1GITTUTORIAL(7) Git Manual GITTUTORIAL(7)
2
3
4
6 gittutorial - A tutorial introduction to Git (for version 1.5.1 or
7 newer)
8
10 git *
11
12
14 This tutorial explains how to import a new project into Git, make
15 changes to it, and share changes with other developers.
16
17 If you are instead primarily interested in using Git to fetch a
18 project, for example, to test the latest version, you may prefer to
19 start with the first two chapters of The Git User’s Manual[1].
20
21 First, note that you can get documentation for a command such as git
22 log --graph with:
23
24 $ man git-log
25
26
27 or:
28
29 $ git help log
30
31
32 With the latter, you can use the manual viewer of your choice; see git-
33 help(1) for more information.
34
35 It is a good idea to introduce yourself to Git with your name and
36 public email address before doing any operation. The easiest way to do
37 so is:
38
39 $ git config --global user.name "Your Name Comes Here"
40 $ git config --global user.email you@yourdomain.example.com
41
42
44 Assume you have a tarball project.tar.gz with your initial work. You
45 can place it under Git revision control as follows.
46
47 $ tar xzf project.tar.gz
48 $ cd project
49 $ git init
50
51
52 Git will reply
53
54 Initialized empty Git repository in .git/
55
56
57 You’ve now initialized the working directory—you may notice a new
58 directory created, named ".git".
59
60 Next, tell Git to take a snapshot of the contents of all files under
61 the current directory (note the .), with git add:
62
63 $ git add .
64
65
66 This snapshot is now stored in a temporary staging area which Git calls
67 the "index". You can permanently store the contents of the index in the
68 repository with git commit:
69
70 $ git commit
71
72
73 This will prompt you for a commit message. You’ve now stored the first
74 version of your project in Git.
75
77 Modify some files, then add their updated contents to the index:
78
79 $ git add file1 file2 file3
80
81
82 You are now ready to commit. You can see what is about to be committed
83 using git diff with the --cached option:
84
85 $ git diff --cached
86
87
88 (Without --cached, git diff will show you any changes that you’ve made
89 but not yet added to the index.) You can also get a brief summary of
90 the situation with git status:
91
92 $ git status
93 # On branch master
94 # Changes to be committed:
95 # (use "git reset HEAD <file>..." to unstage)
96 #
97 # modified: file1
98 # modified: file2
99 # modified: file3
100 #
101
102
103 If you need to make any further adjustments, do so now, and then add
104 any newly modified content to the index. Finally, commit your changes
105 with:
106
107 $ git commit
108
109
110 This will again prompt you for a message describing the change, and
111 then record a new version of the project.
112
113 Alternatively, instead of running git add beforehand, you can use
114
115 $ git commit -a
116
117
118 which will automatically notice any modified (but not new) files, add
119 them to the index, and commit, all in one step.
120
121 A note on commit messages: Though not required, it’s a good idea to
122 begin the commit message with a single short (less than 50 character)
123 line summarizing the change, followed by a blank line and then a more
124 thorough description. The text up to the first blank line in a commit
125 message is treated as the commit title, and that title is used
126 throughout Git. For example, git-format-patch(1) turns a commit into
127 email, and it uses the title on the Subject line and the rest of the
128 commit in the body.
129
131 Many revision control systems provide an add command that tells the
132 system to start tracking changes to a new file. Git’s add command does
133 something simpler and more powerful: git add is used both for new and
134 newly modified files, and in both cases it takes a snapshot of the
135 given files and stages that content in the index, ready for inclusion
136 in the next commit.
137
139 At any point you can view the history of your changes using
140
141 $ git log
142
143
144 If you also want to see complete diffs at each step, use
145
146 $ git log -p
147
148
149 Often the overview of the change is useful to get a feel of each step
150
151 $ git log --stat --summary
152
153
155 A single Git repository can maintain multiple branches of development.
156 To create a new branch named "experimental", use
157
158 $ git branch experimental
159
160
161 If you now run
162
163 $ git branch
164
165
166 you’ll get a list of all existing branches:
167
168 experimental
169 * master
170
171
172 The "experimental" branch is the one you just created, and the "master"
173 branch is a default branch that was created for you automatically. The
174 asterisk marks the branch you are currently on; type
175
176 $ git checkout experimental
177
178
179 to switch to the experimental branch. Now edit a file, commit the
180 change, and switch back to the master branch:
181
182 (edit file)
183 $ git commit -a
184 $ git checkout master
185
186
187 Check that the change you made is no longer visible, since it was made
188 on the experimental branch and you’re back on the master branch.
189
190 You can make a different change on the master branch:
191
192 (edit file)
193 $ git commit -a
194
195
196 at this point the two branches have diverged, with different changes
197 made in each. To merge the changes made in experimental into master,
198 run
199
200 $ git merge experimental
201
202
203 If the changes don’t conflict, you’re done. If there are conflicts,
204 markers will be left in the problematic files showing the conflict;
205
206 $ git diff
207
208
209 will show this. Once you’ve edited the files to resolve the conflicts,
210
211 $ git commit -a
212
213
214 will commit the result of the merge. Finally,
215
216 $ gitk
217
218
219 will show a nice graphical representation of the resulting history.
220
221 At this point you could delete the experimental branch with
222
223 $ git branch -d experimental
224
225
226 This command ensures that the changes in the experimental branch are
227 already in the current branch.
228
229 If you develop on a branch crazy-idea, then regret it, you can always
230 delete the branch with
231
232 $ git branch -D crazy-idea
233
234
235 Branches are cheap and easy, so this is a good way to try something
236 out.
237
239 Suppose that Alice has started a new project with a Git repository in
240 /home/alice/project, and that Bob, who has a home directory on the same
241 machine, wants to contribute.
242
243 Bob begins with:
244
245 bob$ git clone /home/alice/project myrepo
246
247
248 This creates a new directory "myrepo" containing a clone of Alice’s
249 repository. The clone is on an equal footing with the original project,
250 possessing its own copy of the original project’s history.
251
252 Bob then makes some changes and commits them:
253
254 (edit files)
255 bob$ git commit -a
256 (repeat as necessary)
257
258
259 When he’s ready, he tells Alice to pull changes from the repository at
260 /home/bob/myrepo. She does this with:
261
262 alice$ cd /home/alice/project
263 alice$ git pull /home/bob/myrepo master
264
265
266 This merges the changes from Bob’s "master" branch into Alice’s current
267 branch. If Alice has made her own changes in the meantime, then she may
268 need to manually fix any conflicts.
269
270 The "pull" command thus performs two operations: it fetches changes
271 from a remote branch, then merges them into the current branch.
272
273 Note that in general, Alice would want her local changes committed
274 before initiating this "pull". If Bob’s work conflicts with what Alice
275 did since their histories forked, Alice will use her working tree and
276 the index to resolve conflicts, and existing local changes will
277 interfere with the conflict resolution process (Git will still perform
278 the fetch but will refuse to merge --- Alice will have to get rid of
279 her local changes in some way and pull again when this happens).
280
281 Alice can peek at what Bob did without merging first, using the "fetch"
282 command; this allows Alice to inspect what Bob did, using a special
283 symbol "FETCH_HEAD", in order to determine if he has anything worth
284 pulling, like this:
285
286 alice$ git fetch /home/bob/myrepo master
287 alice$ git log -p HEAD..FETCH_HEAD
288
289
290 This operation is safe even if Alice has uncommitted local changes. The
291 range notation "HEAD..FETCH_HEAD" means "show everything that is
292 reachable from the FETCH_HEAD but exclude anything that is reachable
293 from HEAD". Alice already knows everything that leads to her current
294 state (HEAD), and reviews what Bob has in his state (FETCH_HEAD) that
295 she has not seen with this command.
296
297 If Alice wants to visualize what Bob did since their histories forked
298 she can issue the following command:
299
300 $ gitk HEAD..FETCH_HEAD
301
302
303 This uses the same two-dot range notation we saw earlier with git log.
304
305 Alice may want to view what both of them did since they forked. She can
306 use three-dot form instead of the two-dot form:
307
308 $ gitk HEAD...FETCH_HEAD
309
310
311 This means "show everything that is reachable from either one, but
312 exclude anything that is reachable from both of them".
313
314 Please note that these range notation can be used with both gitk and
315 "git log".
316
317 After inspecting what Bob did, if there is nothing urgent, Alice may
318 decide to continue working without pulling from Bob. If Bob’s history
319 does have something Alice would immediately need, Alice may choose to
320 stash her work-in-progress first, do a "pull", and then finally unstash
321 her work-in-progress on top of the resulting history.
322
323 When you are working in a small closely knit group, it is not unusual
324 to interact with the same repository over and over again. By defining
325 remote repository shorthand, you can make it easier:
326
327 alice$ git remote add bob /home/bob/myrepo
328
329
330 With this, Alice can perform the first part of the "pull" operation
331 alone using the git fetch command without merging them with her own
332 branch, using:
333
334 alice$ git fetch bob
335
336
337 Unlike the longhand form, when Alice fetches from Bob using a remote
338 repository shorthand set up with git remote, what was fetched is stored
339 in a remote-tracking branch, in this case bob/master. So after this:
340
341 alice$ git log -p master..bob/master
342
343
344 shows a list of all the changes that Bob made since he branched from
345 Alice’s master branch.
346
347 After examining those changes, Alice could merge the changes into her
348 master branch:
349
350 alice$ git merge bob/master
351
352
353 This merge can also be done by pulling from her own remote-tracking
354 branch, like this:
355
356 alice$ git pull . remotes/bob/master
357
358
359 Note that git pull always merges into the current branch, regardless of
360 what else is given on the command line.
361
362 Later, Bob can update his repo with Alice’s latest changes using
363
364 bob$ git pull
365
366
367 Note that he doesn’t need to give the path to Alice’s repository; when
368 Bob cloned Alice’s repository, Git stored the location of her
369 repository in the repository configuration, and that location is used
370 for pulls:
371
372 bob$ git config --get remote.origin.url
373 /home/alice/project
374
375
376 (The complete configuration created by git clone is visible using git
377 config -l, and the git-config(1) man page explains the meaning of each
378 option.)
379
380 Git also keeps a pristine copy of Alice’s master branch under the name
381 "origin/master":
382
383 bob$ git branch -r
384 origin/master
385
386
387 If Bob later decides to work from a different host, he can still
388 perform clones and pulls using the ssh protocol:
389
390 bob$ git clone alice.org:/home/alice/project myrepo
391
392
393 Alternatively, Git has a native protocol, or can use rsync or http; see
394 git-pull(1) for details.
395
396 Git can also be used in a CVS-like mode, with a central repository that
397 various users push changes to; see git-push(1) and gitcvs-migration(7).
398
400 Git history is represented as a series of interrelated commits. We have
401 already seen that the git log command can list those commits. Note that
402 first line of each git log entry also gives a name for the commit:
403
404 $ git log
405 commit c82a22c39cbc32576f64f5c6b3f24b99ea8149c7
406 Author: Junio C Hamano <junkio@cox.net>
407 Date: Tue May 16 17:18:22 2006 -0700
408
409 merge-base: Clarify the comments on post processing.
410
411
412 We can give this name to git show to see the details about this commit.
413
414 $ git show c82a22c39cbc32576f64f5c6b3f24b99ea8149c7
415
416
417 But there are other ways to refer to commits. You can use any initial
418 part of the name that is long enough to uniquely identify the commit:
419
420 $ git show c82a22c39c # the first few characters of the name are
421 # usually enough
422 $ git show HEAD # the tip of the current branch
423 $ git show experimental # the tip of the "experimental" branch
424
425
426 Every commit usually has one "parent" commit which points to the
427 previous state of the project:
428
429 $ git show HEAD^ # to see the parent of HEAD
430 $ git show HEAD^^ # to see the grandparent of HEAD
431 $ git show HEAD~4 # to see the great-great grandparent of HEAD
432
433
434 Note that merge commits may have more than one parent:
435
436 $ git show HEAD^1 # show the first parent of HEAD (same as HEAD^)
437 $ git show HEAD^2 # show the second parent of HEAD
438
439
440 You can also give commits names of your own; after running
441
442 $ git tag v2.5 1b2e1d63ff
443
444
445 you can refer to 1b2e1d63ff by the name "v2.5". If you intend to share
446 this name with other people (for example, to identify a release
447 version), you should create a "tag" object, and perhaps sign it; see
448 git-tag(1) for details.
449
450 Any Git command that needs to know a commit can take any of these
451 names. For example:
452
453 $ git diff v2.5 HEAD # compare the current HEAD to v2.5
454 $ git branch stable v2.5 # start a new branch named "stable" based
455 # at v2.5
456 $ git reset --hard HEAD^ # reset your current branch and working
457 # directory to its state at HEAD^
458
459
460 Be careful with that last command: in addition to losing any changes in
461 the working directory, it will also remove all later commits from this
462 branch. If this branch is the only branch containing those commits,
463 they will be lost. Also, don’t use git reset on a publicly-visible
464 branch that other developers pull from, as it will force needless
465 merges on other developers to clean up the history. If you need to undo
466 changes that you have pushed, use git revert instead.
467
468 The git grep command can search for strings in any version of your
469 project, so
470
471 $ git grep "hello" v2.5
472
473
474 searches for all occurrences of "hello" in v2.5.
475
476 If you leave out the commit name, git grep will search any of the files
477 it manages in your current directory. So
478
479 $ git grep "hello"
480
481
482 is a quick way to search just the files that are tracked by Git.
483
484 Many Git commands also take sets of commits, which can be specified in
485 a number of ways. Here are some examples with git log:
486
487 $ git log v2.5..v2.6 # commits between v2.5 and v2.6
488 $ git log v2.5.. # commits since v2.5
489 $ git log --since="2 weeks ago" # commits from the last 2 weeks
490 $ git log v2.5.. Makefile # commits since v2.5 which modify
491 # Makefile
492
493
494 You can also give git log a "range" of commits where the first is not
495 necessarily an ancestor of the second; for example, if the tips of the
496 branches "stable" and "master" diverged from a common commit some time
497 ago, then
498
499 $ git log stable..master
500
501
502 will list commits made in the master branch but not in the stable
503 branch, while
504
505 $ git log master..stable
506
507
508 will show the list of commits made on the stable branch but not the
509 master branch.
510
511 The git log command has a weakness: it must present commits in a list.
512 When the history has lines of development that diverged and then merged
513 back together, the order in which git log presents those commits is
514 meaningless.
515
516 Most projects with multiple contributors (such as the Linux kernel, or
517 Git itself) have frequent merges, and gitk does a better job of
518 visualizing their history. For example,
519
520 $ gitk --since="2 weeks ago" drivers/
521
522
523 allows you to browse any commits from the last 2 weeks of commits that
524 modified files under the "drivers" directory. (Note: you can adjust
525 gitk’s fonts by holding down the control key while pressing "-" or
526 "+".)
527
528 Finally, most commands that take filenames will optionally allow you to
529 precede any filename by a commit, to specify a particular version of
530 the file:
531
532 $ git diff v2.5:Makefile HEAD:Makefile.in
533
534
535 You can also use git show to see any such file:
536
537 $ git show v2.5:Makefile
538
539
541 This tutorial should be enough to perform basic distributed revision
542 control for your projects. However, to fully understand the depth and
543 power of Git you need to understand two simple ideas on which it is
544 based:
545
546 · The object database is the rather elegant system used to store the
547 history of your project—files, directories, and commits.
548
549 · The index file is a cache of the state of a directory tree, used to
550 create commits, check out working directories, and hold the various
551 trees involved in a merge.
552
553 Part two of this tutorial explains the object database, the index file,
554 and a few other odds and ends that you’ll need to make the most of Git.
555 You can find it at gittutorial-2(7).
556
557 If you don’t want to continue with that right away, a few other
558 digressions that may be interesting at this point are:
559
560 · git-format-patch(1), git-am(1): These convert series of git commits
561 into emailed patches, and vice versa, useful for projects such as
562 the Linux kernel which rely heavily on emailed patches.
563
564 · git-bisect(1): When there is a regression in your project, one way
565 to track down the bug is by searching through the history to find
566 the exact commit that’s to blame. Git bisect can help you perform a
567 binary search for that commit. It is smart enough to perform a
568 close-to-optimal search even in the case of complex non-linear
569 history with lots of merged branches.
570
571 · gitworkflows(7): Gives an overview of recommended workflows.
572
573 · Everyday Git with 20 Commands Or So[2]
574
575 · gitcvs-migration(7): Git for CVS users.
576
578 gittutorial-2(7), gitcvs-migration(7), gitcore-tutorial(7),
579 gitglossary(7), git-help(1), gitworkflows(7), Everyday Git[2], The Git
580 User’s Manual[1]
581
583 Part of the git(1) suite.
584
586 1. The Git User’s Manual
587 file:///usr/share/doc/git-1.8.3.1/user-manual.html
588
589 2. Everyday Git with 20 Commands Or So
590 file:///usr/share/doc/git-1.8.3.1/everyday.html
591
592
593
594Git 1.8.3.1 11/19/2018 GITTUTORIAL(7)