Re: C-h v fails when a symbol has variable documentation yet is unbound

[Alan Mackenzie]

Good Morning!

On Wed, 9 Nov 2005, Richard M. Stallman wrote:

>This change seems correct.  Would you please install it?

Done.  (My very first commission to the Emacs repository.  :-)

>Meanwhile, why is it necessary for c-syntactic-context to be unbound
>most of the time?  Couldn't it have some innocuous global value?

Hmm.  Thinking about it, yes, it could indeed have been nil.  There are
several places in CC Mode which test (boundp 'c-syntactic-context).  But
there might now also be code in derived modes which test this too.  So it
seems inadvisable to change it.

But it would have been better, right from ancient history, to have used
nil rather than unboundedness here.

-- 
Alan Mackenzie (Munich, Germany)