Re: Tree-sitter integration on feature/tree-sitter

[Eli Zaretskii <eliz@gnu.org>]

> From: Yuan Fu <casouri@gmail.com>
> Date: Sat, 18 Jun 2022 01:25:01 -0700
> Cc: emacs-devel@gnu.org
> 
> >>> Then removing a parser means we actually delete it?
> >> 
> >> Not sure what do you men “delete”.
> > 
> > If creating a parser adds it to the list, then I guessed the semantics
> > of removing from the list is the opposite: having the parser no longer
> > exist, i.e. "delete" it.
> > 
> > But now I'm confused by what you say here:
> > 
> >> Treesit-parser-delete removes the parser from the parser list of a buffer, so it is never kept up-to-date with that buffer again. But you can still do stuff with it until it is gc’ed.
> > 
> > If we already have treesit-parser-delete, and that call removes the
> > parser from the list, then why would we need a function "to remove a
> > parser from the list"?  It sounds like treesit-parser-delete already
> > does it?
> 
> Yeah. There is no other function, treesit-parser-delete deletes and removes the parser.

So you agree with me that a function to remove from the list is not
needed?  Once the list is no longer exposed to Lisp, the way Lisp
programs should manipulate the list is by adding and deleting parsers,
and by asking Emacs to show the list of existing parsers.  Right?

> Though I don’t know how can you make a Lisp_Object “no longer exist”. Normally we just turn #<stuff> into #<deleted stuff>, right?

Yes, we leave the actual deleting to GC.