revisions.txt: mark optional rev arguments with []
[git/git.git] / Documentation / revisions.txt
1 SPECIFYING REVISIONS
2 --------------------
3
4 A revision parameter '<rev>' typically, but not necessarily, names a
5 commit object. It uses what is called an 'extended SHA-1'
6 syntax. Here are various ways to spell object names. The
7 ones listed near the end of this list name trees and
8 blobs contained in a commit.
9
10 NOTE: This document shows the "raw" syntax as seen by git. The shell
11 and other UIs might require additional quoting to protect special
12 characters and to avoid word splitting.
13
14 '<sha1>', e.g. 'dae86e1950b1277e545cee180551750029cfe735', 'dae86e'::
15 The full SHA-1 object name (40-byte hexadecimal string), or
16 a leading substring that is unique within the repository.
17 E.g. dae86e1950b1277e545cee180551750029cfe735 and dae86e both
18 name the same commit object if there is no other object in
19 your repository whose object name starts with dae86e.
20
21 '<describeOutput>', e.g. 'v1.7.4.2-679-g3bee7fb'::
22 Output from `git describe`; i.e. a closest tag, optionally
23 followed by a dash and a number of commits, followed by a dash, a
24 'g', and an abbreviated object name.
25
26 '<refname>', e.g. 'master', 'heads/master', 'refs/heads/master'::
27 A symbolic ref name. E.g. 'master' typically means the commit
28 object referenced by 'refs/heads/master'. If you
29 happen to have both 'heads/master' and 'tags/master', you can
30 explicitly say 'heads/master' to tell Git which one you mean.
31 When ambiguous, a '<refname>' is disambiguated by taking the
32 first match in the following rules:
33
34 . If '$GIT_DIR/<refname>' exists, that is what you mean (this is usually
35 useful only for `HEAD`, `FETCH_HEAD`, `ORIG_HEAD`, `MERGE_HEAD`
36 and `CHERRY_PICK_HEAD`);
37
38 . otherwise, 'refs/<refname>' if it exists;
39
40 . otherwise, 'refs/tags/<refname>' if it exists;
41
42 . otherwise, 'refs/heads/<refname>' if it exists;
43
44 . otherwise, 'refs/remotes/<refname>' if it exists;
45
46 . otherwise, 'refs/remotes/<refname>/HEAD' if it exists.
47 +
48 `HEAD` names the commit on which you based the changes in the working tree.
49 `FETCH_HEAD` records the branch which you fetched from a remote repository
50 with your last `git fetch` invocation.
51 `ORIG_HEAD` is created by commands that move your `HEAD` in a drastic
52 way, to record the position of the `HEAD` before their operation, so that
53 you can easily change the tip of the branch back to the state before you ran
54 them.
55 `MERGE_HEAD` records the commit(s) which you are merging into your branch
56 when you run `git merge`.
57 `CHERRY_PICK_HEAD` records the commit which you are cherry-picking
58 when you run `git cherry-pick`.
59 +
60 Note that any of the 'refs/*' cases above may come either from
61 the '$GIT_DIR/refs' directory or from the '$GIT_DIR/packed-refs' file.
62 While the ref name encoding is unspecified, UTF-8 is preferred as
63 some output processing may assume ref names in UTF-8.
64
65 '@'::
66 '@' alone is a shortcut for `HEAD`.
67
68 '[<refname>]@{<date>}', e.g. 'master@\{yesterday\}', 'HEAD@{5 minutes ago}'::
69 A ref followed by the suffix '@' with a date specification
70 enclosed in a brace
71 pair (e.g. '\{yesterday\}', '{1 month 2 weeks 3 days 1 hour 1
72 second ago}' or '{1979-02-26 18:30:00}') specifies the value
73 of the ref at a prior point in time. This suffix may only be
74 used immediately following a ref name and the ref must have an
75 existing log ('$GIT_DIR/logs/<ref>'). Note that this looks up the state
76 of your *local* ref at a given time; e.g., what was in your local
77 'master' branch last week. If you want to look at commits made during
78 certain times, see `--since` and `--until`.
79
80 '<refname>@{<n>}', e.g. 'master@\{1\}'::
81 A ref followed by the suffix '@' with an ordinal specification
82 enclosed in a brace pair (e.g. '\{1\}', '\{15\}') specifies
83 the n-th prior value of that ref. For example 'master@\{1\}'
84 is the immediate prior value of 'master' while 'master@\{5\}'
85 is the 5th prior value of 'master'. This suffix may only be used
86 immediately following a ref name and the ref must have an existing
87 log ('$GIT_DIR/logs/<refname>').
88
89 '@{<n>}', e.g. '@\{1\}'::
90 You can use the '@' construct with an empty ref part to get at a
91 reflog entry of the current branch. For example, if you are on
92 branch 'blabla' then '@\{1\}' means the same as 'blabla@\{1\}'.
93
94 '@{-<n>}', e.g. '@{-1}'::
95 The construct '@{-<n>}' means the <n>th branch/commit checked out
96 before the current one.
97
98 '[<branchname>]@\{upstream\}', e.g. 'master@\{upstream\}', '@\{u\}'::
99 The suffix '@\{upstream\}' to a branchname (short form '<branchname>@\{u\}')
100 refers to the branch that the branch specified by branchname is set to build on
101 top of (configured with `branch.<name>.remote` and
102 `branch.<name>.merge`). A missing branchname defaults to the
103 current one. These suffixes are also accepted when spelled in uppercase, and
104 they mean the same thing no matter the case.
105
106 '[<branchname>]@\{push\}', e.g. 'master@\{push\}', '@\{push\}'::
107 The suffix '@\{push}' reports the branch "where we would push to" if
108 `git push` were run while `branchname` was checked out (or the current
109 `HEAD` if no branchname is specified). Since our push destination is
110 in a remote repository, of course, we report the local tracking branch
111 that corresponds to that branch (i.e., something in 'refs/remotes/').
112 +
113 Here's an example to make it more clear:
114 +
115 ------------------------------
116 $ git config push.default current
117 $ git config remote.pushdefault myfork
118 $ git checkout -b mybranch origin/master
119
120 $ git rev-parse --symbolic-full-name @{upstream}
121 refs/remotes/origin/master
122
123 $ git rev-parse --symbolic-full-name @{push}
124 refs/remotes/myfork/mybranch
125 ------------------------------
126 +
127 Note in the example that we set up a triangular workflow, where we pull
128 from one location and push to another. In a non-triangular workflow,
129 '@\{push}' is the same as '@\{upstream}', and there is no need for it.
130 +
131 This suffix is also accepted when spelled in uppercase, and means the same
132 thing no matter the case.
133
134 '<rev>{caret}[<n>]', e.g. 'HEAD{caret}, v1.5.1{caret}0'::
135 A suffix '{caret}' to a revision parameter means the first parent of
136 that commit object. '{caret}<n>' means the <n>th parent (i.e.
137 '<rev>{caret}'
138 is equivalent to '<rev>{caret}1'). As a special rule,
139 '<rev>{caret}0' means the commit itself and is used when '<rev>' is the
140 object name of a tag object that refers to a commit object.
141
142 '<rev>{tilde}<n>', e.g. 'master{tilde}3'::
143 A suffix '{tilde}<n>' to a revision parameter means the commit
144 object that is the <n>th generation ancestor of the named
145 commit object, following only the first parents. I.e. '<rev>{tilde}3' is
146 equivalent to '<rev>{caret}{caret}{caret}' which is equivalent to
147 '<rev>{caret}1{caret}1{caret}1'. See below for an illustration of
148 the usage of this form.
149
150 '<rev>{caret}{<type>}', e.g. 'v0.99.8{caret}\{commit\}'::
151 A suffix '{caret}' followed by an object type name enclosed in
152 brace pair means dereference the object at '<rev>' recursively until
153 an object of type '<type>' is found or the object cannot be
154 dereferenced anymore (in which case, barf).
155 For example, if '<rev>' is a commit-ish, '<rev>{caret}\{commit\}'
156 describes the corresponding commit object.
157 Similarly, if '<rev>' is a tree-ish, '<rev>{caret}\{tree\}'
158 describes the corresponding tree object.
159 '<rev>{caret}0'
160 is a short-hand for '<rev>{caret}\{commit\}'.
161 +
162 '<rev>{caret}\{object\}' can be used to make sure '<rev>' names an
163 object that exists, without requiring '<rev>' to be a tag, and
164 without dereferencing '<rev>'; because a tag is already an object,
165 it does not have to be dereferenced even once to get to an object.
166 +
167 '<rev>{caret}\{tag\}' can be used to ensure that '<rev>' identifies an
168 existing tag object.
169
170 '<rev>{caret}{}', e.g. 'v0.99.8{caret}{}'::
171 A suffix '{caret}' followed by an empty brace pair
172 means the object could be a tag,
173 and dereference the tag recursively until a non-tag object is
174 found.
175
176 '<rev>{caret}{/<text>}', e.g. 'HEAD^{/fix nasty bug}'::
177 A suffix '{caret}' to a revision parameter, followed by a brace
178 pair that contains a text led by a slash,
179 is the same as the ':/fix nasty bug' syntax below except that
180 it returns the youngest matching commit which is reachable from
181 the '<rev>' before '{caret}'.
182
183 ':/<text>', e.g. ':/fix nasty bug'::
184 A colon, followed by a slash, followed by a text, names
185 a commit whose commit message matches the specified regular expression.
186 This name returns the youngest matching commit which is
187 reachable from any ref. The regular expression can match any part of the
188 commit message. To match messages starting with a string, one can use
189 e.g. ':/^foo'. The special sequence ':/!' is reserved for modifiers to what
190 is matched. ':/!-foo' performs a negative match, while ':/!!foo' matches a
191 literal '!' character, followed by 'foo'. Any other sequence beginning with
192 ':/!' is reserved for now.
193 Depending on the given text, the shell's word splitting rules might
194 require additional quoting.
195
196 '<rev>:<path>', e.g. 'HEAD:README', ':README', 'master:./README'::
197 A suffix ':' followed by a path names the blob or tree
198 at the given path in the tree-ish object named by the part
199 before the colon.
200 ':path' (with an empty part before the colon)
201 is a special case of the syntax described next: content
202 recorded in the index at the given path.
203 A path starting with './' or '../' is relative to the current working directory.
204 The given path will be converted to be relative to the working tree's root directory.
205 This is most useful to address a blob or tree from a commit or tree that has
206 the same tree structure as the working tree.
207
208 ':<n>:<path>', e.g. ':0:README', ':README'::
209 A colon, optionally followed by a stage number (0 to 3) and a
210 colon, followed by a path, names a blob object in the
211 index at the given path. A missing stage number (and the colon
212 that follows it) names a stage 0 entry. During a merge, stage
213 1 is the common ancestor, stage 2 is the target branch's version
214 (typically the current branch), and stage 3 is the version from
215 the branch which is being merged.
216
217 Here is an illustration, by Jon Loeliger. Both commit nodes B
218 and C are parents of commit node A. Parent commits are ordered
219 left-to-right.
220
221 ........................................
222 G H I J
223 \ / \ /
224 D E F
225 \ | / \
226 \ | / |
227 \|/ |
228 B C
229 \ /
230 \ /
231 A
232 ........................................
233
234 A = = A^0
235 B = A^ = A^1 = A~1
236 C = A^2 = A^2
237 D = A^^ = A^1^1 = A~2
238 E = B^2 = A^^2
239 F = B^3 = A^^3
240 G = A^^^ = A^1^1^1 = A~3
241 H = D^2 = B^^2 = A^^^2 = A~2^2
242 I = F^ = B^3^ = A^^3^
243 J = F^2 = B^3^2 = A^^3^2
244
245
246 SPECIFYING RANGES
247 -----------------
248
249 History traversing commands such as `git log` operate on a set
250 of commits, not just a single commit.
251
252 For these commands,
253 specifying a single revision, using the notation described in the
254 previous section, means the set of commits `reachable` from the given
255 commit.
256
257 A commit's reachable set is the commit itself and the commits in
258 its ancestry chain.
259
260
261 Commit Exclusions
262 ~~~~~~~~~~~~~~~~~
263
264 '{caret}<rev>' (caret) Notation::
265 To exclude commits reachable from a commit, a prefix '{caret}'
266 notation is used. E.g. '{caret}r1 r2' means commits reachable
267 from 'r2' but exclude the ones reachable from 'r1' (i.e. 'r1' and
268 its ancestors).
269
270 Dotted Range Notations
271 ~~~~~~~~~~~~~~~~~~~~~~
272
273 The '..' (two-dot) Range Notation::
274 The '{caret}r1 r2' set operation appears so often that there is a shorthand
275 for it. When you have two commits 'r1' and 'r2' (named according
276 to the syntax explained in SPECIFYING REVISIONS above), you can ask
277 for commits that are reachable from r2 excluding those that are reachable
278 from r1 by '{caret}r1 r2' and it can be written as 'r1..r2'.
279
280 The '...' (three-dot) Symmetric Difference Notation::
281 A similar notation 'r1\...r2' is called symmetric difference
282 of 'r1' and 'r2' and is defined as
283 'r1 r2 --not $(git merge-base --all r1 r2)'.
284 It is the set of commits that are reachable from either one of
285 'r1' (left side) or 'r2' (right side) but not from both.
286
287 In these two shorthand notations, you can omit one end and let it default to HEAD.
288 For example, 'origin..' is a shorthand for 'origin..HEAD' and asks "What
289 did I do since I forked from the origin branch?" Similarly, '..origin'
290 is a shorthand for 'HEAD..origin' and asks "What did the origin do since
291 I forked from them?" Note that '..' would mean 'HEAD..HEAD' which is an
292 empty range that is both reachable and unreachable from HEAD.
293
294 Other <rev>{caret} Parent Shorthand Notations
295 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
296 Three other shorthands exist, particularly useful for merge commits,
297 for naming a set that is formed by a commit and its parent commits.
298
299 The 'r1{caret}@' notation means all parents of 'r1'.
300
301 The 'r1{caret}!' notation includes commit 'r1' but excludes all of its parents.
302 By itself, this notation denotes the single commit 'r1'.
303
304 The '<rev>{caret}-[<n>]' notation includes '<rev>' but excludes the <n>th
305 parent (i.e. a shorthand for '<rev>{caret}<n>..<rev>'), with '<n>' = 1 if
306 not given. This is typically useful for merge commits where you
307 can just pass '<commit>{caret}-' to get all the commits in the branch
308 that was merged in merge commit '<commit>' (including '<commit>'
309 itself).
310
311 While '<rev>{caret}<n>' was about specifying a single commit parent, these
312 three notations also consider its parents. For example you can say
313 'HEAD{caret}2{caret}@', however you cannot say 'HEAD{caret}@{caret}2'.
314
315 Revision Range Summary
316 ----------------------
317
318 '<rev>'::
319 Include commits that are reachable from <rev> (i.e. <rev> and its
320 ancestors).
321
322 '{caret}<rev>'::
323 Exclude commits that are reachable from <rev> (i.e. <rev> and its
324 ancestors).
325
326 '<rev1>..<rev2>'::
327 Include commits that are reachable from <rev2> but exclude
328 those that are reachable from <rev1>. When either <rev1> or
329 <rev2> is omitted, it defaults to `HEAD`.
330
331 '<rev1>\...<rev2>'::
332 Include commits that are reachable from either <rev1> or
333 <rev2> but exclude those that are reachable from both. When
334 either <rev1> or <rev2> is omitted, it defaults to `HEAD`.
335
336 '<rev>{caret}@', e.g. 'HEAD{caret}@'::
337 A suffix '{caret}' followed by an at sign is the same as listing
338 all parents of '<rev>' (meaning, include anything reachable from
339 its parents, but not the commit itself).
340
341 '<rev>{caret}!', e.g. 'HEAD{caret}!'::
342 A suffix '{caret}' followed by an exclamation mark is the same
343 as giving commit '<rev>' and then all its parents prefixed with
344 '{caret}' to exclude them (and their ancestors).
345
346 '<rev>{caret}-<n>', e.g. 'HEAD{caret}-, HEAD{caret}-2'::
347 Equivalent to '<rev>{caret}<n>..<rev>', with '<n>' = 1 if not
348 given.
349
350 Here are a handful of examples using the Loeliger illustration above,
351 with each step in the notation's expansion and selection carefully
352 spelt out:
353
354 ....
355 Args Expanded arguments Selected commits
356 D G H D
357 D F G H I J D F
358 ^G D H D
359 ^D B E I J F B
360 ^D B C E I J F B C
361 C I J F C
362 B..C = ^B C C
363 B...C = B ^F C G H D E B C
364 B^- = B^..B
365 = ^B^1 B E I J F B
366 C^@ = C^1
367 = F I J F
368 B^@ = B^1 B^2 B^3
369 = D E F D G H E F I J
370 C^! = C ^C^@
371 = C ^C^1
372 = C ^F C
373 B^! = B ^B^@
374 = B ^B^1 ^B^2 ^B^3
375 = B ^D ^E ^F B
376 F^! D = F ^I ^J D G H D F
377 ....