develooper Front page | perl.perl5.porters | Postings from May 2016

Re: 5.24.0 + cpan

Thread Previous | Thread Next
From:
Leon Timmermans
Date:
May 18, 2016 12:26
Subject:
Re: 5.24.0 + cpan
Message ID:
CAHhgV8i1EoVmFMJTtLoUG9mZW+XOGxdq=iUiK5epLhT3zA7MvA@mail.gmail.com
On Mon, May 16, 2016 at 9:52 PM, Zefram <zefram@fysh.org> wrote:

> That retains the rather crap behaviour of looking only at the *first*
> instance of "perl" on the line.  It would make more sense to look for
> "perl6" either at *all* instances of "perl" (requiring all to be "perl6"
> to enable chaining) or just in the basename part of the interpreter path.
>

Yeah. IMO the only choices for 5.24.1 are to either revert the patch
entirely or make it a lot more complicated than it currently is. I'm rather
doubtful on doing the latter under the time pressure of a .1; having a
whole year to come up with something or decide to abandon this entirely
sounds like a better idea to me.

But I go back to what I said about this last year, when the "perl6"
> change was originally proposed: the #! hack is obsolete.  We should either
> leave it totally unchanged (as it was pre-5.23) or remove it entirely.
> Changing it for perl6 is just inviting nasty surprises.


Test::Harness in particular tends to rely on this hack, but that could be
fixed (not that T::H is making fixing anything a sinecure). A deprecation
cycle would be appropriate though if we are to remove it.

Leon

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About