Bug #70859 for Classic-Perl: [PATCH] $[

Sat Sep 10 05:01:48 2011 zefram [...] fysh.org - Ticket created

Subject:	[PATCH] $[
Date:	Sat, 10 Sep 2011 10:01:36 +0100
To:	bug-Classic-Perl [...] rt.cpan.org
From:	Zefram <zefram [...] fysh.org>

Attached patch adds support for $[, recently removed from blead. The C:P implementation of $[ is lexically scoped, as the core's has been since Perl 5.000, and works by invoking Array::Base and String::Base. These don't (and can't) perfectly match the core's $[ semantics, but they'll match for simple stuff. The C:P support for $[ is triggered by the Perl version number being at least 5.15.3. You need to modify that to test against current blead, but it'll be correct when applied to released perls. -zefram

Message body is not shown because sender requested not to inline it.

Sun Sep 11 17:43:52 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Sat Sep 10 05:01:48 2011, zefram@fysh.org wrote: Show quoted text

> Attached patch adds support for $[, recently removed from blead. > The C:P implementation of $[ is lexically scoped, as the core's has been > since Perl 5.000, and works by invoking Array::Base and String::Base. > These don't (and can't) perfectly match the core's $[ semantics, but > they'll match for simple stuff. > > The C:P support for $[ is triggered by the Perl version number being at > least 5.15.3. You need to modify that to test against current blead, > but it'll be correct when applied to released perls. > > -zefram

Thank you. I’m not sure when I’ll make a new release, as Jesse Vincent wrote to me recently about putting Classic::Perl or something derived from it in core, presumably to be loaded automatically outside the scope of 5.16. I’m awaiting his response to my questions. We’ll see how it turns out.

Sun Sep 11 17:43:54 2011 The RT System itself - Status changed from 'new' to 'open'

Mon Sep 12 06:18:50 2011 zefram [...] fysh.org - Correspondence added

Subject:	Re: [rt.cpan.org #70859] [PATCH] $[
Date:	Mon, 12 Sep 2011 11:18:38 +0100
To:	Father Chrysostomos via RT <bug-Classic-Perl [...] rt.cpan.org>
From:	Zefram <zefram [...] fysh.org>

I think Classic::Perl itself is insufficiently clean to be blessed by the core. You'd need a substantially revised version or a derivative module. In any case, I'm dubious about automatically invoking anything that doesn't accurately reproduce the former core semantics. Anyway, however that turns out, I don't think you gain anything by holding off releasing a Classic::Perl with $[ support. Seeing what it's like as an operational module would help in deciding what relationship it should have with the core. -zefram

Sun Sep 18 15:58:46 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Mon Sep 12 06:18:50 2011, zefram@fysh.org wrote: Show quoted text

> I think Classic::Perl itself is insufficiently clean to be blessed by > the core. You'd need a substantially revised version or a derivative > module. In any case, I'm dubious about automatically invoking anything > that doesn't accurately reproduce the former core semantics.

Will you accept patches to String::Base and Array::Base that make them closer to $[? Show quoted text

> > Anyway, however that turns out, I don't think you gain anything by holding > off releasing a Classic::Perl with $[ support. Seeing what it's like > as an operational module would help in deciding what relationship it > should have with the core.

I was about to, when I noticed something I missed when I first looked at your patch: Only in Classic-Perl-0.02a.mod0/t: arybase.t If you’ve already taken the time to write a test, may I have it, please? :-)

Sun Sep 18 19:58:55 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Sun Sep 18 15:58:46 2011, SPROUT wrote: Show quoted text

> On Mon Sep 12 06:18:50 2011, zefram@fysh.org wrote:

> > I think Classic::Perl itself is insufficiently clean to be blessed

> by

> > the core. You'd need a substantially revised version or a

> derivative

> > module. In any case, I'm dubious about automatically invoking

> anything

> > that doesn't accurately reproduce the former core semantics.

> > Will you accept patches to String::Base and Array::Base that make them > closer to $[? >

> > > > Anyway, however that turns out, I don't think you gain anything by

> holding

> > off releasing a Classic::Perl with $[ support. Seeing what it's

> like

> > as an operational module would help in deciding what relationship it > > should have with the core.

> > I was about to, when I noticed something I missed when I first looked > at your patch: > > Only in Classic-Perl-0.02a.mod0/t: arybase.t > > If you’ve already taken the time to write a test, may I have it, > please? :-)

I’ve just uploaded 0.03 without tests, but I did test it manually a bit.

Sun Sep 18 19:58:55 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Status changed from 'open' to 'resolved'

Mon Sep 19 03:26:34 2011 zefram [...] fysh.org - Correspondence added

Subject:	Re: [rt.cpan.org #70859] [PATCH] $[
Date:	Mon, 19 Sep 2011 08:26:21 +0100
To:	Father Chrysostomos via RT <bug-Classic-Perl [...] rt.cpan.org>
From:	Zefram <zefram [...] fysh.org>

Father Chrysostomos via RT wrote: Show quoted text

>Will you accept patches to String::Base and Array::Base that make them closer to $[?

Maybe. I'm concerned about incompatibility, but they're young and have no-to-few users. Show quoted text

>Only in Classic-Perl-0.02a.mod0/t: arybase.t

Oops. Attached. -zefram

Message body is not shown because sender requested not to inline it.

Mon Sep 19 03:26:35 2011 The RT System itself - Status changed from 'resolved' to 'open'

Sun Sep 25 23:12:43 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Mon Sep 19 03:26:34 2011, zefram@fysh.org wrote: Show quoted text

> Father Chrysostomos via RT wrote:

> >Will you accept patches to String::Base and Array::Base that make

> them closer to $[? > > Maybe. I'm concerned about incompatibility, but they're young and > have > no-to-few users. >

> >Only in Classic-Perl-0.02a.mod0/t: arybase.t

> > Oops. Attached.

Thank you. I have just included it with version 0.04.

Sun Sep 25 23:12:44 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Status changed from 'open' to 'resolved'

Sat Oct 15 20:47:22 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Mon Sep 19 03:26:34 2011, zefram@fysh.org wrote: Show quoted text

> Father Chrysostomos via RT wrote:

> >Will you accept patches to String::Base and Array::Base that make

> them closer to $[? > > Maybe. I'm concerned about incompatibility, but they're young and > have > no-to-few users.

Actually, I think that Array::Base and String::Base are fine as they are. That way $[ users have the option to upgrade to a less eccentric version of the feature. :-) If you don’t mind, I’m going to copy some code from your modules and tweak it till it behaves more like the historical $[. Show quoted text

>

> >Only in Classic-Perl-0.02a.mod0/t: arybase.t

> > Oops. Attached. > > -zefram

Sat Oct 15 20:47:23 2011 The RT System itself - Status changed from 'resolved' to 'open'

Sat Oct 15 20:47:23 2011 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Status changed from 'open' to 'resolved'