Apple Insert key is & should be where?
Pat LaVarre
to C-h w?
Anybody know?
If this much-discussed key in fact doesn't exist, then is there a
time-honoured GNU Emacs convention for where else to bind
overwrite-mode?
> this much-discussed key
That is, the `emacs` ... inside the terminal app of a current Mac OS X
... replies with text much like "GNU Emacs 20.7.1
(powerpc-apple-darwin1.4) of Sun Sep 2 2001 on diopter" when provoked
by M-x version RET.
When provoked by C-h where-is of overwrite-mode, this `emacs` says
"overwrite-mode is on insert".
I read that text to mean "overwrite-mode" appears bound by default
only to the key labelled Insert ... which, as I said, I don't see on
the keyboard.
> http://members.aol.com/plforth/moforth/20020325/ofemacs.txt
Why overwrite-mode? Well, when I'm editing memory-mapped hardware
like at this link, rather than editing a buffer of Ram, well then by
definition I prefer overwrite-mode.
> Nowadays, nobody would use an overwrite-mode editor, but back then ...
Thanks in advance. Pat LaVarre
P.S. Searches that left me as ignorant as I appear here include:
http://www.google.com/search?q=mac+emacs+%22overwrite-mode%22+insert
http://groups.google.com/groups?q=mac+emacs+%22overwrite-mode%22+insert
http://groups.google.com/groups?q=apple+mac+emacs+faq
http://groups.google.com/groups?q=global-set-key+overwrite-mode
Pat LaVarre
> convention for where ... to bind overwrite-mode
> [on a keyboard that lacks a key labelled Insert]
This (open) question correctly summarises all the rest of this thread
if yes in fact GNU Emacs lies by reporting "overwrite-mode is on
insert" in response to C-h w where-is of overwrite-mode.
Thanks again in advance,
Pat LaVarre
P.S. Thanks for lately mentioning the useful C-q 1 7 7 RET hack.
P.P.S. WHY does C-h w lie?
Eli Zaretskii
>
> > ... is there a time-honoured GNU Emacs
> > convention for where ... to bind overwrite-mode
> > [on a keyboard that lacks a key labelled Insert]
>
> This (open) question correctly summarises all the rest of this thread
> if yes in fact GNU Emacs lies by reporting "overwrite-mode is on
> insert" in response to C-h w where-is of overwrite-mode.
I'm not aware of any other key where this function is bound, except Insert.
> P.P.S. WHY does C-h w lie?
It doesn't lie; Emacs simply has no way of knowing whether your keyboard has
the Insert key.
Piet van Oostrum
EZ> Pat LaVarre wrote:
>>
>> > ... is there a time-honoured GNU Emacs
>> > convention for where ... to bind overwrite-mode
>> > [on a keyboard that lacks a key labelled Insert]
>>
>> This (open) question correctly summarises all the rest of this thread
>> if yes in fact GNU Emacs lies by reporting "overwrite-mode is on
>> insert" in response to C-h w where-is of overwrite-mode.
EZ> I'm not aware of any other key where this function is bound, except Insert.
>> P.P.S. WHY does C-h w lie?
EZ> It doesn't lie; Emacs simply has no way of knowing whether your
EZ> keyboard has the Insert key.
My Apple keyboard has the insert key labelled 'help'. On my iBook, however,
I can't find an equivalent one.
--
Piet van Oostrum <pi...@cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van....@hccnet.nl
Pat LaVarre
> > > convention for where ... to bind overwrite-mode
> > > "overwrite-mode is on insert"
> > > in response to C-h w where-is of overwrite-mode.
> > It doesn't lie; Emacs simply has no way of knowing whether your
> My Apple keyboard has the insert key labelled 'help'.
What does it mean to say a command is on a key ... if it doesn't mean
that the command and the key exist?
Yours in stunning ignorance, with thanks again in advance,
Pat LaVarre
P.S. Didn't Emacs exist with a default of not overwrite-mode a long
time __before__ Insert keys?
Alan Shutko
> What does it mean to say a command is on a key ... if it doesn't mean
> that the command and the key exist?
Unfortunately, it's not always possible to ask the system what keys
are actually on a keyboard. The message is more like "If you manage
to find a way to hit the key the system calls Whatever, I'll give you
this command."
--
Alan Shutko <a...@acm.org> - In a variety of flavors!
Surprise your boss. Get to work on time.
Pat LaVarre
> > > > convention for where ... to bind overwrite-mode ...
> > > [no]
...
> > > > "overwrite-mode is on insert"
> > > > in response to C-h w where-is of overwrite-mode. ...
> > My Apple keyboard has the insert key labelled 'help'.
> Unfortunately, it's not always possible to ask the system
> what keys are actually on a keyboard.
> The message is more like "If you manage to find a way
> to hit the key the system calls Whatever,
> I'll give you this command."
Helpful, thank you, in reply I ask ...
Is it EVER possible for GNU Emacs to know what keys are connected?
What denies us this knowledge? Unix legacy? Or an incomplete port to
Mac OS X of the character-graphic terminal command-line GNU Emacs that
ships with Mac OS X?
I'd say choosing to be ignorant of the key set feels "not Mac like"
because ...
The classic MacOS idea of a keyboard is a physical layout of keys well
enough know to let the Apple.KeyCaps tool draw that layout.
To change the language the system speaks is to change the assumed
labels for each of those keys. This works well visually: change to
"British" and you can find the pounds-sterling symbol on Shift+3 (aka
S-3), change to "Spanish" and you can find an n with a ~ (tilde) on
top on your American ; : key.
People who pop the labels off the keys, rearrange them, and put them
back, end up with a keyboard drawn wrong, ... but the rest of us do
ok.
Curiously yours, thanks in advance,
Pat LaVarre
Eli Zaretskii
>
> Is it EVER possible for GNU Emacs to know what keys are connected?
>
> What denies us this knowledge? Unix legacy? Or an incomplete port to
> Mac OS X of the character-graphic terminal command-line GNU Emacs that
> ships with Mac OS X?
The problem is the lack of a portable API that allows a program to query the
system about available keyboard keys.
Perhaps there's a Mac-specific solution to that, I don't know. But other
systems don't have such facilities, AFAIK.
Pat LaVarre
M-x overwrite-mode RET, or perhaps M-x overw TAB RET, is the least
awful idiomatic suggestion we can conceive??
> the lack of a portable API
> that allows a program to query
> the system about available keyboard keys.
Ouch. Looks like Java left this one out too. I'll go ask the
comp.lang.java.gui folk ...
> a keyboard that lacks a key labelled Insert
By the way, I see now the search:
http://groups.google.com/groups?as_q=overwrite-mode%20is%20on%20insert&as_ugroup=*emacs*
gives some history of people binding C-q to insert a single char in
overwrite-mode but other people preferring to bind C-x C-o to toggle
overwrite-mode ...
... these hits do acknowledge that the C-x C-o alternative we lose its
time-honoured default binding to delete-blank-lines.
C-h b describe-bindings here wrongly suggests that here (M-x version
RET GNU Emacs 20.7.1) no default binding exists for C-i C-m (aka TAB
RET) but correctly suggests that ESC n, Esc o, Esc p are not bound
(aka M-n M-o M-p).
Customising editors over the years, I find I can most easily remember
defining one prefix of my own and then binding the alphabet, e.g. I
could solve this for myself with M-o i. But that's by definition not
a solution I can share broadly - the point of defining my own prefix
is to give me a reasonably large space of my own.
Thanks all for helping me think,
Pat LaVarre
Edward O'Connor
>
> M-x overwrite-mode RET, or perhaps M-x overw TAB RET, is the
> least awful idiomatic suggestion we can conceive??
C-c i sounds nice to me.
> gives some history of people binding C-q to insert a single char
> in overwrite-mode but other people preferring to bind C-x C-o to
> toggle overwrite-mode ...
>
> ... these hits do acknowledge that the C-x C-o alternative we
> lose its time-honoured default binding to delete-blank-lines.
And using C-q thus would clobber the binding for quoted-insert.
> C-h b describe-bindings here wrongly suggests that here (M-x
> version RET GNU Emacs 20.7.1) no default binding exists for C-i
> C-m (aka TAB RET) but correctly suggests that ESC n, Esc o, Esc
> p are not bound (aka M-n M-o M-p).
>
> Customising editors over the years, I find I can most easily
> remember defining one prefix of my own and then binding the
> alphabet, e.g. I could solve this for myself with M-o i. But
> that's by definition not a solution I can share broadly - the
> point of defining my own prefix is to give me a reasonably large
> space of my own.
Generally, you shouldn't rely on something not showing up in the
describe-bindings list as an indicator of whether or not you
should bind a key. And yes, you're right, having your own prefix
to play with is great.
Fortunately, the above issues combine really nicely, since all of
the keybindings of the form C-c <letter> are reserved for the
user, namely, you. Hence my C-c i suggestion earlier.
Ted
--
Edward O'Connor
t...@oconnor.cx
Pat LaVarre
> are reserved for the user, namely, you.
Excellent, thank you.
Can anyone also tell me where in C-h ... I was told that and missed it?
> > > > ... is there no time-honoured GNU Emacs
> > > > convention for where ... to bind overwrite-mode
> > > > on a keyboard that lacks a key labelled Insert ...
> > > yes
Truly? Really? How can this be? Surely Emacs predates Insert keys?
Thanks again in advance, curiously yours,
Pat LaVarre
Kai Großjohann
> Can anyone also tell me where in C-h ... I was told that and missed it?
I'm sure the Emacs manual mentions it. Type C-h i d and then choose
"Emacs" from the list.
kai
--
Silence is foo!
Kevin Rodgers
>
> > From: Edward O'Connor (t...@oconnor.cx)
> > having your own prefix to play with is great.
> > ... all of the keybindings of the form C-c <letter>
> > are reserved for the user, namely, you.
>
> Excellent, thank you.
>
> Can anyone also tell me where in C-h ... I was told that and missed it?
From the Info manual:
|File: emacs, Node: Keymaps, ..., Up: Key Bindings
...
| As a user, you can redefine any key; but it might be best to stick to
|key sequences that consist of `C-c' followed by a letter. These keys
|are "reserved for users,", so they won't conflict with any properly
|designed Emacs extension. If you redefine some other key, your
|
definition may be overridden by certain extensions or major modes which
|redefine the same key.
|File: emacs, Node: Rebinding, ..., Up: Key Bindings
...
| The two-character keys consisting of `C-c' followed by a letter are
|reserved for user customizations. Lisp programs are not supposed to
|define these keys, so the bindings you make for them will be available
|in all major modes and will never get in the way of anything.
> > > > > ... is there no time-honoured GNU Emacs
> > > > > convention for where ... to bind overwrite-mode
> > > > > on a keyboard that lacks a key labelled Insert ...
> > > > yes
>
> Truly? Really? How can this be? Surely Emacs predates Insert keys?
Every command is available via `M-x'. Presumably overwrite-mode wasn't
considered to be frequently used enough to bind to a key.
--
Kevin Rodgers <kev...@ihs.com>
Pat LaVarre
> C-h i d [ info Info-directory ]
...
> From: Kevin Rodgers (kev...@ihs.com)
> From the Info manual ...
> `C-c' followed by a letter.
Ahhh thank you again. For me now likewise, if I have CapsLock off,
then:
C-h i m emacs RET m keymaps RET
leads me to "As a user, you can redefine any key; but it might be best
to stick to key sequences that consist of `C-c' followed by a letter."
I wonder how I missed this. I know I have a friend who favours
binding F1 F2 ... keys.
I imagine myself I was looking for a more completely dedicated prefix.
In the past, reasoning by analogy, if my prefix was C-c, then C-c i
would be my variation on the i key. For example, C-c C-e would be my
variation on C-e end-of-line, whereas C-c i might insert an i with an
accent.
Another example would be to create a prefix for giving the
conventional Windows meanings to keys. C-h b describe-bindings tells
me M-n M-o M-p are unbound by default, so M-o End could mean
end-of-line.
I'm thinking an emulation key prefix, rather than an emulation mode,
would be a less invasive compromise between keyboard conventions than
the:
http://www.cua.dk/cua.html
cited by:
http://members.aol.com/plscsi/emacs/emacs-windows.html
http://www.gnu.org/software/emacs/windows/ntemacs.html
ftp://ftp.gnu.org/gnu/windows/emacs/README
ftp://ftp.gnu.org/gnu/windows/emacs/latest
Pat LaVarre
http://members.aol.com/plscsi/emacs/emacs-deja-vu.html
Pat LaVarre
> > convention for where ... to bind overwrite-mode
> > on a keyboard that lacks a key labelled Insert ...
> Surely Emacs predates Insert keys?
Yes. From above we have: "M-x overwrite-mode RET, or perhaps M-x
overw TAB RET, is the least awful idiomatic suggestion we can
conceive??
Pat LaVarre
P.S. In a separate post I thank you for leading me to:
Pat LaVarre
> > version RET GNU Emacs 20.7.1) no default binding exists for C-i
> > C-m (aka TAB RET) but correctly suggests that ESC n, Esc o, Esc
> > p are not bound (aka M-n M-o M-p).
> ... Generally, you shouldn't rely
> on something not showing up in the describe-bindings list
> as an indicator of whether or not you should bind a key.
Thank you for this caution. Can anyone tell me on what should I rely
instead?
How do I discover what keys Are free?
I see in a different branch of this thread, we find the ( C-h i m
emacs RET m keymaps RET ) guidance "As a user, you can redefine any
key; but it might be best to stick to key sequences that consist of
`C-c' followed by a letter.
I conclude, when practical, I should bind ( C-c a ) thru ( C-c z ).
Glad to hear it, thank you all, but what about when that is not
practical?
How do I discover what keys Are free? How do I judge which will
likely remain free over time, apart from ( C-c a ) thru ( C-c z )?
Thanks again in advance. Pat LaVarre
Pat LaVarre
emacs RET m keymaps RET ) guidance "As a user, you can redefine any
key; but it might be best to stick to key sequences that consist of
<F9> ...
How then should I discover what "a letter" is?
How should I discover what key sequences beginning with C-c are not
bound?
I'm ready to believe a to z are "letter"s. What about A to Z and ñ
and so forth? I'm interested to know how I should answer this myself,
not just what the answer is. I see ...
C-h k describe-key shows me C-c is a prefix without volunteering any
further info.
C-h k describe-key claims falsely that "C-c C-h is undefined". Back
in the real world, C-c C-h gives me an (empty) list of "Global
Bindings Starting with C-c". I gather this list varies by mode.
C-h b describe-bindings, clearly being other than a summary of C-h k
results, says C-c is mode-specific-command-prefix. There hitting RET
to learn more shows me the same text as C-h f describe-function of
mode-specific-command-prefix (the shorthand C-h f mode TAB works too).
C-h f of mode-specific-command-prefix says merely:
Prefix command (definition is a keymap
associating keystrokes with commands).
How should I discover what "a letter" is?
How should I discover what key sequences beginning with C-c are not
bound?
Thanks again in advance, Pat LaVarre
Kai Großjohann
> How then should I discover what "a letter" is?
I think you have more ideas about this than the writers of the
documentation. I'm sure they meant a through z, but probably nobody
thought of A through Z, much less nonascii letters.
Since people didn't think about it, this probably means that you can
use them.
I'll ask the developers.