Re: nobase.test

[Ralf Wildenhues]

Hello Alexandre,

* Alexandre Duret-Lutz wrote on Wed, Sep 06, 2006 at 01:44:34PM CEST:
> >>> "RW" == Ralf Wildenhues <address@hidden> writes:

>  RW> Maybe there is a solution to all cases.  But until then, I don't think
>  RW> it's a good idea to advertise half-solutions, and employ half-tests.
>  RW> And yes, that is very much IMHO.
> 
> Following these lines, it's nonsense to maintain some
> half-support for non-executable install-sh.  If we don't want to
> support this use-case, ${SHELL} should be fully removed from
> every call to install-sh.  

Or it should be fixed for every use case.  Which is what I would love
to see (but am not willing to put work into and test, due to lack of
itch and time).

> And if why stop with install-sh?  Any reason mdate-sh, ylwrap, and
> friends should be handled differently than install-sh?

Oh, there may be other reasons than non-executability, e.g., the desire
to use a decent shell.

> (I'm not being sarcastic here; I just don't care which way the
> balance tilts, as long as it clearly tilts.)

I'm not so sure it clearly tilts.  Let's put the test back in when we
fix $(MKDIR_P) and the other three instances I've mentioned.  But let's
not advertise this as fixed when it isn't.

Cheers,
Ralf