[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Making windows have same number of columns

From: Stephen Berman
Subject: Re: Making windows have same number of columns
Date: Fri, 27 Nov 2020 16:37:38 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/28.0.50 (gnu/linux)

On Thu, 26 Nov 2020 23:05:08 +0100 Stephen Berman <> 

> On Thu, 26 Nov 2020 22:19:48 +0100 wrote:
>> A solution could be to make a defun that performs (text-scale-adjust -)
>> until (window-body-width) is less than a number N.
> Instead of changing the window width, one at first sight plausible idea
> is to check the column at the end the longest line in the buffer after
> each application of text-scale-adjust; unfortunately, this doesn't work,
> because using text-scale-adjust does not change how many columns a line
> takes up: e.g., if the longest line ends at column 82 and
> window-body-width is 72, then after apply text-scale-adjust until the
> longest line appears to be shorter than 72, it still ends at column 82.
> In other words, text-scale-adjust also adjusts the column width.  So
> that approach won't work.  I can't think off hand of a way to get what
> you want, but maybe someone else knows a way.

I did think of a way after all:

(defun srb-shrink-font ()
  "Shrink font to make longest line fit in window width."
    (let ((max 0) line)
      (goto-char (point-min))
      (while (not (eobp))
        (let* ((inhibit-field-text-motion t)
               (col (progn (end-of-line) (current-column))))
          (when (> col max)
            (setq max col)
            (setq line (line-beginning-position)))
      (goto-char line))
    (let ((eol (line-end-position))
          (eovl (lambda () (progn (end-of-visual-line) (point)))))
      (while (> eol (funcall eovl))
        (let ((text-scale-mode-step 1.05))
          (text-scale-decrease 1))))))

But I suspect there will be many cases where the result isn't really
satisfying.  You can experiment with text-scale-mode-step: in the above
I let-bound it to the smallest value that seemed to work reasonably
well, but sometimes it still seems to shrink the font too much and other
times not enough.  I think this has to do with the algorithm
text-scale-adjust uses to change the size, using exponentiation, which
means that sometimes the change will be too small for the available font
sizes and other times too large.  I also didn't implement the case of
increasing the font size, because the only way I could think of to do
that involves increasing too much and then decreasing again, which
probably is visually annoying.

Steve Berman

reply via email to

[Prev in Thread] Current Thread [Next in Thread]