Re: [SPAM UNSURE] Maybe we're taking a wrong approach towards tree-sitter

[Eli Zaretskii]

> From: Andrei Kuznetsov <r12451428287@163.com>
> Cc: manuel@ledu-giraud.fr,  stephen_leake@stephe-leake.org,
>   emacs-devel@gnu.org
> Date: Wed, 28 Jul 2021 21:14:48 +0800
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > TS's code is written in plain C, and doesn't require any regeneration
> > or source modifications.  Anything else is misunderstanding.
> 
> I am confused by TS's documentation, but if my understanding is correct,
> shouldn't it be a parser generator that generates C code?

TS is not a parser generator, it's a parser that accepts the language
grammar from external files.

> In that case, how would users load new parsers or modify existing
> ones?

If you want to modify a TS grammar file, you can (not in C).  But why
would you want to?  The whole point of using TS is NOT to require that
the Emacs development team or Emacs users should know enough about
parsing of the many languages Emacs supports to modify the grammar.
We want another, independent development team to take care of that,
and we want to use the results of their development with minimum
fuss.  Exactly like we do with other libraries developed by other
projects: the image libraries, GnuTLS, HarfBuzz, etc.