Re: [PATCH 2/5] avoid duplicate tests for shell functions

[Paolo Bonzini]

> I'm still thinking about that.  As long as you require a shell fn snippet
> before someone else suggests it, then your solution is perfect.  The
> problem comes that if someone else suggests the snippet first, then it
> still gets duplicated when you later require it.

Yes, that's it.

> I'm almost wondering whether this is a use-case for m4_set.  Both
> DETECT_REQUIRED and DETECT_SUGGESTED are maintaining a set of snippets to
> use, with the further coordination that the output of the DETECT_SUGGESTED
> test should exclude any snippets that are in the required set.

I submitted the imperfect patch because it's a oneliner so it's easy to
back out when we have a better solution.  But this is more of an excuse
than anything else. :-)

Paolo