bug#27761: Crash while using proof-general/company-coq on OS X

[Eli Zaretskii]

> Date: Fri, 04 Aug 2017 10:23:15 +0900
> From: YAMAMOTO Mitsuharu <mituharu@math.s.chiba-u.ac.jp>
> Cc: Денис Редозубов
>  <denis.redozubov@gmail.com>, 27761@debbugs.gnu.org,  jwiegley@gmail.com,
>       charles@aurox.ch
> 
> >>>>> On Wed, 02 Aug 2017 20:32:29 +0300, Eli Zaretskii <eliz@gnu.org> said:
> 
> > Thanks.  The "charpos = 6" part puzzles me, because the string is
> > only 4 characters long.  What does this produce:
> 
> This would be the length of the string "forall", which is composed
> into "∀" by prettify-symbol-mode (see my another post containing
> screenshots).

So you are saying that it->end_charpos, which is 4, is out of sync
with the string that is being processed?  Can you see how did that
happen, by stepping through the code starting from before the first
recursive call to get_next_display_element?

Thanks.