Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

[Stefan Monnier]

>     New macro seq-let, providing destructuring support to seq.el

How 'bout adding support for `seq' to pcase instead?
I.e. add a (seq ...) pattern to pcase.
You can do that with `pcase-defmacro'.


        Stefan

Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

[Nicolas Petton]

[-- Attachment #1: Type: text/plain, Size: 376 bytes --]


Stefan Monnier writes:

>>     New macro seq-let, providing destructuring support to seq.el
>
> How 'bout adding support for `seq' to pcase instead?
> I.e. add a (seq ...) pattern to pcase.
> You can do that with `pcase-defmacro'.

Wouldn't that be for a difference usage? If not, can you provide an
example?

Nico
-- 
Nicolas Petton
http://nicolas-petton.fr

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 512 bytes --]

Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

[Bozhidar Batsov]

[-- Attachment #1: Type: text/plain, Size: 492 bytes --]

seq-let seems pretty useful to me.

On 1 May 2015 at 21:34, Nicolas Petton <nicolas@petton.fr> wrote:

>
> Stefan Monnier writes:
>
> >>     New macro seq-let, providing destructuring support to seq.el
> >
> > How 'bout adding support for `seq' to pcase instead?
> > I.e. add a (seq ...) pattern to pcase.
> > You can do that with `pcase-defmacro'.
>
> Wouldn't that be for a difference usage? If not, can you provide an
> example?
>
> Nico
> --
> Nicolas Petton
> http://nicolas-petton.fr
>

[-- Attachment #2: Type: text/html, Size: 1031 bytes --]

Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

[Stefan Monnier]

> Wouldn't that be for a difference usage? If not, can you provide an
> example?

You could use it as in

   (pcase-let (((seq a b) [1 2 3]))
     (message "%s %s" a b))

which could return "1 2", tho that depends how you'd define the (seq
...) pattern, obviously.


        Stefan

Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

[Nicolas Petton]

[-- Attachment #1: Type: text/plain, Size: 633 bytes --]


Stefan Monnier writes:

>> Wouldn't that be for a difference usage? If not, can you provide an
>> example?
>
> You could use it as in
>
>    (pcase-let (((seq a b) [1 2 3]))
>      (message "%s %s" a b))
>
> which could return "1 2", tho that depends how you'd define the (seq
> ...) pattern, obviously.

I see, it would indeed makes sense to have it.  I'll have a look at how
pcase works internally and how patterns can be defined.

I think having both would make sense though as `seq-let' is IMO simpler
to use.  Maybe one could reuse code from the other.

Nico
-- 
Nicolas Petton
http://nicolas-petton.fr

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 512 bytes --]

Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

[Stefan Monnier]

> I see, it would indeed makes sense to have it.  I'll have a look at how
> pcase works internally and how patterns can be defined.

grep for pcase-defmacro: that should give you a good starting point.

> I think having both would make sense though as `seq-let' is IMO simpler
> to use.  Maybe one could reuse code from the other.

I don't think the `seq' pcase pattern can be defined using seq-let, but
seq-let could easily be defined as a wrapper that expands to a pcase-let.

BTW, another advantage of defining a pcase pattern is that you could then
also do things like

   (pcase-dolist ((seq a b c) my-things)
     ...)


-- Stefan

Re: [Emacs-diffs] seq-let c856843: New macro seq-let, providing destructuring support to seq.el

[Nicolas Petton]

[-- Attachment #1: Type: text/plain, Size: 554 bytes --]


Stefan Monnier writes:

>>     New macro seq-let, providing destructuring support to seq.el
>
> How 'bout adding support for `seq' to pcase instead?
> I.e. add a (seq ...) pattern to pcase.
> You can do that with `pcase-defmacro'.

I pushed in commit 387e1e1 another version that is based on pcase.

As old versions of Emacs don't provide `pcase-defmacro', I kept the old
version of seq-let for backward compatibility.

Stefan, could you review my changes, I'm quite new with pcase.

Nico
-- 
Nicolas Petton
http://nicolas-petton.fr

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 512 bytes --]