Re: git-submodule path computation bug with recursive submodules

[Jens Lehmann <Jens.Lehmann@web.de>]

Am 05.07.2012 16:18, schrieb Phil Hord:
> On Thu, Jul 5, 2012 at 8:09 AM, Bob Halley <halley@play-bow.org> wrote:
>> The first level of submodule clones successfully, but submodule 'foo'
>> contains submodule 'bar', and cloning it fails.
> 
> It fails for me too, running zsh and stock Linux.
> 
> Seems to originate here.
> 
> Commit: 69c3051780d6cacfe242563296160634dc667a90:
> Author: Jens Lehmann <Jens.Lehmann@web.de>
> Date:   Sun Mar 4 22:15:36 2012 +0100
> 
>     submodules: refactor computation of relative gitdir path

Yup, thanks both for reporting and nailing that one down.

This diff fixes the problem for me and should be a portable way to
avoid "pwd -P" or "cd -P", I'll cook up a patch with test for that:


diff --git a/git-submodule.sh b/git-submodule.sh
index 5c61ae2..4a22555 100755
--- a/git-submodule.sh
+++ b/git-submodule.sh
@@ -150,8 +150,8 @@ module_clone()
                die "$(eval_gettext "Clone of '\$url' into submodule path '\$sm_
        fi

-       a=$(cd "$gitdir" && pwd)/
-       b=$(cd "$sm_path" && pwd)/
+       a=$(cd_to_toplevel && cd "$gitdir" && pwd)/
+       b=$(cd_to_toplevel && cd "$sm_path" && pwd)/
        # normalize Windows-style absolute paths to POSIX-style absolute paths
        case $a in [a-zA-Z]:/*) a=/${a%%:*}${a#*:} ;; esac
        case $b in [a-zA-Z]:/*) b=/${b%%:*}${b#*:} ;; esac