On that note someone mentioned to me that all it takes is someone with a
commit bit to be willing to act as release manager for a special release of
older versions to make it happen. I do not have a commit bit, but if I were
willing to act as a release manager for a special maint release of 5.8-5.16
would that be workable? Note I am just asking at this point, not sure I
would be able to commit the time to it yet.

Even then though, chances of everyone upgrading to a new 5.8 are pretty
slim, so I am not sure it would be worth it. I still have to support the
versions with the broken behavior regardless.

On Fri, Dec 12, 2014 at 10:51 AM, Chad Granum wrote:

Ah, yeah, I have done that before (in fact there is a similar one in
Test::More already). I will consider it!

On Thu, Dec 11, 2014 at 2:46 PM, Father Chrysostomos via RT <
perlbug-followup@perl.org> wrote:
On Thu Dec 11 12:45:55 2014, exodist7@gmail.com wrote:
Ok, code similar to what bulk88 suggested seems to fix a majority of
windows woes (I got a vm spun up to test).
BTW, if you need the manual localisation unwound for every type of scope
exit (e.g., goto, next), then a destructor-calling-a-closure may also work,
as in End.pm and B::Hooks::EndOfScope (which are so trivial you could
inline them with just 2-3 lines of code).


Father Chrysostomos

via perlbug: queue: perl5 status: pending release

Search Discussions

Discussion Posts


Follow ups

Related Discussions

Discussion Navigation
viewthread | post
posts ‹ prev | 18 of 20 | next ›
Discussion Overview
groupperl5-porters @
postedDec 11, '14 at 4:10p
activeFeb 6, '15 at 10:06a



site design / logo © 2021 Grokbase