Re: [arensb@ooblick.com: Feature suggestion: split-line]

[storm@cua.dk (Kim F. Storm)]

"Ehud Karni" <ehud@unix.mvs.co.il> writes:

> On Thu, 02 Jan 2003 13:38:26 -0500, Richard Stallman <rms@gnu.org> wrote:
> > 
> >       > The quick 
> >       >           brown fox
> > instead of
> >       > The quick 
> >                   brown fox
> 
> Since `delete-indentation' (aka join-line) is deleting `fill-prefix',
> it is reasonable that `split-line' should add it. `delete-indentation'
> checks for the `fill-prefix' existence. I think `split-line' should
> check it too, i.e. Add IT ONLY IF the current line HAS IT.

I like that!

In message buffers, I'd rather like it to insert message-yank-prefix
with the same semantics!  But I guess a message-split-line would be
better for that; and [remap split-line] to it in the message-mode-map.


Something like this:

(defun message-split-line ()
  "Split current line, moving portion beyond point vertically down.
If fill-prefix is set and the current line has it, put it on the new line"
  (interactive "*")
  (skip-chars-forward " \t")
  (let ((col (current-column))
        (pos (point))
        (prfx (and message-yank-prefix
                   (string= message-yank-prefix
                            (buffer-substring (line-beginning-position)
                                              (+ (line-beginning-position)
                                                 (length message-yank-prefix)))))))
    (newline 1)
    (if prfx (insert-and-inherit message-yank-prefix))
    (indent-to col 0)
    (goto-char pos)))


> BTW. I would have changed the behavior of `open-line' to the same
>      logic, i.e. - Add fill-prefix if the current line has it (instead
>      of its current behavior - adding it only when you are at bolp).

I'd second that too.

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