RE: inputting characters by hexadigit

[Drew Adams]

> I don't know how all that maps to Icicles and your 
> terminology, but what I'm proposing is really very simple.
> I'll explain it differently:
...

Yes, I understood you. I replied because you mentioned Icicles, and with Icicles
your proposal is not needed (it offers no added benefit). I tried to explain
that, but I can explain it better off-list, if you are interested.

Wrt vanilla Emacs, I have nothing particular to say about your proposal.

BTW, what you describe is similar to what Imenu does: use submenus when
appropriate. In my own code (for emacs-lisp buffers), the Imenu submenus are
categories in the sense you use: User Options, Variables, Functions, Macros,
Keys in Maps... (http://www.emacswiki.org/cgi-bin/wiki/imenu%2b.el)

> DA> In Icicles, you can also filter progressively, using 
> DA> multiple patterns, so you can easily filter first with
> DA> category, then with name-in-category, if you like,
> DA> using separate patterns. But you need not do things in 
> DA> that order, and you need not even treat the two separately at all.
> 
> I don't think Icicles is the right way to implement what I'm 
> describing, because it does a lot of other things which are
> fairly intrusive to the user who like the normal completion behavior.

I didn't suggest that Icicles be added to Emacs or that its implementation be
copied. I was letting you know that you can do what you want with Icicles, in
case you're interested. You brought up Icicles; I did not:

> >> When you know what's in the list, isearch,
> >> partial-completion-mode, icicles, etc. work great.

Even when you do not know what is in the list, Icicles works great in your use
case. That was my point. 

That, and letting you know that with Icicles you can add categories (or any
other info) to completion candidates, and you can then filter on (complete
against) the category as well as the rest of the candidate text.