git@vger.kernel.org mailing list mirror (one of many)
 help / Atom feed
From: Stefan Beller <sbeller@google.com>
To: "Billy O'Neal (VC LIBS)" <bion@microsoft.com>
Cc: "git@vger.kernel.org" <git@vger.kernel.org>, "Stephan T. Lavavej" <stl@exchange.microsoft.com>, Cody Miller <codym@microsoft.com>
Subject: Re: Bug report: git reset --hard does not fix submodule worktree
Date: Mon, 6 Nov 2017 12:50:42 -0800
Message-ID: <CAGZ79kZRa6=PNtRoVMzM89ek+9uv9_+C+3-hNAaJ_W=YoeWj-w@mail.gmail.com> (raw)
In-Reply-To: <BN6PR21MB0114B8ECD6E8A158FB4A6FBFCB520@BN6PR21MB0114.namprd21.prod.outlook.com>

On Fri, Nov 3, 2017 at 5:46 PM, Billy O'Neal (VC LIBS)
<bion@microsoft.com> wrote:
> Hello, Git folks. I managed to accidentally break the our test lab by attempting to git mv a directory with a submodule inside. It seems like if a reset does an undo on a mv, the workfold entry should be fixed to put the submodule in its old location. Consider the following sequence of commands:
>
> Setup a git repo with a submodule:
> mkdir metaproject
> mkdir upstream
> cd metaproject
> git init
> cd ..\upstream
> git init
> echo hello > test.txt
> git add -A
> git commit -m "an example commit"
> cd ..\metapoject
> mkdir start
> git submodule add ../upstream start/upstream
> git add -A
> git commit -m "Add submodule in start/upstream."
>
> Move the directory containing the submodule:
> git mv start target
> git add -A
> git commit -m "Moved submodule parent directory."
>
> Check that the worktree got correctly fixed by git mv; this output is as expected:
> type .git\modules\start\upstream\config
> [core]
>         repositoryformatversion = 0
>         filemode = false
>         bare = false
>         logallrefupdates = true
>         symlinks = false
>         ignorecase = true
>         worktree = ../../../../target/upstream
> [remote "origin"]
>         url = C:/Users/bion/Desktop/upstream
>         fetch = +refs/heads/*:refs/remotes/origin/*
> [branch "master"]
>         remote = origin
>         merge = refs/heads/master
>
> Now try to go back to the previous commit using git reset --hard:
> git log --oneline
>  1f560be (HEAD -> master) Moved submodule parent directory.
>  a5977ce Add submodule in start/upstream.
> git reset --hard a5977ce
>  warning: unable to rmdir target/upstream: Directory not empty
>  HEAD is now at a5977ce Add submodule in start/upstream.
>
> Check that the worktree got fixed correctly; it did not:
> type .git\modules\start\upstream\config
> [core]
>         repositoryformatversion = 0
>         filemode = false
>         bare = false
>         logallrefupdates = true
>         symlinks = false
>         ignorecase = true
>         worktree = ../../../../target/upstream
> [remote "origin"]
>         url = C:/Users/bion/Desktop/upstream
>         fetch = +refs/heads/*:refs/remotes/origin/*
> [branch "master"]
>         remote = origin
>         merge = refs/heads/master
>
> Is git reset intended to put the submodule in the right place?
>
> Thanks folks!

git-reset sounds like it ought to put submodules back into another directory.
Historically git-reset did not touch submodules at all; a recent
release introduced
the --recurse-submodules flag for git-reset, which would cover this situation.
However that particular situation (with moving the submodules) seems to be
not covered yet,

./t7112-reset-submodule.sh
...
not ok 69 - git reset --hard: replace submodule with a directory must
fail # TODO known breakage



> If not, is there a command we can run before/after reset to restore consistency?

The submodule "fix-it-all" command would be

    git submodule update --init --recursive

IMHO.

      reply index

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-04  0:46 Billy O'Neal (VC LIBS)
2017-11-06 20:50 ` Stefan Beller [this message]

Reply instructions:

You may reply publically to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: http://vger.kernel.org/majordomo-info.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAGZ79kZRa6=PNtRoVMzM89ek+9uv9_+C+3-hNAaJ_W=YoeWj-w@mail.gmail.com' \
    --to=sbeller@google.com \
    --cc=bion@microsoft.com \
    --cc=codym@microsoft.com \
    --cc=git@vger.kernel.org \
    --cc=stl@exchange.microsoft.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

git@vger.kernel.org mailing list mirror (one of many)

Archives are clonable:
	git clone --mirror https://public-inbox.org/git
	git clone --mirror http://ou63pmih66umazou.onion/git
	git clone --mirror http://czquwvybam4bgbro.onion/git
	git clone --mirror http://hjrcffqmbrq6wope.onion/git

Newsgroups are available over NNTP:
	nntp://news.public-inbox.org/inbox.comp.version-control.git
	nntp://ou63pmih66umazou.onion/inbox.comp.version-control.git
	nntp://czquwvybam4bgbro.onion/inbox.comp.version-control.git
	nntp://hjrcffqmbrq6wope.onion/inbox.comp.version-control.git
	nntp://news.gmane.org/gmane.comp.version-control.git

 note: .onion URLs require Tor: https://www.torproject.org/
       or Tor2web: https://www.tor2web.org/

AGPL code for this site: git clone https://public-inbox.org/ public-inbox