fill length of visual line mode

fill length of visual line mode

[Ivan Kanis]

Hello,

I just enabled visual line mode today. I have started using it to
display gnus article. I noticed that the fill happens at the width of
the Emacs window. I set fill-column to 70 and I was expecting it to wrap
there.

There is a hack [1] by James Wright to make it work by expanding the
right margin of the buffer's window.

I don't feel it should be so complicated to customize this behavior.

Would it be OK to code wrap at fill-column?

Ivan

[1] The code on Emacs wiki:

(defvar visual-wrap-column nil)

(defun set-visual-wrap-column (new-wrap-column &optional buffer)
  "Force visual line wrap at NEW-WRAP-COLUMN in BUFFER (defaults
to current buffer) by setting the right-hand margin on every
window that displays BUFFER.  A value of NIL or 0 for
NEW-WRAP-COLUMN disables this behavior."
  (interactive (list (read-number "New visual wrap column, 0 to disable: "
                                  (or visual-wrap-column fill-column 0))))
  (if (and (numberp new-wrap-column)
           (zerop new-wrap-column))
    (setq new-wrap-column nil))
  (with-current-buffer (or buffer (current-buffer))
    (visual-line-mode t)
    (set (make-local-variable 'visual-wrap-column) new-wrap-column)
    (add-hook 'window-configuration-change-hook
              'update-visual-wrap-column nil t)
    (let ((windows (get-buffer-window-list)))
      (while windows
        (when (window-live-p (car windows))
          (with-selected-window (car windows)
            (update-visual-wrap-column)))
        (setq windows (cdr windows))))))

(defun update-visual-wrap-column ()
  (if (not visual-wrap-column)
    (set-window-margins nil nil)
    (let* ((current-margins (window-margins))
           (right-margin (or (cdr current-margins) 0))
           (current-width (window-width))
           (current-available (+ current-width right-margin)))
      (if (<= current-available visual-wrap-column)
        (set-window-margins nil (car current-margins))
        (set-window-margins nil (car current-margins)
                            (- current-available
  visual-wrap-column))))))

Re: fill length of visual line mode

[Eli Zaretskii]

> From: Ivan Kanis <ivan@kanis.fr>
> Date: Thu, 02 Oct 2014 09:48:12 +0200
> 
> I just enabled visual line mode today. I have started using it to
> display gnus article. I noticed that the fill happens at the width of
> the Emacs window. I set fill-column to 70 and I was expecting it to wrap
> there.

That's not how word-wrap was designed and implemented.  It works more
like Web browsers, i.e. it wraps at the window edge.

> There is a hack [1] by James Wright to make it work by expanding the
> right margin of the buffer's window.
> 
> I don't feel it should be so complicated to customize this behavior.
> 
> Would it be OK to code wrap at fill-column?

Not sure what you mean by your question.  Are you asking whether
patches to implement the wrap at fill-column will be accepted?  If so,
I'm quite sure the answer is YES.

Re: fill length of visual line mode

[Glenn Morris]

Ivan Kanis wrote:

> I just enabled visual line mode today. I have started using it to
> display gnus article. I noticed that the fill happens at the width of
> the Emacs window. I set fill-column to 70 and I was expecting it to wrap
> there.

This is http://debbugs.gnu.org/cgi/bugreport.cgi?bug=18589 .
(Presumably wrote a blog post or something and this is now flavour of
the week.)

> There is a hack [1] by James Wright to make it work by expanding the
> right margin of the buffer's window.

See comments at
http://lists.gnu.org/archive/html/emacs-devel/2013-12/msg00495.html .
    
    The code you quote seems mostly OK, but I don't know who wrote it, so
    I can't just install it, for copyright reasons.
    
    If someone can rewrite it and submit a patch, or if the original author
    can submit the patch, I'd be happy to install the feature.

Re: fill length of visual line mode

[Ivan Kanis]

October, 02 at 18:16 Eli wrote:

>> From: Ivan Kanis <ivan@kanis.fr>
>> Date: Thu, 02 Oct 2014 09:48:12 +0200
>> 
>> I just enabled visual line mode today. I have started using it to
>> display gnus article. I noticed that the fill happens at the width of
>> the Emacs window. I set fill-column to 70 and I was expecting it to wrap
>> there.
>
> That's not how word-wrap was designed and implemented.  It works more
> like Web browsers, i.e. it wraps at the window edge.

It's too long for me at work where I use a tiling window manager and
wide screen.

>> There is a hack [1] by James Wright to make it work by expanding the
>> right margin of the buffer's window.
>> 
>> I don't feel it should be so complicated to customize this behavior.
>> 
>> Would it be OK to code wrap at fill-column?
>
> Not sure what you mean by your question.  Are you asking whether
> patches to implement the wrap at fill-column will be accepted?  If so,
> I'm quite sure the answer is YES.

It is what I meant.

October, 02 at 11:50 Glenn wrote:

> The code you quote seems mostly OK, but I don't know who wrote it, so
> I can't just install it, for copyright reasons.
>     
> If someone can rewrite it and submit a patch, or if the original author
> can submit the patch, I'd be happy to install the feature.

It just boils down to setting the right margin. I have to hide the right
fringe or it looks wrong. FWIW here is how I cobbled for gnus article.

(defun ivan-gnus-hack-visual-line-length ()
  "Set visual line length by expanding the right margin of the
buffer’s window."
  (let ((window (selected-window)))
    (select-window (car (get-buffer-window-list gnus-article-buffer)))
    (set-window-margins nil 0 (- (window-width) 80))
    (set-window-fringes (selected-window) 8 0)
    (visual-line-mode)
    (select-window window)))

(add-hook 'gnus-article-prepare-hook 'ivan-gnus-hack-visual-line-length)

It just feels hackish and wrong. The right thing I feel is to use
fill-column.

Ivan
-- 
You must've hit the wrong any key.
    -- BOFH excuse #52

Re: fill length of visual line mode

[Johan Bockgård]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Ivan Kanis <ivan@kanis.fr>
>> Date: Thu, 02 Oct 2014 09:48:12 +0200
>> 
>> I just enabled visual line mode today. I have started using it to
>> display gnus article. I noticed that the fill happens at the width of
>> the Emacs window. I set fill-column to 70 and I was expecting it to wrap
>> there.
>
> That's not how word-wrap was designed and implemented.  It works more
> like Web browsers, i.e. it wraps at the window edge.

The original word wrap patch by Kim did support wrapping at a specified
column, but this feature was removed when the code was integrated into
Emacs.

Re: fill length of visual line mode

[Stefan Monnier]

> It just boils down to setting the right margin. I have to hide the right
> fringe or it looks wrong.

Better to swap the margin and the fringe with fringes-outside-margins.

> (defun ivan-gnus-hack-visual-line-length ()
>   "Set visual line length by expanding the right margin of the
> buffer’s window."
>   (let ((window (selected-window)))
>     (select-window (car (get-buffer-window-list gnus-article-buffer)))
>     (set-window-margins nil 0 (- (window-width) 80))
>     (set-window-fringes (selected-window) 8 0)
>     (visual-line-mode)
>     (select-window window)))

You want to use `with-selection-window' above.
Additionally, the above will lose its effect if you hide the window and
show it again: you'd need to do it via window-configuration-change-hook.
Other problem: if window-width is smaller than 80 (which I gather is
your target fill-column in the above code), you end up setting
a negative margin.

> It just feels hackish and wrong.  The right thing I feel is to use
> fill-column.

I'm not opposed to a patch that would let it obey fill-column
(optionally).  But it does beg the question: what if the window is
smaller than fill-column?  Should it then fill at the window's edge or
should it continue as if truncate-lines is set?


        Stefan

Re: fill length of visual line mode

[Eli Zaretskii]

> From: Johan Bockgård <bojohan@gnu.org>
> Date: Fri, 03 Oct 2014 23:19:39 +0200
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> >> From: Ivan Kanis <ivan@kanis.fr>
> >> Date: Thu, 02 Oct 2014 09:48:12 +0200
> >> 
> >> I just enabled visual line mode today. I have started using it to
> >> display gnus article. I noticed that the fill happens at the width of
> >> the Emacs window. I set fill-column to 70 and I was expecting it to wrap
> >> there.
> >
> > That's not how word-wrap was designed and implemented.  It works more
> > like Web browsers, i.e. it wraps at the window edge.
> 
> The original word wrap patch by Kim did support wrapping at a specified
> column, but this feature was removed when the code was integrated into
> Emacs.

If so, bringing that back should be easier, as the code to do that is
probably archived somewhere.

Re: fill length of visual line mode

[Harald Hanche-Olsen]

[Stefan Monnier <monnier@iro.umontreal.ca> (2014-10-03 21:20:07 UTC)]

> I'm not opposed to a patch that would let it obey fill-column
> (optionally).  But it does beg the question: what if the window is
> smaller than fill-column?  Should it then fill at the window's edge or
> should it continue as if truncate-lines is set?

Filling at the window's edge is the only reasonable option, IMO.

The behaviour of web browsers was mentioned earlier. But you have the
CSS property max-width to modify that. I have – and frequently use – a
bookmarklet that lets me set the max-width property on the body
element in order to improve readability. It beats adjusting the
browser window itself.

Perhaps emacs could have a variable max-fill-column which, when set,
could govern visual line mode in this way? If set to t, it would mean
use the value of fill-column. Or if set to a number, ignore
fill-column and use the number itself. It shouldn't be allowed to
override fill-column in the other use cases, though. Perhaps
visual-line-max-fill-column would be a better name, if perhaps a bit
unwieldy.

– Harald

Re: fill length of visual line mode

[Ivan Kanis]

October, 03 at 17:20 Stefan wrote:

Hi Stefan,

Let's forget about the code I posted. It's a stupid hack that works for
me.

>> It just feels hackish and wrong.  The right thing I feel is to use
>> fill-column.
>
> But it does beg the question: what if the window is smaller than
> fill-column? Should it then fill at the window's edge or should it
> continue as if truncate-lines is set?

I didn't think of that. I bet users will want either. So a variable
will say what to do.
-- 
Program load too heavy for processor to lift.
    -- BOFH excuse #30

RE: fill length of visual line mode

[Drew Adams]

> Perhaps emacs could have a variable max-fill-column which, when set,
> could govern visual line mode in this way? If set to t, it would
> mean use the value of fill-column. Or if set to a number, ignore
> fill-column and use the number itself. It shouldn't be allowed to
> override fill-column in the other use cases, though. Perhaps
> visual-line-max-fill-column would be a better name, if perhaps a bit
> unwieldy.

If it is specific to visual lines then yes of course, "visual-line"
(or similar) should be in the name.  Users should not be able to
get the impression, by looking only at the name, that this has
something to do with `fill-column' generally.

Re: fill length of visual line mode

[Johan Bockgård]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Johan Bockgård <bojohan@gnu.org>
>> Date: Fri, 03 Oct 2014 23:19:39 +0200
>> 
>> The original word wrap patch by Kim did support wrapping at a specified
>> column, but this feature was removed when the code was integrated into
>> Emacs.
>
> If so, bringing that back should be easier, as the code to do that is
> probably archived somewhere.

http://lists.gnu.org/archive/html/emacs-devel/2005-07/msg01112.html

http://lists.gnu.org/archive/html/emacs-devel/2008-06/msg01544.html

Re: fill length of visual line mode

[Stefan Monnier]

> Let's forget about the code I posted. It's a stupid hack that works for me.
>>> It just feels hackish and wrong.  The right thing I feel is to use
>>> fill-column.
>> But it does beg the question: what if the window is smaller than
>> fill-column? Should it then fill at the window's edge or should it
>> continue as if truncate-lines is set?
> I didn't think of that. I bet users will want either. So a variable
> will say what to do.

I think pursuing the window-configuration-change-hook path is actually
a good idea: if we can get the desired behavior via Elisp code, it's
generally preferable (the C code is hairy enough as it is).  It also
makes it easier to offer configurability (e.g. to choose whether to
center the visible text in the window, or not).


        Stefan

Re: fill length of visual line mode

[Ivan Kanis]

October, 04 at 18:25 Stefan wrote:

>> Let's forget about the code I posted. It's a stupid hack that works for me.
>>>> It just feels hackish and wrong.  The right thing I feel is to use
>>>> fill-column.
>>> But it does beg the question: what if the window is smaller than
>>> fill-column? Should it then fill at the window's edge or should it
>>> continue as if truncate-lines is set?
>> I didn't think of that. I bet users will want either. So a variable
>> will say what to do.
>
> I think pursuing the window-configuration-change-hook path is actually
> a good idea: if we can get the desired behavior via Elisp code, it's
> generally preferable (the C code is hairy enough as it is).  It also
> makes it easier to offer configurability (e.g. to choose whether to
> center the visible text in the window, or not).

OK. I won't do it straight away as the gnus hack is sufficient for me
for now. I am planning to use visual line mode for org in the future. If
I do it I will revisit your suggestions to write some elisp that can be
added to Emacs proper.

Ivan
-- 
Only two things are infinite, the universe and human stupidity, and
I'm not sure about the former.
    -- Albert Einstein