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.
00:26 ufobat_ joined 00:30 ufobat left 02:05 squashable6 left 02:10 squashable6 joined, ChanServ sets mode: +v squashable6 02:28 squashable6 left 02:31 squashable6 joined, ChanServ sets mode: +v squashable6 03:14 squashable6 left
MasterDuke camelia: help 03:17
camelia MasterDuke: Usage: <(star-m|nqp-jvm|nqp-moarvm|debug-cat|p5-to-p6|nqp-js|rakudo-moar|rakudo-jvm|prof-m|perl6|rm|master|r-jvm|nqp|sm|p56|nqp-q|r|nqp-m|star|rakudo|r-m|rj|nqp-mvm|j|p6|m|r-j)(?^::\s(?!OUTPUT)) $perl6_program>
MasterDuke nqp-j: say(is(nqp::sprintf('%17.3g', [3.000000000000e-04]), ' 0.0003', '%17.3g 3.000000000000e-04');) 03:18
r-j: use nqp; say(is(nqp::sprintf('%17.3g', [3.000000000000e-04]), ' 0.0003', '%17.3g 3.000000000000e-04');)
03:18 squashable6 joined, ChanServ sets mode: +v squashable6
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
is used at line 1
03:18
MasterDuke r-j: use nqp; use Test; say(is(nqp::sprintf('%17.3g', [3.000000000000e-04]), ' 0.0003', '%17.3g 3.000000000000e-04');)
camelia java.lang.NullPointerException
MasterDuke bartolin: many thanks, i'm really not firing on all cylinders, but you're pointing out the fixes left and right. i corrected one or two coercion tests that needed a change, now all that's left are two sprintf tests 03:20
bartolin: one of which is `is(nqp::sprintf('%17.3g', [3.000000000000e-04]), ' 0.0003', '%17.3g 3.000000000000e-04');` is now giving me '3.0E-4', but expected value is '0.0003'. github.com/perl6/nqp/blob/master/t...ntf.t#L308 03:24
bartolin: haven't yet figured out if the problem is in sprintf's directive:<g>() github.com/MasterDuke17/nqp/blob/t...#L427-L437 or somewhere else 03:25
04:46 Kaypie is now known as Kaiepi
bartolin MasterDuke: It looks like this special case for the jvm backend (introduced back in 2013 as "ugly hack") is executed: github.com/perl6/nqp/blob/e60119f0...#L326-L330 07:34
MasterDuke: my first suspicion is that the code doesn't work anymore because $num is typed now. With the following change t/hll/06-sprintf.t passes: gist.github.com/usev6/8b21e936a2ee...dd744104e3 07:35
MasterDuke: there is a cleaner solution, probably. But I've to leave for $work now. (BTW, I also see failing tests in t/nqp/083-math.t, but I didn't look at your latest changes.) 07:37
08:05 squashable6 left 08:06 squashable6 joined 08:20 vrurg left 08:21 robertle joined 09:44 pmurias left 10:06 pmurias joined 11:03 lizmat_ joined 11:05 lizmat left 11:12 lizmat_ is now known as lizmat 12:30 AlexDaniel left
|Tux| Rakudo version 2019.03-32-g33e2d7f4c - MoarVM version 2019.03-11-g7ae6684f4
csv-ip5xs0.715 - 0.756
csv-ip5xs-206.135 - 6.220
csv-parser21.914 - 22.263
csv-test-xs-200.432 - 0.435
test7.551 - 7.731
test-t1.804 - 1.856
test-t --race0.824 - 0.901
test-t-2031.855 - 32.498
test-t-20 --race9.868 - 9.988
12:36
13:19 vrurg joined 13:46 sortiz left 14:24 llfourn left 14:25 llfourn joined 14:56 lizmat left 14:58 lizmat joined 15:17 brrt joined 15:23 lizmat left 15:25 lizmat joined 15:26 lizmat left 15:33 lizmat joined 16:40 pmurias left 16:58 TimToady left 16:59 TimToady joined 17:04 brrt left 17:11 dogbert17 left 18:17 robertle left 19:47 [Coke] joined, [Coke] left, [Coke] joined
[Coke] is rememberingn why he hasn't hacked on perl6 docs in a while. 19:47
timotimo :( 19:49
[Coke] hio, timotimo 20:01
20:04 Kaiepi left 20:05 Kaiepi joined
timotimo someone want to help me figure out the zstd code i wrote? 20:05
20:06 leont joined
timotimo if there were a visualizer for zstd byte format, that'd be kind of cool 20:07
maybe i could find out what's going wrong, or where
20:10 brrt joined
timotimo figured it out 20:12
it was, of course, an off-by-one
[Coke] ah, the "I asked the internet and immediately was able to fix it" phenomenon
timotimo not quite 20:17
i've asked a few hours ago already
20:18 dogbert17 joined
[Coke] Ah, but you had to ask here to get it to work. :) 20:20
bartolin MasterDuke: you can forget about the failing tests in t/nqp/083-math.t I mentioned earlier. It turned out they are caused by a recently introduced bug in openjdk8 on FreeBSD (that's what I tested on): bugs.freebsd.org/bugzilla/show_bug...?id=234689 20:21
that was a ... surprising result of my debugging session :) 20:22
20:35 Kaiepi left 20:36 Kaiepi joined 20:46 brrt left 21:36 pmurias joined
[Coke] where is pack defined in rakudo? 21:50
lizmat lib/experimental.pm6 21:51
[Coke] thanks 21:52
22:01 brrt joined 22:09 squashable6 left 22:13 squashable6 joined 22:21 [Coke] left 22:27 pmurias left, pmurias joined 22:50 brrt left 22:51 squashable6 left 22:55 squashable6 joined 23:02 leont left 23:22 pmurias left