6 git-clone - Clone a repository into a new directory
12 'git clone' [--template=<template_directory>]
13 [-l] [-s] [--no-hardlinks] [-q] [-n] [--bare] [--mirror]
14 [-o <name>] [-b <name>] [-u <upload-pack>] [--reference <repository>]
15 [--separate-git-dir|-L <git dir>]
16 [--depth <depth>] [--recursive|--recurse-submodules] [--] <repository>
22 Clones a repository into a newly created directory, creates
23 remote-tracking branches for each branch in the cloned repository
24 (visible using `git branch -r`), and creates and checks out an
25 initial branch that is forked from the cloned repository's
26 currently active branch.
28 After the clone, a plain `git fetch` without arguments will update
29 all the remote-tracking branches, and a `git pull` without
30 arguments will in addition merge the remote master branch into the
31 current master branch, if any.
33 This default configuration is achieved by creating references to
34 the remote branch heads under `refs/remotes/origin` and
35 by initializing `remote.origin.url` and `remote.origin.fetch`
36 configuration variables.
43 When the repository to clone from is on a local machine,
44 this flag bypasses the normal "git aware" transport
45 mechanism and clones the repository by making a copy of
46 HEAD and everything under objects and refs directories.
47 The files under `.git/objects/` directory are hardlinked
48 to save space when possible. This is now the default when
49 the source repository is specified with `/path/to/repo`
50 syntax, so it essentially is a no-op option. To force
51 copying instead of hardlinking (which may be desirable
52 if you are trying to make a back-up of your repository),
53 but still avoid the usual "git aware" transport
54 mechanism, `--no-hardlinks` can be used.
57 Optimize the cloning process from a repository on a
58 local filesystem by copying files under `.git/objects`
63 When the repository to clone is on the local machine,
64 instead of using hard links, automatically setup
65 `.git/objects/info/alternates` to share the objects
66 with the source repository. The resulting repository
67 starts out without any object of its own.
69 *NOTE*: this is a possibly dangerous operation; do *not* use
70 it unless you understand what it does. If you clone your
71 repository using this option and then delete branches (or use any
72 other git command that makes any existing commit unreferenced) in the
73 source repository, some objects may become unreferenced (or dangling).
74 These objects may be removed by normal git operations (such as `git commit`)
75 which automatically call `git gc --auto`. (See linkgit:git-gc[1].)
76 If these objects are removed and were referenced by the cloned repository,
77 then the cloned repository will become corrupt.
79 Note that running `git repack` without the `-l` option in a repository
80 cloned with `-s` will copy objects from the source repository into a pack
81 in the cloned repository, removing the disk space savings of `clone -s`.
82 It is safe, however, to run `git gc`, which uses the `-l` option by
85 If you want to break the dependency of a repository cloned with `-s` on
86 its source repository, you can simply run `git repack -a` to copy all
87 objects from the source repository into a pack in the cloned repository.
89 --reference <repository>::
90 If the reference repository is on the local machine,
91 automatically setup `.git/objects/info/alternates` to
92 obtain objects from the reference repository. Using
93 an already existing repository as an alternate will
94 require fewer objects to be copied from the repository
95 being cloned, reducing network and local storage costs.
97 *NOTE*: see the NOTE for the `--shared` option.
101 Operate quietly. Progress is not reported to the standard
102 error stream. This flag is also passed to the `rsync'
107 Run verbosely. Does not affect the reporting of progress status
108 to the standard error stream.
111 Progress status is reported on the standard error stream
112 by default when it is attached to a terminal, unless -q
113 is specified. This flag forces progress status even if the
114 standard error stream is not directed to a terminal.
118 No checkout of HEAD is performed after the clone is complete.
121 Make a 'bare' GIT repository. That is, instead of
122 creating `<directory>` and placing the administrative
123 files in `<directory>/.git`, make the `<directory>`
124 itself the `$GIT_DIR`. This obviously implies the `-n`
125 because there is nowhere to check out the working tree.
126 Also the branch heads at the remote are copied directly
127 to corresponding local branch heads, without mapping
128 them to `refs/remotes/origin/`. When this option is
129 used, neither remote-tracking branches nor the related
130 configuration variables are created.
133 Set up a mirror of the source repository. This implies `--bare`.
134 Compared to `--bare`, `--mirror` not only maps local branches of the
135 source to local branches of the target, it maps all refs (including
136 remote-tracking branches, notes etc.) and sets up a refspec configuration such
137 that all these refs are overwritten by a `git remote update` in the
142 Instead of using the remote name `origin` to keep track
143 of the upstream repository, use `<name>`.
147 Instead of pointing the newly created HEAD to the branch pointed
148 to by the cloned repository's HEAD, point to `<name>` branch
149 instead. In a non-bare repository, this is the branch that will
152 --upload-pack <upload-pack>::
154 When given, and the repository to clone from is accessed
155 via ssh, this specifies a non-default path for the command
156 run on the other end.
158 --template=<template_directory>::
159 Specify the directory from which templates will be used;
160 (See the "TEMPLATE DIRECTORY" section of linkgit:git-init[1].)
163 Create a 'shallow' clone with a history truncated to the
164 specified number of revisions. A shallow repository has a
165 number of limitations (you cannot clone or fetch from
166 it, nor push from nor into it), but is adequate if you
167 are only interested in the recent history of a large project
168 with a long history, and would want to send in fixes
172 --recurse-submodules::
173 After the clone is created, initialize all submodules within,
174 using their default settings. This is equivalent to running
175 `git submodule update --init --recursive` immediately after
176 the clone is finished. This option is ignored if the cloned
177 repository does not have a worktree/checkout (i.e. if any of
178 `--no-checkout`/`-n`, `--bare`, or `--mirror` is given)
181 --separate-git-dir=<git dir>::
182 Instead of placing the cloned repository where it is supposed
183 to be, place the cloned repository at the specified directory,
184 then make a filesytem-agnostic git symbolic link to there.
185 The result is git repository can be separated from working
190 The (possibly remote) repository to clone from. See the
191 <<URLS,URLS>> section below for more information on specifying
195 The name of a new directory to clone into. The "humanish"
196 part of the source repository is used if no directory is
197 explicitly given (`repo` for `/path/to/repo.git` and `foo`
198 for `host.xz:foo/.git`). Cloning into an existing directory
199 is only allowed if the directory is empty.
207 * Clone from upstream:
210 $ git clone git://git.kernel.org/pub/scm/.../linux-2.6 my2.6
216 * Make a local clone that borrows from the current directory, without checking things out:
219 $ git clone -l -s -n . ../copy
225 * Clone from upstream while borrowing from an existing local directory:
228 $ git clone --reference my2.6 \
229 git://git.kernel.org/pub/scm/.../linux-2.7 \
235 * Create a bare repository to publish your changes to the public:
238 $ git clone --bare -l /home/proj/.git /pub/scm/proj.git
242 * Create a repository on the kernel.org machine that borrows from Linus:
245 $ git clone --bare -l -s /pub/scm/.../torvalds/linux-2.6.git \
246 /pub/scm/.../me/subsys-2.6.git
251 Part of the linkgit:git[1] suite