bug#11718: 24.1.50; `all-completions' returns results with wrong case

[Drew Adams]

> Drew, I have no idea what you're hoping to get.  I already 
> agreed before you even sent a single message in this thread.
> It's not like I'm rejecting a patch or something.

Sorry, it was not clear to me that you agreed.  Glad you do.

FWIW, this is what sounded equivocal (unclear) to me:

> While not strictly wrong, it is indeed an undesirable result.
        ^^^^^^^^^^^^^^^^^^
> I'll try and see how to fix it.

> So while we do want to fix the problem, code should not
> not assume anything about the particular case of the return
  ^^^^^^^^^^^^^^^^^^^
> string (which is only considered to be a "cosmetic" issue).
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

I did hear you say clearly that you want to fix the problem.

But I also (apparently mis-) heard you saying that code should not expect the
case of the return string to correspond to the completions provided by the
COLLECTION arg.  It sounded like your position might be that all bets are off
wrt the result.

Sorry for any noise.  Glad you agree.