bug#66732: tree-sitter fontification doesn't update multi-line syntax reliably

[Eli Zaretskii]

> Date: Tue, 19 Dec 2023 01:10:39 +0200
> Cc: 66732@debbugs.gnu.org, dominik@honnef.co, casouri@gmail.com
> From: Dmitry Gutov <dmitry@gutov.dev>
> 
> On 18/12/2023 21:33, Eli Zaretskii wrote:
> >> From: Stefan Monnier<monnier@iro.umontreal.ca>
> >> Cc: Eli Zaretskii<eliz@gnu.org>,66732@debbugs.gnu.org,
> >>    dominik@honnef.co,  Yuan Fu<casouri@gmail.com>
> >> Date: Mon, 18 Dec 2023 14:12:02 -0500
> >>
> >> I'm surprised that 42 behaves differently from t.
> > The display code in C tests only for nil or non-nil, but jit-lock.el
> > tests against a literal t in a few places.  (And don't forget about
> > the special value 'defer' we use.)
> > 
> >> What we really need here is to force the redisplay to consider that this
> >> part of the buffer needs to be re-rendered.
> > AFAIU, it will only re-render that part of some of the faces changed.
> 
> It seems like in our scenario the region does not get re-rendered even 
> when some of the faces changed in there.

Can't happen.