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 [--dissociate] [--separate-git-dir <git dir>]
16 [--depth <depth>] [--[no-]single-branch]
17 [--recursive | --recurse-submodules] [--] <repository>
23 Clones a repository into a newly created directory, creates
24 remote-tracking branches for each branch in the cloned repository
25 (visible using `git branch -r`), and creates and checks out an
26 initial branch that is forked from the cloned repository's
27 currently active branch.
29 After the clone, a plain `git fetch` without arguments will update
30 all the remote-tracking branches, and a `git pull` without
31 arguments will in addition merge the remote master branch into the
32 current master branch, if any (this is untrue when "--single-branch"
35 This default configuration is achieved by creating references to
36 the remote branch heads under `refs/remotes/origin` and
37 by initializing `remote.origin.url` and `remote.origin.fetch`
38 configuration variables.
45 When the repository to clone from is on a local machine,
46 this flag bypasses the normal "Git aware" transport
47 mechanism and clones the repository by making a copy of
48 HEAD and everything under objects and refs directories.
49 The files under `.git/objects/` directory are hardlinked
50 to save space when possible.
52 If the repository is specified as a local path (e.g., `/path/to/repo`),
53 this is the default, and --local is essentially a no-op. If the
54 repository is specified as a URL, then this flag is ignored (and we
55 never use the local optimizations). Specifying `--no-local` will
56 override the default when `/path/to/repo` is given, using the regular
57 Git transport instead.
60 Force the cloning process from a repository on a local
61 filesystem to copy the files under the `.git/objects`
62 directory instead of using hardlinks. This may be desirable
63 if you are trying to make a back-up of your repository.
67 When the repository to clone is on the local machine,
68 instead of using hard links, automatically setup
69 `.git/objects/info/alternates` to share the objects
70 with the source repository. The resulting repository
71 starts out without any object of its own.
73 *NOTE*: this is a possibly dangerous operation; do *not* use
74 it unless you understand what it does. If you clone your
75 repository using this option and then delete branches (or use any
76 other Git command that makes any existing commit unreferenced) in the
77 source repository, some objects may become unreferenced (or dangling).
78 These objects may be removed by normal Git operations (such as `git commit`)
79 which automatically call `git gc --auto`. (See linkgit:git-gc[1].)
80 If these objects are removed and were referenced by the cloned repository,
81 then the cloned repository will become corrupt.
83 Note that running `git repack` without the `-l` option in a repository
84 cloned with `-s` will copy objects from the source repository into a pack
85 in the cloned repository, removing the disk space savings of `clone -s`.
86 It is safe, however, to run `git gc`, which uses the `-l` option by
89 If you want to break the dependency of a repository cloned with `-s` on
90 its source repository, you can simply run `git repack -a` to copy all
91 objects from the source repository into a pack in the cloned repository.
93 --reference <repository>::
94 If the reference repository is on the local machine,
95 automatically setup `.git/objects/info/alternates` to
96 obtain objects from the reference repository. Using
97 an already existing repository as an alternate will
98 require fewer objects to be copied from the repository
99 being cloned, reducing network and local storage costs.
101 *NOTE*: see the NOTE for the `--shared` option, and also the
102 `--dissociate` option.
105 Borrow the objects from reference repositories specified
106 with the `--reference` options only to reduce network
107 transfer, and stop borrowing from them after a clone is made
108 by making necessary local copies of borrowed objects. This
109 option can also be used when cloning locally from a
110 repository that already borrows objects from another
111 repository--the new repository will borrow objects from the
112 same repository, and this option can be used to stop the
117 Operate quietly. Progress is not reported to the standard
118 error stream. This flag is also passed to the `rsync'
123 Run verbosely. Does not affect the reporting of progress status
124 to the standard error stream.
127 Progress status is reported on the standard error stream
128 by default when it is attached to a terminal, unless -q
129 is specified. This flag forces progress status even if the
130 standard error stream is not directed to a terminal.
134 No checkout of HEAD is performed after the clone is complete.
137 Make a 'bare' Git repository. That is, instead of
138 creating `<directory>` and placing the administrative
139 files in `<directory>/.git`, make the `<directory>`
140 itself the `$GIT_DIR`. This obviously implies the `-n`
141 because there is nowhere to check out the working tree.
142 Also the branch heads at the remote are copied directly
143 to corresponding local branch heads, without mapping
144 them to `refs/remotes/origin/`. When this option is
145 used, neither remote-tracking branches nor the related
146 configuration variables are created.
149 Set up a mirror of the source repository. This implies `--bare`.
150 Compared to `--bare`, `--mirror` not only maps local branches of the
151 source to local branches of the target, it maps all refs (including
152 remote-tracking branches, notes etc.) and sets up a refspec configuration such
153 that all these refs are overwritten by a `git remote update` in the
158 Instead of using the remote name `origin` to keep track
159 of the upstream repository, use `<name>`.
163 Instead of pointing the newly created HEAD to the branch pointed
164 to by the cloned repository's HEAD, point to `<name>` branch
165 instead. In a non-bare repository, this is the branch that will
167 `--branch` can also take tags and detaches the HEAD at that commit
168 in the resulting repository.
170 --upload-pack <upload-pack>::
172 When given, and the repository to clone from is accessed
173 via ssh, this specifies a non-default path for the command
174 run on the other end.
176 --template=<template_directory>::
177 Specify the directory from which templates will be used;
178 (See the "TEMPLATE DIRECTORY" section of linkgit:git-init[1].)
180 --config <key>=<value>::
182 Set a configuration variable in the newly-created repository;
183 this takes effect immediately after the repository is
184 initialized, but before the remote history is fetched or any
185 files checked out. The key is in the same format as expected by
186 linkgit:git-config[1] (e.g., `core.eol=true`). If multiple
187 values are given for the same key, each value will be written to
188 the config file. This makes it safe, for example, to add
189 additional fetch refspecs to the origin remote.
192 Create a 'shallow' clone with a history truncated to the
193 specified number of revisions.
195 --[no-]single-branch::
196 Clone only the history leading to the tip of a single branch,
197 either specified by the `--branch` option or the primary
198 branch remote's `HEAD` points at. When creating a shallow
199 clone with the `--depth` option, this is the default, unless
200 `--no-single-branch` is given to fetch the histories near the
201 tips of all branches.
202 Further fetches into the resulting repository will only update the
203 remote-tracking branch for the branch this option was used for the
204 initial cloning. If the HEAD at the remote did not point at any
205 branch when `--single-branch` clone was made, no remote-tracking
209 --recurse-submodules::
210 After the clone is created, initialize all submodules within,
211 using their default settings. This is equivalent to running
212 `git submodule update --init --recursive` immediately after
213 the clone is finished. This option is ignored if the cloned
214 repository does not have a worktree/checkout (i.e. if any of
215 `--no-checkout`/`-n`, `--bare`, or `--mirror` is given)
217 --separate-git-dir=<git dir>::
218 Instead of placing the cloned repository where it is supposed
219 to be, place the cloned repository at the specified directory,
220 then make a filesystem-agnostic Git symbolic link to there.
221 The result is Git repository can be separated from working
226 The (possibly remote) repository to clone from. See the
227 <<URLS,URLS>> section below for more information on specifying
231 The name of a new directory to clone into. The "humanish"
232 part of the source repository is used if no directory is
233 explicitly given (`repo` for `/path/to/repo.git` and `foo`
234 for `host.xz:foo/.git`). Cloning into an existing directory
235 is only allowed if the directory is empty.
243 * Clone from upstream:
246 $ git clone git://git.kernel.org/pub/scm/.../linux.git my-linux
252 * Make a local clone that borrows from the current directory, without checking things out:
255 $ git clone -l -s -n . ../copy
261 * Clone from upstream while borrowing from an existing local directory:
264 $ git clone --reference /git/linux.git \
265 git://git.kernel.org/pub/scm/.../linux.git \
271 * Create a bare repository to publish your changes to the public:
274 $ git clone --bare -l /home/proj/.git /pub/scm/proj.git
280 Part of the linkgit:git[1] suite