Documentation / gittutorial.txton commit Teach fsck and prune about the new location of temporary objects (0e8316c)
   1gittutorial(7)
   2==============
   3
   4NAME
   5----
   6gittutorial - A tutorial introduction to git (for version 1.5.1 or newer)
   7
   8SYNOPSIS
   9--------
  10git *
  11
  12DESCRIPTION
  13-----------
  14
  15This tutorial explains how to import a new project into git, make
  16changes to it, and share changes with other developers.
  17
  18If you are instead primarily interested in using git to fetch a project,
  19for example, to test the latest version, you may prefer to start with
  20the first two chapters of link:user-manual.html[The Git User's Manual].
  21
  22First, note that you can get documentation for a command such as
  23`git log --graph` with:
  24
  25------------------------------------------------
  26$ man git-log
  27------------------------------------------------
  28
  29It is a good idea to introduce yourself to git with your name and
  30public email address before doing any operation.  The easiest
  31way to do so is:
  32
  33------------------------------------------------
  34$ git config --global user.name "Your Name Comes Here"
  35$ git config --global user.email you@yourdomain.example.com
  36------------------------------------------------
  37
  38
  39Importing a new project
  40-----------------------
  41
  42Assume you have a tarball project.tar.gz with your initial work.  You
  43can place it under git revision control as follows.
  44
  45------------------------------------------------
  46$ tar xzf project.tar.gz
  47$ cd project
  48$ git init
  49------------------------------------------------
  50
  51Git will reply
  52
  53------------------------------------------------
  54Initialized empty Git repository in .git/
  55------------------------------------------------
  56
  57You've now initialized the working directory--you may notice a new
  58directory created, named ".git".
  59
  60Next, tell git to take a snapshot of the contents of all files under the
  61current directory (note the '.'), with 'git-add':
  62
  63------------------------------------------------
  64$ git add .
  65------------------------------------------------
  66
  67This snapshot is now stored in a temporary staging area which git calls
  68the "index".  You can permanently store the contents of the index in the
  69repository with 'git-commit':
  70
  71------------------------------------------------
  72$ git commit
  73------------------------------------------------
  74
  75This will prompt you for a commit message.  You've now stored the first
  76version of your project in git.
  77
  78Making changes
  79--------------
  80
  81Modify some files, then add their updated contents to the index:
  82
  83------------------------------------------------
  84$ git add file1 file2 file3
  85------------------------------------------------
  86
  87You are now ready to commit.  You can see what is about to be committed
  88using 'git-diff' with the --cached option:
  89
  90------------------------------------------------
  91$ git diff --cached
  92------------------------------------------------
  93
  94(Without --cached, 'git-diff' will show you any changes that
  95you've made but not yet added to the index.)  You can also get a brief
  96summary of the situation with 'git-status':
  97
  98------------------------------------------------
  99$ git status
 100# On branch master
 101# Changes to be committed:
 102#   (use "git reset HEAD <file>..." to unstage)
 103#
 104#       modified:   file1
 105#       modified:   file2
 106#       modified:   file3
 107#
 108------------------------------------------------
 109
 110If you need to make any further adjustments, do so now, and then add any
 111newly modified content to the index.  Finally, commit your changes with:
 112
 113------------------------------------------------
 114$ git commit
 115------------------------------------------------
 116
 117This will again prompt you for a message describing the change, and then
 118record a new version of the project.
 119
 120Alternatively, instead of running 'git-add' beforehand, you can use
 121
 122------------------------------------------------
 123$ git commit -a
 124------------------------------------------------
 125
 126which will automatically notice any modified (but not new) files, add
 127them to the index, and commit, all in one step.
 128
 129A note on commit messages: Though not required, it's a good idea to
 130begin the commit message with a single short (less than 50 character)
 131line summarizing the change, followed by a blank line and then a more
 132thorough description.  Tools that turn commits into email, for
 133example, use the first line on the Subject: line and the rest of the
 134commit in the body.
 135
 136Git tracks content not files
 137----------------------------
 138
 139Many revision control systems provide an `add` command that tells the
 140system to start tracking changes to a new file.  Git's `add` command
 141does something simpler and more powerful: 'git-add' is used both for new
 142and newly modified files, and in both cases it takes a snapshot of the
 143given files and stages that content in the index, ready for inclusion in
 144the next commit.
 145
 146Viewing project history
 147-----------------------
 148
 149At any point you can view the history of your changes using
 150
 151------------------------------------------------
 152$ git log
 153------------------------------------------------
 154
 155If you also want to see complete diffs at each step, use
 156
 157------------------------------------------------
 158$ git log -p
 159------------------------------------------------
 160
 161Often the overview of the change is useful to get a feel of
 162each step
 163
 164------------------------------------------------
 165$ git log --stat --summary
 166------------------------------------------------
 167
 168Managing branches
 169-----------------
 170
 171A single git repository can maintain multiple branches of
 172development.  To create a new branch named "experimental", use
 173
 174------------------------------------------------
 175$ git branch experimental
 176------------------------------------------------
 177
 178If you now run
 179
 180------------------------------------------------
 181$ git branch
 182------------------------------------------------
 183
 184you'll get a list of all existing branches:
 185
 186------------------------------------------------
 187  experimental
 188* master
 189------------------------------------------------
 190
 191The "experimental" branch is the one you just created, and the
 192"master" branch is a default branch that was created for you
 193automatically.  The asterisk marks the branch you are currently on;
 194type
 195
 196------------------------------------------------
 197$ git checkout experimental
 198------------------------------------------------
 199
 200to switch to the experimental branch.  Now edit a file, commit the
 201change, and switch back to the master branch:
 202
 203------------------------------------------------
 204(edit file)
 205$ git commit -a
 206$ git checkout master
 207------------------------------------------------
 208
 209Check that the change you made is no longer visible, since it was
 210made on the experimental branch and you're back on the master branch.
 211
 212You can make a different change on the master branch:
 213
 214------------------------------------------------
 215(edit file)
 216$ git commit -a
 217------------------------------------------------
 218
 219at this point the two branches have diverged, with different changes
 220made in each.  To merge the changes made in experimental into master, run
 221
 222------------------------------------------------
 223$ git merge experimental
 224------------------------------------------------
 225
 226If the changes don't conflict, you're done.  If there are conflicts,
 227markers will be left in the problematic files showing the conflict;
 228
 229------------------------------------------------
 230$ git diff
 231------------------------------------------------
 232
 233will show this.  Once you've edited the files to resolve the
 234conflicts,
 235
 236------------------------------------------------
 237$ git commit -a
 238------------------------------------------------
 239
 240will commit the result of the merge. Finally,
 241
 242------------------------------------------------
 243$ gitk
 244------------------------------------------------
 245
 246will show a nice graphical representation of the resulting history.
 247
 248At this point you could delete the experimental branch with
 249
 250------------------------------------------------
 251$ git branch -d experimental
 252------------------------------------------------
 253
 254This command ensures that the changes in the experimental branch are
 255already in the current branch.
 256
 257If you develop on a branch crazy-idea, then regret it, you can always
 258delete the branch with
 259
 260-------------------------------------
 261$ git branch -D crazy-idea
 262-------------------------------------
 263
 264Branches are cheap and easy, so this is a good way to try something
 265out.
 266
 267Using git for collaboration
 268---------------------------
 269
 270Suppose that Alice has started a new project with a git repository in
 271/home/alice/project, and that Bob, who has a home directory on the
 272same machine, wants to contribute.
 273
 274Bob begins with:
 275
 276------------------------------------------------
 277bob$ git clone /home/alice/project myrepo
 278------------------------------------------------
 279
 280This creates a new directory "myrepo" containing a clone of Alice's
 281repository.  The clone is on an equal footing with the original
 282project, possessing its own copy of the original project's history.
 283
 284Bob then makes some changes and commits them:
 285
 286------------------------------------------------
 287(edit files)
 288bob$ git commit -a
 289(repeat as necessary)
 290------------------------------------------------
 291
 292When he's ready, he tells Alice to pull changes from the repository
 293at /home/bob/myrepo.  She does this with:
 294
 295------------------------------------------------
 296alice$ cd /home/alice/project
 297alice$ git pull /home/bob/myrepo master
 298------------------------------------------------
 299
 300This merges the changes from Bob's "master" branch into Alice's
 301current branch.  If Alice has made her own changes in the meantime,
 302then she may need to manually fix any conflicts.  (Note that the
 303"master" argument in the above command is actually unnecessary, as it
 304is the default.)
 305
 306The "pull" command thus performs two operations: it fetches changes
 307from a remote branch, then merges them into the current branch.
 308
 309Note that in general, Alice would want her local changes committed before
 310initiating this "pull".  If Bob's work conflicts with what Alice did since
 311their histories forked, Alice will use her working tree and the index to
 312resolve conflicts, and existing local changes will interfere with the
 313conflict resolution process (git will still perform the fetch but will
 314refuse to merge --- Alice will have to get rid of her local changes in
 315some way and pull again when this happens).
 316
 317Alice can peek at what Bob did without merging first, using the "fetch"
 318command; this allows Alice to inspect what Bob did, using a special
 319symbol "FETCH_HEAD", in order to determine if he has anything worth
 320pulling, like this:
 321
 322------------------------------------------------
 323alice$ git fetch /home/bob/myrepo master
 324alice$ git log -p ..FETCH_HEAD
 325------------------------------------------------
 326
 327This operation is safe even if Alice has uncommitted local changes.
 328
 329After inspecting what Bob did, if there is nothing urgent, Alice may
 330decide to continue working without pulling from Bob.  If Bob's history
 331does have something Alice would immediately need, Alice may choose to
 332stash her work-in-progress first, do a "pull", and then finally unstash
 333her work-in-progress on top of the resulting history.
 334
 335When you are working in a small closely knit group, it is not
 336unusual to interact with the same repository over and over
 337again.  By defining 'remote' repository shorthand, you can make
 338it easier:
 339
 340------------------------------------------------
 341alice$ git remote add bob /home/bob/myrepo
 342------------------------------------------------
 343
 344With this, Alice can perform the first part of the "pull" operation alone using the
 345'git-fetch' command without merging them with her own branch,
 346using:
 347
 348-------------------------------------
 349alice$ git fetch bob
 350-------------------------------------
 351
 352Unlike the longhand form, when Alice fetches from Bob using a
 353remote repository shorthand set up with 'git-remote', what was
 354fetched is stored in a remote tracking branch, in this case
 355`bob/master`.  So after this:
 356
 357-------------------------------------
 358alice$ git log -p master..bob/master
 359-------------------------------------
 360
 361shows a list of all the changes that Bob made since he branched from
 362Alice's master branch.
 363
 364After examining those changes, Alice
 365could merge the changes into her master branch:
 366
 367-------------------------------------
 368alice$ git merge bob/master
 369-------------------------------------
 370
 371This `merge` can also be done by 'pulling from her own remote
 372tracking branch', like this:
 373
 374-------------------------------------
 375alice$ git pull . remotes/bob/master
 376-------------------------------------
 377
 378Note that git pull always merges into the current branch,
 379regardless of what else is given on the command line.
 380
 381Later, Bob can update his repo with Alice's latest changes using
 382
 383-------------------------------------
 384bob$ git pull
 385-------------------------------------
 386
 387Note that he doesn't need to give the path to Alice's repository;
 388when Bob cloned Alice's repository, git stored the location of her
 389repository in the repository configuration, and that location is
 390used for pulls:
 391
 392-------------------------------------
 393bob$ git config --get remote.origin.url
 394/home/alice/project
 395-------------------------------------
 396
 397(The complete configuration created by 'git-clone' is visible using
 398`git config -l`, and the linkgit:git-config[1] man page
 399explains the meaning of each option.)
 400
 401Git also keeps a pristine copy of Alice's master branch under the
 402name "origin/master":
 403
 404-------------------------------------
 405bob$ git branch -r
 406  origin/master
 407-------------------------------------
 408
 409If Bob later decides to work from a different host, he can still
 410perform clones and pulls using the ssh protocol:
 411
 412-------------------------------------
 413bob$ git clone alice.org:/home/alice/project myrepo
 414-------------------------------------
 415
 416Alternatively, git has a native protocol, or can use rsync or http;
 417see linkgit:git-pull[1] for details.
 418
 419Git can also be used in a CVS-like mode, with a central repository
 420that various users push changes to; see linkgit:git-push[1] and
 421linkgit:gitcvs-migration[7].
 422
 423Exploring history
 424-----------------
 425
 426Git history is represented as a series of interrelated commits.  We
 427have already seen that the 'git-log' command can list those commits.
 428Note that first line of each git log entry also gives a name for the
 429commit:
 430
 431-------------------------------------
 432$ git log
 433commit c82a22c39cbc32576f64f5c6b3f24b99ea8149c7
 434Author: Junio C Hamano <junkio@cox.net>
 435Date:   Tue May 16 17:18:22 2006 -0700
 436
 437    merge-base: Clarify the comments on post processing.
 438-------------------------------------
 439
 440We can give this name to 'git-show' to see the details about this
 441commit.
 442
 443-------------------------------------
 444$ git show c82a22c39cbc32576f64f5c6b3f24b99ea8149c7
 445-------------------------------------
 446
 447But there are other ways to refer to commits.  You can use any initial
 448part of the name that is long enough to uniquely identify the commit:
 449
 450-------------------------------------
 451$ git show c82a22c39c   # the first few characters of the name are
 452                        # usually enough
 453$ git show HEAD         # the tip of the current branch
 454$ git show experimental # the tip of the "experimental" branch
 455-------------------------------------
 456
 457Every commit usually has one "parent" commit
 458which points to the previous state of the project:
 459
 460-------------------------------------
 461$ git show HEAD^  # to see the parent of HEAD
 462$ git show HEAD^^ # to see the grandparent of HEAD
 463$ git show HEAD~4 # to see the great-great grandparent of HEAD
 464-------------------------------------
 465
 466Note that merge commits may have more than one parent:
 467
 468-------------------------------------
 469$ git show HEAD^1 # show the first parent of HEAD (same as HEAD^)
 470$ git show HEAD^2 # show the second parent of HEAD
 471-------------------------------------
 472
 473You can also give commits names of your own; after running
 474
 475-------------------------------------
 476$ git tag v2.5 1b2e1d63ff
 477-------------------------------------
 478
 479you can refer to 1b2e1d63ff by the name "v2.5".  If you intend to
 480share this name with other people (for example, to identify a release
 481version), you should create a "tag" object, and perhaps sign it; see
 482linkgit:git-tag[1] for details.
 483
 484Any git command that needs to know a commit can take any of these
 485names.  For example:
 486
 487-------------------------------------
 488$ git diff v2.5 HEAD     # compare the current HEAD to v2.5
 489$ git branch stable v2.5 # start a new branch named "stable" based
 490                         # at v2.5
 491$ git reset --hard HEAD^ # reset your current branch and working
 492                         # directory to its state at HEAD^
 493-------------------------------------
 494
 495Be careful with that last command: in addition to losing any changes
 496in the working directory, it will also remove all later commits from
 497this branch.  If this branch is the only branch containing those
 498commits, they will be lost.  Also, don't use 'git-reset' on a
 499publicly-visible branch that other developers pull from, as it will
 500force needless merges on other developers to clean up the history.
 501If you need to undo changes that you have pushed, use 'git-revert'
 502instead.
 503
 504The 'git-grep' command can search for strings in any version of your
 505project, so
 506
 507-------------------------------------
 508$ git grep "hello" v2.5
 509-------------------------------------
 510
 511searches for all occurrences of "hello" in v2.5.
 512
 513If you leave out the commit name, 'git-grep' will search any of the
 514files it manages in your current directory.  So
 515
 516-------------------------------------
 517$ git grep "hello"
 518-------------------------------------
 519
 520is a quick way to search just the files that are tracked by git.
 521
 522Many git commands also take sets of commits, which can be specified
 523in a number of ways.  Here are some examples with 'git-log':
 524
 525-------------------------------------
 526$ git log v2.5..v2.6            # commits between v2.5 and v2.6
 527$ git log v2.5..                # commits since v2.5
 528$ git log --since="2 weeks ago" # commits from the last 2 weeks
 529$ git log v2.5.. Makefile       # commits since v2.5 which modify
 530                                # Makefile
 531-------------------------------------
 532
 533You can also give 'git-log' a "range" of commits where the first is not
 534necessarily an ancestor of the second; for example, if the tips of
 535the branches "stable-release" and "master" diverged from a common
 536commit some time ago, then
 537
 538-------------------------------------
 539$ git log stable..experimental
 540-------------------------------------
 541
 542will list commits made in the experimental branch but not in the
 543stable branch, while
 544
 545-------------------------------------
 546$ git log experimental..stable
 547-------------------------------------
 548
 549will show the list of commits made on the stable branch but not
 550the experimental branch.
 551
 552The 'git-log' command has a weakness: it must present commits in a
 553list.  When the history has lines of development that diverged and
 554then merged back together, the order in which 'git-log' presents
 555those commits is meaningless.
 556
 557Most projects with multiple contributors (such as the linux kernel,
 558or git itself) have frequent merges, and 'gitk' does a better job of
 559visualizing their history.  For example,
 560
 561-------------------------------------
 562$ gitk --since="2 weeks ago" drivers/
 563-------------------------------------
 564
 565allows you to browse any commits from the last 2 weeks of commits
 566that modified files under the "drivers" directory.  (Note: you can
 567adjust gitk's fonts by holding down the control key while pressing
 568"-" or "+".)
 569
 570Finally, most commands that take filenames will optionally allow you
 571to precede any filename by a commit, to specify a particular version
 572of the file:
 573
 574-------------------------------------
 575$ git diff v2.5:Makefile HEAD:Makefile.in
 576-------------------------------------
 577
 578You can also use 'git-show' to see any such file:
 579
 580-------------------------------------
 581$ git show v2.5:Makefile
 582-------------------------------------
 583
 584Next Steps
 585----------
 586
 587This tutorial should be enough to perform basic distributed revision
 588control for your projects.  However, to fully understand the depth
 589and power of git you need to understand two simple ideas on which it
 590is based:
 591
 592  * The object database is the rather elegant system used to
 593    store the history of your project--files, directories, and
 594    commits.
 595
 596  * The index file is a cache of the state of a directory tree,
 597    used to create commits, check out working directories, and
 598    hold the various trees involved in a merge.
 599
 600Part two of this tutorial explains the object
 601database, the index file, and a few other odds and ends that you'll
 602need to make the most of git. You can find it at linkgit:gittutorial-2[7].
 603
 604If you don't want to continue with that right away, a few other
 605digressions that may be interesting at this point are:
 606
 607  * linkgit:git-format-patch[1], linkgit:git-am[1]: These convert
 608    series of git commits into emailed patches, and vice versa,
 609    useful for projects such as the linux kernel which rely heavily
 610    on emailed patches.
 611
 612  * linkgit:git-bisect[1]: When there is a regression in your
 613    project, one way to track down the bug is by searching through
 614    the history to find the exact commit that's to blame.  Git bisect
 615    can help you perform a binary search for that commit.  It is
 616    smart enough to perform a close-to-optimal search even in the
 617    case of complex non-linear history with lots of merged branches.
 618
 619  * link:everyday.html[Everyday GIT with 20 Commands Or So]
 620
 621  * linkgit:gitcvs-migration[7]: Git for CVS users.
 622
 623SEE ALSO
 624--------
 625linkgit:gittutorial-2[7],
 626linkgit:gitcvs-migration[7],
 627linkgit:gitcore-tutorial[7],
 628linkgit:gitglossary[7],
 629link:everyday.html[Everyday git],
 630link:user-manual.html[The Git User's Manual]
 631
 632GIT
 633---
 634Part of the linkgit:git[1] suite.