Key binding M-g should really be goto-line

Key binding M-g should really be goto-line

[Jari Aalto+mail.emacs]

[please keep CC, I'm not in this list]

I noticed that in CVS lisp/ the M-g mapping reads:

  ldefs-boot.el: (define-key global-map "\M-g" 'facemenu-keymap)

A while ago there was a long discussion:

  http://lists.gnu.org/archive/html/emacs-devel/2004-03/msg00642.html

But I dare to open this question again. I understood that the goal of
keeping M-g where is now is was that in some distant foggy future it might
be used for text editing. 

    http://lists.gnu.org/archive/html/emacs-devel/2004-03/msg00642.html
    From: Richard Stallman

    "I want Emacs to move in the direction of doing word processing.  It
    may take years, but we will get there.  Then commands to specify faces
    will become important, and will need a good key binding.

    I chose the M-g binding for that reason, and the reason continues to
    have force.  So I don't intend to change that binding."

Please, I have been watching this future over 10 years now and in the mean
time hundreds or thousands developers / code writers / designers / students
/ teachers are all using

    M-g to mean goto-line

Is there no hope to see the 'facemenu-keymap moved to another key and have
the goto-line solely for M-g? It really cannot be replaced by M-x compile /
M-x grep or other specialized modes that offer "direct to line" jump
commands. Line numbers can come from many outside sources, like:

    o   Remote sessions; you're editing two copies of Version controlled 
        files
    o   Other programming languages; Like PHP which display errors and
        line numbers in web page itself (can't use Emacs)
    o   Numerous other utilities (like Web link checkers, when you
        edit the file in Emacs)

To make it short: from accessibility point of view, the goto-line
functionality would be far more important and widely used in everyday Emacs
session and ad hoc use, than the current facemenu-keymap functionality. 

Jari

Re: Key binding M-g should really be goto-line

[David Kastrup]

jari.aalto@cante.net (Jari Aalto+mail.emacs) writes:

>     http://lists.gnu.org/archive/html/emacs-devel/2004-03/msg00642.html
>     From: Richard Stallman
>
>     "I want Emacs to move in the direction of doing word processing.  It
>     may take years, but we will get there.  Then commands to specify faces
>     will become important, and will need a good key binding.
>
>     I chose the M-g binding for that reason, and the reason continues to
>     have force.  So I don't intend to change that binding."
>
> Please, I have been watching this future over 10 years now and in the mean
> time hundreds or thousands developers / code writers / designers / students
> / teachers are all using
>
>     M-g to mean goto-line

I'd second that.  And the reason is that it is my opinion that even in
the text processing case M-g has little to none mnemonic value, and
more in the context of text processing functions is needed, anyway,
than just font selection.  If we go that lane, it will more probably
require opening a modifier like "super" or "alt" for it.  Most
keyboards by now readily have enough modifiers available.  We will
need a terminal shortcut for terminals that don't have it, like for
Meta.  Esc-Esc would be one possibility.  It is easier to type than
M-g, and it has more "this is a new class of commands" connotations.
The two bindings already starting with Esc Esc would not suffer much
(The Esc M-: binding is available as M-:, anyway).

I think that would be a good escape route for text processing
commands.  I certainly would like to see M-g for goto-line.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Eli Zaretskii]

> From: David Kastrup <dak@gnu.org>
> Date: Wed, 02 Mar 2005 00:35:09 +0100
> Cc: emacs-devel@gnu.org
> 
> > Please, I have been watching this future over 10 years now and in the mean
> > time hundreds or thousands developers / code writers / designers / students
> > / teachers are all using
> >
> >     M-g to mean goto-line
> 
> I'd second that.

Please, we had an extremely prolonged discussion about this not long
ago, let's not have it again!

Re: Key binding M-g should really be goto-line

[Nick Roberts]

 > Please, we had an extremely prolonged discussion about this not long
 > ago, let's not have it again!

I agree. Its only worth opening again if something has changed since the
last discussion.


Nick

Re: Key binding M-g should really be goto-line

[Miles Bader]

> Please, we had an extremely prolonged discussion about this not long
> ago, let's not have it again!

Same person started that thread too... :-/

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Jari Aalto+mail.emacs]

|  > Please, we had an extremely prolonged discussion about this not long
|  > ago, let's not have it again!
| 
| I agree. Its only worth opening again if something has changed since the
| last discussion.

There is really a need to discuss this, because I see daily situations
where we need the goto-line functionality that Emacs does not offer after
boot. Apache/PHP/Emacs Web development combination is one of them.

The previous discussion did not reach to any consensus on the matter. Emacs
already had M-g to to mean goto-line at some point. Nobody explained good
arguments why current M-g is immovable?

My new suggestion is that

- Move the currently rarely used facemenu-keymap to *another* key.
- Add key binding M-g to mean goto-line as it has been de facto already fo
  long time

I believe these are reasonable feature requests. If we think about the
average Emacs user, I don't know where he would use facemenu-keymap - Most
starters consider Emacs just "another text editor", Like Notepad in
Windows or vi replacement in Unix.

Jari

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Alfred M. Szmidt]

   Nobody explained good arguments why current M-g is immovable?

,----
| http://lists.gnu.org/archive/html/emacs-devel/2004-03/msg00642.html
| From: Richard Stallman
| 
| "I want Emacs to move in the direction of doing word processing.  It
| may take years, but we will get there.  Then commands to specify faces
| will become important, and will need a good key binding.
| 
| I chose the M-g binding for that reason, and the reason continues to
| have force.  So I don't intend to change that binding."
`----

... and I doubt much has changed in the year that has passed since
then.

Jari, you really have to stop arguing from the standpoint "Emacs must
have M-g as goto-line or else".  I really feel that the use of
goto-line to warrant a single char key binding is overblown--specially
for a single keystroke binding!; if people use it that often then one
should write a proper mode for Emacs to ease whatever one is doing
instead of introducing a new keybinding.

But what is wrong with C-x ~ for example? What about allowing for
something like C-u C-u C-l to recenter at a specific line (if it is
just C-u C-u C-l, then prompt the user, if C-u C-u 100 C-l then
recenter at line 100)?  I think this is quite nice, makes sense
keybinding wise.

Happy hacking.

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Miles Bader]

On Wed, 02 Mar 2005 12:30:31 +0100, Alfred M. Szmidt <ams@kemisten.nu> wrote:
> C-u C-u C-l to recenter at a specific line (if it is
> just C-u C-u C-l, then prompt the user, if C-u C-u 100 C-l then
> recenter at line 100)?  I think this is quite nice, makes sense
> keybinding wise.

Two C-u's plus a number...?  That seems pretty convoluted...

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Alfred M. Szmidt]

   > C-u C-u C-l to recenter at a specific line (if it is just C-u C-u
   > C-l, then prompt the user, if C-u C-u 100 C-l then recenter at
   > line 100)?  I think this is quite nice, makes sense keybinding
   > wise.

   Two C-u's plus a number...?  That seems pretty convoluted...

Why do you think that this is convoluted?  It isn't very hard to type.

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Miles Bader]

On Wed, 02 Mar 2005 14:54:30 +0100, Alfred M. Szmidt <ams@kemisten.nu> wrote:
>    > C-u C-u C-l to recenter at a specific line (if it is just C-u C-u
>    > C-l, then prompt the user, if C-u C-u 100 C-l then recenter at
>    > line 100)?  I think this is quite nice, makes sense keybinding
>    > wise.
> 
>    Two C-u's plus a number...?  That seems pretty convoluted...
> 
> Why do you think that this is convoluted?  It isn't very hard to type.

I didn't say it was hard to type -- it doesn't follow typical Emacs
usage.  That makes it hard to remember.  It also just feels "weird"
and kludgey (especially being put on C-l which is _already_ oddly
overloaded).

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Andreas Schwab]

"Alfred M. Szmidt" <ams@kemisten.nu> writes:

> But what is wrong with C-x ~ for example?

This is a rather suboptimal choice because ~ is difficult to type on many
keyboards.  Personally, I have put goto-line on [f1] already a very long
time ago and don't care about any default binding.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Stefan Monnier]

> have M-g as goto-line or else".  I really feel that the use of
> goto-line to warrant a single char key binding is overblown--specially
> for a single keystroke binding!; if people use it that often then one
> should write a proper mode for Emacs to ease whatever one is doing
> instead of introducing a new keybinding.

We've gone through that umpteen times.  It just so happens that "the proper
mode" has not yet been written for each and every case, and that in many
cases it can't be written because the line number is not directly available
to Emacs but has to be typed in by the user (e.g. it's given over the
phone, or it's printed in your web-browser's window, ...).

The reason why people use goto-line is not really relevant: the fact is that
they do use it, even if it means M-x goto-line RET or adding a binding
in their .emacs for it.
I.e. there's a very clear need for it.

Even in this very mailing list of highly experienced Emacs hackers, many
people have mentioned that they have goto-line bound to some key.
I used to have it bound to M-g.  I myself have hacked isearch so that
C-s <number> does goto-line.

Arguing "you should follow the true path of enlightenment because I know
better than you" is just ridiculous,


        Stefan

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Alfred M. Szmidt]

   Arguing "you should follow the true path of enlightenment because I
   know better than you" is just ridiculous,

Look, you obviously didn't read anything of what I wrote, I never
argued or implied that one shouldn't bind a key to goto-line; I even
suggested that it _should_ be bound to a key.

Sighs...

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2 autolearn=no version=3.0.2

[Jari Aalto+mail.linux]

|    Nobody explained good arguments why current M-g is immovable?
| 
| ,----
| | http://lists.gnu.org/archive/html/emacs-devel/2004-03/msg00642.html
| | From: Richard Stallman
| | 
| | "I want Emacs to move in the direction of doing word processing.  It
| | may take years, but we will get there.  Then commands to specify faces
| | will become important, and will need a good key binding.
| | 
| | I chose the M-g binding for that reason, and the reason continues to
| | have force.  So I don't intend to change that binding."
| `----

I meant "immovable" - in literal sense. I'm not suggesting that the
facemenu is removed. But I have big question in my mind why it has to be at
M-g - at binding - which doesn't even sound anything familiar to do with
faces.

| I really feel that the use of
| goto-line to warrant a single char key binding is overblown--specially
| for a single keystroke binding!; if people use it that often then one
| should write a proper mode for Emacs to ease whatever one is doing
| instead of introducing a new keybinding.

What I want to demonstrate is that there is daily need for it; when it is
needed. It's not my invention, while I'm speaker of it here. Perhaps there
is some blindness to see this from another perspective?

Example: How would students/designers use Emacs for PHP coding, where the
remote site displays error messages and line numbers in a web page? 

    Host A          Host B
    Win             Linux, Unix (Apache/PHP)
    client      =>  PHP program
    
    The client A sees Host B's PHP files through SSH, FTP, Version control
    or even SAMBA shares. Program is run at B, but browser at A is used
    to control it.
    
NOTE: PHP is the most popular Web development language ever and it has
skyrocketed since 2000. Now that is has proper OO with 5.x, I assume the
popularity will grow even more.

This is not a small issue. Try coding 10 000 lines of PHP without
goto-line. The most, the daily bread is programming, not the "text editing"
features that M-g is offering.

Would anyone really consider facemenu-keymap to be more important, more
essential than the primitive goto-line functionality? 

Jari

Re: Key binding M-g should really be goto-line

[Kim F. Storm]

jari.aalto@cante.net (Jari Aalto+mail.linux) writes:

> |    Nobody explained good arguments why current M-g is immovable?
> | 
> | ,----
> | | http://lists.gnu.org/archive/html/emacs-devel/2004-03/msg00642.html
> | | From: Richard Stallman
> | | 
> | | "I want Emacs to move in the direction of doing word processing.  It
> | | may take years, but we will get there.  Then commands to specify faces
> | | will become important, and will need a good key binding.
> | | 
> | | I chose the M-g binding for that reason, and the reason continues to
> | | have force.  So I don't intend to change that binding."
> | `----
>
> I meant "immovable" - in literal sense. I'm not suggesting that the
> facemenu is removed. But I have big question in my mind why it has to be at
> M-g - at binding - which doesn't even sound anything familiar to do with
> faces.

I also suggested this little piggy when M-g was discussed last time:

It keeps the current face menu bindings, but in addition allows you to
enter M-g 123 RET to jump to line 123...  Only problem is that it
doesn't show the "Goto line:" prompt until you type the first digit.

(defun goto-line-piggyback ()
  (interactive)
  (goto-line
   (string-to-int
    (read-string "Goto line: "
		 (substring (this-command-keys) -1)))))

(dolist (digit '(?0 ?1 ?2 ?3 ?4 ?5 ?6 ?7 ?8 ?9))
  (define-key facemenu-keymap (vector digit) 'goto-line-piggyback))



BTW, who appends "autolearn=no version=3.0.2" to the subject ?

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2 autolearn=no version=3.0.2

[Alfred M. Szmidt]

   Would anyone really consider facemenu-keymap to be more important,
   more essential than the primitive goto-line functionality?

Yes, the majority doesn't do PHP programming.

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2 autolearn=no version=3.0.2 autolearn=no version=3.0.2

[Jari Aalto]

|    Would anyone really consider facemenu-keymap to be more important,
|    more essential than the primitive goto-line functionality?
| 
| Yes, the majority doesn't do PHP programming.

That paragraph does not say anything about any particular use of
programming languages. Please understand that I only used some examples
where the need would obviously arise. 

We should consider which one is more important for average day to day use;
whatever that be.

Jari

RE: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Drew Adams]

    Example: How would students/designers use Emacs for PHP coding,
    where the remote site displays error messages and line numbers
    in a web page?

Not to distract you all from the burning issue of `M-g', but how about a
poor-man's compile buffer? If the target line number is already displayed in
Emacs somewhere, why bother to type it in to `goto-line'?

It wouldn't help with a Web page display, of course (unless the page were
viewed in Emacs), but it sounds from your use case as if it might be
generally useful to have a command that picks up the line-number from the
text at point (whenever that text can be parsed as a numeral) and does
`goto-line' in buffer `(other-buffer (current-buffer) t)'.

And it might be useful to have an equivalent mouse command: click a
displayed numeral to go to that line in the other buffer.

To be able to specify a different target buffer, you could use a prefix arg,
but you should not have to type anything (even RET) for must uses of the
command - the default target buffer should be what you want. IOW, no default
buffer-name with a prompt; just go to other-buffer.

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Andreas Schwab]

"Drew Adams" <drew.adams@oracle.com> writes:

> It wouldn't help with a Web page display, of course (unless the page were
> viewed in Emacs), but it sounds from your use case as if it might be
> generally useful to have a command that picks up the line-number from the
> text at point (whenever that text can be parsed as a numeral) and does
> `goto-line' in buffer `(other-buffer (current-buffer) t)'.

We already have that, it's compilation-minor-mode.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

RE: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Drew Adams]

    > it might be generally useful to have a command that picks up the
    > line-number from the text at point (whenever that text can be
    > parsed as a numeral) and does `goto-line' in buffer
    > `(other-buffer (current-buffer) t)'.

    We already have that, it's compilation-minor-mode.

Looking at the code for compilation-minor-mode (as I'm inexperienced with
it), I see no connection with what I suggested. That mode appears to work
only in buffers that can be parsed to work with next-error etc.

What I suggested was a simple command to pick up a numeral from any buffer,
regardless of what the numeral might mean in that buffer. It would be
rudimentary, but would do at least what people are doing with `goto-line',
without requiring them to key in the line number. A minor suggestion -
that's all.

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Kevin Rodgers]

Drew Adams wrote:
> Looking at the code for compilation-minor-mode (as I'm inexperienced with
> it), I see no connection with what I suggested. That mode appears to work
> only in buffers that can be parsed to work with next-error etc.
> 
> What I suggested was a simple command to pick up a numeral from any buffer,
> regardless of what the numeral might mean in that buffer. It would be
> rudimentary, but would do at least what people are doing with `goto-line',
> without requiring them to key in the line number. A minor suggestion -
> that's all.

Is this what you have in mind?

(defun goto-line-at-point (&optional buffer)
   "Go to the line whose number is given at point, counting like \
\\[goto-line].
With a prefix arg, prompt for a BUFFER and select it before moving."
   (interactive (list (if current-prefix-arg
                          (read-buffer "Goto buffer: "
                                       (other-buffer (current-buffer) t)
                                       t))))
   (when buffer
     (pop-to-buffer buffer))
   (goto-line (number-at-point)))

If so, maybe a mouse- version would be useful, too.

-- 
Kevin Rodgers

RE: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Drew Adams]

    > What I suggested was a simple command to pick up a numeral
    > from any buffer,
    > regardless of what the numeral might mean in that buffer. It would be
    > rudimentary, but would do at least what people are doing with
    > `goto-line',
    > without requiring them to key in the line number.

    Is this what you have in mind?

    (defun goto-line-at-point (&optional buffer)
       "Go to the line whose number is given at point, counting like \
    \\[goto-line].
    With a prefix arg, prompt for a BUFFER and select it before moving."
       (interactive (list (if current-prefix-arg
                              (read-buffer "Goto buffer: "
                                           (other-buffer (current-buffer) t)
                                           t))))
       (when buffer
         (pop-to-buffer buffer))
       (goto-line (number-at-point)))

I was thinking more like the following. You don't want to have to input the
buffer each time (even hitting RET to get the default). And it's unlikely
that you would want to go to a line in the same buffer in which the line
number appears.

(defun goto-line-at-point (buffer)
  "In another buffer, go to the line whose number is at point.
With prefix argument, you are prompted for the buffer.
Without it, `other-buffer' is used."
  (interactive
   (list (if current-prefix-arg
             (read-buffer
                "Buffer: "
                (other-buffer (current-buffer) t)
                t)
           (other-buffer (current-buffer) t))))
  (let ((lineno (or (number-at-point)
                    (error "No number near cursor"))))
    (unless (wholenump lineno)
            (setq lineno (abs (truncate lineno))))
    (message "Line %s in buffer `%s'" lineno buffer)
    (pop-to-buffer buffer)
    (goto-line lineno)))

    If so, maybe a mouse- version would be useful, too.

Yes, that's what I was suggesting. I don't have a need for such commands,
but it sounded like they might be useful for the use case cited.

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Richard Stallman]

How about making goto-line suggest the number at point
as its default argument?


(defun goto-line (arg &optional buffer)
  "Goto line ARG, counting from line 1 at beginning of buffer.
With just C-u as argument, move point in the most recently displayed
other buffer, and display it."
  (interactive
   (if (and current-prefix-arg (not (consp current-prefix-arg)))
       (list (prefix-numeric-value current-prefix-arg))
     ;; Look for a default, a number in the buffer at point.
     (let ((default
	     (save-excursion
	       (skip-chars-backward "0-9")
	       (if (looking-at "[0-9]")
		   (buffer-substring-no-properties
		    (point)
		    (progn (skip-chars-forward "0-9")
			   (point)))))))
       ;; Read the argument, offering that number (if any) default.
       (list (read-from-minibuffer (format "Goto line (%s): " default)
				   nil nil t
				   'minibuffer-history
				   default)
	     (if (consp current-prefix-arg)
		 (other-buffer (current-buffer) t))))))
  (with-current-buffer (or buffer (current-buffer))
    (save-restriction
      (widen)
      (goto-char 1)
      (if (eq selective-display t)
	  (re-search-forward "[\n\C-m]" nil 'end (1- arg))
	(forward-line (1- arg)))
      (if buffer
	  (let ((window (get-buffer-window buffer)))
	    (if window
		(set-window-point window (point))
	      (display-buffer buffer)))))))

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Miles Bader]

On Fri, 04 Mar 2005 18:45:32 -0500, Richard Stallman <rms@gnu.org> wrote:
> How about making goto-line suggest the number at point
> as its default argument?

That seems very convenient!

I think it may also be convenient if, point is not on a number, use
the first number on the current line as a default.

This is because I think it's common to have point sitting at the
beginning of an error line, and this would save users the effort of
manually positioning point over the actual line number (the heuristic
could fail in some cases if for instance a filename contained digits,
but I think it would often be right).

BTW, also maybe the prompt in the "use last selected buffer" case
should mention the buffer name.

Thanks,

-Miles
-- 
Do not taunt Happy Fun Ball.

RE: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Drew Adams]

    > How about making goto-line suggest the number at point
    > as its default argument?

    That seems very convenient!

I haven't yet received RMS's email, but yes, why not?

    I think it may also be convenient if, point is not on a number, use
    the first number on the current line as a default.

    This is because I think it's common to have point sitting at the
    beginning of an error line, and this would save users the effort of
    manually positioning point over the actual line number (the heuristic
    could fail in some cases if for instance a filename contained digits,
    but I think it would often be right).

    BTW, also maybe the prompt in the "use last selected buffer" case
    should mention the buffer name.

FYI - I don't suggest that `goto-line' itself should be changed this way,
but the code I sent for `goto-line-at-point' (which picks up a number in a
buffer to use as goto-line line number in another buffer), in my version,
uses `number-nearest-point' instead of `number-at-point'. That does what you
suggest wrt a number at line start. `number-nearest-point' is defined here:
http://www.emacswiki.org/elisp/thingatpt-plus.el.

Re: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Miles Bader]

On Fri, 4 Mar 2005 16:49:13 -0800, Drew Adams <drew.adams@oracle.com> wrote:
> FYI - I don't suggest that `goto-line' itself should be changed this way,
> but the code I sent for `goto-line-at-point'

I think goto-line should do it.  Remember, rms's suggestion is that
goto-line use the line number from the buffer as a _default_, not jump
there immediately.  I think that is very desirable, as such heuristics
can guess wrong, and this gives the user a chance to correct a mistake
(and for the many cases where the default is correct, hitting RET to
accept it is quick and painless).

Also it looks like `goto-line' will get a convenient key-binding (M-o)... :-/

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Luc Teirlinck]

Miles Bader wrote:

   Also it looks like `goto-line' will get a convenient key-binding (M-o)...

In that case, it might confuse the user that this binding will be
shadowed in gnus (gnus-server-open-all-servers), dired-x
(dired-omit-mode), ibuffer (ibuffer-visit-buffer-1-window) ans ses
(ses-insert-column).

Sincerely,

Luc.

Re: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Reiner Steib]

On Sat, Mar 05 2005, Luc Teirlinck wrote:

> Miles Bader wrote:
>
>    Also it looks like `goto-line' will get a convenient key-binding (M-o)...
>
> In that case, it might confuse the user that this binding will be
> shadowed in gnus (gnus-server-open-all-servers), [...]

I don't think that anyone needs a key binding for `goto-line' in Gnus'
server buffer (which is usually just a few lines).

Apart from this specific example, my understanding is that it is
normal for major (and minor) modes to override global key bindings
(info "(emacs)Local Keymaps").

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

RE: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Drew Adams]

    > FYI - I don't suggest that `goto-line' itself should be
    > changed this way, but the code I sent for `goto-line-at-point'

    I think goto-line should do it.

What is "it"?  You cut off the rest of my sentence where I said what I meant
by "changed this way": use `number-nearest-point'. I suspect that your "it"
is not my "this way", but I'm not sure what you mean.

RMS has already said that he is not interested in such nearest-point stuff
in Emacs. That's why I changed it to `number-at-point' in the code I sent,
and that's why I said that I don't propose it for `goto-line' itelf. I
mentioned it only "FYI", for the original poster, in particular.

Your suggestion to pick up the line number at bol might be acceptable to
RMS, however. I think that the main pb he had with the "nearest" stuff was
that its scope was not sufficiently constrained. That is not a pb with bol.

Re: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Miles Bader]

On Fri, 4 Mar 2005 18:00:51 -0800, Drew Adams <drew.adams@oracle.com> wrote:
> Your suggestion to pick up the line number at bol might be acceptable to
> RMS, however. I think that the main pb he had with the "nearest" stuff was
> that its scope was not sufficiently constrained. That is not a pb with bol.

I suggested it use the first number on the current line, even if it is
not at the beginning of the line.  Because it only searches one line
(in a very predictable way), the result is used only as a default, and
it happens to fit a very common usage scenario, I think this is an
acceptable and useful heuristic.  YMMV.

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Stefan Monnier]

> Also it looks like `goto-line' will get a convenient key-binding (M-o)... :-/

I really wish it weren't M-o but M-g instead.
Many people already use M-g, it's mnemonic (in English), XEmacs uses M-g, ...


        Stefan

Re: Key binding M-g should really be goto-lineautolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Miles Bader]

On Sat, 05 Mar 2005 14:00:42 -0500, Stefan Monnier
<monnier@iro.umontreal.ca> wrote:
> > Also it looks like `goto-line' will get a convenient key-binding (M-o)... :-/
> 
> I really wish it weren't M-o but M-g instead.
> Many people already use M-g, it's mnemonic (in English), XEmacs uses M-g, ...

and heeeeere we go again...

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line

[Johan Bockgård]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> Also it looks like `goto-line' will get a convenient key-binding
>> (M-o)... :-/
>
> I really wish it weren't M-o but M-g instead.

Maybe i misunderstood, but it looked to me like RMS did agree to use
M-g:

    Jari> Is there no hope to see the 'facemenu-keymap moved to
    Jari> another key and have the goto-line solely for M-g?

    RMS> What other key do you suggest? M-o would be ok, if people
    RMS> want.

-- 
Johan Bockgård

Re: Key binding M-g should really be goto-line

[Miles Bader]

On Sat, 05 Mar 2005 21:22:28 +0100, Johan Bockgård
<bojohan+news@dd.chalmers.se> wrote:
> > I really wish it weren't M-o but M-g instead.
> 
> Maybe i misunderstood, but it looked to me like RMS did agree to use
> M-g:
> 
>     Jari> Is there no hope to see the 'facemenu-keymap moved to
>     Jari> another key and have the goto-line solely for M-g?
> 
>     RMS> What other key do you suggest? M-o would be ok, if people
>     RMS> want.

You may be right; I don't really care which one it is, the point I was
trying to make is the same.

[Though using M-g would have the advantage of greatly reducing the
amount of whining on this list...]

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line

[David Kastrup]

Miles Bader <snogglethorpe@gmail.com> writes:

> On Sat, 05 Mar 2005 21:22:28 +0100, Johan Bockgård
> <bojohan+news@dd.chalmers.se> wrote:
>> > I really wish it weren't M-o but M-g instead.
>> 
>> Maybe i misunderstood, but it looked to me like RMS did agree to use
>> M-g:
>> 
>>     Jari> Is there no hope to see the 'facemenu-keymap moved to
>>     Jari> another key and have the goto-line solely for M-g?
>> 
>>     RMS> What other key do you suggest? M-o would be ok, if people
>>     RMS> want.
>
> You may be right; I don't really care which one it is, the point I was
> trying to make is the same.
>
> [Though using M-g would have the advantage of greatly reducing the
> amount of whining on this list...]

That would be reason alone.  So since the rationale of "M-g" was IIRC
"it's available" and since Richard seems to have agreed to using "M-o"
for facemenu-keymap, let us just wait for him to reconfirm and end the
whining about this matter once and for all, starting now.

Hopefully.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Kim F. Storm]

Richard Stallman <rms@gnu.org> writes:

> How about making goto-line suggest the number at point
> as its default argument?

I thought we had a feature freeze :-|

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Miles Bader]

On Sun, 06 Mar 2005 22:15:12 +0100, Kim F. Storm <storm@cua.dk> wrote:
> > How about making goto-line suggest the number at point
> > as its default argument?
> 
> I thought we had a feature freeze :-|

It's called "jugement"...

[Massive hacking of display code (or even, really, minor hacking of
display code) == "no fucking way"; minor tweak to prompt of
"goto-line" == "who cares?"]

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Miles Bader]

> It's called "jugement"...

Some people may also use the word "judgement"...

Egg
on
face

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line autolearn=noversion=3.0.2 autolearn=no version=3.0.2

[Andreas Schwab]

"Drew Adams" <drew.adams@oracle.com> writes:

>     > it might be generally useful to have a command that picks up the
>     > line-number from the text at point (whenever that text can be
>     > parsed as a numeral) and does `goto-line' in buffer
>     > `(other-buffer (current-buffer) t)'.
>
>     We already have that, it's compilation-minor-mode.
>
> Looking at the code for compilation-minor-mode (as I'm inexperienced with
> it), I see no connection with what I suggested. That mode appears to work
> only in buffers that can be parsed to work with next-error etc.

Sorry, I misunderstood your suggestion.

> What I suggested was a simple command to pick up a numeral from any buffer,
> regardless of what the numeral might mean in that buffer. It would be
> rudimentary, but would do at least what people are doing with `goto-line',
> without requiring them to key in the line number. A minor suggestion -
> that's all.

You already don't have to type the number, just mark, copy and paste it.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Eli Zaretskii]

> CC: Eli Zaretskii <eliz@gnu.org>, emacs-devel@gnu.org
> From: jari.aalto@cante.net (Jari Aalto+mail.emacs)
> Date: Wed, 2 Mar 2005 10:31:13 +0200
> 
> |  > Please, we had an extremely prolonged discussion about this not long
> |  > ago, let's not have it again!
> | 
> | I agree. Its only worth opening again if something has changed since the
> | last discussion.
> 
> There is really a need to discuss this

It's hardly surprising that you say this, since you were the one who
started that discussion the last time.

> I see daily situations where we need the goto-line functionality
> that Emacs does not offer after boot.

Then add a line to your .emacs and be done with that.

> My new suggestion is that
> 
> - Move the currently rarely used facemenu-keymap to *another* key.
> - Add key binding M-g to mean goto-line as it has been de facto already fo
>   long time

How is that different from your old suggestion?

Re: Key binding M-g should really be goto-line

[Alfred M. Szmidt]

Since Richard really wants facemenu-keymap as a keybinding, what about
making a compromise?  Bind goto-line to M-o, or bind facemenu-keymap
to M-o and goto-line to M-g?  Or maybe modify a existing keybinding so
that C-u ... does goto-line (maybe M-g is a candidate for this?)? Or
bind goto-line to something like C-x ~.

FWIW, I fail to see why people want goto-line, I never used it,
facemenu-keymap atleast has a potential use already, and in the
future; whereas goto-line is just a crutch.  And I fail to see the
fetish people have over keybidnings, specially M-g.

Cheers.

Re: Key binding M-g should really be goto-line

[Ralf Angeli]

* Alfred M. Szmidt (2005-03-02) writes:

> FWIW, I fail to see why people want goto-line, I never used it,
> facemenu-keymap atleast has a potential use already, and in the
> future; whereas goto-line is just a crutch.  And I fail to see the
> fetish people have over keybidnings, specially M-g.

One thing I did very soon after starting to use Emacs was to bind
`M-g' to `goto-line'.  I am using it often when dealing with patches
generated with `diff' and wanting to look at the original source
before applying a patch.

-- 
Ralf

Re: Key binding M-g should really be goto-line

[Alfred M. Szmidt]

   I am using it often when dealing with patches generated with `diff'
   and wanting to look at the original source before applying a patch.

That is what `diff-mode' and C-c C-c (or <RET>) is for.  So once
again, I fail to see what use goto-line has for other then in really
awkward situations.

Cheers.

Re: Key binding M-g should really be goto-line

[Ralf Angeli]

* Alfred M. Szmidt (2005-03-02) writes:

>    I am using it often when dealing with patches generated with `diff'
>    and wanting to look at the original source before applying a patch.
>
> That is what `diff-mode' and C-c C-c (or <RET>) is for.  So once
> again, I fail to see what use goto-line has for other then in really
> awkward situations.

I have a keyboard shortcut which opens Emacs with a Dired buffer in
the directory I am usually editing and I find it more efficient to
open the file in question from there and just jump to the line in
question with `M-g' than to activate diff-mode for a patch I got via
email and having to type in the path to the file the patch applies to.
You could argue that such things don't happen often enough to warrant
a special key binding, but I find it very convenient having it.

-- 
Ralf

Re: Key binding M-g should really be goto-line

[David Kastrup]

Ralf Angeli <angeli@iwi.uni-sb.de> writes:

> * Alfred M. Szmidt (2005-03-02) writes:
>
>> FWIW, I fail to see why people want goto-line, I never used it,
>> facemenu-keymap atleast has a potential use already, and in the
>> future; whereas goto-line is just a crutch.  And I fail to see the
>> fetish people have over keybidnings, specially M-g.
>
> One thing I did very soon after starting to use Emacs was to bind
> `M-g' to `goto-line'.  I am using it often when dealing with patches
> generated with `diff' and wanting to look at the original source
> before applying a patch.

Pressing C-c C-c in diff-mode should be easier.  Whenever this topic
has come up, one of the arguments was "whatever you are trying to do,
Emacs should have a better way to do it than M-g".  And in many cases
this might be correct.  But I don't think that the best way to teach
about refined features is to not provide simpler features.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Robert J. Chassell]

In today's GNU Emacs CVS snapshot, Wed, 2005 Mar  2  11:50 UTC
GNU Emacs 22.0.50.13 (i686-pc-linux-gnu, GTK+ Version 2.6.2)
started with

    /usr/local/src/emacs/src/emacs -Q


M-o is undefined;

C-x C-g is undefined;

C-x g runs the command insert-register-compatibility-binding.



Alfred M. Szmidt wrote

    FWIW, I fail to see why people want goto-line, I never used it, ...

This is yet another example of how people are different.  I use
goto-line frequently and cannot see working without it.  Fifteen or
eighteen years ago I bound it to C-c C-g and have used that binding
ever since.

There are a great many different actions people want to do, and not
that many keybindings.

This suggests that if goto-line is bound to a key chord by default, it
perhaps should be bound to a potentially less used key chord than to a
potentially more used key chord.

--
    Robert J. Chassell
    bob@rattlesnake.com                         GnuPG Key ID: 004B4AC8
    http://www.rattlesnake.com                  http://www.teak.cc

Re: Key binding M-g should really be goto-line autolearn=no version=3.0.2

[Jari Aalto+mail.linux]

| Since Richard really wants facemenu-keymap as a keybinding, what about
| making a compromise?  Bind goto-line to M-o, or bind facemenu-keymap
| to M-o and goto-line to M-g?  Or maybe modify a existing keybinding so
| that C-u ... does goto-line (maybe M-g is a candidate for this?)? Or
| bind goto-line to something like C-x ~.

I'd welcome gladly:

M-o   for facemenu-keymap  
M-g   for goto-line


| FWIW, I fail to see why people want goto-line, I never used it,
| facemenu-keymap atleast has a potential use already, and in the
| future; whereas goto-line is just a crutch.  And I fail to see the
| fetish people have over keybidnings, specially M-g.

It's not fetish. It's practical. As I showed. Add to it doing remote
compilation on site (A) when you edit the source at (B) and both sites
share the code through version control. (So you update B; refresh A and;
recompile at A)

Or try mixing Win32 environment with other hosts where not all the power of
Linux/Unix's remote handling is available. 

Line numbers coming from different sources like after cron job, where
site's links have been checked and you edit the files in another host.

But the typical situation is Web programming like PHP and others, where
line number is crucial in order to reach the error. 

It's just everyday work. All cannot be done from within Emacs; only handful
of people might be able to do that; the majority of users certainly consider
goto-line more useful than facemenu-keymap.

Jari

Re: Key binding M-g should really be goto-line

[Luc Teirlinck]

Jari Aalto wrote:

       "I want Emacs to move in the direction of doing word processing.  It
       may take years, but we will get there.  Then commands to specify faces
       will become important, and will need a good key binding.

       I chose the M-g binding for that reason, and the reason continues to
       have force.  So I don't intend to change that binding."

   Please, I have been watching this future over 10 years now

I personally do not have strong opinions on the goto-line binding.  I
personally use `M-x g-l RET' (with partial completion mode).

In as far as the current state of Emacs' word processing capabilities
are concerned:

I am personally not a "word processor guy", but while proofreading the
Emacs manual I took a close look at Enriched mode.

Enriched mode used to have bugs that nearly made it unusable, except
for very basic stuff.  I recently checked and corrected the
documentation and corrected bugs I found.  I believe that Enriched
mode currently largely "works" in the sense that it pretty much
correctly does what it claims to be able to do.  _But_ it only
supports the text/enriched format _and_ it does so in a way conforming
to RFC 1563, which has been obsolete since 1996.  I believe that
text/enriched should be updated for RFC 1896.  (Assuming that this is
still the most up to date standard.  It was when I last checked.)
Importantly, other formats should be supported.

Last time we discussed this, at least one person seemed to interested
in supporting additional formats.  If such plans would actually
materialize (of course, there is no guarantee of that until it
actually happens), it could bring the future of "Emacs as a word
processor" a lot closer.  Updating for RFC 1896 would seem to be less
complex and require less work than supporting new formats.

Sincerely,

Luc.

Re: Key binding M-g should really be goto-line

[Richard Stallman]

      I believe that
    text/enriched should be updated for RFC 1896.  (Assuming that this is
    still the most up to date standard.  It was when I last checked.)
    Importantly, other formats should be supported.

I would very much appreciate that work.

Re: Key binding M-g should really be goto-line

[Kai Großjohann]

Once again, I suggest C-x g as a binding for goto-line.  The old C-x g
binding is still available as C-x r g, if I'm not mistaken.

Kai

Re: Key binding M-g should really be goto-line

[Jari Aalto+mail.emacs]

| The following message is a courtesy copy of an article
| that has been posted to gmane.emacs.devel as well.
| 
| Once again, I suggest C-x g as a binding for goto-line.  The old C-x g
| binding is still available as C-x r g, if I'm not mistaken.

goto-line is so often used (when you need it; like during PHP coding), that
putting it behind two key presses is not practical. People would need to
map it to some fast key and we'd back where we started.

Jari

Re: Key binding M-g should really be goto-line

[Alfred M. Szmidt]

   goto-line is so often used (when you need it; like during PHP
   coding)

You have only noted PHP as needing goto-line, this doesn't warrant
binding it globally.  What about binding M-g in php-mode to goto-line?

Re: Key binding M-g should really be goto-line

[Josh Varner]

On Wed, 02 Mar 2005 14:52:07 +0100, Alfred M. Szmidt <ams@kemisten.nu> wrote:
>    goto-line is so often used (when you need it; like during PHP
>    coding)
> 
> You have only noted PHP as needing goto-line, this doesn't warrant
> binding it globally.  What about binding M-g in php-mode to goto-line?
> 
When the more advanced context sensitive goto-line like behaviors
break down having the primitive easily accessible is incredibly
helpful. For example on my system gdb mode has problems finding the
right header files when stepping through code, so I have to go to
those lines manually.

It would also be informative to take a look at some of the
distributions out there. Redhat's default .emacs binds goto-line to
Ctrl-x g. Given that this is RH 9.0, but I would not be surprised if
this is still in the default for latest Fedora. But how often do you
or did you receive complaints about losing
insert-register-compatibility-binding.

Josh

Re: Key binding M-g should really be goto-line

[Alfred M. Szmidt]

   For example on my system gdb mode has problems finding the right
   header files when stepping through code, so I have to go to those
   lines manually.

Then this is a bug in gdb-mode, not the fault of a missing keybinding.

Could people stop arguing why a keybinding is needed for goto-line?  I
think everyone already agrees that people see a need for it; for
whatever reason they may have.

M-g is a bad choice for goto-line, since it isn't used that often (and
please don't come up with reaons for facemenu-keymap).

C-x g is also bad, it has a already existing keybinding since a long
time.

C-u C-u C-l, C-x C-g, C-s NN, all seem like good candidates.  Can some
Emacs developer just pick one, and define it to goto-line and kill
this stupid thread?

   But how often do you or did you receive complaints about losing
   insert-register-compatibility-binding.

I will raise a stink about it.

Re: Key binding M-g should really be goto-line

[Reiner Steib]

On Wed, Mar 02 2005, Alfred M. Szmidt wrote:

> C-x g is also bad, it has a already existing keybinding since a long
> time.
>
> C-u C-u C-l, C-x C-g, C-s NN, all seem like good candidates.  Can some
> Emacs developer just pick one, and define it to goto-line and kill
> this stupid thread?

IMHO, `C-x C-g' is a bad choice, because it is supposed to abort (C-g)
an unintended `C-x'.  I'd vote for `C-x g' or `C-M-g'.

Bye, Reiner.                       ...who has bound <f5> to `goto-line'.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: Key binding M-g should really be goto-line

[Alfred M. Szmidt]

   > C-x g is also bad, it has a already existing keybinding since a
   > long time.
   >
   > C-u C-u C-l, C-x C-g, C-s NN, all seem like good candidates.  Can
   > some Emacs developer just pick one, and define it to goto-line
   > and kill this stupid thread?

   IMHO, `C-x C-g' is a bad choice, because it is supposed to abort
   (C-g) an unintended `C-x'.  I'd vote for `C-x g' or `C-M-g'.

I oppose to C-x g strongly, as for C-x C-g, I think you are wrong
there.  C-x C-g produces "C-x C-g is undefined", ditto for things like
"C-x v C-g".  And then, you always have ESC ESC ESC to the rescue. :-)

Happy hacking!

Re: Key binding M-g should really be goto-line

[Johan Bockgård]

Reiner Steib <reinersteib+gmane@imap.cc> writes:

> I'd vote for `C-x g' or `C-M-g'.

C-M-g may not work in a terminal/console.

-- 
Johan Bockgård

Re: Key binding M-g should really be goto-line

[Gaetan Leurent]

Johan Bockgård wrote on 02 Mar 2005 20:01:27 +0100:

>> I'd vote for `C-x g' or `C-M-g'.
>
> C-M-g may not work in a terminal/console.

We already have C-M-% that doesn't work in a terminal.

If you need to use one of those functions in a terminal, you just define
them on some other key and that's all.

I think `C-x g´ or `C-M-g´ is a good choice for the default binding, and
anyway people will just redefine it somewhere if they want to (I have it
on M-g and I think I'll keep that binding).

-- 
Gaëtan LEURENT

Re: Key binding M-g should really be goto-line

[Miles Bader]

On Wed, 02 Mar 2005 21:23:50 +0100, Gaetan Leurent
<gaetan.leurent@ens.fr> wrote:
> I think `C-x g´ or `C-M-g´ is a good choice for the default binding, and
> anyway people will just redefine it somewhere if they want to (I have it
> on M-g and I think I'll keep that binding).

C-x bindings are much easier to type if the following key is also a
control character though.

-MIles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line

[Alfred M. Szmidt]

   C-M-g may not work in a terminal/console.

<escape> C-g should work, which is the same.

Re: Key binding M-g should really be goto-line

[Jari Aalto]

|    goto-line is so often used (when you need it; like during PHP
|    coding)
| 
| You have only noted PHP as needing goto-line, this doesn't warrant
| binding it globally.  What about binding M-g in php-mode to goto-line?

I presented several examples. Please refer to other posts. The sources for
need for line numbers are so diverse, that one can imagine lot of uses.

Let's all mediate a while for several programming languages, automated
tools, shell scripts, syntax checkers, error messages ...

Jari

Re: Key binding M-g should really be goto-line

[Richard Stallman]

    Is there no hope to see the 'facemenu-keymap moved to another key and have
    the goto-line solely for M-g?

What other key do you suggest?  M-o would be ok, if people want.

      If we go that lane, it will more probably
    require opening a modifier like "super" or "alt" for it.  Most
    keyboards by now readily have enough modifiers available.

I don't think that is true.  Most keyboards that I have seen have only
CTRL, META and SHIFT.

Re: Key binding M-g should really be goto-line

[Jari Aalto]

|     Is there no hope to see the 'facemenu-keymap moved to another key and have
|     the goto-line solely for M-g?
| 
| What other key do you suggest?  M-o would be ok, if people want.

M-o is fine and looks well accessible.
 
Jari

Re: Key binding M-g should really be goto-line

[Juri Linkov]

I see `goto-line' finally bound to M-g in CVS.  Good news!
After such a big change, it is time now for a few improvements.

1. Before this change, font-lock-fontify-block was bound to `M-g M-g'.
Now it is bound to `M-o M-g'.  I suspect that the reason to bind it
to `M-g M-g' was to make it easier to run this frequent command
by typing the same key twice.  If this is true then now perhaps it
should be rebound to `M-o M-o'.

2. goto-line is not too frequent command to deserve the sole
M-g key.  There are many other goto-related commands that could
share the same mnemonics and have the common M-g prefix key.
Since goto-line is still the most frequent among them, it could
have also the most easily typed key binding `M-g M-g'.

Other commands that deserve a key binding with a `M-g' prefix are:

2.1. next-error (alias goto-next-locus), previous-error

The current key binding C-x ` is too hard to type on
many keyboards, and it is the constant cause of complaints.
The prefix key M-g will allow to add natural key bindings
for these commands and to reduce complaints in future.

2.2. dired-goto-file

This command is used to move point to the specified file name in
dired buffers.  It would be useful to run this command not only
in dired buffers, but in any buffer, with a new key binding.

2.3. goto-char is much less used command than goto-line,
but sometimes it's needed and could have a key binding too.

2.4. The M-g prefix key will also reserve a space for adding
other goto-related commands later.  Similarly to facemenu-keymap
which currently displays the message after typing M-o:

Set face: default, bold, italic, l = bold-italic, underline, Other...

M-g could display a message:

Go to: line, M-g = line, char, file, next error, prev error, Other...

and have a keymap with the following key bindings:

M-g M-g   - goto-line
M-g l     - goto-line
M-g c     - goto-char
M-g f     - dired-goto-file
M-g n     - next-error (goto-next-locus)
M-g M-n   - next-error (goto-next-locus)
M-g p     - previous-error
M-g M-p   - previous-error

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: Key binding M-g should really be goto-line

[Miles Bader]

On Wed, 09 Mar 2005 09:05:33 +0200, Juri Linkov <juri@jurta.org> wrote:
> 2. goto-line is not too frequent command to deserve the sole
> M-g key.  There are many other goto-related commands that could
> share the same mnemonics and have the common M-g prefix key.

Hmmm, not a bad idea.  I especially like the fact that next-error &c
could have a less annoying binding that C-x ` (also completely
unintuitive).

> Since goto-line is still the most frequent among them, it could
> have also the most easily typed key binding `M-g M-g'.

It would be good to also have a plain "g" binding for goto-line, for
those people that use ESC instead of a real meta key -- it's much
easier to hit `ESC g g' than  `ESC g ESC g' (you've already done this
for the other "meta" commands in the keymap).

[I suppose it's people that use ESC that might object most strenuously
to the prefix-key idea.]

-Miles

p.s. I have M-g bound to `fill-paragraph' myself... :-)
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line

[Kim F. Storm]

Juri Linkov <juri@jurta.org> writes:

> 2. goto-line is not too frequent command to deserve the sole
> M-g key.  There are many other goto-related commands that could
> share the same mnemonics and have the common M-g prefix key.

Brilliant ideas!!  I second ALL of your suggestions.

One command I use quite often is "find-file-and-line-near-point" which
takes something like FILE:LINE "near" the cursor and jumps to LINE in
FILE.  

I use this when I look at backtraces in bug-reports (of course, I
already have code which can do this, but a logical binding would be
nice).

M-g j would be an excellent binding for this.

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Key binding M-g should really be goto-line

[Juri Linkov]

storm@cua.dk (Kim F. Storm) writes:
> One command I use quite often is "find-file-and-line-near-point"
> which takes something like FILE:LINE "near" the cursor and jumps to
> LINE in FILE.

This is what `goto-line' is supposed to do.  Since it was already
completely modified recently with the complex logic of getting
a line number from the buffer and jumping to the most recently
displayed other buffer, why not modify it further to get a file
name from the buffer (before colon and the line number) and to
visit it before moving to the specified line number.

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: Key binding M-g should really be goto-line

[David Kastrup]

Juri Linkov <juri@jurta.org> writes:

> I see `goto-line' finally bound to M-g in CVS.  Good news!  After
> such a big change, it is time now for a few improvements.

Can you spell "Can-Of-Worms"?  Can you spell "feature freeze"?

> 1. Before this change, font-lock-fontify-block was bound to `M-g
> M-g'.  Now it is bound to `M-o M-g'.  I suspect that the reason to
> bind it to `M-g M-g' was to make it easier to run this frequent
> command by typing the same key twice.  If this is true then now
> perhaps it should be rebound to `M-o M-o'.

So far, I can agree.

> 2. goto-line is not too frequent command to deserve the sole
> M-g key.  There are many other goto-related commands that could
> share the same mnemonics and have the common M-g prefix key.

Forget it.  No precedence, no previous desire, not fitting the feature
freeze.  Is this a plot to distract people from releasing?

> Other commands that deserve a key binding with a `M-g' prefix are:
>
> 2.1. next-error (alias goto-next-locus), previous-error
>
> The current key binding C-x ` is too hard to type on
> many keyboards, and it is the constant cause of complaints.

C-x ` completely sucks as a keybinding on many international
keyboards.  We need a new keybinding alternative at some point of
time.  But M-g is clean out, in my opinion.  Really.

My proposal for an C-x ` alternative would be C-x ? which appears to
be free at the moment.  ? is a frequent character in all languages I
know, and so it should be more accessible than ` on most keyboards.
It also is connotated with "error" somewhat.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Miles Bader]

> My proposal for an C-x ` alternative would be C-x ? which appears to
> be free at the moment.  ? is a frequent character in all languages I
> know, and so it should be more accessible than ` on most keyboards.
> It also is connotated with "error" somewhat.

That's really no better than C-x ` so what's the point (the "also
connotated with error" thing is so tenuous as to be worthless)? 
Either we should change it to something good, or just leave it as is
until the next release.

[My personal binding for next-error is `C-x C-n', which just
completely rocks; but Juri's suggestion is pretty good too.]

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line

[David Kastrup]

Miles Bader <snogglethorpe@gmail.com> writes:

>> My proposal for an C-x ` alternative would be C-x ? which appears to
>> be free at the moment.  ? is a frequent character in all languages I
>> know, and so it should be more accessible than ` on most keyboards.
>> It also is connotated with "error" somewhat.
>
> That's really no better than C-x ` so what's the point (the "also
> connotated with error" thing is so tenuous as to be worthless)?

As a user of an American keyboard, you would not understand.  On
almost all keyboards of the world, getting ` is at best tedious, at
worst impossible or at least error-prone, usually yielding an accent
instead of a backquote.  For example, on a German keyboard you get `
by typing Shift-= SPC (where = means the American position for =), so
the whole sequence becomes C-x Shift-= SPC.

On a French keyboard, it is Right-Alt-7.  And so on.

> Either we should change it to something good, or just leave it as is
> until the next release.

For most languages of the world, ` is one heck of a nuisance.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Kim F. Storm]

David Kastrup <dak@gnu.org> writes:

> Juri Linkov <juri@jurta.org> writes:
>
>> 2. goto-line is not too frequent command to deserve the sole
>> M-g key.  There are many other goto-related commands that could
>> share the same mnemonics and have the common M-g prefix key.
>
> Forget it.  No precedence, 

Huh?  M-g was a prefix key before the change...

>                            no previous desire, not fitting the feature
> freeze.  Is this a plot to distract people from releasing?

IMO, it is a reasonable opportunity to DTRT before locking M-g down as
a single command-key only.

> C-x ` completely sucks as a keybinding on many international
> keyboards.  We need a new keybinding alternative at some point of
> time.  But M-g is clean out, in my opinion.  Really.

Why?  I think it is excellent.  Good mnemonic.

> My proposal for an C-x ` alternative would be C-x ? which appears to
> be free at the moment.  ? is a frequent character in all languages I
> know, and so it should be more accessible than ` on most keyboards.

That is C-x S-? on my keyboard which is still cumbersome for
something you type very frequently.

M-g-n (hold M while pressing g + n) is just so much better, and
allows you to go back as well M-g-p ...

And it is also easy to repeat M-g-n-g-n-g-p (ups, got one too far).

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Key binding M-g should really be goto-line

[Piet van Oostrum]

>>>>> storm@cua.dk (Kim F. Storm) (KFS) wrote:

>KFS> David Kastrup <dak@gnu.org> writes:
>>> Juri Linkov <juri@jurta.org> writes:
>>> 
>>>> 2. goto-line is not too frequent command to deserve the sole
>>>> M-g key.  There are many other goto-related commands that could
>>>> share the same mnemonics and have the common M-g prefix key.
>>> 
>>> Forget it.  No precedence, 

>KFS> Huh?  M-g was a prefix key before the change...

Was it? In gnus it is bound to  gnus-summary-rescan-group. At least in the
version a month or so ago.
-- 
Piet van Oostrum <piet@cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: piet@vanoostrum.org

Re: Key binding M-g should really be goto-line

[Miles Bader]

On Thu, 10 Mar 2005 10:18:49 +0100, Piet van Oostrum <piet@cs.uu.nl> wrote:
> >KFS> Huh?  M-g was a prefix key before the change...
> 
> Was it? In gnus it is bound to  gnus-summary-rescan-group. At least in the
> version a month or so ago.

Yup.  You're allowed to bind over prefix keys... :-)

Of course the usefulness of Gnus' local binding will have to be
re-evaluated with the new global binding in mind (the old global
binding of M-g was not at all useful for Gnus).

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line

[Kim F. Storm]

Piet van Oostrum <piet@cs.uu.nl> writes:

>>>>>> storm@cua.dk (Kim F. Storm) (KFS) wrote:
>
>>KFS> David Kastrup <dak@gnu.org> writes:
>>>> Juri Linkov <juri@jurta.org> writes:
>>>> 
>>>>> 2. goto-line is not too frequent command to deserve the sole
>>>>> M-g key.  There are many other goto-related commands that could
>>>>> share the same mnemonics and have the common M-g prefix key.
>>>> 
>>>> Forget it.  No precedence, 
>
>>KFS> Huh?  M-g was a prefix key before the change...
>
> Was it? In gnus it is bound to  gnus-summary-rescan-group. At least in the
> version a month or so ago.

I'm talking about the global binding, not some buffer local binding.

IMO, the gnus M-g binding should definitely be removed in the
*Article* buffer, as is makes good sense to use goto-line to another
buffer based on a file:linenumber found in a mail message (e.g. a
backtrace in a bug report).

The M-g binding can remain in gnus summary buffers if people prefer...

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Key binding M-g should really be goto-line

[Richard Stallman]

I think I will move the goto-line binding to M-g M-g,
making it a prefix key, so that in the future we can
put on other commands.

Re: Key binding M-g should really be goto-line

[Romain Francoise]

Juri Linkov <juri@jurta.org> writes:

> M-g could display a message:

> Go to: line, M-g = line, char, file, next error, prev error, Other...

> and have a keymap with the following key bindings:

> M-g M-g   - goto-line
> M-g l     - goto-line
> M-g c     - goto-char
> M-g f     - dired-goto-file
> M-g n     - next-error (goto-next-locus)
> M-g M-n   - next-error (goto-next-locus)
> M-g p     - previous-error
> M-g M-p   - previous-error

This is a very good idea.  However, Dired users are probably used to M-g
not being a prefix key since it's been bound to dired-goto-file in
Dired-X for as long as I can remember (more than 11 years, says CVS).

That being said, I'm very much in favor of this change--especially the
bindings you suggest for next-error and previous-error.

-- 
Romain Francoise <romain@orebokech.com> | With your feet in the air and
it's a miracle -- http://orebokech.com/ | your head on the ground.

Re: Key binding M-g should really be goto-line

[David Kastrup]

Romain Francoise <romain@orebokech.com> writes:

> Juri Linkov <juri@jurta.org> writes:
>
>> M-g could display a message:
>
>> Go to: line, M-g = line, char, file, next error, prev error, Other...
>
>> and have a keymap with the following key bindings:
>
>> M-g M-g   - goto-line
>> M-g l     - goto-line
>> M-g c     - goto-char
>> M-g f     - dired-goto-file
>> M-g n     - next-error (goto-next-locus)
>> M-g M-n   - next-error (goto-next-locus)
>> M-g p     - previous-error
>> M-g M-p   - previous-error
>
> This is a very good idea.

I am afraid that I have a hard time not to agree after thinking it
over.

> However, Dired users are probably used to M-g not being a prefix key
> since it's been bound to dired-goto-file in Dired-X for as long as I
> can remember (more than 11 years, says CVS).

Well, one could put M-g M-g on that in dired mode, but have M-g M-g
mean goto-line in most other modes.

It seems absurd to have dired-goto-file as a global binding.

> That being said, I'm very much in favor of this change--especially
> the bindings you suggest for next-error and previous-error.

Well, yes.  It makes just too much sense to throw the idea out the
door.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Miles Bader]

On Wed, 09 Mar 2005 23:34:54 +0100, David Kastrup <dak@gnu.org> wrote:
> It seems absurd to have dired-goto-file as a global binding.

Actually I often want to jump to a file's dired entry while visiting
the file.  It's not like it's all _that_ hard to do now (just do `C-x
C-f RET' to visit the cur directory, and look for the file), but the
concept is not completely absurd.

-Miles
-- 
Do not taunt Happy Fun Ball.

Re: Key binding M-g should really be goto-line

[David Kastrup]

Miles Bader <snogglethorpe@gmail.com> writes:

> On Wed, 09 Mar 2005 23:34:54 +0100, David Kastrup <dak@gnu.org> wrote:
>> It seems absurd to have dired-goto-file as a global binding.
>
> Actually I often want to jump to a file's dired entry while visiting
> the file.  It's not like it's all _that_ hard to do now (just do `C-x
> C-f RET' to visit the cur directory, and look for the file), but the
> concept is not completely absurd.

Granted.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Johan Bockgård]

Miles Bader <snogglethorpe@gmail.com> writes:

> Actually I often want to jump to a file's dired entry while visiting
> the file. It's not like it's all _that_ hard to do now (just do `C-x
> C-f RET' to visit the cur directory, and look for the file), but the
> concept is not completely absurd.

Indeed not. It's called `dired-jump' (bound to C-x C-j when dired-x is
loaded).

-- 
Johan Bockgård

Re: Key binding M-g should really be goto-line

[Juri Linkov]

bojohan+news@dd.chalmers.se (Johan Bockgård) writes:
> Indeed not. It's called `dired-jump' (bound to C-x C-j when dired-x
> is loaded).

It is good as a global binding, but it has one limitation: it can't jump
to an arbitrary file, it jumps only to buffer-file-name when called
not from a dired buffer.  It would be useful to move `dired-jump'
to dired.el and to modify it to ask the file name with the default
set to buffer-file-name (or to create a new similar function and
bind it to M-g f).

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: Key binding M-g should really be goto-line

[Romain Francoise]

Juri Linkov <juri@jurta.org> writes:

> It would be useful to move `dired-jump' to dired.el and to modify it
> to ask the file name with the default set to buffer-file-name (or to
> create a new similar function and bind it to M-g f).

I can second the suggestion to move `dired-jump' to dired.el, it's one
of the most useful commands in Dired.  But jumping to an unrelated file
doesn't sound that intuitive to me, and I definitely don't want to have
to confirm the command each time I use it.

-- 
Romain Francoise <romain@orebokech.com> | But you don't care for me and
it's a miracle -- http://orebokech.com/ | I don't care now for anybody.

Re: Key binding M-g should really be goto-line

[David Kastrup]

Romain Francoise <romain@orebokech.com> writes:

> Juri Linkov <juri@jurta.org> writes:
>
>> It would be useful to move `dired-jump' to dired.el and to modify it
>> to ask the file name with the default set to buffer-file-name (or to
>> create a new similar function and bind it to M-g f).
>
> I can second the suggestion to move `dired-jump' to dired.el, it's
> one of the most useful commands in Dired.  But jumping to an
> unrelated file doesn't sound that intuitive to me, and I definitely
> don't want to have to confirm the command each time I use it.

It would be somewhat consistent with typical Emacs applications if the
file name was prompted for only when a prefix argument was given.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Romain Francoise]

David Kastrup <dak@gnu.org> writes:

> It would be somewhat consistent with typical Emacs applications if the
> file name was prompted for only when a prefix argument was given.

Unfortunately, the prefix argument is already used to jump to the file
in another window.

-- 
Romain Francoise <romain@orebokech.com> | Why did you kill that poor
it's a miracle -- http://orebokech.com/ | old man, melody? She said,
                                        | "He was never good to me"

Re: Key binding M-g should really be goto-line

[Andreas Schwab]

Romain Francoise <romain@orebokech.com> writes:

> David Kastrup <dak@gnu.org> writes:
>
>> It would be somewhat consistent with typical Emacs applications if the
>> file name was prompted for only when a prefix argument was given.
>
> Unfortunately, the prefix argument is already used to jump to the file
> in another window.

The prompt could be triggered by a specific prefix argument like C-u 0
(aka M-0 aka C-0).

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Key binding M-g should really be goto-line

[Kim F. Storm]

Andreas Schwab <schwab@suse.de> writes:

> Romain Francoise <romain@orebokech.com> writes:
>
>> David Kastrup <dak@gnu.org> writes:
>>
>>> It would be somewhat consistent with typical Emacs applications if the
>>> file name was prompted for only when a prefix argument was given.
>>
>> Unfortunately, the prefix argument is already used to jump to the file
>> in another window.
>
> The prompt could be triggered by a specific prefix argument like C-u 0
> (aka M-0 aka C-0).

If M-g is a prefix, putting it on a different key, e.g. M-g j, is
much easier than messing with special prefix args.


-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Key binding M-g should really be goto-line

[Romain Francoise]

Juri Linkov <juri@jurta.org> writes:

> I see `goto-line' finally bound to M-g in CVS.  Good news!
> After such a big change, it is time now for a few improvements.

Now that Richard has changed M-g to be a prefix key, can we implement
this plan?

We already have:

M-g M-g   - goto-line
M-g g     - goto-line

The candidates are:

M-g c     - goto-char
M-g n     - next-error (goto-next-locus)
M-g M-n   - next-error (goto-next-locus)
M-g p     - previous-error
M-g M-p   - previous-error

Undecided:

M-g f     - dired-goto-file? (doesn't work in non-Dired buffers)
M-g j     - Kim's find-file-and-line-near-point?
M-g j     - dired-jump? (or a similar function which prompts for a file
            name)

-- 
Romain Francoise <romain@orebokech.com> | Why did you kill that poor
it's a miracle -- http://orebokech.com/ | old man, melody? She said,
                                        | "He was never good to me"

Re: Key binding M-g should really be goto-line

[Richard Stallman]

    Now that Richard has changed M-g to be a prefix key, can we implement
    this plan?

I would rather put this aside until after the release.
I would like a small group to organize to make a plan
for what to do with M-g, and discuss the question outside
of this list.  Would those who want to do this please do so?

Re: Key binding M-g should really be goto-line

[David Kastrup]

Richard Stallman <rms@gnu.org> writes:

>     Now that Richard has changed M-g to be a prefix key, can we
>     implement this plan?
>
> I would rather put this aside until after the release.

Discussing the full population of this binding, I might agree here.
However, part of this proposal was due to C-x ` being very burdensome
to type on many international keyboards.  I don't think it makes sense
to introduce M-g as a prefix without populating at least those slots
which have made this proposal compelling.

So I think that the next release should at the very least contain the
(rather uncontentious) M-g M-p, M-g p, M-g M-n and M-g n bindings.
All the rest is nice in various degrees, but those seem rather
essential.

> I would like a small group to organize to make a plan
> for what to do with M-g, and discuss the question outside
> of this list.  Would those who want to do this please do so?

I'd be willing to participate in such an off-list discussion if it
proves necessary or beneficial.  Let's raise the bar by demanding that
anybody discussing it should also be willing to implement his
proposals including documentation.  Since we are talking about
functionality that seems quite convenient to the average user, this
includes the help sheets which probably should be checked for accuracy
before release, anyhow, requiring speakers of different languages.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Key binding M-g should really be goto-line

[Romain Francoise]

Richard Stallman <rms@gnu.org> writes:

> I would rather put this aside until after the release.

I was hoping we could at least add the "candidates" from my list before
the release.  New bindings for previous-error and next-error would be
immediately useful and everyone involved in the discussion so far seemed
to agree.

> I would like a small group to organize to make a plan for what to do
> with M-g, and discuss the question outside of this list.  Would those
> who want to do this please do so?

We can just postpone this discussion to after the release.

-- 
Romain Francoise <romain@orebokech.com> | How long will it be before he
it's a miracle -- http://orebokech.com/ | sees you own his legs but his
                                        | mind is free?

Re: Key binding M-g should really be goto-line

[Richard Stallman]

    > I would like a small group to organize to make a plan for what to do
    > with M-g, and discuss the question outside of this list.  Would those
    > who want to do this please do so?

    We can just postpone this discussion to after the release.

We could, but that's not quite the issue.  I'm not talking about when
to have this discussion, I'm talking about where.  Having it on this
list would be distracting, so I would like interested people to
discuss the question outside this list.

We could do that later instead of now, if people remember the decision
to discuss this off the list.