develooper Front page | perl.perl5.porters | Postings from February 2017

Re: [perl #130467] Default perl builds to not include . in @INC(default_inc_excludes_dot)

Thread Previous | Thread Next
From:
Sawyer X
Date:
February 8, 2017 11:02
Subject:
Re: [perl #130467] Default perl builds to not include . in @INC(default_inc_excludes_dot)
Message ID:
029ec148-7dcd-c8ac-643a-76712dd0d2f4@gmail.com


On 02/06/2017 08:40 PM, Todd Rinaldo via RT wrote:
> So I have cancelled the pull requests for Module::Build, Module::Build::Tiny, and ExtUtils::MakeMaker. It appears that the majority of issues can be addressed by:
>
> 1. Merge this release into core: https://metacpan.org/release/LEONT/Test-Harness-3.37_01
> 2. Make it so Module::Install is always installed to the local perl so that use inc::Module::Install is always available outside of the local directory.
> 3. Track and fix what remains broken. At this time I've reported about 20 modules that are broken. There are definitely more but I think it's still a manageable number.
>
> If we can solve number 2, then I can't think of anything blocking the suggested action in this ticket of changing the default build option to not include . in @INC.
>
> One concern raised so far is that while shipping M::I with perl would solve the problem, it might indicate that p5p endorses the use of Module::Install. Karen has pointed out to me that we already explicitly discourage new use of this module so that should not be a significant issue?

Including Module::Install *does* implicitly encourage it. The only way
would be to document very very clearly that it should not be used. At
the same time, this also implies an adoption of this by p5p, even if to
slowly deprecate it.

This raises another question on my part: Is the intention for this to be
for good, or for a period? If for a period, how long of a period?

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