bug-gnu-emacs
[Top][All Lists]
Advanced

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

bug#51715: defface forms not having dynamic value


From: irenezerafa
Subject: bug#51715: defface forms not having dynamic value
Date: Tue, 09 Nov 2021 14:17:03 +0000

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐

On Tuesday, November 9th, 2021 at 1:26 PM, Eli Zaretskii <eliz@gnu.org> wrote:

> > Date: Tue, 09 Nov 2021 11:41:10 +0000
> > From: irenezerafa via "Bug reports for GNU Emacs,
> > the Swiss army knife of text editors" bug-gnu-emacs@gnu.org
> >
> > I am getting into quite messy code because my macro creates
> > defface forms. Those do not have a dynamic value, but use the
> > value they get at the time they are evaluated. So using something
> > like a variable purple-intense does not help, because its value
> > would be a reference to one value, not a function that returns a
> > value (and having a function as a value would not be possible for
> > a face).
> >
> > Because the purpose of faces is to be configurable by the user,
> > could there be some improvements to easily tackle such problems?
>
> Please tell more. Which face attributes did you want to make dynamic,
> and how?

I take rainbow-delimiters as example.  Specifically, the

(eval-when-compile
  (defmacro rainbow-delimiters--define-depth-faces ()
    (let ( (faces '())
           (dark-colors  [ "#ff62d4"  "#3fdfd0"  "#fba849"  "#9f80ff"
            "#4fe42f"  "#fe6060"  "#4fafff"  "#f0dd60" "#ffffff" ])
           (light-colors [ "#a8007f"  "#005f88"  "#904200"  "#7f10d0"
            "#006800"  "#b60000"  "#1f1fce"  "#605b00" "#000000"]) )
      (dotimes (i 9)
        (push `( defface ,(intern (format "rainbow-delimiters-depth-%d-face" 
(1+ i)))
                 '( (default (:inherit rainbow-delimiters-base-face))
                    ( ((class color) (background dark))
                      :foreground ,(aref dark-colors  i))
                    ( ((class color) (background light))
                      :foreground ,(aref light-colors i)) )
                 ,(format "Nested delimiter face, depth %d." (1+ i))
                 :group 'rainbow-delimiters-faces )
              faces))
      `(progn ,@faces)) ))

(rainbow-delimiters--define-depth-faces)

Now, suppose I want to use a colour scheme from modus-themes.

(require 'modus-themes)

(eval-when-compile
  (defmacro rainbow-delimiters--define-depth-faces ()
    (let ( (faces '())
           (dark-colors (rainbow-delimiters-modus-vivendi-intense-colours
                         (list magenta-intense cyan-intense orange-intense
                               purple-intense green-intense red-intense
                               blue-intense yellow-intense fg-main)))
           (light-colors (rainbow-delimiters-modus-operandi-intense-colours
                          (list magenta-intense cyan-intense orange-intense
                                purple-intense green-intense red-intense
                                blue-intense yellow-intense fg-main))) )
      (dotimes (i 9)
        (push `(defface ,(intern (format "rainbow-delimiters-depth-%d-face" (1+ 
i)))
                 '( (default (:inherit rainbow-delimiters-base-face))
                    ( ((class color) (background light))
                      :foreground ,(nth i light-colors))
                    ( ((class color) (background dark))
                      :foreground ,(nth i dark-colors)) )
                 ,(format "Nested delimiter face, depth %d." (1+ i))
                 :group 'rainbow-delimiters-faces)
              faces))
      `(progn ,@faces)) ))

This would not be workable in practice.  Because the dependency only
matters when your macro is being evaluated and the faces are reified.
Since you have a package, that will be the moment the package gets
required. So you are making it a dependency for everyone.

Ideally, one could simply add the colour values directly so it would
work everywhere, or use the modus-themes dependency for those who are
using modus-themes.








reply via email to

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