#perl6-dev on 12 July 2019 - Raku Programming Language Log

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.

jnthn

m: dd &await.candidates; EVAL 'use v6.c; dd &await.candidates'

00:00

camelia

(multi sub await { #`(Sub|38935760) ... }, multi sub await (Any:U $x) { #`(Sub|38935912) ... }, multi sub await (Awaitable:D \a) { #`(Sub|38936064) ... }, multi sub await (*@awaitables) { #`(Sub|38936216) ... })
(multi sub await { #`(Sub|38935760) .…

vrurg

I'm spectesting without EVAL loading CORE.

jnthn

m: dd &await.candidates.map(*.signature); EVAL 'use v6.c; dd &await.candidates.map(*.signature)'

camelia

(:(), :(Any:U $x), :(Awaitable:D \a), :(*@awaitables)).Seq
(:(), :(Any:U $x), :(Awaitable:D \a), :(*@awaitables)).Seq

jnthn

Those should be different if it actually wedged 6.c stuff in there

00:01

vrurg

Ah, it doesn't reveal clearly. It's my fix for CORE loading which does it.

jnthn

m: sub foo() { EVAL 'whenever Supply.interal(1) { }' }; supply { foo() }

00:02

camelia

( no output )

jnthn

m: sub foo() { EVAL 'whenever Supply.interal(1) { }' }; supply { foo() }.tap

camelia

5===SORRY!5=== Error while compiling /home/camelia/EVAL_0
Cannot have a 'whenever' block outside the scope of a 'supply' or 'react' block
at /home/camelia/EVAL_0:1
------> 3whenever7⏏5 Supply.interal(1) { }

jnthn

m: sub foo() { EVAL 'use v6.c; whenever Supply.interal(1) { }' }; supply { foo() }.tap

camelia

No such method 'interal' for invocant of type 'Supply'. Did you mean 'interval'?
in block <unit> at EVAL_0 line 1
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1

jnthn

Hm, so it applies the 6.c parser semantics but without the 6.c setting, in whatever commit is running here. That's...not liable to end so well. :)

00:03

vrurg

m: use v6.d; use nqp; say nqp::p6box_s(nqp::getcomp(q<perl6>).language_version()); EVAL q|use v6.c; use nqp; say nqp::p6box_s(nqp::getcomp(q<perl6>).language_version())|; say nqp::p6box_s(nqp::getcomp(q<perl6>).language_version());

camelia

6.d
6.c
6.c

vrurg

I think everything would work smoothly if EVAL creates own instance of compiler. But I don't know if this would work.

00:05

jnthn

m: await start for ^1000 { EVAL q|use v6.c; use nqp; die unless '6.c' eq nqp::p6box_s(nqp::getcomp(q<perl6>).language_version())| }, start { for ^1000 { EVAL q|use v6.d; use nqp; die unless '6.d' eq '' nqp::p6box_s(nqp::getcomp(q<perl6>).language_version())|| }

camelia

5===SORRY!5=== Error while compiling <tmp>
Missing required term after infix
at <tmp>:1
------> 3getcomp(q<perl6>).language_version())|| 7⏏5}
expecting any of:
prefix
term

jnthn

00:06

camelia

An operation first awaited:
in block <unit> at <tmp> line 1

Died with the exception:
5===SORRY!5=== Error while compiling /home/camelia/EVAL_1
Two terms in a row
at /home/camelia/EVAL_1:1
------> 3se v6.d; use nqp…

jnthn

huh

00:07

vrurg

m: await ^1000 .map: {start '{my $x}'.EVAL}; say "all good"

camelia

===SORRY!===
Decoder may not be used concurrently

vrurg

This is zoffix sample. But in his case the error was different.

jnthn

But yeah, the current language can't be stuck as a property in Perl6::Compiler

Perl6::Compiler isn't meant to be stateful like that

00:08

Should really be hung off the language braid at compile time...and maybe annoted on UNIT for runtime, though that brings back the whole "what is UNIT in EVAL" kind of question.

Sleep time for me, alas. 'night o/

00:10

vrurg

Unfortunately, that's a kind of question beyond my knowledge too. g'night! o/

AlexDaniel

.tell jnthn what do you think about this commit? The same test is in 6.c-errata (so it needs a change also). github.com/perl6/roast/commit/d60f...0250243809

00:20

yoleaux

AlexDaniel: I'll pass your message to jnthn.

Geth

roast/6.c-errata: e96297869f | (Vadim Belman)++ (committed by Aleks-Daniel Jakimenko-Aleksejev) | S14-roles/parameterized-mixin.t
Remove invalid fudged tests

These two tests apparently must not pass.

00:36

00:52 pamplemousse joined

vrurg

.tell jnthn I think I have R#3040 finalized and ready for merge. Would you please have another look at it?

01:06

yoleaux

vrurg: I'll pass your message to jnthn.

synopsebot

R#3040 [open]: github.com/rakudo/rakudo/pull/3040 Implement support for defining new symbols in CORE.e.setting

01:32 epony left

AlexDaniel

and another, much simpler one: github.com/rakudo/rakudo/issues/3046

02:11

02:21 epony joined 02:52 pamplemousse left 04:52 vrurg left 04:53 vrurg joined 04:54 tbrowder left, tbrowder joined 06:03 squashable6 left 06:08 squashable6 joined, ChanServ sets mode: +v squashable6 06:10 squashable6 left 06:13 squashable6 joined, ChanServ sets mode: +v squashable6 06:39 patrickb joined

[Tux]

modules.perl6.org/search/?q=CSV

06:40

1. modules.perl6.org/dist/Text::CSV:cpan:HMBRAND

2. modules.perl6.org/dist/Text::CSV:gi...nd%20(Tux)

How do I force the second with zef?

$ zef install Text::CSV github

===> Searching for: github

===> Updating cpan mirror: raw.githubusercontent.com/ugexe/Pe...cpan1.json

===> Updating p6c mirror: raw.githubusercontent.com/ugexe/Pe.../p6c1.json

===> Updated p6c mirror: raw.githubusercontent.com/ugexe/Pe.../p6c1.json

===> Updated cpan mirror: raw.githubusercontent.com/ugexe/Pe...cpan1.json

No candidates found matching identity: github

$ zef install Text::CSV github

===> Searching for: github

===> Updating cpan mirror: raw.githubusercontent.com/ugexe/Pe...cpan1.json

===> Updating p6c mirror: raw.githubusercontent.com/ugexe/Pe.../p6c1.json

===> Updated p6c mirror: raw.githubusercontent.com/ugexe/Pe.../p6c1.json

===> Updated cpan mirror: raw.githubusercontent.com/ugexe/Pe...cpan1.json

No candidates found matching identity: github

$ zef install Text::CSV github_ 'github:H.Merijn Brand (Tux)'

06:41

Don't understand identity: github:H.Merijn Brand (Tux)

lizmat, I Cc'd you on a mail that describes this problem in Dutch to someone moving to perl6 :)

06:43

06:57 robertle_ joined 07:41 Tux__ joined, |Tux| left 08:22 Tux__ is now known as |Tux|

|Tux|

Rakudo version 2019.03.1-683-g20e74837a - MoarVM version 2019.07-5-gdefe0a959

csv-ip5xs	0.703	-	0.716
csv-ip5xs-20	5.282	-	5.380
csv-parser	22.181	-	22.409
csv-test-xs-20	0.433	-	0.435
test	6.881	-	7.017
test-t	1.740	-	1.790
test-t --race	0.847	-	0.861
test-t-20	27.512	-	28.203
test-t-20 --race	9.255	-	9.280

08:22

10:22 AlexDaniel left, AlexDaniel joined

lizmat

Files=1275, Tests=108166, 207 wallclock secs (26.93 usr 8.01 sys + 2853.95 cusr 280.60 csys = 3169.49 CPU)

12:05

timotimo

[Tux]: please try "Text::CSV:ver<0.009>:auth<github:Tux>"

which is what "zef list | grep Text" shows

12:06

lizmat

oh?

timotimo

actually it also has Text::CSV:ver<0.008>:auth<github:Tux> in the list

but from cpan

anyway, you can --/cpan in zef, iirc

or is that only for "zef update"?

12:07

lizmat

not sure...

Geth

rakudo: 7668301dd6 | (Elizabeth Mattijsen)++ | src/core/Main.pm6
Further refine test for showing message

- prompted by R#3046
- only show message if nothing has been said on $*OUT or $*ERR

12:13

synopsebot

R#3046 [open]: github.com/rakudo/rakudo/issues/3046 [BLOCKER][regression] Unnecessary warnings on stderr

Geth

rakudo: tbrowder self-unassigned Diagnostic or definition problem? github.com/rakudo/rakudo/issues/3036
f13ac2c036 | (Stefan Seifert)++ | 2 files

ASN::META parses a definition with a grammar and creates classes and methods from the parsed description. This leads to a StrLexRef getting used as a method name which ends up in the serialization context. Decont the names of added methods to ensure we have a plain old string.
Fixes GH #3045

12:18

12:59 lizmat_ joined 13:00 evalable6 left, notable6 left, statisfiable6 left 13:01 lizmat left 13:05 AlexDaniel left, AlexDaniel joined 13:13 pamplemousse joined 13:19 lizmat_ is now known as lizmat 13:33 lucasb joined 13:59 evalable6 joined 14:01 notable6 joined, ChanServ sets mode: +v notable6 14:02 statisfiable6 joined 14:41 pamplemousse left

[Coke] 's flight was canceled last night, ended up renting a car and driving six hours instead.

14:54

lizmat

wow

15:01

[Coke]

beat the alternative, which was not being able to fly out for another 24h

15:04

lizmat

yeah... good to have such an alternative...

timotimo

i hope the airline paid for the rental

lizmat

for transatlantic flights, the car rental option doesn't work as well

timotimo

oh, yeah

15:05

ship rental isn't as affordable, and also not nearly fast enough to be a viable alternative

[Coke]

airline gave a refund on the flight, car rental was cheaper, and it's all work travel

15:07

15:14 patrickb left 15:19 nubie joined 15:21 robertle_ left

timotimo

15:33

synopsebot

RAKUDO#3045 [open]: github.com/rakudo/rakudo/issues/3045 [BLOCKER][regression] Cro::LDAP regression since 2019.03.1

lizmat

nine++ # good catch!

16:04

Geth

rakudo/release-2019.07: 600ece3b7e | (Stefan Seifert)++ | 2 files
Fix "Missing serialize REPR function for REPR NativeRef (StrLexRef)"

ASN::META parses a definition with a grammar and creates classes and methods from the parsed description. This leads to a StrLexRef getting used as a method name which ends up in the serialization context. Decont the names of added methods to ensure we have a plain old string.
Fixes GH #3045

16:06

Kaiepi

releasable6, status

16:10

releasable6

Kaiepi, Next release will happen when it's ready. R6 is down. At least 2 blockers. 666 out of 685 commits logged

Kaiepi, Details: gist.github.com/950a7c40c825c1f319...de82aea67c

AlexDaniel

oh, right in the release branch, too! Thanks!

tbrowder

hi, all

16:12

yoleaux

8 Jul 2019 15:27Z <jmerelo> tbrowder: thanks!Ç

tbrowder

i need to create a doc to show the unicode details of the STD opener chars in Rakudo repo src/Perl6/Grammar.nqp. i can get the details ok, but i also need tp