bug#56637: 28.1.90; [FR] Allow a way around font-lock-mode being unconditionally disabled in " *hidden*" buffers

[Eli Zaretskii <eliz@gnu.org>]

> From: Ihor Radchenko <yantar92@gmail.com>
> Date: Tue, 19 Jul 2022 12:15:33 +0800
> 
> font-lock-mode contains the follow clause:
> 
>   ;; Don't turn on Font Lock mode if we don't have a display (we're running a
>   ;; batch job) or if the buffer is invisible (the name starts with a space).
>   (when (or noninteractive (eq (aref (buffer-name) 0) ?\s))
>     (setq font-lock-mode nil))
> 
> which interferes with Org mode's fontification code in some edge cases.
> 
> Org mode fontifies code blocks using "hidden" temporary buffers with
> their name starting from " ": " *org-src-fontification:major-mode*".
> 
> The buffers are kept around to avoid extra major-mode loading overheads
> and the fontification is not using (font-lock-ensure).
> 
> (font-lock-ensure) usually fontifies the buffer correctly even though
> font-lock-mode is disabled. However, not always.
> 
> We just got some edge case
> https://teddit.net/r/orgmode/comments/w2b0tw/syntax_highlighting_in_orgsource_blocks/igqdx18/
> with web-mode, which directly sets 'font-lock-face text properties.
> 
> Without font-lock-mode being enabled, 'font-lock-face is not
> automatically remapped to 'face text property and hence some part of the
> fontification may not be reflected by simply looking into 'face
> property.
> 
> While Org can work around this particular case by looking into
> 'font-lock-face, I am not sure if disabled font-lock may not trigger
> similar issues in other cases.
> 
> It would be nice if we could somehow enable font-lock inside " *hidden*"
> buffers. Would it be possible?

Depends on the "somehow" part.  I generally prefer that obscure corner
cases be solved in the software which creates those obscure cases.  In
this case, it seems like web-mode is that place?  I'm not sure why
you say Org should solve this, but perhaps I didn't understand the
description well enough.  But if this indeed needs to be solved in
Org, how about not using buffers whose names begin with a space?