Mailing List Archive: PSC #043, 2021-10-29

PSC #043, 2021-10-29

perl.p5p at rjbs

Nov 8, 2021, 6:53 PM

Post #1 of 11 (426 views)

Porters,

[ *Oops, it appears I never sent this draft. *]

The PSC had a Zoom call on Friday after a week off. It was pretty low-key, but topics of note included:
* moving work on signatures forward, about which expect more soon
* required next steps on RFCs: namely, Rik and Paul to merge some MRs and push along some things that still seem plausible for v5.36
I am most especially keen to move forward with something to allow easy k/v iteration of an array and its indexes with 2-item foreach iteration. Others on the call had other favorite things to learn on, as one might expect!

We never did finish our review of the big "quirks" doc, but have put it on a back burner as we focus on getting things done before we start heading into the start-of-2022 code freeze.

--
rjbs

Re: PSC #043, 2021-10-29 [ In reply to ]

davidnicol at gmail

Nov 9, 2021, 7:42 AM

Post #2 of 11 (426 views)

I'm guessing
use Acme::TieArrayToIterableHashref 'itarr';
while (my ($i,$v) = each %{ itarr \@SomeArray }){
...

isn't sufficiently easy? Because that would be trivial to set up.

On Mon, Nov 8, 2021 at 8:54 PM Ricardo Signes <perl.p5p@rjbs.manxome.org>
wrote:

> Porters,
>
> [ *Oops, it appears I never sent this draft. *]
>
> The PSC had a Zoom call on Friday after a week off. It was pretty
> low-key, but topics of note included:
>
> - moving work on signatures forward, about which expect more soon
> - required next steps on RFCs: namely, Rik and Paul to merge some MRs
> and push along some things that still seem plausible for v5.36
>
> I am most especially keen to move forward with something to allow easy k/v
> iteration of an array and its indexes with 2-item foreach iteration.
> Others on the call had other favorite things to learn on, as one might
> expect!
>
> We never did finish our review of the big "quirks" doc, but have put it on
> a back burner as we focus on getting things done before we start heading
> into the start-of-2022 code freeze.
>
> --
> rjbs
>

--
"Lay off that whiskey, and let that cocaine be!" -- Johnny Cash

Re: PSC #043, 2021-10-29 [ In reply to ]

grinnz at gmail

Nov 9, 2021, 7:52 AM

Post #3 of 11 (426 views)

On Tue, Nov 9, 2021 at 10:43 AM David Nicol <davidnicol@gmail.com> wrote:

>
>
> I'm guessing
> use Acme::TieArrayToIterableHashref 'itarr';
> while (my ($i,$v) = each %{ itarr \@SomeArray }){
> ...
>
> isn't sufficiently easy? Because that would be trivial to set up.
>
>
> On Mon, Nov 8, 2021 at 8:54 PM Ricardo Signes <perl.p5p@rjbs.manxome.org>
> wrote:
>
>> I am most especially keen to move forward with something to allow easy
>> k/v iteration of an array and its indexes **with 2-item foreach iteration.**
>>
>
Emphasis mine.

-Dan

Re: PSC #043, 2021-10-29 [ In reply to ]

scott at perturb

Nov 9, 2021, 9:41 AM

Post #4 of 11 (426 views)

I am anxiously awaiting the future of signatures. The "experimental"
signatures we have already are the best of any language I've used. I
really hope they can become first class citizens soon.

Put me down as a +1 for signatures

- Scott

On 11/8/2021 6:53 PM, Ricardo Signes wrote:
> Porters,
>
> [ /Oops, it appears I never sent this draft. /]
>
> The PSC had a Zoom call on Friday after a week off. It was pretty
> low-key, but topics of note included:
>
> * moving work on signatures forward, about which expect more soon
>

Re: PSC #043, 2021-10-29 [ In reply to ]

kimoto.yuki at gmail

Nov 9, 2021, 5:19 PM

Post #5 of 11 (426 views)

2021-11-9 11:54 Ricardo Signes <perl.p5p@rjbs.manxome.org> wrote:

>
> - moving work on signatures forward, about which expect more soon
>
> I hope for a little more discussion to reach a better idea and consensus.

Re: PSC #043, 2021-10-29 [ In reply to ]

rwp.primary at gmail

Nov 9, 2021, 11:28 PM

Post #6 of 11 (426 views)

please, make non-experimental default signatures happen for v5.36

thanks

On Wed, Nov 10, 2021 at 5:20 AM Yuki Kimoto <kimoto.yuki@gmail.com> wrote:

>
> 2021-11-9 11:54 Ricardo Signes <perl.p5p@rjbs.manxome.org> wrote:
>
>>
>> - moving work on signatures forward, about which expect more soon
>>
>> I hope for a little more discussion to reach a better idea and consensus.
>
>
>

Re: PSC #043, 2021-10-29 [ In reply to ]

perl5-porters at perl

Nov 10, 2021, 8:52 AM

Post #7 of 11 (426 views)

Any reason why "state" still needs "use v5.10"?

https://perldoc.perl.org/functions/state

Cheers,
Brett

* Elvin Aslanov <rwp.primary@gmail.com> [2021-11-10 11:28:24 +0400]:

> please, make non-experimental default signatures happen for v5.36
>
> thanks
>
> On Wed, Nov 10, 2021 at 5:20 AM Yuki Kimoto <kimoto.yuki@gmail.com> wrote:
>
> >
> > 2021-11-9 11:54 Ricardo Signes <perl.p5p@rjbs.manxome.org> wrote:
> >
> >>
> >> - moving work on signatures forward, about which expect more soon
> >>
> >> I hope for a little more discussion to reach a better idea and consensus.
> >
> >
> >

--
--
oodler@cpan.org
oodler577@sdf-eu.org
SDF-EU Public Access UNIX System - http://sdfeu.org
irc.perl.org #openmp #pdl #native

Re: PSC #043, 2021-10-29 [ In reply to ]

grinnz at gmail

Nov 10, 2021, 8:57 AM

Post #8 of 11 (426 views)

On Wed, Nov 10, 2021 at 11:52 AM Oodler 577 via perl5-porters <
perl5-porters@perl.org> wrote:

> Any reason why "state" still needs "use v5.10"?
>
> https://perldoc.perl.org/functions/state

As any new keyword does, because it will break code that defines or uses a
"sub state".

-Dan

Re: PSC #043, 2021-10-29 [ In reply to ]

perl5-porters at perl

Nov 10, 2021, 9:01 AM

Post #9 of 11 (426 views)

* Dan Book <grinnz@gmail.com> [2021-11-10 11:57:05 -0500]:

> On Wed, Nov 10, 2021 at 11:52 AM Oodler 577 via perl5-porters <
> perl5-porters@perl.org> wrote:
>
> > Any reason why "state" still needs "use v5.10"?
> >
> > https://perldoc.perl.org/functions/state
>
>
> As any new keyword does, because it will break code that defines or uses a
> "sub state".
>
> -Dan

Thank you.

Makes sense. The documentation isn't clear. May I assume a PR
would be welcome to update this? Any information on this policy
regarding new keywords? I am curious and it would be nice to
make reference to this in any PR I might make.

Cheers,
Brett

--
--
oodler@cpan.org
oodler577@sdf-eu.org
SDF-EU Public Access UNIX System - http://sdfeu.org
irc.perl.org #openmp #pdl #native

Re: PSC #043, 2021-10-29 [ In reply to ]

davidnicol at gmail

Nov 10, 2021, 9:15 AM

Post #10 of 11 (426 views)

could we just make core keywords overwritable on a package basis? This is
with 5.26:

$ perl -le 'sub my { print 5 }; my'
syntax error at -e line 1, at EOF
Execution of -e aborted due to compilation errors.

$ perl -le 'sub state { print 5 }; state'
5

$ perl -le ' use v5.10 ; sub state { print 5 }; state'
syntax error at -e line 1, at EOF
Execution of -e aborted due to compilation errors.

> As any new keyword does, because it will break code that defines or uses a
> > "sub state".
>
--
"Lay off that whiskey, and let that cocaine be!" -- Johnny Cash

Re: PSC #043, 2021-10-29 [ In reply to ]

perl5-porters at perl

Nov 10, 2021, 3:24 PM

Post #11 of 11 (426 views)

On Thu, 11 Nov 2021 at 01:16, David Nicol <davidnicol@gmail.com> wrote:

>
> could we just make core keywords overwritable on a package basis? This is
> with 5.26:
>

That would still break cases like this, I think?

perl -E'use feature qw(state); state $x; sub state { ... }'