Re: master 431b098: * lisp/emacs-lisp/pcase.el (let): Reimplement as a pcase macro

Re: master 431b098: * lisp/emacs-lisp/pcase.el (let): Reimplement as a pcase macro

[Basil L. Contovounesios]

monnier@iro.umontreal.ca (Stefan Monnier) writes:

> branch: master
> commit 431b098a206d27a2dff6a88312c28c36926f90e9
> Author: Stefan Monnier <monnier@iro.umontreal.ca>
> Commit: Stefan Monnier <monnier@iro.umontreal.ca>
>
>     * lisp/emacs-lisp/pcase.el (let): Reimplement as a pcase macro
>     
>     (pcase--macroexpand, pcase--u1): Remove handling of `let` from
>     `pcase`s core.

Does this mean (info "(elisp) pcase Macro") has to be updated, e.g. to
document (let ...) alongside (rx ...) as another "built-in" pattern, or
can we still pretend (let ...) is a "core" pattern for documentation
purposes?

Thanks,

-- 
Basil

Re: master 431b098: * lisp/emacs-lisp/pcase.el (let): Reimplement as a pcase macro

[Stefan Monnier]

> Does this mean (info "(elisp) pcase Macro") has to be updated, e.g. to
> document (let ...) alongside (rx ...) as another "built-in" pattern, or
> can we still pretend (let ...) is a "core" pattern for documentation
> purposes?

It's indeed not a "core" pattern any more, but just like with
specialforms-vs-macros, the difference between core patterns and built-in
patterns is just an internal implementation detail, and the way I read
the Texinfo doc, I don't see such a distinction: `rx` is (oddly) placed
together with `and` and `or`, which are both core patterns.

BTW, regarding the `grok/pcase` example, maybe the doc should make it
more clear that this example is a contrived piece of code intended to
combine as many of `pcase`s features as possible, and doesn't intend to
encourage using such style (which I personally find horrendous ;-).


        Stefan