问题
Is it possible to check out subdirectories of a repository in Git?
Imagine I am setting up a new WordPress installation. I will create two new directories for my plugin and theme customization:
wordpress/wp-content/plugins/myplugins/wordpress/wp-content/themes/mytheme/
I want to maintain these directories via Git. In Subversion, I would accomplish this by having trunk/myplugins/ and trunk/mytheme/ directories and checking out subdirectories. Does Git have a way to accomplish the same task using a single repository?
I could just be missing the boat on some Git paradigm, as a long time SVN user with little exposure to Git.
Edit: Multiple branches storing different content is an interesting way to handle this.
回答1:
Sparse checkouts are now in Git 1.7.
Also see the question “Is it possible to do a sparse checkout without checking out the whole repository first?”.
Note that sparse checkouts still require you to download the whole repository, even though some of the files Git downloads won't end up in your working tree.
回答2:
There is no real way to do that in git. And if you won’t be making changes that affect both trees at once as a single work unit, there is no good reason to use a single repository for both. I thought I would miss this Subversion feature, but I found that creating repositories has so little administrative mental overhead (simply due to the fact that repositories are stored right next to their working copy, rather than requiring me to explicitly pick some place outside of the working copy) that I got used to just making lots of small single-purpose repositories.
If you insist (or really need it), though, you could make a git repository with just mytheme and myplugins directories and symlink those from within the WordPress install.
MDCore wrote:
making a commit to, e.g., mytheme will increment the revision number for myplugin
Note that this is not a concern for git, if you do decide to put both directories in a single repository, because git does away entirely with the concept of monotonically increasing revision numbers of any form.
The sole criterion for what things to put together in a single repository in git is whether it constitutes a single unit, ie. in your case whether there are changes where it does not make sense to look at the edits in each directory in isolation. If you have changes where you need to edit files in both directories at once and the edits belong together, they should be one repository. If not, then don’t glom them together.
Git really really wants you to use separate repositories for separate entities.
submodules
Submodules do not address the desire to keep both directories in one repository, because they would actually enforce having a separate repository for each directory, which are then brought together in another repository using submodules. Worse, since the directories inside the WordPress install are not direct subdirectories of the same directory and are also part of a hierarchy with many other files, using the per-directory repositories as submodules in a unified repository would offer no benefit whatsoever, because the unified repository would not reflect any use case/need.
回答3:
One thing I don't like about sparse checkouts, is that if you want to checkout a subdirectory that is a few directories deep, your directory structure must contain all directories leading to it.
How I work around this is to clone the repo in a place that is not my workspace and then create a symbolic link in my workspace directory to the subdirectory in the repository. Git works like this quite nicely because things like git status will display the change files relative to your current working directory.
回答4:
Actually, "narrow" or "partial" or "sparse" checkouts are under current, heavy development for Git. Note, you'll still have the full repository under .git. So, the other two posts are current for the current state of Git but it looks like we will be able to do sparse checkouts eventually. Checkout the mailing lists if you're interested in more details -- they're changing rapidly.
回答5:
git clone --filter from Git 2.19
This option will actually skip fetching unneeded objects from the server:
git clone --depth 1 --no-checkout --filter=blob:none \
"file://$(pwd)/server_repo" local_repo
cd local_repo
git checkout master -- mdir/
The server should be configured with:
git config --local uploadpack.allowfilter 1
git config --local uploadpack.allowanysha1inwant 1
There is no server support as of v2.19.0, but it can already be locally tested.
file://$(path) is required to overcome git clone protocol shenanigans: How to shallow clone a local git repository with a relative path?
Remember that --depth 1 already implies --single-branch, see also: How do I clone a single branch in Git?
TODO: --filter=blob:none skips all blobs, but still fetches all tree objects. But on a normal repo, this should be tiny compared to the files themselves, so this is already good enough. Asked at: https://www.spinics.net/lists/git/msg342006.html Devs replied a --filter=tree:0 is in the works to do that.
The format of --filter is documented on man git-rev-list.
An extension was made to the Git remote protocol to support this feature.
Docs on Git tree:
- https://github.com/git/git/blob/v2.19.0/Documentation/technical/partial-clone.txt
- https://github.com/git/git/blob/v2.19.0/Documentation/rev-list-options.txt#L720
- https://github.com/git/git/blob/v2.19.0/t/t5616-partial-clone.sh
Test it out
#!/usr/bin/env bash
set -eu
list-objects() (
git rev-list --all --objects
echo "master commit SHA: $(git log -1 --format="%H")"
echo "mybranch commit SHA: $(git log -1 --format="%H")"
git ls-tree master
git ls-tree mybranch | grep mybranch
git ls-tree master~ | grep root
)
# Reproducibility.
export GIT_COMMITTER_NAME='a'
export GIT_COMMITTER_EMAIL='a'
export GIT_AUTHOR_NAME='a'
export GIT_AUTHOR_EMAIL='a'
export GIT_COMMITTER_DATE='2000-01-01T00:00:00+0000'
export GIT_AUTHOR_DATE='2000-01-01T00:00:00+0000'
rm -rf server_repo local_repo
mkdir server_repo
cd server_repo
# Create repo.
git init --quiet
git config --local uploadpack.allowfilter 1
git config --local uploadpack.allowanysha1inwant 1
# First commit.
# Directories present in all branches.
mkdir d1 d2
printf 'd1/a' > ./d1/a
printf 'd1/b' > ./d1/b
printf 'd2/a' > ./d2/a
printf 'd2/b' > ./d2/b
# Present only in root.
mkdir 'root'
printf 'root' > ./root/root
git add .
git commit -m 'root' --quiet
# Second commit only on master.
git rm --quiet -r ./root
mkdir 'master'
printf 'master' > ./master/master
git add .
git commit -m 'master commit' --quiet
# Second commit only on mybranch.
git checkout -b mybranch --quiet master~
git rm --quiet -r ./root
mkdir 'mybranch'
printf 'mybranch' > ./mybranch/mybranch
git add .
git commit -m 'mybranch commit' --quiet
echo "# List and identify all objects"
list-objects
echo
# Restore master.
git checkout --quiet master
cd ..
# Clone. Don't checkout for now, only .git/ dir.
git clone --depth 1 --quiet --no-checkout --filter=blob:none "file://$(pwd)/server_repo" local_repo
cd local_repo
# List missing objects from master.
echo "# Missing objects after --no-checkout"
git rev-list --all --quiet --objects --missing=print
echo
echo "# Git checkout fails without internet"
mv ../server_repo ../server_repo.off
! git checkout master
echo
echo "# Git checkout fetches the missing directory from internet"
mv ../server_repo.off ../server_repo
git checkout master -- d1/
echo
echo "# Missing objects after checking out d1"
git rev-list --all --quiet --objects --missing=print
GitHub upstream.
Output in Git v2.19:
# List and identify all objects
c6fcdfaf2b1462f809aecdad83a186eeec00f9c1
fc5e97944480982cfc180a6d6634699921ee63ec
7251a83be9a03161acde7b71a8fda9be19f47128
62d67bce3c672fe2b9065f372726a11e57bade7e
b64bf435a3e54c5208a1b70b7bcb0fc627463a75 d1
308150e8fddde043f3dbbb8573abb6af1df96e63 d1/a
f70a17f51b7b30fec48a32e4f19ac15e261fd1a4 d1/b
84de03c312dc741d0f2a66df7b2f168d823e122a d2
0975df9b39e23c15f63db194df7f45c76528bccb d2/a
41484c13520fcbb6e7243a26fdb1fc9405c08520 d2/b
7d5230379e4652f1b1da7ed1e78e0b8253e03ba3 master
8b25206ff90e9432f6f1a8600f87a7bd695a24af master/master
ef29f15c9a7c5417944cc09711b6a9ee51b01d89
19f7a4ca4a038aff89d803f017f76d2b66063043 mybranch
1b671b190e293aa091239b8b5e8c149411d00523 mybranch/mybranch
c3760bb1a0ece87cdbaf9a563c77a45e30a4e30e
a0234da53ec608b54813b4271fbf00ba5318b99f root
93ca1422a8da0a9effc465eccbcb17e23015542d root/root
master commit SHA: fc5e97944480982cfc180a6d6634699921ee63ec
mybranch commit SHA: fc5e97944480982cfc180a6d6634699921ee63ec
040000 tree b64bf435a3e54c5208a1b70b7bcb0fc627463a75 d1
040000 tree 84de03c312dc741d0f2a66df7b2f168d823e122a d2
040000 tree 7d5230379e4652f1b1da7ed1e78e0b8253e03ba3 master
040000 tree 19f7a4ca4a038aff89d803f017f76d2b66063043 mybranch
040000 tree a0234da53ec608b54813b4271fbf00ba5318b99f root
# Missing objects after --no-checkout
?f70a17f51b7b30fec48a32e4f19ac15e261fd1a4
?8b25206ff90e9432f6f1a8600f87a7bd695a24af
?41484c13520fcbb6e7243a26fdb1fc9405c08520
?0975df9b39e23c15f63db194df7f45c76528bccb
?308150e8fddde043f3dbbb8573abb6af1df96e63
# Git checkout fails without internet
fatal: '/home/ciro/bak/git/test-git-web-interface/other-test-repos/partial-clone.tmp/server_repo' does not appear to be a git repository
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.
# Git checkout fetches the missing directory from internet
remote: Enumerating objects: 1, done.
remote: Counting objects: 100% (1/1), done.
remote: Total 1 (delta 0), reused 0 (delta 0)
Receiving objects: 100% (1/1), 45 bytes | 45.00 KiB/s, done.
remote: Enumerating objects: 1, done.
remote: Counting objects: 100% (1/1), done.
remote: Total 1 (delta 0), reused 0 (delta 0)
Receiving objects: 100% (1/1), 45 bytes | 45.00 KiB/s, done.
# Missing objects after checking out d1
?8b25206ff90e9432f6f1a8600f87a7bd695a24af
?41484c13520fcbb6e7243a26fdb1fc9405c08520
?0975df9b39e23c15f63db194df7f45c76528bccb
Conclusions: all blobs from outside of d1/ are missing.
Note that root/root and mybranch/mybranch are also missing, but --depth 1 hides that from the list of missing files. If you remove --depth 1, then they show on the list of missing files.
回答6:
As your edit points out, you can use two separate branches to store the two separate directories. This does keep them both in the same repository, but you still can't have commits spanning both directory trees. If you have a change in one that requires a change in the other, you'll have to do those as two separate commits, and you open up the possibility that a pair of checkouts of the two directories can go out of sync.
If you want to treat the pair of directories as one unit, you can use 'wordpress/wp-content' as the root of your repo and use .gitignore file at the top level to ignore everything but the two subdirectories of interest. This is probably the most reasonable solution at this point.
Sparse checkouts have been allegedly coming for two years now, but there's still no sign of them in the git development repo, nor any indication that the necessary changes will ever arrive there. I wouldn't count on them.
回答7:
You can't checkout a single directory of a repository because the entire repository is handled by the single .git folder in the root of the project instead of subversion's myriad of .svn directories.
The problem with working on plugins in a single repository is that making a commit to, e.g., mytheme will increment the revision number for myplugin, so even in subversion it is better to use separate repositories.
The subversion paradigm for sub-projects is svn:externals which translates somewhat to submodules in git (but not exactly in case you've used svn:externals before.)
回答8:
There is an inspiration here. Just utilize shell regex or git regex.
git checkout commit_id */*.bat # *.bat in 1-depth subdir exclude current dir, shell regex
git checkout commit_id '*.bat' # *.bat in all subdir include current dir, git regex
Use quotation to escape shell regex interpretation and pass wildcards to git.
The first one is not recursive, only files in 1-depth subdir. But the second one is recursive.
As for your situation, the following may be enough.
git checkout master */*/wp-content/*/*
git checkout master '*/wp-content/*'
Just hack the lines as required.
来源：https://stackoverflow.com/questions/180052/checkout-subdirectories-in-git