bug#7350: 24.0.50; make vc-deduce-backend smarter

[Bob Rogers <rogers@rgrjr.dyndns.org>]

   From: Stefan Monnier <monnier@iro.umontreal.ca>
   Date: Fri, 12 Nov 2010 08:48:30 -0500

   >    It can definitely be generalized, but I'd rather stick to buffers where
   >    there's a clear association with a particular file or directory.
   >    E.g. *Help* buffers aren't good candidates.
   > Fair enough.  Seems odd that default-directory is not nil for these.

   There are all kinds of reasons why we don't like default-directory to
   be nil.

   > Works for me.  It doesn't cover the cases where I'm looking at a
   > generated file, or test output, but (in my workflow anyway) those cases
   > are rarer.

   I've installed the patch into the trunk (i.e. not for Emacs-23.3 but
   Emacs-24) and added compilation-mode to shell-mode, since it's also
   clearly linked to a particular location in the file-system.

Great; thank you.

   >    This was the result of a trade-off (get rid of one RET since it's
   >    almost never used).  But I guess we could/should prompt the user for
   >    a file/dir rather than signal "File is not under version control" or
   >    some such error.

   > That would be much nicer, especially since the underlying functionality
   > (i.e. support for vc-diff of directories) still works.

   Patch welcome.

Hmm.  TRT would be to figure this out in the "interactive" form, so that
repeat-complex-command works, but that would change the args to vc-diff.
How big a change are you willing to contemplate?

   > P.S.  I mailed the signed copyright assignment today, so FSF should
   > have it tomorrow.

   Great (tho, OT1H I can't remember for which patch this is, and OTOH the
   FSF itself usually takes a little while to process those mail).

	   Stefan

NP; it was on my mind, so I just thought I'd mention it.

					-- Bob