Perl 5.34.0 released [LWN.net]

Version 5.34.0 of the Perl language has been released. "Perl 5.34.0 represents approximately 11 months of development since Perl 5.32.0 and contains approximately 280,000 lines of changes across 2,100 files from 78 authors." See this page for a list of changes; they include a new try/catch syntax, a new octal syntax, and many improvements to various modules.

From:		Sawyer X <xsawyerx-AT-gmail.com>
To:		Perl5 Porters <perl5-porters-AT-perl.org>
Subject:		Perl 5.34.0 is now available!
Date:		Thu, 20 May 2021 22:14:37 +0200
Message-ID:		<CAMvkq_RA=sgZGiU6np51m8ueAfF0vA1bJd6Vm9F8EqBECW3zYw@mail.gmail.com>
Cc:		noc-AT-metacpan.org

  Our greatest power as nations and individuals is not the ability to
employ assault weapons, suicide bombers, and drones to destroy each
other.
  The greater more creative powers with which we may arm ourselves are
grace and compassion sufficient enough to love and save each other.

    -- Aberjhani

We are thrilled to announce perl v5.34.0, the first stable release of
version 34 of Perl 5.

You will soon be able to download Perl 5.34.0 from your favorite CPAN
mirror or find it at:

https://metacpan.org/release/XSAWYERX/perl-5.34.0/

SHA1 digests for this release are:

  870b4e1ffe33f94a4996a4acbbcebb8422c5e73a  perl-5.34.0.tar.gz
  d461e206a1dca5e79d39e77debf0b564f6d77d37  perl-5.34.0.tar.xz

You can find a full list of changes in the file "perldelta.pod"
located in the "pod" directory inside the release and on the web at

https://metacpan.org/pod/release/XSAWYERX/perl-5.34.0/pod...

Perl 5.34.0 represents approximately 11 months of development since
Perl 5.32.0 and contains approximately 280,000 lines of changes across
2,100 files from 78 authors.

Excluding auto-generated files, documentation and release tools, there
were approximately 150,000 lines of changes to 1,300 .pm, .t, .c and
.h files.

Perl continues to flourish into its fourth decade thanks to a vibrant
community of users and developers. The following people are known to
have contributed the improvements that became Perl 5.34.0:

Aaron Crane, Adam Hartley, Andy Dougherty, Ben Cornett, Branislav
Zahradník, brian d foy, Chris 'BinGOs' Williams, Christian Walde
(Mithaldu), Craig A. Berry, Dagfinn Ilmari Mannsåker, Dan Book, Daniel
Böhmer, Daniel Laügt, Dan Kogai, David Cantrell, David Mitchell,
Dominic Hamon, E. Choroba, Ed J, Eric Herman, Eric Lindblad, Eugene
Alvin Villar, Felipe Gasper, Giovanni Tataranni, Graham Knop, Graham
Ollis, Hauke D, H.Merijn Brand, Hugo van der Sanden, Ichinose Shogo,
Ivan Baidakou, Jae Bradley, James E Keenan, Jason McIntosh, jkahrman,
John Karr, John Lightsey, Kang-min Liu, Karen Etheridge, Karl
Williamson, Keith Thompson, Leon Timmermans, Marc Reisner, Marcus
Holland-Moritz, Max Maischein, Michael G Schwern, Nicholas Clark,
Nicolas R., Paul Evans, Petr Písař, raiph, Renee Baecker, Ricardo
Signes, Richard Leach, Romano, Ryan Voots, Samanta Navarro, Samuel
Thibault, Sawyer X, Scott Baker, Sergey Poznyakoff, Sevan Janiyan,
Shirakata Kentaro, Shlomi Fish, Sisyphus, Sizhe Zhao, Steve Hay, TAKAI
Kousuke, Thibault Duponchelle, Todd Rinaldo, Tomasz Konojacki, Tom
Hukins, Tom Stellard, Tony Cook, vividsnow, Yves Orton, Zakariyya
Mughal, Михаил Козачков.

The list above is almost certainly incomplete as it is automatically
generated from version control history. In particular, it does not
include the names of the (very much appreciated) contributors who
reported issues to the Perl bug tracker.

Many of the changes included in this version originated in the CPAN
modules included in Perl's core. We're grateful to the entire CPAN
community for helping Perl to flourish.

For a more complete list of all of Perl's historical contributors,
please see the AUTHORS file in the Perl source distribution.

We expect to release the next development version imminently. The next
major stable release of Perl should appear in the first half of 2022.

Stay safe,
In hugs and bugs,
Sawyer X.

to post comments

Perl 5.34.0 released

Posted May 22, 2021 19:49 UTC (Sat) by pebolle (guest, #35204) [Link] (20 responses)

brian d foy
Chris 'BinGOs' Williams
Christian Walde (Mithaldu)
Ed J
Hauke D
Sawyer X
vividsnow

I rather like perl and I have no doubt the above handles refer to smart people. But I'd really prefer it if they stopped imitating rock stars, Brazilian soccer players, writers, etc.

It's all a bit juvenile.

Perl 5.34.0 released

Posted May 23, 2021 1:19 UTC (Sun) by atai (subscriber, #10977) [Link]

Probably "Sawyer X" does not know the real names of these people.

Perl 5.34.0 released

Posted May 23, 2021 4:00 UTC (Sun) by sub2LWN (subscriber, #134200) [Link]

And "Sisyphus" :-) Once he discovered the fine art of code golf he realized he could push his boulder with a one-liner, and he's been contributing to CPAN ever since!!

Perl 5.34.0 released

Posted May 23, 2021 7:21 UTC (Sun) by flussence (guest, #85566) [Link]

On a scale of IBM policing which email addresses its employees are allowed to use in the kernel MAINTAINERS file to Oracle harassing Pidgin devs like they're a personal helpdesk, this identity-policing comment is way off to the right. I hope you're just speaking out of sheer ignorance and not actual bigotry.

Perl 5.34.0 released

Posted May 23, 2021 8:41 UTC (Sun) by oldtomas (guest, #72579) [Link]

"I rather like perl..."

If the above is your only problem with your favourite language, life seems to be gentle to you.

Snark aside: I strongly support anyone calling themselves the way they deem right and think I (or you) have no right to second guess that decision.

Perl 5.34.0 released

Posted May 23, 2021 9:42 UTC (Sun) by niner (guest, #26151) [Link] (1 responses)

So what's wrong with Brian D. Foy name? Abbreviating the middle name seems to be very common in the US. I find it a bit harsh to tell the Foys that naming their son Brian was juvenile.

Perl 5.34.0 released

Posted May 23, 2021 10:12 UTC (Sun) by pebolle (guest, #35204) [Link]

> So what's wrong with Brian D. Foy name?

You mean brian d foy?

Perl 5.34.0 released

Posted May 23, 2021 11:54 UTC (Sun) by amacater (subscriber, #790) [Link] (3 responses)

Some people use IRC nicks, some people may have had that same identity since the dawn of the Internet. Some people may want to disambiguate themselves - in England, there are a lot of people with the family name Smith. A very few may have reasons to cut their professional career away from their Free software career and use a psudonym: some people may just fancy a different name. A few still use university IDs from, perhaps, 25 years ago because that's how they got known.
[And some of them are here on LWN: LWN is harder because you _can't_ tie contributors here to names in other projects conclusively or straightforwardly.]

Provided the identity is consistent and the code contributions and fixes can be traced to a single individual or team in case of problem: where's the harm? It's a little like pronouns: address people the way they want to be called and don't stress about it on their behalf? brian d foy may just have found that was convenient and more readable than Brian D. Foy and quicker and easier to type.

amacater@[wherever] for the last 28 years (and it was entirely by chance that that wasn't something random/mythological - galactic as part of my most common email address was because I was slightly interested in astronomy .... )
I left university in 1984 just before PCs arrived on the scene otherwise I might have been one of those still using a university
ID string.

Perl 5.34.0 released

Posted May 23, 2021 12:40 UTC (Sun) by pebolle (guest, #35204) [Link] (2 responses)

> where's the harm?

perl seems currently to be stuck in a, well, Debianesque phase. Heated discussions, people stepping down, elections, people taking the time to write and read very, very long messages to lists. Some of it was reported on LWN. Perhaps all. The totally botched perl5 to perl6 transition (for lack of better term) probably fuels it quite a bit. If this continues too long it might turn into a farce. Perhaps it already has.

So my thinking is that accepting, or perhaps even encouraging, nicks is a sign that people are investing emotionally into their roles in a group. The nick is far more than just an account name. It certainly looks like that to me.

So the nicks remind me that we might be looking here at a shrinking group of people getting ever more erratic. (That must have happened before on the internet.) Let's hope this group can prove me wrong.

Perl 5.34.0 released

Posted May 23, 2021 17:54 UTC (Sun) by dvdeug (subscriber, #10998) [Link] (1 responses)

In the creation of software, or pretty much anything, either "people are investing emotionally into their roles in a group" or people are getting a paycheck or similar compensation. As long as nobody is paying for Perl development, you should hope they're emotionally invested.

Perl 5.34.0 released

Posted May 24, 2021 13:26 UTC (Mon) by nix (subscriber, #2304) [Link]

Quite. And it's odd, really... the *other* poor reason people give for banning pseudonyms and forcing legal names everywhere is that the pseudonyms encourage trolling and bad behaviour because, uh, the names aren't "real" so people aren't invested in them and can burn them at whim.

Both these things seem unlikely to be true at the same time.

Perl 5.34.0 released

Posted May 24, 2021 17:48 UTC (Mon) by jkingweb (subscriber, #113039) [Link] (8 responses)

Perhaps you missed the next paragraphs which say that the list was auto-generated from version-control history. This is how those people chose to identify themselves when they set up their software. An unknown number could be drive-by hobbyists or professionals toiling in their free time with no regularly involvement with Perl.

As a hobbyist I sign my code "J. King" because I'm not particularly fond of my given name; does that make me unprofessional? It's still my name, and even if I used my full name I know of at least six people more famous than me with the same name, so it wouldn't necessary help to identify me better.

Also, you'll find pseudonyms all over the Internet; Perl is by no means uncommon here. Anyway, "vividsnow" is quite poetic, I think. It would be a shame to lose such a conduit of expression.

Perl 5.34.0 released

Posted May 24, 2021 23:24 UTC (Mon) by pebolle (guest, #35204) [Link] (7 responses)

> It would be a shame to lose such a conduit of expression.

Using these juvenile handles while fixing uses after free, double frees, typos, off by one errors or adding new syntax, etc., etc., is a rather odd way to express oneself.

This is coding. Not writing or singing. (Even most artists use their real name.)

Perl 5.34.0 released

Posted May 24, 2021 23:38 UTC (Mon) by dskoll (subscriber, #1630) [Link] (4 responses)

This is a really odd thing to get hung up about. Is pebolle your real name?

Perl 5.34.0 released

Posted May 24, 2021 23:54 UTC (Mon) by pebolle (guest, #35204) [Link] (2 responses)

No, it's my account name at lwn.net and a few other sites, ISP's etc. (Does lwn.net even allow dots and spaces in a username?) If you're curious you can do:

git log --author pebolle

on some random free software repositories to find my real name. (I'm sure it's much closer to my real name than Sawyer X is to his.) You can even "reveal" my real name here, but that would be a bit at odds with the sentiment of your comment.

Perl 5.34.0 released

Posted May 25, 2021 0:09 UTC (Tue) by pebolle (guest, #35204) [Link] (1 responses)

> If you're curious you can do:
>
> git log --author pebolle
>
> on some random free software repositories to find my real name.

Searching on github.com works great to do that. I just checked.

Perl 5.34.0 released

Posted May 25, 2021 0:59 UTC (Tue) by dskoll (subscriber, #1630) [Link]

On Github, I easily found vividsnow and xsawyerx. So again... what's the issue? These are all real people. The fact that their online names don't conform to your notion of what a conventional name is supposed to look like is really unimportant.

Perl 5.34.0 released

Posted May 26, 2021 15:31 UTC (Wed) by oldtomas (guest, #72579) [Link]

Exactly my feeling. Those folks are doing a great job, and judging that by some ersatz "professionalism" criteria is -- unprofessional itself.

It's as if I decided whether someone is good at prime number theory based on whether that person wears a tie or not.

Perl 5.34.0 released

Posted May 25, 2021 7:59 UTC (Tue) by mpr22 (subscriber, #60784) [Link] (1 responses)

> Using these juvenile handles

Aaaaaand you're using that word again.

Hey Wikiquote, you got something for me?

"When I was ten, I read fairy tales in secret and would have been ashamed if I had been found doing so. Now that I am fifty I read them openly. When I became a man I put away childish things, including the fear of childishness and the desire to be very grown up." — C. S. Lewis, "On Three Ways of Writing for Children" (1952)

Perl 5.34.0 released

Posted May 25, 2021 17:14 UTC (Tue) by Wol (subscriber, #4433) [Link]

What about my nic? Childish, or (based on) my real name?

BOTH, actually.

Cheers,
Wol

Perl 5.34.0 released

Posted May 28, 2021 13:04 UTC (Fri) by jschrod (subscriber, #1646) [Link]

I think calling oneself "pebolle" and complaining about other nicks is the summit of juvenility -- that height cannot be reached by people like brian d foy who does substantial work under this name on Perl for more than 2 decades.

Perl 5.34.0 released

Posted May 24, 2021 13:49 UTC (Mon) by nix (subscriber, #2304) [Link] (2 responses)

I'm a little surprised there's so little disruptive stuff in this Perl release. It must be the quietest release in many years (and it's not a bad thing to have one of those now and again). I wonder if they broke ABI for any reason but habit this time :)

I guess the perl5-porters time was all spent with arguing over Perl 7. To good end, though: the new Perl 7 plan (stay compatible but allow a 'use perl 7;' to switch on a bunch of new stuff) seems much better than the previous plan, which would have inflicted compatibility pain on every single user and probably forced everyone to drag around both Perl 5 and 7 for many years (and God knows what it would have meant for programs embedding Perl). The new plan seems much more like the approach previous Perl 5 versions took, and much more like the editions approach taken by Rust, which was a demonstrable success and allows all sorts of arbitrarily incompatible changes without breaking anything whatsoever for existing users, with a tiny constant cost to people using the new stuff (say 'use perl 7;' or perl 8 or whatever).

Perl 5.34.0 released

Posted May 25, 2021 16:11 UTC (Tue) by simcop2387 (subscriber, #101710) [Link] (1 responses)

The ABI is not expected to be stable even among point releases with perl. This means that for every upgrade you have to rebuild any native code that you're using. This is because of the way that XS modules (as perl cause them) are deeply tied to the internal functions, state, and behavior of the perl interpreter. In fact they're able to change how parsing, lexing, and other internal bits work. That's both a good and bad thing since it allows a lot of really neat behavior but it prevents there being a stable ABI since any change or fix usually can cause a change in any internal structure of the interpreter.

Perl 5.34.0 released

Posted May 28, 2021 20:03 UTC (Fri) by nix (subscriber, #2304) [Link]

ABI breaks are at the very least *extremely minimal* in point releases and a strong attempt is made to minimize them. I can't recall ever running into one, or at least not into the *effects* of one. (I think there might have been one involving regexes back in the 5.24 era? But that's so long ago I can't really remember).

Debian and several other distros nail the major.minor version number into the privlib/vendorlib directories for basically this reason. It is routine for distros to rebuild Perl modules only when non-point releases come out, so if there were routine significant ABI breaks I suspect the distros would have got a bit unhappy by now.