Re: Testing redisplay code in batch

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: emacs-devel@gnu.org
> Date: Wed, 23 Sep 2020 14:37:50 -0400
> 
> Indeed.  In the past I was thinking (abstractly) of much more demanding
> circumstances, but for this specific test, all I need is to know the
> `window-start`, so the corresponding data is already exposed to Lisp.
> I don't know if that makes it easy enough to implement it without
> significant work.

What about running a second instance of Emacs as a subprocess,
starting it without -batch?  On a Posix system, that would use a PTY,
which supposedly will look to the sub-Emacs as a text-mode display.
Then at least the features that don't require GUI capabilities could
be tested.  Does that make sense?