C-x a C-x C-h

C-x a C-x C-h

[Deniz Dogan]

When I hit `C-x a C-x' in the latest Emacs, it awaits further input,
as if it was a prefix key sequence. However, `C-x a C-x C-h' reveals
that there are no commands bound to anything beginning with that
prefix. Why?

Output:

\fGlobal Bindings Starting With C-x a C-x:
key             binding
---             -------

[back]


-- 
Deniz Dogan

RE: C-x a C-x C-h

[Drew Adams]

> When I hit `C-x a C-x' in the latest Emacs, it awaits further input,
> as if it was a prefix key sequence. However, `C-x a C-x C-h' reveals
> that there are no commands bound to anything beginning with that
> prefix. Why?

Good question. I wonder too. Same thing in Emacs 22, 21, and 20.
And C-x C-x C-x C-x C-x ... just keeps on keepin on.

`C-x a' is bound to `abbrev-map', but its value is this:

(keymap
 (110 . expand-jump-to-next-slot)
 (112 . expand-jump-to-previous-slot)
 (39 . expand-abbrev)
 (101 . expand-abbrev)
 (45 . inverse-add-global-abbrev)
 (105 keymap
      (108 . inverse-add-mode-abbrev)
      (103 . inverse-add-global-abbrev))
 (43 . add-mode-abbrev)
 (103 . add-global-abbrev)
 (1 . add-mode-abbrev)
 (108 . add-mode-abbrev))

No entry for C-x, that I can see.

And in English:

key             binding
---             -------

C-a		add-mode-abbrev
'		expand-abbrev
+		add-mode-abbrev
-		inverse-add-global-abbrev
e		expand-abbrev
g		add-global-abbrev
i		Prefix Command
l		add-mode-abbrev
n		expand-jump-to-next-slot
p		expand-jump-to-previous-slot

i g		inverse-add-global-abbrev
i l		inverse-add-mode-abbrev

I don't see anything that would explain this in the Lisp sources, grepping for
`abbrev-map'. I'm curious what the answer will be.

Re: C-x a C-x C-h

[Deniz Dogan]

2009/8/27 Drew Adams <drew.adams@oracle.com>:
>> When I hit `C-x a C-x' in the latest Emacs, it awaits further input,
>> as if it was a prefix key sequence. However, `C-x a C-x C-h' reveals
>> that there are no commands bound to anything beginning with that
>> prefix. Why?
>
> Good question. I wonder too. Same thing in Emacs 22, 21, and 20.
> And C-x C-x C-x C-x C-x ... just keeps on keepin on.

That doesn't happen here, it only gets to `C-x a C-x C-x', after which
any further keypresses will say the sequence undefined.

-- 
Deniz Dogan

RE: C-x a C-x C-h

[Drew Adams]

> >> When I hit `C-x a C-x' in the latest Emacs, it awaits 
> further input,
> >> as if it was a prefix key sequence. However, `C-x a C-x 
> C-h' reveals
> >> that there are no commands bound to anything beginning with that
> >> prefix. Why?
> >
> > Good question. I wonder too. Same thing in Emacs 22, 21, and 20.
> > And C-x C-x C-x C-x C-x ... just keeps on keepin on.
> 
> That doesn't happen here, it only gets to `C-x a C-x C-x', after which
> any further keypresses will say the sequence undefined.

I wasn't quite right.

On Emacs 20 and 21, you can just keep repeating C-x forever, with no apparent
effect.

On Emac 22 and 23, it seems that C-x a C-x C-x C-x calls
`exchange-point-and-mark', and every other `C-x' after that calls it again. So
here too I can keep repeating C-x C-x..., but each such pair invokes
`exchange-point-and-mark'.

Re: C-x a C-x C-h

[Deniz Dogan]

2009/8/27 Drew Adams <drew.adams@oracle.com>:
>> >> When I hit `C-x a C-x' in the latest Emacs, it awaits
>> further input,
>> >> as if it was a prefix key sequence. However, `C-x a C-x
>> C-h' reveals
>> >> that there are no commands bound to anything beginning with that
>> >> prefix. Why?
>> >
>> > Good question. I wonder too. Same thing in Emacs 22, 21, and 20.
>> > And C-x C-x C-x C-x C-x ... just keeps on keepin on.
>>
>> That doesn't happen here, it only gets to `C-x a C-x C-x', after which
>> any further keypresses will say the sequence undefined.
>
> I wasn't quite right.
>
> On Emacs 20 and 21, you can just keep repeating C-x forever, with no apparent
> effect.
>
> On Emac 22 and 23, it seems that C-x a C-x C-x C-x calls
> `exchange-point-and-mark', and every other `C-x' after that calls it again. So
> here too I can keep repeating C-x C-x..., but each such pair invokes
> `exchange-point-and-mark'.
>

You're right, I tried it without having set mark. It wasn't undefined here.

-- 
Deniz Dogan

RE: C-x a C-x C-h

[Drew Adams]

> You're right, I tried it without having set mark. It wasn't 
> undefined here.

In fact, that's how I discovered that it ran `exchange-point-and-mark'. I too
got the error that mark was not set, turned on `debug-on-error', and saw that
`exchange-point-and-mark' was what was invoked.

Re: C-x a C-x C-h

[Andreas Schwab]

Deniz Dogan <deniz.a.m.dogan@gmail.com> writes:

> When I hit `C-x a C-x' in the latest Emacs, it awaits further input,
> as if it was a prefix key sequence. However, `C-x a C-x C-h' reveals
> that there are no commands bound to anything beginning with that
> prefix. Why?

Probably because C-x is a prefix key in key-translation-map.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."