Re: Git mv -- submodule -- recursive

[Stefan Beller]

On Mon, Jun 27, 2016 at 1:07 PM, Bart Bogaerts  wrote:
> Hi,
>
> I tested this on git version 2.9 and it still fails (exactly the same
> behaviour as on the stackoverflow post; also the workarounded I posted
> there still works).

I see. the commit I referenced was solving a slightly different problem, than
what we do here. That commit only fixed the issue for submodules in a nested
directory.

The important detail what is different in this case is having another
nested submodule.

If you only had

- git-repo a
-- subdirectory 2015
--- git-submodule b
-- git-submodule c

(quoted from SO)

then `git mv c 2015/` is expected to work as well as
`mkdir d && git mv 2015 d/` with submodule b being nested in 2015.

but as you have a submodule inside of c, moving c is buggy.

Stefan
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Git mv -- submodule -- recursive

[Bart Bogaerts]

Hi,

I tested this on git version 2.9 and it still fails (exactly the same
behaviour as on the stackoverflow post; also the workarounded I posted
there still works).
Some output showing the bug follows below:


bartb@EB-Latitude-E5450 ~/Documents/papers $ git status
On branch master
Your branch is up-to-date with 'origin/master'.
nothing to commit, working directory clean
bartb@EB-Latitude-E5450 ~/Documents/papers $ git --version
git version 2.9.0
bartb@EB-Latitude-E5450 ~/Documents/papers $ git mv fo-c-revisited/ 2016
bartb@EB-Latitude-E5450 ~/Documents/papers $ git status
fatal: Not a git repository:
idp-latex/../../.git/modules/fo-c-revisited/modules/idp-latex
fatal: 'git status --porcelain' failed in submodule 2016/fo-c-revisited
bartb@EB-Latitude-E5450 ~/Documents/papers $ rm
2016/fo-c-revisited/idp-latex/.git
bartb@EB-Latitude-E5450 ~/Documents/papers $ git submodule update
bartb@EB-Latitude-E5450 ~/Documents/papers $ git status
On branch master
Your branch is up-to-date with 'origin/master'.
Changes to be committed:
  (use "git reset HEAD ..." to unstage)

modified:   .gitmodules
renamed:fo-c-revisited -> 2016/fo-c-revisited

bartb@EB-Latitude-E5450 ~/Documents/papers $

2016-06-27 19:57 GMT+03:00 Stefan Beller :
> On Sun, Jun 26, 2016 at 11:01 PM, Bart Bogaerts  
> wrote:
>> With a repo structured as follows
>>
>> main-files
>> |- submod
>> |- subsubmodule
>>
>> git mv submod newlocation
>>
>> does not do what it is supposed to do. It actually breaks the git repository.
>> It can be fixed easily.
>> A complete description of the bug, including a workaround, can be found on
>> http://stackoverflow.com/q/32782382/2274140
>
> Which version of Git are you using?
> I think this is fixed in a127331cd81233 (mv: allow moving nested
> submodules, 2016-04-19), which is first included in v2.8.3
> (or v2.9 and later).
>
> Thanks,
> Stefan
>
>>
>> --
>> Bart Bogaerts
>> --
>> To unsubscribe from this list: send the line "unsubscribe git" in
>> the body of a message to majord...@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html



-- 
Bart Bogaerts
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Git mv -- submodule -- recursive

[Stefan Beller]

On Sun, Jun 26, 2016 at 11:01 PM, Bart Bogaerts  wrote:
> With a repo structured as follows
>
> main-files
> |- submod
> |- subsubmodule
>
> git mv submod newlocation
>
> does not do what it is supposed to do. It actually breaks the git repository.
> It can be fixed easily.
> A complete description of the bug, including a workaround, can be found on
> http://stackoverflow.com/q/32782382/2274140

Which version of Git are you using?
I think this is fixed in a127331cd81233 (mv: allow moving nested
submodules, 2016-04-19), which is first included in v2.8.3
(or v2.9 and later).

Thanks,
Stefan

>
> --
> Bart Bogaerts
> --
> To unsubscribe from this list: send the line "unsubscribe git" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Git mv -- submodule -- recursive

[Bart Bogaerts]

With a repo structured as follows

main-files
|- submod
|- subsubmodule

git mv submod newlocation

does not do what it is supposed to do. It actually breaks the git repository.
It can be fixed easily.
A complete description of the bug, including a workaround, can be found on
http://stackoverflow.com/q/32782382/2274140

-- 
Bart Bogaerts
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: git-mv-submodule

[Jens Lehmann]

Am 21.12.2013 10:48, schrieb fREW Schmidt:
 Hello all,
 
 I was on a plane, moving around some of the many (30ish) submodules in
 my dotfiles and got really annoyed at how much work it is (move the
 dir, remove old from git, add new to git, fix .gitmodules, fix
 .git/config, fix all the parts of the submodule config) so I wrote a
 perl script to work for the most common case.
 
 As far as I know it should work for anyone not doing Something Weird,
 ie manually fiddling with their submodules.  The main case it does not
 support that I'd like to in the future is submodules containing
 submodules, and also at some point I'd like to wrap git mv to invoke
 this script on demand automatically.

Thanks for sharing! Form a cursory look over your perl script it
looks like it does what stock git mv will do since 1.8.5 (except
for changing the name of the submodule, which I would not advise
to do when only moving the submodule location in the work tree).
--
To unsubscribe from this list: send the line unsubscribe git in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: git-mv-submodule

[fREW Schmidt]

On Sat, Dec 21, 2013 at 05:08:59PM +0100, Jens Lehmann wrote:
 Am 21.12.2013 10:48, schrieb fREW Schmidt:
 Thanks for sharing! Form a cursory look over your perl script it
 looks like it does what stock git mv will do since 1.8.5 (except
 for changing the name of the submodule, which I would not advise
 to do when only moving the submodule location in the work tree).

See, I thought I read that in the changelog; unfortunately I don'g
thing it does the final set of book-keeping (changing the .git file if
you changed the depth of the submodule in the mv and changing the path
of the worktree in the actual git repo in .git/modules)

I'd love to be wrong on that as this script is clearly not perfect.  I
think my second little script in my previous email re git submodule
bugs shows the issue.  I'll include it here for simplicity though:

 mkdir -p test/a test/b
 cd test/a
 git init
 touch a.txt
 git add a.txt
 git ci -m 'initial commit'
 cd ../b
 git init
 mkdir c
 touch c/c.txt
 git submodule add ../a c/a
 git ci -m 'initial commit'
 git mv c d
 git status

-- 
fREW Schmidt
http://blog.afoolishmanifesto.com


pgpZne3BJMPSW.pgp
Description: PGP signature

git-mv-submodule

[fREW Schmidt]

Hello all,

I was on a plane, moving around some of the many (30ish) submodules in
my dotfiles and got really annoyed at how much work it is (move the
dir, remove old from git, add new to git, fix .gitmodules, fix
.git/config, fix all the parts of the submodule config) so I wrote a
perl script to work for the most common case.

As far as I know it should work for anyone not doing Something Weird,
ie manually fiddling with their submodules.  The main case it does not
support that I'd like to in the future is submodules containing
submodules, and also at some point I'd like to wrap git mv to invoke
this script on demand automatically.

Note that this script requires perl 5.10.1, released in 2009.  If you
are stuck with something inferior to that you can comment out the
version at the top and the autodie usage and it should work further
back, but won't be quite as robust.

Enjoy!
-- 
fREW Schmidt
http://blog.afoolishmanifesto.com
#!/usr/bin/env perl

use 5.10.1;
use strict;
use warnings;

use autodie;
use File::Basename 'dirname';
use File::Path 'make_path';

die you must pass both a from and a to unless @ARGV == 2;

my ($from, $to) = @ARGV;

die you have changes in your working copy!
   unless is_clean();

# move the real dir
make_path(dirname($to));
safe_system('mv', $from, $to);

# move the git dir (not really that important)
make_path(dirname(.git/modules/$to));
safe_system('mv', .git/modules/$from, .git/modules/$to);

# update .gitmodules and .git/config book keeping
spew($_, slurp($_) =~ s/\Q$from\E/$to/gr)
   for qw( .gitmodules .git/config );

my $dir_count = scalar split qr(/), $to =~ s(/$)()r;

my $derp = ('../' x (2 + $dir_count)) . $to;

# update .git/modules/$to/config book keeping
spew(
   .git/modules/$to/config,
   slurp(.git/modules/$to/config) =~ s/worktree.*/worktree = $derp/gr
);

# update $to book keeping
spew(
   $to/.git,
   'gitdir: ' . ('../' x $dir_count) . .git/modules/$to
);

safe_system(qw( git add -A ), $from, $to, '.gitmodules' );

sub safe_system {
   system(@_);

   die @_ exited poorly :(
  if $?  8;
}

sub safe_capture {
   my $ret = qx(@_);

   die @_ exited poorly :(
  if $?  8;

   return $ret;
}

sub slurp {
   open my $fh, '', $_[0];
   local $/ = undef;
   scalar $fh
}

sub spew {
   open my $fh, '', $_[0];
   print {$fh} $_[1]
}

sub is_clean { !safe_capture(qw(git status --porcelain)) }


pgpEyJW_kdh5r.pgp
Description: PGP signature