develooper Front page | perl.perl5.porters | Postings from March 2022

PSC #059 2022-03-17

From:
Paul "LeoNerd" Evans
Date:
March 22, 2022 21:18
Subject:
PSC #059 2022-03-17
Message ID:
20220322211742.0b14da61@shy.leonerd.org.uk
Present: Neil, Paul, Rik

## List Culture

We reviewed a draft email Neil had written from last week's actions.
Further edits will be made and posted soon.
(update: that happened)


## Configure Changes for Disabling Taint Mode

Neil has made updates to four dual-life CPAN dists that ship with perl
core, as the unit tests require some alterations to cope with a
taint-disabled Perl build. PRs have been sent upstream, awaiting merge
and release to make new CPAN modules that we can bring back into core
perl.

We spent some time discussing the wider concept of how to handle
CPAN-first dual-life modules in general. Sometimes core changes will
cause breakages in unit tests in bundled modules. For core-first
modules we can easily fix these, but CPAN-first ones are more
difficult. We didn't really reach any firm conclusions - it's something
to look into more another day.


## In Flight for 5.36

We looked at what's still up in the air for 5.36. There's only really
one main question left: What to do about the "created_as_..." functions
for builtin.
  RFC PR:  https://github.com/Perl/RFCs/pull/13
  Code PR: https://github.com/Perl/perl5/pull/19524

We spent a while discussing this and got somewhat sidetracked into
thoughts of type systems and why this isn't one, but after reigning
that back in we got back to realising that the main point of contention
with this one is simply the name. We also are mildly concerned that
some of the discussion might really be people wanting the functions to
be something other than what they're intended for. A lot of the
discussion on the mailing list was in reference to the Pre-RFC post,
rather than the actual RFC or the subsequent code PR that was written
to create it. Paul will post to the list again to try to narrow down
the discussion on this one back to the specific use-case intended here,
with particular reference to its documentation that draws attention to
this. With any luck we can get some agreement in time to get it in for
5.36.


## Release Schedule and Release Manager for 5.36

Speaking of, we then looked at what the release schedule looks like for
the leadup to 5.36. Sawyer has volunteered to do 5.35.10, and Rik will
do 5.36. We don't yet have a volunteer for the final development
release, 5.35.11, due on Wednesday 20th April. Would someone like to
volunteer?


## Blockers

With only a few minutes left, we moved on to discuss the set of
release-blocker bugs we currently have. The github "BBC" label had (at
time of discussion) 30 issues relating to it.

  https://github.com/Perl/perl5/issues?q=is%3Aissue+is%3Aopen+label%3ABBC

However, many of them have a (negative?) tag called "non-5.36-blocker";
once those are removed only 11 remained. We also seem to lack an actual
"Blocker" tag, though there is one called "Release Blocker", but it
only has one issue. Rik will post an email to discuss a workflow for
managing these.


-- 
Paul "LeoNerd" Evans

leonerd@leonerd.org.uk      |  https://metacpan.org/author/PEVANS
http://www.leonerd.org.uk/  |  https://www.tindie.com/stores/leonerd/



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