2 A tour of git: the basics
6 This document is a modified version of a document originally titled
7 "Distributed revision control with Mercurial" and originally authored
8 by Bryan O’Sullivan. The original document was obtained from
9 <http://hgbook.red-bean.com/>.
11 Copyright © 2006, 2007 Bryan O’Sullivan.
13 This material may be distributed only subject to the terms and
14 conditions set forth in version 1.0 of the Open Publication
15 License. Please refer to Appendix D for the license text.
17 As this is a modified version, the name of Bryan O'Sullivan is used
18 only to properly credit him with the original text. The appearance of
19 his name here explicitly does not assert or imply his endorsement of
20 this modified document.
22 Portions Copyright © 2007 Carl Worth.
24 Changes made by Carl include the following:
27 * Convert from HTML to markdown source syntax
28 * Eliminate all content except Chapter 2 and Appendix D
29 * Eliminate line numbers from examples
30 * Modified to describe git instead of mercurial
32 The source of this modified version can be obtained via git:
34 git clone git://cworth.org/git/hgbook-git
38 git clone http://cworth.org/git/hgbook-git
40 and can be [browsed online](http://git.cworth.org/git/hgbook-git)
42 ### 2.1 Installing git on your system
44 Prebuilt binary packages of git are available for many popular
45 operating systems. These make it easy to start using git on your
50 Because each Linux distribution has its own packaging tools, policies,
51 and rate of development, it’s difficult to give a comprehensive set of
52 instructions on how to install git binaries. The version of
53 git that you will end up with can vary depending on how active
54 the person is who maintains the package for your distribution.
56 To keep things simple, I will focus on installing git from the
57 command line under the most popular Linux distributions. Most of these
58 distributions provide graphical package managers that will let you
59 install git with a single click. The package name to look for is
60 often git, but is sometimes git-core, (due to an unfortunate name
61 with git, meaning GNU Interactive Tools).
65 apt-get install git-core
81 apt-get install git-core
85 A git-core package is available through
86 [macports](http://macports.org). Once macports is enabled, the command
93 Git has long been available as part of cygwin, and works reasonably
94 well in that environment. Some people find cygwin a particularly
95 inelegant approach to running git and would prefer a "native"
96 solution. To this end, the [msysgit
97 project](http://code.google.com/p/msysgit/) is rapidly putting
98 together a solution including various packages with full
99 installers. These include GitMe, a package to install the entire
100 development environment necessary to work on improving the msysgit
101 port of git, and WinGit, a package for installing just git itself
102 without the development environment, (still in Alpha as of September
105 ### 2.2 Getting started
107 To begin, we’ll use the “git version” command to find out whether git
108 is actually installed properly. Versions 1.5 and newer of git are much
109 more friendly to new users than versions 1.4 and older. If you aren't
110 yet running version 1.5 or newer, it's highly recommended that you
116 #### 2.2.1 Built-in help
118 Git provides a built-in help system. This is invaluable for those
119 times when you find yourself stuck trying to remember how to run a
120 command. If you are completely stuck, simply run “git help”; it will
121 print a brief list of commonly-used commands, along with a description
122 of what each does. If you ask for help on a specific command (such as
123 "git help init"), it prints more detailed information. [XXX: Does "git
124 help <foo>" work universally as a built-in or does it expect man to be
125 present and just call out to "man git-<foo>"?]
127 [XXX: The original hgbook includes the complete output of "hg
128 help init" at this point. I'm not including the corresponding
129 "git help init" output as it would be excessively long. The
130 description alone is quite reasonable, (other than a
131 not-too-helpful aside about the obsolete git-init-db command),
132 but it only comes after a full screen's worth of options
133 details. Might it make sense to have a more summarized help
134 output for "git help <foo>" than all of the documentation
135 available for git-<foo>? And perhaps alos provide a "git -v
136 help" similar to "hg -v help" for more?]
138 ### 2.3 Working with a repository
140 In git, everything happens inside a repository. The repository
141 for a project contains all of the files that “belong to” that project,
142 along with a historical record of the project’s files.
144 There’s nothing particularly magical about a repository; it is simply
145 a directory tree in your filesystem that git treats as
146 special. You can rename or delete a repository any time you like,
147 using either the command line or your file browser.
149 #### 2.3.1 Making a local copy of a repository
151 Copying a repository is just a little bit special. While you could use
152 a normal file copying command to make a copy of a repository, it’s
153 best to use a built-in command that git provides. This command
154 is called “git clone”, because it creates an identical copy of an
157 $ git clone git://cworth.org/git/hello
158 Initialized empty Git repository in /tmp/hello/.git/
159 remote: Generating pack...
160 remote: Done counting 15 objects.
161 remote: Deltifying 15 objects...
162 remote: 100% (15/15) done
163 remote: Total 15 (delta 2), reused 15 (delta remote: 2)
164 Indexing 15 objects...
166 Resolving 2 deltas...
169 If for some reason you are prevented from talking on the git: port,
170 then there is also the capability to clone a repository (less
171 efficiently) over http:
173 $ git clone http://cworth.org/git/hello
174 Initialized empty Git repository in /tmp/hello/.git/
175 Getting alternates list for http://cworth.org/git/hello
176 Getting pack list for http://cworth.org/git/hello
177 Getting index for pack 04ecb061314ecbd60fa0610ecf55a1cbf85ea294
178 Getting pack 04ecb061314ecbd60fa0610ecf55a1cbf85ea294
179 which contains a1a0e8b392b17caf50325498df54802fe3c03710
180 walk a1a0e8b392b17caf50325498df54802fe3c03710
181 walk 72d4f10e4a27dbb09ace1503c20dbac1912ee451
182 walk 13ed136b983a9c439eddeea8a1c2076cffbb685f
183 walk 0a633bf58b45fcf1a8299d3c82cd1fd26d3f48f2
184 walk db7117a9dd9a6e57e8632ea5848e1101eee0fbde
186 If our clone succeeded, we should now have a local directory called
187 hello. This directory will contain some files.
191 drwxr-xr-x 3 cworth cworth 4096 2007-09-27 16:40 hello
195 These files have the same contents and history in our repository as
196 they do in the repository we cloned.
198 Every git repository is complete, self-contained, and
199 independent. It contains its own private copy of a project’s files and
200 history. A cloned repository remembers the location of the repository
201 it was cloned from, but it does not communicate with that repository,
202 or any other, unless you tell it to.
204 What this means for now is that we’re free to experiment with our
205 repository, safe in the knowledge that it’s a private “sandbox” that
206 won’t affect anyone else.
208 #### 2.3.2 What’s in a repository?
210 When we take a more detailed look inside a repository, we can see that
211 it contains a directory named .git. This is where git keeps all
212 of its metadata for the repository.
216 . .. .git hello.c Makefile
218 The contents of the .git directory and its subdirectories are private
219 to git. Every other file and directory in the repository is
220 yours to do with as you please.
222 To introduce a little terminology, the .git directory is the “real”
223 repository, and all of the files and directories that coexist with it
224 are said to live in the working directory. An easy way to remember the
225 distinction is that the repository contains the history of your
226 project, while the working directory contains a snapshot of your
227 project at a particular point in history.
229 ### 2.4 A tour through history
231 One of the first things we might want to do with a new, unfamiliar
232 repository is understand its history. The “git log” command gives us a
236 commit a1a0e8b392b17caf50325498df54802fe3c03710
237 Author: Bryan O'Sullivan <bos@serpentine.com>
238 Date: Tue Sep 6 15:43:07 2005 -0700
242 commit 72d4f10e4a27dbb09ace1503c20dbac1912ee451
243 Author: Bryan O'Sullivan <bos@serpentine.com>
244 Date: Tue Sep 6 13:15:58 2005 -0700
246 Get make to generate the final binary from a .o file.
248 commit 13ed136b983a9c439eddeea8a1c2076cffbb685f
249 Author: Bryan O'Sullivan <bos@serpentine.com>
250 Date: Tue Sep 6 13:15:43 2005 -0700
252 Introduce a typo into hello.c.
254 commit 0a633bf58b45fcf1a8299d3c82cd1fd26d3f48f2
255 Author: Bryan O'Sullivan <mpm@selenic.com>
256 Date: Fri Aug 26 01:21:28 2005 -0700
260 commit db7117a9dd9a6e57e8632ea5848e1101eee0fbde
261 Author: Bryan O'Sullivan <mpm@selenic.com>
262 Date: Fri Aug 26 01:20:50 2005 -0700
264 Create a standard "hello, world" program
266 By default, this command prints a brief paragraph of output for each
267 change to the project that was recorded. In git terminology, we
268 call each of these recorded events a commit.
270 The fields in a record of output from “git log” are as follows.
272 * commit This field consists of a string of 40 hexadecimal characters.
273 This is a unique identifier for referring to particular commits.
274 * Author The identity of the person who authored the commit. This
275 field consist of two sub-fields for the user's name and email
276 address, (or at least an email-like idenitifer). Note that git
277 stores a separate "Committer" field for the person who commited
278 the change, (since often an author will email a change to a
279 maintainer that commits it). The "git log" command doesn't display
280 the Committer, but other git tools do.
281 * Date The date and time on which the commit was authored, (again
282 stored separately from the date the change was committed).
283 timezone in which it was created. (The date and time are displayed
284 in the timezone of the person who created the commit.)
285 * commit message The text message that the creator of the commit
286 entered to describe the commit, (generally a one-line summary
287 followed by more supporting text).
289 The default output printed by “git log” is purely a summary; it is
290 missing a lot of detail.
292 #### 2.4.1 Commits, revisions, and talking to other people
294 As English is a notoriously sloppy language, and computer science has
295 a hallowed history of terminological confusion (why use one term when
296 four will do?), revision control has a variety of words and phrases
297 that mean the same thing. If you are talking about git history
298 with other people, you will find that what we have called a “commit”
299 is often called a "revision". In other systems, a similar notion
300 is referred to as a "changeset". You might even see abbreviations of
301 these terms such as "rev", "change", or even "cset".
303 While it may not matter much what word you use to refer to the concept
304 of “a commit”, it's important to know how to name “a specific
305 commit”. We have already seen one means of referring to a particular
306 commit, the 40-character hexadecimal string shown by "git log". These
307 commit identifiers are powerful because they are permanent, unique
308 identifiers that always identify the same commit in any copy of a
309 repository. If two users are examining a working directory associated
310 with the same commit identifier, then those two users have precisely
311 the same contents in all files, and exactly the same history leading
314 So there are places where it is often important to archive the
315 complete commit identifier, (perhaps in bug-tracking systems to
316 indicate a specific commit that fixes a bug, for example). But often,
317 in more casual settings, it's more convenient to use abbreviated
318 commit identifiers. Git accept any unique prefix of a commit
319 identifier, (and for reasonably-sized project the first 8 or 10
320 characters are almost always unique).
322 And unlike the permanent commit identifiers, git also provides
323 transient means of identifying commits. In fact, in day-to-day use of
324 git, you will probably use these names more than commit
325 identifiers. One example is branch names, (such as the default
326 "master" branch in any git repository), or any project-specific branch
327 names such as "stable", "experimental", or "crazy-insane-changes". Git
328 also provides a special name "HEAD" which always refers to the current
331 #### 2.4.2 Naming related commits
333 Git offers simple ways to name revisions that are related to
334 particular revisions in the history. One syntax is the ~ suffix which
335 refers to the parent of a commit, or if followed by a number, to the
336 Nth parent. For example, since "HEAD" refers to the most recent commit
337 in the current branch, "HEAD~", refers to the previous commit, and
338 "HEAD~2" refers to two commits back in the history.
340 Another useful syntax is .. which can be used to specify a range of
341 commits. So "origin..master" specifies everything that has been
342 committed to master since it diverged from origin.
344 #### 2.4.3 Viewing specific revisions
346 You can use "git log" to explore the range syntax just introduced. For
347 example, to see a list of the most recent 3 revisions you can use
348 "HEAD~3..", (the destination of the range is implicitly HEAD in this
352 commit a1a0e8b392b17caf50325498df54802fe3c03710
353 Author: Bryan O'Sullivan <bos@serpentine.com>
354 Date: Tue Sep 6 15:43:07 2005 -0700
358 commit 72d4f10e4a27dbb09ace1503c20dbac1912ee451
359 Author: Bryan O'Sullivan <bos@serpentine.com>
360 Date: Tue Sep 6 13:15:58 2005 -0700
362 Get make to generate the final binary from a .o file.
364 commit 13ed136b983a9c439eddeea8a1c2076cffbb685f
365 Author: Bryan O'Sullivan <bos@serpentine.com>
366 Date: Tue Sep 6 13:15:43 2005 -0700
368 Introduce a typo into hello.c.
370 #### 2.4.4 Other log filters
372 Besides filtering by commit identifiers, git allows you to easily
373 filter the log output according to which files (or directories) are
374 modified by listing them after "--" wihch is necessary to distinguish
375 commit names from file names:
377 $ git log -- Makefile
378 commit 72d4f10e4a27dbb09ace1503c20dbac1912ee451
379 Author: Bryan O'Sullivan <bos@serpentine.com>
380 Date: Tue Sep 6 13:15:58 2005 -0700
382 Get make to generate the final binary from a .o file.
384 commit 0a633bf58b45fcf1a8299d3c82cd1fd26d3f48f2
385 Author: Bryan O'Sullivan <mpm@selenic.com>
386 Date: Fri Aug 26 01:21:28 2005 -0700
390 And "git log" can also filter based on the dates at which commits were
393 $ git log --since="2 weeks ago" --until="yesterday"
395 Another useful option is -n or --max-count which, unsurprisingly,
396 limits the maximum number of commits to be displayed.
398 #### 2.4.5 More detailed information
400 While the default information printed by “git log” is useful if you
401 already know what you’re looking for, you may need to see more details
402 of the change, such as the "diffstat" information with --stat:
404 $ git log --stat --max-count=3
405 commit a1a0e8b392b17caf50325498df54802fe3c03710
406 Author: Bryan O'Sullivan <bos@serpentine.com>
407 Date: Tue Sep 6 15:43:07 2005 -0700
412 1 files changed, 2 insertions(+), 6 deletions(-)
414 commit 72d4f10e4a27dbb09ace1503c20dbac1912ee451
415 Author: Bryan O'Sullivan <bos@serpentine.com>
416 Date: Tue Sep 6 13:15:58 2005 -0700
418 Get make to generate the final binary from a .o file.
421 1 files changed, 2 insertions(+), 0 deletions(-)
423 commit 13ed136b983a9c439eddeea8a1c2076cffbb685f
424 Author: Bryan O'Sullivan <bos@serpentine.com>
425 Date: Tue Sep 6 13:15:43 2005 -0700
427 Introduce a typo into hello.c.
430 1 files changed, 1 insertions(+), 1 deletions(-)
432 Or perhaps you'd like to see the actual patch content of each change,
433 which you can get with -p. That commit with the word typo in its name
434 looks suspicous, so let's tak a closer look. Remember that we can name
435 it as master~3, HEAD~3, or any prefix of its commit identifier, (such
438 $ git log -p -n 1 13ed136b
439 commit 13ed136b983a9c439eddeea8a1c2076cffbb685f
440 Author: Bryan O'Sullivan <bos@serpentine.com>
441 Date: Tue Sep 6 13:15:43 2005 -0700
443 Introduce a typo into hello.c.
445 diff --git a/hello.c b/hello.c
446 index ed55ec0..80b260c 100644
451 int main(int argc, char **argv)
453 - printf("hello, world!\n");
454 + printf("hello, world!\");
458 Of course, wanting to see all this information for a single commit is
459 such a common operation that it's given its own name in git, "git
460 show". So "git show 13ed136b" is a much easier way to get exactly the
464 commit 13ed136b983a9c439eddeea8a1c2076cffbb685f
465 Author: Bryan O'Sullivan <bos@serpentine.com>
466 Date: Tue Sep 6 13:15:43 2005 -0700
468 Introduce a typo into hello.c.
470 diff --git a/hello.c b/hello.c
471 index ed55ec0..80b260c 100644
476 int main(int argc, char **argv)
478 - printf("hello, world!\n");
479 + printf("hello, world!\");
483 ### 2.5 All about command options
485 Let’s take a brief break from exploring git commands to discuss
486 a pattern in the way that they work; you may find this useful to keep
487 in mind as we continue our tour.
489 Git has a consistent and straightforward approach to dealing
490 with the options that you can pass to commands. It follows the
491 conventions for options that are common to modern Linux and Unix
494 * Most options have long names. For example, as we’ve already seen,
495 the “git log" command accepts a --max-count=<number> option.
496 * Some options have short, single-character names. Often these are
497 aliases for long commands, (such as "-n <number>" instead of
498 --max-count=<number>), but sometimes the option exists in
499 short-form with no long-form equivalent, (such as -p). [XXX: It
500 wouldn't hurt to fix this by adding --patch, etc. right?]
501 * Long options start with two dashes (e.g. --max-count), while short
502 options start with one (e.g. -n).
504 * Option naming and usage is consistent across commands. For
505 example, every command that lets you specify a commit identifier
506 or range will accept the same expressions, (HEAD~3,
507 origin..master, 72d4f10e, etc), while any command that can be
508 limited by paths will accept the same expressions ("-- doc/
511 Many commands that print output of some kind can be made more quiet by
512 passing the -q or --quiet options.
514 ### 2.6 Making and reviewing changes
516 Now that we have a grasp of viewing history in git, let’s take a
517 look at making some changes and examining them.
519 The first thing we’ll do is isolate our experiment in a repository of
520 its own. We use the “git clone” command, but we don’t need to clone a
521 copy of the remote repository. Since we already have a copy of it
522 locally, we can just clone that instead. This is much faster than
523 cloning over the network, and cloning a local repository uses less
524 disk space in most cases, too.
527 $ git clone hello my-hello
528 Initialized empty Git repository in /tmp/my-hello/.git/
531 [XXX We say "empty" here, (presumably from the git-init part),
532 but shouldn't the command also report the succesful clone
533 which makes it non-empty? And what the heck does "0 blocks"
536 As an aside, it’s often good practice to keep a “pristine” copy of a
537 remote repository around, which you can then make temporary clones of
538 to create sandboxes for each task you want to work on. This lets you
539 work on multiple tasks in parallel, each isolated from the others
540 until it’s complete and you’re ready to integrate it back. Because
541 local clones are so cheap, there’s almost no overhead to cloning and
542 destroying repositories whenever you want.
544 Alternatively, you can achieve much the same effect by creating
545 multiple branches in a single repository, (but we won't go into detail
546 on how to do that in this chapter). Some people greatly appreciate
547 having multiple branches in a single repository rather than having
548 many repositories cluttering up their filesystem. Other people prefer
549 the ability to have working-tree changes, and intermediate build
550 files, etc. each isolated in a separate repository per branch. Both
551 modes are very well-supported by git, so it's really a matter of which
552 you find most appropriate at any time given your tastes and project
555 In our my-hello repository, we have a file hello.c that contains the
556 classic “hello, world” program. Let’s use the ancient and venerable
557 sed command to edit this file so that it prints a second line of
558 output. (I’m only using sed to do this because it’s easy to write a
559 scripted example this way. Since you’re not under the same constraint,
560 you probably won’t want to use sed; simply use your preferred text
561 editor to do the same thing.)
563 $ sed -i '/printf/a\\tprintf("hello again!\\n");' hello.c
565 The “git status” command will tell us what git knows about the files
572 # Changed but not updated:
573 # (use "git add <file>..." to update what will be committed)
577 no changes added to commit (use "git add" and/or "git commit -a")
579 We see that “git status” command prints a line with "modified" for
580 hello.c. The “git status” command will not print any output for files
581 that have not been modified.
583 Notice that we didn’t need to inform git that we were going to modify
584 the file before we started, or that we had modified the file after we
585 were done; it was able to figure this out itself.
587 It’s a little bit helpful to know that we’ve modified hello.c, but we
588 might prefer to know exactly what changes we’ve made to it. To do
589 this, we use the “git diff” command.
592 diff --git a/hello.c b/hello.c
593 index 9a3ff79..6d28887 100644
597 int main(int argc, char **argv)
599 printf("hello, world!\");
600 + printf("hello again!\n");
604 ### 2.7 Recording changes in a new commit
606 We can modify files, build and test our changes, and use “git status”
607 and “git diff” to review our changes, until we’re satisfied with what
608 we’ve done and arrive at a natural stopping point where we want to
609 record our work in a new commit.
611 The “git commit” command lets us create a new changeset; we’ll usually
612 refer to this as “making a commit” or “committing”.
614 #### 2.7.1 Setting up a username
616 When you try to run “git commit” for the first time, it might not do
617 exactly what you want. Git records your name and address with each
618 change that you commit, (as both author and committer unless you tell
619 it otherwise), so that you and others will later be able to tell who
620 made each change. Git tries to automatically figure out a sensible
621 name and address to attribute to both author and committer. It will
622 attempt each of the following methods, in order, (stopping for each field as soon as a value is found):
624 1. If you specify a --author option to the “git commit” command on
625 the command line, followed by a "Real Name <email@example.com>"
626 string, then this name and addresss will be used for the author
627 fields. The committer fields will still be determined as
628 below. This option is very helpful for when applying a commit
629 originally authored by someone other than yourself.
630 2. If any of the GIT_AUTHOR_NAME, GIT_AUTHOR_EMAIL,
631 GIT_COMMITTER_NAME, or GIT_COMMITER_EMAIL environment variables
632 are set, then those values will be used for the corresponding
634 3. If you have a file in your home directory called .gitconfig, with
635 name or email settings in the [user] section, then these values
636 will be used to set any remaining author and committer
637 fields. For more details on the contents of this file, refer to
639 4. If you have a file in the local repository called .git/config,
640 again with name or email settings in the [user] section, then
641 these values will be used to set any remaining author and
643 5. If you have set the EMAIL environment variable, this will be used
644 to set author and committer email addresses if still unset.
645 6. git will query your system to find out your real name from
646 available GECOS field and your username, hostname, and domain to
647 construct an email address, (or at least an identifier resembling
650 If all of these mechanisms fail, "git commit" will fail, printing an
651 error message instructing you how to use "git config" to tell git your
652 name and email address.
654 You should think of the GIT_AUTHOR/COMMITER_NAME/EMAIL environment
655 variables and the --author option to the “git commit” command as ways
656 to override git’s default selection. For normal use, the simplest and
657 most robust way to set your information is by creating a .gitconfig
658 file, (either manually or with the "git config" command); see below
661 ##### Creating a git configuration file
663 To set your name and email address, just use the following commands:
665 git config --global user.name "Your Name"
666 git config --global user.email "you@example.com"
668 The --global option means that this command will set global
669 information, (affecting all repositories on this machine), in the
670 .gitconfig file in your home directory. Alternately, you could omit
671 the --global which would make the change take effect only in the local
672 repository. This is convenient if you want to have different email
673 addresses associated with different projects, for example.
675 Of course, git's configuration file is a simple-to-edit plain-text
676 file, so instead of using the above commands, you can also just edit
677 the files directly. Use your favorite editor to create a file called
678 .gitconfig in your home directory, (or if you ran the above commands
679 then it will be there already). The initial contents of your
680 .gitconfig should look like this.
682 # This is a git configuration file.
685 email = you@example.com
687 Similarly, you can make a repository-specific configuration by editing
688 .git/config in the local repository. It will already have some
689 sections present, (created by the "git clone"), just add a [user]
692 The “[user]” line begins a section of the config file, so you can read
693 the “name = ...” line as meaning “set the value of the name item in
694 the user section”. This is the same notion expressed with the
695 "user.name" syntax on the git-config command line. A section
696 continues until a new section begins, or the end of the file. Git
697 ignores empty lines and treats any text from “#” to the end of a line
700 ##### Choosing a user name
702 You can use any text you like as the value of the name and email
703 configuration items, since this information is for reading by other
704 people, not for interpreting by git. It is conventional to use a valid
705 email address, but some, (notably Linus Torvalds, the original author
706 of git), actually like the default user@hostname convention that git
707 falls back on without any additional information. There's no
708 requirement that the email address actually be valid, and perhaps it's
709 useful to be reminded which machine was used to create particular
712 #### 2.7.2 Writing a commit message
714 When we commit a change, git drops us into a text editor to
715 enter a message that will describe the modifications we’ve made in
716 this commit. This is called the commit message. It will be a record
717 for readers of what we did and why, and it will be printed by “git log”
718 after we’ve finished committing.
722 Note: The -a on the command-line instructs git to commit all changes
723 to tracked files. Without this, "git commit" will only commit changes
724 that have been previously staged for committing with "git add
725 file". The most common usage is to commit with "git commit -a" and
726 only use "git add file; git commit" when there is a need to commit
727 only some subset of changes that have been made.
729 The editor that the “git commit” command drops us into will contain an
730 empty line, followed by a number of lines starting with “#”.
733 # Please enter the commit message for your changes.
734 # (Comment lines starting with '#' will not be included)
736 # Changes to be committed:
737 # (use "git reset HEAD <file>..." to unstage)
742 git ignores the lines that start with “#”; it uses them only
743 to tell us which files it’s recording changes to. Modifying or
744 deleting these lines has no effect.
746 #### 2.7.3 Writing a good commit message
748 A good commit message will generally have a single line that
749 summarizes the commit, a blank line, and then one or more pargraphs
750 with supporting detail. Since many tools only print the first line of
751 a commit message by default, it’s important that the first line stands
754 One example of a first-line-only viewer is "git log
755 --pretty=short". Other examples include graphical history viewers such
756 as gitk and gitview, and web-based viewers such as gitweb and cgit.
758 Here’s a real example of a commit message that doesn’t follow
759 this guideline, and hence has a summary that is not readable.
761 $ git log --pretty=short
762 commit 3ef5535144da88a854f7930503845cd44506c2e2
763 Author: Censored Person <censored.person@example.org>
765 include buildmeister/commondefs. Add an exports and install
767 As far as the remainder of the contents of the commit message are
768 concerned, there are no hard-and-fast rules. git itself doesn’t
769 interpret or care about the contents of the commit message, though
770 your project may have policies that dictate a certain kind of
773 My personal preference is for short, but informative, commit messages
774 that tell me something that I can’t figure out with a quick glance at
775 the output of “git log -p".
777 #### 2.7.4 Aborting a commit
779 If you decide that you don’t want to commit while in the middle of
780 editing a commit message, simply exit from your editor without saving
781 the file that it’s editing. This will cause nothing to happen to
782 either the repository or the working directory.
784 #### 2.7.5 Admiring our new handiwork
786 Once we’ve finished the commit, we can use the “git show” command to
787 display the commit we just created. As discussed previously, this
788 command produces output that is identical to “git log -p”, but for
789 only a single revision, (and the most recent revision by default):
792 commit 018cfb742be6176443ffddac454e593e802ddf3e
793 Author: Carl Worth <cworth@cworth.org>
794 Date: Thu Sep 27 23:55:00 2007 -0700
796 Added an extra line of output.
798 If I would have been clever I would have fixed that old typo
801 diff --git a/hello.c b/hello.c
802 index 9a3ff79..6d28887 100644
806 int main(int argc, char **argv)
808 printf("hello, world!\");
809 + printf("hello again!\n");
813 Note that you will not see the same commit identifier for your commit,
814 even if the change you made is identical to mine. The commit
815 identifier incorporates not only the contents of the files, but commit
816 message, the author and committer names and emails, and the author and
817 commit dates. (OK, so now you probably know enough to be able to guess
818 the right command to produce a commit with exactly the commit
819 identifier shown above. Can you do it?)
821 #### 2.7.6 Fixing up a broken commit (before anyone else sees it)
823 So now that we've cloned a local repository, made a change to the
824 code, setup our name and email address, and made a commit with a
825 careful message, we're just about ready to share our change with the
826 world. But wait, we forgot to try to compile it didn't we?
829 cc -c -o hello.o hello.c
830 hello.c:10:9: warning: missing terminating " character
831 hello.c:10:9: warning: missing terminating " character
832 hello.c: In function ‘main’:
833 hello.c:10: error: missing terminating " character
834 hello.c:11: error: expected ‘)’ before ‘;’ token
835 hello.c:13: warning: passing argument 1 of ‘printf’ makes pointer from integer without a cast
836 hello.c:13: error: expected ‘;’ before ‘}’ token
837 make: *** [hello.o] Error 1
839 Oh look. The code's broken and doesn't compile. We don't want to share
840 code in this state. For situations where you notice one tiny detail
841 that got left out of the last commit, (a silly syntax error, a
842 misspelling in a comment or commit messsage), git provides a very
843 handy tool for just changing the last commit.
845 So fix that typo, (a missing 'n' between the '\' and the '"'), with
846 your editor or with something like this:
848 sed -i 's/\\"/\\n"/' hello.c
850 And then you can just amend the previous commit rather than creating a
851 new one with the --amend option to "git commit":
853 $ git commit -a --amend
855 Note that we use -a to include the code change here. And that helps
856 point out a situation where "git commit" is useful without the -a
857 option, "git commit --amend" is a useful command for amend just the
858 last commit message, without committing any new code changes, even if
859 some files have been modified in the working tree.
861 And here's the final result:
864 commit 839b58d021c618bd0e1d336d4d5878a0082672e6
865 Author: Carl Worth <cworth@cworth.org>
866 Date: Thu Sep 27 23:55:00 2007 -0700
868 Added an extra line of output and fixed the typo bug.
870 diff --git a/hello.c b/hello.c
871 index 9a3ff79..ca750e0 100644
876 int main(int argc, char **argv)
878 - printf("hello, world!\");
879 + printf("hello, world!\n");
880 + printf("hello again!\n");
884 I can't help but point out that this really was a poor example for
885 --amend. The end result is a single commit that does two independent
886 things, (fixes one bug and adds one new feature). It's much better to
887 create a code history where each commit makes an independent change,
888 (and as small as possible). This is important for several reasons:
890 * Small changes are easier to review
892 * Independent changes are easier to split up if only part of the
893 series gets accepted "upstream" for one reason or another.
895 * The smaller the changes are the more useful the history will be
896 when actually using the history, not just viewing it. This is
897 particularly important when doing "git bisect"---that's a powerful
898 tool for isolating the single commit that introduces a bug. And
899 it's much more powerful if the commit it isolates is as small as
902 So it's a good thing this document is available under a license that
903 allows for distribution of modified versions. Someone should clean up
904 the --amend example to not teach bad habits like I did above. [Note:
905 All this bad-habit stuff was introduced by me, and was not present in
906 Bryan's original chapter. -Carl]
908 ### 2.8 Sharing changes
910 We mentioned earlier that repositories in git are
911 self-contained. This means that the commit we just created exists
912 only in our my-hello repository. Let’s look at a few ways that we can
913 propagate this change into other repositories.
915 #### 2.8.1 Pulling changes from another repository
917 To get started, let’s clone our original hello repository, which does
918 not contain the change we just committed. We’ll call our temporary
919 repository hello-pull.
922 $ git clone hello hello-pull
923 Initialized empty Git repository in /tmp/hello-pull/.git/
926 We could use the “git pull” command to apply changes from my-hello to
927 our master branch in hello-pull. However, blindly pulling unknown
928 changes into a repository is a somewhat scary prospect. The "git pull"
929 command is coneptually the combination of two commands, "git fetch"
930 and "git merge"; we can run those separately to examine the changes
931 before applying them locally. First we do the fetch:
934 $ git fetch ../my-hello
935 remote: Generating pack...
936 Unpacking 3 objects...
938 remote: Done counting 5 objects.
939 Result has 3 objects.
940 Deltifying 3 objects...
941 100% remote: (3/3) done
942 Total 3 (delta 1), reused 0 (delta 0)
944 The fetched commits (or commit in this case) are available as the name
945 FETCH_HEAD. [XXX: Shouldn't git-fetch print that name out to the user
946 if the user didn't provide a specific branch name to fetch into.] And
947 the difference between what we had before and what exists on
948 FETCH_HEAD can easily be examined with the ..FETCH_HEAD range
951 $ git log ..FETCH_HEAD
952 commit 839b58d021c618bd0e1d336d4d5878a0082672e6
953 Author: Carl Worth <cworth@cworth.org>
954 Date: Thu Sep 27 23:55:00 2007 -0700
956 Added an extra line of output and fixed the typo bug.
958 Since these commits actually exist in the local repository now, we
959 don't need to fetch or pull them from the remote repository again---we
960 can now use "git merge" to apply the previously fetched commits. (A
961 mercurial user might notice here that git does not have the race
962 condition between "hg incoming" and "hg pull" that mercurial has since
963 the commits are fetched only once.)
965 $ git merge FETCH_HEAD
966 Updating a1a0e8b..839b58d
969 1 files changed, 2 insertions(+), 1 deletions(-)
971 Notice that "git merge" reports that our branch pointer has been
972 updated from a1a0e8b to 839b58d. Also, this is a "fast forward"
973 meaning that the new commits are a linear sequence on top of the
974 commit we already hand. In other words, there wasn't any divergence
975 between these two repositories so no actual "merge" commit was
978 This separation of fetch and merge is useful when you need to
979 carefully review some changes before applying them. But often you're
980 in a situation where you know you trust the remote repository and you
981 simply want to pull those changes as conveniently as possible, (no
982 extra commands, no typing a magic name like FETCH_HEAD). This is the
983 case when the tracking upstream development of a project with git. And
984 in that case, the above steps are as simple as just executing "git
985 pull". So let's repeat all that the simpler way:
988 $ git clone hello hello-tracking
989 Initialized empty Git repository in /tmp/hello-tracking/.git/
992 $ git pull ../my-hello
993 remote: Generating pack...
994 remote: Done counting 5 objects.
995 Result has 3 objects.
996 Deltifying 3 objects...
997 Unpacking 3 objects...
998 remote: 100% (3/3) done
999 Total 3 (delta 1), reused 0 (delta 0)
1001 Updating a1a0e8b..839b58d
1004 1 files changed, 2 insertions(+), 1 deletions(-)
1006 It should be plain to see that the "git pull" command really did the
1007 combined sequence of "git fetch" and "git merge". Also, if you want to
1008 pull from the same repository you cloned from originally, (which is
1009 the common case for the upstream-tracking scenario), then "git pull"
1010 with no explicit repository is suffcient, and it will default to
1011 pulling from the same repository as the original clone.
1013 [XXX: The structure of the preceding section follows that of the
1014 original hgbook. But an alternate structure that arranged to pull from
1015 the originally cloned repository (as would be common) would allow for
1016 more straightforward use of git's features. For example, instead of
1017 the silly FETCH_HEAD stuff it would allow for "git fetch" and "git log
1018 master..origin" to be a very nice replacement for "hg
1019 incoming". Similarly, below, "git log origin..master" would make a
1020 nice replacement for "hg outgoing" which is something I didn't offer
1021 at all. One could also use git's remotes with the myriad repositories
1022 as used here, but it would require doing things like "git remote add
1023 <some-name> ../hello-pull" and that seems like a bit much to introduce
1024 for a turorial of this level. If nothing else, if the above section
1025 seems a little intimidating, understand that it's because things are
1026 not presented in the most natural "git way", (and I'm a little too
1027 tired to fix it tonight).]
1029 #### 2.8.2 Checking out previous revisions
1031 If any users of mercurial are reading this, they might wonder if
1032 there's a need for the equivalent of "hg update" after doing a "git
1033 pull". And the answer is no. Unlike mercurial, "git pull" and "git
1034 merge" will automatically update the workind-directory files as
1037 But there's another function provided by "hg update" which is to
1038 update the working-directory files to a particular revision. In git,
1039 this functionality is provided by the "git checkout" command. To
1040 checkout a particular branch, tag, or an arbitrary revions, simply
1041 give the appropriate name to the "git checkout" command. For example,
1042 to examine the files as they existed before the original typo
1043 introduction, we could do:
1045 $ git checkout 0a633bf5
1046 Note: moving to "0a633bf5" which isn't a local branch
1047 If you want to create a new branch from this checkout, you may do so
1048 (now or later) by using -b with the checkout command again. Example:
1049 git checkout -b <new_branch_name>
1050 HEAD is now at 0a633bf... Create a makefile
1052 The note that git gives us is to indicate that we are checking out a
1053 non-branch revision. This is perfectly fine if we are just exploring
1054 history, but if we actually wanted to use this revision as the basis
1055 for new commits, we would first have to create a new branch name as it
1058 For now, let's return back to the tip of the master branch by just
1059 checking it out again:
1061 $ git checkout master
1062 Previous HEAD position was 0a633bf... Create a makefile
1063 Switched to branch "master"
1065 #### 2.8.3 Pushing changes to another repository
1067 Git lets us push changes to another repository, from the repository
1068 we’re currently visiting. As with previous examples, above, we’ll
1069 first create a temporary repository to push our changes into. But
1070 instead of using "git clone", this time we'll use "git init" to make a
1071 repository from an empty directory. We do this to create a "bare"
1072 repository which is simply a repository that has no working-directory
1073 files associated with it. In general, you should only push to bare
1080 Initialized empty Git repository in /tmp/hello-push/
1082 And then we'll go back to our my-hello repository to perform the
1083 push. Since this is our very first push into this repository we need
1084 to tell git which branches to push. The easiest way to do this is to
1085 use --all to indicate all branches:
1088 $ git push ../hello-push --all
1089 updating 'refs/heads/master'
1090 from 0000000000000000000000000000000000000000
1091 to 839b58d021c618bd0e1d336d4d5878a0082672e6
1093 Done counting 18 objects.
1094 Deltifying 18 objects...
1096 Writing 18 objects...
1098 Total 18 (delta 3), reused 0 (delta 0)
1099 Unpacking 18 objects...
1101 refs/heads/master: 0000000000000000000000000000000000000000 -> 839b58d021c618bd0e1d336d4d5878a0082672e6
1103 For subsequent pushes we don't need to specify --all as "git push"
1104 will push all branches that exist in both the local and remote
1107 What happens if we try to pull or push changes and the receiving
1108 repository already has those changes? Nothing too exciting.
1110 $ git push ../hello-push
1111 Everything up-to-date
1113 #### 2.8.4 Sharing changes over a network
1115 The commands we have covered in the previous few sections are not
1116 limited to working with local repositories. Each works in exactly the
1117 same fashion over a network connection; simply pass in a URL or an ssh
1118 host:/path/name specification instead of a local path.
1121 Open Publication License
1123 Version 1.0, 8 June 1999
1125 ### D.1 Requirements on both unmodified and modified versions
1127 The Open Publication works may be reproduced and distributed in whole
1128 or in part, in any medium physical or electronic, provided that the
1129 terms of this license are adhered to, and that this license or an
1130 incorporation of it by reference (with any options elected by the
1131 author(s) and/or publisher) is displayed in the reproduction.
1133 Proper form for an incorporation by reference is as follows:
1135 Copyright (c) year by author’s name or designee. This material may be
1136 distributed only subject to the terms and conditions set forth in the
1137 Open Publication License, vx.y or later (the latest version is
1138 presently available at
1139 [http://www.opencontent.org/openpub/][http://www.opencontent.org/openpub/]).
1141 The reference must be immediately followed with any options elected by
1142 the author(s) and/or publisher of the document (see section D.6).
1144 Commercial redistribution of Open Publication-licensed material is
1147 Any publication in standard (paper) book form shall require the
1148 citation of the original publisher and author. The publisher and
1149 author’s names shall appear on all outer surfaces of the book. On all
1150 outer surfaces of the book the original publisher’s name shall be as
1151 large as the title of the work and cited as possessive with respect to
1156 The copyright to each Open Publication is owned by its author(s) or
1159 ### D.3 Scope of license
1161 The following license terms apply to all Open Publication works,
1162 unless otherwise explicitly stated in the document.
1164 Mere aggregation of Open Publication works or a portion of an Open
1165 Publication work with other works or programs on the same media shall
1166 not cause this license to apply to those other works. The aggregate
1167 work shall contain a notice specifying the inclusion of the Open
1168 Publication material and appropriate copyright notice.
1170 Severability. If any part of this license is found to be unenforceable
1171 in any jurisdiction, the remaining portions of the license remain in
1174 No warranty. Open Publication works are licensed and provided “as is”
1175 without warranty of any kind, express or implied, including, but not
1176 limited to, the implied warranties of merchantability and fitness for
1177 a particular purpose or a warranty of non-infringement.
1179 ### D.4 Requirements on modified works
1181 All modified versions of documents covered by this license, including
1182 translations, anthologies, compilations and partial documents, must
1183 meet the following requirements:
1185 1. The modified version must be labeled as such.
1186 2. The person making the modifications must be identified and the
1187 modifications dated.
1188 3. Acknowledgement of the original author and publisher if
1189 applicable must be retained according to normal academic citation
1191 4. The location of the original unmodified document must be identified.
1192 5. The original author’s (or authors’) name(s) may not be used to
1193 assert or imply endorsement of the resulting document without the
1194 original author’s (or authors’) permission.
1196 ### D.5 Good-practice recommendations
1198 In addition to the requirements of this license, it is requested from
1199 and strongly recommended of redistributors that:
1201 1. If you are distributing Open Publication works on hardcopy or
1202 CD-ROM, you provide email notification to the authors of your
1203 intent to redistribute at least thirty days before your
1204 manuscript or media freeze, to give the authors time to provide
1205 updated documents. This notification should describe
1206 modifications, if any, made to the document.
1207 2. All substantive modifications (including deletions) be either
1208 clearly marked up in the document or else described in an
1209 attachment to the document.
1210 3. Finally, while it is not mandatory under this license, it is
1211 considered good form to offer a free copy of any hardcopy and
1212 CD-ROM expression of an Open Publication-licensed work to its
1215 ### D.6 License options
1217 The author(s) and/or publisher of an Open Publication-licensed
1218 document may elect certain options by appending language to the
1219 reference to or copy of the license. These options are considered part
1220 of the license instance and must be included with the license (or its
1221 incorporation by reference) in derived works.
1223 1. To prohibit distribution of substantively modified versions
1224 without the explicit permission of the author(s). “Substantive
1225 modification” is defined as a change to the semantic content of
1226 the document, and excludes mere changes in format or
1227 typographical corrections.
1229 To accomplish this, add the phrase “Distribution of substantively
1230 modified versions of this document is prohibited without the
1231 explicit permission of the copyright holder.” to the license
1234 2. To prohibit any publication of this work or derivative works in
1235 whole or in part in standard (paper) book form for commercial
1236 purposes is prohibited unless prior permission is obtained from
1237 the copyright holder.
1239 To accomplish this, add the phrase “Distribution of the work or
1240 derivative of the work in any standard (paper) book form is
1241 prohibited unless prior permission is obtained from the copyright
1242 holder.” to the license reference or copy.