Re: [PATCH] var(win32): do report the GIT_SHELL_PATH that is actually used

[Phillip Wood <phillip.wood123@gmail.com>]

Hi Johannes

On 08/07/2024 14:02, Johannes Schindelin via GitGitGadget wrote:
> From: Johannes Schindelin <johannes.schindelin@gmx.de>

Thanks for putting a patch together so quickly

>   static char *shell_path(int ident_flag UNUSED)
>   {
> +#ifdef WIN32
> +	char *p = locate_in_PATH("sh");

If I'm reading is_busybox_applet() (which only exists in 
git-for-windows) correctly then this will return "busybox.exe" under 
mingit-busybox rather than ash.exe, so the calling program would have to 
know to set argv[0] (which is likely not possible unless the calling 
program is written in C) or pass "sh" as the first argument. As the code 
to support busybox does not exist upstream I guess that's best handled 
downstream.

Best Wishes

Phillip

> +	convert_slashes(p);
> +	return p;
> +#else
>   	return xstrdup(SHELL_PATH);
> +#endif
>   }
>   
>   static char *git_attr_val_system(int ident_flag UNUSED)
> diff --git a/t/t0007-git-var.sh b/t/t0007-git-var.sh
> index ff4fd9348cc..9fc58823873 100755
> --- a/t/t0007-git-var.sh
> +++ b/t/t0007-git-var.sh
> @@ -157,7 +157,7 @@ test_expect_success POSIXPERM 'GIT_SHELL_PATH points to a valid executable' '
>   test_expect_success MINGW 'GIT_SHELL_PATH points to a suitable shell' '
>   	shellpath=$(git var GIT_SHELL_PATH) &&
>   	case "$shellpath" in
> -	*sh) ;;
> +	[A-Z]:/*/sh.exe) test -f "$shellpath";;
>   	*) return 1;;
>   	esac
>   '
> 
> base-commit: 06e570c0dfb2a2deb64d217db78e2ec21672f558