git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Duy Nguyen <pclouds@gmail.com>
To: Christian Spanier <cspanier@boxie.eu>
Cc: Git Mailing List <git@vger.kernel.org>,
	Phillip Wood <phillip.wood@dunelm.org.uk>
Subject: Re: bug: git pull may delete untracked files in submodule without notice
Date: Fri, 3 May 2019 17:28:48 +0700	[thread overview]
Message-ID: <CACsJy8ArJ9Lc1pA7SbJ1OE6hK+rprXpZViLDPUKPMcWG7Bre3A@mail.gmail.com> (raw)
In-Reply-To: <2e321dbe-42a3-7516-52aa-4bc50a3c403e@boxie.eu>

On Fri, May 3, 2019 at 5:25 PM Christian Spanier <cspanier@boxie.eu> wrote:
>
> Hi,
>
> I found a bug where Git may delete untracked files without notice in
> certain situations. This bug effects Git 2.21.0 both on Linux and Windows.
> In summary this happens when git pull merges a commit that replaces a
> submodule folder with a symlink. Any files within the folder are deleted
> without notice.

Maybe it's not the same, maybe it is. But Phillip recently did some
work protecting untracked files and I think he touched a test case
about submodule. Adding him so he can check, if he has time.

> Check out the script below for details.
>
> This happend on some developer's machine and deleted a repository
> containing about 200GiB of files and tons of uncommited local scripts,
> log files and whatever, just because some other dev accidentally
> commited a temporary change.
>
> Greetings,
> Christian Spanier
>
> ##### PREPARATION #####
>
> # New empty repository #1
> mkdir rep1
> cd rep1
> git init --bare .
> cd ..
>
> # New empty repository #2
> mkdir rep2
> cd rep2
> git init --bare .
> cd ..
>
> # Clone repository #1 and create initial commit
> git clone rep1 clone_rep1_user1
> cd clone_rep1_user1
> touch README
> git add README
> git commit -m "initial commit"
> git push
> cd ..
>
> # Clone repository #2 and create initial commit
> git clone rep2 clone_rep2
> cd clone_rep2
> touch README
> git add README
> git commit -m "initial commit"
> git push
> cd ..
>
> # Add repository #2 as a submodule to repository #1
> cd clone_rep1_user1
> git submodule add ../rep2
> git commit -m "add submodule"
> git push
> cd ..
>
> # User 2 also clones repository #1 and #2 recursively
> git clone --recursive rep1 clone_rep1_user2
>
> # User 2 starts working in his folder and adds an important local file
> which is
> # not yet committed inside the submodule folder.
> cd clone_rep1_user2/rep2
> echo "important work" > uncommitted_file
> cd ../../
>
> # Meanwhile, user 1 temporarily switch out folder /clone_rep1_user1/rep2
> with a
> # symbolic link to a different folder (for whatever reason, maybe a copy
> of an
> # older version or anything).
> mkdir rep2_alternative
> cd clone_rep1_user1
> mv rep2 ../rep2_backup
> ln -s ../rep2_alternative rep2
> # On Windows this can be done with 'mklink /D rep2 ../rep2_alternative',
> # which requires admin privileges. The bug is not reproducible when
> # using a directory junction with 'mklink /D /J ...'.
>
> # He does some work on rep1 but then accidently adds the symbolic link
> to his
> # next commit and pushes the changes. Notice the typechange of rep2.
> echo "some" > work
> git status
> # On branch master
> # Your branch is up to date with 'origin/master'.
> #
> # Changes not staged for commit:
> #   (use "git add <file>..." to update what will be committed)
> #   (use "git checkout -- <file>..." to discard changes in working
> directory)
> #
> #         typechange: rep2
> #
> # Untracked files:
> #   (use "git add <file>..." to include in what will be committed)
> #
> #         work
> #
> # no changes added to commit (use "git add" and/or "git commit -a")
> git add .
> git commit -m "do some work"
> git push
> cd ..
>
> # NOW THE BUG:
>
> # User 2 pulls the changes and loses his important work in
> # rep2/uncommitted_file because Git replaces the folder with a symlink
> # without checking for modified or uncommited files!
> # He should get an error in this case!
> cd clone_rep1_user2
> git pull
> cat rep2/uncommitted_file
> # cat: rep2/uncommitted_file: Not a directory
> # "important work" in rep2/uncommitted_file is gone :(



-- 
Duy

  reply	other threads:[~2019-05-03 10:29 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-03  8:02 bug: git pull may delete untracked files in submodule without notice Christian Spanier
2019-05-03 10:28 ` Duy Nguyen [this message]
2019-05-03 15:10   ` Phillip Wood

Reply instructions:

You may reply publicly 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=CACsJy8ArJ9Lc1pA7SbJ1OE6hK+rprXpZViLDPUKPMcWG7Bre3A@mail.gmail.com \
    --to=pclouds@gmail.com \
    --cc=cspanier@boxie.eu \
    --cc=git@vger.kernel.org \
    --cc=phillip.wood@dunelm.org.uk \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).