Re: Question about composite.c

[Gerry Agbobada <gagbobada@gmail.com>]

Hello,

It took me a good while, but I found a case where :
- I can reproduce with =emacs -q=  (Reproducing infinite looping in
composition_get_gstring with an empty non-nil string)
- prettify-symbols-mode is not enabled

The example includes the code of eros.el package, which creates an
overlay when calling eval-last-sexp. The overlay is supposed to start
with `=>`, and when I have this in my composition-function-table it
triggers the loop.
As explained in the file linked below, with the cursor on `(+ 3 2)|`
I'd expect `C-x C-e` to create an overlay with `=> 5`. Instead, it
loops indefinitely and the only way out is to kill -9 emacs (kill
-USR2 doesn't help)

Question is : is there something I should do from elisp point of view
(as a user of eros, and composition-function-table) to avoid this
problem ? I'm still not sure about the code path that I'm following to
lead to this point. :(

I'm writing this from gmail, I wasn't able to paste the file in the
editor respecting the newlines, so I pushed it in a gist... I'd prefer
the code to be archived in the mailing list, but until I find a better
mail client than gmail online one, I won't be able to do it sorry.

For the time being, it is here
https://gist.github.com/gagbo/13239f12cfd6b90cce32b1c255f2553d (325
lines long)
Feel free to paste it in the conversation if you feel it is safer.
Someday I'll stop using Gmail I suppose.

Best regards
Gerry

Le mer. 22 janv. 2020 à 09:55, Gerry Agbobada <gagbobada@gmail.com> a écrit :
>
> > Can you show examples of those, if not their exhaustive list?
>
> Between iosevka, Fira Code and Cascadia Code, ligatures that may
> happen starting with * are :
> - *
> - **
> - ***
> - */
> - *>
> - *)
>
> iosevka : https://raw.githubusercontent.com/be5invis/Iosevka/master/images/ligations.png
> FiraCode : https://github.com/tonsky/FiraCode/blob/master/showcases/all_ligatures.png
> Cascadia : https://github.com/microsoft/cascadia-code/wiki/Coding-ligature-coverage
>
> > I don't see why it would be easier to access/modify a separate table.
>
> Fair enough, I'll probably remove the indeirection once I find out
> my elisp issue
>
> > 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.
> > [...]
> > I think the important part is to understand why Emacs is trying to
> > compose an empty string.
>
> I'm trying to do this to understand better yes. In my "bloated" emacs
> setup, I still have the "trying to compose an empty string" issue without
> prettify-symbols-mode, so I'll just continue playing with my short .el file
> until I can find a way to properly reproduce.
>
> Gerry