bug#70525: [PATCH] Make auto-reveal customizations easier to extend

[Arash Esbati]

Hi Paul,

Paul Nelson <ultrono@gmail.com> writes:

> I think what you describe is indeed the plan, but I'll elaborate just
> in case.  I had in mind that external packages would add to
> *-reveal-commands via setup functions, with lines such as
>
>   (add-to-list 'preview-auto-reveal-commands #'tex-parens-down-list)
>
> Users can similarly add their own commands in their own config.  This
> is simpler than adding to the internals of the lisp form in the
> current default value of *-reveal.
>
> Of course, this is only for users who use the default setting.  Other
> users can continue to customize the *-reveal behavior however they
> prefer.

Thanks.  In this case, I think we can have a sort of API by patching the
function `TeX-fold-auto-reveal-p' and declaring a standard var like this:

--8<---------------cut here---------------start------------->8---
(defvar-local TeX-fold-auto-reveal-external-commands nil
  "List of external commands which may cause a fold to be revealed.
This is list is intended for external packages ...")

(defun TeX-fold-auto-reveal-p (mode)
  "Decide whether to auto-reveal.
Return non-nil if folded region should be auto-opened.
See `TeX-fold-auto-reveal' for definitions of MODE."
  (cond ((symbolp mode)
         (and (boundp mode)
              (symbol-value mode)))
        ;; Clause modified:
        ((and (consp mode)
              (null TeX-fold-auto-reveal-external-commands))
         (apply (car mode) (cdr mode)))
        ;; Clause added:
        ((and (consp mode) TeX-fold-auto-reveal-external-commands)
         (apply (car mode) (cons (caadr mode)
                                 (append TeX-fold-auto-reveal-external-commands
                                         (cdadr mode)))))
        (t mode)))
--8<---------------cut here---------------end--------------->8---

Does this make sense?  The question would then be: Should
`TeX-fold-auto-reveal-external-commands' be a regualar var or a custom
one?

> I'm not sure I understand what you mean when you say "that value is
> used when *-reveal is set".  The value would be used every time the
> lisp form in *-reveal is evaluated by the preview/folding libraries.

Yes, sorry, I should have looked more carefully.  I'm not that familiar
with the code, so bear with me.

Best, Arash