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 [--depth <depth>] [--recursive] [--] <repository> [<directory>]
20 Clones a repository into a newly created directory, creates
21 remote-tracking branches for each branch in the cloned repository
22 (visible using `git branch -r`), and creates and checks out an
23 initial branch that is forked from the cloned repository's
24 currently active branch.
26 After the clone, a plain `git fetch` without arguments will update
27 all the remote-tracking branches, and a `git pull` without
28 arguments will in addition merge the remote master branch into the
29 current master branch, if any.
31 This default configuration is achieved by creating references to
32 the remote branch heads under `refs/remotes/origin` and
33 by initializing `remote.origin.url` and `remote.origin.fetch`
34 configuration variables.
41 When the repository to clone from is on a local machine,
42 this flag bypasses the normal "git aware" transport
43 mechanism and clones the repository by making a copy of
44 HEAD and everything under objects and refs directories.
45 The files under `.git/objects/` directory are hardlinked
46 to save space when possible. This is now the default when
47 the source repository is specified with `/path/to/repo`
48 syntax, so it essentially is a no-op option. To force
49 copying instead of hardlinking (which may be desirable
50 if you are trying to make a back-up of your repository),
51 but still avoid the usual "git aware" transport
52 mechanism, `--no-hardlinks` can be used.
55 Optimize the cloning process from a repository on a
56 local filesystem by copying files under `.git/objects`
61 When the repository to clone is on the local machine,
62 instead of using hard links, automatically setup
63 `.git/objects/info/alternates` to share the objects
64 with the source repository. The resulting repository
65 starts out without any object of its own.
67 *NOTE*: this is a possibly dangerous operation; do *not* use
68 it unless you understand what it does. If you clone your
69 repository using this option and then delete branches (or use any
70 other git command that makes any existing commit unreferenced) in the
71 source repository, some objects may become unreferenced (or dangling).
72 These objects may be removed by normal git operations (such as `git commit`)
73 which automatically call `git gc --auto`. (See linkgit:git-gc[1].)
74 If these objects are removed and were referenced by the cloned repository,
75 then the cloned repository will become corrupt.
77 Note that running `git repack` without the `-l` option in a repository
78 cloned with `-s` will copy objects from the source repository into a pack
79 in the cloned repository, removing the disk space savings of `clone -s`.
80 It is safe, however, to run `git gc`, which uses the `-l` option by
83 If you want to break the dependency of a repository cloned with `-s` on
84 its source repository, you can simply run `git repack -a` to copy all
85 objects from the source repository into a pack in the cloned repository.
87 --reference <repository>::
88 If the reference repository is on the local machine,
89 automatically setup `.git/objects/info/alternates` to
90 obtain objects from the reference repository. Using
91 an already existing repository as an alternate will
92 require fewer objects to be copied from the repository
93 being cloned, reducing network and local storage costs.
95 *NOTE*: see the NOTE for the `--shared` option.
99 Operate quietly. Progress is not reported to the standard
100 error stream. This flag is also passed to the `rsync'
105 Run verbosely. Does not affect the reporting of progress status
106 to the standard error stream.
109 Progress status is reported on the standard error stream
110 by default when it is attached to a terminal, unless -q
111 is specified. This flag forces progress status even if the
112 standard error stream is not directed to a terminal.
116 No checkout of HEAD is performed after the clone is complete.
119 Make a 'bare' GIT repository. That is, instead of
120 creating `<directory>` and placing the administrative
121 files in `<directory>/.git`, make the `<directory>`
122 itself the `$GIT_DIR`. This obviously implies the `-n`
123 because there is nowhere to check out the working tree.
124 Also the branch heads at the remote are copied directly
125 to corresponding local branch heads, without mapping
126 them to `refs/remotes/origin/`. When this option is
127 used, neither remote-tracking branches nor the related
128 configuration variables are created.
131 Set up a mirror of the source repository. This implies `--bare`.
132 Compared to `--bare`, `--mirror` not only maps local branches of the
133 source to local branches of the target, it maps all refs (including
134 remote-tracking branches, notes etc.) and sets up a refspec configuration such
135 that all these refs are overwritten by a `git remote update` in the
140 Instead of using the remote name `origin` to keep track
141 of the upstream repository, use `<name>`.
145 Instead of pointing the newly created HEAD to the branch pointed
146 to by the cloned repository's HEAD, point to `<name>` branch
147 instead. In a non-bare repository, this is the branch that will
150 --upload-pack <upload-pack>::
152 When given, and the repository to clone from is accessed
153 via ssh, this specifies a non-default path for the command
154 run on the other end.
156 --template=<template_directory>::
157 Specify the directory from which templates will be used;
158 (See the "TEMPLATE DIRECTORY" section of linkgit:git-init[1].)
161 Create a 'shallow' clone with a history truncated to the
162 specified number of revisions. A shallow repository has a
163 number of limitations (you cannot clone or fetch from
164 it, nor push from nor into it), but is adequate if you
165 are only interested in the recent history of a large project
166 with a long history, and would want to send in fixes
170 After the clone is created, initialize all submodules within,
171 using their default settings. This is equivalent to running
172 `git submodule update --init --recursive` immediately after
173 the clone is finished. This option is ignored if the cloned
174 repository does not have a worktree/checkout (i.e. if any of
175 `--no-checkout`/`-n`, `--bare`, or `--mirror` is given)
178 The (possibly remote) repository to clone from. See the
179 <<URLS,URLS>> section below for more information on specifying
183 The name of a new directory to clone into. The "humanish"
184 part of the source repository is used if no directory is
185 explicitly given (`repo` for `/path/to/repo.git` and `foo`
186 for `host.xz:foo/.git`). Cloning into an existing directory
187 is only allowed if the directory is empty.
195 * Clone from upstream:
198 $ git clone git://git.kernel.org/pub/scm/.../linux-2.6 my2.6
204 * Make a local clone that borrows from the current directory, without checking things out:
207 $ git clone -l -s -n . ../copy
213 * Clone from upstream while borrowing from an existing local directory:
216 $ git clone --reference my2.6 \
217 git://git.kernel.org/pub/scm/.../linux-2.7 \
223 * Create a bare repository to publish your changes to the public:
226 $ git clone --bare -l /home/proj/.git /pub/scm/proj.git
230 * Create a repository on the kernel.org machine that borrows from Linus:
233 $ git clone --bare -l -s /pub/scm/.../torvalds/linux-2.6.git \
234 /pub/scm/.../me/subsys-2.6.git
240 Written by Linus Torvalds <torvalds@osdl.org>
245 Documentation by Junio C Hamano and the git-list <git@vger.kernel.org>.
250 Part of the linkgit:git[1] suite