1036 Delta compression will not be attempted for blobs for paths with
1037 the attribute delta set to false.
1038 Viewing files in GUI tools
1040 encoding
1041 The value of this attribute specifies the character encoding that
1042 should be used by GUI tools (e.g. gitk(1) and git-gui(1)) to
1043 display the contents of the relevant file. Note that due to
1044 performance considerations gitk(1) does not use this attribute
1045 unless you manually enable per-file encodings in its options.
1046 If this attribute is not set or has an invalid value, the value of
1048 the gui.encoding configuration variable is used instead (See git-
1049 config(1)).
1050
1052 You do not want any end-of-line conversions applied to, nor textual
1053 diffs produced for, any binary file you track. You would need to
1054 specify e.g.
1055 *.jpg -text -diff
1057 but that may become cumbersome, when you have many attributes. Using
1060 macro attributes, you can define an attribute that, when set, also sets
1061 or unsets a number of other attributes at the same time. The system
1062 knows a built-in macro attribute, binary:
1063 *.jpg binary
1065 Setting the "binary" attribute also unsets the "text" and "diff"
1068 attributes as above. Note that macro attributes can only be "Set",
1069 though setting one might have the effect of setting or unsetting other
1070 attributes or even returning other attributes to the "Unspecified"
1071 state.
1072
1074 Custom macro attributes can be defined only in top-level gitattributes
1075 files ($GIT_DIR/info/attributes, the .gitattributes file at the top
1076 level of the working tree, or the global or system-wide gitattributes
1077 files), not in .gitattributes files in working tree subdirectories. The
1078 built-in macro attribute "binary" is equivalent to:
1079 [attr]binary -diff -merge -text
1081
1084 If you have these three gitattributes file:
1085 (in $GIT_DIR/info/attributes)
1087 a* foo !bar -baz
1089 (in .gitattributes)
1091 abc foo bar baz
1092 (in t/.gitattributes)
1094 ab* merge=filfre
1095 abc -foo -bar
1096 *.c frotz
1097 the attributes given to path t/abc are computed as follows:
1100 1. By examining t/.gitattributes (which is in the same directory as
1102 the path in question), Git finds that the first line matches.
1103 merge attribute is set. It also finds that the second line matches,
1104 and attributes foo and bar are unset.
1105 2. Then it examines .gitattributes (which is in the parent directory),
1107 and finds that the first line matches, but t/.gitattributes file
1108 already decided how merge, foo and bar attributes should be given
1109 to this path, so it leaves foo and bar unset. Attribute baz is set.
1110 3. Finally it examines $GIT_DIR/info/attributes. This file is used to
1112 override the in-tree settings. The first line is a match, and foo
1113 is set, bar is reverted to unspecified state, and baz is unset.
1114 As the result, the attributes assignment to t/abc becomes:
1116 foo set to true
1118 bar unspecified
1119 baz set to false
1120 merge set to string value "filfre"
1121 frotz unspecified
1122
1125 git-check-attr(1).
1126
1128 Part of the git(1) suite
1129Git 2.24.1 12/10/2019 GITATTRIBUTES(5)