Re: rosie/libpexl library for regex pattern composition

[Helmut Eller <eller.helmut@gmail.com>]

On Mon, Jul 29 2024, Danny McClanahan wrote:

[...]
> I had not considered the possibility of exposing a module API for
> extensible parsing engines as opposed to building it directly into
> Emacs like the current regex-emacs engine, but I definitely appreciate
> the idea of decoupling from a specific framework, especially with an
> eye to reducing the maintenance burden. From my current understanding,
> the benefits of building Rosie/PEXL directly into Emacs would be:
> (1) could take advantage of e.g. SIMD search by directly accessing the
> gap buffer memory,

It depends on how the module API exposes the buffer text.  E.g. I would
imagine that a Rust function that receives the text as two &[u8],
i.e. two byte array slices borrowed read-only for the duration of a
call, could have similar efficiency as something built directly into
Emacs.

> (2) since it would be reliably available to all Emacs users, could
> form a composeable elisp search API.

I think Rosie/PEXL would stay an optional feature for a long time.

[...]
> Does that correctly outline the goals/desires for lisp-level matcher
> construction?

I don't know.  This sounds like an new API would be needed and the
difficulty, beside designing the API, would be to make it popular;
usually the most difficult part.  The old API would still be used for a
long time.

I think the Emacs community is mostly a do-ocracy, i.e., people work on
things they are interested in or know how to do; in contrast to work on
goals/ideas that somebody else defined.  That's quite ok as long as the
boring tasks get done eventually.  And I would suggest the same here: do
what you think is useful and that you know how to do.