samcv ok pushed a fix to at least not make it panic. so it won't die if in a try block 00:00
ugh hate when i read this part in unicode spec 00:01
"Further customization of this rule may be necessary for best behavior of emoji zwj sequences"
that means i have to customize it to perform properly :)
BenGoldberg .tell timotimo Where is tr documented? Why can't it simply count the number of letters changed at the same time as it changes them? 00:03
yoleaux BenGoldberg: I'll pass your message to timotimo.
BenGoldberg m: say "test" ~~ tr/t/t/;
camelia Cannot modify an immutable Str (test)
in block <unit> at <tmp> line 1
BenGoldberg m: say "test" ~~ tr:r/t/t/;
camelia Cannot modify an immutable Str (test)
in block <unit> at <tmp> line 1
samcv well tr is mentioned here docs.perl6.org/language/5to6-perlfunc#tr///
timotimo BenGoldberg: it uses a low-level instruction that Just Doesn't Do That, but if you come up with something that isn't drastically slower, go right ahead 00:04
yoleaux 00:03Z <BenGoldberg> timotimo: Where is tr documented? Why can't it simply count the number of letters changed at the same time as it changes them?
samcv and there's the .trans method i guess
BenGoldberg m: my $mutate = "test"; dd $mutate ~~ tr/t/t/;
camelia StrDistance.new(before => "test", after => "test")
BenGoldberg m: my $mutate = "test"; say +$mutate ~~ tr/t/t/;
camelia Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3ā5test' (indicated by ā)
in block <unit> at <tmp> line 1
BenGoldberg m: my $mutate = "test"; say +($mutate ~~ tr/t/t/);
camelia 0
BenGoldberg Now, in perl5, that would have printed 2, the number of 't's in $mutate. 00:05
m: my $mutate = "test"; say +($mutate ~~ tr/es/se/); 00:06
camelia 2
travis-ci Rakudo build failed. Elizabeth Mattijsen 'Remove push-all from R:Q:Quanty-kv 00:13
travis-ci.org/rakudo/rakudo/builds/250966477 github.com/rakudo/rakudo/compare/3...c8e18790a4
buggable [travis build above] āœ“ All failures are due to timeout (0), missing build log (0), GitHub connectivity (1), or failed make test (0).
samcv u: ???? 00:19
unicodable6 samcv, U+1F6CC SLEEPING ACCOMMODATION [So] (????)
samcv unidump: ???? 00:20
unicodable6 samcv, gist.github.com/f3d30b5d3b9be8d277...a2fdf7bcb4
samcv well instead of trying to fix that weird emoji test in GraphemeBreakTest.t i'm going to try and fix the *actual* emoji that don't work (though those are not in the test) and hopefully will end up fixing that synthetic case 00:21
yay. from 275 to 230 failed emoji tests! woo 00:42
emoji emjoi tests not graphemeclusterbreak.t emjoi ones
so that's 45 *actual* emoji that now have the correct .chars 00:44
unidump: 0x1F4BB 00:49
unicodable6 samcv, gist.github.com/3959b019e8b86e0f40...c159ba935f
samcv argh ok. wtf. so the computer has no GCB and no special emoji properties but i'm supposed to not break inside of it. hm 00:51
"Further customization of this rule may be necessary" strikes again! 00:52
AlexDaniel ? 00:55
samcv oh just the rules they give you for emoji are incomplete 00:58
they give you certain information and then just say further customization may be needed
to have it work
as i'm going to need to save that i'm in an emoji sequence somehow, because the 3rd codeponint in the sequence doesn't has GCB=Other and Emoji_Modifier = False as well 00:59
which kind of frustrates me :\
Zoffix moritz: what's the link to your chart of Perl 6 books? Would be handy to list it under Books: perl6.org/resources/ 04:08
perl6book.com/
Added in github.com/perl6/perl6.org/commit/...cb64e7bebd 04:10
[Tux] This is Rakudo version 2017.06-170-gb0c8e1879 built on MoarVM version 2017.06-37-g4e29e4c7 06:22
csv-ip5xs 2.629
test 12.749
test-t 4.129 - 4.152
csv-parser 12.677
AlexDaniel alright, whateverable tests no longer pass after upgrading rakudoā€¦ interesting 06:55
but this seems to be consistent, always the same issue every run 06:56
I'll investigate tomorrow
.oO( and until then enjoy slightly unreliable bots! )
06:57
moritz do they throw in a snarky remark now and then? :-) 07:00
lizmat Files=1209, Tests=64162, 221 wallclock secs (13.37 usr 4.97 sys + 1344.29 cusr 140.62 csys = 1503.25 CPU) 08:20
stmuk_ I'm upgrading pl6anet.org to FreeBSD 11 so it may go *poof* for a bit 08:32
lizmat afk for most of the rest of the day& 08:47
Geth roast: 9227bc3d84 | (Samantha McVey)++ | S15-unicode-information/uniprop.t
Add test for MoarVM Issue #566 .uniprop overflow
10:20
Zoffix New blog post: "The Hot New Language Named Rakudo": perl6.party/post/The-Hot-New-Langu...med-Rakudo 10:36
stmuk_ Fo6.d for Thought *groan* 10:38
nine So we rename Perl 6 to Fo6.d? Sounds odd, but at least there won't be any confusion with anything... 10:39
Quick poll: without looking it up, in a few words, what is KDE? 10:46
ilmari a destktop environment based on Qt
nine will wait for a couple more minutes for answers 10:47
Zoffix A bloated desktop environment that assails you with all sorts of notifications and crap 10:48
stmuk_ something people didn't use because of the licencing and when that was changed noone much used it still apart from SuSE
tadzik I'd just say "a desktop environment for unix-ish systems"
nine All of the above answers are completely wrong. 10:49
Zoffix hehe cool
So what is it?
nine What all of you are talking about is the Plasma desktop.
tadzik I guess they call themselves "platform" these days
nine KDE has been the name of a free software community, since 2009!
ilmari nine: no, I was talking about the pre-plasma one that was just called KDE ;-P 10:50
stmuk_ about the last time I used KDE :P
ilmari hasn't used KDE since 4.2 or something
nine Point being: the KDE has a whole marketing _team_ and yet, 8 years have not been enough to push the name change anywhere near the general audience.
tadzik well, two-thirds of its name literally say "desktop environment" 10:51
unless it's like PERL and Perl
Zoffix Yeah, having a marketing team doesn't magically let you change people's minds at will :)
stmuk_ how many people also say "cute" rather than QT?
nine tadzik: it doesn't anymore
stmuk_: AFAIK that pronounciation is actually correct 10:52
tadzik heh, I visited their fosdem stand twice and haven't noticed 10:53
stmuk_ a name is really something other people call you and its very hard to change once established 10:55
nine stmuk_: IOW: one is not using a name, one is given a name. And people usually stick with the first suggestion for one. 11:03
stmuk_ Rimmer: "Friends call me Ace" 11:05
Zoffix takes another weekend off away from humans :) 12:43
See you Monday \o
llfourn wishes Zoffix and the bots a good weekend 13:59
Geth nqp: da0dda809a | (Jonathan Worthington)++ | tools/build/MOAR_REVISION
Bump to latest MoarVM version.
15:37
Ā¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...3-g45b008f
rakudo/nom: 1dda86222b | (Jonathan Worthington)++ | tools/build/NQP_REVISION
Bump for MoarVM updates.

Includes a fix for RT #131384.
15:38
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131384
Geth Ā¦ rakudo/nom: version bump brought these changes: github.com/perl6/nqp/compare/2017....6-gda0dda8
roast: 891e365c00 | (Jonathan Worthington)++ | S32-io/io-handle.t
Test to cover RT #131384.
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131384
jnthn m: use Test; my \a = -42; is Int(a), -42, "Sigilless variable does not confuse coercion type parsing" 16:00
camelia ===SORRY!===
No compile-time value for a
Geth rakudo/nom: c76d9324a9 | (Jonathan Worthington)++ | src/Perl6/Grammar.nqp
Don't trip over Type(sigilles-var).

If it can't be handled as a type, then we fall back to trying to parse it as a value. This worked fine in most cases, except it would try to get the compile time value for a sigilless var, which it won't have. Fixes RT #131611.
16:04
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131611
roast: c182968a83 | (Jonathan Worthington)++ | S12-coercion/coercion-types.t
Test for RT #131611.
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131611
jnthn Well, there's a random fix to finish off the week :) 16:17
AlexDaniel NeuralAnomaly: when? 16:18
NeuralAnomaly: help
NeuralAnomaly AlexDaniel, stats | blockers
AlexDaniel NeuralAnomaly: stats
NeuralAnomaly AlexDaniel, [āœ˜] Next release will be in 1 day and 1 week. Since last release, there are 22 new still-open tickets (1 unreviewed and 0 blockers) and 175 unreviewed commits. See perl6.fail/release/stats for details
travis-ci Rakudo build passed. Jonathan Worthington 'Bump for MoarVM updates. 16:50
travis-ci.org/rakudo/rakudo/builds/251206039 github.com/rakudo/rakudo/compare/b...da86222b4e
AlexDaniel timotimo: hey, are you there? 16:51
timotimo more or less, AlexDaniel 17:04
AlexDaniel timotimo: ah wellā€¦ I thought maybe you can somehow help me with bisectable segv :| 17:05
timotimo i'll be snipping up veggies for the next while, and then those veggies will be eaten for another while 17:06
what have you already figured out?
AlexDaniel timotimo: ā€¦ that there's a SEGVā€¦ now I'm a bit lost
timotimo OK 17:07
AlexDaniel I mean, that it segfaults sometimes
I can reproduce it
timotimo: ah, also I know that this is a regression 17:09
timotimo segfaults are likely moarvm stuff, could you bisect bisectable? :) 17:13
AlexDaniel .tell [Coke] case in point (re perl6 web app serving docs): irclog.perlgeek.de/perl6-dev/2017-...i_14840912 17:14
yoleaux AlexDaniel: I'll pass your message to [Coke].
AlexDaniel uhhhh, wellā€¦
not really, actually :)
but I can narrow it down to the actual line where this happens I think
noā€¦ I don't know nowā€¦ now it segfaults at a completely different moment 17:18
travis-ci Rakudo build passed. Jonathan Worthington 'Don't trip over Type(sigilles-var). 17:26
travis-ci.org/rakudo/rakudo/builds/251214786 github.com/rakudo/rakudo/compare/1...6d9324a9f6
AlexDaniel FFFFFFFFFFuuuuuuuuuuuuuā€¦ 17:45
yes, the bots are completely busted :) 17:46
dogbert17 ZofBot: you there 19:03
ZofBot dogbert17, Whereas now we can do the replacement char support and only have to extend one lower-level API to do it
dogbert17 at least one bot seems to be working :) 19:05
timotimo AlexDaniel: nom finished! 19:10
Geth 6.d-prep: b6476fcce6 | (Sylvain Colinet)++ | NativeCall.md
Add some idea about NativeCall
19:40
AlexDaniel timotimo: \o/ 20:29
timotimo i was afk again for a bit, but now i'm here 20:40
AlexDaniel yeah, me too
anyway, you probably expect me to catch it under valgrind, right? 20:41
with debugging symbols enabled
timotimo that'd probably be a good step forwards 20:52
i wonder how good it works to transfer rr recordings between computers?
AlexDaniel timotimo: mm why, can't you just connect to the server? 20:53
timotimo oh, i guess i could 20:54
does rr run there?
AlexDaniel timotimo: now yes 20:56
timotimo cool
it'd be better if we ran the moar process directly rather than the perl6-m script 20:57
AlexDaniel timotimo: alright, what does that mean? 20:58
timotimo ugh perl6-m is a rakudobrew script, it's loooooong 20:59
okay 21:00
we want this line:
rr record /home/bisectable/.rakudobrew/moar-nom/install/bin/moar --execname="$0" --libpath="/home/bisectable/.rakudobrew/moar-nom/install/share/nqp/lib" --libpath="/home/bisectable/.rakudobrew/moar-nom/install/share/nqp/lib" --libpath="/home/bisectable/.rakudobrew/moar-nom/install/share/perl6/lib" --libpath="/home/bisectable/.rakudobrew/moar-nom/install/share/perl6/runtime"
/home/bisectable/.rakudobrew/moar-nom/install/share/perl6/runtime/perl6.moarvm *arguments go here*
... lines
AlexDaniel timotimo: by the way, we can move to #whateverable so that others don't have to witness my stupidity :D 21:02
timotimo sure
it's my stupidity, too!
[Coke] is Proc::Async favored over "run" ? 21:28
timotimo it's easier to deadlock with run compared to p::a 21:30
[Coke] what about SEGV? (getting a segv with a script using run now, trying to golf it) 21:36
samcv time to close some RT's 21:44
AlexDaniel [Coke]: great! 21:45
[Coke]: do you have a short example to reproduce it?
[Coke] "trying to golf it" 21:49
AlexDaniel ah, I'm blind :) 22:04
jnthn [Coke]: run is implemented in terms of Proc::Async 22:34
(So getting the problem expressed just in terms of Proc::Async cuts out a whole layer were the problems could be) 22:36