Perl 6 language and compiler development | Logs at | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by AlexDaniel on 12 June 2018.
MasterDuke nqp: nqp::add_i(1) 01:52
camelia Arg count 1 doesn't equal required operand count 3 for op 'add_i'
at gen/moar/stage2/QAST.nqp:1695 (/home/camelia/rakudo-m-inst-2/share/nqp/lib/QAST.moarvm:compile_mastop)
from gen/moar/stage2/QAST.nqp:1907 (/home/camelia/rakudo-m-inst-2/share/nqp/lib/…
MasterDuke nqp: nqp::add_i(1, 1)
camelia ( no output )
MasterDuke why does it say 3 in the error?
AlexDaniel weekly: reportable: 02:02
notable6 AlexDaniel, Noted!
AlexDaniel MasterDuke: hm, it does accept 3 02:04
nqp: nqp::say(nqp::add_i(3,4,5))
camelia 9
AlexDaniel nqp: nqp::say(nqp::add_i(4,5))
camelia 9
AlexDaniel shrugs
02:06 Kaiepi left
AlexDaniel hmmm, there's Sake for c# and swift 02:08
MasterDuke nqp: my int $a := 0; my int $b := 4; my int $c := 10; say(nqp::add_i($a, $b, $c)); say($a) 02:09
camelia 14
AlexDaniel also [S]cripted-r[ake] -- a JavaScript build tool similar to rake, or make.
so… 6ake? 02:10
MasterDuke you could rename ours to psake, with a silent 'p'
AlexDaniel yes, pee-sake
:) 02:11
that's not a bad idea at all
and something has to be done, I think
6akefile will stand out in alphabetically sorted file lists, which is nice 02:13
psake will only make sense with psix I think
6-ake-file /o\ 02:14
6file? 02:15
AlexDaniel goes to bed for some inspiration
02:31 Kaiepi joined
AlexDaniel weekly: AlexDaniel invites everyone to participate in the development of Sake ( ). Sake is a build utility inspired by rake. It is in early stages of development but is already usable for many things. The project was started by perlpilot++ 02:34
notable6 AlexDaniel, Noted!
Kaiepi will take a look
benjikun that's sweet, looking at it now 03:38
04:15 lizmat joined 04:23 MasterDuke left 04:51 skids left 05:00 brrt joined
nine So, it's a replacement for make? 06:10
06:14 ufobat_ joined 06:23 Kaypie joined, Kaiepi left
lizmat notable6: weekly 06:30
notable6 lizmat, Sorry kid, that's not my department.
lizmat weekly: test
notable6 lizmat, Noted!
lizmat notable6: weekly
hmmmm :-(
notable6: weekly 06:38
benjikun the bot developed a mind of its own 06:40
perl6 made skynet 06:41
brrt AlexDaniel: no :-P 07:12
as in, i'm not helping to replace make, i'd rather see perl6 work well with makefiles 07:17
07:59 |Tux| joined 08:11 dogbert17 left 08:43 pmurias joined 09:02 brrt left 09:29 Kaypie left
|Tux| Rakudo version 2018.06-25-ge9351cbaa - MoarVM version 2018.06-20-g888e1cf07
csv-ip5xs0.879 - 0.889
csv-ip5xs-207.323 - 7.409
csv-parser22.782 - 23.187
csv-test-xs-200.451 - 0.457
test8.588 - 8.759
test-t2.230 - 2.238
test-t --race0.926 - 0.949
test-t-2039.598 - 40.689
test-t-20 --race13.493 - 14.177
I::P5 still fails
nine |Tux|: I cannot reproduce any failure :/ 09:48
10:20 benjikun left 10:46 Zoffix joined
Zoffix lizmat: here are the notes in the bot's file: 10:46
11:58 scovit joined 11:59 Zoffix left 12:03 lizmat left 12:08 j3nnn1 joined 12:58 notable6 left, notable6 joined
AlexDaniel notable6: weekly 12:58
notable6 AlexDaniel, 6 notes:
AlexDaniel .tell brrt well, if you look at you'll see that pretty much every language has its own alternative 13:00
yoleaux AlexDaniel: I'll pass your message to brrt.
AlexDaniel .tell brrt which in many cases is very convenient given that you can write some simple things using *the same language* that you already know 13:01
yoleaux AlexDaniel: I'll pass your message to brrt.
stmuk programmers write make replacements like singers sing songs about love!
;) 13:02
AlexDaniel exactly :)
but these replacements can at least handle filenames with spaces properly…
Geth nqp/truffle: 074d81c9c4 | (Paweł Murias)++ | 3 files
[truffle] Avoid using unneeded DSL magic
nqp/truffle: 4278a59a29 | (Paweł Murias)++ | 4 files
[truffle] Enable taking of positional parameters and add some simplistic casting
nqp/truffle: f2560958c1 | (Paweł Murias)++ | nqp-truffle.nqp
[truffle] Parse with a simplified NQP setting.

Pass t/nqp/001-literals.t
timotimo .tell MasterDuke the add_i op does tak 3 registers, one of them is the target register for the calculation 13:24
yoleaux timotimo: I'll pass your message to MasterDuke.
13:32 skids joined 13:37 brrt joined 13:50 lizmat joined 14:11 greppable6 left 14:12 greppable6 joined 14:17 raschipi joined
Geth rakudo: W4anD0eR96++ created pull request #1962:
Keep .perl consistency among Set/Mix/Bag
[Coke] wonders if anyone else is considering targeting .NET core 15:29
15:35 brrt left
|Tux| nine, might it be that I have a perl5 -Duselongdouble? 15:55
geekosaur thought it built its own, specifically to ensure the -D options were as I::P5 needs them to be? 15:59
16:20 donaldh joined 16:28 donaldh left, donaldh joined 16:30 donaldh left 16:31 donaldh joined 16:32 donaldh left, donaldh joined 16:42 brrt joined
brrt AlexDaniel: I respectfully, but absolutely, disagree 16:45
AlexDaniel brrt: that's ok, but with what? You disagree that it can be convenient to describe some make-like tasks in perl 6? 16:48
raschipi They are always a pain for people trying to actually make use of the language. Everything should be either make (which has it's warts, but we're used to dealing with them) or meson 16:50
brrt AlexDaniel: yes, I disagree with that exact premise. But I'm a bit too distracted to go into detail 16:51
I have zero authority on the matter 16:52
You go do what you want to do. Have fun :-)
jnthn hasn't yet encountered any complex build system that didn't make him grumpy :) 16:53
raschipi The main problem is having to learn new ones. 16:54
jnthn gradle making me work in this sort-or-Java-but-not-quite was kinda annoying :)
Though if you copy/paste/squint it almost is Java :) 16:55
17:00 donaldh left 17:03 buggable left 17:04 buggable joined, ChanServ sets mode: +v buggable
brrt Personally I detest gradle 17:10
but that is a story for another day
lizmat and another Perl 6 Weekly hits the Net: 17:29
notable6: reset 17:35
notable6 lizmat, No notes for “reset”
lizmat notable6: weekly reset
notable6 lizmat, Moved existing notes to “weekly_2018-06-25T17:35:26Z”
lizmat Zoffix++ # handing me the notes when notable6 was not responsive
17:36 lizmat left 18:04 Ven`` joined 18:20 ilogger2 joined 18:26 ilogger2 joined
samcv i wanted to choose perl 5 and C but i could not 18:26
jnthn: what needs to happen to get goto support? 18:28
18:30 dct joined, brrt joined 18:56 benjikun joined 19:07 benjikun2 joined 19:09 benjikun left, benjikun2 is now known as benjikun 19:21 brrt left 19:27 lizmat joined 20:20 Ven`` joined
jnthn samcv: Goto within a block or out of a block: just do it with a control exception (and let spesh turn it into a low-level goto where possible). Goto into a block or between blocks: even with the zero arg constraint that's still a bit tricky; I never came up with a design I liked for it. 20:24
[Coke] could me make it a compile time error if you tried to goto too far? 20:25
*we 20:26
jnthn It's all lexical so yes, we can statically detect gotos that we refuse to do
And of course doing the limited form may well be useful enough
Or rather, better than nothing 20:27
walk, bbiab
[Coke] sweet. Now you just need a decent error message. :)
20:41 lizmat left 21:02 Ven`` left, Ven`` joined 21:11 Ven`` left 22:06 evalable6 joined, ChanServ sets mode: +v evalable6, quotable6 joined, ChanServ sets mode: +v quotable6, reportable6 joined, notable6 joined, ChanServ sets mode: +v reportable6, ChanServ sets mode: +v notable6, benchable6 joined, committable6 joined 23:20 BenGoldberg joined 23:21 skids joined