»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, std:, or /msg camelia p6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by masak on 12 May 2015.
00:15 BenGoldberg joined 00:24 laouji joined 00:28 llfourn joined 00:32 llfourn left 00:39 laouji left, laouji joined 00:42 laouji left, laouji joined 01:03 danaj joined
ShimmerFairy smls: what Pod::To::HTML test are you referring to? I fixed issue #8 last night. 01:03
01:09 __blackcat joined 01:16 yeahnoob joined 01:19 cognominal left 01:32 dayangkun joined 01:45 laouji_ joined, laouji left 01:51 smls left 02:29 noganex joined, llfourn joined 02:34 llfourn left 02:42 cognominal joined 02:55 BenGoldberg left 03:01 dayangkun left 03:03 dayangkun joined 03:04 MadcapJake left 03:15 khw joined 03:26 aborazmeh left 03:34 cognominal left
ShimmerFairy .tell smls I see what the issue you were likely mentioning was, and fixed it. Turns out the tests in Pod::To::HTML were testing for the wrong result :) 03:34
yoleaux ShimmerFairy: I'll pass your message to smls.
03:35 llfourn joined 03:46 laouji_ left
dalek kudo/glr: eae3eac | skids++ | src/core/Any-iterable-methods.pm:
GLRify .pairup
03:47
kudo/glr: 3f78fa4 | lizmat++ | src/core/Any-iterable-methods.pm:
Merge pull request #499 from skids/glr

GLRify .pairup
03:47 laouji joined 03:51 laouji left
dalek kudo/glr: 2707db2 | lizmat++ | src/core/Mu.pm:
Introduce a temporary "gimme" method

Because of irclog.perlgeek.de/perl6/2015-08-17#i_11068906 , it seems to be better to be specific in why the 'gimme' method is not available anymore, instead of just 'gimme not found'.
This can of course go around Christmas, or possible even in September.
04:04
04:05 gfldex joined 04:13 __blackcat is now known as Loren
skids So what's the plan regarding *@ and **@ again? *@ seems to currenty flatten, so are we to change all/most of the API to **@ (and update spec and doc as well?.) A lot of spectest failures are due to this. 04:21
I guess we have to be careful to find those few functions that want flattening.
Or are the tests wrong and we'll be keeping *@/flattening on most sub forms? 04:22
04:30 Loren left 04:32 roguelazer joined 04:35 cognominal joined
JimmyZ skids: ping 04:36
04:37 roguelazer left, roguelazer joined 04:42 aborazmeh joined, aborazmeh left, aborazmeh joined
dalek kudo/glr: 4dccb48 | lizmat++ | src/core/asyncops.pm:
Unbreak earliest { }

Still investigating why, but it appears there is a subtle difference between @a.pick(*) and @a.pick(+@a) in the earliest context. Which is strange, as they share most of the code. In any case, this makes at least S17-supply/list.t pass its test without breaking with:
   This Seq has already been iterated, and its values consumed
04:43
lizmat skids: they way I understand it, is that *@a flattens always 04:44
and **@a does not flatten, *unless* there is only one parameter specified
m: my @a = [1,2,3]; say @a.perl 04:45
GLRelia rakudo-moar 2707db: OUTPUT«[1, 2, 3]␤»
camelia rakudo-moar 375435: OUTPUT«[[1, 2, 3]]<>␤»
lizmat m: my @a = [1,2,3],[4,5]; say @a.perl
GLRelia rakudo-moar 2707db: OUTPUT«[$[1, 2, 3], $[4, 5]]␤»
camelia rakudo-moar 375435: OUTPUT«[[1, 2, 3], [4, 5]]<>␤»
lizmat (storing values has **@a semantics)
m: my @a; @a.push: [1,2,3]; say @a.perl 04:46
GLRelia rakudo-moar 2707db: OUTPUT«[1, 2, 3]␤»
camelia rakudo-moar 375435: OUTPUT«[[1, 2, 3]]<>␤»
lizmat m: my @a; @a.push: [1,2,3],[4,5]; say @a.perl
GLRelia rakudo-moar 2707db: OUTPUT«[[1, 2, 3], [4, 5]]␤»
camelia rakudo-moar 375435: OUTPUT«[[1, 2, 3], [4, 5]]<>␤»
lizmat Array.push has **@ semantics
skids m: my @a; push(@a, 1, [0,1], 1).perl.say; 04:48
GLRelia rakudo-moar 4dccb4: OUTPUT«[1, 0, 1, 1]␤»
camelia rakudo-moar 375435: OUTPUT«[1, [0, 1], 1]<>␤»
skids but function style stays flattenning? Or is that a TBD?
JimmyZ skids: I think I got a bit better version gist.github.com/zhuomingliang/0d00...a2e851d77, not sure whether it is retrogression or not 04:51
dalek kudo/glr: facd0ef | lizmat++ | src/core/Array.pm:
Give push()/unshift() same semantics as method

Spotted by skids++
04:53
lizmat sightseeing again&
skids There are lots of those, the -index subs. I just didn't know whether to change them or the tests.
lizmat please change them, not the tests :-) 04:54
really afk now& 04:55
04:55 araujo left
skids JimmyZ: Yeah that's what happens when you get your head too far into "translating" the existing implementation. Feel free to file a PR (I don't have a commit bit.) 04:57
04:57 laouji joined
JimmyZ skids: thanks for review 04:58
skids Well, I'm probably too tired for that to constitute a "review" but eyeballing it it looks OK.
JimmyZ just got up ;) 04:59
skids has to get up 3 hours early on Wed so should probably get to bed now. 05:02
05:02 skids left 05:09 khw left 05:14 ShimmerFairy joined 05:39 Sgeo joined 05:47 brrt joined 05:50 brrt left 05:53 testform joined, testform left 05:58 TEttinger left 05:59 TEttinger joined 06:01 cognominal left 06:02 brrt joined 06:03 diana_olhovik_ joined 06:06 blackcat_ joined
brrt good * 06:06
JimmyZ \o brrt 06:11
brrt JimmyZ: will you be going to YAPC::Asia?
JimmyZ brrt: I won't 06:13
brrt too bad
neither will i, of course :-)
JimmyZ it is hard to go to japanese from china :(
ShimmerFairy
.oO(How much data must one send to Japan before they're practically in Japan?)
06:14
06:21 JimmyZ joined 06:30 laouji left 06:31 laouji joined 06:40 RabidGravy joined 06:41 rindolf joined
nine GLRious morning #perl6! 06:46
259/1047 spec test file failing 06:47
06:50 laben joined, Ven joined
laben hello p6'ers! 06:50
06:51 domidumont joined
nine This test looks quite dubious to me: ok(«$x $y» === <a b>, "«$x $y» interpolation works correctly"); 06:54
How can those two be the same object? Both are Lists containing similiar strings, but they cannot be the same list, can they? 06:55
laben nqp: nqp::say('works')
ouch, is there no nqp bot?
06:56 domidumont left
JimmyZ nine: I have a pr for GLR 06:56
06:56 domidumont joined, GLRelia joined
GLRelia nqp-jvm: OUTPUT«Can't exec "./rakudo-inst/bin/nqp-j": No such file or directory at lib/EvalbotExecuter.pm line 193.␤exec (./rakudo-inst/bin/nqp-j /tmp/tmpfile) failed: No such file or directory␤Server error occurred! Closing Link: li1360-252.members.linode.com (Quit: GLRelia)␤…» 06:56
06:56 camelia joined
GLRelia ..nqp-parrot: OUTPUT«Can't exec "./rakudo-inst/bin/nqp-p": No such file or directory at lib/EvalbotExecuter.pm line 193.␤exec (./rakudo-inst/bin/nqp-p /tmp/tmpfile) failed: No such file or directory␤Lost connection to server irc.freenode.org.␤» 06:57
..nqp-moarvm: OUTPUT«works␤»
nine JimmyZ: link?
laben: seems like it's just a bit sleepy :) 06:58
camelia ..nqp-moarvm: OUTPUT«works␤»
06:58 ChanServ sets mode: +v camelia 07:04 rindolf left
arnsholt nine: I think your === test is bogus, yeah 07:04
ShimmerFairy Yeah, I'm not sure === is the right operator (try eqv) 07:05
Ven o/, #perl6! 07:06
[Tux] test 50000 43.661 43.571 07:08
test-t 50000 42.230 42.139
maettu blogs.perl.org/users/swiss_perl_wor...kshop.html 07:10
dalek kudo/glr: c0ac618 | (Stefan Seifert)++ | src/core/Seq.pm:
Give Seq a .perl for improved Test failure diagnostics
ast/glr: 6c6f0b1 | (Stefan Seifert)++ | S02-literals/quoting.t:
Fix bogus quoting test

  «$x $y» === <a b> can hardly become True.
Both are Lists containing similiar strings, but they cannot be the same list. Use eqv instead. Thanks to arnsholt++ for clarification.
07:12
nine ShimmerFairy++ as well :)
ShimmerFairy nine: are you sure your implementation of Seq.perl won't consume the Seq in the process? 07:15
07:16 rindolf joined
nine ShimmerFairy: oh I'm sure it does. Same as .gist. I just don't see the harm. .perl is usually used for debugging and in addition is supposed to return something that can EVAL back into the original object. For that we have to consume it. 07:17
ShimmerFairy of course not, I just wanted to make sure it doesn't consume the Seq where it's not supposed to :)
And if you consume it in .perl, you render the Seq useless. So if that .list consumes it, you'll have to change it so that it clones the Seq and exhausts the clone instead. 07:18
nine The irony is, that instead of expected: Seq.new, got: Seq.new I now get expected: ("foo", "hate").iterator, got: ("foo", "hate").iterator
07:19 darutoko joined
nine ShimmerFairy: Numeric, Int, Str, Stringy, fmt, gist, AT-POS and EXISTS-POS all consume the iterator. None clone. 07:21
ShimmerFairy nine: AFAIK that's a mistake. Seq shouldn't be consumed by things like gist and perl, at least. 07:22
that is, if you stick a say $seq.perl; in the middle of your code while debugging, that shouldn't break subsequent use of the Seq.
laben ShimmerFairy: about our dicussion yesterday, CORE code assumes that if both a class and a role have implemented the same only method, the role one gets skipped entirely. see Bridge in Int and in Real (Int does Real, btw) 07:26
nine ShimmerFairy: I'm not sure I'm really qualified to have this discussion.
ShimmerFairy: I certainly see your point. On the other hand, consuming a Seq can have side effects, can't it? Cloning therefore could lead to even more surprises. 07:27
ShimmerFairy I don't know of any side-effects it would have, but the point is that sticking a say $seq.perl in your code when debugging shouldn't suddenly give an empty Seq down the line. Debugging tools are not typically supposed to affect the outcome of code :P 07:28
laben: that's an interesting example. The signatures are textually different, but the role and class both use their own name there. Since the role's stuff "becomes" the class' stuff, in most ways, that might not count as a conflict. Hm. 07:30
jdv79 well its lazy so what is one supposed to do?
ShimmerFairy (look at the "ambiguous call" error you get when you have an identical multi method defined in a role and a doing class: the candidates printed both use the _class'_ name) 07:31
jdv79 maybe .list it at least to not lose it
nine m: my $a = [1, 2].map: {say $_; $_}; @b = $a.list;
GLRelia rakudo-moar c0ac61: OUTPUT«5===SORRY!5=== Error while compiling /tmp/pfPD0FLpXA␤Variable '@b' is not declared␤at /tmp/pfPD0FLpXA:1␤------> 3my $a = [1, 2].map: {say $_; $_}; 7⏏5@b = $a.list;␤»
camelia rakudo-moar 375435: OUTPUT«5===SORRY!5=== Error while compiling /tmp/iqR_HIahjU␤Variable '@b' is not declared␤at /tmp/iqR_HIahjU:1␤------> 3my $a = [1, 2].map: {say $_; $_}; 7⏏5@b = $a.list;␤»
nine m: my $a = [1, 2].map: {say $_; $_}; my @b = $a.list;
camelia rakudo-moar 375435: OUTPUT«1␤2␤»
GLRelia rakudo-moar c0ac61: OUTPUT«1␤2␤»
nine ShimmerFairy: side effect ^^^ 07:32
07:33 Ven left
laben ShimmerFairy: consider that the implemented method in the class has a narrower sig that the role "interface" one... 07:35
ShimmerFairy laben: true. I think you'd have to use ::?CLASS if you really meant the class that will be doing the role. 07:36
Although perhaps the invocant argument needs to be handled differently than other arguments, who knows? 07:37
laben ShimmerFairy: you mean that while composing a method from a role, it should check which param is Role and allow it to be Class when comparing the sig? 07:39
ShimmerFairy My current intuition says that the something like method foo(Role $a:) would be overridden by method(Class $a:) , since a 'Role' invocant wouldn't make sense in the class anyway (I don't think).
laben: at least for the invocant specifier (which just describes the object itself, e.g. class Bar { method foo(Bar:D:) } says that Bar.foo needs to be called on a "real" (instantiated) Bar object) 07:40
laben ShimmerFairy: mh, i think we can do that just about only to the sel param, yes. messing with the others is sure to mess up what the class writer means to achieve 07:41
ShimmerFairy Basically, I think role R { method M(R: R $a, R $b) { } }; class C { method M(C: R $a, R $b) { } } would be just fine, but (say) M(C: C $a, C $b) would cause a conflict of multiple 'only' methods.
(er, with a 'does R' on the class, duh :P) 07:42
laben ShimmerFairy: understood, makes sense to me
ShimmerFairy laben: yeah, I think if the role means to specify whatever class will be doing the role, it'll have to use ::?CLASS (or however its spelled) in the other arguments.
07:43 AlexDaniel joined
laben ShimmerFairy: unfortunately, i think other conflict could be seen if class specifies those other params to be C rather than R 07:44
ShimmerFairy just for the sake of making sure it's clear, M(C: C $a, C $b) would only work (in my thinking) if the role declared that same signature, or more generically, M(C, ::?CLASS $a, ::?CLASS $b) (again however ::?CLASS is actually spelled)
laben: yes. If the role says it expects an argument of its own kind to appear $thing.foo(HERE), then the class can't narrow down that expectation. But the invocant argument specifies the thing you're calling the method on, and that naturally changes in role composition. 07:46
laben iiuc we would preclude this from working: role R {method M(R: R $a, R $b){}}; class C does R { method M(C: C $a, C $b){}}
ShimmerFairy: ok, i'm clear on this now 07:47
ShimmerFairy Yes. The R: will naturally become a C: when you do role composition, so the conflict detector should check for that. The other Rs are asking for any kind of R-doing thing, so you can't spell that out as C instead.
07:48 abraxxa joined
ShimmerFairy (the invocant won't get a free pass though; foo(Role:D:) and foo(Role:U:) are different invocants, for instance) 07:48
07:57 bjz joined 08:01 araujo joined 08:06 bjz left 08:07 laouji left
dalek kudo/glr: 7f6c2d6 | (Stefan Seifert)++ | src/core/Seq.pm:
Implement eqv for Seq to fix is-deeply tests of Seqs

Fixes is-deeply [1,2].sort, [2,1].sort;
08:08
nine 257/1047 test files fail now
08:09 laouji joined 08:10 rindolf left
TimToady m: say (1,2) === (1,2) 08:10
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar c0ac61: OUTPUT«False␤»
TimToady this is probably why «$x $y» === <a b> is expected to match, () makes an immutable tuple pre-GLR if all the bits are immutable 08:12
not sure why we're undoing this...
would be nice to have an immutable tuple type... 08:13
ShimmerFairy I thought Lists were immutable in GLR?
TimToady yes, but they might contain mutables 08:14
ShimmerFairy couldn't pre-GLR contain mutables too? :)
TimToady yes, in which case it didn't proviede a value type
ShimmerFairy If you wanted a truly immutable list type, how would you detect and disallow mutable things? 08:15
TimToady we might need an explicit tuple composer for efficiency 08:16
ShimmerFairy I think it would be nice to have an immutable counterpart for Positional things (we seem to tend to want that already, what with Buf/Blob, Set/SetHash, etc.) 08:17
08:20 zakharyas joined 08:25 CIAvash joined 08:33 bjz joined 08:40 virtualsue joined 08:42 JimmyZ_ joined
JimmyZ_ nine: PR #500 08:42
08:42 blackcat_ left 08:43 JimmyZ_ left 08:44 blackcat_ joined
jnthn TimToady: Everything in Parcel simply went away, but bits of it can be incorporated back into List 08:44
yoleaux 17 Aug 2015 21:21Z <nine> jnthn: is ([1,2] X~ <a b>), '1 2a 1 2b', '[] does not flatten'; cannot work with the nqp::istype(elem, Iterable) distinction between item and list because no matter how much you itemize [1,2], it will stay Iterable and thus be iterated over.
jnthn .tell nine yeah, I think that test has to be bogus by now, thanks to the single item rule
yoleaux jnthn: I'll pass your message to nine.
nine jnthn: I removed the test in question in the mean time
yoleaux 08:44Z <jnthn> nine: yeah, I think that test has to be bogus by now, thanks to the single item rule
jnthn ah, k 08:45
jnthn just woke up and is soon going to the airport :)
08:45 virtualsue left
jnthn TimToady: The trickier one will be that List and Parcel had quite different ideas about smartmatching 08:45
I'll delegate deciding that bit to you ;)
laben jnthn: what type have the decont'ed code_objs passed to add_method? is it not Routine? 08:46
jnthn laben: The ones the compiler passes will typically be Method/Regex
Well, or Submethod
laben jnthn: do these have no .code method?
jnthn The MOP cares little though, it assumes you're giving it something invokable
What would a .code method do? 08:47
08:47 pochi joined
jnthn They already *care* code objects. 08:47
uh, *are*
laben i wanted to get the signatures, thought i would do .code.signature
jnthn Just .signature?
laben maybe i just need to do .signature, then?
jnthn: thanks
jnthn :)
ShimmerFairy My first idea is that Lists could apply operators like === in a chaining fashion, i.e. [===] @a-List 08:48
nine jnthn: I fixed my X and meta-X problems by removing code that I don't know what it was there for: github.com/rakudo/rakudo/commit/bf657a1f5d and github.com/rakudo/rakudo/commit/44cb3de1a5
jnthn ShimmerFairy: Well, the ticky bit is that I think we did some things to List to work out well with m:g/.../ 08:49
08:50 telex left 08:52 telex joined 08:53 luiz_lha joined, espadrine joined, RabidGravy left 08:54 luiz_lha is now known as Guest91257 08:56 Guest91257 is now known as luiz_lha 08:57 virtualsue joined 08:58 andreoss joined
andreoss can i see an AST of some chunk of code? 09:00
nine JimmyZ: would be nice if the commit message told us in what way the new version is better.
andreoss or at least something like STD.parse("...").ast?
laben perl6 --target=ast ? 09:01
andreoss oh. thanks
laben andreoss: check perl6 -h since it list some interesting feature (--profile, --stagestats) 09:02
jnthn PRG -> MUC -> HND & 09:03
09:06 andreoss left 09:07 JimmyZ_ joined
JimmyZ_ nine: this pr is closer to the origin one, and a bit faster 09:07
nine: github.com/rakudo/rakudo/commit/ea...0ee83bce87 # see this one 09:08
dalek kudo/glr: da520cf | (Jimmy Zhuo)++ | src/core/Any-iterable-methods.pm:
a bit better version of pairup
09:13
kudo/glr: 6b97f53 | niner++ | src/core/Any-iterable-methods.pm:
Merge pull request #500 from zhuomingliang/patch-2

a bit better version of pairup
timotimo o/ 09:14
laben what does a null Signature obj mean in nqp-land? no params? 09:18
_itz_ m: my @a=<1 2>; say \@a 09:19
camelia rakudo-moar 375435: OUTPUT«Potential difficulties:␤ To pass an array, hash or sub to a function in Perl 6, just pass it as is.␤ For other uses of Perl 5's ref operator consider binding with ::= instead.␤ Parenthesize as \(...) if you intended a capture of a single var…»
GLRelia rakudo-moar 6b97f5: OUTPUT«Potential difficulties:␤ To pass an array, hash or sub to a function in Perl 6, just pass it as is.␤ For other uses of Perl 5's ref operator consider binding with ::= instead.␤ Parenthesize as \(...) if you intended a capture of a single var…»
09:23 JimmyZ_ left 09:24 bin_005 joined 09:27 yeahnoob left
ShimmerFairy Oh hey blog post! :D shimmeryfairy.wordpress.com/2015/0...in-rakudo/ 09:39
09:42 rurban joined 09:46 FROGGS_ joined 09:47 RabidGravy joined 09:48 rurban_ joined
JimmyZ ShimmerFairy: nice 09:50
I would like to see the new POD 6 parser 09:51
ShimmerFairy thanks. It'll take some time, though :) 09:53
nine Well, it's also about time :) 09:54
JimmyZ : 09:57
ShimmerFairy I remember back when rakudo didn't even have a Pod parser, and I tried foolishly tried writing one. But it was outside of rakudo, and I didn't want to play along with all the Pod:: classes as mentioned. 09:59
So it's kinda funny and surreal to find myself doing the same exact thing a few years later, but for totally different reasons :P
JimmyZ A few year ago, rakud core is full of PIR :) 10:01
*was 10:02
ShimmerFairy I remember that too :)
JimmyZ not mention the compiling time was 10min 10:03
10:03 TEttinger left
jdv79 not to mention startup was a real demotivator 10:20
laben how do you dump an object in nqp? cant find a way to inspect easily :( 10:26
JimmyZ .DUMP ? 10:27
laben you mean $c.DUMP or nqp::dump($c)? 10:28
JimmyZ some objects have .dump() 10:32
laben cannot find method... i would be happy with just inspecting the type 10:34
10:52 Ven joined 10:55 bjz left 10:56 bjz joined 11:04 bjz left 11:05 bjz joined 11:06 bjz left 11:08 rindolf joined 11:15 g4 joined 11:16 leont joined 11:20 dayangkun_ joined 11:21 virtualsue left, FROGGS_ left 11:22 dayangkun left 11:30 bin_005 left 11:35 leont left 11:38 FROGGS joined, dayangkun_ left 11:41 laouji left 11:45 bjz joined 11:53 leont joined 11:55 ShimmerFairy left, smls joined 11:58 CIAvash left 12:03 colomon_ joined, colomon left, colomon_ is now known as colomon 12:07 colomon_ joined, colomon left, colomon_ is now known as colomon
dalek kudo/glr: fde07e0 | (Stefan Seifert)++ | src/core/Kernel.pm:
Fix $*KERNEL.signals on non-win32
12:08
12:09 ShimmerFairy joined 12:10 brrt left
[Coke] does a git bisect to find a bug reporting on an RT... and bisect tells me it started failing after I reported the failure. O_o 12:11
colomon is github.com/perl6/specs/blob/master...-draft.pod still a valid GLR doc? 12:12
Ven m: say "foobar".IO.e.WHAT;
GLRelia rakudo-moar fde07e: OUTPUT«(Bool)␤»
camelia rakudo-moar 375435: OUTPUT«(Bool)␤»
Ven m: say "foobar".IO.d.WHAT;
camelia rakudo-moar 375435: OUTPUT«(Failure)␤»
GLRelia rakudo-moar fde07e: OUTPUT«(Failure)␤»
Ven m: use fatal; say "foobar".IO.d;
camelia rakudo-moar 375435: OUTPUT«Failed to find 'foobar' while trying to do '.d'␤ in block <unit> at /tmp/Gvjat5qs3y:1␤␤Actually thrown at:␤ in block <unit> at /tmp/Gvjat5qs3y:1␤␤»
GLRelia rakudo-moar fde07e: OUTPUT«Failed to find 'foobar' while trying to do '.d'␤ in block <unit> at /tmp/_I7VkMHeie:1␤␤Actually thrown at:␤ in block <unit> at /tmp/_I7VkMHeie:1␤␤»
Ven ^ i'm surprised by this behavior. I guess I'm supposed to do `foo.IO ~~ :d`? but why does .d return a Failure and not a Bool? 12:13
nine colomon: it's somewhat out of date, but it may still help understanding the changes
colomon: when reading, you have to s/List/Array/g and s/Parcel/List/g 12:14
ShimmerFairy colomon: if you don't have jnthn's "exploration" gist, I think that would be a more trustworthy source
m: say "foobar".IO ~~ :d 12:15
camelia rakudo-moar 375435: OUTPUT«False␤»
GLRelia rakudo-moar fde07e: OUTPUT«False␤»
ShimmerFairy Ven: I think it returns a Failure because it tests to see if the given path is a directory; if it's not a real path it can't exactly tell you anything about it, now can it? :) 12:16
colomon ShimmerFairy: link?
Ven ShimmerFairy: I'm just surprised from .e and ~~ :e being the same but .d and ~~ :d being different
ShimmerFairy Ven: and the :d actually tests a method call for its value
colomon: gist.github.com/jnthn/aa370f8b32ef98e4e7c9
colomon ShimmerFairy++
ShimmerFairy m: say "foobar.IO ~~ :!d
GLRelia rakudo-moar fde07e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/4qVd3U0iuk␤Unable to parse expression in double quotes; couldn't find final '"' ␤at /tmp/4qVd3U0iuk:1␤------> 3say "foobar.IO ~~ :!d7⏏5<EOL>␤ expecting any of:␤ argument list␤ …»
camelia rakudo-moar 375435: OUTPUT«5===SORRY!5=== Error while compiling /tmp/cNFlb6FL77␤Unable to parse expression in double quotes; couldn't find final '"' ␤at /tmp/cNFlb6FL77:1␤------> 3say "foobar.IO ~~ :!d7⏏5<EOL>␤ expecting any of:␤ argument list␤ …»
ShimmerFairy m: say "foobar".IO ~~ :!d
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar fde07e: OUTPUT«True␤»
ShimmerFairy m: say "foobar".IO ~~ :d(Failure)
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar fde07e: OUTPUT«True␤»
ShimmerFairy m: say "foobar".IO ~~ :d(False) 12:17
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar fde07e: OUTPUT«True␤»
ShimmerFairy m: say "foobar".IO ~~ :d(Bool)
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar fde07e: OUTPUT«True␤»
ShimmerFairy heh
seems I don't understand the adverb on the right-hand side yet :) 12:18
smls looks like it just negates the answer if the adverb value is false? 12:21
yoleaux 03:34Z <ShimmerFairy> smls: I see what the issue you were likely mentioning was, and fixed it. Turns out the tests in Pod::To::HTML were testing for the wrong result :)
ShimmerFairy ?."{X.key}" === ?X.value
looks like it coerces to boolean for the Any ~~ Pair case 12:22
nine m: my ($c, $d) = 1, 2; my (:$a, :$b) = :$c, :$d; say $a; say $b; 12:23
camelia rakudo-moar 375435: OUTPUT«c => 1␤d => 2␤»
GLRelia rakudo-moar fde07e: OUTPUT«c => 1␤d => 2␤»
smls ShimmerFairy: Yeah, Pod::To:HTML installs fine again, thanks!
ShimmerFairy smls: in theory I should've caught it in the first place, but reinstalling panda and everything just to run the test files is very annoying, so I ended up not bothering :P 12:24
smls: Also, I made a blog post about my plans around Pod. I can link it again if you don't see it in the backlogs. 12:27
jdv79 how come that post isnt on pl6anet? 12:28
12:28 chansen_ joined
ShimmerFairy Because my new blog hasn't been set up there yet :) 12:28
12:29 colomon left
ShimmerFairy ...Except it is? Hmmm 12:29
jdv79 nine++ # glring machine 12:30
smls ShimmerFairy: I can't seem to find it; please link it again... :P 12:32
ShimmerFairy smls: shimmeryfairy.wordpress.com/2015/0...in-rakudo/
smls thanks
12:33 pnu joined 12:34 ggherdov joined
jdv79 anybody know of any cpanish plans? 12:39
singular or plural? use p5's or separate? stuff like that. 12:40
ShimmerFairy There were plans to move to CPAN at some point, but I do not know where those plans are status-wise. 12:41
jdv79 also, can a version be a git hash? 12:42
ShimmerFairy still isn't sure having P5 and P6 modules live so close together on one system is a good idea, fwiw 12:43
jdv79 def big pros and cons
obv pro is a hell of a lot less work. stuff for free! 12:44
or at least cheaper
dalek ast/glr: 5d9c53d | (Stefan Seifert)++ | S02-types/WHICH.t:
Replace tests for no longer existing classes by new GLR classes
[Coke] ISTR lots of work was done to map out how to safely put sixperl things on cpan. 12:45
ShimmerFairy People who know both languages probably feel differently, but for me it's super weird to insist on having Perl 5 and Perl 6 code be handled by the same system, when they aren't at all the same language.
jdv79 where is it
[Coke] ShimmerFairy: insist?
It's less weird than having to build cpan from scratch.
we don't have to reinvent every where. 12:46
*wheel
jdv79 i just think about how much work and resources it would take
[Coke] jdv79: I'm remembering a snippet of conversation that occurred sometime in the last 5 years.
ShimmerFairy [Coke]: well, it seemed last time that people were really keen on trying to slap stuff onto CPAN, since it's such a 5 thing. I still find it odd that our RT system is the same as P5's, for example.
jdv79 the cpan ecosystem isnt simple or tiny
|Tux| still no panda on glr
[Coke] there was an early draft that was the result of a TPF grant that went nowhere. There was a more recent discussion that occurred, I think, at a perl qa hackathon. I *think* lizmat was involved. 12:47
ShimmerFairy: "why reinvent the wheel"
jdv79 lizmat: ?
ShimmerFairy [Coke]: I don't remember much from when CPAN stuff was happening here, but I vaguely recall that the setup was very much "try to fit it on CPAN in a way that won't cause problems", which makes P6 seem like an unimportant afterthought
[Coke] having gone through this on parrot before, there's a great desire to do everything ourselves, but then that leaves little time to do the actual thing we're trying to do. 12:48
|Tux| jdv79, markov was also working on CPAN6 iirc
nine |Tux|: no, but it's much further along. I also have a glr branch of panda on github.com/niner/panda
[Coke] ShimmerFairy: I think you're projecting.
ShimmerFairy [Coke]: I'd much prefer a setup that was appropriate for Perl 6, not a setup that tries to shoehorn itself into an existing system
12:48 chansen_ left 12:49 ggherdov left, pnu left
[Coke] ShimmerFairy: That's great. we have a pretty limited set of volunteers and hardware infrastructure. Is that really a good use of our time? 12:49
|Tux| is sooooooo eager to see possible gain of GLR on Text::CSV :)
jdv79 thats akin to saying lets reinvent github kinda
no thanks
ShimmerFairy [Coke]: Like I said, people who know 5 and 6 would probably feel different from me. But as someone who knows 6 and nothing of 5, it's always strange when people try to stuff both languages into the same room.
nine ShimmerFairy: I was at the QAH and was advocating for using CPAN for 5 and 6 (still am). There was interest there for sure. 12:50
[Coke] ... the other room is -already built-. :)
it's got a lot of nice furniture.
jdv79 its not really about p5 and p6. its more about cpan being state of thecart.
nine |Tux|: then please help making glr a reality :)
jdv79 the art
ShimmerFairy I did eventually sign up for a CPAN account, but IIRC I've never liked how I had to provide a full name (I only relented when I saw I could get by with initials) 12:51
And I've not touched that CPAN account, and for various reasons would like an account with different names attached to it now :)
jdv79 trivial details in context. 12:52
ShimmerFairy [Coke]: there's also the fact that CPAN feels really... old, last time I looked at it. Even if I were programming Perl 5, I suspect I wouldn't use CPAN to publish my stuff. (It's just... bleh)
jdv79 but if you want to impl p6 cpan, testers, pause, mirrors, metacpan, etc... plesse:)
please
|Tux| nine, how? 12:53
|Tux| more or less sees his major attribution to perl6 in doing deep testing 12:54
and feedback
and hugging lizmat
[Coke] all noble tasks, thanks!
jdv79 ShimmerFairy-- # no love for cpan?!
nine |Tux|: my first attempt at helping was finding out how much of Inc.pm we can run before crashing. That was just a week ago. It's really 90 % just mechanical work and 10 % banging one's head. 12:55
For the NYI methods I usually find a close enough method to use as a template. 12:56
ShimmerFairy jdv79: why no, it just doesn't impress me.
|Tux| works with Oracle: enoungh headbanging already
jdv79 cpan is why alot, perhaps most, use p5!
geez
gotta respect the cpan... 12:57
ShimmerFairy (also, please refrain from --'ing people, especially over their opinions. That's just rude.)
|Tux| ShimmerFairy++ 12:58
jdv79 i did include a :)
nine ShimmerFairy: metacpan.org looks quite fresh to me :) It's just pause.cpan.org that could use a re-vamp. But I only see that when uploading a dist and that's not all that often. Heavy contributors have that step automated anyway.
|Tux| no, I do not have that automated 12:59
all hand-work
ShimmerFairy nine: yeah, admittedly it's PAUSE that looks and acts like it's straight out of the 80s. _Emailing_ people to _manually_ make an account? Demanding a full name from people, who may like me not enjoy giving out personal info freely?
|Tux| but that is mainly because my pre-release tests take so much time
ShimmerFairy nine: I suppose in reality, I just don't want Perl 6 to be a second-class citizen on CPAN (which as I recall the latest plan for CPAN involved making sure not to upset P5's tools, which is kinda lame to me) 13:00
RabidGravy well cpan is twenty years old
dalek ast/glr: c1fb12f | (Stefan Seifert)++ | S02-types/compact.t:
Fix a "This type cannot unbox to a native integer" error

Need to .list the Seq returned by .map before assigning to uint8 @buffer
leont |Tux|: yeah, mine too :-)
ShimmerFairy Perl 6 is kinda a big deal; too bad if your P5 CPAN tools need a little updating :P 13:01
|Tux| my LT in Granada will be about pre-release actions
DrForr My big thing there is making perl6-1.0 available on perlbrew, that would go a ways toward adoption. 13:02
nine ShimmerFairy: I for sure don't want Perl 6 to be a second-class citizen on CPAN. I also think, that we can work together to avoid that. Of course, backwards compatibility is and always has been an issue. But certainly one worth the gain.
ShimmerFairy nine: sure, for example I wouldn't dream of suggesting "if P6 modules have to be in Perl6/ subdirs on accounts, then you better be equal and move P5 modules to a Perl5 subdir !!!" :) 13:03
13:04 pmurias joined
ShimmerFairy nine: how involved are you with CPAN, ooc? 13:04
nine I would very much like to see our tools working together nicely. I am advocating for Perl 6 modules to have Perl 5 dependencies after all ;) 13:05
pmurias ShimmerFairy: I don't think asking to provide their full names when creating CPAN accounts is unreasonable 13:06
ShimmerFairy nine: though I'll never be caught dead using P5 modules in P6 code :D
nine ShimmerFairy: as someone who wants to see the Perls growing closer together again, I had a couple of discussions about CPAN at the QA Hackathon this year. A very productive one with Andreas, the primary maintainer of PAUSE.
mst ShimmerFairy: your disappointingly combative attitude is the sort of thing that led to a lot of perl5 people hoping perl6 wouldn't use CPAN at all
"I don't want to be a second class citizen of the thing somebody else already built and maintained for 20 years for me"
well, ok, then don't use it at all 13:07
ShimmerFairy pmurias: I've historically been a _very_ private person, so requiring a full name is pretty unreasonable for me (I'm more comfortable with my name now, funny enough, but still)
jdv79 yeah...:(
mst me, I'll keep working with the people who actually care about perl5 and perl6
jdv79 cant we all just be friends?
sorry
mst also, the perl5 tooling is full of cruft but the perl6 tooling is currently essentially the one to throw away 13:08
so there's plenty of room for things to work together later
ShimmerFairy mst: Sorry for coming off as combative, it's just that as someone who enjoys Perl 6, I'm generally frustrated when Perl 6 isn't treated as its own thing (being shoehorned into CPAN instead of making CPAN support the two equally, the specs assuming P5 knowledge, etc.)
who enjoys Perl 6 with no knowledge of Perl 5, to be clear
mst 'making CPAN support' - so you want perl5 to be second class instead
which is why you're getting pushback 13:09
you can't -make- a volunteer run system do anything
either be nice, or build your own
ShimmerFairy mst: I'm not saying that at all, I'm saying both should be supported equally.
mst either be nice, or build your own
yes, you are, because you're expecting somebody else to do the work
either you play nice with what's already there, or you put the work in to build something else
jdv79 it may not be possible but 2nd class is still bettervthan no class.
ShimmerFairy mst: proposing equality is not at all like proposing a flipping of the tables.
mst you're not listening 13:10
ShimmerFairy I just don't want Perl 6 support to be stuffed into whatever corner of CPAN it can get crammed into, that's all.
mst proposing "equality" when the work to support perl5 has already been done and you're expecting other people to do the work to support perl6
is not constructive
DrForr "some things are more equal than others."
ShimmerFairy nine: do you think the existing cpan website would show P6 stuff too, or would separate site(s) be better? 13:11
nine With ShimmerFairy already being on the defense, I suggest letting this discussion rest for now.
mst entitled children are no fun anyway
ShimmerFairy: *PLONK*
nine ShimmerFairy: in my world, CPAN would show Perl modules.
RabidGravy at some point some one who cares enough will make something happen
ShimmerFairy mst: Insisting that I have to personally do all the work just to have the privilege to suggest things is very poisonous 13:12
13:13 rmgk_ joined, rmgk_ is now known as rmgk
jdv79 so glad i brought this up. 13:13
ShimmerFairy nine: yeah, I wonder what it would look like to retool cpan and metacpan to let you browse both languages, as well as what it would take to retool modules.perl6.org (though that could just be a portal to CPAN if those sites get retooled)
nine ShimmerFairy: the problem becomes much smaller once you think of it in terms of one language 13:15
ShimmerFairy jdv79: I hope don't feel bad about it :) I admit I was being combative for a bit, I've noticed lately I tend to start out like that when I discover I'm the only one with my view on things.
pmurias mst: re combative I think that's a really small majority
* combative attitude to perl5/perl6 13:16
lizmat is back from sightseeing
will backlog before reacting to anything :-)
mst pmurias: yes. I was really surprised to see somebody doing that 13:17
everybody else has grown up already
ShimmerFairy nine: sure, if you allow for the C/C++ analogy (of which I make use of C++ only as well, appropriately), then of course C and C++ can and should intermingle as much as feasible. I just happen to be someone who codes in only C++, so I don't tend to see the C side of stuff.
mst five or so years ago it was rather more of a problem
jdv79 i am justba bit concerned we've not got past the whole 5 6 hostility thing:(
i hoped that was wellbbehind us
ShimmerFairy jdv79: in reality I think it would be a good thing for 5 and 6 to get more friendly. I just happen to have no use for 5, so my opinions tend to be heavily 6-based :) 13:18
mst the refusal to let the two languages have their own names means it never well be
RabidGravy well right now there are 361 P6 modules by 87 distinct authors in the perl 6 ecosystem, not really sure how much further the current mechanism will scale
mst but, OTOH, we can get pretty close without
DrForr Shrug, that's why I'm writing p5-6 bridges, hoping to get more people at least interested.
nine jdv79: I stopped fighting my brother and found him to be one of the most awesome persons I know when I was about 17. So maybe it'll take a year or two more ;)
mst basically, occasionally you get some foolish child who starts blaming the other 'side' for not magically inventing extra volunteers
then you stomp on it
then you move on 13:19
ShimmerFairy RabidGravy: of course, if we stuck with panda I can't imagine keeping it to the ecosystem repo
mst: I hope you're still not insinuating that I specifically am a child.
13:19 lizmat left
RabidGravy but it will annoy someone at some point and someone will do something and at that point someone will consider the options for bringing P6 and P5 modules under one fold 13:20
leont isn't sure why p6 on CPAN is supposed to make sense. He does think a CPAN inspired system would be a terribly good idea.
dalek kudo/glr: 67e9f32 | (Stefan Seifert)++ | src/core/Hash.pm:
Fix positional support in hash('foo', 1) Hash constructor

Need to manually flatten now.
nine 254/1047 spec test files failing
tadzik leont: can we figure out a list of things that CPAN does either wrong or not fitting to Perl 6 that we want changed?
leont I mean, CPAN is essentially little more than an indexer and a distribution system. The indexer can't be shared, and the distribution system is easily copied. 13:21
mst right
I think 'piggyback on the distribution stuff temporarily' isn't a bad plan
and eventually you'll end up with your own indexer
and at some point you can pick a new distribution system
nine leont: I don't see anything about metacpan.org that I don't want for Perl 6
mst will work fine
jdv79 no. its testers and search also at least
plus other more minor aspects 13:22
leont nine: metacpan is a layer on top of CPAN, it's not tied to the core of it at all
13:22 lizmat joined
leont tadzik: when doing things over, I'd make the index a dynamic one (e.g. webservice based), instead of the current set of files 13:23
mst there's a difference between cpan-the-thing and cpan-the-ecosystem
jdv79 i was talking eco
tadzik leont: will you be at SPW next week?
jdv79 the thing isnt as valuable 13:24
obviously
leont No, I won't even be online next week
Some of the ecosystem can probably easily be adapted to perl6. CPAN-Testers is ultimately a fancy value-storage engine. If you write a Perl6 smoker you can probably easily reuse the centralized bits..
jdv79 YES 13:25
leont It might need some adaptation (wrt namespacing p5 and p6 separately), but other than that􏿽x85
jdv79 maybe we should assemble all the past discussions
nine m: my $hashref = {:a(1), :b(2), :c(3)}; my %h; %h = $hashref; say %h.perl; 13:26
camelia rakudo-moar 375435: OUTPUT«Odd number of elements found where hash initializer expected␤ in block <unit> at /tmp/NKuKiMTev9:1␤␤»
GLRelia rakudo-moar fde07e: OUTPUT«{:a(1), :b(2), :c(3)}␤»
lizmat leont: the *big* difference is that Perl 6 has multiple backends
so this multiplies times the number of OS's
jdv79 thatvcan just be folded into platform dimension, right?
leont I don't know metabase well, I suspect the main changes would be on the web-facing side of it
RabidGravy the cpan.io/ref/cpan/history.html might help to focus on the how and why CPAN got like it is right now
ShimmerFairy lizmat: do you have an on-hand link to a list of how P6 support on CPAN was planned? IIRC It was a specific subdir you had to put stuff in, but that's about it. 13:27
mst metabase is basically a joke anyway
it's some weird recursive mongodb storage that got prematurely generalised
so long as you don't breathe hard on it, it probably won't break
RabidGravy :)
lizmat ShimmerFairy: if a distribution contains the file META6.info (or .json, I might have that wrong in my memory)
mst but it's a fragile implementation of an incredibly silly design
lizmat PAUSE will put that distribution into a possibly newly created Perl6 subdirectory 13:28
the distrubution will *not* be indexed for Perl 5
jdv79 really? work has already been done?
lizmat yes, that has been done 13:29
ShimmerFairy lizmat: as I recall it was the forced subdir that I didn't like, even though it is just a stinkin' subdir :) (and if PAUSE can detect a P6 module from P5, why can't the rest do the same? :P)
RabidGravy sounds perfectly reasonable as something to get it started
lizmat the delay is *really* on the Perl 6 side
jdv79 thats fantastic!
woohoo
dalek ast/glr: b265042 | (Stefan Seifert)++ | S02-types/assigning-refs.t:
Fix test not conforming to its description

  jnthn++ already changed the test's description to reflect that %h = $h
indeed should not work. The test itself was still unchanged.
nine 253 to go...
lizmat the problem is really on the Perl 6 side, getting CURLI to work as specced
jdv79 ah...
lizmat I hope we can nail that at the SPW Perl 6 hackathon
ShimmerFairy I wonder if panda should be retooled to work off CPAN, as long as it's getting upgraded to GLR anyway... :) 13:30
ShimmerFairy looks up CURLI
jdv79 the path is set
very nice
lizmat ShimmerFairy: panda should become a client of CURLI.install
and CURLI.install should Do All The Right Things(TM)
jdv79 i found the cur stuff under specced 13:31
is there more info somewhere else?
leont Where can I find thsi CURLI?
lizmat jdv79: CompUnitRepo::Local::Installation
jdv79 for example how is a remote cur supposed to work
lizmat the P6 equivalent of the installed modules list of P5
jdv79 no. remote.
leont Ah 13:32
lizmat sorry, was an answer to ShimmerFairy
jdv79 the totality of the design space
lizmat leont
*sigh*
nine ShimmerFairy: there's not much to change in panda for GLR. Mostly a couple of "flat"s here and there.
leont Panda needs more work, unless it changed since last time I looked at it
ShimmerFairy lizmat: ah, an abbreviation of a P6 term (I thought it was a P5/CPAN thing). I'll add it to S99 if you don't beat me to it :P
jdv79 oh. sorry.
leont Its extensibility is kind of sucky IMO
ShimmerFairy nine: I figured, but as long as people are already working on it... :)
lizmat ShimmerFairy: I'm pretty beat being out there in the 33C heat for most of the afternoon / evening
it's going to be a cold night here in Tokyo, with lows of 26C :-) 13:33
ShimmerFairy lizmat: eee, sounds unfun (not a fan of warm weather myself)
RabidGravy shocking
jdv79 its 33c here too in istanbil
nine ShimmerFairy: as I'm currently the only one working on panda/glr I won't take on any extra work ;) Especially since I haven't done anything for my YAPC talk in a week...
jdv79 killing me, really
lizmat jdv79: know the feeling :-( 13:34
RabidGravy actually quite chilly in London :-\
lizmat yeah, in NL we had a record cold for August as well
15C max for August is pretty unheard of
nine lizmat: last time I was in Japan, I met an 86 year old Japanese. He spends two months every year on Hawaii, because he just can't stand the hot and humid Japanese summers :) 13:35
lizmat and lots and lots of rain: I wonder how long it will take before people realize we're going to have a monsoon season in NL in the future
nine: sounds like a smart guy
jdv79 i was in japan a month ago and it rained almost the whole time 13:36
blackcat_ panda is too slow, 'panda --help' will take 2s
13:36 rarara joined
dalek ecs: 5b5a934 | ShimmerFairy++ | S99-glossary.pod:
Define CURLI in glossary
13:36
13:36 aborazmeh left
DrForr On my laptop it takes a full 5 seconds :/ 13:36
13:37 chansen_ joined
ShimmerFairy lizmat: I definitely think it'd be nice if we were able to migrate o'er to CPAN before christmas, at least we've got things planned already :) 13:37
jdv79 im not sure why its a blocker for xmas but it would elevate p6's maturity maybe. 13:39
ShimmerFairy not a blocker, just a Nice Thing™ :)
jdv79 sure
blackcat_ 5s.. its a so long time..
DrForr I'm sure it'll speed up. Shrug. 13:40
RabidGravy I think it's something that's more likely to actually happen after more people start using P6 and the current arrangments for the ecosystem become unmanageable
jdv79 huh. ok. 13:41
ShimmerFairy I think so too, but on the other hand it'd be nice if panda+ecosystem didn't blow up before we made it all work.
jdv79 it would be nice if rakudo was cpanable
lizmat ShimmerFairy jdv79 Please read S11 and S22 :-)
jdv79 would be a big step forward i think 13:42
ok
mst I put rakudo on cpan
jdv79 wut
13:42 skids joined
mst eventually I deleted it because nobody was using it or reporting bugs 13:42
I'll probably try it again at some point, see if anybody cares this time
lizmat CPAN would basically be one of the content providers and possibly recommendation manager
blackcat_ Yeah , or how to do use..
13:42 pnu joined
ShimmerFairy lizmat: I'll be sure to, it's about time I understood the module system :) (And maybe help work on it afterwards, amongst the other Perl 6 stuff I'm planning on ☺) 13:42
lizmat ++ShimmerFairy 13:43
mst jdv79: you used to be able to 'cpanm Rakudo::Star'
but I got no complaints, no help, no users, and nobody reminded me to update it after releases
so I figured I'd try again later
jdv79 id like to see itvsoon
13:43 ggherdov joined
jdv79 but im at a cafe on a phone so maybe later 13:44
13:44 g4 left
dalek kudo/nom: ffbad1d | PerlJam++ | docs/announce/2015.08.md:
draft release announcement
13:45
13:46 lucasb joined
masak I think I would use cpanm to install Perl 6 modules if I could. 13:46
lizmat PerlJam: are you aware that jnthn would like to release with GLR merged? perhaps not on Thu but later >?
RabidGravy PerlJam++ # releasey
PerlJam lizmat: aye
lizmat okidoki :-) 13:47
PerlJam So ... what's a good name for this release? :)
jdv79 later &
ShimmerFairy masak: I'd rather use a P6 script to install P6 modules for my P6 programs :P
PerlJam: pre-pre-Christmas? :)
DrForr cpanm and adding perl6-1.0.0 to the perlbrew list would be tops on my personal list.
dalek ast/glr: a27d6dc | (Stefan Seifert)++ | S02-types/capture.t:
Fix another test broken by immutable tuples going away.

  () made an immutable tuple pre-GLR if all the bits are immutable.
Though List itself is immutable, its elements don't have to be, so two captures can no longer be the same object if they are constructed separately. We have to use eqv instead of === now.
smls ShimmerFairy: I posted 2 comments on your blog :)
nine 252 to go... 13:48
RabidGravy :)
leont mst: I did report one bug in Rakudo::Star right after you released it!
lizmat nine++ # can't be said emough
PerlJam ShimmerFairy: I'd don't think there's a pre-pre--Christmas PM group out there :)
mst leont: yeah, sorry
anyway, I -will- revive it somehow at some point
leont Can't say I really used it though, it was a nice experiment
mst it's a fun exercise in doing terrible things with MakeMaker if nothing else
RabidGravy (doing terrible things with MakeMaker)++ 13:49
smls nine: Why can't two instances of \(1,2,3) be the same object?
lucasb I like the idea of fixing the release date at the 25th of every month, and forgetting about tuesdays and thursdays... :)
smls the three integers are all immutable, and List is (shallowly) immutable 13:50
why should the \ change that?
13:50 g4 joined
smls m: say (1, 2, 3) == (1, 2, 3) 13:51
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«True␤»
Ven ShimmerFairy: where did you blog? :)
nine smls: well "can't" is too strong a wording. They are no longer. The compiler probably could detect the situation and re-use the object (it did pre-GLR).
ShimmerFairy Ven: shimmeryfairy.wordpress.com/2015/0...in-rakudo/
Ven smls: (..) == (..) checks number of elements, though
nine m: say (1, 2, 3) === (1, 2, 3)
Ven ShimmerFairy++ # thanks!
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«False␤»
smls :(
nine m: say (1, 2, 3) eqv (1, 2, 3)
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«True␤»
moritz \o 13:52
smls I thought (1, 2, 3) === (1, 2, 3) was an intentional feature, not some side-effect/hack
Ven ShimmerFairy: did you contact the previous parser's auhor?
nine The test I changed really was concerned with equivalence more than object identity. It used === to check if the tested object looked exactly like the reference object. But that's eqv. It doesn't have to _be_ the reference object.
smls Oh, moritz is back?
ShimmerFairy Ven: no, and I'm not sure who that would be 13:53
Ven ShimmerFairy: not sure. I guess you'd need to git blame to find out...
moritz smls: I am
blackcat_ m: say (1, 2, 4) == (1, 2, 3)
GLRelia rakudo-moar 67e9f3: OUTPUT«True␤»
camelia rakudo-moar 375435: OUTPUT«True␤»
Ven ShimmerFairy: I think hoelzro++ did a bit of Pod
ShimmerFairy I think so too, and he was here yesterday :)
nine smls: well I'd say bringing back (1, 2, 3) === (1, 2, 3) is above my pay grade :) No idea how even to begin trying to do that. For now I just change tests that are actually not concerned with this particular feature, but are testing something else and should be content with eqv. 13:54
smls nine: Understood.
lizmat nine smls : I don't think they can ever be the same, as the containers can be changed, so it is not a value based object 13:56
smls the *contents* of child containers can change
but both would still have the same container as child 13:57
hence "shallow" immutability
nine lizmat: then I will bow to your understanding of these things and happily continue replacing === with eqv when I'm sure it's ok :)
lizmat nine++ :-) 13:58
mst adjusts the lispometer
lizmat m: (1,2,3) === (1,2,3)
tadzik ShimmerFairy: re panda using CPAN, I may consider just switching it one day and thus force the entire ecosystem to adapt :P
camelia rakudo-moar 375435: OUTPUT«WARNINGS:␤Useless use of "===" in expression "(1,2,3) === (1,2,3)" in sink context (line 1)␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«WARNINGS:␤Useless use of "===" in expression "(1,2,3) === (1,2,3)" in sink context (line 1)␤»
lizmat m: say (1,2,3) === (1,2,3)
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«False␤»
tadzik and since it will need to adapt to GLR anyway...
ShimmerFairy tadzik: sounds like my kind of plan :) 13:59
lizmat m: my @a = 1,2,3; my @b = @a; say @a === @b
camelia rakudo-moar 375435: OUTPUT«False␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«False␤»
ShimmerFairy tadzik: and like I said, as long as we currently have eyes on panda already... :P
lizmat m: my $a = (1,2,3); my $b = $a; say $a === $b
nine tadzik: may I have a commit bit to panda? Then I could push my glr branch there in case someone wants to play around with it.
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«True␤»
smls lizmat: Right, Array is not meant to be immutable / value-ish. But List is, I think.
tadzik nine: sure
lizmat m: my $a = (1,2,3); my $b = (1,2,3); say $a === $b 14:00
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«False␤»
tadzik nine: last time I checked panda deps weren't ready yet though
lizmat m: my $a = (1,2,3); my $b = (1,2,3); say $a === $b; say $a.WHICH; say $b.WHICH
camelia rakudo-moar 375435: OUTPUT«True␤Parcel|(Int|1)(Int|2)(Int|3)␤Parcel|(Int|1)(Int|2)(Int|3)␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«False␤List|69409040␤List|69409088␤»
b2gills ShimmerFairy: When working with Perl 5, if the code isn't on CPAN it doesn't exist. Also I like that getting an account isn't automated, it helps prevent drive-by uploads that don't follow the guidelines, and having many uploads with slight variations ( I think I read this is a problem with Ruby Gems ).
nine tadzik: if that was before yesterday, you're long out of date ;)
tadzik nine: I think it was yesterday :P You now have a commit bit 14:01
lizmat m: my $a = (1,2,3); my $b = $a; $b[2] = 4; say $a === $b
camelia rakudo-moar 375435: OUTPUT«Cannot modify an immutable Int␤ in block <unit> at /tmp/sjFlSCf5iw:1␤␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«Cannot modify an immutable Int␤ in block <unit> at /tmp/ZXmvIOfLLd:1␤␤»
lizmat if they *are* immutable Int's then List.WHICH should reflect so
which would fix ===
ShimmerFairy b2gills: fair enough on the CPAN being the only place where stuff exists, but I still think manual account creation is a very antiquated-looking thing :)
lizmat ShimmerFairy: that's because it's 20+ years old :-) 14:02
leont In practice, the manual account creation is not that much of an issue
RabidGravy b2gills, it definitely is a problem with npm.js
leont The bar to upload something is still really low, as is proven my certain people 14:03
ShimmerFairy lizmat: yes, and PAUSE's web (not-)design reflects that best :P
nine tadzik: thanks! pushed
14:03 rindolf left
leont TBH, I rarely use PAUSE's web-interface anymore CPAN::Uploader/cpan-upload is useful like that 14:03
mst mostly uses it for mangling permissions for other people 14:04
14:05 xinming left
smls m: my $mutable = Buf.new; say set("immutable", $mutable) === set("immutable", $mutable); 14:05
camelia rakudo-moar 375435: OUTPUT«True␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«True␤»
smls ^^ lizmat: Set/Mag/Mix also happily accept mutable child elements, without breaking ===
List could do the same
lizmat smls: could you show me an example of that ? 14:06
smls === on the set simply checks === on the children
and === vor mutable types, does object identity
lizmat no, it does an eq on the .WHICH 14:07
RabidGravy I've only got three more P6 modules until I've got more than I have on CPAN
lizmat (currently)
and the .WHICH of a set is a concatenation of all the .WHICHes of the elements 14:08
smls lizmat: Doesn't the example I gave show it?
lizmat which is sub-optimal, but there you go
RabidGravy++
RabidGravy or maybe I should just delete Linux::Svgalib from CPAN as I'm pretty sure it doesn't work any more
;-)
leont should add more of his modules to the ecosystem, even if the others are currently fairly minor
RabidGravy the more the merrier leont :) 14:09
smls lizmat: Those are implementation details :P 14:10
The gist is that Set's are immutable, and if two sets have the same elements as defined by ===, the sets themselves compare positively with ===. 14:11
lizmat smls: ok, :-) 14:12
smls And I was under the impression that like pre-GLR Parcel, post-GLR List should be like that too.
Thus, my surprise at nine++'s commits message... :P 14:13
Though only jnthn and TimToady will know for sure, what post-GLR List is supposed to be like... ;) 14:14
ShimmerFairy smls: replied to your comments :)
dalek kudo/nom: ff0fb76 | lizmat++ | docs/ChangeLog:
ChangeLog tweaks
14:16
14:18 rindolf joined, lucasb left
laben if panda could get modules from CPAN, we would need a Perl6 module to upload there and a good part of the work would be done :) 14:25
14:26 davercc`` joined
laben meanwhile can i ask you to think of the distro packagers, please don't make their work harder than it needs to be 14:26
leont is using @foo􏿽xBB.bar a lot as short hand for @foo.map(*.bar), even when I don't care about it being hyper. Is there any reason why this is a bad idea? I understand that post-glr the performance cost is minimal 14:28
At least in single thread, but I'm not sure it really pays off in multi when the method is essentially an accessor
lizmat leont: for small @foo in hot loops, >>.'s overhead (estimated at about 5%) might be an issue 14:29
laben leont: can you make sure you're single threaded? otherwise the results order is not assured to be the same as input order, as with .map
lizmat otherwise, on multi-core machines, I don't think there's any other theoretical drawback 14:30
ShimmerFairy leont: » doesn't guarantee evaluation order (it's currently scrambled specifically to discourage this when it becomes more problematic), and if the method you're hyper-ing can't handle being multi-threaded (for whatever reason), there's another reason not to use »
lizmat laben: >> ensures same order at the end
14:30 zakharyas left
leont ShimmerFairy: I know, but they're accessors, so that shouldn't be an issue. It's more than I'm not actively trying to distribute the work over multiple cores, as the method should be fast 14:30
lizmat m: my @a = 1,2,3; @a>>.say # *not* in order, but say is a side-effect here
GLRelia rakudo-moar 67e9f3: OUTPUT«3␤1␤2␤»
camelia rakudo-moar ffbad1: OUTPUT«3␤1␤2␤»
14:31 cognominal joined
laben lizmat: there was some discussion yesterday or the day before that it was not assured, unless on a single thread. word was by jnthn too 14:31
lizmat m: my @a = 1,2,3; my @b = @a>>.Str; say @b.perl
GLRelia rakudo-moar 67e9f3: OUTPUT«["1", "2", "3"]␤»
camelia rakudo-moar ffbad1: OUTPUT«["1", "2", "3"]<>␤»
14:31 rurban_ left
lizmat ^^ result *is* assured to be in order 14:31
laben: that would surprise me a *lot* ! 14:32
PerlJam too
ShimmerFairy it's "race" that out-of-orders the results
lizmat ShimmerFairy: indeed
ShimmerFairy "hyper" specifically takes the time to make sure things are properly ordered
lizmat feels a bit like Teal'C
ShimmerFairy Maybe we should have a "double hyper" for race :) @foo»».really-don't-care
moritz can anybody give me a short summary of the GLR status? 14:33
laben m: say (1, 5, -7, 5)>>.sin;
camelia rakudo-moar ffbad1: OUTPUT«0.841470984807897 -0.958924274663138 -0.656986598718789 -0.958924274663138␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«0.841470984807897 -0.958924274663138 -0.656986598718789 -0.958924274663138␤»
RabidGravy lizmat, you had a gold thing painted on your forehead?
laben m: say (1, 5, -7, 5)>>.sin;
camelia rakudo-moar ffbad1: OUTPUT«0.841470984807897 -0.958924274663138 -0.656986598718789 -0.958924274663138␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«0.841470984807897 -0.958924274663138 -0.656986598718789 -0.958924274663138␤»
lizmat RabidGravy: not recently, no :-)
laben m: say (1, 5, -7, 5).map(sin);
camelia rakudo-moar ffbad1: OUTPUT«5===SORRY!5=== Error while compiling /tmp/DWXZTopVoB␤Unsupported use of bare "sin"; in Perl 6 please use .sin if you meant $_, or use an explicit invocant or argument␤at /tmp/DWXZTopVoB:1␤------> 3say (1, 5, -7, 5).map(sin7⏏5);␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«5===SORRY!5=== Error while compiling /tmp/9_0SePDQX6␤Unsupported use of bare "sin"; in Perl 6 please use .sin if you meant $_, or use an explicit invocant or argument␤at /tmp/9_0SePDQX6:1␤------> 3say (1, 5, -7, 5).map(sin7⏏5);␤»
laben m: say (1, 5, -7, 5).map(&sin);
camelia rakudo-moar ffbad1: OUTPUT«0.841470984807897 -0.958924274663138 -0.656986598718789 -0.958924274663138␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«0.841470984807897 -0.958924274663138 -0.656986598718789 -0.958924274663138␤»
nine moritz: 252/1047 spec test files failing 14:34
laben m: my @a = 1,2,3; @a>>.say
camelia rakudo-moar ffbad1: OUTPUT«3␤1␤2␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«3␤1␤2␤»
14:34 g4 left
laben yeah sorry, i remembered wrong, final result is ordered but obviously not the side-effects 14:34
you know what, a perl6 faq with these little thing explained could very well be of much use 14:36
PerlJam laben: github.com/perl6/faq Add it if it's not already there. 14:37
:)
nine m: my @a is rw; say @a.^name; 14:38
camelia rakudo-moar ffbad1: OUTPUT«5===SORRY!5=== Error while compiling /tmp/vUzoPwQjcb␤Can't use unknown trait 'is rw' in a variable declaration.␤at /tmp/vUzoPwQjcb:1␤------> 3my @a is rw7⏏5; say @a.^name;␤ expecting any of:␤ TypeObject␤ default␤ …»
GLRelia rakudo-moar 67e9f3: OUTPUT«5===SORRY!5=== Error while compiling /tmp/NA0u7cVdAf␤Can't use unknown trait 'is rw' in a variable declaration.␤at /tmp/NA0u7cVdAf:1␤------> 3my @a is rw7⏏5; say @a.^name;␤ expecting any of:␤ TypeObject␤ default␤ …»
laben PerlJam: a bit strange that perl6 faq are generated by a perl5 script from markdown :) 14:41
nine laben: rakudo does not even build without perl5 14:42
ShimmerFairy Hopefully with my work that will soon be a Perl 6 FAQ generated by a Perl 6 script from Pod6 :)
laben nine: ok but building a html file surely is not as complicated as building rakudo...
PerlJam ShimmerFairy: how soon is soon? Are you trying to get it done by christmas?
ShimmerFairy PerlJam: I should hope by Christmas, I don't want stale parsing code in Rakudo any longer than necessary :) 14:43
laben btw what i think hurts more is that it's all in one file instead of a collection
moritz nine: thanks 14:44
b2gills m: my @a = ^5; say gather { take @a; @a[0,1] = @a[1,0]; take @a }.list.perl
camelia rakudo-moar ffbad1: OUTPUT«(1, 0, 2, 3, 4, 1, 0, 2, 3, 4)␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«([0, 1, 2, 3, 4], [1, 1, 2, 3, 4])␤»
b2gills m: my @a = ^5; say gather { take @a; @a[0,1] = @ = @a[1,0]; take @a }.list.perl
camelia rakudo-moar ffbad1: OUTPUT«(1, 0, 2, 3, 4, 1, 0, 2, 3, 4)␤»
GLRelia rakudo-moar 67e9f3: OUTPUT«([0, 1, 2, 3, 4], [1, 0, 2, 3, 4])␤»
lizmat feels like take() has the wrong sig 14:46
skids wonders if the CURLI pun to "curl" was ever intentional or just happened accidentally
nine m: sub foo() { my $foo = 2; my $bar = 1; :$foo, :$bar }; my (:$foo, :$bar) := foo; say $foo;
GLRelia rakudo-moar 67e9f3: OUTPUT«Too many positionals passed; expected 0 arguments but got 2␤ in block <unit> at /tmp/thERgH1CTu:1␤␤»
camelia rakudo-moar ffbad1: OUTPUT«2␤»
lizmat skids: it was accidental 14:47
ShimmerFairy PerlJam: I'm starting work on it now, and I plan on making it worth putting on github within the next few days, optimistically. (I'm admittedly still not familiar with my changed work ethic :3)
skids lizmat: serendipitous punning++
ShimmerFairy I initially thought it somehow had something to do with `curl` at first :P .oO(Why is CPAN dependent on a unix tool's stuff?) 14:49
14:49 rurban left
ShimmerFairy
.oO(Cpan URL Installation)
14:50
DrForr Now all we need to do is port Mo and we'll be set. 14:51
lizmat
.oO( it will only take a mo :-)
ShimmerFairy Mo? 14:52
DrForr That way we'll have Larry, CURLI and Mo.
ShimmerFairy ah :)
ugexe the only thing that has stopped me from adding cpan as an authority source is a pure perl6 decompression module
ShimmerFairy
.oO(Should we rename panda to 'shemp' then?)
14:54 rmgk_ joined, rmgk left, rmgk_ is now known as rmgk 14:55 labster_ joined, Brock joined 14:59 araujo left, frew left, labster_ is now known as labster
hoelzro o/ #perl6 15:01
15:01 Util joined 15:04 oetiker joined, Guest6532 joined 15:05 fling joined
RabidGravy Harr! 15:06
15:06 diana_olhovik_ left 15:07 f3ew joined 15:10 mr-foobar joined, carlin joined 15:12 silug_ joined, mattp_ joined, mtj_ joined, frew joined 15:13 araujo joined, CIAvash joined 15:15 araujo left, isBEKaml joined
dalek kudo/glr: b5baf30 | (Stefan Seifert)++ | src/core/metaops.pm:
Bring back a part of HYPER

Not all cases supported and the whole sub may benefit from a redesign around the new Iterator API but it's enough for now to fix File::Find's tests.
15:16
isBEKaml [Coke]: is your daily test run on pugs back online?
15:16 araujo joined
RabidGravy is there a way of limiting the scope of an attribute trait to a class to which some role has been applied? (or something similar) 15:17
15:18 rurban joined
RabidGravy obviously I could fake it by over-riding the attributes compose so that it can check out the class its in, but silly 15:19
[Coke] lizmat++ # backlog first attitude. Hard to stick to. 15:22
lizmat yeah :-( 15:23
skids lizmat: WRT foo.?bar, the problem comes when someone who does not understand .? then modifies it to foo.?bar.foobar which will die in the undef case. 15:26
lizmat well, then they should start to understand :-)
skids I only appreciate that because I had to clear one of those out of the Exception code :-) 15:27
laben i'm messing around composition but sometimes when i try to get a signature with .signature, it returns a not-null and concrete object that i cannot inspect and nqp::what returns ''. can someone give me a hint?
lizmat you're getting an NQP level object in a Perl6 world 15:29
most likely
it's a crack between realities! :-)
laben: example code ? 15:30
laben lizmat: i tried to nqp::decont it too
lizmat maybe there's a bug afoot
laben to no avail
hoelzro I feel like the exchange of NQP and Perl 6 objects between the two worlds is one of the hardest parts about hacking on Rakudo
lizmat hoelzro: agree
laben compose_method_table in RoleToClassApplier.nqp, i'm trying to inspect the method code before add_method'ing it 15:31
[Coke] poitns out that his last name means Christmas in at least one language.
hoelzro ~ 15:32
oops
laben i do $_.value.signature and sometimes get that obj out
nine [Coke]: oh does the fight for the release manager slot for christmas start already? ;)
moritz is glad to let somebody else do the honors
[Coke] lizmat++ # SG reference. 15:34
isBEKaml: I tried to re-enable the pugs line, but cabal isn't available on hack, and I didn't figure out how to install it yet. 15:35
isBEKaml [Coke]: ah, who admins hack? 15:36
15:36 pnu left, chansen_ left, ggherdov left
[Coke] sheepishly raises his hand. 15:36
i have sudo, anyway.
isBEKaml hehheh
timotimo me, too *cough* 15:37
that's documented in the infrastructure docs repository on github
isBEKaml perhaps I can help?
[Coke] guesses "apt-get install <mumble>", and pokes.
moritz cabal-install maybe
isBEKaml apt-get install cabal-install #guesses 15:38
moritz does so
there's now a 'cabal' binary on hack.
isBEKaml moritz: now, you'd also need stack
nine I'm afk for a while, teaching how to fly this baby: segelflieger-linz.at/static/1/7/8/1...E-5502.jpg 15:39
moritz isBEKaml: what's stack?
isBEKaml moritz: github.com/commercialhaskell/stack...ads#debian
15:40 telex left
isBEKaml pug.hs readme also mentions it 15:40
moritz they didn't seriously name a debian package "stack", did they? 15:41
isBEKaml haskell-stack, I guess.
It should be available on testing or experimental repos
moritz well, hack is Jessie, which is stable now
isBEKaml *shrugs* I haven't used debian for a few years now 15:42
It was etch when I was using it
15:42 telex joined
[Coke] pugs build dies with unrecognized option `--extra-prog-path=/home/coke/.cabal/bin' 15:42
isBEKaml what's the cabal version? 15:43
[Coke] oh. one thing I could totally do is setup a daily glr run.
moritz seems I have to add that fpcomplete.com debian repo
[Coke] cabal-install version 1.20.0.3
isBEKaml 7.10 ships with 1.22, I think
[Coke] "cabal-install is already the newest version." 15:44
moritz ghc 7.6.3 is installe 15:45
d
15:46 mr-foobar left 15:47 chansen_ joined
isBEKaml ok, pugs is known to build with 7.6+ too. 15:47
[Coke] known? 15:48
This is post-resurrection?
isBEKaml yeah, it's mentioned somewhere in the repo. The only thing having a problem with it is windows (I raised an issue a while back -- not checked since)
15:48 mr-foobar joined
[Coke] (while back) it's been broken for a while back. 15:49
isBEKaml I know. Somebody was good enough to fix them all, thanks to stack
15:50 mr-foobar left, mr-foobar joined
isBEKaml let me try building it here, without using stack. I got ghc 7.8 15:51
15:52 abraxxa left 15:53 pnu joined
dalek rl6-roast-data: bc8600a | coke++ | bin/doit:
remove merge marker, fix typo
15:53
rl6-roast-data: ad94419 | coke++ | bin/ (2 files):
Don't actually run pugs on the daily

still can't run it automatically on hack yet.
15:54 ggherdov joined
skids m: (1, Failure.new, 2).Array; 15:57
camelia ( no output )
GLRelia rakudo-moar b5baf3: OUTPUT«Failed␤␤Actually thrown at:␤ in block <unit> at /tmp/jO0Z1GPAuZ:1␤␤»
skids OK I guess that's my next mission.
15:58 domidumont left
[Coke] trips over github.com/minimaxir/big-list-of-n...ty-strings 16:00
smls moritz
sorry
laben [Coke]: someone's redditing :)
16:00 [Sno] left
[Coke] not me, but sure. 16:00
laben i was just reading that and considering, how useful would it be to make it into asingle test 16:01
a single test file 16:02
RabidGravy well a big list of naughty javascript strings 16:05
laben the unicode ones could make for a good test for Unicode normalization 16:06
RabidGravy It might make a good test for a json parser too 16:07
leont Fun. Probably discovered some issues in YAML::Tiny while writing the p6 equivalent􏿽x85 16:08
16:13 rurban left, domidumont joined 16:16 rurban joined 16:18 khw joined, blackcat_ left 16:19 blackcat_ joined 16:32 ilbot3 joined 16:33 Averna joined 16:35 carlin left 16:37 carlin joined, carlin is now known as Guest85772 16:39 [Sno] joined 16:41 mohij joined 16:52 pmurias left 16:56 leont left 17:00 Guest85772 left 17:01 isBEKaml left
laben does a table of NQP types exist? 17:03
arnsholt Don't think so
What kind of types are you looking for? 17:04
laben trying to debug an object i get when calling .signature on a candidate method
it's not-null, concrete, not Signature, not Code, not Block, .... not Hash, not LIst, not anything i could think of 17:05
arnsholt What about Capture?
laben is it a type?
arnsholt Until GLR lands, sure 17:06
(A Perl 6 type, though, not an NQP one)
dalek rl6-roast-data: a00684a | coke++ | bin/doit:
fix more merge cruft.

Previous version basically undid the memory restrictions for most builds.
rl6-roast-data: 436def1 | coke++ | / (5 files):
add (temporary) daily run of rakudo/glr branch
laben do i need to decont it before checking? 17:07
arnsholt Dunno. Worth trying if you're stumped
moritz laben: just print its .^name 17:08
laben moritz: i'm in NQP land, doesn't print anything even with nqp::what()
[Coke] daily runs have been broken since I tried to bring pugs back without paying close attention; doing a run now, which will include numbers from rakudo/glr
moritz laben: nqp::print($obj.HOW.name($obj)) 17:09
laben i'm crossing fingers here
[Coke] n: my $obj := 3; nqp::print($obj.HOW.name($obj))
17:09 GLRelia left
camelia niecza v24-109-g48a8de3: OUTPUT«sh: mono: command not found␤» 17:09
[Coke] nqp: my $obj := 3; nqp::print($obj.HOW.name($obj)) 17:10
17:10 camelia left
RabidGravy oops 17:10
17:11 GLRelia joined
GLRelia niecza : OUTPUT«Can't chdir to '/home/steve/niecza': No such file or directory at lib/EvalbotExecuter.pm line 171.␤ EvalbotExecuter::_auto_execute(HASH(0x1b1ff70), "my \$obj := 3; nqp::print(\$obj.HOW.name(\$obj))", GLOB(0x1ec6e68), "/tmp/WzKNGrRqjb", "niecza") called at lib/Evalb…» 17:11
17:11 camelia joined
camelia nqp-moarvm: OUTPUT«BOOTInt» 17:12
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«BOO"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia ..nqp-jvm: OUTPUT«(signal ABRT)#␤# There is insufficient memory for the Java Runtime Environment to continue.␤# pthread_getattr_np␤# An error report file with more information is saved as:␤# /tmp/jvm-2601/hs_error.log␤»
..nqp-parrot: OUTPUT«Can't exec "./rakudo-inst/bin/nqp-p": No such file or directory at lib/EvalbotExecuter.pm line 193.␤exec (./rakudo-inst/bin/nqp-p /tmp/tmpfile) failed: No such file or directory␤Server error occurred! Closing Link: 78.47.60.231 (Quit: camelia)␤Lost connectio…»
17:12 ChanServ sets mode: +v camelia
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…» 17:12
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
ugexe ooo
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…» 17:13
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
laben wth
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
moritz two bots playing pingpong
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
laben omg
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…» 17:14
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen…»
GLRelia nqp-moarvm: OUTPUT«Confused at line 2, near "OUTPUT«Con"␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/steve/sandbox/perl6/rakudo/install/share/nqp/lib/nqp.moarvm:…»
17:14 camelia left, camelia joined
flussence oh geez. 17:14
[Coke] hey, that was fun. :|
laben the spam O.o
moritz ok, who runs GLRelia? 17:15
lizmat moritz: perhaps kick them?
*phew* :-)
[Coke] I learned this the hard way on a system that predated irc a while ago, bots should ignore each other. :|
moritz lizmat: I logged in to the camelia host, and killed the process
ShimmerFairy camelia should really be set up to run a glr branch real soon
flussence moritz++
[Coke] moritz++
lizmat moritz++
ShimmerFairy moritz++
17:15 ChanServ sets mode: +v camelia
[Coke] well, soon it won't matter. 17:16
laben moritz++ the #perl6 saviour
[Coke] apologies for being the one that triggered that.
flussence
.oO( methinks the output shouldn't contain a valid input prefix there )
laben maybe prefix it like [npq-moarvm]
or en.wikipedia.org/wiki/Special:Searc...search=nqp 17:17
flussence or s:1st/:/\c[$some-right-arrow]/
laben omg what's that
[ [ nqp ] ]
RabidGravy JSON::Tiny doesn't think that blns.json is valid json 17:18
everything else I've thrown at it does
dalek albot: 3235487 | moritz++ | evalbot.pl:
Disallow programs starting with "OUTPUT"

for primitive loop protection
17:19
moritz evalbot control restart
17:19 camelia left, GLRelia left, camelia joined
laben RabidGravy: did you try the others perl6 modules? like JSON::Fast 17:20
RabidGravy nope not yet
moritz evalbot control pull
17:20 ChanServ sets mode: +v camelia
camelia (success) 17:20
moritz evalbot control restart
17:20 camelia left 17:21 camelia joined
laben RabidGravy: there were some control chars, maybe JSON::Tiny doesnt accept them 17:21
flussence json has some weird rules regarding unicode, I wouldn't be surprised if that's the reason. I had to fix JSON::Tiny a long time ago to understand escape-code utf16 surrogate pairs because I had to process a file using them...
17:22 ChanServ sets mode: +v camelia
moritz flussence: I chose to leave them when I originally implemented JSON::Tiny; now I think it's feature-complete 17:22
flussence yeah, can't say it's given me any other roadbumps :)
moritz nqp-m: say 42; # intentionally triggering a syntax error to see if the bots still loop
camelia nqp-moarvm: OUTPUT«Confused at line 2, near "say 42; # "␤ at gen/moar/stage2/NQPHLL.nqp:516 (/home/camelia/rakudo-inst-2/share/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-inst-2/share/nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen/…»
Ven m: say /:s ^#/; 17:23
camelia rakudo-moar ff0fb7: OUTPUT«5===SORRY!5===␤Regex not terminated.␤at /tmp/HaJ1w8xSWV:1␤------> 3say /:s ^#/;7⏏5<EOL>␤Unable to parse regex; couldn't find final '/'␤at /tmp/HaJ1w8xSWV:1␤------> 3say /:s ^#/;7⏏5<EOL>␤ expecting any of:␤ infix st…»
Ven uh, I'm dumb.
moritz the # introduces a comment.
17:24 blackcat_ left
laben anyway, i'm going wrong here somewhere : "if !nqp::istype($a, Signature) {nqp::die('DIE: $a is ' ~ $a.HOW.name($a));}" it's DIEing telling me that $a is indeed a Signature... 17:24
moritz laben: is that early in the bootstrap process, maybe? 17:25
Ven moritz: yeah, I already said I was dumb :)
laben in metamodel yeah
shouldnt bee too early
s/bee/be/
moritz laben: also, the result of nqp::which(Signature) and nqp::which(nqp::what($a)) might be interesting (if they are identical or not)
laben moritz: O.o what is that? not in ops.markdown, i'm gonna try it right away 17:27
Ven still wishes for a .grep-not at times... 17:28
subset Comment of Str where /^'#'/; lines.grep-not(Comment) 17:29
laben uhm, i get "No registered operation handler for 'which'"
moritz laben: it's bullshit; use nqp::objectid instead
laben basically you're suspecting two different type object for the same type? 17:31
yeah they're different
flussence m: say lines.grep(/:i ^<[aeiou]>/) 17:32
camelia rakudo-moar ff0fb7: OUTPUT«Agus dhá chéad slán ag an Eireagal ard ina stua os cionn caor is coll; I mo dhiaidh bhí gleanntáin ghlas’ Ghaoth Dobhair, is beag nár bhris mo chroí. Ag taisteal dom amach trí chnoic Ghleann Domhain ’s an Mucais ar mo chúl Ach trom lámh Gall,…»
flussence m: say lines.grep( ! /:i ^<[aeiou]>/)
camelia rakudo-moar ff0fb7: OUTPUT«Method 'match' not found for invocant of class 'Any'␤ in block <unit> at /tmp/ugYCa8S0Ba:1␤␤»
smls i.imgur.com/u06mCCc.png -- new perl6.org mockup revision. For feedback/bikeshedding, please join the #perl6-site channel (which is now also logged, moritz++).
flussence m: say lines.grep(* !~~ /:i ^<[aeiou]>/)
camelia rakudo-moar ff0fb7: OUTPUT«5===SORRY!5=== Error while compiling /tmp/jRdvHjgpWJ␤Variable '&infix:<!~~>' is not declared␤at /tmp/jRdvHjgpWJ:1␤------> 3say lines.grep(* 7⏏5!~~ /:i ^<[aeiou]>/)␤»
flussence m: say lines.grep(* !~ /:i ^<[aeiou]>/) 17:33
camelia rakudo-moar ff0fb7: OUTPUT«5===SORRY!5=== Error while compiling /tmp/m_Gn0dVqD7␤Unsupported use of !~ to do negated pattern matching; in Perl 6 please use !~~␤at /tmp/m_Gn0dVqD7:1␤------> 3say lines.grep(* !~ 7⏏5/:i ^<[aeiou]>/)␤»
flussence wha?
Ven smls: I like it a lot. maybe we could get a few more examples in, and toggle between them (I'm thinking about racket-lang.org here)
smls: maybe a few too many technical words? I'm not sure people should be lost on our homepage if they don't know what a garbage collector or what gradual typing :) 17:35
(admittedly, GC is well-known)
smls Well it's a programming language homepage; Many visitors will already know about programming, but will want to know what *kind* of programming language Perl 6 is. 17:37
Ven smls: fair enough :)
smls And those who don;t know the technical terms, can write them off as buzzwords and ignore them for now... ;)
Ven (I still stand behind the "multiple examples" idea, though I'm not sure which ones. Maybe we could use perl6advent possts?)
smls Ven: Let's take this to the #perl6-site channel 17:39
laben moritz: this different Signature object doesn't have attributes $!arity nor $!params... 17:41
17:47 zakharyas joined, coffee` joined 17:55 Ven left, Averna left
RabidGravy nope, JSON::Fast can't parse it either : "can't parse objects starting in "̡͓̞ͅ yet." 17:55
[Coke] maaaaan is the jvm spectest slow. 18:07
arnsholt Yeah, the VM startup is a killer
[Coke] I'm using the eval server. 18:09
nine -awaz 18:11
18:12 rindolf left
nine -awaz 18:15
Argh...stupid KDE not remembering my konsole keyboard layout setting bug
18:15 TEttinger joined 18:20 espadrine left 18:27 spider-mario joined
moritz laben: then either it's a bootstrap or a container issue 18:34
laben moritz: i was messing around in RoleToClassApplier, trying to check signatures before adding the methods. dunno if you want to investigate deeper, i don't have time anymore for that nor would i know how to proceed 18:36
18:37 virtualsue joined, virtualsue left
pyrimidine varlogrant.blogspot.com/2015/08/i-a...perl6.html # :( 18:37
18:38 xinming joined, virtualsue joined 18:39 zakharyas left 18:40 GLRelia joined 18:41 GLRelia left
[Coke] power skimmed that, but he's under the same misapprehension that perl 6 too the brains of the perl 5 community away to hack on six. (the rest seems to be "I'm happy with perl 5and probably can't switch to perl 6 for work", which is fine. I could say the same thing about coldfusion vs. perl 6. 18:42
18:42 GLRelia joined, CIAvash left
vendethiel pyrimidine: I read that blog post as "I'm happy with perl5, and I wouldn't move to perl6 just like I wouldn't move to ruby, haskell, js, etc, because i love perl5" 18:42
_itz_ m: 1/0 18:43
GLRelia ( no output )
camelia ( no output )
vendethiel m: say (1/0).WHAT
GLRelia rakudo-moar b5baf3: OUTPUT«(Rat)␤»
camelia rakudo-moar ff0fb7: OUTPUT«(Rat)␤»
18:45 llfourn left
_itz_ GLRelia has the evalbot patch now 18:45
BTW if and when camelia gets GLR I will kill it 18:48
18:51 virtualsue left 18:54 abaugher joined
moritz _itz_: maybe you could also patch GLRelia to only respond to the rakudo target (m, r, j) 18:55
pyrimidine vendethiel: I read the same, but I also got that Perl6 left him kinda burned (maybe not intentionally, but still). 18:56
nine _itz_: until then, you're providing a useful service. Thank you!
vendethiel pyrimidine: yeah, I can see that. I'm not a perl5 guy, so I can't really say
moritz _itz_: the double response for nqp* targets for example is a bit redundant
_itz_ moritz: thanks for the feedback. I'll hack the code about a bit tomorrow 18:59
vendethiel [Coke]: do you have any good coldfusion learning ressources? 19:00
(preferably using the script-like thing, not html tags, I guess)
nine m: 1...1
GLRelia rakudo-moar b5baf3: OUTPUT«Type check failed in binding; expected 'Positional' but got 'Seq'␤ in block <unit> at /tmp/5q1V9b0wTS:1␤␤»
camelia ( no output )
nine Another 5 character GLR fail ;)
[Coke] vendethiel: nope, just years of annoying, hands on experience. 19:04
happy to answer questions off channel if you ahve something specific.
vendethiel no, i'm just always up to learn the basics of another language :) 19:06
[Coke] the scripting is an afterthought which gets slightly better each release, as more tag-based stuff is made available. variable scoping is evil, whitespace output is evil, everything is case insensitive except when it's not... 19:13
(e.g. you can not declare a variable in a method... in an object... that can be invoked multiple times across multiple requests, and isn't thread safe. and that's the -default-. 19:14
vendethiel uhm. seems confusing alright :) thanks 19:15
flussence sounds like php :D 19:16
nine probably worse though 19:17
flussence well, true, php doesn't have thread safety to worry about...
(which is a good thing IMO) 19:18
19:18 davercc`` left 19:19 davercc`` joined, bin_005 joined 19:22 yqt joined
nine Can someone explain to me what exactly the purpose of the following code is? my @right := nqp::iscont(right) ?? [right] !! (right,).list.flat; 19:23
m: ({$_},).list.flat.perl.say; 19:27
GLRelia rakudo-moar b5baf3: OUTPUT«(-> ;; $_? is parcel { #`(Block|70044240) ... }).iterator␤»
camelia rakudo-moar ff0fb7: OUTPUT«(-> ;; $_? is parcel { #`(Block|78245208) ... },)␤»
nine m: ({$_},).list.perl.say;
GLRelia rakudo-moar b5baf3: OUTPUT«(-> ;; $_? is parcel { #`(Block|74452008) ... })␤»
camelia rakudo-moar ff0fb7: OUTPUT«(-> ;; $_? is parcel { #`(Block|59064888) ... },)␤»
smls Who manages pl6anet.org ? I'm setting up a Perl 6 blog... :) 19:28
do I need an RSS feed for my blog, or does pl6anet also understand Atom?
dalek ast: b92897b | usev6++ | S02-names-vars/names.t:
Tighten scope of fudging; refer to existing ticket
19:30
_itz_ smls: submit a pull req for github.com/stmuk/pl6anet.org/blob/...perlanetrc 19:36
it understands both atom and rss
skids nine: that looks like a "single item rule" thing. 19:37
smls ok
nine skids: thanks. I think, I can just use the standard idiom for handling that 19:38
laben pyrimidine: does the majority of p5'ers think that way about p6? that would be quite disheartening :-( 19:42
19:43 Psyche^ joined 19:46 domidumont left 19:47 Averna joined
dalek kudo/nom: 2dae2c3 | ugexe++ | src/core/CompUnitRepo/Local/Installation.pm:
bug fixes for CURLI .files and bin wrapper

Fixes `gimmie` failure from bin/ wrapper loop iteration. Allows a bin/script to be with `$id` 0 to be a valid candidate
19:50
kudo/nom: 93418e9 | FROGGS++ | src/core/CompUnitRepo/Local/Installation.pm:
Merge pull request #501 from ugexe/patch-2

bug fixes for CURLI .files and bin wrapper
nine ugexe: I wonder why you replaced >> with map? 19:51
19:56 espadrine joined 19:58 AlexDaniel left 20:02 mohij left
ugexe the compiler told me to 20:18
gist.github.com/ugexe/2557bc1413d4a86bb58e
nine ugexe: but you didn't use gimme directly. Some other core code still does and has to be refactored. Using >> is still fine. 20:19
ugexe bin/scripts can be busted in the mean time or it can use .map. I didn't think it was a controversial choice 20:20
20:20 abaugher left 20:22 abaugher joined
nine ugexe: I just want to avoid any misunderstanding WRT the gimme error message 20:25
japhb What is the "single item rule"? I feel like I kn[oe]w it under a different name ... 20:30
20:31 darutoko left
ugexe m: say [[1]].perl # probably this 20:32
GLRelia rakudo-moar b5baf3: OUTPUT«[1]␤»
camelia rakudo-moar 93418e: OUTPUT«[[1]]␤»
nine Really, why the hell do we need constructs like [] ... [] giving []??
dalek kudo/glr: abf079b | (Stefan Seifert)++ | src/core/operators.pm:
Unbust sequence some more

Passes almost all spectests except for the last 3
20:34
rl6-roast-data: 77ea3c9 | coke++ | / (10 files):
today (automated commit)
20:37
[Coke] rakudo.moar-glr,2015-08-18,93.27%,a27d6dc,110851,8386,688,1121,121046,120087,b5baf30, 20:39
^^
rakudo.jvm,2015-08-18,31.46%,cddf3a2,37384,942,815,1420,40561,118164,ff0fb76, 20:40
nine Only 224/1047 spec test files failing now! :) SEQUENCE seems to have been a big one
smls m: dd ([] ... [])
GLRelia rakudo-moar abf079: OUTPUT«Cannot get sequence endpoint from an empty list (use * or :!elems instead?)␤ in block <unit> at /tmp/hb6xpC0m3_:1␤␤»
camelia rakudo-moar 93418e: OUTPUT«([],)␤»
nine m: say 1 ... 10 20:41
GLRelia rakudo-moar abf079: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
camelia rakudo-moar 93418e: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
[Coke] nine; github.com/coke/perl6-roast-data/b...ummary.out has a slightly out of date run that might have more diagnostic output than a spectest run.
nine oh, GLRelia is up to date :)
smls Heh. I guess it's because the ... op doesn't care what its LHS is (except in the case of numbere) and just calls .succ until it reaches the target?
20:41 llfourn joined
smls which in the [] case, is zero times... :P 20:42
pyrimidine laben: I don't think that's always the case, but I have heard similar sentiments from others 20:43
nine smls: that would explain why ... does that on nom, but now why we have spec tests making sure that it does so
[Coke] throws another glr run together for grins.
smls nine: True.
m: dd ([] .. []).list 20:45
camelia rakudo-moar 93418e: OUTPUT«(0,)␤»
GLRelia rakudo-moar abf079: OUTPUT«Range iterator NYI␤ in block <unit> at /tmp/zSRwSvXiDa:1␤␤»
20:46 llfourn left
nine Well the spec test seems to be more about [] ... [] falling into an endless loop: rt.perl.org/SelfService/Display.html?id=114326 21:01
Anyway, time for shutting down now. Good night #perl6! 21:10
21:16 laben left 21:23 RabidGravy left
dalek rl6-roast-data: ac769bb | coke++ | / (2 files):
Do another daily run to get latest glr fixes
21:23
[Coke] ^^ there, latest test run for people to work off of. 21:24
21:35 Averna left, skids left 21:36 Averna joined 21:42 llfourn joined 21:43 [Sno] left 21:44 bjz left 21:45 bjz joined 21:46 llfourn left 21:52 Averna left 21:53 smls left 21:58 leont joined 22:05 rurban left 22:12 abaugher left 22:13 abaugher joined, abaugher left 22:14 abaugher joined, abaugher left 22:16 leont left 22:17 abaugher joined, bjz left 22:19 bin_005 left 22:35 [Sno] joined 22:37 bbkr left 22:38 yqt left 22:43 espadrine left 22:50 skids joined
dalek c: 19c69e8 | Carlin++ | lib/Language/nativecall.pod:
add missing bracket
23:01
23:01 telex left 23:02 telex joined
dalek c: e7262d1 | Carlin++ | lib/Language/nativecall.pod:
fix variable name
23:05
23:15 aborazmeh joined, aborazmeh left, aborazmeh joined 23:20 LordVorp joined 23:30 spider-mario left 23:32 aborazmeh left 23:42 aborazmeh joined, aborazmeh left, aborazmeh joined 23:43 llfourn joined 23:47 llfourn left
dalek kudo-star-daily: f536938 | coke++ | log/MoarVM- (5 files):
today (automated commit)
23:55