6 gitattributes - defining attributes per path
10 $GIT_DIR/info/attributes, .gitattributes
16 A `gitattributes` file is a simple text file that gives
17 `attributes` to pathnames.
19 Each line in `gitattributes` file is of form:
21 pattern attr1 attr2 ...
23 That is, a pattern followed by an attributes list,
24 separated by whitespaces. When the pattern matches the
25 path in question, the attributes listed on the line are given to
28 Each attribute can be in one of these states for a given path:
32 The path has the attribute with special value "true";
33 this is specified by listing only the name of the
34 attribute in the attribute list.
38 The path has the attribute with special value "false";
39 this is specified by listing the name of the attribute
40 prefixed with a dash `-` in the attribute list.
44 The path has the attribute with specified string value;
45 this is specified by listing the name of the attribute
46 followed by an equal sign `=` and its value in the
51 No pattern matches the path, and nothing says if
52 the path has or does not have the attribute, the
53 attribute for the path is said to be Unspecified.
55 When more than one pattern matches the path, a later line
56 overrides an earlier line. This overriding is done per
57 attribute. The rules how the pattern matches paths are the
58 same as in `.gitignore` files; see linkgit:gitignore[5].
59 Unlike `.gitignore`, negative patterns are forbidden.
61 When deciding what attributes are assigned to a path, Git
62 consults `$GIT_DIR/info/attributes` file (which has the highest
63 precedence), `.gitattributes` file in the same directory as the
64 path in question, and its parent directories up to the toplevel of the
65 work tree (the further the directory that contains `.gitattributes`
66 is from the path in question, the lower its precedence). Finally
67 global and system-wide files are considered (they have the lowest
70 When the `.gitattributes` file is missing from the work tree, the
71 path in the index is used as a fall-back. During checkout process,
72 `.gitattributes` in the index is used and then the file in the
73 working tree is used as a fall-back.
75 If you wish to affect only a single repository (i.e., to assign
76 attributes to files that are particular to
77 one user's workflow for that repository), then
78 attributes should be placed in the `$GIT_DIR/info/attributes` file.
79 Attributes which should be version-controlled and distributed to other
80 repositories (i.e., attributes of interest to all users) should go into
81 `.gitattributes` files. Attributes that should affect all repositories
82 for a single user should be placed in a file specified by the
83 `core.attributesFile` configuration option (see linkgit:git-config[1]).
84 Its default value is $XDG_CONFIG_HOME/git/attributes. If $XDG_CONFIG_HOME
85 is either not set or empty, $HOME/.config/git/attributes is used instead.
86 Attributes for all users on a system should be placed in the
87 `$(prefix)/etc/gitattributes` file.
89 Sometimes you would need to override an setting of an attribute
90 for a path to `Unspecified` state. This can be done by listing
91 the name of the attribute prefixed with an exclamation point `!`.
97 Certain operations by Git can be influenced by assigning
98 particular attributes to a path. Currently, the following
99 operations are attributes-aware.
101 Checking-out and checking-in
102 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
104 These attributes affect how the contents stored in the
105 repository are copied to the working tree files when commands
106 such as 'git checkout' and 'git merge' run. They also affect how
107 Git stores the contents you prepare in the working tree in the
108 repository upon 'git add' and 'git commit'.
113 This attribute enables and controls end-of-line normalization. When a
114 text file is normalized, its line endings are converted to LF in the
115 repository. To control what line ending style is used in the working
116 directory, use the `eol` attribute for a single file and the
117 `core.eol` configuration variable for all text files.
118 Note that `core.autocrlf` overrides `core.eol`
122 Setting the `text` attribute on a path enables end-of-line
123 normalization and marks the path as a text file. End-of-line
124 conversion takes place without guessing the content type.
128 Unsetting the `text` attribute on a path tells Git not to
129 attempt any end-of-line conversion upon checkin or checkout.
131 Set to string value "auto"::
133 When `text` is set to "auto", the path is marked for automatic
134 end-of-line conversion. If Git decides that the content is
135 text, its line endings are converted to LF on checkin.
136 When the file has been commited with CRLF, no conversion is done.
140 If the `text` attribute is unspecified, Git uses the
141 `core.autocrlf` configuration variable to determine if the
142 file should be converted.
144 Any other value causes Git to act as if `text` has been left
150 This attribute sets a specific line-ending style to be used in the
151 working directory. It enables end-of-line conversion without any
152 content checks, effectively setting the `text` attribute.
154 Set to string value "crlf"::
156 This setting forces Git to normalize line endings for this
157 file on checkin and convert them to CRLF when the file is
160 Set to string value "lf"::
162 This setting forces Git to normalize line endings to LF on
163 checkin and prevents conversion to CRLF when the file is
166 Backwards compatibility with `crlf` attribute
167 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
169 For backwards compatibility, the `crlf` attribute is interpreted as
172 ------------------------
176 ------------------------
178 End-of-line conversion
179 ^^^^^^^^^^^^^^^^^^^^^^
181 While Git normally leaves file contents alone, it can be configured to
182 normalize line endings to LF in the repository and, optionally, to
183 convert them to CRLF when files are checked out.
185 Here is an example that will make Git normalize .txt, .vcproj and .sh
186 files, ensure that .vcproj files have CRLF and .sh files have LF in
187 the working directory, and prevent .jpg files from being normalized
188 regardless of their content.
190 ------------------------
193 *.vcproj text eol=crlf
196 ------------------------
198 Other source code management systems normalize all text files in their
199 repositories, and there are two ways to enable similar automatic
200 normalization in Git.
202 If you simply want to have CRLF line endings in your working directory
203 regardless of the repository you are working with, you can set the
204 config variable "core.autocrlf" without using any attributes.
206 ------------------------
209 ------------------------
211 This does not force normalization of all text files, but does ensure
212 that text files that you introduce to the repository have their line
213 endings normalized to LF when they are added, and that files that are
214 already normalized in the repository stay normalized.
216 If you want to interoperate with a source code management system that
217 enforces end-of-line normalization, or you simply want all text files
218 in your repository to be normalized, you should instead set the `text`
219 attribute to "auto" for _all_ files.
221 ------------------------
223 ------------------------
225 This ensures that all files that Git considers to be text will have
226 normalized (LF) line endings in the repository. The `core.eol`
227 configuration variable controls which line endings Git will use for
228 normalized files in your working directory; the default is to use the
229 native line ending for your platform, or CRLF if `core.autocrlf` is
232 NOTE: When `text=auto` normalization is enabled in an existing
233 repository, any text files containing CRLFs should be normalized. If
234 they are not they will be normalized the next time someone tries to
235 change them, causing unfortunate misattribution. From a clean working
238 -------------------------------------------------
239 $ echo "* text=auto" >>.gitattributes
240 $ rm .git/index # Remove the index to force Git to
241 $ git reset # re-scan the working directory
242 $ git status # Show files that will be normalized
244 $ git add .gitattributes
245 $ git commit -m "Introduce end-of-line normalization"
246 -------------------------------------------------
248 If any files that should not be normalized show up in 'git status',
249 unset their `text` attribute before running 'git add -u'.
251 ------------------------
253 ------------------------
255 Conversely, text files that Git does not detect can have normalization
258 ------------------------
260 ------------------------
262 If `core.safecrlf` is set to "true" or "warn", Git verifies if
263 the conversion is reversible for the current setting of
264 `core.autocrlf`. For "true", Git rejects irreversible
265 conversions; for "warn", Git only prints a warning but accepts
266 an irreversible conversion. The safety triggers to prevent such
267 a conversion done to the files in the work tree, but there are a
268 few exceptions. Even though...
270 - 'git add' itself does not touch the files in the work tree, the
271 next checkout would, so the safety triggers;
273 - 'git apply' to update a text file with a patch does touch the files
274 in the work tree, but the operation is about text files and CRLF
275 conversion is about fixing the line ending inconsistencies, so the
276 safety does not trigger;
278 - 'git diff' itself does not touch the files in the work tree, it is
279 often run to inspect the changes you intend to next 'git add'. To
280 catch potential problems early, safety triggers.
286 When the attribute `ident` is set for a path, Git replaces
287 `$Id$` in the blob object with `$Id:`, followed by the
288 40-character hexadecimal blob object name, followed by a dollar
289 sign `$` upon checkout. Any byte sequence that begins with
290 `$Id:` and ends with `$` in the worktree file is replaced
291 with `$Id$` upon check-in.
297 A `filter` attribute can be set to a string value that names a
298 filter driver specified in the configuration.
300 A filter driver consists of a `clean` command and a `smudge`
301 command, either of which can be left unspecified. Upon
302 checkout, when the `smudge` command is specified, the command is
303 fed the blob object from its standard input, and its standard
304 output is used to update the worktree file. Similarly, the
305 `clean` command is used to convert the contents of worktree file
308 One use of the content filtering is to massage the content into a shape
309 that is more convenient for the platform, filesystem, and the user to use.
310 For this mode of operation, the key phrase here is "more convenient" and
311 not "turning something unusable into usable". In other words, the intent
312 is that if someone unsets the filter driver definition, or does not have
313 the appropriate filter program, the project should still be usable.
315 Another use of the content filtering is to store the content that cannot
316 be directly used in the repository (e.g. a UUID that refers to the true
317 content stored outside Git, or an encrypted content) and turn it into a
318 usable form upon checkout (e.g. download the external content, or decrypt
319 the encrypted content).
321 These two filters behave differently, and by default, a filter is taken as
322 the former, massaging the contents into more convenient shape. A missing
323 filter driver definition in the config, or a filter driver that exits with
324 a non-zero status, is not an error but makes the filter a no-op passthru.
326 You can declare that a filter turns a content that by itself is unusable
327 into a usable content by setting the filter.<driver>.required configuration
330 For example, in .gitattributes, you would assign the `filter`
333 ------------------------
335 ------------------------
337 Then you would define a "filter.indent.clean" and "filter.indent.smudge"
338 configuration in your .git/config to specify a pair of commands to
339 modify the contents of C programs when the source files are checked
340 in ("clean" is run) and checked out (no change is made because the
343 ------------------------
347 ------------------------
349 For best results, `clean` should not alter its output further if it is
350 run twice ("clean->clean" should be equivalent to "clean"), and
351 multiple `smudge` commands should not alter `clean`'s output
352 ("smudge->smudge->clean" should be equivalent to "clean"). See the
353 section on merging below.
355 The "indent" filter is well-behaved in this regard: it will not modify
356 input that is already correctly indented. In this case, the lack of a
357 smudge filter means that the clean filter _must_ accept its own output
358 without modifying it.
360 If a filter _must_ succeed in order to make the stored contents usable,
361 you can declare that the filter is `required`, in the configuration:
363 ------------------------
365 clean = openssl enc ...
366 smudge = openssl enc -d ...
368 ------------------------
370 Sequence "%f" on the filter command line is replaced with the name of
371 the file the filter is working on. A filter might use this in keyword
372 substitution. For example:
374 ------------------------
376 clean = git-p4-filter --clean %f
377 smudge = git-p4-filter --smudge %f
378 ------------------------
381 Interaction between checkin/checkout attributes
382 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
384 In the check-in codepath, the worktree file is first converted
385 with `filter` driver (if specified and corresponding driver
386 defined), then the result is processed with `ident` (if
387 specified), and then finally with `text` (again, if specified
390 In the check-out codepath, the blob content is first converted
391 with `text`, and then `ident` and fed to `filter`.
394 Merging branches with differing checkin/checkout attributes
395 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
397 If you have added attributes to a file that cause the canonical
398 repository format for that file to change, such as adding a
399 clean/smudge filter or text/eol/ident attributes, merging anything
400 where the attribute is not in place would normally cause merge
403 To prevent these unnecessary merge conflicts, Git can be told to run a
404 virtual check-out and check-in of all three stages of a file when
405 resolving a three-way merge by setting the `merge.renormalize`
406 configuration variable. This prevents changes caused by check-in
407 conversion from causing spurious merge conflicts when a converted file
408 is merged with an unconverted file.
410 As long as a "smudge->clean" results in the same output as a "clean"
411 even on files that are already smudged, this strategy will
412 automatically resolve all filter-related conflicts. Filters that do
413 not act in this way may cause additional merge conflicts that must be
423 The attribute `diff` affects how Git generates diffs for particular
424 files. It can tell Git whether to generate a textual patch for the path
425 or to treat the path as a binary file. It can also affect what line is
426 shown on the hunk header `@@ -k,l +n,m @@` line, tell Git to use an
427 external command to generate the diff, or ask Git to convert binary
428 files to a text format before generating the diff.
432 A path to which the `diff` attribute is set is treated
433 as text, even when they contain byte values that
434 normally never appear in text files, such as NUL.
438 A path to which the `diff` attribute is unset will
439 generate `Binary files differ` (or a binary patch, if
440 binary patches are enabled).
444 A path to which the `diff` attribute is unspecified
445 first gets its contents inspected, and if it looks like
446 text and is smaller than core.bigFileThreshold, it is treated
447 as text. Otherwise it would generate `Binary files differ`.
451 Diff is shown using the specified diff driver. Each driver may
452 specify one or more options, as described in the following
453 section. The options for the diff driver "foo" are defined
454 by the configuration variables in the "diff.foo" section of the
458 Defining an external diff driver
459 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
461 The definition of a diff driver is done in `gitconfig`, not
462 `gitattributes` file, so strictly speaking this manual page is a
463 wrong place to talk about it. However...
465 To define an external diff driver `jcdiff`, add a section to your
466 `$GIT_DIR/config` file (or `$HOME/.gitconfig` file) like this:
468 ----------------------------------------------------------------
471 ----------------------------------------------------------------
473 When Git needs to show you a diff for the path with `diff`
474 attribute set to `jcdiff`, it calls the command you specified
475 with the above configuration, i.e. `j-c-diff`, with 7
476 parameters, just like `GIT_EXTERNAL_DIFF` program is called.
477 See linkgit:git[1] for details.
480 Defining a custom hunk-header
481 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
483 Each group of changes (called a "hunk") in the textual diff output
484 is prefixed with a line of the form:
488 This is called a 'hunk header'. The "TEXT" portion is by default a line
489 that begins with an alphabet, an underscore or a dollar sign; this
490 matches what GNU 'diff -p' output uses. This default selection however
491 is not suited for some contents, and you can use a customized pattern
494 First, in .gitattributes, you would assign the `diff` attribute
497 ------------------------
499 ------------------------
501 Then, you would define a "diff.tex.xfuncname" configuration to
502 specify a regular expression that matches a line that you would
503 want to appear as the hunk header "TEXT". Add a section to your
504 `$GIT_DIR/config` file (or `$HOME/.gitconfig` file) like this:
506 ------------------------
508 xfuncname = "^(\\\\(sub)*section\\{.*)$"
509 ------------------------
511 Note. A single level of backslashes are eaten by the
512 configuration file parser, so you would need to double the
513 backslashes; the pattern above picks a line that begins with a
514 backslash, and zero or more occurrences of `sub` followed by
515 `section` followed by open brace, to the end of line.
517 There are a few built-in patterns to make this easier, and `tex`
518 is one of them, so you do not have to write the above in your
519 configuration file (you still need to enable this with the
520 attribute mechanism, via `.gitattributes`). The following built in
521 patterns are available:
523 - `ada` suitable for source code in the Ada language.
525 - `bibtex` suitable for files with BibTeX coded references.
527 - `cpp` suitable for source code in the C and C++ languages.
529 - `csharp` suitable for source code in the C# language.
531 - `fortran` suitable for source code in the Fortran language.
533 - `fountain` suitable for Fountain documents.
535 - `html` suitable for HTML/XHTML documents.
537 - `java` suitable for source code in the Java language.
539 - `matlab` suitable for source code in the MATLAB language.
541 - `objc` suitable for source code in the Objective-C language.
543 - `pascal` suitable for source code in the Pascal/Delphi language.
545 - `perl` suitable for source code in the Perl language.
547 - `php` suitable for source code in the PHP language.
549 - `python` suitable for source code in the Python language.
551 - `ruby` suitable for source code in the Ruby language.
553 - `tex` suitable for source code for LaTeX documents.
556 Customizing word diff
557 ^^^^^^^^^^^^^^^^^^^^^
559 You can customize the rules that `git diff --word-diff` uses to
560 split words in a line, by specifying an appropriate regular expression
561 in the "diff.*.wordRegex" configuration variable. For example, in TeX
562 a backslash followed by a sequence of letters forms a command, but
563 several such commands can be run together without intervening
564 whitespace. To separate them, use a regular expression in your
565 `$GIT_DIR/config` file (or `$HOME/.gitconfig` file) like this:
567 ------------------------
569 wordRegex = "\\\\[a-zA-Z]+|[{}]|\\\\.|[^\\{}[:space:]]+"
570 ------------------------
572 A built-in pattern is provided for all languages listed in the
576 Performing text diffs of binary files
577 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
579 Sometimes it is desirable to see the diff of a text-converted
580 version of some binary files. For example, a word processor
581 document can be converted to an ASCII text representation, and
582 the diff of the text shown. Even though this conversion loses
583 some information, the resulting diff is useful for human
584 viewing (but cannot be applied directly).
586 The `textconv` config option is used to define a program for
587 performing such a conversion. The program should take a single
588 argument, the name of a file to convert, and produce the
589 resulting text on stdout.
591 For example, to show the diff of the exif information of a
592 file instead of the binary information (assuming you have the
593 exif tool installed), add the following section to your
594 `$GIT_DIR/config` file (or `$HOME/.gitconfig` file):
596 ------------------------
599 ------------------------
601 NOTE: The text conversion is generally a one-way conversion;
602 in this example, we lose the actual image contents and focus
603 just on the text data. This means that diffs generated by
604 textconv are _not_ suitable for applying. For this reason,
605 only `git diff` and the `git log` family of commands (i.e.,
606 log, whatchanged, show) will perform text conversion. `git
607 format-patch` will never generate this output. If you want to
608 send somebody a text-converted diff of a binary file (e.g.,
609 because it quickly conveys the changes you have made), you
610 should generate it separately and send it as a comment _in
611 addition to_ the usual binary diff that you might send.
613 Because text conversion can be slow, especially when doing a
614 large number of them with `git log -p`, Git provides a mechanism
615 to cache the output and use it in future diffs. To enable
616 caching, set the "cachetextconv" variable in your diff driver's
619 ------------------------
623 ------------------------
625 This will cache the result of running "exif" on each blob
626 indefinitely. If you change the textconv config variable for a
627 diff driver, Git will automatically invalidate the cache entries
628 and re-run the textconv filter. If you want to invalidate the
629 cache manually (e.g., because your version of "exif" was updated
630 and now produces better output), you can remove the cache
631 manually with `git update-ref -d refs/notes/textconv/jpg` (where
632 "jpg" is the name of the diff driver, as in the example above).
634 Choosing textconv versus external diff
635 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
637 If you want to show differences between binary or specially-formatted
638 blobs in your repository, you can choose to use either an external diff
639 command, or to use textconv to convert them to a diff-able text format.
640 Which method you choose depends on your exact situation.
642 The advantage of using an external diff command is flexibility. You are
643 not bound to find line-oriented changes, nor is it necessary for the
644 output to resemble unified diff. You are free to locate and report
645 changes in the most appropriate way for your data format.
647 A textconv, by comparison, is much more limiting. You provide a
648 transformation of the data into a line-oriented text format, and Git
649 uses its regular diff tools to generate the output. There are several
650 advantages to choosing this method:
652 1. Ease of use. It is often much simpler to write a binary to text
653 transformation than it is to perform your own diff. In many cases,
654 existing programs can be used as textconv filters (e.g., exif,
657 2. Git diff features. By performing only the transformation step
658 yourself, you can still utilize many of Git's diff features,
659 including colorization, word-diff, and combined diffs for merges.
661 3. Caching. Textconv caching can speed up repeated diffs, such as those
662 you might trigger by running `git log -p`.
665 Marking files as binary
666 ^^^^^^^^^^^^^^^^^^^^^^^
668 Git usually guesses correctly whether a blob contains text or binary
669 data by examining the beginning of the contents. However, sometimes you
670 may want to override its decision, either because a blob contains binary
671 data later in the file, or because the content, while technically
672 composed of text characters, is opaque to a human reader. For example,
673 many postscript files contain only ASCII characters, but produce noisy
674 and meaningless diffs.
676 The simplest way to mark a file as binary is to unset the diff
677 attribute in the `.gitattributes` file:
679 ------------------------
681 ------------------------
683 This will cause Git to generate `Binary files differ` (or a binary
684 patch, if binary patches are enabled) instead of a regular diff.
686 However, one may also want to specify other diff driver attributes. For
687 example, you might want to use `textconv` to convert postscript files to
688 an ASCII representation for human viewing, but otherwise treat them as
689 binary files. You cannot specify both `-diff` and `diff=ps` attributes.
690 The solution is to use the `diff.*.binary` config option:
692 ------------------------
696 ------------------------
698 Performing a three-way merge
699 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
704 The attribute `merge` affects how three versions of a file are
705 merged when a file-level merge is necessary during `git merge`,
706 and other commands such as `git revert` and `git cherry-pick`.
710 Built-in 3-way merge driver is used to merge the
711 contents in a way similar to 'merge' command of `RCS`
712 suite. This is suitable for ordinary text files.
716 Take the version from the current branch as the
717 tentative merge result, and declare that the merge has
718 conflicts. This is suitable for binary files that do
719 not have a well-defined merge semantics.
723 By default, this uses the same built-in 3-way merge
724 driver as is the case when the `merge` attribute is set.
725 However, the `merge.default` configuration variable can name
726 different merge driver to be used with paths for which the
727 `merge` attribute is unspecified.
731 3-way merge is performed using the specified custom
732 merge driver. The built-in 3-way merge driver can be
733 explicitly specified by asking for "text" driver; the
734 built-in "take the current branch" driver can be
735 requested with "binary".
738 Built-in merge drivers
739 ^^^^^^^^^^^^^^^^^^^^^^
741 There are a few built-in low-level merge drivers defined that
742 can be asked for via the `merge` attribute.
746 Usual 3-way file level merge for text files. Conflicted
747 regions are marked with conflict markers `<<<<<<<`,
748 `=======` and `>>>>>>>`. The version from your branch
749 appears before the `=======` marker, and the version
750 from the merged branch appears after the `=======`
755 Keep the version from your branch in the work tree, but
756 leave the path in the conflicted state for the user to
761 Run 3-way file level merge for text files, but take
762 lines from both versions, instead of leaving conflict
763 markers. This tends to leave the added lines in the
764 resulting file in random order and the user should
765 verify the result. Do not use this if you do not
766 understand the implications.
769 Defining a custom merge driver
770 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
772 The definition of a merge driver is done in the `.git/config`
773 file, not in the `gitattributes` file, so strictly speaking this
774 manual page is a wrong place to talk about it. However...
776 To define a custom merge driver `filfre`, add a section to your
777 `$GIT_DIR/config` file (or `$HOME/.gitconfig` file) like this:
779 ----------------------------------------------------------------
781 name = feel-free merge driver
782 driver = filfre %O %A %B %L %P
784 ----------------------------------------------------------------
786 The `merge.*.name` variable gives the driver a human-readable
789 The `merge.*.driver` variable's value is used to construct a
790 command to run to merge ancestor's version (`%O`), current
791 version (`%A`) and the other branches' version (`%B`). These
792 three tokens are replaced with the names of temporary files that
793 hold the contents of these versions when the command line is
794 built. Additionally, %L will be replaced with the conflict marker
797 The merge driver is expected to leave the result of the merge in
798 the file named with `%A` by overwriting it, and exit with zero
799 status if it managed to merge them cleanly, or non-zero if there
802 The `merge.*.recursive` variable specifies what other merge
803 driver to use when the merge driver is called for an internal
804 merge between common ancestors, when there are more than one.
805 When left unspecified, the driver itself is used for both
806 internal merge and the final merge.
808 The merge driver can learn the pathname in which the merged result
809 will be stored via placeholder `%P`.
812 `conflict-marker-size`
813 ^^^^^^^^^^^^^^^^^^^^^^
815 This attribute controls the length of conflict markers left in
816 the work tree file during a conflicted merge. Only setting to
817 the value to a positive integer has any meaningful effect.
819 For example, this line in `.gitattributes` can be used to tell the merge
820 machinery to leave much longer (instead of the usual 7-character-long)
821 conflict markers when merging the file `Documentation/git-merge.txt`
822 results in a conflict.
824 ------------------------
825 Documentation/git-merge.txt conflict-marker-size=32
826 ------------------------
829 Checking whitespace errors
830 ~~~~~~~~~~~~~~~~~~~~~~~~~~
835 The `core.whitespace` configuration variable allows you to define what
836 'diff' and 'apply' should consider whitespace errors for all paths in
837 the project (See linkgit:git-config[1]). This attribute gives you finer
842 Notice all types of potential whitespace errors known to Git.
843 The tab width is taken from the value of the `core.whitespace`
844 configuration variable.
848 Do not notice anything as error.
852 Use the value of the `core.whitespace` configuration variable to
853 decide what to notice as error.
857 Specify a comma separate list of common whitespace problems to
858 notice in the same format as the `core.whitespace` configuration
868 Files and directories with the attribute `export-ignore` won't be added to
874 If the attribute `export-subst` is set for a file then Git will expand
875 several placeholders when adding this file to an archive. The
876 expansion depends on the availability of a commit ID, i.e., if
877 linkgit:git-archive[1] has been given a tree instead of a commit or a
878 tag then no replacement will be done. The placeholders are the same
879 as those for the option `--pretty=format:` of linkgit:git-log[1],
880 except that they need to be wrapped like this: `$Format:PLACEHOLDERS$`
881 in the file. E.g. the string `$Format:%H$` will be replaced by the
891 Delta compression will not be attempted for blobs for paths with the
892 attribute `delta` set to false.
895 Viewing files in GUI tools
896 ~~~~~~~~~~~~~~~~~~~~~~~~~~
901 The value of this attribute specifies the character encoding that should
902 be used by GUI tools (e.g. linkgit:gitk[1] and linkgit:git-gui[1]) to
903 display the contents of the relevant file. Note that due to performance
904 considerations linkgit:gitk[1] does not use this attribute unless you
905 manually enable per-file encodings in its options.
907 If this attribute is not set or has an invalid value, the value of the
908 `gui.encoding` configuration variable is used instead
909 (See linkgit:git-config[1]).
912 USING MACRO ATTRIBUTES
913 ----------------------
915 You do not want any end-of-line conversions applied to, nor textual diffs
916 produced for, any binary file you track. You would need to specify e.g.
922 but that may become cumbersome, when you have many attributes. Using
923 macro attributes, you can define an attribute that, when set, also
924 sets or unsets a number of other attributes at the same time. The
925 system knows a built-in macro attribute, `binary`:
931 Setting the "binary" attribute also unsets the "text" and "diff"
932 attributes as above. Note that macro attributes can only be "Set",
933 though setting one might have the effect of setting or unsetting other
934 attributes or even returning other attributes to the "Unspecified"
938 DEFINING MACRO ATTRIBUTES
939 -------------------------
941 Custom macro attributes can be defined only in top-level gitattributes
942 files (`$GIT_DIR/info/attributes`, the `.gitattributes` file at the
943 top level of the working tree, or the global or system-wide
944 gitattributes files), not in `.gitattributes` files in working tree
945 subdirectories. The built-in macro attribute "binary" is equivalent
949 [attr]binary -diff -merge -text
956 If you have these three `gitattributes` file:
958 ----------------------------------------------------------------
959 (in $GIT_DIR/info/attributes)
966 (in t/.gitattributes)
970 ----------------------------------------------------------------
972 the attributes given to path `t/abc` are computed as follows:
974 1. By examining `t/.gitattributes` (which is in the same
975 directory as the path in question), Git finds that the first
976 line matches. `merge` attribute is set. It also finds that
977 the second line matches, and attributes `foo` and `bar`
980 2. Then it examines `.gitattributes` (which is in the parent
981 directory), and finds that the first line matches, but
982 `t/.gitattributes` file already decided how `merge`, `foo`
983 and `bar` attributes should be given to this path, so it
984 leaves `foo` and `bar` unset. Attribute `baz` is set.
986 3. Finally it examines `$GIT_DIR/info/attributes`. This file
987 is used to override the in-tree settings. The first line is
988 a match, and `foo` is set, `bar` is reverted to unspecified
989 state, and `baz` is unset.
991 As the result, the attributes assignment to `t/abc` becomes:
993 ----------------------------------------------------------------
997 merge set to string value "filfre"
999 ----------------------------------------------------------------
1004 linkgit:git-check-attr[1].
1008 Part of the linkgit:git[1] suite