| blob | 873ec36605f6777518f54e26bebcd2ea4f7ac8ab |
2 <html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><title>Git User Manual</title><link rel="stylesheet" type="text/css" href="docbook-xsl.css"><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot"></head><body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"><div lang="en" class="book"><div class="titlepage"><div><div><h1 class="title"><a name="id-1"></a>Git User Manual</h1></div><div><div class="revhistory"><table style="border-style:solid; width:100%;" summary="Revision History"><tr><th align="left" valign="top" colspan="2"><b>Revision History</b></th></tr><tr><td align="left"></td><td align="left">2024-04-15</td></tr></table></div></div></div><hr></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="preface"><a href="#_introduction">Introduction</a></span></dt><dt><span class="chapter"><a href="#repositories-and-branches">1. Repositories and Branches</a></span></dt><dd><dl><dt><span class="section"><a href="#how-to-get-a-git-repository">How to get a Git repository</a></span></dt><dt><span class="section"><a href="#how-to-check-out">How to check out a different version of a project</a></span></dt><dt><span class="section"><a href="#understanding-commits">Understanding History: Commits</a></span></dt><dd><dl><dt><span class="section"><a href="#understanding-reachability">Understanding history: commits, parents, and reachability</a></span></dt><dt><span class="section"><a href="#history-diagrams">Understanding history: History diagrams</a></span></dt><dt><span class="section"><a href="#what-is-a-branch">Understanding history: What is a branch?</a></span></dt></dl></dd><dt><span class="section"><a href="#manipulating-branches">Manipulating branches</a></span></dt><dt><span class="section"><a href="#detached-head">Examining an old version without creating a new branch</a></span></dt><dt><span class="section"><a href="#examining-remote-branches">Examining branches from a remote repository</a></span></dt><dt><span class="section"><a href="#how-git-stores-references">Naming branches, tags, and other references</a></span></dt><dt><span class="section"><a href="#Updating-a-repository-With-git-fetch">Updating a repository with git fetch</a></span></dt><dt><span class="section"><a href="#fetching-branches">Fetching branches from other repositories</a></span></dt></dl></dd><dt><span class="chapter"><a href="#exploring-git-history">2. Exploring Git history</a></span></dt><dd><dl><dt><span class="section"><a href="#using-bisect">How to use bisect to find a regression</a></span></dt><dt><span class="section"><a href="#naming-commits">Naming commits</a></span></dt><dt><span class="section"><a href="#creating-tags">Creating tags</a></span></dt><dt><span class="section"><a href="#browsing-revisions">Browsing revisions</a></span></dt><dt><span class="section"><a href="#generating-diffs">Generating diffs</a></span></dt><dt><span class="section"><a href="#viewing-old-file-versions">Viewing old file versions</a></span></dt><dt><span class="section"><a href="#history-examples">Examples</a></span></dt><dd><dl><dt><span class="section"><a href="#counting-commits-on-a-branch">Counting the number of commits on a branch</a></span></dt><dt><span class="section"><a href="#checking-for-equal-branches">Check whether two branches point at the same history</a></span></dt><dt><span class="section"><a href="#finding-tagged-descendants">Find first tagged version including a given fix</a></span></dt><dt><span class="section"><a href="#showing-commits-unique-to-a-branch">Showing commits unique to a given branch</a></span></dt><dt><span class="section"><a href="#making-a-release">Creating a changelog and tarball for a software release</a></span></dt><dt><span class="section"><a href="#Finding-commits-With-given-Content">Finding commits referencing a file with given content</a></span></dt></dl></dd></dl></dd><dt><span class="chapter"><a href="#Developing-With-git">3. Developing with Git</a></span></dt><dd><dl><dt><span class="section"><a href="#telling-git-your-name">Telling Git your name</a></span></dt><dt><span class="section"><a href="#creating-a-new-repository">Creating a new repository</a></span></dt><dt><span class="section"><a href="#how-to-make-a-commit">How to make a commit</a></span></dt><dt><span class="section"><a href="#creating-good-commit-messages">Creating good commit messages</a></span></dt><dt><span class="section"><a href="#ignoring-files">Ignoring files</a></span></dt><dt><span class="section"><a href="#how-to-merge">How to merge</a></span></dt><dt><span class="section"><a href="#resolving-a-merge">Resolving a merge</a></span></dt><dd><dl><dt><span class="section"><a href="#conflict-resolution">Getting conflict-resolution help during a merge</a></span></dt></dl></dd><dt><span class="section"><a href="#undoing-a-merge">Undoing a merge</a></span></dt><dt><span class="section"><a href="#fast-forwards">Fast-forward merges</a></span></dt><dt><span class="section"><a href="#fixing-mistakes">Fixing mistakes</a></span></dt><dd><dl><dt><span class="section"><a href="#reverting-a-commit">Fixing a mistake with a new commit</a></span></dt><dt><span class="section"><a href="#fixing-a-mistake-by-rewriting-history">Fixing a mistake by rewriting history</a></span></dt><dt><span class="section"><a href="#checkout-of-path">Checking out an old version of a file</a></span></dt><dt><span class="section"><a href="#interrupted-work">Temporarily setting aside work in progress</a></span></dt></dl></dd><dt><span class="section"><a href="#ensuring-good-performance">Ensuring good performance</a></span></dt><dt><span class="section"><a href="#ensuring-reliability">Ensuring reliability</a></span></dt><dd><dl><dt><span class="section"><a href="#checking-for-corruption">Checking the repository for corruption</a></span></dt><dt><span class="section"><a href="#recovering-lost-changes">Recovering lost changes</a></span></dt></dl></dd></dl></dd><dt><span class="chapter"><a href="#sharing-development">4. Sharing development with others</a></span></dt><dd><dl><dt><span class="section"><a href="#getting-updates-With-git-pull">Getting updates with git pull</a></span></dt><dt><span class="section"><a href="#submitting-patches">Submitting patches to a project</a></span></dt><dt><span class="section"><a href="#importing-patches">Importing patches to a project</a></span></dt><dt><span class="section"><a href="#public-repositories">Public Git repositories</a></span></dt><dd><dl><dt><span class="section"><a href="#setting-up-a-public-repository">Setting up a public repository</a></span></dt><dt><span class="section"><a href="#exporting-via-git">Exporting a Git repository via the Git protocol</a></span></dt><dt><span class="section"><a href="#exporting-via-http">Exporting a git repository via HTTP</a></span></dt><dt><span class="section"><a href="#pushing-changes-to-a-public-repository">Pushing changes to a public repository</a></span></dt><dt><span class="section"><a href="#forcing-push">What to do when a push fails</a></span></dt><dt><span class="section"><a href="#setting-up-a-shared-repository">Setting up a shared repository</a></span></dt><dt><span class="section"><a href="#setting-up-gitweb">Allowing web browsing of a repository</a></span></dt></dl></dd><dt><span class="section"><a href="#how-to-get-a-git-repository-with-minimal-history">How to get a Git repository with minimal history</a></span></dt><dt><span class="section"><a href="#sharing-development-examples">Examples</a></span></dt><dd><dl><dt><span class="section"><a href="#maintaining-topic-branches">Maintaining topic branches for a Linux subsystem maintainer</a></span></dt></dl></dd></dl></dd><dt><span class="chapter"><a href="#cleaning-up-history">5. Rewriting history and maintaining patch series</a></span></dt><dd><dl><dt><span class="section"><a href="#patch-series">Creating the perfect patch series</a></span></dt><dt><span class="section"><a href="#using-git-rebase">Keeping a patch series up to date using git rebase</a></span></dt><dt><span class="section"><a href="#rewriting-one-commit">Rewriting a single commit</a></span></dt><dt><span class="section"><a href="#reordering-patch-series">Reordering or selecting from a patch series</a></span></dt><dt><span class="section"><a href="#interactive-rebase">Using interactive rebases</a></span></dt><dt><span class="section"><a href="#patch-series-tools">Other tools</a></span></dt><dt><span class="section"><a href="#problems-With-rewriting-history">Problems with rewriting history</a></span></dt><dt><span class="section"><a href="#bisect-merges">Why bisecting merge commits can be harder than bisecting linear history</a></span></dt></dl></dd><dt><span class="chapter"><a href="#advanced-branch-management">6. Advanced branch management</a></span></dt><dd><dl><dt><span class="section"><a href="#fetching-individual-branches">Fetching individual branches</a></span></dt><dt><span class="section"><a href="#fetch-fast-forwards">git fetch and fast-forwards</a></span></dt><dt><span class="section"><a href="#forcing-fetch">Forcing git fetch to do non-fast-forward updates</a></span></dt><dt><span class="section"><a href="#remote-branch-configuration">Configuring remote-tracking branches</a></span></dt></dl></dd><dt><span class="chapter"><a href="#git-concepts">7. Git concepts</a></span></dt><dd><dl><dt><span class="section"><a href="#the-object-database">The Object Database</a></span></dt><dd><dl><dt><span class="section"><a href="#commit-object">Commit Object</a></span></dt><dt><span class="section"><a href="#tree-object">Tree Object</a></span></dt><dt><span class="section"><a href="#blob-object">Blob Object</a></span></dt><dt><span class="section"><a href="#trust">Trust</a></span></dt><dt><span class="section"><a href="#tag-object">Tag Object</a></span></dt><dt><span class="section"><a href="#pack-files">How Git stores objects efficiently: pack files</a></span></dt><dt><span class="section"><a href="#dangling-objects">Dangling objects</a></span></dt><dt><span class="section"><a href="#recovering-from-repository-corruption">Recovering from repository corruption</a></span></dt></dl></dd><dt><span class="section"><a href="#the-index">The index</a></span></dt></dl></dd><dt><span class="chapter"><a href="#submodules">8. Submodules</a></span></dt><dd><dl><dt><span class="section"><a href="#pitfalls-with-submodules">Pitfalls with submodules</a></span></dt></dl></dd><dt><span class="chapter"><a href="#low-level-operations">9. Low-level Git operations</a></span></dt><dd><dl><dt><span class="section"><a href="#object-manipulation">Object access and manipulation</a></span></dt><dt><span class="section"><a href="#the-workflow">The Workflow</a></span></dt><dd><dl><dt><span class="section"><a href="#working-directory-to-index">working directory → index</a></span></dt><dt><span class="section"><a href="#index-to-object-database">index → object database</a></span></dt><dt><span class="section"><a href="#object-database-to-index">object database → index</a></span></dt><dt><span class="section"><a href="#index-to-working-directory">index → working directory</a></span></dt><dt><span class="section"><a href="#tying-it-all-together">Tying it all together</a></span></dt></dl></dd><dt><span class="section"><a href="#examining-the-data">Examining the data</a></span></dt><dt><span class="section"><a href="#merging-multiple-trees">Merging multiple trees</a></span></dt><dt><span class="section"><a href="#merging-multiple-trees-2">Merging multiple trees, continued</a></span></dt></dl></dd><dt><span class="chapter"><a href="#hacking-git">10. Hacking Git</a></span></dt><dd><dl><dt><span class="section"><a href="#object-details">Object storage format</a></span></dt><dt><span class="section"><a href="#birdview-on-the-source-code">A birds-eye view of Git’s source code</a></span></dt></dl></dd><dt><span class="chapter"><a href="#glossary">11. Git Glossary</a></span></dt><dd><dl><dt><span class="section"><a href="#git-explained">Git explained</a></span></dt></dl></dd><dt><span class="appendix"><a href="#git-quick-start">A. Git Quick Reference</a></span></dt><dd><dl><dt><span class="section"><a href="#quick-creating-a-new-repository">Creating a new repository</a></span></dt><dt><span class="section"><a href="#managing-branches">Managing branches</a></span></dt><dt><span class="section"><a href="#exploring-history">Exploring history</a></span></dt><dt><span class="section"><a href="#making-changes">Making changes</a></span></dt><dt><span class="section"><a href="#merging">Merging</a></span></dt><dt><span class="section"><a href="#sharing-your-changes">Sharing your changes</a></span></dt><dt><span class="section"><a href="#repository-maintenance">Repository maintenance</a></span></dt></dl></dd><dt><span class="appendix"><a href="#todo">B. Notes and todo list for this manual</a></span></dt><dd><dl><dt><span class="section"><a href="#todo-list">Todo list</a></span></dt></dl></dd></dl></div><div class="preface"><div class="titlepage"><div><div><h1 class="title"><a name="_introduction"></a>Introduction</h1></div></div></div><p>Git is a fast distributed revision control system.</p><p>This manual is designed to be readable by someone with basic UNIX
3 command-line skills, but no previous knowledge of Git.</p><p><a class="xref" href="#repositories-and-branches" title="Chapter 1. Repositories and Branches">Chapter 1, <i>Repositories and Branches</i></a> and <a class="xref" href="#exploring-git-history" title="Chapter 2. Exploring Git history">Chapter 2, <i>Exploring Git history</i></a> explain how
4 to fetch and study a project using git—read these chapters to learn how
5 to build and test a particular version of a software project, search for
6 regressions, and so on.</p><p>People needing to do actual development will also want to read
7 <a class="xref" href="#Developing-With-git" title="Chapter 3. Developing with Git">Chapter 3, <i>Developing with Git</i></a> and <a class="xref" href="#sharing-development" title="Chapter 4. Sharing development with others">Chapter 4, <i>Sharing development with others</i></a>.</p><p>Further chapters cover more specialized topics.</p><p>Comprehensive reference documentation is available through the man
8 pages, or <a class="ulink" href="git-help.html" target="_top">git-help(1)</a> command. For example, for the command
9 <code class="literal">git clone <repo></code>, you can either use:</p><pre class="screen">$ man git-clone</pre><p>or:</p><pre class="screen">$ git help clone</pre><p>With the latter, you can use the manual viewer of your choice; see
10 <a class="ulink" href="git-help.html" target="_top">git-help(1)</a> for more information.</p><p>See also <a class="xref" href="#git-quick-start" title="Appendix A. Git Quick Reference">Appendix A, <i>Git Quick Reference</i></a> for a brief overview of Git commands,
11 without any explanation.</p><p>Finally, see <a class="xref" href="#todo" title="Appendix B. Notes and todo list for this manual">Appendix B, <i>Notes and todo list for this manual</i></a> for ways that you can help make this manual more
12 complete.</p></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="repositories-and-branches"></a>Chapter 1. Repositories and Branches</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#how-to-get-a-git-repository">How to get a Git repository</a></span></dt><dt><span class="section"><a href="#how-to-check-out">How to check out a different version of a project</a></span></dt><dt><span class="section"><a href="#understanding-commits">Understanding History: Commits</a></span></dt><dd><dl><dt><span class="section"><a href="#understanding-reachability">Understanding history: commits, parents, and reachability</a></span></dt><dt><span class="section"><a href="#history-diagrams">Understanding history: History diagrams</a></span></dt><dt><span class="section"><a href="#what-is-a-branch">Understanding history: What is a branch?</a></span></dt></dl></dd><dt><span class="section"><a href="#manipulating-branches">Manipulating branches</a></span></dt><dt><span class="section"><a href="#detached-head">Examining an old version without creating a new branch</a></span></dt><dt><span class="section"><a href="#examining-remote-branches">Examining branches from a remote repository</a></span></dt><dt><span class="section"><a href="#how-git-stores-references">Naming branches, tags, and other references</a></span></dt><dt><span class="section"><a href="#Updating-a-repository-With-git-fetch">Updating a repository with git fetch</a></span></dt><dt><span class="section"><a href="#fetching-branches">Fetching branches from other repositories</a></span></dt></dl></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="how-to-get-a-git-repository"></a>How to get a Git repository</h2></div></div></div><p>It will be useful to have a Git repository to experiment with as you
13 read this manual.</p><p>The best way to get one is by using the <a class="ulink" href="git-clone.html" target="_top">git-clone(1)</a> command to
14 download a copy of an existing repository. If you don’t already have a
15 project in mind, here are some interesting examples:</p><pre class="screen"> # Git itself (approx. 40MB download):
16 $ git clone git://git.kernel.org/pub/scm/git/git.git
17 # the Linux kernel (approx. 640MB download):
18 $ git clone git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git</pre><p>The initial clone may be time-consuming for a large project, but you
19 will only need to clone once.</p><p>The clone command creates a new directory named after the project
20 (<code class="literal">git</code> or <code class="literal">linux</code> in the examples above). After you cd into this
21 directory, you will see that it contains a copy of the project files,
24 about the history of the project.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="how-to-check-out"></a>How to check out a different version of a project</h2></div></div></div><p>Git is best thought of as a tool for storing the history of a collection
25 of files. It stores the history as a compressed collection of
26 interrelated snapshots of the project’s contents. In Git each such
27 version is called a <a class="link" href="#def_commit">commit</a>.</p><p>Those snapshots aren’t necessarily all arranged in a single line from
28 oldest to newest; instead, work may simultaneously proceed along
30 merge and diverge.</p><p>A single Git repository can track development on multiple branches. It
32 latest commit on each branch; the <a class="ulink" href="git-branch.html" target="_top">git-branch(1)</a> command shows
34 * master</pre><p>A freshly cloned repository contains a single branch head, by default
35 named "master", with the working directory initialized to the state of
36 the project referred to by that branch head.</p><p>Most projects also use <a class="link" href="#def_tag">tags</a>. Tags, like heads, are
37 references into the project’s history, and can be listed using the
38 <a class="ulink" href="git-tag.html" target="_top">git-tag(1)</a> command:</p><pre class="screen">$ git tag -l
39 v2.6.11
40 v2.6.11-tree
41 v2.6.12
42 v2.6.12-rc2
43 v2.6.12-rc3
44 v2.6.12-rc4
45 v2.6.12-rc5
46 v2.6.12-rc6
47 v2.6.13
48 ...</pre><p>Tags are expected to always point at the same version of a project,
49 while heads are expected to advance as development progresses.</p><p>Create a new branch head pointing to one of these versions and check it
50 out using <a class="ulink" href="git-switch.html" target="_top">git-switch(1)</a>:</p><pre class="screen">$ git switch -c new v2.6.13</pre><p>The working directory then reflects the contents that the project had
51 when it was tagged v2.6.13, and <a class="ulink" href="git-branch.html" target="_top">git-branch(1)</a> shows two
52 branches, with an asterisk marking the currently checked-out branch:</p><pre class="screen">$ git branch
53 master
55 the current branch to point at v2.6.17 instead, with</p><pre class="screen">$ git reset --hard v2.6.17</pre><p>Note that if the current branch head was your only reference to a
56 particular point in history, then resetting that branch may leave you
57 with no way to find the history it used to point to; so use this command
58 carefully.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="understanding-commits"></a>Understanding History: Commits</h2></div></div></div><p>Every change in the history of a project is represented by a commit.
59 The <a class="ulink" href="git-show.html" target="_top">git-show(1)</a> command shows the most recent commit on the
61 commit 17cf781661e6d38f737f15f53ab552f1e95960d7
65 Remove duplicate getenv(DB_ENVIRONMENT) call
67 Noted by Tony Luck.
69 diff --git a/init-db.c b/init-db.c
71 --- a/init-db.c
72 +++ b/init-db.c
75 int main(int argc, char **argv)
76 {
77 - char *sha1_dir = getenv(DB_ENVIRONMENT), *path;
78 + char *sha1_dir, *path;
79 int len, i;
81 if (mkdir(".git", 0755) < 0) {</pre><p>As you can see, a commit shows who made the latest change, what they
83 "SHA-1 id", shown on the first line of the <code class="literal">git show</code> output. You can usually
84 refer to a commit by a shorter name, such as a tag or a branch name, but this
85 longer name can also be useful. Most importantly, it is a globally unique
86 name for this commit: so if you tell somebody else the object name (for
87 example in email), then you are guaranteed that name will refer to the same
88 commit in their repository that it does in yours (assuming their repository
89 has that commit at all). Since the object name is computed as a hash over the
90 contents of the commit, you are guaranteed that the commit can never change
91 without its name also changing.</p><p>In fact, in <a class="xref" href="#git-concepts" title="Chapter 7. Git concepts">Chapter 7, <i>Git concepts</i></a> we shall see that everything stored in Git
92 history, including file data and directory contents, is stored in an object
93 with a name that is a hash of its contents.</p><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="understanding-reachability"></a>Understanding history: commits, parents, and reachability</h3></div></div></div><p>Every commit (except the very first commit in a project) also has a
94 parent commit which shows what happened before this commit.
95 Following the chain of parents will eventually take you back to the
96 beginning of the project.</p><p>However, the commits do not form a simple list; Git allows lines of
97 development to diverge and then reconverge, and the point where two
98 lines of development reconverge is called a "merge". The commit
99 representing a merge can therefore have more than one parent, with
100 each parent representing the most recent commit on one of the lines
101 of development leading to that point.</p><p>The best way to see how this works is using the <a class="ulink" href="gitk.html" target="_top">gitk(1)</a>
102 command; running gitk now on a Git repository and looking for merge
103 commits will help understand how Git organizes history.</p><p>In the following, we say that commit X is "reachable" from commit Y
104 if commit X is an ancestor of commit Y. Equivalently, you could say
105 that Y is a descendant of X, or that there is a chain of parents
106 leading from commit Y to commit X.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="history-diagrams"></a>Understanding history: History diagrams</h3></div></div></div><p>We will sometimes represent Git history using diagrams like the one
107 below. Commits are shown as "o", and the links between them with
108 lines drawn with - / and \. Time goes left to right:</p><pre class="literallayout"> o--o--o <-- Branch A
109 /
110 o--o--o <-- master
111 \
112 o--o--o <-- Branch B</pre><p>If we need to talk about a particular commit, the character "o" may
113 be replaced with another letter or number.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="what-is-a-branch"></a>Understanding history: What is a branch?</h3></div></div></div><p>When we need to be precise, we will use the word "branch" to mean a line
115 to the most recent commit on a branch. In the example above, the branch
116 head named "A" is a pointer to one particular commit, but we refer to
117 the line of three commits leading up to that point as all being part of
119 "branch" both for branches and for branch heads.</p></div></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="manipulating-branches"></a>Manipulating branches</h2></div></div></div><p>Creating, deleting, and modifying branches is quick and easy; here’s
120 a summary of the commands:</p><div class="variablelist"><dl class="variablelist"><dt><span class="term">
122 </span></dt><dd>
123 list all branches.
126 </span></dt><dd>
128 point in history as the current branch.
131 </span></dt><dd>
134 including using a branch name or a tag name.
137 </span></dt><dd>
139 merged in its upstream branch or contained in the current branch,
140 this command will fail with a warning.
143 </span></dt><dd>
144 delete the branch <code class="literal"><branch></code> irrespective of its merged status.
147 </span></dt><dd>
152 </span></dt><dd>
153 create a new branch <code class="literal"><new></code> referencing <code class="literal"><start-point></code>, and
154 check it out.
156 branch. In fact, Git uses a file named <code class="literal">HEAD</code> in the <code class="literal">.git</code> directory
158 ref: refs/heads/master</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="detached-head"></a>Examining an old version without creating a new branch</h2></div></div></div><p>The <code class="literal">git switch</code> command normally expects a branch head, but will also
159 accept an arbitrary commit when invoked with --detach; for example,
160 you can check out the commit referenced by a tag:</p><pre class="screen">$ git switch --detach v2.6.17
161 Note: checking out 'v2.6.17'.
163 You are in 'detached HEAD' state. You can look around, make experimental
164 changes and commit them, and you can discard any commits you make in this
165 state without impacting any branches by performing another switch.
167 If you want to create a new branch to retain commits you create, you may
168 do so (now or later) by using -c with the switch command again. Example:
170 git switch -c new_branch_name
172 HEAD is now at 427abfa Linux v2.6.17</pre><p>The HEAD then refers to the SHA-1 of the commit instead of to a branch,
174 427abfa28afedffadfca9dd8b067eb6d36bac53f
175 $ git branch
176 * (detached from v2.6.17)
177 master</pre><p>In this case we say that the HEAD is "detached".</p><p>This is an easy way to check out a particular version without having to
178 make up a name for the new branch. You can still create a new branch
179 (or tag) for this version later if you decide to.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="examining-remote-branches"></a>Examining branches from a remote repository</h2></div></div></div><p>The "master" branch that was created at the time you cloned is a copy
180 of the HEAD in the repository that you cloned from. That repository
181 may also have had other branches, though, and your local repository
182 keeps branches which track each of those remote branches, called
183 remote-tracking branches, which you
184 can view using the <code class="literal">-r</code> option to <a class="ulink" href="git-branch.html" target="_top">git-branch(1)</a>:</p><pre class="screen">$ git branch -r
185 origin/HEAD
186 origin/html
187 origin/maint
188 origin/man
189 origin/master
190 origin/next
191 origin/seen
193 for short. The branches of this repository are called "remote
194 branches" from our point of view. The remote-tracking branches listed
195 above were created based on the remote branches at clone time and will
196 be updated by <code class="literal">git fetch</code> (hence <code class="literal">git pull</code>) and <code class="literal">git push</code>. See
197 <a class="xref" href="#Updating-a-repository-With-git-fetch" title="Updating a repository with git fetch">the section called “Updating a repository with git fetch”</a> for details.</p><p>You might want to build on one of these remote-tracking branches
198 on a branch of your own, just as you would for a tag:</p><pre class="screen">$ git switch -c my-todo-copy origin/todo</pre><p>You can also check out <code class="literal">origin/todo</code> directly to examine it or
199 write a one-off patch. See <a class="link" href="#detached-head" title="Examining an old version without creating a new branch">detached head</a>.</p><p>Note that the name "origin" is just the name that Git uses by default
200 to refer to the repository that you cloned from.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="how-git-stores-references"></a>Naming branches, tags, and other references</h2></div></div></div><p>Branches, remote-tracking branches, and tags are all references to
201 commits. All references are named with a slash-separated path name
203 shorthand:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
204 The branch <code class="literal">test</code> is short for <code class="literal">refs/heads/test</code>.
206 The tag <code class="literal">v2.6.18</code> is short for <code class="literal">refs/tags/v2.6.18</code>.
208 <code class="literal">origin/master</code> is short for <code class="literal">refs/remotes/origin/master</code>.
209 </li></ul></div><p>The full name is occasionally useful if, for example, there ever
210 exists a tag and a branch with the same name.</p><p>(Newly created refs are actually stored in the <code class="literal">.git/refs</code> directory,
211 under the path given by their name. However, for efficiency reasons
212 they may also be packed together in a single file; see
213 <a class="ulink" href="git-pack-refs.html" target="_top">git-pack-refs(1)</a>).</p><p>As another useful shortcut, the "HEAD" of a repository can be referred
214 to just using the name of that repository. So, for example, "origin"
215 is usually a shortcut for the HEAD branch in the repository "origin".</p><p>For the complete list of paths which Git checks for references, and
216 the order it uses to decide which to choose when there are multiple
217 references with the same shorthand name, see the "SPECIFYING
218 REVISIONS" section of <a class="ulink" href="gitrevisions.html" target="_top">gitrevisions(7)</a>.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="Updating-a-repository-With-git-fetch"></a>Updating a repository with git fetch</h2></div></div></div><p>After you clone a repository and commit a few changes of your own, you
219 may wish to check the original repository for updates.</p><p>The <code class="literal">git-fetch</code> command, with no arguments, will update all of the
220 remote-tracking branches to the latest version found in the original
221 repository. It will not touch any of your own branches—not even the
222 "master" branch that was created for you on clone.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="fetching-branches"></a>Fetching branches from other repositories</h2></div></div></div><p>You can also track branches from repositories other than the one you
223 cloned from, using <a class="ulink" href="git-remote.html" target="_top">git-remote(1)</a>:</p><pre class="screen">$ git remote add staging git://git.kernel.org/.../gregkh/staging.git
224 $ git fetch staging
225 ...
226 From git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
227 * [new branch] master -> staging/master
228 * [new branch] staging-linus -> staging/staging-linus
229 * [new branch] staging-next -> staging/staging-next</pre><p>New remote-tracking branches will be stored under the shorthand name
230 that you gave <code class="literal">git remote add</code>, in this case <code class="literal">staging</code>:</p><pre class="screen">$ git branch -r
231 origin/HEAD -> origin/master
232 origin/master
233 staging/master
234 staging/staging-linus
235 staging/staging-next</pre><p>If you run <code class="literal">git fetch <remote></code> later, the remote-tracking branches
236 for the named <code class="literal"><remote></code> will be updated.</p><p>If you examine the file <code class="literal">.git/config</code>, you will see that Git has added
238 ...
239 [remote "staging"]
240 url = git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git
241 fetch = +refs/heads/*:refs/remotes/staging/*
242 ...</pre><p>This is what causes Git to track the remote’s branches; you may modify
243 or delete these configuration options by editing <code class="literal">.git/config</code> with a
244 text editor. (See the "CONFIGURATION FILE" section of
245 <a class="ulink" href="git-config.html" target="_top">git-config(1)</a> for details.)</p></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="exploring-git-history"></a>Chapter 2. Exploring Git history</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#using-bisect">How to use bisect to find a regression</a></span></dt><dt><span class="section"><a href="#naming-commits">Naming commits</a></span></dt><dt><span class="section"><a href="#creating-tags">Creating tags</a></span></dt><dt><span class="section"><a href="#browsing-revisions">Browsing revisions</a></span></dt><dt><span class="section"><a href="#generating-diffs">Generating diffs</a></span></dt><dt><span class="section"><a href="#viewing-old-file-versions">Viewing old file versions</a></span></dt><dt><span class="section"><a href="#history-examples">Examples</a></span></dt><dd><dl><dt><span class="section"><a href="#counting-commits-on-a-branch">Counting the number of commits on a branch</a></span></dt><dt><span class="section"><a href="#checking-for-equal-branches">Check whether two branches point at the same history</a></span></dt><dt><span class="section"><a href="#finding-tagged-descendants">Find first tagged version including a given fix</a></span></dt><dt><span class="section"><a href="#showing-commits-unique-to-a-branch">Showing commits unique to a given branch</a></span></dt><dt><span class="section"><a href="#making-a-release">Creating a changelog and tarball for a software release</a></span></dt><dt><span class="section"><a href="#Finding-commits-With-given-Content">Finding commits referencing a file with given content</a></span></dt></dl></dd></dl></div><p>Git is best thought of as a tool for storing the history of a
246 collection of files. It does this by storing compressed snapshots of
247 the contents of a file hierarchy, together with "commits" which show
248 the relationships between these snapshots.</p><p>Git provides extremely flexible and fast tools for exploring the
249 history of a project.</p><p>We start with one specialized tool that is useful for finding the
250 commit that introduced a bug into a project.</p><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="using-bisect"></a>How to use bisect to find a regression</h2></div></div></div><p>Suppose version 2.6.18 of your project worked, but the version at
251 "master" crashes. Sometimes the best way to find the cause of such a
252 regression is to perform a brute-force search through the project’s
253 history to find the particular commit that caused the problem. The
254 <a class="ulink" href="git-bisect.html" target="_top">git-bisect(1)</a> command can help you do this:</p><pre class="screen">$ git bisect start
255 $ git bisect good v2.6.18
256 $ git bisect bad master
257 Bisecting: 3537 revisions left to test after this
258 [65934a9a028b88e83e2b0f8b36618fe503349f8e] BLOCK: Make USB storage depend on SCSI rather than selecting it [try #6]</pre><p>If you run <code class="literal">git branch</code> at this point, you’ll see that Git has
259 temporarily moved you in "(no branch)". HEAD is now detached from any
260 branch and points directly to a commit (with commit id 65934) that
263 Bisecting: 1769 revisions left to test after this
264 [7eff82c8b1511017ae605f0c99ac275a7e21b867] i2c-core: Drop useless bitmaskings</pre><p>checks out an older version. Continue like this, telling Git at each
265 stage whether the version it gives you is good or bad, and notice
266 that the number of revisions left to test is cut approximately in
268 the guilty commit. You can then examine the commit with
269 <a class="ulink" href="git-show.html" target="_top">git-show(1)</a>, find out who wrote it, and mail them your bug
270 report with the commit id. Finally, run</p><pre class="screen">$ git bisect reset</pre><p>to return you to the branch you were on before.</p><p>Note that the version which <code class="literal">git bisect</code> checks out for you at each
271 point is just a suggestion, and you’re free to try a different
272 version if you think it would be a good idea. For example,
273 occasionally you may land on a commit that broke something unrelated;
274 run</p><pre class="screen">$ git bisect visualize</pre><p>which will run gitk and label the commit it chose with a marker that
275 says "bisect". Choose a safe-looking commit nearby, note its commit
276 id, and check it out with:</p><pre class="screen">$ git reset --hard fb47ddb2db</pre><p>then test, run <code class="literal">bisect good</code> or <code class="literal">bisect bad</code> as appropriate, and
277 continue.</p><p>Instead of <code class="literal">git bisect visualize</code> and then <code class="literal">git reset --hard
278 fb47ddb2db</code>, you might just want to tell Git that you want to skip
279 the current commit:</p><pre class="screen">$ git bisect skip</pre><p>In this case, though, Git may not eventually be able to tell the first
280 bad one between some first skipped commits and a later bad commit.</p><p>There are also ways to automate the bisecting process if you have a
281 test script that can tell a good from a bad commit. See
282 <a class="ulink" href="git-bisect.html" target="_top">git-bisect(1)</a> for more information about this and other <code class="literal">git
283 bisect</code> features.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="naming-commits"></a>Naming commits</h2></div></div></div><p>We have seen several ways of naming commits already:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
284 40-hexdigit object name
286 branch name: refers to the commit at the head of the given
287 branch
289 tag name: refers to the commit pointed to by the given tag
290 (we’ve seen branches and tags are special cases of
291 <a class="link" href="#how-git-stores-references" title="Naming branches, tags, and other references">references</a>).
293 HEAD: refers to the head of the current branch
295 <a class="ulink" href="gitrevisions.html" target="_top">gitrevisions(7)</a> man page for the complete list of ways to
296 name revisions. Some examples:</p><pre class="screen">$ git show fb47ddb2 # the first few characters of the object name
297 # are usually enough to specify it uniquely
298 $ git show HEAD^ # the parent of the HEAD commit
299 $ git show HEAD^^ # the grandparent
300 $ git show HEAD~4 # the great-great-grandparent</pre><p>Recall that merge commits may have more than one parent; by default,
301 <code class="literal">^</code> and <code class="literal">~</code> follow the first parent listed in the commit, but you can
303 $ git show HEAD^2 # show the second parent of HEAD</pre><p>In addition to HEAD, there are several other special names for
304 commits:</p><p>Merges (to be discussed later), as well as operations such as
305 <code class="literal">git reset</code>, which change the currently checked-out commit, generally
306 set ORIG_HEAD to the value HEAD had before the current operation.</p><p>The <code class="literal">git fetch</code> operation always stores the head of the last fetched
308 specifying a local branch as the target of the operation</p><pre class="screen">$ git fetch git://example.com/proj.git theirbranch</pre><p>the fetched commits will still be available from FETCH_HEAD.</p><p>When we discuss merges we’ll also see the special name MERGE_HEAD,
309 which refers to the other branch that we’re merging in to the current
310 branch.</p><p>The <a class="ulink" href="git-rev-parse.html" target="_top">git-rev-parse(1)</a> command is a low-level command that is
311 occasionally useful for translating some name for a commit to the object
313 e05db0fd4f31dde7005f075a84f96b360d05984b</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="creating-tags"></a>Creating tags</h2></div></div></div><p>We can also create a tag to refer to a particular commit; after
314 running</p><pre class="screen">$ git tag stable-1 1b2e1d63ff</pre><p>You can use <code class="literal">stable-1</code> to refer to the commit 1b2e1d63ff.</p><p>This creates a "lightweight" tag. If you would also like to include a
315 comment with the tag, and possibly sign it cryptographically, then you
316 should create a tag object instead; see the <a class="ulink" href="git-tag.html" target="_top">git-tag(1)</a> man page
317 for details.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="browsing-revisions"></a>Browsing revisions</h2></div></div></div><p>The <a class="ulink" href="git-log.html" target="_top">git-log(1)</a> command can show lists of commits. On its
318 own, it shows all commits reachable from the parent commit; but you
319 can also make more specific requests:</p><pre class="screen">$ git log v2.5.. # commits since (not reachable from) v2.5
320 $ git log test..master # commits reachable from master but not test
321 $ git log master..test # ...reachable from test but not master
322 $ git log master...test # ...reachable from either test or master,
323 # but not both
325 $ git log Makefile # commits which modify Makefile
326 $ git log fs/ # ... which modify any file under fs/
327 $ git log -S'foo()' # commits which add or remove any file data
328 # matching the string 'foo()'</pre><p>And of course you can combine all of these; the following finds
329 commits since v2.5 which touch the <code class="literal">Makefile</code> or any file under <code class="literal">fs</code>:</p><pre class="screen">$ git log v2.5.. Makefile fs/</pre><p>You can also ask git log to show patches:</p><pre class="screen">$ git log -p</pre><p>See the <code class="literal">--pretty</code> option in the <a class="ulink" href="git-log.html" target="_top">git-log(1)</a> man page for more
330 display options.</p><p>Note that git log starts with the most recent commit and works
331 backwards through the parents; however, since Git history can contain
332 multiple independent lines of development, the particular order that
333 commits are listed in may be somewhat arbitrary.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="generating-diffs"></a>Generating diffs</h2></div></div></div><p>You can generate diffs between any two versions using
334 <a class="ulink" href="git-diff.html" target="_top">git-diff(1)</a>:</p><pre class="screen">$ git diff master..test</pre><p>That will produce the diff between the tips of the two branches. If
335 you’d prefer to find the diff from their common ancestor to test, you
336 can use three dots instead of two:</p><pre class="screen">$ git diff master...test</pre><p>Sometimes what you want instead is a set of patches; for this you can
337 use <a class="ulink" href="git-format-patch.html" target="_top">git-format-patch(1)</a>:</p><pre class="screen">$ git format-patch master..test</pre><p>will generate a file with a patch for each commit reachable from test
338 but not from master.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="viewing-old-file-versions"></a>Viewing old file versions</h2></div></div></div><p>You can always view an old version of a file by just checking out the
339 correct revision first. But sometimes it is more convenient to be
340 able to view an old version of a single file without checking
341 anything out; this command does that:</p><pre class="screen">$ git show v2.5:fs/locks.c</pre><p>Before the colon may be anything that names a commit, and after it
342 may be any path to a file tracked by Git.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="history-examples"></a>Examples</h2></div></div></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="counting-commits-on-a-branch"></a>Counting the number of commits on a branch</h3></div></div></div><p>Suppose you want to know how many commits you’ve made on <code class="literal">mybranch</code>
343 since it diverged from <code class="literal">origin</code>:</p><pre class="screen">$ git log --pretty=oneline origin..mybranch | wc -l</pre><p>Alternatively, you may often see this sort of thing done with the
344 lower-level command <a class="ulink" href="git-rev-list.html" target="_top">git-rev-list(1)</a>, which just lists the SHA-1’s
345 of all the given commits:</p><pre class="screen">$ git rev-list origin..mybranch | wc -l</pre></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="checking-for-equal-branches"></a>Check whether two branches point at the same history</h3></div></div></div><p>Suppose you want to check whether two branches point at the same point
346 in history.</p><pre class="screen">$ git diff origin..master</pre><p>will tell you whether the contents of the project are the same at the
347 two branches; in theory, however, it’s possible that the same project
348 contents could have been arrived at by two different historical
350 e05db0fd4f31dde7005f075a84f96b360d05984b
351 $ git rev-list master
352 e05db0fd4f31dde7005f075a84f96b360d05984b</pre><p>Or you could recall that the <code class="literal">...</code> operator selects all commits
353 reachable from either one reference or the other but not
354 both; so</p><pre class="screen">$ git log origin...master</pre><p>will return no commits when the two branches are equal.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="finding-tagged-descendants"></a>Find first tagged version including a given fix</h3></div></div></div><p>Suppose you know that the commit e05db0fd fixed a certain problem.
355 You’d like to find the earliest tagged release that contains that
356 fix.</p><p>Of course, there may be more than one answer—if the history branched
357 after commit e05db0fd, then there could be multiple "earliest" tagged
358 releases.</p><p>You could just visually inspect the commits since e05db0fd:</p><pre class="screen">$ gitk e05db0fd..</pre><p>or you can use <a class="ulink" href="git-name-rev.html" target="_top">git-name-rev(1)</a>, which will give the commit a
359 name based on any tag it finds pointing to one of the commit’s
361 e05db0fd tags/v1.5.0-rc1^0~23</pre><p>The <a class="ulink" href="git-describe.html" target="_top">git-describe(1)</a> command does the opposite, naming the
362 revision using a tag on which the given commit is based:</p><pre class="screen">$ git describe e05db0fd
363 v1.5.0-rc0-260-ge05db0f</pre><p>but that may sometimes help you guess which tags might come after the
364 given commit.</p><p>If you just want to verify whether a given tagged version contains a
365 given commit, you could use <a class="ulink" href="git-merge-base.html" target="_top">git-merge-base(1)</a>:</p><pre class="screen">$ git merge-base e05db0fd v1.5.0-rc1
366 e05db0fd4f31dde7005f075a84f96b360d05984b</pre><p>The merge-base command finds a common ancestor of the given commits,
367 and always returns one or the other in the case where one is a
368 descendant of the other; so the above output shows that e05db0fd
369 actually is an ancestor of v1.5.0-rc1.</p><p>Alternatively, note that</p><pre class="screen">$ git log v1.5.0-rc1..e05db0fd</pre><p>will produce empty output if and only if v1.5.0-rc1 includes e05db0fd,
370 because it outputs only commits that are not reachable from v1.5.0-rc1.</p><p>As yet another alternative, the <a class="ulink" href="git-show-branch.html" target="_top">git-show-branch(1)</a> command lists
371 the commits reachable from its arguments with a display on the left-hand
372 side that indicates which arguments that commit is reachable from.
373 So, if you run something like</p><pre class="screen">$ git show-branch e05db0fd v1.5.0-rc0 v1.5.0-rc1 v1.5.0-rc2
374 ! [e05db0fd] Fix warnings in sha1_file.c - use C99 printf format if
375 available
379 ...</pre><p>then a line like</p><pre class="screen">+ ++ [e05db0fd] Fix warnings in sha1_file.c - use C99 printf format if
381 and from v1.5.0-rc2, and not from v1.5.0-rc0.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="showing-commits-unique-to-a-branch"></a>Showing commits unique to a given branch</h3></div></div></div><p>Suppose you would like to see all the commits reachable from the branch
382 head named <code class="literal">master</code> but not from any other head in your repository.</p><p>We can list all the heads in this repository with
383 <a class="ulink" href="git-show-ref.html" target="_top">git-show-ref(1)</a>:</p><pre class="screen">$ git show-ref --heads
384 bf62196b5e363d73353a9dcf094c59595f3153b7 refs/heads/core-tutorial
385 db768d5504c1bb46f63ee9d6e1772bd047e05bf9 refs/heads/maint
386 a07157ac624b2524a059a3414e99f6f44bebc1e7 refs/heads/master
388 1e87486ae06626c2f31eaa63d26fc0fd646c8af2 refs/heads/tutorial-fixes</pre><p>We can get just the branch-head names, and remove <code class="literal">master</code>, with
389 the help of the standard utilities cut and grep:</p><pre class="screen">$ git show-ref --heads | cut -d' ' -f2 | grep -v '^refs/heads/master'
390 refs/heads/core-tutorial
391 refs/heads/maint
392 refs/heads/tutorial-2
393 refs/heads/tutorial-fixes</pre><p>And then we can ask to see all the commits reachable from master
394 but not from these other heads:</p><pre class="screen">$ gitk master --not $( git show-ref --heads | cut -d' ' -f2 |
395 grep -v '^refs/heads/master' )</pre><p>Obviously, endless variations are possible; for example, to see all
396 commits reachable from some head but not from any tag in the repository:</p><pre class="screen">$ gitk $( git show-ref --heads ) --not $( git show-ref --tags )</pre><p>(See <a class="ulink" href="gitrevisions.html" target="_top">gitrevisions(7)</a> for explanations of commit-selecting
397 syntax such as <code class="literal">--not</code>.)</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="making-a-release"></a>Creating a changelog and tarball for a software release</h3></div></div></div><p>The <a class="ulink" href="git-archive.html" target="_top">git-archive(1)</a> command can create a tar or zip archive from
398 any version of a project; for example:</p><pre class="screen">$ git archive -o latest.tar.gz --prefix=project/ HEAD</pre><p>will use HEAD to produce a gzipped tar archive in which each filename
400 the output file extension if possible, see <a class="ulink" href="git-archive.html" target="_top">git-archive(1)</a> for
401 details.</p><p>Versions of Git older than 1.7.7 don’t know about the <code class="literal">tar.gz</code> format,
402 you’ll need to use gzip explicitly:</p><pre class="screen">$ git archive --format=tar --prefix=project/ HEAD | gzip >latest.tar.gz</pre><p>If you’re releasing a new version of a software project, you may want
403 to simultaneously make a changelog to include in the release
404 announcement.</p><p>Linus Torvalds, for example, makes new kernel releases by tagging them,
405 then running:</p><pre class="screen">$ release-script 2.6.12 2.6.13-rc6 2.6.13-rc7</pre><p>where release-script is a shell script that looks like:</p><pre class="screen">#!/bin/sh
409 echo "# git tag v$new"
410 echo "git archive --prefix=linux-$new/ v$new | gzip -9 > ../linux-$new.tar.gz"
411 echo "git diff v$stable v$new | gzip -9 > ../patch-$new.gz"
412 echo "git log --no-merges v$new ^v$last > ../ChangeLog-$new"
413 echo "git shortlog --no-merges v$new ^v$last > ../ShortLog"
414 echo "git diff --stat --summary -M v$last v$new > ../diffstat-$new"</pre><p>and then he just cut-and-pastes the output commands after verifying that
415 they look OK.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="Finding-commits-With-given-Content"></a>Finding commits referencing a file with given content</h3></div></div></div><p>Somebody hands you a copy of a file, and asks which commits modified a
416 file such that it contained the given content either before or after the
417 commit. You can find out with this:</p><pre class="screen">$ git log --raw --abbrev=40 --pretty=oneline |
418 grep -B 1 `git hash-object filename`</pre><p>Figuring out why this works is left as an exercise to the (advanced)
419 student. The <a class="ulink" href="git-log.html" target="_top">git-log(1)</a>, <a class="ulink" href="git-diff-tree.html" target="_top">git-diff-tree(1)</a>, and
420 <a class="ulink" href="git-hash-object.html" target="_top">git-hash-object(1)</a> man pages may prove helpful.</p></div></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="Developing-With-git"></a>Chapter 3. Developing with Git</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#telling-git-your-name">Telling Git your name</a></span></dt><dt><span class="section"><a href="#creating-a-new-repository">Creating a new repository</a></span></dt><dt><span class="section"><a href="#how-to-make-a-commit">How to make a commit</a></span></dt><dt><span class="section"><a href="#creating-good-commit-messages">Creating good commit messages</a></span></dt><dt><span class="section"><a href="#ignoring-files">Ignoring files</a></span></dt><dt><span class="section"><a href="#how-to-merge">How to merge</a></span></dt><dt><span class="section"><a href="#resolving-a-merge">Resolving a merge</a></span></dt><dd><dl><dt><span class="section"><a href="#conflict-resolution">Getting conflict-resolution help during a merge</a></span></dt></dl></dd><dt><span class="section"><a href="#undoing-a-merge">Undoing a merge</a></span></dt><dt><span class="section"><a href="#fast-forwards">Fast-forward merges</a></span></dt><dt><span class="section"><a href="#fixing-mistakes">Fixing mistakes</a></span></dt><dd><dl><dt><span class="section"><a href="#reverting-a-commit">Fixing a mistake with a new commit</a></span></dt><dt><span class="section"><a href="#fixing-a-mistake-by-rewriting-history">Fixing a mistake by rewriting history</a></span></dt><dt><span class="section"><a href="#checkout-of-path">Checking out an old version of a file</a></span></dt><dt><span class="section"><a href="#interrupted-work">Temporarily setting aside work in progress</a></span></dt></dl></dd><dt><span class="section"><a href="#ensuring-good-performance">Ensuring good performance</a></span></dt><dt><span class="section"><a href="#ensuring-reliability">Ensuring reliability</a></span></dt><dd><dl><dt><span class="section"><a href="#checking-for-corruption">Checking the repository for corruption</a></span></dt><dt><span class="section"><a href="#recovering-lost-changes">Recovering lost changes</a></span></dt></dl></dd></dl></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="telling-git-your-name"></a>Telling Git your name</h2></div></div></div><p>Before creating any commits, you should introduce yourself to Git.
421 The easiest way to do so is to use <a class="ulink" href="git-config.html" target="_top">git-config(1)</a>:</p><pre class="screen">$ git config --global user.name 'Your Name Comes Here'
422 $ git config --global user.email 'you@yourdomain.example.com'</pre><p>Which will add the following to a file named <code class="literal">.gitconfig</code> in your
424 name = Your Name Comes Here
425 email = you@yourdomain.example.com</pre><p>See the "CONFIGURATION FILE" section of <a class="ulink" href="git-config.html" target="_top">git-config(1)</a> for
426 details on the configuration file. The file is plain text, so you can
427 also edit it with your favorite editor.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="creating-a-new-repository"></a>Creating a new repository</h2></div></div></div><p>Creating a new repository from scratch is very easy:</p><pre class="screen">$ mkdir project
428 $ cd project
429 $ git init</pre><p>If you have some initial content (say, a tarball):</p><pre class="screen">$ tar xzvf project.tar.gz
430 $ cd project
431 $ git init
432 $ git add . # include everything below ./ in the first commit:
433 $ git commit</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="how-to-make-a-commit"></a>How to make a commit</h2></div></div></div><p>Creating a new commit takes three steps:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem">
434 Making some changes to the working directory using your
435 favorite editor.
437 Telling Git about your changes.
439 Creating the commit using the content you told Git about
440 in step 2.
442 times as you want: in order to keep track of what you want committed
443 at step 3, Git maintains a snapshot of the tree’s contents in a
444 special staging area called "the index."</p><p>At the beginning, the content of the index will be identical to
446 the difference between the HEAD and the index, should therefore
447 produce no output at that point.</p><p>Modifying the index is easy:</p><p>To update the index with the contents of a new or modified file, use</p><pre class="screen">$ git add path/to/file</pre><p>To remove a file from the index and from the working tree, use</p><pre class="screen">$ git rm path/to/file</pre><p>After each step you can verify that</p><pre class="screen">$ git diff --cached</pre><p>always shows the difference between the HEAD and the index file—this
448 is what you’d commit if you created the commit now—and that</p><pre class="screen">$ git diff</pre><p>shows the difference between the working tree and the index file.</p><p>Note that <code class="literal">git add</code> always adds just the current contents of a file
449 to the index; further changes to the same file will be ignored unless
450 you run <code class="literal">git add</code> on the file again.</p><p>When you’re ready, just run</p><pre class="screen">$ git commit</pre><p>and Git will prompt you for a commit message and then create the new
451 commit. Check to make sure it looks like what you expected with</p><pre class="screen">$ git show</pre><p>As a special shortcut,</p><pre class="screen">$ git commit -a</pre><p>will update the index with any files that you’ve modified or removed
452 and create a commit, all in one step.</p><p>A number of commands are useful for keeping track of what you’re
453 about to commit:</p><pre class="screen">$ git diff --cached # difference between HEAD and the index; what
454 # would be committed if you ran "commit" now.
455 $ git diff # difference between the index file and your
456 # working directory; changes that would not
457 # be included if you ran "commit" now.
458 $ git diff HEAD # difference between HEAD and working tree; what
459 # would be committed if you ran "commit -a" now.
460 $ git status # a brief per-file summary of the above.</pre><p>You can also use <a class="ulink" href="git-gui.html" target="_top">git-gui(1)</a> to create commits, view changes in
461 the index and the working tree files, and individually select diff hunks
462 for inclusion in the index (by right-clicking on the diff hunk and
463 choosing "Stage Hunk For Commit").</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="creating-good-commit-messages"></a>Creating good commit messages</h2></div></div></div><p>Though not required, it’s a good idea to begin the commit message
464 with a single short (no more than 50 characters) line summarizing the
465 change, followed by a blank line and then a more thorough
466 description. The text up to the first blank line in a commit
467 message is treated as the commit title, and that title is used
468 throughout Git. For example, <a class="ulink" href="git-format-patch.html" target="_top">git-format-patch(1)</a> turns a
469 commit into email, and it uses the title on the Subject line and the
470 rest of the commit in the body.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="ignoring-files"></a>Ignoring files</h2></div></div></div><p>A project will often generate files that you do <span class="emphasis"><em>not</em></span> want to track with Git.
471 This typically includes files generated by a build process or temporary
472 backup files made by your editor. Of course, <span class="emphasis"><em>not</em></span> tracking files with Git
473 is just a matter of <span class="emphasis"><em>not</em></span> calling <code class="literal">git add</code> on them. But it quickly becomes
474 annoying to have these untracked files lying around; e.g. they make
475 <code class="literal">git add .</code> practically useless, and they keep showing up in the output of
476 <code class="literal">git status</code>.</p><p>You can tell Git to ignore certain files by creating a file called
477 <code class="literal">.gitignore</code> in the top level of your working directory, with contents
479 # Ignore any file named foo.txt.
480 foo.txt
481 # Ignore (generated) html files,
482 *.html
483 # except foo.html which is maintained by hand.
484 !foo.html
485 # Ignore objects and archives.
486 *.[oa]</pre><p>See <a class="ulink" href="gitignore.html" target="_top">gitignore(5)</a> for a detailed explanation of the syntax. You can
487 also place .gitignore files in other directories in your working tree, and they
488 will apply to those directories and their subdirectories. The <code class="literal">.gitignore</code>
489 files can be added to your repository like any other files (just run <code class="literal">git add
490 .gitignore</code> and <code class="literal">git commit</code>, as usual), which is convenient when the exclude
491 patterns (such as patterns matching build output files) would also make sense
492 for other users who clone your repository.</p><p>If you wish the exclude patterns to affect only certain repositories
493 (instead of every repository for a given project), you may instead put
494 them in a file in your repository named <code class="literal">.git/info/exclude</code>, or in any
496 Some Git commands can also take exclude patterns directly on the
497 command line. See <a class="ulink" href="gitignore.html" target="_top">gitignore(5)</a> for the details.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="how-to-merge"></a>How to merge</h2></div></div></div><p>You can rejoin two diverging branches of development using
498 <a class="ulink" href="git-merge.html" target="_top">git-merge(1)</a>:</p><pre class="screen">$ git merge branchname</pre><p>merges the development in the branch <code class="literal">branchname</code> into the current
499 branch.</p><p>A merge is made by combining the changes made in <code class="literal">branchname</code> and the
500 changes made up to the latest commit in your current branch since
501 their histories forked. The work tree is overwritten by the result of
502 the merge when this combining is done cleanly, or overwritten by a
503 half-merged results when this combining results in conflicts.
504 Therefore, if you have uncommitted changes touching the same files as
505 the ones impacted by the merge, Git will refuse to proceed. Most of
506 the time, you will want to commit your changes before you can merge,
507 and if you don’t, then <a class="ulink" href="git-stash.html" target="_top">git-stash(1)</a> can take these changes
508 away while you’re doing the merge, and reapply them afterwards.</p><p>If the changes are independent enough, Git will automatically complete
509 the merge and commit the result (or reuse an existing commit in case
510 of <a class="link" href="#fast-forwards" title="Fast-forward merges">fast-forward</a>, see below). On the other hand,
511 if there are conflicts—for example, if the same file is
512 modified in two different ways in the remote branch and the local
513 branch—then you are warned; the output may look something like this:</p><pre class="screen">$ git merge next
515 Auto-merged file.txt
516 CONFLICT (content): Merge conflict in file.txt
517 Automatic merge failed; fix conflicts and then commit the result.</pre><p>Conflict markers are left in the problematic files, and after
518 you resolve the conflicts manually, you can update the index
519 with the contents and run Git commit, as you normally would when
520 creating a new file.</p><p>If you examine the resulting commit using gitk, you will see that it
521 has two parents, one pointing to the top of the current branch, and
522 one to the top of the other branch.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="resolving-a-merge"></a>Resolving a merge</h2></div></div></div><p>When a merge isn’t resolved automatically, Git leaves the index and
523 the working tree in a special state that gives you all the
524 information you need to help resolve the merge.</p><p>Files with conflicts are marked specially in the index, so until you
525 resolve the problem and update the index, <a class="ulink" href="git-commit.html" target="_top">git-commit(1)</a> will
527 file.txt: needs merge</pre><p>Also, <a class="ulink" href="git-status.html" target="_top">git-status(1)</a> will list those files as "unmerged", and the
528 files with conflicts will have conflict markers added, like this:</p><pre class="screen"><<<<<<< HEAD:file.txt
529 Hello world
530 =======
531 Goodbye
532 >>>>>>> 77976da35a11db4580b80ae27e8d65caf5208086:file.txt</pre><p>All you need to do is edit the files to resolve the conflicts, and then</p><pre class="screen">$ git add file.txt
533 $ git commit</pre><p>Note that the commit message will already be filled in for you with
534 some information about the merge. Normally you can just use this
535 default message unchanged, but you may add additional commentary of
536 your own if desired.</p><p>The above is all you need to know to resolve a simple merge. But Git
537 also provides more information to help resolve conflicts:</p><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="conflict-resolution"></a>Getting conflict-resolution help during a merge</h3></div></div></div><p>All of the changes that Git was able to merge automatically are
538 already added to the index file, so <a class="ulink" href="git-diff.html" target="_top">git-diff(1)</a> shows only
540 diff --cc file.txt
542 --- a/file.txt
543 +++ b/file.txt
545 ++<<<<<<< HEAD:file.txt
546 +Hello world
547 ++=======
548 + Goodbye
549 ++>>>>>>> 77976da35a11db4580b80ae27e8d65caf5208086:file.txt</pre><p>Recall that the commit which will be committed after we resolve this
550 conflict will have two parents instead of the usual one: one parent
551 will be HEAD, the tip of the current branch; the other will be the
552 tip of the other branch, which is stored temporarily in MERGE_HEAD.</p><p>During the merge, the index holds three versions of each file. Each of
553 these three "file stages" represents a different version of the file:</p><pre class="screen">$ git show :1:file.txt # the file in a common ancestor of both branches
554 $ git show :2:file.txt # the version from HEAD.
555 $ git show :3:file.txt # the version from MERGE_HEAD.</pre><p>When you ask <a class="ulink" href="git-diff.html" target="_top">git-diff(1)</a> to show the conflicts, it runs a
556 three-way diff between the conflicted merge results in the work tree with
558 mixed (in other words, when a hunk’s merge results come only from stage 2,
559 that part is not conflicting and is not shown. Same for stage 3).</p><p>The diff above shows the differences between the working-tree version of
561 each line by a single <code class="literal">+</code> or <code class="literal">-</code>, it now uses two columns: the first
562 column is used for differences between the first parent and the working
563 directory copy, and the second for differences between the second parent
564 and the working directory copy. (See the "COMBINED DIFF FORMAT" section
565 of <a class="ulink" href="git-diff-files.html" target="_top">git-diff-files(1)</a> for a details of the format.)</p><p>After resolving the conflict in the obvious way (but before updating the
567 diff --cc file.txt
569 --- a/file.txt
570 +++ b/file.txt
572 - Hello world
573 -Goodbye
575 first parent, deleted "Goodbye" from the second parent, and added
576 "Goodbye world", which was previously absent from both.</p><p>Some special diff options allow diffing the working directory against
578 $ git diff --base file.txt # same as the above
580 $ git diff --ours file.txt # same as the above
582 $ git diff --theirs file.txt # same as the above.</pre><p>When using the <span class="emphasis"><em>ort</em></span> merge strategy (the default), before updating the working
583 tree with the result of the merge, Git writes a ref named AUTO_MERGE
584 reflecting the state of the tree it is about to write. Conflicted paths with
585 textual conflicts that could not be automatically merged are written to this
586 tree with conflict markers, just as in the working tree. AUTO_MERGE can thus be
587 used with <a class="ulink" href="git-diff.html" target="_top">git-diff(1)</a> to show the changes you’ve made so far to resolve
588 conflicts. Using the same example as above, after resolving the conflict we
590 diff --git a/file.txt b/file.txt
592 --- a/file.txt
593 +++ b/file.txt
595 -<<<<<<< HEAD:file.txt
596 -Hello world
597 -=======
598 -Goodbye
600 +Goodbye world</pre><p>Notice that the diff shows we deleted the conflict markers and both versions of
601 the content line, and wrote "Goodbye world" instead.</p><p>The <a class="ulink" href="git-log.html" target="_top">git-log(1)</a> and <a class="ulink" href="gitk.html" target="_top">gitk(1)</a> commands also provide special help
603 $ gitk --merge</pre><p>These will display all commits which exist only on HEAD or on
604 MERGE_HEAD, and which touch an unmerged file.</p><p>You may also use <a class="ulink" href="git-mergetool.html" target="_top">git-mergetool(1)</a>, which lets you merge the
605 unmerged files using external tools such as Emacs or kdiff3.</p><p>Each time you resolve the conflicts in a file and update the index:</p><pre class="screen">$ git add file.txt</pre><p>the different stages of that file will be "collapsed", after which
606 <code class="literal">git diff</code> will (by default) no longer show diffs for that file.</p></div></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="undoing-a-merge"></a>Undoing a merge</h2></div></div></div><p>If you get stuck and decide to just give up and throw the whole mess
607 away, you can always return to the pre-merge state with</p><pre class="screen">$ git merge --abort</pre><p>Or, if you’ve already committed the merge that you want to throw away,</p><pre class="screen">$ git reset --hard ORIG_HEAD</pre><p>However, this last command can be dangerous in some cases—never
608 throw away a commit you have already committed if that commit may
609 itself have been merged into another branch, as doing so may confuse
610 further merges.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="fast-forwards"></a>Fast-forward merges</h2></div></div></div><p>There is one special case not mentioned above, which is treated
611 differently. Normally, a merge results in a merge commit, with two
612 parents, one pointing at each of the two lines of development that
613 were merged.</p><p>However, if the current branch is an ancestor of the other—so every commit
614 present in the current branch is already contained in the other branch—then Git
615 just performs a "fast-forward"; the head of the current branch is moved forward
616 to point at the head of the merged-in branch, without any new commits being
617 created.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="fixing-mistakes"></a>Fixing mistakes</h2></div></div></div><p>If you’ve messed up the working tree, but haven’t yet committed your
618 mistake, you can return the entire working tree to the last committed
619 state with</p><pre class="screen">$ git restore --staged --worktree :/</pre><p>If you make a commit that you later wish you hadn’t, there are two
620 fundamentally different ways to fix the problem:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem">
621 You can create a new commit that undoes whatever was done
622 by the old commit. This is the correct thing if your
623 mistake has already been made public.
625 You can go back and modify the old commit. You should
626 never do this if you have already made the history public;
627 Git does not normally expect the "history" of a project to
628 change, and cannot correctly perform repeated merges from
629 a branch that has had its history changed.
630 </li></ol></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="reverting-a-commit"></a>Fixing a mistake with a new commit</h3></div></div></div><p>Creating a new commit that reverts an earlier change is very easy;
631 just pass the <a class="ulink" href="git-revert.html" target="_top">git-revert(1)</a> command a reference to the bad
632 commit; for example, to revert the most recent commit:</p><pre class="screen">$ git revert HEAD</pre><p>This will create a new commit which undoes the change in HEAD. You
633 will be given a chance to edit the commit message for the new commit.</p><p>You can also revert an earlier change, for example, the next-to-last:</p><pre class="screen">$ git revert HEAD^</pre><p>In this case Git will attempt to undo the old change while leaving
634 intact any changes made since then. If more recent changes overlap
635 with the changes to be reverted, then you will be asked to fix
636 conflicts manually, just as in the case of <a class="link" href="#resolving-a-merge" title="Resolving a merge">resolving a merge</a>.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="fixing-a-mistake-by-rewriting-history"></a>Fixing a mistake by rewriting history</h3></div></div></div><p>If the problematic commit is the most recent commit, and you have not
637 yet made that commit public, then you may just
638 <a class="link" href="#undoing-a-merge" title="Undoing a merge">destroy it using <code class="literal">git reset</code></a>.</p><p>Alternatively, you
639 can edit the working directory and update the index to fix your
640 mistake, just as if you were going to <a class="link" href="#how-to-make-a-commit" title="How to make a commit">create a new commit</a>, then run</p><pre class="screen">$ git commit --amend</pre><p>which will replace the old commit by a new commit incorporating your
641 changes, giving you a chance to edit the old commit message first.</p><p>Again, you should never do this to a commit that may already have
642 been merged into another branch; use <a class="ulink" href="git-revert.html" target="_top">git-revert(1)</a> instead in
643 that case.</p><p>It is also possible to replace commits further back in the history, but
644 this is an advanced topic to be left for
645 <a class="link" href="#cleaning-up-history" title="Chapter 5. Rewriting history and maintaining patch series">another chapter</a>.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="checkout-of-path"></a>Checking out an old version of a file</h3></div></div></div><p>In the process of undoing a previous bad change, you may find it
646 useful to check out an older version of a particular file using
647 <a class="ulink" href="git-restore.html" target="_top">git-restore(1)</a>. The command</p><pre class="screen">$ git restore --source=HEAD^ path/to/file</pre><p>replaces path/to/file by the contents it had in the commit HEAD^, and
648 also updates the index to match. It does not change branches.</p><p>If you just want to look at an old version of the file, without
649 modifying the working directory, you can do that with
650 <a class="ulink" href="git-show.html" target="_top">git-show(1)</a>:</p><pre class="screen">$ git show HEAD^:path/to/file</pre><p>which will display the given version of the file.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="interrupted-work"></a>Temporarily setting aside work in progress</h3></div></div></div><p>While you are in the middle of working on something complicated, you
651 find an unrelated but obvious and trivial bug. You would like to fix it
652 before continuing. You can use <a class="ulink" href="git-stash.html" target="_top">git-stash(1)</a> to save the current
653 state of your work, and after fixing the bug (or, optionally after doing
654 so on a different branch and then coming back), unstash the
655 work-in-progress changes.</p><pre class="screen">$ git stash push -m "work in progress for foo feature"</pre><p>This command will save your changes away to the <code class="literal">stash</code>, and
656 reset your working tree and the index to match the tip of your
657 current branch. Then you can make your fix as usual.</p><pre class="screen">... edit and test ...
658 $ git commit -a -m "blorpl: typofix"</pre><p>After that, you can go back to what you were working on with
659 <code class="literal">git stash pop</code>:</p><pre class="screen">$ git stash pop</pre></div></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="ensuring-good-performance"></a>Ensuring good performance</h2></div></div></div><p>On large repositories, Git depends on compression to keep the history
660 information from taking up too much space on disk or in memory. Some
661 Git commands may automatically run <a class="ulink" href="git-gc.html" target="_top">git-gc(1)</a>, so you don’t
662 have to worry about running it manually. However, compressing a large
664 to avoid automatic compression kicking in when it is not convenient.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="ensuring-reliability"></a>Ensuring reliability</h2></div></div></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="checking-for-corruption"></a>Checking the repository for corruption</h3></div></div></div><p>The <a class="ulink" href="git-fsck.html" target="_top">git-fsck(1)</a> command runs a number of self-consistency checks
665 on the repository, and reports on any problems. This may take some
667 dangling commit 7281251ddd2a61e38657c827739c57015671a6b3
668 dangling commit 2706a059f258c6b245f298dc4ff2ccd30ec21a63
669 dangling commit 13472b7c4b80851a1bc551779171dcb03655e9b5
670 dangling blob 218761f9d90712d37a9c5e36f406f92202db07eb
671 dangling commit bf093535a34a4d35731aa2bd90fe6b176302f14f
672 dangling commit 8e4bec7f2ddaa268bef999853c25755452100f8e
673 dangling tree d50bb86186bf27b681d25af89d3b5b68382e4085
674 dangling tree b24c2473f1fd3d91352a624795be026d64c8841f
675 ...</pre><p>You will see informational messages on dangling objects. They are objects
676 that still exist in the repository but are no longer referenced by any of
677 your branches, and can (and will) be removed after a while with <code class="literal">gc</code>.
678 You can run <code class="literal">git fsck --no-dangling</code> to suppress these messages, and still
679 view real errors.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="recovering-lost-changes"></a>Recovering lost changes</h3></div></div></div><div class="section"><div class="titlepage"><div><div><h4 class="title"><a name="reflogs"></a>Reflogs</h4></div></div></div><p>Say you modify a branch with <a class="link" href="#fixing-mistakes" title="Fixing mistakes"><code class="literal">git reset --hard</code></a>,
680 and then realize that the branch was the only reference you had to
682 previous values of each branch. So in this case you can still find the
683 old history using, for example,</p><pre class="screen">$ git log master@{1}</pre><p>This lists the commits reachable from the previous version of the
685 that accepts a commit, not just with <code class="literal">git log</code>. Some other examples:</p><pre class="screen">$ git show master@{2} # See where the branch pointed 2,
687 $ gitk master@{yesterday} # See where it pointed yesterday,
688 $ gitk master@{"1 week ago"} # ... or last week
689 $ git log --walk-reflogs master # show reflog entries for master</pre><p>A separate reflog is kept for the HEAD, so</p><pre class="screen">$ git show HEAD@{"1 week ago"}</pre><p>will show what HEAD pointed to one week ago, not what the current branch
690 pointed to one week ago. This allows you to see the history of what
692 pruned. See <a class="ulink" href="git-reflog.html" target="_top">git-reflog(1)</a> and <a class="ulink" href="git-gc.html" target="_top">git-gc(1)</a> to learn
693 how to control this pruning, and see the "SPECIFYING REVISIONS"
694 section of <a class="ulink" href="gitrevisions.html" target="_top">gitrevisions(7)</a> for details.</p><p>Note that the reflog history is very different from normal Git history.
695 While normal history is shared by every repository that works on the
696 same project, the reflog history is not shared: it tells you only about
697 how the branches in your local repository have changed over time.</p></div><div class="section"><div class="titlepage"><div><div><h4 class="title"><a name="dangling-object-recovery"></a>Examining dangling objects</h4></div></div></div><p>In some situations the reflog may not be able to save you. For example,
698 suppose you delete a branch, then realize you need the history it
699 contained. The reflog is also deleted; however, if you have not yet
700 pruned the repository, then you may still be able to find the lost
702 <a class="xref" href="#dangling-objects" title="Dangling objects">the section called “Dangling objects”</a> for the details.</p><pre class="screen">$ git fsck
703 dangling commit 7281251ddd2a61e38657c827739c57015671a6b3
704 dangling commit 2706a059f258c6b245f298dc4ff2ccd30ec21a63
705 dangling commit 13472b7c4b80851a1bc551779171dcb03655e9b5
706 ...</pre><p>You can examine
707 one of those dangling commits with, for example,</p><pre class="screen">$ gitk 7281251ddd --not --all</pre><p>which does what it sounds like: it says that you want to see the commit
708 history that is described by the dangling commit(s), but not the
709 history that is described by all your existing branches and tags. Thus
710 you get exactly the history reachable from that commit that is lost.
711 (And notice that it might not be just one commit: we only report the
712 "tip of the line" as being dangling, but there might be a whole deep
713 and complex commit history that was dropped.)</p><p>If you decide you want the history back, you can always create a new
714 reference pointing to it, for example, a new branch:</p><pre class="screen">$ git branch recovered-branch 7281251ddd</pre><p>Other types of dangling objects (blobs and trees) are also possible, and
715 dangling objects can arise in other situations.</p></div></div></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="sharing-development"></a>Chapter 4. Sharing development with others</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#getting-updates-With-git-pull">Getting updates with git pull</a></span></dt><dt><span class="section"><a href="#submitting-patches">Submitting patches to a project</a></span></dt><dt><span class="section"><a href="#importing-patches">Importing patches to a project</a></span></dt><dt><span class="section"><a href="#public-repositories">Public Git repositories</a></span></dt><dd><dl><dt><span class="section"><a href="#setting-up-a-public-repository">Setting up a public repository</a></span></dt><dt><span class="section"><a href="#exporting-via-git">Exporting a Git repository via the Git protocol</a></span></dt><dt><span class="section"><a href="#exporting-via-http">Exporting a git repository via HTTP</a></span></dt><dt><span class="section"><a href="#pushing-changes-to-a-public-repository">Pushing changes to a public repository</a></span></dt><dt><span class="section"><a href="#forcing-push">What to do when a push fails</a></span></dt><dt><span class="section"><a href="#setting-up-a-shared-repository">Setting up a shared repository</a></span></dt><dt><span class="section"><a href="#setting-up-gitweb">Allowing web browsing of a repository</a></span></dt></dl></dd><dt><span class="section"><a href="#how-to-get-a-git-repository-with-minimal-history">How to get a Git repository with minimal history</a></span></dt><dt><span class="section"><a href="#sharing-development-examples">Examples</a></span></dt><dd><dl><dt><span class="section"><a href="#maintaining-topic-branches">Maintaining topic branches for a Linux subsystem maintainer</a></span></dt></dl></dd></dl></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="getting-updates-With-git-pull"></a>Getting updates with git pull</h2></div></div></div><p>After you clone a repository and commit a few changes of your own, you
716 may wish to check the original repository for updates and merge them
717 into your own work.</p><p>We have already seen <a class="link" href="#Updating-a-repository-With-git-fetch" title="Updating a repository with git fetch">how to keep remote-tracking branches up to date</a> with <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a>,
718 and how to merge two branches. So you can merge in changes from the
720 $ git merge origin/master</pre><p>However, the <a class="ulink" href="git-pull.html" target="_top">git-pull(1)</a> command provides a way to do this in
721 one step:</p><pre class="screen">$ git pull origin master</pre><p>In fact, if you have <code class="literal">master</code> checked out, then this branch has been
723 origin repository. So often you can
724 accomplish the above with just a simple</p><pre class="screen">$ git pull</pre><p>This command will fetch changes from the remote branches to your
725 remote-tracking branches <code class="literal">origin/*</code>, and merge the default branch into
726 the current branch.</p><p>More generally, a branch that is created from a remote-tracking branch
727 will pull
728 by default from that branch. See the descriptions of the
729 <code class="literal">branch.<name>.remote</code> and <code class="literal">branch.<name>.merge</code> options in
730 <a class="ulink" href="git-config.html" target="_top">git-config(1)</a>, and the discussion of the <code class="literal">--track</code> option in
731 <a class="ulink" href="git-checkout.html" target="_top">git-checkout(1)</a>, to learn how to control these defaults.</p><p>In addition to saving you keystrokes, <code class="literal">git pull</code> also helps you by
732 producing a default commit message documenting the branch and
733 repository that you pulled from.</p><p>(But note that no such commit will be created in the case of a
734 <a class="link" href="#fast-forwards" title="Fast-forward merges">fast-forward</a>; instead, your branch will just be
735 updated to point to the latest commit from the upstream branch.)</p><p>The <code class="literal">git pull</code> command can also be given <code class="literal">.</code> as the "remote" repository,
736 in which case it just merges in a branch from the current repository; so
738 $ git merge branch</pre><p>are roughly equivalent.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="submitting-patches"></a>Submitting patches to a project</h2></div></div></div><p>If you just have a few changes, the simplest way to submit them may
739 just be to send them as patches in email:</p><p>First, use <a class="ulink" href="git-format-patch.html" target="_top">git-format-patch(1)</a>; for example:</p><pre class="screen">$ git format-patch origin</pre><p>will produce a numbered series of files in the current directory, one
740 for each patch in the current branch but not in <code class="literal">origin/HEAD</code>.</p><p><code class="literal">git format-patch</code> can include an initial "cover letter". You can insert
741 commentary on individual patches after the three dash line which
744 <code class="literal">git format-patch --notes</code> will include the commit’s notes in a similar
745 manner.</p><p>You can then import these into your mail client and send them by
746 hand. However, if you have a lot to send at once, you may prefer to
747 use the <a class="ulink" href="git-send-email.html" target="_top">git-send-email(1)</a> script to automate the process.
748 Consult the mailing list for your project first to determine
749 their requirements for submitting patches.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="importing-patches"></a>Importing patches to a project</h2></div></div></div><p>Git also provides a tool called <a class="ulink" href="git-am.html" target="_top">git-am(1)</a> (am stands for
750 "apply mailbox"), for importing such an emailed series of patches.
751 Just save all of the patch-containing messages, in order, into a
752 single mailbox file, say <code class="literal">patches.mbox</code>, then run</p><pre class="screen">$ git am -3 patches.mbox</pre><p>Git will apply each patch in order; if any conflicts are found, it
753 will stop, and you can fix the conflicts as described in
754 "<a class="link" href="#resolving-a-merge" title="Resolving a merge">Resolving a merge</a>". (The <code class="literal">-3</code> option tells
755 Git to perform a merge; if you would prefer it just to abort and
756 leave your tree and index untouched, you may omit that option.)</p><p>Once the index is updated with the results of the conflict
757 resolution, instead of creating a new commit, just run</p><pre class="screen">$ git am --continue</pre><p>and Git will create the commit for you and continue applying the
758 remaining patches from the mailbox.</p><p>The final result will be a series of commits, one for each patch in
759 the original mailbox, with authorship and commit log message each
760 taken from the message containing each patch.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="public-repositories"></a>Public Git repositories</h2></div></div></div><p>Another way to submit changes to a project is to tell the maintainer
761 of that project to pull the changes from your repository using
762 <a class="ulink" href="git-pull.html" target="_top">git-pull(1)</a>. In the section "<a class="link" href="#getting-updates-With-git-pull" title="Getting updates with git pull">Getting updates with <code class="literal">git pull</code></a>" we described this as a way to get
763 updates from the "main" repository, but it works just as well in the
764 other direction.</p><p>If you and the maintainer both have accounts on the same machine, then
765 you can just pull changes from each other’s repositories directly;
766 commands that accept repository URLs as arguments will also accept a
768 $ git pull /path/to/other/repository</pre><p>or an ssh URL:</p><pre class="screen">$ git clone ssh://yourhost/~you/repository</pre><p>For projects with few developers, or for synchronizing a few private
769 repositories, this may be all you need.</p><p>However, the more common way to do this is to maintain a separate public
770 repository (usually on a different host) for others to pull changes
771 from. This is usually more convenient, and allows you to cleanly
772 separate private work in progress from publicly visible work.</p><p>You will continue to do your day-to-day work in your personal
773 repository, but periodically "push" changes from your personal
774 repository into your public repository, allowing other developers to
775 pull from that repository. So the flow of changes, in a situation
776 where there is one other developer with a public repository, looks
778 your personal repo ------------------> your public repo
779 ^ |
780 | |
781 | you pull | they pull
782 | |
783 | |
784 | they push V
785 their public repo <------------------- their repo</pre><p>We explain how to do this in the following sections.</p><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="setting-up-a-public-repository"></a>Setting up a public repository</h3></div></div></div><p>Assume your personal repository is in the directory <code class="literal">~/proj</code>. We
786 first create a new clone of the repository and tell <code class="literal">git daemon</code> that it
788 $ touch proj.git/git-daemon-export-ok</pre><p>The resulting directory proj.git contains a "bare" git repository—it is
789 just the contents of the <code class="literal">.git</code> directory, without any files checked out
790 around it.</p><p>Next, copy <code class="literal">proj.git</code> to the server where you plan to host the
791 public repository. You can use scp, rsync, or whatever is most
792 convenient.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="exporting-via-git"></a>Exporting a Git repository via the Git protocol</h3></div></div></div><p>This is the preferred method.</p><p>If someone else administers the server, they should tell you what
794 appear at. You can then skip to the section
795 "<a class="link" href="#pushing-changes-to-a-public-repository" title="Pushing changes to a public repository">Pushing changes to a public repository</a>", below.</p><p>Otherwise, all you need to do is start <a class="ulink" href="git-daemon.html" target="_top">git-daemon(1)</a>; it will
796 listen on port 9418. By default, it will allow access to any directory
797 that looks like a Git directory and contains the magic file
799 arguments will further restrict the exports to those paths.</p><p>You can also run <code class="literal">git daemon</code> as an inetd service; see the
800 <a class="ulink" href="git-daemon.html" target="_top">git-daemon(1)</a> man page for details. (See especially the
801 examples section.)</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="exporting-via-http"></a>Exporting a git repository via HTTP</h3></div></div></div><p>The Git protocol gives better performance and reliability, but on a
802 host with a web server set up, HTTP exports may be simpler to set up.</p><p>All you need to do is place the newly created bare Git repository in
803 a directory that is exported by the web server, and make some
804 adjustments to give web clients some extra information they need:</p><pre class="screen">$ mv proj.git /home/you/public_html/proj.git
805 $ cd proj.git
806 $ git --bare update-server-info
807 $ mv hooks/post-update.sample hooks/post-update</pre><p>(For an explanation of the last two lines, see
808 <a class="ulink" href="git-update-server-info.html" target="_top">git-update-server-info(1)</a> and <a class="ulink" href="githooks.html" target="_top">githooks(5)</a>.)</p><p>Advertise the URL of <code class="literal">proj.git</code>. Anybody else should then be able to
809 clone or pull from that URL, for example with a command line like:</p><pre class="screen">$ git clone http://yourserver.com/~you/proj.git</pre><p>(See also
810 <a class="ulink" href="howto/setup-git-server-over-http.html" target="_top">setup-git-server-over-http</a>
811 for a slightly more sophisticated setup using WebDAV which also
812 allows pushing over HTTP.)</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="pushing-changes-to-a-public-repository"></a>Pushing changes to a public repository</h3></div></div></div><p>Note that the two techniques outlined above (exporting via
813 <a class="link" href="#exporting-via-http" title="Exporting a git repository via HTTP">http</a> or <a class="link" href="#exporting-via-git" title="Exporting a Git repository via the Git protocol">git</a>) allow other
814 maintainers to fetch your latest changes, but they do not allow write
815 access, which you will need to update the public repository with the
816 latest changes created in your private repository.</p><p>The simplest way to do this is using <a class="ulink" href="git-push.html" target="_top">git-push(1)</a> and ssh; to
817 update the remote branch named <code class="literal">master</code> with the latest state of your
818 branch named <code class="literal">master</code>, run</p><pre class="screen">$ git push ssh://yourserver.com/~you/proj.git master:master</pre><p>or just</p><pre class="screen">$ git push ssh://yourserver.com/~you/proj.git master</pre><p>As with <code class="literal">git fetch</code>, <code class="literal">git push</code> will complain if this does not result in a
819 <a class="link" href="#fast-forwards" title="Fast-forward merges">fast-forward</a>; see the following section for details on
820 handling this case.</p><p>Note that the target of a <code class="literal">push</code> is normally a
822 repository that has a checked-out working tree, but a push to update the
823 currently checked-out branch is denied by default to prevent confusion.
824 See the description of the receive.denyCurrentBranch option
825 in <a class="ulink" href="git-config.html" target="_top">git-config(1)</a> for details.</p><p>As with <code class="literal">git fetch</code>, you may also set up configuration options to
826 save typing; so, for example:</p><pre class="screen">$ git remote add public-repo ssh://yourserver.com/~you/proj.git</pre><p>adds the following to <code class="literal">.git/config</code>:</p><pre class="screen">[remote "public-repo"]
827 url = yourserver.com:proj.git
828 fetch = +refs/heads/*:refs/remotes/example/*</pre><p>which lets you do the same push with just</p><pre class="screen">$ git push public-repo master</pre><p>See the explanations of the <code class="literal">remote.<name>.url</code>,
829 <code class="literal">branch.<name>.remote</code>, and <code class="literal">remote.<name>.push</code> options in
830 <a class="ulink" href="git-config.html" target="_top">git-config(1)</a> for details.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="forcing-push"></a>What to do when a push fails</h3></div></div></div><p>If a push would not result in a <a class="link" href="#fast-forwards" title="Fast-forward merges">fast-forward</a> of the
831 remote branch, then it will fail with an error like:</p><pre class="screen"> ! [rejected] master -> master (non-fast-forward)
832 error: failed to push some refs to '...'
833 hint: Updates were rejected because the tip of your current branch is behind
834 hint: its remote counterpart. Integrate the remote changes (e.g.
835 hint: 'git pull ...') before pushing again.
836 hint: See the 'Note about fast-forwards' in 'git push --help' for details.</pre><p>This can happen, for example, if you:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
840 (as in <a class="xref" href="#fixing-a-mistake-by-rewriting-history" title="Fixing a mistake by rewriting history">the section called “Fixing a mistake by rewriting history”</a>), or
843 in <a class="xref" href="#using-git-rebase" title="Keeping a patch series up to date using git rebase">the section called “Keeping a patch series up to date using git rebase”</a>).
844 </li></ul></div><p>You may force <code class="literal">git push</code> to perform the update anyway by preceding the
845 branch name with a plus sign:</p><pre class="screen">$ git push ssh://yourserver.com/~you/proj.git +master</pre><p>Note the addition of the <code class="literal">+</code> sign. Alternatively, you can use the
846 <code class="literal">-f</code> flag to force the remote update, as in:</p><pre class="screen">$ git push -f ssh://yourserver.com/~you/proj.git master</pre><p>Normally whenever a branch head in a public repository is modified, it
847 is modified to point to a descendant of the commit that it pointed to
848 before. By forcing a push in this situation, you break that convention.
849 (See <a class="xref" href="#problems-With-rewriting-history" title="Problems with rewriting history">the section called “Problems with rewriting history”</a>.)</p><p>Nevertheless, this is a common practice for people that need a simple
850 way to publish a work-in-progress patch series, and it is an acceptable
851 compromise as long as you warn other developers that this is how you
852 intend to manage the branch.</p><p>It’s also possible for a push to fail in this way when other people have
853 the right to push to the same repository. In that case, the correct
854 solution is to retry the push after first updating your work: either by a
855 pull, or by a fetch followed by a rebase; see the
856 <a class="link" href="#setting-up-a-shared-repository" title="Setting up a shared repository">next section</a> and
857 <a class="ulink" href="gitcvs-migration.html" target="_top">gitcvs-migration(7)</a> for more.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="setting-up-a-shared-repository"></a>Setting up a shared repository</h3></div></div></div><p>Another way to collaborate is by using a model similar to that
858 commonly used in CVS, where several developers with special rights
859 all push to and pull from a single shared repository. See
860 <a class="ulink" href="gitcvs-migration.html" target="_top">gitcvs-migration(7)</a> for instructions on how to
861 set this up.</p><p>However, while there is nothing wrong with Git’s support for shared
862 repositories, this mode of operation is not generally recommended,
863 simply because the mode of collaboration that Git supports—by
864 exchanging patches and pulling from public repositories—has so many
865 advantages over the central shared repository:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
866 Git’s ability to quickly import and merge patches allows a
867 single maintainer to process incoming changes even at very
869 an easy way for that maintainer to delegate this job to other
870 maintainers while still allowing optional review of incoming
871 changes.
873 Since every developer’s repository has the same complete copy
874 of the project history, no repository is special, and it is
875 trivial for another developer to take over maintenance of a
876 project, either by mutual agreement, or because a maintainer
877 becomes unresponsive or difficult to work with.
879 The lack of a central group of "committers" means there is
880 less need for formal decisions about who is "in" and who is
881 "out".
882 </li></ul></div></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="setting-up-gitweb"></a>Allowing web browsing of a repository</h3></div></div></div><p>The gitweb cgi script provides users an easy way to browse your
883 project’s revisions, file contents and logs without having to install
884 Git. Features like RSS/Atom feeds and blame/annotation details may
885 optionally be enabled.</p><p>The <a class="ulink" href="git-instaweb.html" target="_top">git-instaweb(1)</a> command provides a simple way to start
886 browsing the repository using gitweb. The default server when using
887 instaweb is lighttpd.</p><p>See the file gitweb/INSTALL in the Git source tree and
888 <a class="ulink" href="gitweb.html" target="_top">gitweb(1)</a> for instructions on details setting up a permanent
889 installation with a CGI or Perl capable server.</p></div></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="how-to-get-a-git-repository-with-minimal-history"></a>How to get a Git repository with minimal history</h2></div></div></div><p>A <a class="link" href="#def_shallow_clone">shallow clone</a>, with its truncated
890 history, is useful when one is interested only in recent history
891 of a project and getting full history from the upstream is
892 expensive.</p><p>A <a class="link" href="#def_shallow_clone">shallow clone</a> is created by specifying
893 the <a class="ulink" href="git-clone.html" target="_top">git-clone(1)</a> <code class="literal">--depth</code> switch. The depth can later be
894 changed with the <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a> <code class="literal">--depth</code> switch, or full
895 history restored with <code class="literal">--unshallow</code>.</p><p>Merging inside a <a class="link" href="#def_shallow_clone">shallow clone</a> will work as long
896 as a merge base is in the recent history.
897 Otherwise, it will be like merging unrelated histories and may
898 have to result in huge conflicts. This limitation may make such
899 a repository unsuitable to be used in merge based workflows.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="sharing-development-examples"></a>Examples</h2></div></div></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="maintaining-topic-branches"></a>Maintaining topic branches for a Linux subsystem maintainer</h3></div></div></div><p>This describes how Tony Luck uses Git in his role as maintainer of the
900 IA64 architecture for the Linux kernel.</p><p>He uses two public branches:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
901 A "test" tree into which patches are initially placed so that they
902 can get some exposure when integrated with other ongoing development.
903 This tree is available to Andrew for pulling into -mm whenever he
904 wants.
906 A "release" tree into which tested patches are moved for final sanity
907 checking, and as a vehicle to send them upstream to Linus (by sending
908 him a "please pull" request.)
910 containing a logical grouping of patches.</p><p>To set this up, first create your work tree by cloning Linus’s public
911 tree:</p><pre class="screen">$ git clone git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git work
912 $ cd work</pre><p>Linus’s tree will be stored in the remote-tracking branch named origin/master,
913 and can be updated using <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a>; you can track other
914 public trees using <a class="ulink" href="git-remote.html" target="_top">git-remote(1)</a> to set up a "remote" and
915 <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a> to keep them up to date; see
916 <a class="xref" href="#repositories-and-branches" title="Chapter 1. Repositories and Branches">Chapter 1, <i>Repositories and Branches</i></a>.</p><p>Now create the branches in which you are going to work; these start out
917 at the current tip of origin/master branch, and should be set up (using
918 the <code class="literal">--track</code> option to <a class="ulink" href="git-branch.html" target="_top">git-branch(1)</a>) to merge changes in from
920 $ git branch --track release origin/master</pre><p>These can be easily kept up to date using <a class="ulink" href="git-pull.html" target="_top">git-pull(1)</a>.</p><pre class="screen">$ git switch test && git pull
921 $ git switch release && git pull</pre><p>Important note! If you have any local changes in these branches, then
922 this merge will create a commit object in the history (with no local
923 changes Git will simply do a "fast-forward" merge). Many people dislike
924 the "noise" that this creates in the Linux history, so you should avoid
925 doing this capriciously in the <code class="literal">release</code> branch, as these noisy commits
926 will become part of the permanent history when you ask Linus to pull
927 from the release branch.</p><p>A few configuration variables (see <a class="ulink" href="git-config.html" target="_top">git-config(1)</a>) can
928 make it easy to push both branches to your public tree. (See
929 <a class="xref" href="#setting-up-a-public-repository" title="Setting up a public repository">the section called “Setting up a public repository”</a>.)</p><pre class="screen">$ cat >> .git/config <<EOF
930 [remote "mytree"]
931 url = master.kernel.org:/pub/scm/linux/kernel/git/aegl/linux.git
932 push = release
933 push = test
934 EOF</pre><p>Then you can push both the test and release trees using
935 <a class="ulink" href="git-push.html" target="_top">git-push(1)</a>:</p><pre class="screen">$ git push mytree</pre><p>or push just one of the test and release branches using:</p><pre class="screen">$ git push mytree test</pre><p>or</p><pre class="screen">$ git push mytree release</pre><p>Now to apply some patches from the community. Think of a short
936 snappy name for a branch to hold this patch (or related group of
937 patches), and create a new branch from a recent stable tag of
938 Linus’s branch. Picking a stable base for your branch will:
939 1) help you: by avoiding inclusion of unrelated and perhaps lightly
940 tested changes
941 2) help future bug hunters that use <code class="literal">git bisect</code> to find problems</p><pre class="screen">$ git switch -c speed-up-spinlocks v2.6.35</pre><p>Now you apply the patch(es), run some tests, and commit the change(s). If
942 the patch is a multi-part series, then you should apply each as a separate
943 commit to this branch.</p><pre class="screen">$ ... patch ... test ... commit [ ... patch ... test ... commit ]*</pre><p>When you are happy with the state of this change, you can merge it into the
944 "test" branch in preparation to make it public:</p><pre class="screen">$ git switch test && git merge speed-up-spinlocks</pre><p>It is unlikely that you would have any conflicts here … but you might if you
945 spent a while on this step and had also pulled new versions from upstream.</p><p>Sometime later when enough time has passed and testing done, you can pull the
946 same branch into the <code class="literal">release</code> tree ready to go upstream. This is where you
947 see the value of keeping each patch (or patch series) in its own branch. It
948 means that the patches can be moved into the <code class="literal">release</code> tree in any order.</p><pre class="screen">$ git switch release && git merge speed-up-spinlocks</pre><p>After a while, you will have a number of branches, and despite the
949 well chosen names you picked for each of them, you may forget what
950 they are for, or what status they are in. To get a reminder of what
951 changes are in a specific branch, use:</p><pre class="screen">$ git log linux..branchname | git shortlog</pre><p>To see whether it has already been merged into the test or release branches,
952 use:</p><pre class="screen">$ git log test..branchname</pre><p>or</p><pre class="screen">$ git log release..branchname</pre><p>(If this branch has not yet been merged, you will see some log entries.
953 If it has been merged, then there will be no output.)</p><p>Once a patch completes the great cycle (moving from test to release,
954 then pulled by Linus, and finally coming back into your local
955 <code class="literal">origin/master</code> branch), the branch for this change is no longer needed.
956 You detect this when the output from:</p><pre class="screen">$ git log origin..branchname</pre><p>is empty. At this point the branch can be deleted:</p><pre class="screen">$ git branch -d branchname</pre><p>Some changes are so trivial that it is not necessary to create a separate
957 branch and then merge into each of the test and release branches. For
959 merge that into the <code class="literal">test</code> branch.</p><p>After pushing your work to <code class="literal">mytree</code>, you can use
960 <a class="ulink" href="git-request-pull.html" target="_top">git-request-pull(1)</a> to prepare a "please pull" request message
962 $ git request-pull origin mytree release</pre><p>Here are some of the scripts that simplify all this even further.</p><pre class="screen">==== update script ====
963 # Update a branch in my Git tree. If the branch to be updated
964 # is origin, then pull from kernel.org. Otherwise merge
965 # origin/master branch into test|release branch
967 case "$1" in
968 test|release)
970 ;;
971 origin)
972 before=$(git rev-parse refs/remotes/origin/master)
973 git fetch origin
974 after=$(git rev-parse refs/remotes/origin/master)
975 if [ $before != $after ]
976 then
977 git log $before..$after | git shortlog
978 fi
979 ;;
980 *)
982 exit 1
983 ;;
985 # Merge a branch into either the test or release branch
987 pname=$0
989 usage()
990 {
992 exit 1
993 }
995 git show-ref -q --verify -- refs/heads/"$1" || {
997 usage
998 }
1000 case "$2" in
1001 test|release)
1003 then
1005 exit 1
1006 fi
1008 ;;
1009 *)
1010 usage
1011 ;;
1013 # report on status of my ia64 Git tree
1015 gb=$(tput setab 2)
1016 rb=$(tput setab 1)
1017 restore=$(tput setab 9)
1019 if [ `git rev-list test..release | wc -c` -gt 0 ]
1020 then
1021 echo $rb Warning: commits in release that are not in test $restore
1022 git log test..release
1023 fi
1025 for branch in `git show-ref --heads | sed 's|^.*/||'`
1026 do
1027 if [ $branch = test -o $branch = release ]
1028 then
1029 continue
1030 fi
1032 echo -n $gb ======= $branch ====== $restore " "
1033 status=
1034 for ref in test release origin/master
1035 do
1036 if [ `git rev-list $ref..$branch | wc -c` -gt 0 ]
1037 then
1039 fi
1040 done
1041 case $status in
1042 trl)
1043 echo $rb Need to pull into test $restore
1044 ;;
1045 rl)
1046 echo "In test"
1047 ;;
1048 l)
1049 echo "Waiting for linus"
1050 ;;
1051 "")
1052 echo $rb All done $restore
1053 ;;
1054 *)
1055 echo $rb "<$status>" $restore
1056 ;;
1057 esac
1058 git log origin/master..$branch | git shortlog
1059 done</pre></div></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="cleaning-up-history"></a>Chapter 5. Rewriting history and maintaining patch series</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#patch-series">Creating the perfect patch series</a></span></dt><dt><span class="section"><a href="#using-git-rebase">Keeping a patch series up to date using git rebase</a></span></dt><dt><span class="section"><a href="#rewriting-one-commit">Rewriting a single commit</a></span></dt><dt><span class="section"><a href="#reordering-patch-series">Reordering or selecting from a patch series</a></span></dt><dt><span class="section"><a href="#interactive-rebase">Using interactive rebases</a></span></dt><dt><span class="section"><a href="#patch-series-tools">Other tools</a></span></dt><dt><span class="section"><a href="#problems-With-rewriting-history">Problems with rewriting history</a></span></dt><dt><span class="section"><a href="#bisect-merges">Why bisecting merge commits can be harder than bisecting linear history</a></span></dt></dl></div><p>Normally commits are only added to a project, never taken away or
1060 replaced. Git is designed with this assumption, and violating it will
1061 cause Git’s merge machinery (for example) to do the wrong thing.</p><p>However, there is a situation in which it can be useful to violate this
1062 assumption.</p><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="patch-series"></a>Creating the perfect patch series</h2></div></div></div><p>Suppose you are a contributor to a large project, and you want to add a
1063 complicated feature, and to present it to the other developers in a way
1064 that makes it easy for them to read your changes, verify that they are
1065 correct, and understand why you made each change.</p><p>If you present all of your changes as a single patch (or commit), they
1066 may find that it is too much to digest all at once.</p><p>If you present them with the entire history of your work, complete with
1067 mistakes, corrections, and dead ends, they may be overwhelmed.</p><p>So the ideal is usually to produce a series of patches such that:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem">
1068 Each patch can be applied in order.
1070 Each patch includes a single logical change, together with a
1071 message explaining the change.
1073 No patch introduces a regression: after applying any initial
1074 part of the series, the resulting project still compiles and
1075 works, and has no bugs that it didn’t have before.
1077 The complete series produces the same end result as your own
1078 (probably much messier!) development process did.
1079 </li></ol></div><p>We will introduce some tools that can help you do this, explain how to
1080 use them, and then explain some of the problems that can arise because
1081 you are rewriting history.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="using-git-rebase"></a>Keeping a patch series up to date using git rebase</h2></div></div></div><p>Suppose that you create a branch <code class="literal">mywork</code> on a remote-tracking branch
1082 <code class="literal">origin</code>, and create some commits on top of it:</p><pre class="screen">$ git switch -c mywork origin
1083 $ vi file.txt
1084 $ git commit
1085 $ vi otherfile.txt
1086 $ git commit
1087 ...</pre><p>You have performed no merges into mywork, so it is just a simple linear
1088 sequence of patches on top of <code class="literal">origin</code>:</p><pre class="literallayout"> o--o--O <-- origin
1089 \
1090 a--b--c <-- mywork</pre><p>Some more interesting work has been done in the upstream project, and
1091 <code class="literal">origin</code> has advanced:</p><pre class="literallayout"> o--o--O--o--o--o <-- origin
1092 \
1093 a--b--c <-- mywork</pre><p>At this point, you could use <code class="literal">pull</code> to merge your changes back in;
1094 the result would create a new merge commit, like this:</p><pre class="literallayout"> o--o--O--o--o--o <-- origin
1095 \ \
1096 a--b--c--m <-- mywork</pre><p>However, if you prefer to keep the history in mywork a simple series of
1097 commits without any merges, you may instead choose to use
1098 <a class="ulink" href="git-rebase.html" target="_top">git-rebase(1)</a>:</p><pre class="screen">$ git switch mywork
1099 $ git rebase origin</pre><p>This will remove each of your commits from mywork, temporarily saving
1100 them as patches (in a directory named <code class="literal">.git/rebase-apply</code>), update mywork to
1101 point at the latest version of origin, then apply each of the saved
1102 patches to the new mywork. The result will look like:</p><pre class="literallayout"> o--o--O--o--o--o <-- origin
1103 \
1104 a'--b'--c' <-- mywork</pre><p>In the process, it may discover conflicts. In that case it will stop
1105 and allow you to fix the conflicts; after fixing conflicts, use <code class="literal">git add</code>
1106 to update the index with those contents, and then, instead of
1107 running <code class="literal">git commit</code>, just run</p><pre class="screen">$ git rebase --continue</pre><p>and Git will continue applying the rest of the patches.</p><p>At any point you may use the <code class="literal">--abort</code> option to abort this process and
1108 return mywork to the state it had before you started the rebase:</p><pre class="screen">$ git rebase --abort</pre><p>If you need to reorder or edit a number of commits in a branch, it may
1110 squash commits, as well as marking them for individual editing during
1111 the rebase. See <a class="xref" href="#interactive-rebase" title="Using interactive rebases">the section called “Using interactive rebases”</a> for details, and
1112 <a class="xref" href="#reordering-patch-series" title="Reordering or selecting from a patch series">the section called “Reordering or selecting from a patch series”</a> for alternatives.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="rewriting-one-commit"></a>Rewriting a single commit</h2></div></div></div><p>We saw in <a class="xref" href="#fixing-a-mistake-by-rewriting-history" title="Fixing a mistake by rewriting history">the section called “Fixing a mistake by rewriting history”</a> that you can replace the
1113 most recent commit using</p><pre class="screen">$ git commit --amend</pre><p>which will replace the old commit by a new commit incorporating your
1114 changes, giving you a chance to edit the old commit message first.
1115 This is useful for fixing typos in your last commit, or for adjusting
1116 the patch contents of a poorly staged commit.</p><p>If you need to amend commits from deeper in your history, you can
1117 use <a class="link" href="#interactive-rebase" title="Using interactive rebases">interactive rebase’s <code class="literal">edit</code> instruction</a>.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="reordering-patch-series"></a>Reordering or selecting from a patch series</h2></div></div></div><p>Sometimes you want to edit a commit deeper in your history. One
1119 and then reset the state to before the patches:</p><pre class="screen">$ git format-patch origin
1120 $ git reset --hard origin</pre><p>Then modify, reorder, or eliminate patches as needed before applying
1121 them again with <a class="ulink" href="git-am.html" target="_top">git-am(1)</a>:</p><pre class="screen">$ git am *.patch</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="interactive-rebase"></a>Using interactive rebases</h2></div></div></div><p>You can also edit a patch series with an interactive rebase. This is
1122 the same as <a class="link" href="#reordering-patch-series" title="Reordering or selecting from a patch series">reordering a patch series using <code class="literal">format-patch</code></a>, so use whichever interface you like best.</p><p>Rebase your current HEAD on the last commit you want to retain as-is.
1123 For example, if you want to reorder the last 5 commits, use:</p><pre class="screen">$ git rebase -i HEAD~5</pre><p>This will open your editor with a list of steps to be taken to perform
1125 pick fa1afe1 The oneline of the next commit
1126 ...
1128 # Rebase c0ffeee..deadbee onto c0ffeee
1129 #
1130 # Commands:
1131 # p, pick = use commit
1132 # r, reword = use commit, but edit the commit message
1133 # e, edit = use commit, but stop for amending
1134 # s, squash = use commit, but meld into previous commit
1135 # f, fixup = like "squash", but discard this commit's log message
1136 # x, exec = run command (the rest of the line) using shell
1137 #
1138 # These lines can be re-ordered; they are executed from top to bottom.
1139 #
1140 # If you remove a line here THAT COMMIT WILL BE LOST.
1141 #
1142 # However, if you remove everything, the rebase will be aborted.
1143 #
1144 # Note that empty commits are commented out</pre><p>As explained in the comments, you can reorder commits, squash them
1145 together, edit commit messages, etc. by editing the list. Once you
1146 are satisfied, save the list and close your editor, and the rebase
1147 will begin.</p><p>The rebase will stop where <code class="literal">pick</code> has been replaced with <code class="literal">edit</code> or
1148 when a step in the list fails to mechanically resolve conflicts and
1149 needs your help. When you are done editing and/or resolving conflicts
1152 --abort</code>. Even after the rebase is complete, you can still recover
1153 the original branch by using the <a class="link" href="#reflogs" title="Reflogs">reflog</a>.</p><p>For a more detailed discussion of the procedure and additional tips,
1154 see the "INTERACTIVE MODE" section of <a class="ulink" href="git-rebase.html" target="_top">git-rebase(1)</a>.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="patch-series-tools"></a>Other tools</h2></div></div></div><p>There are numerous other tools, such as StGit, which exist for the
1155 purpose of maintaining a patch series. These are outside of the scope of
1156 this manual.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="problems-With-rewriting-history"></a>Problems with rewriting history</h2></div></div></div><p>The primary problem with rewriting the history of a branch has to do
1157 with merging. Suppose somebody fetches your branch and merges it into
1158 their branch, with a result something like this:</p><pre class="literallayout"> o--o--O--o--o--o <-- origin
1159 \ \
1160 t--t--t--m <-- their branch:</pre><p>Then suppose you modify the last three commits:</p><pre class="literallayout"> o--o--o <-- new head of origin
1161 /
1162 o--o--O--o--o--o <-- old head of origin</pre><p>If we examined all this history together in one repository, it will
1164 /
1165 o--o--O--o--o--o <-- old head of origin
1166 \ \
1167 t--t--t--m <-- their branch:</pre><p>Git has no way of knowing that the new head is an updated version of
1168 the old head; it treats this situation exactly the same as it would if
1169 two developers had independently done the work on the old and new heads
1170 in parallel. At this point, if someone attempts to merge the new head
1171 in to their branch, Git will attempt to merge together the two (old and
1172 new) lines of development, instead of trying to replace the old by the
1173 new. The results are likely to be unexpected.</p><p>You may still choose to publish branches whose history is rewritten,
1174 and it may be useful for others to be able to fetch those branches in
1175 order to examine or test them, but they should not attempt to pull such
1176 branches into their own work.</p><p>For true distributed development that supports proper merging,
1177 published branches should never be rewritten.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="bisect-merges"></a>Why bisecting merge commits can be harder than bisecting linear history</h2></div></div></div><p>The <a class="ulink" href="git-bisect.html" target="_top">git-bisect(1)</a> command correctly handles history that
1178 includes merge commits. However, when the commit that it finds is a
1179 merge commit, the user may need to work harder than usual to figure out
1180 why that commit introduced a problem.</p><p>Imagine this history:</p><pre class="literallayout"> ---Z---o---X---...---o---A---C---D
1181 \ /
1182 o---o---Y---...---o---B</pre><p>Suppose that on the upper line of development, the meaning of one
1183 of the functions that exists at Z is changed at commit X. The
1184 commits from Z leading to A change both the function’s
1185 implementation and all calling sites that exist at Z, as well
1186 as new calling sites they add, to be consistent. There is no
1187 bug at A.</p><p>Suppose that in the meantime on the lower line of development somebody
1188 adds a new calling site for that function at commit Y. The
1189 commits from Z leading to B all assume the old semantics of that
1190 function and the callers and the callee are consistent with each
1191 other. There is no bug at B, either.</p><p>Suppose further that the two development lines merge cleanly at C,
1192 so no conflict resolution is required.</p><p>Nevertheless, the code at C is broken, because the callers added
1193 on the lower line of development have not been converted to the new
1194 semantics introduced on the upper line of development. So if all
1195 you know is that D is bad, that Z is good, and that
1196 <a class="ulink" href="git-bisect.html" target="_top">git-bisect(1)</a> identifies C as the culprit, how will you
1197 figure out that the problem is due to this change in semantics?</p><p>When the result of a <code class="literal">git bisect</code> is a non-merge commit, you should
1198 normally be able to discover the problem by examining just that commit.
1199 Developers can make this easy by breaking their changes into small
1200 self-contained commits. That won’t help in the case above, however,
1201 because the problem isn’t obvious from examination of any single
1202 commit; instead, a global view of the development is required. To
1203 make matters worse, the change in semantics in the problematic
1204 function may be just one small part of the changes in the upper
1205 line of development.</p><p>On the other hand, if instead of merging at C you had rebased the
1206 history between Z to B on top of A, you would have gotten this
1207 linear history:</p><pre class="literallayout"> ---Z---o---X--...---o---A---o---o---Y*--...---o---B*--D*</pre><p>Bisecting between Z and D* would hit a single culprit commit Y*,
1208 and understanding why Y* was broken would probably be easier.</p><p>Partly for this reason, many experienced Git users, even when
1209 working on an otherwise merge-heavy project, keep the history
1210 linear by rebasing against the latest upstream version before
1211 publishing.</p></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="advanced-branch-management"></a>Chapter 6. Advanced branch management</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#fetching-individual-branches">Fetching individual branches</a></span></dt><dt><span class="section"><a href="#fetch-fast-forwards">git fetch and fast-forwards</a></span></dt><dt><span class="section"><a href="#forcing-fetch">Forcing git fetch to do non-fast-forward updates</a></span></dt><dt><span class="section"><a href="#remote-branch-configuration">Configuring remote-tracking branches</a></span></dt></dl></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="fetching-individual-branches"></a>Fetching individual branches</h2></div></div></div><p>Instead of using <a class="ulink" href="git-remote.html" target="_top">git-remote(1)</a>, you can also choose just
1212 to update one branch at a time, and to store it locally under an
1213 arbitrary name:</p><pre class="screen">$ git fetch origin todo:my-todo-work</pre><p>The first argument, <code class="literal">origin</code>, just tells Git to fetch from the
1214 repository you originally cloned from. The second argument tells Git
1216 store it locally under the name <code class="literal">refs/heads/my-todo-work</code>.</p><p>You can also fetch branches from other repositories; so</p><pre class="screen">$ git fetch git://example.com/proj.git master:example-master</pre><p>will create a new branch named <code class="literal">example-master</code> and store in it the
1218 already have a branch named example-master, it will attempt to
1219 <a class="link" href="#fast-forwards" title="Fast-forward merges">fast-forward</a> to the commit given by example.com’s
1220 master branch. In more detail:</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="fetch-fast-forwards"></a>git fetch and fast-forwards</h2></div></div></div><p>In the previous example, when updating an existing branch, <code class="literal">git fetch</code>
1221 checks to make sure that the most recent commit on the remote
1222 branch is a descendant of the most recent commit on your copy of the
1223 branch before updating your copy of the branch to point at the new
1224 commit. Git calls this process a <a class="link" href="#fast-forwards" title="Fast-forward merges">fast-forward</a>.</p><p>A fast-forward looks something like this:</p><pre class="literallayout"> o--o--o--o <-- old head of the branch
1225 \
1226 o--o--o <-- new head of the branch</pre><p>In some cases it is possible that the new head will <span class="strong"><strong>not</strong></span> actually be
1227 a descendant of the old head. For example, the developer may have
1228 realized a serious mistake was made and decided to backtrack,
1229 resulting in a situation like:</p><pre class="literallayout"> o--o--o--o--a--b <-- old head of the branch
1230 \
1231 o--o--o <-- new head of the branch</pre><p>In this case, <code class="literal">git fetch</code> will fail, and print out a warning.</p><p>In that case, you can still force Git to update to the new head, as
1232 described in the following section. However, note that in the
1233 situation above this may mean losing the commits labeled <code class="literal">a</code> and <code class="literal">b</code>,
1234 unless you’ve already created a reference of your own pointing to
1235 them.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="forcing-fetch"></a>Forcing git fetch to do non-fast-forward updates</h2></div></div></div><p>If git fetch fails because the new head of a branch is not a
1236 descendant of the old head, you may force the update with:</p><pre class="screen">$ git fetch git://example.com/proj.git +master:refs/remotes/example/master</pre><p>Note the addition of the <code class="literal">+</code> sign. Alternatively, you can use the <code class="literal">-f</code>
1237 flag to force updates of all the fetched branches, as in:</p><pre class="screen">$ git fetch -f origin</pre><p>Be aware that commits that the old version of example/master pointed at
1238 may be lost, as we saw in the previous section.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="remote-branch-configuration"></a>Configuring remote-tracking branches</h2></div></div></div><p>We saw above that <code class="literal">origin</code> is just a shortcut to refer to the
1239 repository that you originally cloned from. This information is
1240 stored in Git configuration variables, which you can see using
1241 <a class="ulink" href="git-config.html" target="_top">git-config(1)</a>:</p><pre class="screen">$ git config -l
1242 core.repositoryformatversion=0
1243 core.filemode=true
1244 core.logallrefupdates=true
1245 remote.origin.url=git://git.kernel.org/pub/scm/git/git.git
1246 remote.origin.fetch=+refs/heads/*:refs/remotes/origin/*
1247 branch.master.remote=origin
1248 branch.master.merge=refs/heads/master</pre><p>If there are other repositories that you also use frequently, you can
1249 create similar configuration options to save typing; for example,</p><pre class="screen">$ git remote add example git://example.com/proj.git</pre><p>adds the following to <code class="literal">.git/config</code>:</p><pre class="screen">[remote "example"]
1250 url = git://example.com/proj.git
1251 fetch = +refs/heads/*:refs/remotes/example/*</pre><p>Also note that the above configuration can be performed by directly
1252 editing the file <code class="literal">.git/config</code> instead of using <a class="ulink" href="git-remote.html" target="_top">git-remote(1)</a>.</p><p>After configuring the remote, the following three commands will do the
1253 same thing:</p><pre class="screen">$ git fetch git://example.com/proj.git +refs/heads/*:refs/remotes/example/*
1254 $ git fetch example +refs/heads/*:refs/remotes/example/*
1255 $ git fetch example</pre><p>See <a class="ulink" href="git-config.html" target="_top">git-config(1)</a> for more details on the configuration
1256 options mentioned above and <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a> for more details on
1257 the refspec syntax.</p></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="git-concepts"></a>Chapter 7. Git concepts</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#the-object-database">The Object Database</a></span></dt><dd><dl><dt><span class="section"><a href="#commit-object">Commit Object</a></span></dt><dt><span class="section"><a href="#tree-object">Tree Object</a></span></dt><dt><span class="section"><a href="#blob-object">Blob Object</a></span></dt><dt><span class="section"><a href="#trust">Trust</a></span></dt><dt><span class="section"><a href="#tag-object">Tag Object</a></span></dt><dt><span class="section"><a href="#pack-files">How Git stores objects efficiently: pack files</a></span></dt><dt><span class="section"><a href="#dangling-objects">Dangling objects</a></span></dt><dt><span class="section"><a href="#recovering-from-repository-corruption">Recovering from repository corruption</a></span></dt></dl></dd><dt><span class="section"><a href="#the-index">The index</a></span></dt></dl></div><p>Git is built on a small number of simple but powerful ideas. While it
1258 is possible to get things done without understanding them, you will find
1259 Git much more intuitive if you do.</p><p>We start with the most important, the <a class="link" href="#def_object_database">object database</a> and the <a class="link" href="#def_index">index</a>.</p><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="the-object-database"></a>The Object Database</h2></div></div></div><p>We already saw in <a class="xref" href="#understanding-commits" title="Understanding History: Commits">the section called “Understanding History: Commits”</a> that all commits are stored
1261 represent the history of a project is stored in objects with such names.
1262 In each case the name is calculated by taking the SHA-1 hash of the
1263 contents of the object. The SHA-1 hash is a cryptographic hash function.
1264 What that means to us is that it is impossible to find two different
1265 objects with the same name. This has a number of advantages; among
1266 others:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1267 Git can quickly determine whether two objects are identical or not,
1268 just by comparing names.
1270 Since object names are computed the same way in every repository, the
1271 same content stored in two repositories will always be stored under
1272 the same name.
1274 Git can detect errors when it reads an object, by checking that the
1275 object’s name is still the SHA-1 hash of its contents.
1276 </li></ul></div><p>(See <a class="xref" href="#object-details" title="Object storage format">the section called “Object storage format”</a> for the details of the object formatting and
1277 SHA-1 calculation.)</p><p>There are four different types of objects: "blob", "tree", "commit", and
1278 "tag".</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1282 "blob" objects into a directory structure. In addition, a tree object
1283 can refer to other tree objects, thus creating a directory hierarchy.
1285 A <a class="link" href="#def_commit_object">"commit" object</a> ties such directory hierarchies
1287 commit contains the object name of exactly one tree designating the
1288 directory hierarchy at the time of the commit. In addition, a commit
1289 refers to "parent" commit objects that describe the history of how we
1290 arrived at that directory hierarchy.
1293 used to sign other objects. It contains the object name and type of
1294 another object, a symbolic name (of course!) and, optionally, a
1295 signature.
1296 </li></ul></div><p>The object types in some more detail:</p><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="commit-object"></a>Commit Object</h3></div></div></div><p>The "commit" object links a physical state of a tree with a description
1298 <a class="ulink" href="git-show.html" target="_top">git-show(1)</a> or <a class="ulink" href="git-log.html" target="_top">git-log(1)</a> to examine your favorite
1300 commit 2be7fcb4764f2dbcee52635b91fedb1b3dcf7ab4
1301 tree fb3a8bdd0ceddd019615af4d57a53f43d8cee2bf
1302 parent 257a84d9d02e90447b149af58b271c19405edb6a
1306 Fix misspelling of 'suppress' in docs
1308 Signed-off-by: Junio C Hamano <gitster@pobox.com></pre><p>As you can see, a commit is defined by:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1309 a tree: The SHA-1 name of a tree object (as defined below), representing
1310 the contents of a directory at a certain point in time.
1312 parent(s): The SHA-1 name(s) of some number of commits which represent the
1313 immediately previous step(s) in the history of the project. The
1314 example above has one parent; merge commits may have more than
1315 one. A commit with no parents is called a "root" commit, and
1316 represents the initial revision of a project. Each project must have
1317 at least one root. A project can also have multiple roots, though
1318 that isn’t common (or necessarily a good idea).
1320 an author: The name of the person responsible for this change, together
1321 with its date.
1323 a committer: The name of the person who actually created the commit,
1324 with the date it was done. This may be different from the author, for
1325 example, if the author was someone who wrote a patch and emailed it
1326 to the person who used it to create the commit.
1328 a comment describing this commit.
1329 </li></ul></div><p>Note that a commit does not itself contain any information about what
1330 actually changed; all changes are calculated by comparing the contents
1331 of the tree referred to by this commit with the trees associated with
1332 its parents. In particular, Git does not attempt to record file renames
1333 explicitly, though it can identify cases where the existence of the same
1334 file data at changing paths suggests a rename. (See, for example, the
1335 <code class="literal">-M</code> option to <a class="ulink" href="git-diff.html" target="_top">git-diff(1)</a>).</p><p>A commit is usually created by <a class="ulink" href="git-commit.html" target="_top">git-commit(1)</a>, which creates a
1336 commit whose parent is normally the current HEAD, and whose tree is
1337 taken from the content currently stored in the index.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="tree-object"></a>Tree Object</h3></div></div></div><p>The ever-versatile <a class="ulink" href="git-show.html" target="_top">git-show(1)</a> command can also be used to
1338 examine tree objects, but <a class="ulink" href="git-ls-tree.html" target="_top">git-ls-tree(1)</a> will give you more
1348 ...</pre><p>As you can see, a tree object contains a list of entries, each with a
1349 mode, object type, SHA-1 name, and name, sorted by name. It represents
1350 the contents of a single directory tree.</p><p>The object type may be a blob, representing the contents of a file, or
1351 another tree, representing the contents of a subdirectory. Since trees
1352 and blobs, like all other objects, are named by the SHA-1 hash of their
1353 contents, two trees have the same SHA-1 name if and only if their
1354 contents (including, recursively, the contents of all subdirectories)
1355 are identical. This allows Git to quickly determine the differences
1356 between two related tree objects, since it can ignore any entries with
1357 identical object names.</p><p>(Note: in the presence of submodules, trees may also have commits as
1358 entries. See <a class="xref" href="#submodules" title="Chapter 8. Submodules">Chapter 8, <i>Submodules</i></a> for documentation.)</p><p>Note that the files all have mode 644 or 755: Git actually only pays
1359 attention to the executable bit.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="blob-object"></a>Blob Object</h3></div></div></div><p>You can use <a class="ulink" href="git-show.html" target="_top">git-show(1)</a> to examine the contents of a blob; take,
1360 for example, the blob in the entry for <code class="literal">COPYING</code> from the tree above:</p><pre class="screen">$ git show 6ff87c4664
1362 Note that the only valid version of the GPL as far as this project
1363 is concerned is _this_ particular version of the license (ie v2, not
1364 v2.2 or v3.x or whatever), unless explicitly otherwise stated.
1366 to anything else or have attributes of any kind.</p><p>Since the blob is entirely defined by its data, if two files in a
1367 directory tree (or in multiple different versions of the repository)
1368 have the same contents, they will share the same blob object. The object
1369 is totally independent of its location in the directory tree, and
1370 renaming a file does not change the object that file is associated with.</p><p>Note that any tree or blob object can be examined using
1371 <a class="ulink" href="git-show.html" target="_top">git-show(1)</a> with the <revision>:<path> syntax. This can
1372 sometimes be useful for browsing the contents of a tree that is not
1373 currently checked out.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="trust"></a>Trust</h3></div></div></div><p>If you receive the SHA-1 name of a blob from one source, and its contents
1374 from another (possibly untrusted) source, you can still trust that those
1375 contents are correct as long as the SHA-1 name agrees. This is because
1376 the SHA-1 is designed so that it is infeasible to find different contents
1377 that produce the same hash.</p><p>Similarly, you need only trust the SHA-1 name of a top-level tree object
1378 to trust the contents of the entire directory that it refers to, and if
1379 you receive the SHA-1 name of a commit from a trusted source, then you
1380 can easily verify the entire history of commits reachable through
1381 parents of that commit, and all of those contents of the trees referred
1382 to by those commits.</p><p>So to introduce some real trust in the system, the only thing you need
1383 to do is to digitally sign just <span class="emphasis"><em>one</em></span> special note, which includes the
1384 name of a top-level commit. Your digital signature shows others
1385 that you trust that commit, and the immutability of the history of
1386 commits tells others that they can trust the whole history.</p><p>In other words, you can easily validate a whole archive by just
1387 sending out a single email that tells the people the name (SHA-1 hash)
1388 of the top commit, and digitally sign that email using something
1389 like GPG/PGP.</p><p>To assist in this, Git also provides the tag object…</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="tag-object"></a>Tag Object</h3></div></div></div><p>A tag object contains an object, object type, tag name, the name of the
1390 person ("tagger") who created the tag, and a message, which may contain
1391 a signature, as can be seen using <a class="ulink" href="git-cat-file.html" target="_top">git-cat-file(1)</a>:</p><pre class="screen">$ git cat-file tag v1.5.0
1392 object 437b1b20df4b356c9342dac8d38849f24ef44f27
1393 type commit
1394 tag v1.5.0
1397 GIT 1.5.0
1398 -----BEGIN PGP SIGNATURE-----
1399 Version: GnuPG v1.4.6 (GNU/Linux)
1401 iD8DBQBF0lGqwMbZpPMRm5oRAuRiAJ9ohBLd7s2kqjkKlq1qqC57SbnmzQCdG4ui
1402 nLE/L9aUXdWeTFPron96DLA=
1403 =2E+0
1404 -----END PGP SIGNATURE-----</pre><p>See the <a class="ulink" href="git-tag.html" target="_top">git-tag(1)</a> command to learn how to create and verify tag
1405 objects. (Note that <a class="ulink" href="git-tag.html" target="_top">git-tag(1)</a> can also be used to create
1406 "lightweight tags", which are not tag objects at all, but just simple
1407 references whose names begin with <code class="literal">refs/tags/</code>).</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="pack-files"></a>How Git stores objects efficiently: pack files</h3></div></div></div><p>Newly created objects are initially created in a file named after the
1408 object’s SHA-1 hash (stored in <code class="literal">.git/objects</code>).</p><p>Unfortunately this system becomes inefficient once a project has a
1410 6930 objects, 47620 kilobytes</pre><p>The first number is the number of objects which are kept in
1411 individual files. The second is the amount of space taken up by
1412 those "loose" objects.</p><p>You can save space and make Git faster by moving these loose objects in
1413 to a "pack file", which stores a group of objects in an efficient
1414 compressed format; the details of how pack files are formatted can be
1415 found in <a class="ulink" href="gitformat-pack.html" target="_top">gitformat-pack(5)</a>.</p><p>To put the loose objects into a pack, just run git repack:</p><pre class="screen">$ git repack
1416 Counting objects: 6020, done.
1417 Delta compression using up to 4 threads.
1420 Total 6020 (delta 4070), reused 0 (delta 0)</pre><p>This creates a single "pack file" in .git/objects/pack/
1421 containing all currently unpacked objects. You can then run</p><pre class="screen">$ git prune</pre><p>to remove any of the "loose" objects that are now contained in the
1422 pack. This will also remove any unreferenced objects (which may be
1424 You can verify that the loose objects are gone by looking at the
1425 <code class="literal">.git/objects</code> directory or by running</p><pre class="screen">$ git count-objects
1426 0 objects, 0 kilobytes</pre><p>Although the object files are gone, any commands that refer to those
1427 objects will work exactly as they did before.</p><p>The <a class="ulink" href="git-gc.html" target="_top">git-gc(1)</a> command performs packing, pruning, and more for
1428 you, so is normally the only high-level command you need.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="dangling-objects"></a>Dangling objects</h3></div></div></div><p>The <a class="ulink" href="git-fsck.html" target="_top">git-fsck(1)</a> command will sometimes complain about dangling
1429 objects. They are not a problem.</p><p>The most common cause of dangling objects is that you’ve rebased a
1430 branch, or you have pulled from somebody else who rebased a branch—see
1431 <a class="xref" href="#cleaning-up-history" title="Chapter 5. Rewriting history and maintaining patch series">Chapter 5, <i>Rewriting history and maintaining patch series</i></a>. In that case, the old head of the original
1432 branch still exists, as does everything it pointed to. The branch
1433 pointer itself just doesn’t, since you replaced it with another one.</p><p>There are also other situations that cause dangling objects. For
1434 example, a "dangling blob" may arise because you did a <code class="literal">git add</code> of a
1435 file, but then, before you actually committed it and made it part of the
1436 bigger picture, you changed something else in that file and committed
1437 that <span class="strong"><strong>updated</strong></span> thing—the old state that you added originally ends up
1438 not being pointed to by any commit or tree, so it’s now a dangling blob
1440 there are criss-cross merges and thus more than one merge base (which is
1441 fairly unusual, but it does happen), it will generate one temporary
1442 midway tree (or possibly even more, if you had lots of criss-crossing
1443 merges and more than two merge bases) as a temporary internal merge
1444 base, and again, those are real objects, but the end result will not end
1445 up pointing to them, so they end up "dangling" in your repository.</p><p>Generally, dangling objects aren’t anything to worry about. They can
1446 even be very useful: if you screw something up, the dangling objects can
1447 be how you recover your old tree (say, you did a rebase, and realized
1448 that you really didn’t want to—you can look at what dangling objects
1449 you have, and decide to reset your head to some old dangling state).</p><p>For commits, you can just use:</p><pre class="screen">$ gitk <dangling-commit-sha-goes-here> --not --all</pre><p>This asks for all the history reachable from the given commit but not
1450 from any branch, tag, or other reference. If you decide it’s something
1451 you want, you can always create a new reference to it, e.g.,</p><pre class="screen">$ git branch recovered-branch <dangling-commit-sha-goes-here></pre><p>For blobs and trees, you can’t do the same, but you can still examine
1452 them. You can just do</p><pre class="screen">$ git show <dangling-blob/tree-sha-goes-here></pre><p>to show what the contents of the blob were (or, for a tree, basically
1453 what the <code class="literal">ls</code> for that directory was), and that may give you some idea
1454 of what the operation was that left that dangling object.</p><p>Usually, dangling blobs and trees aren’t very interesting. They’re
1455 almost always the result of either being a half-way mergebase (the blob
1456 will often even have the conflict markers from a merge in it, if you
1457 have had conflicting merges that you fixed up by hand), or simply
1458 because you interrupted a <code class="literal">git fetch</code> with ^C or something like that,
1459 leaving <span class="emphasis"><em>some</em></span> of the new objects in the object database, but just
1460 dangling and useless.</p><p>Anyway, once you are sure that you’re not interested in any dangling
1461 state, you can just prune all unreachable objects:</p><pre class="screen">$ git prune</pre><p>and they’ll be gone. (You should only run <code class="literal">git prune</code> on a quiescent
1462 repository—it’s kind of like doing a filesystem fsck recovery: you
1463 don’t want to do that while the filesystem is mounted.
1464 <code class="literal">git prune</code> is designed not to cause any harm in such cases of concurrent
1465 accesses to a repository but you might receive confusing or scary messages.)</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="recovering-from-repository-corruption"></a>Recovering from repository corruption</h3></div></div></div><p>By design, Git treats data trusted to it with caution. However, even in
1466 the absence of bugs in Git itself, it is still possible that hardware or
1467 operating system errors could corrupt data.</p><p>The first defense against such problems is backups. You can back up a
1468 Git directory using clone, or just using cp, tar, or any other backup
1469 mechanism.</p><p>As a last resort, you can search for the corrupted objects and attempt
1470 to replace them by hand. Back up your repository before attempting this
1471 in case you corrupt things even more in the process.</p><p>We’ll assume that the problem is a single missing or corrupted blob,
1472 which is sometimes a solvable problem. (Recovering missing trees and
1473 especially commits is <span class="strong"><strong>much</strong></span> harder).</p><p>Before starting, verify that there is corruption, and figure out where
1474 it is with <a class="ulink" href="git-fsck.html" target="_top">git-fsck(1)</a>; this may be time-consuming.</p><p>Assume the output looks like this:</p><pre class="screen">$ git fsck --full --no-dangling
1475 broken link from tree 2d9263c6d23595e7cb2a21e5ebbb53655278dff8
1476 to blob 4b9458b3786228369c63936db65827de3cc06200
1477 missing blob 4b9458b3786228369c63936db65827de3cc06200</pre><p>Now you know that blob 4b9458b3 is missing, and that the tree 2d9263c6
1478 points to it. If you could find just one copy of that missing blob
1479 object, possibly in some other repository, you could move it into
1481 still examine the tree that pointed to it with <a class="ulink" href="git-ls-tree.html" target="_top">git-ls-tree(1)</a>,
1482 which might output something like:</p><pre class="screen">$ git ls-tree 2d9263c6d23595e7cb2a21e5ebbb53655278dff8
1484 100644 blob ebf9bf84da0aab5ed944264a5db2a65fe3a3e883 .mailmap
1485 100644 blob ca442d313d86dc67e0a2e5d584b465bd382cbf5c COPYING
1486 ...
1488 ...</pre><p>So now you know that the missing blob was the data for a file named
1490 say it’s in <code class="literal">somedirectory</code>. If you’re lucky the missing copy might be
1491 the same as the copy you have checked out in your working tree at
1493 <a class="ulink" href="git-hash-object.html" target="_top">git-hash-object(1)</a>:</p><pre class="screen">$ git hash-object -w somedirectory/myfile</pre><p>which will create and store a blob object with the contents of
1494 somedirectory/myfile, and output the SHA-1 of that object. if you’re
1495 extremely lucky it might be 4b9458b3786228369c63936db65827de3cc06200, in
1496 which case you’ve guessed right, and the corruption is fixed!</p><p>Otherwise, you need more information. How do you tell which version of
1497 the file has been lost?</p><p>The easiest way to do this is with:</p><pre class="screen">$ git log --raw --all --full-history -- somedirectory/myfile</pre><p>Because you’re asking for raw output, you’ll now get something like</p><pre class="screen">commit abc
1498 Author:
1499 Date:
1500 ...
1501 :100644 100644 4b9458b newsha M somedirectory/myfile
1504 commit xyz
1505 Author:
1506 Date:
1508 ...
1509 :100644 100644 oldsha 4b9458b M somedirectory/myfile</pre><p>This tells you that the immediately following version of the file was
1511 You also know the commit messages that went with the change from oldsha
1512 to 4b9458b and with the change from 4b9458b to newsha.</p><p>If you’ve been committing small enough changes, you may now have a good
1513 shot at reconstructing the contents of the in-between state 4b9458b.</p><p>If you can do that, you can now recreate the missing object with</p><pre class="screen">$ git hash-object -w <recreated-file></pre><p>and your repository is good again!</p><p>(Btw, you could have ignored the <code class="literal">fsck</code>, and started with doing a</p><pre class="screen">$ git log --raw --all</pre><p>and just looked for the sha of the missing object (4b9458b) in that
1514 whole thing. It’s up to you—Git does <span class="strong"><strong>have</strong></span> a lot of information, it is
1515 just missing one particular blob version.</p></div></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="the-index"></a>The index</h2></div></div></div><p>The index is a binary file (generally kept in <code class="literal">.git/index</code>) containing a
1516 sorted list of path names, each with permissions and the SHA-1 of a blob
1517 object; <a class="ulink" href="git-ls-files.html" target="_top">git-ls-files(1)</a> can show you the contents of the index:</p><pre class="screen">$ git ls-files --stage
1523 ...
1526 100644 d5de8292e05e7c36c4b68857c1cf9855e3d2f70a 0 xdiff/xutils.h</pre><p>Note that in older documentation you may see the index called the
1528 properties:</p><div class="orderedlist"><ol class="orderedlist" type="1"><li class="listitem"><p class="simpara">
1529 The index contains all the information necessary to generate a single
1530 (uniquely determined) tree object.
1531 </p><p class="simpara">For example, running <a class="ulink" href="git-commit.html" target="_top">git-commit(1)</a> generates this tree object
1532 from the index, stores it in the object database, and uses it as the
1534 The index enables fast comparisons between the tree object it defines
1535 and the working tree.
1537 the last modified time). This data is not displayed above, and is not
1538 stored in the created tree object, but it can be used to determine
1539 quickly which files in the working directory differ from what was
1540 stored in the index, and thus save Git from having to read all of the
1542 It can efficiently represent information about merge conflicts
1543 between different tree objects, allowing each pathname to be
1544 associated with sufficient information about the trees involved that
1545 you can create a three-way merge between them.
1546 </p><p class="simpara">We saw in <a class="xref" href="#conflict-resolution" title="Getting conflict-resolution help during a merge">the section called “Getting conflict-resolution help during a merge”</a> that during a merge the index can
1547 store multiple versions of a single file (called "stages"). The third
1548 column in the <a class="ulink" href="git-ls-files.html" target="_top">git-ls-files(1)</a> output above is the stage
1549 number, and will take on values other than 0 for files with merge
1550 conflicts.</p></li></ol></div><p>The index is thus a sort of temporary staging area, which is filled with
1551 a tree which you are in the process of working on.</p><p>If you blow the index away entirely, you generally haven’t lost any
1552 information as long as you have the name of the tree that it described.</p></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="submodules"></a>Chapter 8. Submodules</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#pitfalls-with-submodules">Pitfalls with submodules</a></span></dt></dl></div><p>Large projects are often composed of smaller, self-contained modules. For
1553 example, an embedded Linux distribution’s source tree would include every
1554 piece of software in the distribution with some local modifications; a movie
1555 player might need to build against a specific, known-working version of a
1556 decompression library; several independent programs might all share the same
1557 build scripts.</p><p>With centralized revision control systems this is often accomplished by
1558 including every module in one single repository. Developers can check out
1559 all modules or only the modules they need to work with. They can even modify
1560 files across several modules in a single commit while moving things around
1561 or updating APIs and translations.</p><p>Git does not allow partial checkouts, so duplicating this approach in Git
1562 would force developers to keep a local copy of modules they are not
1563 interested in touching. Commits in an enormous checkout would be slower
1564 than you’d expect as Git would have to scan every directory for changes.
1565 If modules have a lot of local history, clones would take forever.</p><p>On the plus side, distributed revision control systems can much better
1566 integrate with external sources. In a centralized model, a single arbitrary
1567 snapshot of the external project is exported from its own revision control
1568 and then imported into the local revision control on a vendor branch. All
1569 the history is hidden. With distributed revision control you can clone the
1570 entire external history and much more easily follow development and re-merge
1571 local changes.</p><p>Git’s submodule support allows a repository to contain, as a subdirectory, a
1572 checkout of an external project. Submodules maintain their own identity;
1573 the submodule support just stores the submodule repository location and
1574 commit ID, so other developers who clone the containing project
1575 ("superproject") can easily clone all the submodules at the same revision.
1576 Partial checkouts of the superproject are possible: you can tell Git to
1577 clone none, some or all of the submodules.</p><p>The <a class="ulink" href="git-submodule.html" target="_top">git-submodule(1)</a> command is available since Git 1.5.3. Users
1578 with Git 1.5.2 can look up the submodule commits in the repository and
1579 manually check them out; earlier versions won’t recognize the submodules at
1580 all.</p><p>To see how submodule support works, create four example
1582 $ cd ~/git
1583 $ for i in a b c d
1584 do
1585 mkdir $i
1586 cd $i
1587 git init
1589 git add $i.txt
1590 git commit -m "Initial commit, submodule $i"
1591 cd ..
1592 done</pre><p>Now create the superproject and add all the submodules:</p><pre class="screen">$ mkdir super
1593 $ cd super
1594 $ git init
1595 $ for i in a b c d
1596 do
1597 git submodule add ~/git/$i $i
1598 done</pre><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p>Do not use local URLs here if you plan to publish your superproject!</p></div><p>See what files <code class="literal">git submodule</code> created:</p><pre class="screen">$ ls -a
1599 . .. .git .gitmodules a b c d</pre><p>The <code class="literal">git submodule add <repo> <path></code> command does a couple of things:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1600 It clones the submodule from <code class="literal"><repo></code> to the given <code class="literal"><path></code> under the
1601 current directory and by default checks out the master branch.
1603 It adds the submodule’s clone path to the <a class="ulink" href="gitmodules.html" target="_top">gitmodules(5)</a> file and
1604 adds this file to the index, ready to be committed.
1606 It adds the submodule’s current commit ID to the index, ready to be
1607 committed.
1608 </li></ul></div><p>Commit the superproject:</p><pre class="screen">$ git commit -m "Add submodules a, b, c and d."</pre><p>Now clone the superproject:</p><pre class="screen">$ cd ..
1609 $ git clone super cloned
1610 $ cd cloned</pre><p>The submodule directories are there, but they’re empty:</p><pre class="screen">$ ls -a a
1611 . ..
1612 $ git submodule status
1613 -d266b9873ad50488163457f025db7cdd9683d88b a
1614 -e81d457da15309b4fef4249aba9b50187999670d b
1615 -c1536a972b9affea0f16e0680ba87332dc059146 c
1616 -d96249ff5d57de5de093e6baff9e0aafa5276a74 d</pre><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p>The commit object names shown above would be different for you, but they
1617 should match the HEAD commit object names of your repositories. You can check
1618 it by running <code class="literal">git ls-remote ../a</code>.</p></div><p>Pulling down the submodules is a two-step process. First run <code class="literal">git submodule
1619 init</code> to add the submodule repository URLs to <code class="literal">.git/config</code>:</p><pre class="screen">$ git submodule init</pre><p>Now use <code class="literal">git submodule update</code> to clone the repositories and check out the
1621 $ cd a
1622 $ ls -a
1623 . .. .git a.txt</pre><p>One major difference between <code class="literal">git submodule update</code> and <code class="literal">git submodule add</code> is
1624 that <code class="literal">git submodule update</code> checks out a specific commit, rather than the tip
1625 of a branch. It’s like checking out a tag: the head is detached, so you’re not
1627 * (detached from d266b98)
1628 master</pre><p>If you want to make a change within a submodule and you have a detached head,
1629 then you should create or checkout a branch, make your changes, publish the
1630 change within the submodule, and then update the superproject to reference the
1631 new commit:</p><pre class="screen">$ git switch master</pre><p>or</p><pre class="screen">$ git switch -c fix-up</pre><p>then</p><pre class="screen">$ echo "adding a line again" >> a.txt
1632 $ git commit -a -m "Updated the submodule from within the superproject."
1633 $ git push
1634 $ cd ..
1635 $ git diff
1636 diff --git a/a b/a
1638 --- a/a
1639 +++ b/a
1641 -Subproject commit d266b9873ad50488163457f025db7cdd9683d88b
1642 +Subproject commit 261dfac35cb99d380eb966e102c1197139f7fa24
1643 $ git add a
1644 $ git commit -m "Updated submodule a."
1645 $ git push</pre><p>You have to run <code class="literal">git submodule update</code> after <code class="literal">git pull</code> if you want to update
1646 submodules, too.</p><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="pitfalls-with-submodules"></a>Pitfalls with submodules</h2></div></div></div><p>Always publish the submodule change before publishing the change to the
1647 superproject that references it. If you forget to publish the submodule change,
1649 $ echo i added another line to this file >> a.txt
1650 $ git commit -a -m "doing it wrong this time"
1651 $ cd ..
1652 $ git add a
1653 $ git commit -m "Updated submodule a again."
1654 $ git push
1655 $ cd ~/git/cloned
1656 $ git pull
1657 $ git submodule update
1658 error: pathspec '261dfac35cb99d380eb966e102c1197139f7fa24' did not match any file(s) known to git.
1659 Did you forget to 'git add'?
1660 Unable to checkout '261dfac35cb99d380eb966e102c1197139f7fa24' in submodule path 'a'</pre><p>In older Git versions it could be easily forgotten to commit new or modified
1661 files in a submodule, which silently leads to similar problems as not pushing
1662 the submodule changes. Starting with Git 1.7.0 both <code class="literal">git status</code> and <code class="literal">git diff</code>
1663 in the superproject show submodules as modified when they contain new or
1664 modified files to protect against accidentally committing such a state. <code class="literal">git
1665 diff</code> will also add a <code class="literal">-dirty</code> to the work tree side when generating patch
1666 output or used with the <code class="literal">--submodule</code> option:</p><pre class="screen">$ git diff
1667 diff --git a/sub b/sub
1668 --- a/sub
1669 +++ b/sub
1671 -Subproject commit 3f356705649b5d566d97ff843cf193359229a453
1672 +Subproject commit 3f356705649b5d566d97ff843cf193359229a453-dirty
1673 $ git diff --submodule
1674 Submodule sub 3f35670..3f35670-dirty:</pre><p>You also should not rewind branches in a submodule beyond commits that were
1675 ever recorded in any superproject.</p><p>It’s not safe to run <code class="literal">git submodule update</code> if you’ve made and committed
1676 changes within a submodule without checking out a branch first. They will be
1678 module a
1679 $ echo line added from private2 >> a.txt
1680 $ git commit -a -m "line added inside private2"
1681 $ cd ..
1682 $ git submodule update
1683 Submodule path 'a': checked out 'd266b9873ad50488163457f025db7cdd9683d88b'
1684 $ cd a
1685 $ cat a.txt
1686 module a</pre><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p>The changes are still visible in the submodule’s reflog.</p></div><p>If you have uncommitted changes in your submodule working tree, <code class="literal">git
1687 submodule update</code> will not overwrite them. Instead, you get the usual
1688 warning about not being able switch from a dirty branch.</p></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="low-level-operations"></a>Chapter 9. Low-level Git operations</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#object-manipulation">Object access and manipulation</a></span></dt><dt><span class="section"><a href="#the-workflow">The Workflow</a></span></dt><dd><dl><dt><span class="section"><a href="#working-directory-to-index">working directory → index</a></span></dt><dt><span class="section"><a href="#index-to-object-database">index → object database</a></span></dt><dt><span class="section"><a href="#object-database-to-index">object database → index</a></span></dt><dt><span class="section"><a href="#index-to-working-directory">index → working directory</a></span></dt><dt><span class="section"><a href="#tying-it-all-together">Tying it all together</a></span></dt></dl></dd><dt><span class="section"><a href="#examining-the-data">Examining the data</a></span></dt><dt><span class="section"><a href="#merging-multiple-trees">Merging multiple trees</a></span></dt><dt><span class="section"><a href="#merging-multiple-trees-2">Merging multiple trees, continued</a></span></dt></dl></div><p>Many of the higher-level commands were originally implemented as shell
1689 scripts using a smaller core of low-level Git commands. These can still
1690 be useful when doing unusual things with Git, or just as a way to
1691 understand its inner workings.</p><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="object-manipulation"></a>Object access and manipulation</h2></div></div></div><p>The <a class="ulink" href="git-cat-file.html" target="_top">git-cat-file(1)</a> command can show the contents of any object,
1692 though the higher-level <a class="ulink" href="git-show.html" target="_top">git-show(1)</a> is usually more useful.</p><p>The <a class="ulink" href="git-commit-tree.html" target="_top">git-commit-tree(1)</a> command allows constructing commits with
1693 arbitrary parents and trees.</p><p>A tree can be created with <a class="ulink" href="git-write-tree.html" target="_top">git-write-tree(1)</a> and its data can be
1694 accessed by <a class="ulink" href="git-ls-tree.html" target="_top">git-ls-tree(1)</a>. Two trees can be compared with
1695 <a class="ulink" href="git-diff-tree.html" target="_top">git-diff-tree(1)</a>.</p><p>A tag is created with <a class="ulink" href="git-mktag.html" target="_top">git-mktag(1)</a>, and the signature can be
1696 verified by <a class="ulink" href="git-verify-tag.html" target="_top">git-verify-tag(1)</a>, though it is normally simpler to
1697 use <a class="ulink" href="git-tag.html" target="_top">git-tag(1)</a> for both.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="the-workflow"></a>The Workflow</h2></div></div></div><p>High-level operations such as <a class="ulink" href="git-commit.html" target="_top">git-commit(1)</a> and
1699 between the working tree, the index, and the object database. Git
1700 provides low-level operations which perform each of these steps
1701 individually.</p><p>Generally, all Git operations work on the index file. Some operations
1702 work <span class="strong"><strong>purely</strong></span> on the index file (showing the current state of the
1703 index), but most operations move data between the index file and either
1704 the database or the working directory. Thus there are four main
1705 combinations:</p><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="working-directory-to-index"></a>working directory → index</h3></div></div></div><p>The <a class="ulink" href="git-update-index.html" target="_top">git-update-index(1)</a> command updates the index with
1706 information from the working directory. You generally update the
1707 index information by just specifying the filename you want to update,
1708 like so:</p><pre class="screen">$ git update-index filename</pre><p>but to avoid common mistakes with filename globbing etc., the command
1709 will not normally add totally new entries or remove old entries,
1710 i.e. it will normally just update existing cache entries.</p><p>To tell Git that yes, you really do realize that certain files no
1711 longer exist, or that new files should be added, you
1712 should use the <code class="literal">--remove</code> and <code class="literal">--add</code> flags respectively.</p><p>NOTE! A <code class="literal">--remove</code> flag does <span class="emphasis"><em>not</em></span> mean that subsequent filenames will
1713 necessarily be removed: if the files still exist in your directory
1714 structure, the index will be updated with their new status, not
1715 removed. The only thing <code class="literal">--remove</code> means is that update-index will be
1716 considering a removed file to be a valid thing, and if the file really
1717 does not exist any more, it will update the index accordingly.</p><p>As a special case, you can also do <code class="literal">git update-index --refresh</code>, which
1718 will refresh the "stat" information of each index to match the current
1719 stat information. It will <span class="emphasis"><em>not</em></span> update the object status itself, and
1720 it will only update the fields that are used to quickly test whether
1721 an object still matches its old backing store object.</p><p>The previously introduced <a class="ulink" href="git-add.html" target="_top">git-add(1)</a> is just a wrapper for
1722 <a class="ulink" href="git-update-index.html" target="_top">git-update-index(1)</a>.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="index-to-object-database"></a>index → object database</h3></div></div></div><p>You write your current index file to a "tree" object with the program</p><pre class="screen">$ git write-tree</pre><p>that doesn’t come with any options—it will just write out the
1723 current index into the set of tree objects that describe that state,
1724 and it will return the name of the resulting top-level tree. You can
1725 use that tree to re-generate the index at any time by going in the
1726 other direction:</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="object-database-to-index"></a>object database → index</h3></div></div></div><p>You read a "tree" file from the object database, and use that to
1727 populate (and overwrite—don’t do this if your index contains any
1728 unsaved state that you might want to restore later!) your current
1729 index. Normal operation is just</p><pre class="screen">$ git read-tree <SHA-1 of tree></pre><p>and your index file will now be equivalent to the tree that you saved
1730 earlier. However, that is only your <span class="emphasis"><em>index</em></span> file: your working
1731 directory contents have not been modified.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="index-to-working-directory"></a>index → working directory</h3></div></div></div><p>You update your working directory from the index by "checking out"
1732 files. This is not a very common operation, since normally you’d just
1733 keep your files updated, and rather than write to your working
1734 directory, you’d tell the index files about the changes in your
1735 working directory (i.e. <code class="literal">git update-index</code>).</p><p>However, if you decide to jump to a new version, or check out somebody
1736 else’s version, or just restore a previous tree, you’d populate your
1737 index file with read-tree, and then you need to check out the result
1738 with</p><pre class="screen">$ git checkout-index filename</pre><p>or, if you want to check out all of the index, use <code class="literal">-a</code>.</p><p>NOTE! <code class="literal">git checkout-index</code> normally refuses to overwrite old files, so
1739 if you have an old version of the tree already checked out, you will
1740 need to use the <code class="literal">-f</code> flag (<span class="emphasis"><em>before</em></span> the <code class="literal">-a</code> flag or the filename) to
1741 <span class="emphasis"><em>force</em></span> the checkout.</p><p>Finally, there are a few odds and ends which are not purely moving
1742 from one representation to the other:</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a name="tying-it-all-together"></a>Tying it all together</h3></div></div></div><p>To commit a tree you have instantiated with <code class="literal">git write-tree</code>, you’d
1743 create a "commit" object that refers to that tree and the history
1744 behind it—most notably the "parent" commits that preceded it in
1746 before a certain change was made. However, sometimes it can have two
1747 or more parent commits, in which case we call it a "merge", due to the
1748 fact that such a commit brings together ("merges") two or more
1749 previous states represented by other commits.</p><p>In other words, while a "tree" represents a particular directory state
1750 of a working directory, a "commit" represents that state in time,
1751 and explains how we got there.</p><p>You create a commit object by giving it the tree that describes the
1752 state at the time of the commit, and a list of parents:</p><pre class="screen">$ git commit-tree <tree> -p <parent> [(-p <parent2>)...]</pre><p>and then giving the reason for the commit on stdin (either through
1753 redirection from a pipe or file, or by just typing it at the tty).</p><p><code class="literal">git commit-tree</code> will return the name of the object that represents
1754 that commit, and you should save it away for later use. Normally,
1755 you’d commit a new <code class="literal">HEAD</code> state, and while Git doesn’t care where you
1756 save the note about that state, in practice we tend to just write the
1757 result to the file pointed at by <code class="literal">.git/HEAD</code>, so that we can always see
1758 what the last committed state was.</p><p>Here is a picture that illustrates how various pieces fit together:</p><pre class="screen"> commit-tree
1759 commit obj
1760 +----+
1761 | |
1762 | |
1763 V V
1764 +-----------+
1765 | Object DB |
1766 | Backing |
1767 | Store |
1768 +-----------+
1769 ^
1770 write-tree | |
1771 tree obj | |
1772 | | read-tree
1773 | | tree obj
1774 V
1775 +-----------+
1776 | Index |
1777 | "cache" |
1778 +-----------+
1779 update-index ^
1780 blob obj | |
1781 | |
1782 checkout-index -u | | checkout-index
1783 stat | | blob obj
1784 V
1785 +-----------+
1786 | Working |
1787 | Directory |
1788 +-----------+</pre></div></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="examining-the-data"></a>Examining the data</h2></div></div></div><p>You can examine the data represented in the object database and the
1789 index with various helper tools. For every object, you can use
1790 <a class="ulink" href="git-cat-file.html" target="_top">git-cat-file(1)</a> to examine details about the
1791 object:</p><pre class="screen">$ git cat-file -t <objectname></pre><p>shows the type of the object, and once you have the type (which is
1792 usually implicit in where you find the object), you can use</p><pre class="screen">$ git cat-file blob|tree|commit|tag <objectname></pre><p>to show its contents. NOTE! Trees have binary content, and as a result
1793 there is a special helper for showing that content, called
1796 tend to be small and fairly self-explanatory. In particular, if you
1798 you can do</p><pre class="screen">$ git cat-file commit HEAD</pre><p>to see what the top commit was.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="merging-multiple-trees"></a>Merging multiple trees</h2></div></div></div><p>Git can help you perform a three-way merge, which can in turn be
1799 used for a many-way merge by repeating the merge procedure several
1800 times. The usual situation is that you only do one three-way merge
1801 (reconciling two lines of history) and commit the result, but if
1802 you like to, you can merge several branches in one go.</p><p>To perform a three-way merge, you start with the two commits you
1803 want to merge, find their closest common parent (a third commit),
1804 and compare the trees corresponding to these three commits.</p><p>To get the "base" for the merge, look up the common parent of two
1805 commits:</p><pre class="screen">$ git merge-base <commit1> <commit2></pre><p>This prints the name of a commit they are both based on. You should
1806 now look up the tree objects of those commits, which you can easily
1807 do with</p><pre class="screen">$ git cat-file commit <commitname> | head -1</pre><p>since the tree object information is always the first line in a commit
1809 tree, aka the common tree, and the two "result" trees, aka the branches
1810 you want to merge), you do a "merge" read into the index. This will
1811 complain if it has to throw away your old index contents, so you should
1812 make sure that you’ve committed those—in fact you would normally
1813 always do a merge against your last commit (which should thus match what
1814 you have in your current index anyway).</p><p>To do the merge, do</p><pre class="screen">$ git read-tree -m -u <origtree> <yourtree> <targettree></pre><p>which will do all trivial merge operations for you directly in the
1815 index file, and you can just write the result out with
1816 <code class="literal">git write-tree</code>.</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="merging-multiple-trees-2"></a>Merging multiple trees, continued</h2></div></div></div><p>Sadly, many merges aren’t trivial. If there are files that have
1817 been added, moved or removed, or if both branches have modified the
1818 same file, you will be left with an index tree that contains "merge
1819 entries" in it. Such an index tree can <span class="emphasis"><em>NOT</em></span> be written out to a tree
1820 object, and you will have to resolve any such merge clashes using
1821 other tools before you can write out the result.</p><p>You can examine such index state with <code class="literal">git ls-files --unmerged</code>
1823 $ git ls-files --unmerged
1826 100644 cc44c73eb783565da5831b4d820c962954019b69 3 hello.c</pre><p>Each line of the <code class="literal">git ls-files --unmerged</code> output begins with
1828 filename. The <span class="emphasis"><em>stage number</em></span> is Git’s way to say which tree it
1830 the <code class="literal">HEAD</code> tree, and stage 3 to the <code class="literal">$target</code> tree.</p><p>Earlier we said that trivial merges are done inside
1832 from <code class="literal">$orig</code> to <code class="literal">HEAD</code> or <code class="literal">$target</code>, or if the file changed
1833 from <code class="literal">$orig</code> to <code class="literal">HEAD</code> and <code class="literal">$orig</code> to <code class="literal">$target</code> the same way,
1836 <code class="literal">$orig</code> to <code class="literal">HEAD</code> and <code class="literal">$orig</code> to <code class="literal">$target</code> in a different way.
1837 You could resolve this by running your favorite 3-way merge
1838 program, e.g. <code class="literal">diff3</code>, <code class="literal">merge</code>, or Git’s own merge-file, on
1839 the blob objects from these three stages yourself, like this:</p><pre class="screen">$ git cat-file blob 263414f >hello.c~1
1842 $ git merge-file hello.c~2 hello.c~1 hello.c~3</pre><p>This would leave the merge result in <code class="literal">hello.c~2</code> file, along
1843 with conflict markers if there are conflicts. After verifying
1844 the merge result makes sense, you can tell Git what the final
1846 $ git update-index hello.c</pre><p>When a path is in the "unmerged" state, running <code class="literal">git update-index</code> for
1847 that path tells Git to mark the path resolved.</p><p>The above is the description of a Git merge at the lowest level,
1848 to help you understand what conceptually happens under the hood.
1849 In practice, nobody, not even Git itself, runs <code class="literal">git cat-file</code> three times
1851 stages to temporary files and calls a "merge" script on it:</p><pre class="screen">$ git merge-index git-merge-one-file hello.c</pre><p>and that is what higher level <code class="literal">git merge -s resolve</code> is implemented with.</p></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="hacking-git"></a>Chapter 10. Hacking Git</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#object-details">Object storage format</a></span></dt><dt><span class="section"><a href="#birdview-on-the-source-code">A birds-eye view of Git’s source code</a></span></dt></dl></div><p>This chapter covers internal details of the Git implementation which
1852 probably only Git developers need to understand.</p><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="object-details"></a>Object storage format</h2></div></div></div><p>All objects have a statically determined "type" which identifies the
1853 format of the object (i.e. how it is used, and how it can refer to other
1854 objects). There are currently four different object types: "blob",
1856 characteristics: they are all deflated with zlib, and have a header
1857 that not only specifies their type, but also provides size information
1858 about the data in the object. It’s worth noting that the SHA-1 hash
1859 that is used to name the object is the hash of the original data
1860 plus this header, so <code class="literal">sha1sum</code> <span class="emphasis"><em>file</em></span> does not match the object name
1861 for <span class="emphasis"><em>file</em></span> (the earliest versions of Git hashed slightly differently
1862 but the conclusion is still the same).</p><p>The following is a short example that demonstrates how these hashes
1863 can be generated manually:</p><p>Let’s assume a small text file with some simple content:</p><pre class="screen">$ echo "Hello world" >hello.txt</pre><p>We can now manually generate the hash Git would use for this file:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1864 The object we want the hash for is of type "blob" and its size is
1865 12 bytes.
1867 Prepend the object header to the file content and feed this to
1870 802992c4220de19a90767f3000a79a31b98d0df7 -</pre><p>This manually constructed hash can be verified using <code class="literal">git hash-object</code>
1871 which of course hides the addition of the header:</p><pre class="screen">$ git hash-object hello.txt
1872 802992c4220de19a90767f3000a79a31b98d0df7</pre><p>As a result, the general consistency of an object can always be tested
1873 independently of the contents or the type of the object: all objects can
1874 be validated by verifying that (a) their hashes match the content of the
1875 file and (b) the object successfully inflates to a stream of bytes that
1876 forms a sequence of
1877 <code class="literal"><ascii-type-without-space> + <space> + <ascii-decimal-size> +
1878 <byte\0> + <binary-object-data></code>.</p><p>The structured objects can further have their structure and
1879 connectivity to other objects verified. This is generally done with
1881 of all objects, and verifies their internal consistency (in addition
1882 to just verifying their superficial consistency through the hash).</p></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="birdview-on-the-source-code"></a>A birds-eye view of Git’s source code</h2></div></div></div><p>It is not always easy for new developers to find their way through Git’s
1883 source code. This section gives you a little guidance to show where to
1884 start.</p><p>A good place to start is with the contents of the initial commit, with:</p><pre class="screen">$ git switch --detach e83c5163</pre><p>The initial revision lays the foundation for almost everything Git has
1885 today (even though details may differ in a few places), but is small
1886 enough to read in one sitting.</p><p>Note that terminology has changed since that revision. For example, the
1887 README in that revision uses the word "changeset" to describe what we
1888 now call a <a class="link" href="#def_commit_object">commit</a>.</p><p>Also, we do not call it "cache" any more, but rather "index"; however,
1889 the file is still called <code class="literal">read-cache.h</code>.</p><p>If you grasp the ideas in that initial commit, you should check out a
1890 more recent version and skim <code class="literal">read-cache-ll.h</code>, <code class="literal">object.h</code> and <code class="literal">commit.h</code>.</p><p>In the early days, Git (in the tradition of UNIX) was a bunch of programs
1891 which were extremely simple, and which you used in scripts, piping the
1892 output of one into another. This turned out to be good for initial
1893 development, since it was easier to test new things. However, recently
1894 many of these parts have become builtins, and some of the core has been
1895 "libified", i.e. put into libgit.a for performance, portability reasons,
1896 and to avoid code duplication.</p><p>By now, you know what the index is (and find the corresponding data
1897 structures in <code class="literal">read-cache-ll.h</code>), and that there are just a couple of
1898 object types (blobs, trees, commits and tags) which inherit their
1900 (and thus, you can cast e.g. <code class="literal">(struct object *)commit</code> to achieve the
1901 <span class="emphasis"><em>same</em></span> as <code class="literal">&commit->object</code>, i.e. get at the object name and flags).</p><p>Now is a good point to take a break to let this information sink in.</p><p>Next step: get familiar with the object naming. Read <a class="xref" href="#naming-commits" title="Naming commits">the section called “Naming commits”</a>.
1902 There are quite a few ways to name an object (and not only revisions!).
1903 All of these are handled in <code class="literal">sha1_name.c</code>. Just have a quick look at
1905 functions like <code class="literal">get_sha1_basic()</code> or the likes.</p><p>This is just to get you into the groove for the most libified part of Git:
1906 the revision walker.</p><p>Basically, the initial version of <code class="literal">git log</code> was a shell script:</p><pre class="screen">$ git-rev-list --pretty $(git-rev-parse --default HEAD "$@") | \
1907 LESS=-S ${PAGER:-less}</pre><p>What does this mean?</p><p><code class="literal">git rev-list</code> is the original version of the revision walker, which
1908 <span class="emphasis"><em>always</em></span> printed a list of revisions to stdout. It is still functional,
1909 and needs to, since most new Git commands start out as scripts using
1910 <code class="literal">git rev-list</code>.</p><p><code class="literal">git rev-parse</code> is not as important any more; it was only used to filter out
1911 options that were relevant for the different plumbing commands that were
1912 called by the script.</p><p>Most of what <code class="literal">git rev-list</code> did is contained in <code class="literal">revision.c</code> and
1913 <code class="literal">revision.h</code>. It wraps the options in a struct named <code class="literal">rev_info</code>, which
1914 controls how and what revisions are walked, and more.</p><p>The original job of <code class="literal">git rev-parse</code> is now taken by the function
1915 <code class="literal">setup_revisions()</code>, which parses the revisions and the common command-line
1916 options for the revision walker. This information is stored in the struct
1917 <code class="literal">rev_info</code> for later consumption. You can do your own command-line option
1918 parsing after calling <code class="literal">setup_revisions()</code>. After that, you have to call
1919 <code class="literal">prepare_revision_walk()</code> for initialization, and then you can get the
1920 commits one by one with the function <code class="literal">get_revision()</code>.</p><p>If you are interested in more details of the revision walking process,
1922 <code class="literal">git show v1.3.0~155^2~4</code> and scroll down to that function (note that you
1923 no longer need to call <code class="literal">setup_pager()</code> directly).</p><p>Nowadays, <code class="literal">git log</code> is a builtin, which means that it is <span class="emphasis"><em>contained</em></span> in the
1924 command <code class="literal">git</code>. The source side of a builtin is</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1925 a function called <code class="literal">cmd_<bla></code>, typically defined in <code class="literal">builtin/<bla.c></code>
1926 (note that older versions of Git used to have it in <code class="literal">builtin-<bla>.c</code>
1929 an entry in the <code class="literal">commands[]</code> array in <code class="literal">git.c</code>, and
1931 an entry in <code class="literal">BUILTIN_OBJECTS</code> in the <code class="literal">Makefile</code>.
1932 </li></ul></div><p>Sometimes, more than one builtin is contained in one source file. For
1933 example, <code class="literal">cmd_whatchanged()</code> and <code class="literal">cmd_log()</code> both reside in <code class="literal">builtin/log.c</code>,
1934 since they share quite a bit of code. In that case, the commands which are
1935 <span class="emphasis"><em>not</em></span> named like the <code class="literal">.c</code> file in which they live have to be listed in
1936 <code class="literal">BUILT_INS</code> in the <code class="literal">Makefile</code>.</p><p><code class="literal">git log</code> looks more complicated in C than it does in the original script,
1937 but that allows for a much greater flexibility and performance.</p><p>Here again it is a good point to take a pause.</p><p>Lesson three is: study the code. Really, it is the best way to learn about
1938 the organization of Git (after you know the basic concepts).</p><p>So, think about something which you are interested in, say, "how can I
1939 access a blob just knowing the object name of it?". The first step is to
1940 find a Git command with which you can do it. In this example, it is either
1941 <code class="literal">git show</code> or <code class="literal">git cat-file</code>.</p><p>For the sake of clarity, let’s stay with <code class="literal">git cat-file</code>, because it</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1942 is plumbing, and
1944 was around even in the initial commit (it literally went only through
1945 some 20 revisions as <code class="literal">cat-file.c</code>, was renamed to <code class="literal">builtin/cat-file.c</code>
1946 when made a builtin, and then saw less than 10 versions).
1947 </li></ul></div><p>So, look into <code class="literal">builtin/cat-file.c</code>, search for <code class="literal">cmd_cat_file()</code> and look what
1949 if (argc != 3)
1950 usage("git cat-file [-t|-s|-e|-p|<type>] <sha1>");
1951 if (get_sha1(argv[2], sha1))
1952 die("Not a valid object name %s", argv[2]);</pre><p>Let’s skip over the obvious details; the only really interesting part
1953 here is the call to <code class="literal">get_sha1()</code>. It tries to interpret <code class="literal">argv[2]</code> as an
1954 object name, and if it refers to an object which is present in the current
1955 repository, it writes the resulting SHA-1 into the variable <code class="literal">sha1</code>.</p><p>Two things are interesting here:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
1956 <code class="literal">get_sha1()</code> returns 0 on <span class="emphasis"><em>success</em></span>. This might surprise some new
1957 Git hackers, but there is a long tradition in UNIX to return different
1958 negative numbers in case of different errors—and 0 on success.
1960 the variable <code class="literal">sha1</code> in the function signature of <code class="literal">get_sha1()</code> is <code class="literal">unsigned
1963 commit. Note that whenever a SHA-1 is passed as <code class="literal">unsigned char *</code>, it
1964 is the binary representation, as opposed to the ASCII representation in
1966 </li></ul></div><p>You will see both of these things throughout the code.</p><p>Now, for the meat:</p><pre class="screen"> case 0:
1967 buf = read_object_with_reference(sha1, argv[1], &size, NULL);</pre><p>This is how you read a blob (actually, not only a blob, but any type of
1968 object). To know how the function <code class="literal">read_object_with_reference()</code> actually
1971 the source.</p><p>To find out how the result can be used, just read on in <code class="literal">cmd_cat_file()</code>:</p><pre class="screen"> write_or_die(1, buf, size);</pre><p>Sometimes, you do not know where to look for a feature. In many such cases,
1972 it helps to search through the output of <code class="literal">git log</code>, and then <code class="literal">git show</code> the
1973 corresponding commit.</p><p>Example: If you know that there was some test case for <code class="literal">git bundle</code>, but
1974 do not remember where it was (yes, you <span class="emphasis"><em>could</em></span> <code class="literal">git grep bundle t/</code>, but that
1975 does not illustrate the point!):</p><pre class="screen">$ git log --no-merges t/</pre><p>In the pager (<code class="literal">less</code>), just search for "bundle", go a few lines back,
1976 and see that it is in commit 18449ab0. Now just copy this object name,
1977 and paste it into the command line</p><pre class="screen">$ git show 18449ab0</pre><p>Voila.</p><p>Another example: Find out what to do in order to make some script a
1978 builtin:</p><pre class="screen">$ git log --no-merges --diff-filter=A builtin/*.c</pre><p>You see, Git is actually the best tool to find out about the source of Git
1979 itself!</p></div></div><div class="chapter"><div class="titlepage"><div><div><h1 class="title"><a name="glossary"></a>Chapter 11. Git Glossary</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#git-explained">Git explained</a></span></dt></dl></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="git-explained"></a>Git explained</h2></div></div></div><div class="variablelist"><dl class="variablelist"><dt><span class="term">
1981 </span></dt><dd>
1984 from another object database, which is called an "alternate".
1987 </span></dt><dd>
1988 A bare repository is normally an appropriately
1989 named <a class="link" href="#def_directory">directory</a> with a <code class="literal">.git</code> suffix that does not
1990 have a locally checked-out copy of any of the files under
1991 revision control. That is, all of the Git
1992 administrative and control files that would normally be present in the
1995 and no other files are present and checked out. Usually publishers of
1996 public repositories make bare repositories available.
1999 </span></dt><dd>
2003 </span></dt><dd>
2004 A "branch" is a line of development. The most recent
2006 that branch. The tip of the branch is <a class="link" href="#def_ref">referenced</a> by a branch
2008 is done on the branch. A single Git
2015 </span></dt><dd>
2019 </span></dt><dd>
2020 A list of objects, where each <a class="link" href="#def_object">object</a> in the list contains
2021 a reference to its successor (for example, the successor of a
2022 <a class="link" href="#def_commit">commit</a> could be one of its <a class="link" href="#def_parent">parents</a>).
2025 </span></dt><dd>
2027 store changes, but states, it really does not make sense to use the term
2028 "changesets" with Git.
2031 </span></dt><dd>
2032 The action of updating all or part of the
2033 <a class="link" href="#def_working_tree">working tree</a> with a <a class="link" href="#def_tree_object">tree object</a>
2036 <a class="link" href="#def_index">index</a> and <a class="link" href="#def_HEAD">HEAD</a> if the whole working tree has
2040 </span></dt><dd>
2042 changes out of a series of changes (typically commits) and record them
2043 as a new series of changes on top of a different codebase. In Git, this is
2044 performed by the "git cherry-pick" command to extract the change introduced
2049 </span></dt><dd>
2052 <a class="link" href="#def_head">head</a>. Also see "<a class="link" href="#def_dirty">dirty</a>".
2056 As a noun: A single point in the
2057 Git history; the entire history of a project is represented as a
2058 set of interrelated commits. The word "commit" is often
2059 used by Git in the same places other revision control systems
2063 state in the Git history, by creating a new commit representing the current
2064 state of the <a class="link" href="#def_index">index</a> and advancing <a class="link" href="#def_HEAD">HEAD</a>
2067 </span></dt><dd>
2071 This structure is the definitive commit graph. The
2072 graph can be represented in other ways, e.g. the
2076 </span></dt><dd>
2077 The "commit-graph" (normally hyphenated) file is a supplemental
2079 which accelerates commit graph walks. The "commit-graph" file is
2080 stored either in the .git/objects/info directory or in the info
2081 directory of an alternate object database.
2084 </span></dt><dd>
2086 particular <a class="link" href="#def_revision">revision</a>, such as <a class="link" href="#def_parent">parents</a>, committer,
2089 revision.
2092 </span></dt><dd>
2093 A <a class="link" href="#def_commit_object">commit object</a> or an <a class="link" href="#def_object">object</a> that
2094 can be recursively <a class="link" href="#def_dereference">dereferenced</a> to a commit object.
2095 The following are all commit-ishes:
2096 a commit object,
2098 object,
2099 a tag object that points to a tag object that points to a
2100 commit object,
2101 etc.
2104 </span></dt><dd>
2105 Fundamental data structures and utilities of Git. Exposes only limited
2106 source code management tools.
2109 </span></dt><dd>
2110 Directed acyclic graph. The <a class="link" href="#def_commit_object">commit objects</a> form a
2111 directed acyclic graph, because they have parents (directed), and the
2116 </span></dt><dd>
2119 dangling object has no references to it from any
2120 reference or <a class="link" href="#def_object">object</a> in the <a class="link" href="#def_repository">repository</a>.
2124 Referring to a <a class="link" href="#def_symref">symbolic ref</a>: the action of accessing the
2126 dereferencing involves repeating the aforementioned process on the
2127 resulting ref until a non-symbolic reference is found.
2128 </p><p class="simpara">Referring to a <a class="link" href="#def_tag_object">tag object</a>: the action of accessing the
2129 <a class="link" href="#def_object">object</a> a tag points at. Tags are recursively dereferenced by
2130 repeating the operation on the result object until the result has either a
2131 specified <a class="link" href="#def_object_type">object type</a> (where applicable) or any non-"tag"
2132 object type. A synonym for "recursive dereference" in the context of tags is
2133 "<a class="link" href="#def_peel">peel</a>".</p><p class="simpara">Referring to a <a class="link" href="#def_commit_object">commit object</a>: the action of accessing
2134 the commit’s tree object. Commits cannot be dereferenced recursively.</p><p class="simpara">Unless otherwise specified, "dereferencing" as it used in the context of Git
2140 history HEAD represents operate on the history leading to the
2141 tip of the branch the HEAD points at. However, Git also
2144 particular branch. The HEAD in such a state is called
2145 "detached".
2148 while the HEAD is detached. They update the HEAD to point at the tip
2149 of the updated history without affecting any branch. Commands that
2150 update or inquire information <span class="emphasis"><em>about</em></span> the current branch (e.g. <code class="literal">git
2151 branch --set-upstream-to</code> that sets what remote-tracking branch the
2152 current branch integrates with) obviously do not work, as there is no
2155 </span></dt><dd>
2156 The list you get with "ls" :-)
2159 </span></dt><dd>
2161 it contains modifications which have not been <a class="link" href="#def_commit">committed</a> to the current
2165 </span></dt><dd>
2170 </span></dt><dd>
2171 A fast-forward is a special type of <a class="link" href="#def_merge">merge</a> where you have a
2175 <a class="link" href="#def_commit">commit</a> but instead just update your branch to point at the same
2176 revision as the branch you are merging. This will happen frequently on a
2181 </span></dt><dd>
2186 and to get them, too. See also <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a>.
2189 </span></dt><dd>
2190 Linus Torvalds originally designed Git to be a user space file system,
2191 i.e. the infrastructure to hold files and directories. That ensured the
2192 efficiency and speed of Git.
2195 </span></dt><dd>
2199 </span></dt><dd>
2201 points at the directory that is the real repository.
2202 For proper use see <a class="ulink" href="git-worktree.html" target="_top">git-worktree(1)</a> or <a class="ulink" href="git-submodule.html" target="_top">git-submodule(1)</a>.
2203 For syntax see <a class="ulink" href="gitrepository-layout.html" target="_top">gitrepository-layout(5)</a>.
2207 Grafts enable two otherwise different lines of development to be joined
2208 together by recording fake ancestry information for commits. This way
2209 you can make Git pretend the set of <a class="link" href="#def_parent">parents</a> a <a class="link" href="#def_commit">commit</a> has
2210 is different from what was recorded when the commit was
2213 transferring objects between repositories; see <a class="ulink" href="git-replace.html" target="_top">git-replace(1)</a>
2216 </span></dt><dd>
2220 </span></dt><dd>
2221 A <a class="link" href="#def_ref">named reference</a> to the <a class="link" href="#def_commit">commit</a> at the tip of a
2223 <code class="literal">$GIT_DIR/refs/heads/</code> directory, except when using packed refs. (See
2227 </span></dt><dd>
2228 The current <a class="link" href="#def_branch">branch</a>. In more detail: Your <a class="link" href="#def_working_tree">working tree</a> is normally derived from the state of the tree
2229 referred to by HEAD. HEAD is a reference to one of the
2232 references an arbitrary commit.
2235 </span></dt><dd>
2239 </span></dt><dd>
2240 During the normal execution of several Git commands, call-outs are made
2241 to optional scripts that allow a developer to add functionality or
2242 checking. Typically, the hooks allow for a command to be pre-verified
2243 and potentially aborted, and allow for a post-notification after the
2244 operation is done. The hook scripts are found in the
2246 removing the <code class="literal">.sample</code> suffix from the filename. In earlier versions
2247 of Git you had to make them executable.
2250 </span></dt><dd>
2251 A collection of files with stat information, whose contents are stored
2252 as objects. The index is a stored version of your
2253 <a class="link" href="#def_working_tree">working tree</a>. Truth be told, it can also contain a second, and even
2254 a third version of a working tree, which are used
2258 </span></dt><dd>
2259 The information regarding a particular file, stored in the
2262 the index contains multiple versions of that file).
2265 </span></dt><dd>
2268 "master" is created, and becomes the active branch. In most
2269 cases, this contains the local development, though that is
2270 purely by convention and is not required.
2274 As a verb: To bring the contents of another
2277 case where the merged-in branch is from a different repository,
2279 and then merging the result into the current branch. This
2280 combination of fetch and merge operations is called a
2282 that identifies changes made since the branches diverged, and
2283 then applies all those changes together. In cases where changes
2284 conflict, manual intervention may be required to complete the
2285 merge.
2286 </p><p class="simpara">As a noun: unless it is a <a class="link" href="#def_fast_forward">fast-forward</a>, a
2287 successful merge results in the creation of a new <a class="link" href="#def_commit">commit</a>
2288 representing the result of the merge, and having as
2289 <a class="link" href="#def_parent">parents</a> the tips of the merged <a class="link" href="#def_branch">branches</a>.
2290 This commit is referred to as a "merge commit", or sometimes just a
2293 </span></dt><dd>
2294 The unit of storage in Git. It is uniquely identified by the
2296 object cannot be changed.
2299 </span></dt><dd>
2305 </span></dt><dd>
2309 </span></dt><dd>
2311 object name is usually represented by a 40 character
2315 </span></dt><dd>
2317 "<a class="link" href="#def_tree_object">tree</a>", "<a class="link" href="#def_tag_object">tag</a>" or
2322 </span></dt><dd>
2323 To <a class="link" href="#def_merge">merge</a> more than two <a class="link" href="#def_branch">branches</a>.
2326 </span></dt><dd>
2329 such an operation, the commit first created becomes a commit
2330 without a parent, starting a new history.
2333 </span></dt><dd>
2335 at least one upstream project which they track. By default
2337 will be fetched into <a class="link" href="#def_remote_tracking_branch">remote-tracking branches</a> named
2338 origin/name-of-upstream-branch, which you can see using
2342 </span></dt><dd>
2343 Only update and add files to the working directory, but don’t
2344 delete them, similar to how <span class="emphasis"><em>cp -R</em></span> would update the contents
2345 in the destination directory. This is the default mode in a
2347 <a class="link" href="#def_index">index</a> or a <a class="link" href="#def_tree-ish">tree-ish</a>. In
2348 contrast, no-overlay mode also deletes tracked files not
2352 </span></dt><dd>
2353 A set of objects which have been compressed into one file (to save space
2354 or to transmit them efficiently).
2357 </span></dt><dd>
2358 The list of identifiers, and other information, of the objects in a
2359 <a class="link" href="#def_pack">pack</a>, to assist in efficiently accessing the contents of a
2360 pack.
2364 Pattern used to limit paths in Git commands.
2367 and many other commands to
2368 limit the scope of operations to some subset of the tree or
2369 working tree. See the documentation of each command for whether
2370 paths are relative to the current directory or toplevel. The
2371 pathspec syntax is as follows:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
2372 any path matches itself
2374 the pathspec up to the last slash represents a
2375 directory prefix. The scope of that pathspec is
2376 limited to that subtree.
2378 the rest of the pathspec is a pattern for the remainder
2379 of the pathname. Paths relative to the directory
2380 prefix will be matched against that pattern using fnmatch(3);
2381 in particular, <span class="emphasis"><em>*</em></span> and <span class="emphasis"><em>?</em></span> <span class="emphasis"><em>can</em></span> match directory separators.
2383 in the Documentation subtree,
2384 including Documentation/chapter_1/figure_1.jpg.</p><p class="simpara">A pathspec that begins with a colon <code class="literal">:</code> has special meaning. In the
2385 short form, the leading colon <code class="literal">:</code> is followed by zero or more "magic
2386 signature" letters (which optionally is terminated by another colon <code class="literal">:</code>),
2387 and the remainder is the pattern to match against the path.
2388 The "magic signature" consists of ASCII symbols that are neither
2389 alphanumeric, glob, regex special characters nor colon.
2390 The optional colon that terminates the "magic signature" can be
2391 omitted if the pattern begins with a character that does not belong to
2392 "magic signature" symbol set and is not a colon.</p><p class="simpara">In the long form, the leading colon <code class="literal">:</code> is followed by an open
2393 parenthesis <code class="literal">(</code>, a comma-separated list of zero or more "magic words",
2394 and a close parentheses <code class="literal">)</code>, and the remainder is the pattern to match
2395 against the path.</p><p class="simpara">A pathspec with only a colon means "there is no pathspec". This form
2396 should not be combined with other pathspec.</p><div class="variablelist"><dl class="variablelist"><dt><span class="term">
2397 top
2398 </span></dt><dd>
2399 The magic word <code class="literal">top</code> (magic signature: <code class="literal">/</code>) makes the pattern
2400 match from the root of the working tree, even when you are
2401 running the command from inside a subdirectory.
2403 literal
2404 </span></dt><dd>
2405 Wildcards in the pattern such as <code class="literal">*</code> or <code class="literal">?</code> are treated
2406 as literal characters.
2408 icase
2409 </span></dt><dd>
2410 Case insensitive match.
2412 glob
2414 Git treats the pattern as a shell glob suitable for
2415 consumption by fnmatch(3) with the FNM_PATHNAME flag:
2416 wildcards in the pattern will not match a / in the pathname.
2417 For example, "Documentation/*.html" matches
2419 or "tools/perf/Documentation/perf.html".
2420 </p><p class="simpara">Two consecutive asterisks ("<code class="literal">**</code>") in patterns matched against
2421 full pathname may have special meaning:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
2424 "<code class="literal">foo</code>" anywhere, the same as pattern "<code class="literal">foo</code>". "<code class="literal">**/foo/bar</code>"
2432 A slash followed by two consecutive asterisks then a slash
2434 matches "<code class="literal">a/b</code>", "<code class="literal">a/x/b</code>", "<code class="literal">a/x/y/b</code>" and so on.
2436 Other consecutive asterisks are considered invalid.
2437 </p><p class="simpara">Glob magic is incompatible with literal magic.</p></li></ul></div></dd><dt><span class="term">
2438 attr
2441 requirements", all of which must be met in order for the
2442 path to be considered a match; this is in addition to the
2443 usual non-magic pathspec pattern matching.
2446 these forms:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
2447 "<code class="literal">ATTR</code>" requires that the attribute <code class="literal">ATTR</code> be set.
2449 "<code class="literal">-ATTR</code>" requires that the attribute <code class="literal">ATTR</code> be unset.
2451 "<code class="literal">ATTR=VALUE</code>" requires that the attribute <code class="literal">ATTR</code> be
2454 "<code class="literal">!ATTR</code>" requires that the attribute <code class="literal">ATTR</code> be
2455 unspecified.
2457 obtained from working tree, not from the given tree object.</p></li></ul></div></dd><dt><span class="term">
2458 exclude
2459 </span></dt><dd>
2460 After a path matches any non-exclude pathspec, it will be run
2463 is no non-exclude pathspec, the exclusion is applied to the
2464 result set as if invoked without any pathspec.
2467 </span></dt><dd>
2469 of the logical predecessor(s) in the line of development, i.e. its
2470 parents.
2473 </span></dt><dd>
2478 </span></dt><dd>
2480 routines that help select changes that add or delete a given text
2481 string. With the <code class="literal">--pickaxe-all</code> option, it can be used to view the full
2483 particular line of text. See <a class="ulink" href="git-diff.html" target="_top">git-diff(1)</a>.
2486 </span></dt><dd>
2490 </span></dt><dd>
2491 Cute name for programs and program suites depending on
2497 </span></dt><dd>
2501 unusual refs.
2504 </span></dt><dd>
2506 like refs for the purposes of rev-parse, but which are treated
2507 specially by git. Pseudorefs both have names that are all-caps,
2508 and always start with a line consisting of a
2510 pseudoref, because it is sometimes a symbolic ref. They might
2514 be symbolic refs, and never have reflogs. They also cannot be
2515 updated through the normal ref update machinery. Instead,
2516 they are updated by directly writing to the files. However,
2518 MERGE_HEAD</code> will work.
2521 </span></dt><dd>
2522 Pulling a <a class="link" href="#def_branch">branch</a> means to <a class="link" href="#def_fetch">fetch</a> it and
2523 <a class="link" href="#def_merge">merge</a> it. See also <a class="ulink" href="git-pull.html" target="_top">git-pull(1)</a>.
2526 </span></dt><dd>
2528 <a class="link" href="#def_head_ref">head ref</a> from a remote <a class="link" href="#def_repository">repository</a>,
2529 find out if it is an ancestor to the branch’s local
2530 head ref, and in that case, putting all
2532 head ref, and which are missing from the remote
2533 repository, into the remote
2536 ancestor to the local head, the push fails.
2539 </span></dt><dd>
2541 "reachable" from that commit. More
2543 another if we can reach the one from the other by a <a class="link" href="#def_chain">chain</a>
2546 <a class="link" href="#def_tree_object">trees</a> to the trees or <a class="link" href="#def_blob_object">blobs</a>
2547 that they contain.
2550 </span></dt><dd>
2551 Reachability bitmaps store information about the
2553 a packfile, or a multi-pack index (MIDX), to speed up object search.
2554 The bitmaps are stored in a ".bitmap" file. A repository may have at
2555 most one bitmap file in use. The bitmap file may belong to either one
2556 pack, or the repository’s multi-pack index (if it exists).
2559 </span></dt><dd>
2562 to the result.
2566 A name that begins with <code class="literal">refs/</code> (e.g. <code class="literal">refs/heads/master</code>)
2569 For convenience, a ref can sometimes be abbreviated when used
2570 as an argument to a Git command; see <a class="ulink" href="gitrevisions.html" target="_top">gitrevisions(7)</a>
2571 for details.
2574 Different subhierarchies are used for different purposes (e.g. the
2575 <code class="literal">refs/heads/</code> hierarchy is used to represent local branches).</p><p class="simpara">There are a few special-purpose refs that do not begin with <code class="literal">refs/</code>.
2578 </span></dt><dd>
2579 A reflog shows the local "history" of a ref. In other words,
2580 it can tell you what the 3rd last revision in <span class="emphasis"><em>this</em></span> repository
2582 yesterday 9:14pm. See <a class="ulink" href="git-reflog.html" target="_top">git-reflog(1)</a> for details.
2585 </span></dt><dd>
2591 </span></dt><dd>
2593 project but resides somewhere else. To communicate with remotes,
2597 </span></dt><dd>
2600 <span class="emphasis"><em>refs/remotes/foo/bar</em></span> (indicating that it tracks a branch named
2601 <span class="emphasis"><em>bar</em></span> in a remote named <span class="emphasis"><em>foo</em></span>), and matches the right-hand-side of
2603 branch should not contain direct modifications or have local
2604 commits made to it.
2607 </span></dt><dd>
2611 accompanied by meta data from one or more <a class="link" href="#def_porcelain">porcelains</a>. A
2612 repository can share an object database with other repositories
2616 </span></dt><dd>
2617 The action of fixing up manually what a failed automatic
2621 </span></dt><dd>
2625 </span></dt><dd>
2626 To throw away part of the development, i.e. to assign the
2627 <a class="link" href="#def_head">head</a> to an earlier <a class="link" href="#def_revision">revision</a>.
2630 </span></dt><dd>
2631 Source code management (tool).
2634 </span></dt><dd>
2635 "Secure Hash Algorithm 1"; a cryptographic hash function.
2636 In the context of Git used as a synonym for <a class="link" href="#def_object_name">object name</a>.
2639 </span></dt><dd>
2641 but the phrase makes it more explicit that it was created by
2645 </span></dt><dd>
2647 history some of whose <a class="link" href="#def_commit">commits</a> have <a class="link" href="#def_parent">parents</a> cauterized away (in other
2648 words, Git is told to pretend that these commits do not have the
2649 parents, even though they are recorded in the <a class="link" href="#def_commit_object">commit object</a>). This is sometimes useful when you are interested only in the
2650 recent history of a project even though the real history recorded in the
2651 upstream is much larger. A shallow repository
2652 is created by giving the <code class="literal">--depth</code> option to <a class="ulink" href="git-clone.html" target="_top">git-clone(1)</a>, and
2653 its history can be later deepened with <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a>.
2656 </span></dt><dd>
2662 A ref that has different semantics than normal refs. These refs can be
2663 accessed via normal Git commands but may not behave the same as a
2664 normal ref in some cases.
2665 </p><p class="simpara">The following special refs are known to Git:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
2666 "<code class="literal">FETCH_HEAD</code>" is written by <a class="ulink" href="git-fetch.html" target="_top">git-fetch(1)</a> or <a class="ulink" href="git-pull.html" target="_top">git-pull(1)</a>. It
2667 may refer to multiple object IDs. Each object ID is annotated with metadata
2668 indicating where it was fetched from and its fetch status.
2670 "<code class="literal">MERGE_HEAD</code>" is written by <a class="ulink" href="git-merge.html" target="_top">git-merge(1)</a> when resolving merge
2671 conflicts. It contains all commit IDs which are being merged.
2674 </span></dt><dd>
2676 separate project inside another repository (the latter of
2680 </span></dt><dd>
2683 The superproject knows about the names of (but does not hold
2684 copies of) commit objects of the contained submodules.
2687 </span></dt><dd>
2689 itself, it is of the format <span class="emphasis"><em>ref: refs/some/thing</em></span> and when referenced,
2691 <span class="emphasis"><em><a class="link" href="#def_HEAD">HEAD</a></em></span> is a prime example of a symref. Symbolic references
2692 are manipulated with the <a class="ulink" href="git-symbolic-ref.html" target="_top">git-symbolic-ref(1)</a> command.
2695 </span></dt><dd>
2696 A <a class="link" href="#def_ref">ref</a> under <code class="literal">refs/tags/</code> namespace that points to an
2697 object of an arbitrary type (typically a tag points to either a
2698 <a class="link" href="#def_tag_object">tag</a> or a <a class="link" href="#def_commit_object">commit object</a>).
2702 in Git’s context). A tag is most typically used to mark a particular
2706 </span></dt><dd>
2707 An <a class="link" href="#def_object">object</a> containing a <a class="link" href="#def_ref">ref</a> pointing to
2708 another object, which can contain a message just like a
2710 signature, in which case it is called a "signed tag object".
2713 </span></dt><dd>
2715 identify a conceptual line of development. Since branches are very easy
2716 and inexpensive, it is often desirable to have several small branches
2717 that each contain very well defined concepts or small incremental yet
2718 related changes.
2721 </span></dt><dd>
2722 Either a <a class="link" href="#def_working_tree">working tree</a>, or a <a class="link" href="#def_tree_object">tree object</a> together with the dependent <a class="link" href="#def_blob_object">blob</a> and tree objects
2723 (i.e. a stored representation of a working tree).
2726 </span></dt><dd>
2727 An <a class="link" href="#def_object">object</a> containing a list of file names and modes along
2728 with refs to the associated blob and/or tree objects. A
2729 <a class="link" href="#def_tree">tree</a> is equivalent to a <a class="link" href="#def_directory">directory</a>.
2732 </span></dt><dd>
2733 A <a class="link" href="#def_tree_object">tree object</a> or an <a class="link" href="#def_object">object</a> that can
2738 The following are all tree-ishes:
2740 a tree object,
2742 a tag object that points to a tag object that points to a tree
2743 object,
2744 etc.
2747 </span></dt><dd>
2748 The <a class="link" href="#def_HEAD">HEAD</a> can point at a <a class="link" href="#def_branch">branch</a>
2749 that does not yet exist and that does not have any commit on
2750 it yet, and such a branch is called an unborn branch. The
2751 most typical way users encounter an unborn branch is by
2752 creating a repository anew without cloning from elsewhere.
2753 The HEAD would point at the <span class="emphasis"><em>main</em></span> (or <span class="emphasis"><em>master</em></span>, depending
2754 on your configuration) branch that is yet to be born. Also
2755 some operations can get you on an unborn branch with their
2759 </span></dt><dd>
2764 </span></dt><dd>
2765 An <a class="link" href="#def_object">object</a> which is not <a class="link" href="#def_reachable">reachable</a> from a
2766 <a class="link" href="#def_branch">branch</a>, <a class="link" href="#def_tag">tag</a>, or any other reference.
2769 </span></dt><dd>
2771 question (or the branch in question is rebased onto). It is configured
2773 of <span class="emphasis"><em>A</em></span> is <span class="emphasis"><em>origin/B</em></span> sometimes we say "<span class="emphasis"><em>A</em></span> is tracking <span class="emphasis"><em>origin/B</em></span>".
2776 </span></dt><dd>
2777 The tree of actual checked out files. The working tree normally
2779 plus any local changes that you have made but not yet committed.
2782 </span></dt><dd>
2783 A repository can have zero (i.e. bare repository) or one or
2784 more worktrees attached to it. One "worktree" consists of a
2785 "working tree" and repository metadata, most of which are
2786 shared among other worktrees of a single repository, and
2787 some of which are maintained separately per worktree
2788 (e.g. the index, HEAD and pseudorefs like MERGE_HEAD,
2789 per-worktree refs and per-worktree configuration file).
2790 </dd></dl></div></div></div><div class="appendix"><div class="titlepage"><div><div><h1 class="title"><a name="git-quick-start"></a>Appendix A. Git Quick Reference</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#quick-creating-a-new-repository">Creating a new repository</a></span></dt><dt><span class="section"><a href="#managing-branches">Managing branches</a></span></dt><dt><span class="section"><a href="#exploring-history">Exploring history</a></span></dt><dt><span class="section"><a href="#making-changes">Making changes</a></span></dt><dt><span class="section"><a href="#merging">Merging</a></span></dt><dt><span class="section"><a href="#sharing-your-changes">Sharing your changes</a></span></dt><dt><span class="section"><a href="#repository-maintenance">Repository maintenance</a></span></dt></dl></div><p>This is a quick summary of the major commands; the previous chapters
2791 explain how these work in more detail.</p><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="quick-creating-a-new-repository"></a>Creating a new repository</h2></div></div></div><p>From a tarball:</p><pre class="screen">$ tar xzf project.tar.gz
2792 $ cd project
2793 $ git init
2794 Initialized empty Git repository in .git/
2795 $ git add .
2796 $ git commit</pre><p>From a remote repository:</p><pre class="screen">$ git clone git://example.com/pub/project.git
2797 $ cd project</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="managing-branches"></a>Managing branches</h2></div></div></div><pre class="screen">$ git branch # list all local branches in this repo
2798 $ git switch test # switch working directory to branch "test"
2799 $ git branch new # create branch "new" starting at current HEAD
2800 $ git branch -d new # delete branch "new"</pre><p>Instead of basing a new branch on current HEAD (the default), use:</p><pre class="screen">$ git branch new test # branch named "test"
2802 $ git branch new HEAD^ # commit before the most recent
2803 $ git branch new HEAD^^ # commit before that
2804 $ git branch new test~10 # ten commits before tip of branch "test"</pre><p>Create and switch to a new branch at the same time:</p><pre class="screen">$ git switch -c new v2.6.15</pre><p>Update and examine branches from the repository you cloned from:</p><pre class="screen">$ git fetch # update
2805 $ git branch -r # list
2806 origin/master
2807 origin/next
2808 ...
2809 $ git switch -c masterwork origin/master</pre><p>Fetch a branch from a different repository, and give it a new
2810 name in your repository:</p><pre class="screen">$ git fetch git://example.com/project.git theirbranch:mybranch
2811 $ git fetch git://example.com/project.git v2.6.15:mybranch</pre><p>Keep a list of repositories you work with regularly:</p><pre class="screen">$ git remote add example git://example.com/project.git
2812 $ git remote # list remote repositories
2813 example
2814 origin
2815 $ git remote show example # get details
2816 * remote example
2817 URL: git://example.com/project.git
2818 Tracked remote branches
2819 master
2820 next
2821 ...
2822 $ git fetch example # update branches from example
2823 $ git branch -r # list all remote branches</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="exploring-history"></a>Exploring history</h2></div></div></div><pre class="screen">$ gitk # visualize and browse history
2824 $ git log # list all commits
2825 $ git log src/ # ...modifying src/
2827 $ git log master..test # ...in branch test, not in branch master
2828 $ git log test..master # ...in branch master, but not in test
2829 $ git log test...master # ...in one branch, not in both
2830 $ git log -S'foo()' # ...where difference contain "foo()"
2832 $ git log -p # show patches as well
2833 $ git show # most recent commit
2835 $ git diff v2.6.15..HEAD # diff with current head
2838 $ git show v2.6.15:a.txt # look at old version of a.txt</pre><p>Search for regressions:</p><pre class="screen">$ git bisect start
2839 $ git bisect bad # current version is bad
2840 $ git bisect good v2.6.13-rc2 # last known good revision
2841 Bisecting: 675 revisions left to test after this
2842 # test here, then:
2843 $ git bisect good # if this revision is good, or
2844 $ git bisect bad # if this revision is bad.
2845 # repeat until done.</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="making-changes"></a>Making changes</h2></div></div></div><p>Make sure Git knows who to blame:</p><pre class="screen">$ cat >>~/.gitconfig <<\EOF
2846 [user]
2847 name = Your Name Comes Here
2848 email = you@yourdomain.example.com
2849 EOF</pre><p>Select file contents to include in the next commit, then make the
2851 $ git add b.txt # new file
2852 $ git rm c.txt # old file
2853 $ git commit</pre><p>Or, prepare and create the commit in one step:</p><pre class="screen">$ git commit d.txt # use latest content only of d.txt
2854 $ git commit -a # use latest content of all tracked files</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="merging"></a>Merging</h2></div></div></div><pre class="screen">$ git merge test # merge branch "test" into the current branch
2855 $ git pull git://example.com/project.git master
2856 # fetch and merge in remote branch
2857 $ git pull . test # equivalent to git merge test</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="sharing-your-changes"></a>Sharing your changes</h2></div></div></div><p>Importing or exporting patches:</p><pre class="screen">$ git format-patch origin..HEAD # format a patch for each commit
2858 # in HEAD but not in origin
2859 $ git am mbox # import patches from the mailbox "mbox"</pre><p>Fetch a branch in a different Git repository, then merge into the
2860 current branch:</p><pre class="screen">$ git pull git://example.com/project.git theirbranch</pre><p>Store the fetched branch into a local branch before merging into the
2861 current branch:</p><pre class="screen">$ git pull git://example.com/project.git theirbranch:mybranch</pre><p>After creating commits on a local branch, update the remote
2862 branch with your commits:</p><pre class="screen">$ git push ssh://example.com/project.git mybranch:theirbranch</pre><p>When remote and local branch are both named "test":</p><pre class="screen">$ git push ssh://example.com/project.git test</pre><p>Shortcut version for a frequently used remote repository:</p><pre class="screen">$ git remote add example ssh://example.com/project.git
2863 $ git push example test</pre></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="repository-maintenance"></a>Repository maintenance</h2></div></div></div><p>Check for corruption:</p><pre class="screen">$ git fsck</pre><p>Recompress, remove unused cruft:</p><pre class="screen">$ git gc</pre></div></div><div class="appendix"><div class="titlepage"><div><div><h1 class="title"><a name="todo"></a>Appendix B. Notes and todo list for this manual</h1></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl class="toc"><dt><span class="section"><a href="#todo-list">Todo list</a></span></dt></dl></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="todo-list"></a>Todo list</h2></div></div></div><p>This is a work in progress.</p><p>The basic requirements:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
2864 It must be readable in order, from beginning to end, by someone
2865 intelligent with a basic grasp of the UNIX command line, but without
2866 any special knowledge of Git. If necessary, any other prerequisites
2867 should be specifically mentioned as they arise.
2869 Whenever possible, section headings should clearly describe the task
2870 they explain how to do, in language that requires no more knowledge
2871 than necessary: for example, "importing patches into a project" rather
2873 </li></ul></div><p>Think about how to create a clear chapter dependency graph that will
2874 allow people to get to important topics without necessarily reading
2875 everything in between.</p><p>Scan <code class="literal">Documentation/</code> for other stuff left out; in particular:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
2876 howto’s
2880 hooks
2883 </li></ul></div><p>Scan email archives for other stuff left out</p><p>Scan man pages to see if any assume more background than this manual
2884 provides.</p><p>Add more good examples. Entire sections of just cookbook examples
2885 might be a good idea; maybe make an "advanced examples" section a
2886 standard end-of-chapter section?</p><p>Include cross-references to the glossary, where appropriate.</p><p>Add a section on working with other version control systems, including
2887 CVS, Subversion, and just imports of series of release tarballs.</p><p>Write a chapter on using plumbing and writing scripts.</p><p>Alternates, clone -reference, etc.</p><p>More on recovery from repository corruption. See:
2888 <a class="ulink" href="https://lore.kernel.org/git/Pine.LNX.4.64.0702272039540.12485@woody.linux-foundation.org/" target="_top">https://lore.kernel.org/git/Pine.LNX.4.64.0702272039540.12485@woody.linux-foundation.org/</a>
2889 <a class="ulink" href="https://lore.kernel.org/git/Pine.LNX.4.64.0702141033400.3604@woody.linux-foundation.org/" target="_top">https://lore.kernel.org/git/Pine.LNX.4.64.0702141033400.3604@woody.linux-foundation.org/</a></p></div></div></div></body></html>