develooper Front page | perl.perl5.porters | Postings from March 2023

Re: strictness and "use 5.whatever"

Thread Previous | Thread Next
Zefram via perl5-porters
March 20, 2023 04:44
Re: strictness and "use 5.whatever"
Message ID:
Paul "LeoNerd" Evans wrote:
>Yes; that is exactly the point. We wanted it to have (what you call)
>the "hard stricture" behaviour in 5.36.

Whatever you wanted, "use v5.36" on actual Perl 5.36 actually exhibits
soft stricture behaviour.  That's the established meaning of "use v5.36",
and to change it to hard stricture would be a break of backcompat.

>                                        It's just that my
>implementation of it was sloppy and broke the behaviour of the lower
>`use VERSION` requests as well. Your patch fixes the older versions,
>but reverts `use v5.36` back to that older behaviour.

I'm not clear what code you're referring to as your implementation,
nor what is your reference point for judging what my patch "fixes" and
"reverts".  I see no evidence of the stricture meaning of "use v5.16"
having ever changed in released Perl versions since Perl 5.15.6.  My patch
deliberately doesn't change the behaviour of "use v5.36", "use v5.16",
"use v5.10", or any other version declaration <=5.36.  Perhaps you're
seeing my patch as "fixing" and "reverting" these versions because you're
seeing it as a replacement for some other patch, planned but never applied
to blead, that would have changed the effects of those declarations?


The change of the stricture behaviour threshold from 5.37 to 5.35 would
be a break of backcompat, and is a bad idea.


Thread Previous | Thread Next Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at | Group listing | About