Re: Question about composite.c

[Eli Zaretskii <eliz@gnu.org>]

> From: Gerry Agbobada <gagbobada@gmail.com>
> Date: Tue, 21 Jan 2020 19:57:07 +0100
> Cc: emacs-devel@gnu.org
> 
> Therefore, the main goal was to "catch common programming ligatures
> starting with ?*"

Can you show examples of those, if not their exhaustive list?

> > Also, why did you need to use a separate char-table instead of
> > composition-function-table?
> 
> I used another char-table so the variable is easily accessible to modify
> parts of it. Then again, maybe a poor design decision, but just a design
> decision.

I don't see why it would be easier to access/modify a separate table.

> > It's not just an empty string, it's a _unibyte_ string.  How did that
> > happen?
> 
> That's the main part I was trying to understand, since I could not
> reproduce it in an easy way, I was wondering if an empty string can be
> unibyte or multibyte. I have my answer now :)

If you post a complete reproduction recipe, starting from "emacs -Q",
perhaps I will be able to help you understand what is the immediate
cause of the problem.

> > There's no problem with empty strings here, as long as they are
> > multibyte.
> 
> That the main point I don't understand : is there a way to mark an empty
> string as a multibyte string ?

We could look into that, but why would an empty string end up in
composition-get-gstring anyway? there's no meaningful way of composing
such a string.

> At least now I know I can continue investigating on the elisp side, thanks for
> your answers !

I think the important part is to understand why Emacs is trying to
compose an empty string.