Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
00:00 reportable6 left 00:02 reportable6 joined, ChanServ sets mode: +v reportable6, kiwi_65 joined 00:09 leont left
Geth rakudo/master: 8 commits pushed by (Vadim Belman)++ 00:09
roast: 7b3cce6ed9 | (Vadim Belman)++ | 4 files
Add tests for CORE::v6<rev> pseudo-packages

Test for namespaces referring their respective COREs and for valid multi-dispatching over symbols from different COREs.
perl6/problem-solving#80		 00:11
roast: e412b6f75e | (Vadim Belman)++ (committed using GitHub Web editor) | 4 files
Merge pull request #567 from vrurg/problem-solving-80

Add tests for CORE::v6<rev> pseudo-packages
Test for namespaces referring their respective COREs and for valid multi-dispatching over symbols from different COREs.
  - perl6/problem-solving#80
  - rakudo/rakudo#3112
synopsebot RAKUDO#3112 [closed]: github.com/rakudo/rakudo/pull/3112 [6.e] Implement CORE::v6<lang-rev> pseudo-packages
00:37 RageGauge joined
Geth roast: vrurg++ created pull request #568:
Fix for incorrect number of tests		 00:39
roast: 021bca6cd0 | (Vadim Belman)++ | S02-names/pseudo-6e.t
Fix for incorrect number of tests

Reverting a left-over of debugging process.
roast: 83adc85292 | (Vadim Belman)++ (committed using GitHub Web editor) | S02-names/pseudo-6e.t
Merge pull request #568 from vrurg/problem-solving-80

Fix for incorrect number of tests
kiwi_65 To the channel: timotimo has confirmed that my supposed "dramatic bug" significant DoS is no such thing, just a plain old bug. Sorry about the false alarm. 00:41
RageGauge I'm the one who found the bug, glad to know it's not something I need to lose sleep over. 00:42
timotimo thanks for kicking our butt regarding vulnerability report infrastructure/plans in any case :) 00:45
vrurg github.com/perl6/6.d-prep/issues/10 – can somebody confirm that this was implemented? So far, scannng through Actions doesn't reveal any 'dynamic' alongside to $_ 00:47
timotimo ah, yes, $_ is no longer dynamic in 6.d 00:48
vrurg Ok, I'm closing the ticket. Asking because there is a test for `$CALLER::_` and I wonder wether it shall be dropped altogether or modified for `CALLER::MY::`? 00:49
I think dropping it makes more sense because those aspects are covered by S02-names/pseudo*.t 00:50
RageGauge I think this is the first time I've ever come across a legitimate bug in a language, every other time it's been a bug in my code. Since I'm already here, thank you, to all of you who have made Perl6. I've used and loved Perl 5 for over a decade. Learning Perl6 has been like falling in love again. 00:51
timotimo \o/ 00:52
vrurg RageGauge: oh know this feeling. :) 00:53
*I know
01:06 MasterDuke left 01:16 travis-ci joined
travis-ci Rakudo build passed. Vadim Belman 'Merge pull request #3112 from vrurg/problem-solving-80 01:16
travis-ci.org/rakudo/rakudo/builds/574096947 github.com/rakudo/rakudo/compare/d...8859089521
01:16 travis-ci left 01:19 lucasb left 02:09 RageGauge left 03:07 squashable6 left, squashable6 joined, ChanServ sets mode: +v squashable6 03:10 kiwi_65 left
Kaiepi bisectable6, use nqp; class FooHOW does Metamodel::MethodContainer { method new_type() { my $meta := self.new; my $foo := nqp::settypehll(nqp::newtype($meta, "Uninstantiable"), "perl6"); $foo } }; my $foo := FooHOW.new_type; say $foo.^methods 04:29
bisectable6 Kaiepi, Bisecting by output (old=2015.12 new=6a88590) because on both starting points the exit code is 1
Kaiepi, bisect log: gist.github.com/343768e938081c5a60...87714afaa7 04:30
Kaiepi, (2016-03-02) github.com/rakudo/rakudo/commit/fd...7e379b5038
Kaiepi weird
i'm guessing perl 6 classes can't do nqp roles or something? 04:32
wait, that's not right, Metamodel::Naming works just fine 04:37
05:33 robertle left 06:00 reportable6 left 06:01 reportable6 joined 06:19 fake_space_whale left 06:41 rba left 06:42 rba joined 06:45 rba left, rba joined 07:20 leont joined 07:45 leont left
kawaii AlexDaniel: started a Blin run this morning :) 08:15
08:16 robertle joined
Kaiepi kawaii lemme know if any networking modules end up breaking 08:19
08:47 cygx joined
cygx o/ 08:47
I neew some help manually getting an object into and then again out of a serialization context 08:48
the code in question: gist.github.com/cygx/5fa39ff2c3a12...5ac16d6aca
what'S missing? 08:49
do I need to load the module that contains $bc.WHAT?
jnthn cygx: Yes 08:57
Or at least make sure something did so
cygx can that happen in the mainline, or do I have to put that into QAST::CompUnit.pre_deserialize? 08:59
Geth rakudo: 0dd6573a45 | (Jonathan Worthington)++ | src/vm/moar/spesh-plugins.nqp
Add a missing guard to the Nil assignment plugin

Fixes a regression introduced by c7056f2071.		 09:00
roast: 1f526efa70 | (Jonathan Worthington)++ | S02-types/nil.t
Test to cover assign works after assigning Nil

Covers a bug where we ended up fixated on the Nil defaulting behavior.		 09:03
jnthn cygx: pre_deserialize; that runs before the mainline 09:04
cygx that seems to have done something, there's a new error: No current compiling SC 09:06
lizmat .tell SmokeMachine I've already suggested many alternatives, "London Perl Community Workshop" being only one of them 09:09
tellable6 lizmat, I'll pass your message to SmokeMachine
lizmat .tell SmokeMachine or to take the Barcelona example "London Perl and Friends Workshop" 09:10
tellable6 lizmat, I'll pass your message to SmokeMachine
lizmat .tell SmokeMachine see also blogs.perl.org/users/swiss_perl_wor...uture.html
tellable6 lizmat, I'll pass your message to SmokeMachine
SmokeMachine lizmat: yes, I saw that... but going there with the original name wouldn't it be good to convince the people to change the name for the next ones? 09:12
lizmat I've already talked extensively with some of the organizers... and that was not enough. Some statement needs to be made
SmokeMachine lizmat: I'm not trying to start any kind of flame war: does the name of the event is really that important? If it would like to receive perl6 talks, isn't it enough? 09:16
lizmat it's about how it is seen in the world, *before* knowing what type of talks will be given
"Perl" in the world stands for "Perl 5"
example: look at the schedule of PerlCon and count how many talks have just "Perl" in their title, and are about "Perl 5" 09:17
and this is *inside* the echo chamber!
SmokeMachine lizmat: I'm asking because changing name can be a long process... 09:18
lizmat SmokeMachine: why ?
SmokeMachine lizmat: just see how long it is taking the process to change perl6 name 09:19
lizmat: first is the discussion of if it should change
lizmat: than what should be the new name
lizmat: etc...
lizmat I've expressed my opinion to the LPW organizers many months ago already
I feel it was not taken seriously 09:20
SmokeMachine lizmat: ok, I got it...
lizmat the only way I can indicate that I'm serious about it, is to vote with my feet
SmokeMachine lizmat: yes... that make sense 09:21
s 09:22
tellable6 2019-08-20T09:09:34Z #perl6-dev <lizmat> SmokeMachine I've already suggested many alternatives, "London Perl Community Workshop" being only one of them
2019-08-20T09:10:11Z #perl6-dev <lizmat> SmokeMachine or to take the Barcelona example "London Perl and Friends Workshop"
2019-08-20T09:10:39Z #perl6-dev <lizmat> SmokeMachine see also blogs.perl.org/users/swiss_perl_wor...uture.html
lizmat hmnmm.... tellable still a bit on the blink 09:23
SmokeMachine yes... it seems to be...
kawaii hmm, so the Star Makefile references a file `tools/build/gen-version.pl` which doesn't seem to exist within the `rakudo` repository 09:24
lizmat vrurg probably moved it somewhere else 09:25
kawaii lizmat: github.com/rakudo/rakudo/commit/c8...2f97770fb8 09:27
:)
well, I have a tarball, time to see if it works! 09:32
hmm, looks like more has changed in the rakudo process than I first thought www.irccloud.com/pastebin/uHvhCbq7/ 09:35
lizmat m: sub a() { "foo" }; dd "&&a()"
camelia "foo"
lizmat surprised that this works... now lets add some more &
m: sub a() { "foo" }; dd "&&&&&&&&&&&&&&&&&&&&&&&&&&&a()"
camelia "foo"
lizmat m: sub a() { "foo" }; dd "&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&a()" 09:36
camelia "foo"
lizmat hmmmm...
now, drop the "a()"
m: sub a() { "foo" }; dd "&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&"
09:40 camelia left 09:42 camelia joined
camelia (timeout) 09:43
09:43 ChanServ sets mode: +v camelia
jnthn oops :) 09:56
lizmat same for % and @ BTW 09:59
m: sub a() { "foo" }; dd "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$"
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of $$ variable; in Perl 6 please use $*PID
at <tmp>:1
------> 3$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$"7⏏5<EOL>
lizmat m: sub a() { "foo" }; use isms <Perl5>; dd "$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$"
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in double quotes; couldn't find final '"' (corresponding starter was at line 1)
at <tmp>:1
------> 3$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$"7⏏5<EOL>
expecting …
lizmat hmmm... 10:00
in any case, feels we need a token for <sigil>+ followed by a non-alpha ?
jnthn Well, it's just a contextualizer 10:02
lizmat m: my $a = 42; say "$$$$$$$$$$$$$$a" # should we even allow for that ?
camelia 42 10:03
jnthn Sure, it's fine, though useless
lizmat m: my @a = 42,666; say "@@@@@@@@@@@@@@@@@@@@@@@@@@a[]" # should we even allow for that ?
camelia 42 666
jnthn It's just short for @(@(@(....@a[]...)))
lizmat it's a potential vector of attack :-(
jnthn ? 10:04
If you're already getting to choose what code to run, you've already won.
Still worth looking into the parser blow-up, of course. 10:07
Since it's a poor language user experience
lizmat indeed
10:08 travis-ci joined
travis-ci Rakudo build failed. Jonathan Worthington 'Add a missing guard to the Nil assignment plugin 10:08
travis-ci.org/rakudo/rakudo/builds/574240012 github.com/rakudo/rakudo/compare/6...d6573a45f6
10:08 travis-ci left 10:12 samebchase joined 10:27 squashable6 left 10:32 squashable6 joined
cygx this thing I'm trying to do would be a lot simpler if there was a bootbytearray... 10:37
I'll make do without, but is that something we might want to have? 10:38
jnthn I'm not really keen on having more BOOTWhatever 10:39
cygx is there another way to pass around some bytes that doesn't drag in a whole other bunch of stuff via its type object? 10:42
11:19 cygx left 11:30 cygx joined
Geth problem-solving/path-to-raku: b9c597b3d7 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Add mention of pragmas and other places needing adaptation.
 11:40
12:00 reportable6 left 12:01 reportable6 joined
timotimo kawaii: yeah, rakudo/tools/build/NQP_REVISION has moved away, it's now in rakudo/build/templates/ or something like that 12:15
cygx not sure how to debug this: gist.github.com/cygx/b92999e516932...69c40f8a98
somehow, there's still a reference to *something* outside the SC 12:16
timotimo cygx: you should be able to just quickly create a vmarray of uint8 with nqp::newtype
is there a reason you don't want a native-typed array there? 12:17
cygx timotimo: because that is associated with a p6-level type object? 12:18
timotimo you're already creating an array type yourself though 12:19
13:01 lucasb joined 13:06 cygx left 13:45 |Tux| left 13:50 |Tux| joined
Geth problem-solving/path-to-raku: 87d43430ed | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Separate next release from language release
 14:31
14:32 brrt joined
Geth problem-solving/path-to-raku: 535d778c66 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Specify when "Perl 6" should be referenced
 14:34
problem-solving/path-to-raku: dd6054cddd | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
NQP stays Not Quit Perl
 14:37
problem-solving/path-to-raku: d14804e6b0 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Elaborate a bit about changing external references
 14:40
problem-solving/path-to-raku: ce377bd9c6 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Remove mention of marketing repo

We need new material, that's the important thing. Of course that will wind up in a repo.
FWIW, I mentioned a new repo because I thought we had lost the repo with Zoffix' departure.		 14:45
14:46 [TuxCM] joined
Geth problem-solving/path-to-raku: 2f5794b458 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Elaborated on special attention
 14:48
problem-solving/path-to-raku: 1b89762c28 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Elaborate on repo name changes and documentation changes
 14:53
¦ problem-solving: AlexDaniel assigned to JJ Issue Specify documentation URLs github.com/perl6/problem-solving/issues/93 14:55
problem-solving/path-to-raku: 2b237b5ca8 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Elaborate on effects on running sub-projects
 14:57
[TuxCM] Rakudo version 2019.07.1-186-g0dd6573a4 - MoarVM version 2019.07.1-117-g7f8283424
csv-ip5xs0.684 - 0.705
csv-ip5xs-205.062 - 5.224
csv-parser20.846 - 21.380
csv-test-xs-200.428 - 0.439
test6.790 - 7.015
test-t1.713 - 1.728
test-t --race0.805 - 0.878
test-t-2029.391 - 30.025
test-t-20 --race9.093 - 9.107
 15:01
15:04 |Tux| left 15:05 robertle left 15:08 cygx joined, |Tux| joined
cygx still can't get even a minimal example to work :( 15:09
cf gist.github.com/cygx/b92999e516932...69c40f8a98
15:21 brrt left 15:26 pamplemousse joined 15:27 [TuxCM] left
Geth problem-solving/path-to-raku: f3d057acc1 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Elaborate a bit on community healing
 15:34
problem-solving/path-to-raku: 6b0041002e | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Add social media section
 15:43
¦ problem-solving: JJ self-unassigned X::MyModule::Foo or MyModule::X::Foo ? github.com/perl6/problem-solving/issues/57 15:48
¦ problem-solving: AlexDaniel self-assigned X::MyModule::Foo or MyModule::X::Foo ? github.com/perl6/problem-solving/issues/57 15:51
15:52 robertle joined 16:10 brrt joined 16:28 cygx left 16:31 chloekek joined 16:39 brrt left
rba I've installed a znc (irc boucner; wiki.znc.in/ZNC). Does one like to jump on? Still do testing though. Yet this is my second installation and the first one was rocket stable, yet I needed it installed on a system with gcc >= 4.9 for a znc push module. 16:43
Let me know your nick and I will add it as a user. 16:44
vrurg Writing in QAST feels like writing in assembler... 16:54
17:15 brrt joined
Geth problem-solving/path-to-raku: 3118a7eb71 | (Elizabeth Mattijsen)++ | solutions/language/PATH-TO-RAKU.md
Add historical matters
 17:17
[Coke] is anyone looking at a .net framework backend for NQP? 17:19
er, Core. 17:20
brrt not me atm 17:21
17:25 [TuxCM] joined 17:39 brrt left 18:00 reportable6 left 18:04 reportable6 joined
AlexDaniel lizmat: no, I don't think it's going to work this way 18:14
lizmat: github.com/ugexe/zef/search?utf8=%...&type=
lizmat looks like keeping $*PERL working, should be enough ? 18:16
AlexDaniel or maybe we should delay all deprecation messages until 6.f 18:17
lizmat and keep supporting "perl" in META6.json
that feels.... wrong
feels like deprecating something after it has not been documented as deprecated
Perl 5 has bad experience with that wrt deprecating smart match 18:18
AlexDaniel lizmat: sure, but it's an actual problem 18:27
lizmat: if you drop a deprecation on users in one language release, any module maintainer who has deprecated code will have to add code for backward compatibility 18:28
which is not bad, but it's work
whereas if you do it slower, then in some cases it'd be possible to get away with just changing the code to the new stuff and that's it 18:29
18:40 MasterDuke joined
MasterDuke looks like we need nqp/rakudo bumps to pull in brrt++'s floating point jit support 18:41
Geth nqp: 49fdab2e9f | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/MOAR_REVISION
[MoarVM Bump] Brings 45 commits

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...g7f8283424 7f8283424 [JIT] Handle NaN in FP branching comparisons 6c6225534 [JIT] Fix floating point comparisons 7f8061a5b [JIT] Support floating point comparisons 27e9f979b [JIT] Functions that return floating point values ... (41 more lines)		 18:43
rakudo: cd8463a853 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/templates/NQP_REVISION
[NQP Bump] Brings 4 commits

NQP bump brought: github.com/perl6/nqp/compare/2019....g49fdab2e9 49fdab2e9 [MoarVM Bump] Brings 45 commits 68da2436a [examples/rubyish] - bring up-to-date with current NQP 413c257c0 Remove obsolete pumpking mention 8778cff06 [js] Fix serialization/deserialization of some integers
AlexDaniel MasterDuke: thanks ;)
19:01 chloekek left 19:44 Ven`` joined 19:53 travis-ci joined
travis-ci Rakudo build passed. Aleks-Daniel Jakimenko-Aleksejev '[NQP Bump] Brings 4 commits 19:53
travis-ci.org/rakudo/rakudo/builds/574501714 github.com/rakudo/rakudo/compare/0...8463a85351
19:53 travis-ci left
vrurg timotimo: what name would you suggest for nqp::p6 op which returns caller's CORE context? I'm thinking of nqp::p6clientcorectx because it acts as CLIENT:: pseudo-package. 19:57
20:13 pamplemousse left 20:14 Ven`` left 20:21 Ven`` joined 20:31 Ven`` left 20:39 chloekek joined
dogbert17 vrurg: wrt M#1155 exactly what cmdline were you running? 21:09
synopsebot M#1155 [open]: github.com/MoarVM/MoarVM/issues/1155 SIGSEGV with heap profiling.
Geth rakudo: usev6++ created pull request #3130:
Add some concreteness/definedness checks		 21:10
tbrowder .ask nine are you available for some inline:perl5 help? 21:21
tellable6 tbrowder, I'll pass your message to nine
tbrowder .tell nine see nonworking p6 code at gist gist.github.com/tbrowder/890714b6c...e2ca088e24 21:22
tellable6 tbrowder, I'll pass your message to nine
21:51 leont joined 21:55 leont left 21:59 robertle left 22:12 chloekek left
vrurg dogbert17: there was a script on the command line, otherwise – nothing. so, eventually, it was 'perl6 --profile --profile-kind=heap somescript.p6' 22:35
dogbert17: it was probably fixed, I can't reproduce. 22:36
dogbert17 vrurg: that's why I was asking, timotimo committed a fix and I was curious if that fixed your problem 22:37
vrurg dogbert17: Close it. 22:39
thanks!

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!