a simple convenience function

a simple convenience function

[Paul Pogonyshev]

I'm not sure this issue wasn't raised already, but what do
you think about adding this simple function to Emasc?  It
could just sit there for optional binding through `~/.emacs'
by those who like this behaviour.  Or we could make behaviour
of the Home key customizeable.

Quick explanation: in the line

	  some t[>>point<<]ext here

after first evaluation of proposed function we get

	  [>>point<<]some text here

and after second evaluation we get

[>>point<<]	  some text here


Paul


--- indent.el	24 Oct 2004 23:58:43 +0300	1.59
+++ indent.el	14 Nov 2004 23:28:35 +0200	
@@ -329,6 +329,21 @@ line, but does not move past any whitesp
   (if (memq (current-justification) '(center right))
       (skip-chars-forward " \t")))
 
+(defun beginning-of-line-smart (&optional n)
+  "Move to the beginning of the text on this line, or to the beginning of the line.
+More exactly, if the point is already at the beginning of the
+line's text (as defined by `beginning-of-line-text'), it is
+placed at the very beginning of the line.  Otherwise it is moved
+to the text beginning.
+
+With optional argument, different from 1, behave identically to
+`beginning-of-line-text'."
+  (interactive "p")
+  (let ((current-point (point)))
+    (beginning-of-line-text n)
+    (when (= (point) current-point)
+      (beginning-of-line))))
+
 (defvar indent-region-function nil
   "Short cut function to indent region using `indent-according-to-mode'.
 A value of nil means really run `indent-according-to-mode' on each line.")

Re: a simple convenience function

[Miles Bader]

On Sun, Nov 14, 2004 at 11:36:00PM +0200, Paul Pogonyshev wrote:
> I'm not sure this issue wasn't raised already, but what do
> you think about adding this simple function to Emasc? 

There's already `back-to-indentation', bound to M-m by default.

I don't think the repeat-functionality you described seems really worth
adding another function.

-Miles
-- 
"Whatever you do will be insignificant, but it is very important that
 you do it."  Mahatma Ghandi

Re: a simple convenience function

[Lennart Borgman]

Though I found the behaviour suggested by Paul much more mnemonic when bound
to the Home key. It is also found in some other editors.

- Lennart

PS: Maybe I should write it as

(defun home-or-back-to-indentation()
  (interactive)
  (if (bolp) (back-to-indentation) (beginning-of-line)))

(define-key global-map [home] 'home-or-back-to-indentation)


----- Original Message ----- 
From: "Miles Bader" <miles@gnu.org>
To: "Paul Pogonyshev" <pogonyshev@gmx.net>
Cc: <emacs-devel@gnu.org>
Sent: Monday, November 15, 2004 12:34 AM
Subject: Re: a simple convenience function


> On Sun, Nov 14, 2004 at 11:36:00PM +0200, Paul Pogonyshev wrote:
> > I'm not sure this issue wasn't raised already, but what do
> > you think about adding this simple function to Emasc?
>
> There's already `back-to-indentation', bound to M-m by default.
>
> I don't think the repeat-functionality you described seems really worth
> adding another function.
>
> -Miles
> -- 
> "Whatever you do will be insignificant, but it is very important that
>  you do it."  Mahatma Ghandi
>
>
> _______________________________________________
> Emacs-devel mailing list
> Emacs-devel@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-devel
>

Re: a simple convenience function

[Paul Pogonyshev]

> Though I found the behaviour suggested by Paul much more mnemonic when
> bound to the Home key. It is also found in some other editors.
>
> - Lennart
>
> PS: Maybe I should write it as
>
> (defun home-or-back-to-indentation()
>   (interactive)
>   (if (bolp) (back-to-indentation) (beginning-of-line)))
>
> (define-key global-map [home] 'home-or-back-to-indentation)

Perhaps, but this way you get it the way round: it will first jump
to the beginning of line and then to the beginning of the text.

Paul

Re: a simple convenience function

[Dhruva Krishnamurthy]

Hello,


On Mon, 15 Nov 2004 03:49:57 +0200, Paul Pogonyshev <pogonyshev@gmx.net> wrote:
> 
> 
> 
> > Though I found the behaviour suggested by Paul much more mnemonic when
> > bound to the Home key. It is also found in some other editors.
> >
> > - Lennart
> >
> > PS: Maybe I should write it as
> >
> > (defun home-or-back-to-indentation()
> >   (interactive)
> >   (if (bolp) (back-to-indentation) (beginning-of-line)))
> >
> > (define-key global-map [home] 'home-or-back-to-indentation)
> 
> Perhaps, but this way you get it the way round: it will first jump
> to the beginning of line and then to the beginning of the text.
> 

I have a slightly fancy stuff which I use: On first hit on HOME, it
finds whether it is closer to beginning of line or begining of text
and jumps to the closer. Same with end of line. An experienced elisp
geek could clean up significantly... I am learning to walk/crawl in
the elisp ground.

;;-----------------------------------------------------------------------------
;; Intelligent HOME to toggle between 1st text and ZERO column
;;-----------------------------------------------------------------------------
(defun intelli-home()
  "User method to toggle between 1st text and ZERO column"
  (interactive)
  (catch 'ret
    (let ((pos (point)) (blpos (line-beginning-position)) (btpos 0))

      ;; If at begin of line
      (if (= blpos pos)
	  (progn
	    (beginning-of-line-text)
	    (throw 'ret t)))

      (beginning-of-line-text)
      (setq btpos (point))

      ;; If at begin of text
      (if (= btpos pos)
	  (progn
	    (beginning-of-line)
	    (throw 'ret t)))

      ;; Cursor after first text
      (if (> pos btpos)
	  (throw 'ret t))

      ;; Cursor in between begin of line and first text
      ;; Closer to first text
      (if (> (- pos blpos) (- btpos pos))
	  (throw 'ret t))

      ;; Closer to begin line
      (if (or (< (- pos blpos) (- btpos pos))
	      (= (- pos blpos) (- btpos pos)))
	  (progn
	    (beginning-of-line)
	    (throw 'ret t))))))

;;-----------------------------------------------------------------------------
;; Move to end of line text (as beginning-of-line-text)
;;-----------------------------------------------------------------------------
(defun end-of-line-text()
  "User method to simulate beginning-of-line-text for end of line"
  (interactive)
  (end-of-line)
  (while (looking-at "[ \t\n]+")
    (backward-char 1)))

;;-----------------------------------------------------------------------------
;; Intelligent END to toggle between Last text and Last column
;;-----------------------------------------------------------------------------
(defun intelli-end()
  "User method to toggle between Last text and Last column"
  (interactive)
  (catch 'ret
    (let ((pos (point)) (elpos (line-end-position)) (etpos 0))

      ;; If at begin of line
      (if (= elpos pos)
	  (progn
	    (end-of-line-text)
	    (throw 'ret t)))

      (end-of-line)
      (setq etpos (point))

      ;; If at begin of text
      (if (= etpos pos)
	  (progn
	    (end-of-line)
	    (throw 'ret t)))

      ;; Cursor after first text
      (if (< pos etpos)
	  (throw 'ret t))

      ;; Cursor in between begin of line and first text
      ;; Closer to first text
      (if (< (- pos elpos) (- etpos pos))
	  (throw 'ret t))

      ;; Closer to begin line
      (if (or (> (- pos elpos) (- etpos pos))
	      (= (- pos elpos) (- etpos pos)))
	  (progn
	    (end-of-line)
	    (throw 'ret t))))))

;; Key bindings
(global-set-key [end]  'intelli-end)
(global-set-key [home] 'intelli-home)

- dhruva
-- 
Proud FSF member: #1935
http://schemer.fateback.com/

Re: a simple convenience function

[Lennart Borgman]

Yes, my intention was to not change the default behaviour for default usage
so to say, but I realize now that a little bit more is needed to achieve
that.

- Lennart


----- Original Message ----- 
From: "Paul Pogonyshev" <pogonyshev@gmx.net>
To: "Lennart Borgman" <lennart.borgman.073@student.lu.se>
Cc: "Miles Bader" <miles@gnu.org>; <emacs-devel@gnu.org>
Sent: Monday, November 15, 2004 2:49 AM
Subject: Re: a simple convenience function



> Though I found the behaviour suggested by Paul much more mnemonic when
> bound to the Home key. It is also found in some other editors.
>
> - Lennart
>
> PS: Maybe I should write it as
>
> (defun home-or-back-to-indentation()
>   (interactive)
>   (if (bolp) (back-to-indentation) (beginning-of-line)))
>
> (define-key global-map [home] 'home-or-back-to-indentation)

Perhaps, but this way you get it the way round: it will first jump
to the beginning of line and then to the beginning of the text.

Paul

Re: a simple convenience function

[Paul Pogonyshev]

> On Sun, Nov 14, 2004 at 11:36:00PM +0200, Paul Pogonyshev wrote:
> > I'm not sure this issue wasn't raised already, but what do
> > you think about adding this simple function to Emasc?
>
> There's already `back-to-indentation', bound to M-m by default.
>
> I don't think the repeat-functionality you described seems really worth
> adding another function.

I got used to it years ago when I was still working under Windows with
MS Visual Studio.  When I switched to Emacs, it was about the first
things I customized with Elisp.

Paul

Re: a simple convenience function

[Thien-Thi Nguyen]

Paul Pogonyshev <pogonyshev@gmx.net> writes:

   (defun beginning-of-line-smart (&optional n) ...)

see also codeline.el:

http://www.glug.org/people/ttn/software/ttn-pers-elisp/dist-lisp-index.html

it behaves similarly, and additionally in the other direction.

thi

Re: a simple convenience function

[Kai Grossjohann]

Paul Pogonyshev <pogonyshev@gmx.net> writes:

> +(defun beginning-of-line-smart (&optional n)
> +  "Move to the beginning of the text on this line, or to the beginning of the line.
> +More exactly, if the point is already at the beginning of the
> +line's text (as defined by `beginning-of-line-text'), it is
> +placed at the very beginning of the line.  Otherwise it is moved
> +to the text beginning.
> +
> +With optional argument, different from 1, behave identically to
> +`beginning-of-line-text'."

I used to have something similar, except that I changed the <home> key
to move to beginning of line, then beginning of window, then beginning
of buffer.  But the behavior of that function was different: where it
moved depended on how often you hit the <home> key.  When you first
hit it, it moved to beginning of line, if you then immediately press
it again, it moves to beginning of window, and so on.

I found this to be better because I knew I could always hit <home>
twice to end up at the beginning of the window, and I didn't have to
check whether point was already at the beginning of the line.

But for the proposed binding, it's more difficult: one would expect
C-a to move to the left, so its behavior needs to depend on the
position of point, I think.

What do others think about the tradeoff between the "never move right"
and the "make behavior predictable without looking at text"
constraints?

Kai

Re: a simple convenience function

[Alex Schroeder]

Kai Grossjohann <kai@emptydomain.de> writes:

> What do others think about the tradeoff between the "never move right"
> and the "make behavior predictable without looking at text"
> constraints?

I really like your predictable behavior.  I never thought about it
that way, but now that you put it this way, I think "dwim" in this
situation should be independent of where point is in the text.  It
should depend on what the user did, instead.

I'm not sure I like your "beginning of window" step, however, because
I usually hop through buffers in paragraphs, sexps, or defuns, or
whatever the text-based unit is.  Using a display-based unit is weird.

Alex.
-- 
.O.  http://www.emacswiki.org/alex/
..O  Schroeder's fifth law:
OOO  Never accept more work than you can handle in one night of hacking.

Re: a simple convenience function

[Lennart Borgman]

----- Original Message ----- 
From: "Alex Schroeder" <alex@emacswiki.org>


> Kai Grossjohann <kai@emptydomain.de> writes:
>
> > What do others think about the tradeoff between the "never move right"
> > and the "make behavior predictable without looking at text"
> > constraints?
>
> I really like your predictable behavior.

Agree. I would not be able to use it if it were not predictable. I type
first and look later...

- Lennart

Re: a simple convenience function

[Kai Grossjohann]

Alex Schroeder <alex@emacswiki.org> writes:

> I'm not sure I like your "beginning of window" step, however, because
> I usually hop through buffers in paragraphs, sexps, or defuns, or
> whatever the text-based unit is.  Using a display-based unit is weird.

I only explained what my function did.  After using it for a while, I
found out that I always hit it once or thrice, but never twice.

I also think that perhaps going to the beginning of the defun is a
good idea as an intermediary step.

Kai

Re: a simple convenience function

[Kai Grossjohann]

Alex Schroeder <alex@emacswiki.org> writes:

> Kai Grossjohann <kai@emptydomain.de> writes:
>
>> What do others think about the tradeoff between the "never move right"
>> and the "make behavior predictable without looking at text"
>> constraints?
>
> I really like your predictable behavior.  I never thought about it
> that way, but now that you put it this way, I think "dwim" in this
> situation should be independent of where point is in the text.  It
> should depend on what the user did, instead.

The only problem is that if you always move to the beginning of the
text on the first press, then C-a might move right.  And that is very
strange in left-to-right languages.

Gnus has dwimish C-a in message-mode: It goes to the beginning of the
header text, or to the beginning of the line, when in a message header.
The Gnus behavior is also to move right in some circumstances.  It's
strange.

Kai

Re: a simple convenience function

[Kai Grossjohann]

Alex Schroeder <alex@emacswiki.org> writes:

> I'm not sure I like your "beginning of window" step, however, because
> I usually hop through buffers in paragraphs, sexps, or defuns, or
> whatever the text-based unit is.  Using a display-based unit is weird.

Well, a line is a display-based unit, so having the C-a behavior
depend on the display seems logical.

But I also found out that I never used the beginning of window step.

Kai