Re: [PATCH] Pass CONFIG_SHELL down if provided

[Eric Blake]

Eric Blake <ebb9 <at> byu.net> writes:

> In the following patch, the test fails without the changes in lib/, but
> passes afterwards.  An alternate patch would be to output
> #!${CONFIG_SHELL-/bin/sh} rather than #!$SHELL in AS_INIT_GENERATED.
> Which would you prefer?  I like the idea of making SHELL sane during any
> m4sh script (ie. this patch variant), so that constructs like '$SHELL
> $build_aux/config.sub' will work.

I went ahead and applied my patch, with one modification - it is 
_AS_DETECT_BETTER_SHELL and not AS_INIT that should set SHELL.

-- 
Eric Blake