bug#42248: 27.0.91; With enchant-2.2.8 from Guix, Flyspell errors out or gives lots of false positives

[Eli Zaretskii <eliz@gnu.org>]

> From: Jorge P. de Morais Neto <jorge+list@disroot.org>
> Date: Wed, 19 Aug 2020 22:35:19 -0300
> Cc: 42248@debbugs.gnu.org
> 
> > Here, "doesn't" gives the wrong result, but not the number "2015".
> 
> Yes.  I don't remember how exactly I got a false positive on "2015", but I
> tried your trimmed recipe and here too there is no false positive on
> "2015" with flyspell-mode.  There is, however, a false positive with
> ispell-buffer.  Should I open a new bug report?
> 
> The recipe:
> 
> 0. emacs -Q
> 1. Evaluate
>     (progn
>            (setq ispell-program-name "enchant-2")
>            (ispell-change-dictionary "en_US")
>            (insert "doesn't 2015")
>            (ispell-buffer))

Do you see the same problem when you invoke Enchant from the shell
prompt?

The examples you provided all work fine for me, although not with
Enchant, so I'm inclined to think this is an Enchant problem.  Or
maybe they modified the interface, making it no longer compatible with
ispell.el.  In the latter case, someone with access to Enchant will
have to tell us how to modify ispell.el to adapt to the Enchant
changes.

The error message shown in your original report seems to indicate that
our invocation of Enchant is somehow incorrect, but there isn't enough
information there to understand what is incorrect and how to fix that.

Thanks.