Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2013-10-28

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 jnthn No, I'm not sure where hits it
00:00 timotimo i'll let the branch rest for a little longer then and drive home for now
00:00 jnthn Can you add a debugging .WHAT on the thing that decont gives back?
00:00 timotimo can do
00:00 dmol joined #perl6
00:03 timotimo oh, i'm still building for two backends
00:03 diakopter jnthn: won't the .WHAT box it automatically?
00:03 timotimo this could take a few seconds longer ...
00:04 diakopter jnthn: oh nm, I suppose that's fine :)
00:05 timotimo i wrote it in the parrot block instead of the !parrot block >_>
00:07 jnthn diakopter: Oh, I just meant the .WHAT to help us debug it
00:07 Mouq std: say 1 Δ 2; BEGIN sub infix:<Δ>(\a, \b) { &[-](b,a) } # Should this work?
00:07 camelia std 8adbc60: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Two terms in a row at /tmp/3Ir3SuQy3Q line 1:â�¤------> [32msay 1 [33mâ��[31mΔ 2; BEGIN sub infix:<Δ>(\a, \b) { &[-]([0mâ�¤    expecting any of:â�¤      feed_separatorâ�¤ infix or meta-infixâ�¤    infixed functionâ�¤       statement modif…»
00:07 Mouq r-j: say 1 Δ 2; BEGIN sub infix:<Δ>(\a, \b) { &[-](b,a) } # Should this work?
00:07 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:07 Mouq r-p: say 1 Δ 2; BEGIN sub infix:<Δ>(\a, \b) { &[-](b,a) } # Should this work?
00:07 camelia rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/UqH02Sj4s8â�¤Two terms in a rowâ�¤at /tmp/UqH02Sj4s8:1â�¤------> [32msay 1 [33mâ��[31mΔ 2; BEGIN sub infix:<Δ>(\a, \b) { &[-]([0mâ�¤    expecting any of:â�¤        postfixâ�¤        infix st…»
00:07 jnthn That looks like the most pointless ues of BEGIN ever... :P
00:07 jnthn It doesn't work 'cus BEGIN takes a blorst and statementlist contextualizes %*LANG, iirc
00:08 jnthn oh, no, it's a statement
00:08 jnthn hmm
00:08 jnthn oh..
00:08 jnthn But of course
00:08 Mouq jnthn: I ask because it has implications for something like this:
00:08 jnthn You still don't define the operator
00:08 Mouq std: sub infix:["\x227c"](\a, \b) { 1 + 2 }; say 1 ≼ 2;
00:08 camelia std 8adbc60: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Confused at /tmp/W0Jjey5g97 line 1:â�¤------> [32mfix:["\x227c"](\a, \b) { 1 + 2 }; say 1 [33mâ��[31m≼ 2;[0mâ�¤    expecting any of:â�¤  feed_separatorâ�¤ infix or meta-infixâ�¤    infixed functionâ�¤       statement modifier loopâ�¤P…»
00:08 jnthn I mean, the BEGIN is pointless because it's a sub declaration which is already compile time
00:09 jnthn The Δ still needs to be declared so the parser knows about it
00:09 Mouq I already knew it wouldn't work because of how the parser works
00:09 jnthn I think the latest one you just mentioned should work, though
00:10 jnthn I mean, "\x227c" is constant at compile time.
00:10 jnthn I don't see how the two relate, fwiw :)
00:10 jnthn The first is about one-pass parsing, the second is about what counts as constant, afaict.
00:10 timotimo running into some trouble with debugging
00:12 Psyche^_ joined #perl6
00:13 Mouq jnthn: Alright. So then something like: sub infix:["{chr(4*2207)}"](\a, \b) { 1 + 2 }; say 1 ≼ 2; shouldn't work
00:14 jnthn Mouq: I think that counts as pushing your luck... :)
00:14 jnthn Mouq: but TimToady will probably come and tell us, "yeah, that should work" :P
00:15 Mouq r: eval "sub infix:[{chr(4*2207)}](\a, \b) { 1 + 2 }; say 1 ≼ 2;" # :D
00:15 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:15 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling eval_0�Preceding context expects a term, but found infix ≼ instead�at eval_0:1�------> [32msub infix:[≼[33m�[31m](, ) 3; say 1 ≼ 2;[0m�»
00:15 Mouq r: eval "sub infix:<{chr(4*2207)}>(\a, \b) { 1 + 2 }; say 1 ≼ 2;" # :D
00:15 Mouq r: eval "sub infix:<{chr(4*2207)}>(\\a, \\b) { 1 + 2 }; say 1 ≼ 2;" # :D
00:15 Mouq Bluh
00:16 Mouq r: eval "sub infix:<{chr(4*2207)}>(\\a, \\b) \{ 1 + 2 \}; say 1 ≼ 2;" # :D
00:16 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:16 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling eval_0â�¤Malformed parameterâ�¤at eval_0:1â�¤------> [32msub infix:<≼>([33mâ��[31m, ) 3; say 1 ≼ 2;[0mâ�¤    expecting any of:â�¤        colon pairâ�¤        quote wordsâ�¤Â»
00:16 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:16 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling eval_0â�¤Missing blockâ�¤at eval_0:1â�¤------> [32msub infix:<≼>(\a, \b) [33mâ��[31m3; say 1 ≼ 2;[0mâ�¤    expecting any of:â�¤        new term to be definedâ�¤        constraintâ�¤Â»
00:16 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:16 camelia ..rakudo-parrot 3cef56: OUTPUT«3␤»
00:19 * timotimo drives home now
00:26 Mouq Is anyone going to look at Arcterus' PR for Rakudo? https://github.com/rakudo/rakudo/pull/217 His stuff is kind of getting ignored for whatever reason
00:28 jnthn Mouq: It's meant to be on my plate but my plate is overflowing as usual :/
00:29 jnthn Last time I reviewed the stuff it looked overall sane
00:30 pippo r: my @array; for ^10_000 {@array[$_]=$_}; say @array
00:30 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:30 camelia ..rakudo-parrot 3cef56: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 …»
00:31 pippo r: my @array; for ^10_000 {@array[$_]=$_}; say @array
00:31 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:31 camelia ..rakudo-parrot 3cef56: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 …»
00:35 pippo my @array; for ^1_000 {@array[$_]=$_}; my $file = @array.join(";"); my @array_clone=$file.split(";"); say @array_clone
00:35 pippo r: my @array; for ^1_000 {@array[$_]=$_}; my $file = @array.join(";"); my @array_clone=$file.split(";"); say @array_clone
00:35 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:35 camelia ..rakudo-parrot 3cef56: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 …»
00:36 pippo r: my @array; for ^10_000 {@array[$_]=$_}; my $file = @array.join(";"); my @array_clone=$file.split(";"); say @array_clone
00:36 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:36 camelia ..rakudo-parrot 3cef56: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 …»
00:37 pippo r: my @array; for ^10_000 {@array[$_]=$_}; my $file = @array.join(";"); my @array_clone=$file.split(";"); say @array_clone
00:38 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:38 camelia ..rakudo-parrot 3cef56: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 …»
00:38 pippo left #perl6
00:48 timotimo i'm trying to add debug stuff to the creation of the buildplan, but that's kinda exploding
00:49 timotimo nqp-j: say(nqp:isnull(nqp:null_s()));
00:49 camelia nqp-jvm: OUTPUT«Confused at line 2, near "say(nqp:is"␤  in panic (gen/jvm/stage2/NQPHLL.nqp:377)␤  in comp_unit (gen/jvm/stage2/NQP.nqp:922)␤  in TOP (gen/jvm/stage2/NQP.nqp:820)␤  in parse (gen/jvm/stage2/QRegex.nqp:1247)␤  in parse (gen/jvm/stage2/NQPHLL.nqp:1376)␤  in…»
00:50 timotimo nqp-j: say(nqp::isnull(nqp::null_s()));
00:50 camelia nqp-jvm: OUTPUT«0␤»
00:51 BenGoldberg r-p: my @a is default(42); say @a[5]
00:51 camelia rakudo-parrot 3cef56: OUTPUT«42␤»
00:51 BenGoldberg r-p: my @a is default (42); say @a[5]
00:51 camelia rakudo-parrot 3cef56: OUTPUT«True␤»
00:51 * BenGoldberg is confused
00:51 BenGoldberg Why does that space change it to True?
00:54 BenGoldberg r-j: 0
00:54 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:54 timotimo a mere say in the create buildplan gives me a null pointer exception :|
00:55 timotimo ah, but an nqp::say seems to work
00:55 Mouq rn: say "a"~~token a { <sym> }
00:55 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:55 camelia ..rakudo-parrot 3cef56: OUTPUT«Useless declaration of a has-scoped method in mainline␤「」␤ sym => 「」␤␤»
00:55 camelia ..niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m===[0m��Methods must be used in some kind of package at /tmp/Ortg2lMSTS line 1:�------> [32msay "a"~~token a [33m�[31m{ <sym> }[0m��<sym> is only valid in multiregexes at /tmp/Ortg2lMSTS line 1:�------> [32msa…»
00:55 BenGoldberg r-j: my $t = token { <sym> }; say "a" ~~ $t;
00:56 BenGoldberg r-p: my $t = token { <sym> }; say "a" ~~ $t;
00:56 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
00:56 camelia rakudo-parrot 3cef56: OUTPUT«「」␤ sym => 「」␤␤»
00:56 colomon joined #perl6
00:56 BenGoldberg r-p: say "a" ~~ token { <sym> };
00:56 camelia rakudo-parrot 3cef56: OUTPUT«「」␤ sym => 「」␤␤»
00:56 BenGoldberg r-p: say "a" ~~ token a { <sym> };
00:56 camelia rakudo-parrot 3cef56: OUTPUT«Useless declaration of a has-scoped method in mainline␤「」␤ sym => 「」␤␤»
00:56 Mouq n: grammar A {token a { <sym> }}; A.parse("a")
00:56 camelia niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤<sym> is only valid in multiregexes at /tmp/JGRdZPIluw line 1:â�¤------> [32mgrammar A {token a { <sym[33mâ��[31m> }}; A.parse("a")[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib…»
00:57 Mouq Niecza gets a cookie :)
00:57 BenGoldberg n: my $t = token { <sym> }; say "a" ~~ $t;
00:57 camelia niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤<sym> is only valid in multiregexes at /tmp/0wgJbrdMmz line 1:â�¤------> [32mmy $t = token { <sym[33mâ��[31m> }; say "a" ~~ $t;[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/COR…»
00:57 BenGoldberg n: my $t = rule { <sym> }; say "a" ~~ $t;
00:58 camelia niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤<sym> is only valid in multiregexes at /tmp/csPYjiURBm line 1:â�¤------> [32mmy $t = rule { <sym[33mâ��[31m> }; say "a" ~~ $t;[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE…»
00:59 woosley joined #perl6
01:00 Mouq std: grammar A {token a { <sym> }};
01:00 camelia std 8adbc60: OUTPUT«ok 00:01 124m␤»
01:01 Mouq n: grammar A {token a { <sym> }; token sym { "This shouldn't blow up" } };
01:01 camelia niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤<sym> is only valid in multiregexes at /tmp/5oK41JSCcq line 1:â�¤------> [32mgrammar A {token a { <sym[33mâ��[31m> }; token sym { "This shouldn't blow up[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/…»
01:01 Mouq Niecza gets half a cookie
01:02 Mouq Hmm. Maybe
01:07 diakopter I just realized.. the more you spam #perl6, Programming Perl 6 is delayed by 1 more second...
01:07 diakopter <WINK>
01:22 timotimo wtf, this makes no sense
01:24 timotimo unless of course there is some kind of recursion that i'm not seeing
01:24 timotimo doesn't seem to be the case
01:25 timotimo i've sprinkled the whole method with debug says, and it seems to not happen inside the while loop at all, even though that's what i see in the backtrace
01:29 diakopter ETOOMANYSPRINKLES
01:33 timotimo so ... apparently the posat that gets the code is the culprit? but ... whaaat?
01:37 timotimo so, the result of 4 + $typespec is not unboxable to a native int?
01:37 timotimo r-j: say nqp::objprimspec(my int $x).WHAT;
01:37 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
01:38 timotimo bluh
01:39 benabik r-j: # timeout
01:39 timotimo using nqp::add_i instead of + seems to fix that little bit, great! >_>
01:39 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
01:39 benabik rakudo:
01:40 Mouq benabik: camelia doesn't bother if you don't give it a program
01:42 timotimo jnthn: Cannot unbox a type object is the result of nqp::say(nqp::decont(%attrinit{$key_name}).WHAT)
01:42 timotimo oh, i may need to just-say that
01:44 benabik Mouq: I was just trying to find the minimal program that times out.  I guess less than just a comment isn't going to happen.
01:45 Mouq r-j: (timeout)
01:45 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
01:45 Mouq rnp: say (timeout)
01:46 Mouq rnp: say "(timeout)" # dummy
01:46 Mouq Do we no longer have rnp mapped?
01:46 Mouq rn: say "(timeout)" # dummy
01:46 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
01:46 camelia ..rakudo-parrot 3cef56, niecza v24-98-g473bd20: OUTPUT«(timeout)␤»
01:47 timotimo great, i got it to work on rakudo-jvm
01:47 timotimo the only thing wrong with the code was the 4 + $typecode
01:49 benabik Finally!  I have nqp-[pmj] and rakudo-[pj] installed.
01:49 Mouq nqp: grammar A { token TOP { <a>* }; proto token a {*}; token a:!sym { <sym> } }; say(A.parse("000000")) #heh
01:49 camelia nqp-moarvm, nqp-jvm, nqp-parrot: OUTPUT«000000␤»
01:49 timotimo oh you silly :)
01:51 * Mouq can't figure out where, but a:!b becomes a:b<0> and a:b becomes a:b<1> somewhere in the parse stage of NQP
01:52 Mouq Or... somewhere
01:53 timotimo easy
01:53 timotimo method colonpair in src/NQP/Actions.nqp
01:53 timotimo it takes the second branch and puts the IVal there with !$<not>
01:54 Mouq Okay, but how does that value end up in %*RX<name>?
01:54 Mouq Oh
01:54 Mouq It actually does take the .ast
01:54 * Mouq is dumb
01:54 timotimo where in the spectests do i put tests for the buildplan?
01:55 timotimo ah, easy
01:56 timotimo S12-construction seems to be the place
01:57 PZt joined #perl6
02:03 SmokeMachine joined #perl6
02:09 SmokeMachine joined #perl6
02:10 timotimo i should hit the hay and continue this tomorrow.
02:25 wayland-mobile joined #perl6
02:29 dalek nqp: 1931744 | Mouq++ | src/ (3 files):
02:29 dalek nqp: Allow multiregexes of the form category:usage
02:29 dalek nqp:
02:29 dalek nqp: Before, we always had to use "category:sym<usage>" even if "usage" was
02:29 dalek nqp: entirely <ident>
02:29 dalek nqp: review: https://github.com/perl6/nqp/commit/193174476a
02:34 Mouq nqp: grammar A { rule TOP { <fruit> + }; proto token fruit {*}; token fruit:sym<banana> { <sym> s? };}; say(A.parse("banana bananas banana"))
02:34 camelia nqp-moarvm, nqp-jvm, nqp-parrot: OUTPUT«banana bananas banana␤»
02:35 Mouq fruit:sym<banana> can now just be fruit:banana, is all that does. Of course, it's not bootstrapped and won't work in Rakudo Perl, so the only place you can really use it ATM is the Rakudo NQP source
03:01 lue I still get those failures. :(
03:02 lue which means my test modifications didn't get pulled. Considering the returned value is correct, I say the spectest passed successfully, but I'd like further input before moving to modules-test
03:04 Mouq lue: "[S05-mass] Fix punctuation tests."?
03:04 lue yes.
03:07 BenGoldberg r: print "(timeout)"
03:07 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«(timeout)»
03:07 wayland-mobile left #perl6
03:07 BenGoldberg :)
03:07 BenGoldberg r: 1
03:08 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
03:08 camelia ..rakudo-parrot 3cef56: ( no output )
03:24 dansamo joined #perl6
03:26 lue Spectest has two failures, one's my fault, and the other can't be changed unless someone moves tags around in rakudo to get a rakudo that pulls the updated specs. I'll go through the rest of Step 10 and then wait for someone to respond.
03:40 lue Welp, DBIish failed. (That's the only broken one by test)
03:41 lue ✘ ⑩ (Release PAUSED until further notice)
03:46 lue jnthn, moritz, other release pros: what should I do about https://gist.github.com/lue/7188705 (ignore the access failure, I have a noatime mount) and the failed DBIish module test? I did fix the failing charset test (¶ and § *are* punctuation), but it appears R* isn't pulling the fixed version.
04:46 * TimToady would like to point out to various and sundry that infix:<{...}> or infix:<xdeaf> will never interpolate inside <>
04:46 TimToady we've seen that braino a couple times in the last several days now
04:47 lue infix:«{...}» works though, right?
04:48 TimToady BenGoldberg: re http://irclog.perlgeek.de/​perl6/2013-10-28#i_7774302 a trait is just sugar for an colonpair, so without an argument defaults to True; dunno what the (42) is doing other than maybe poking a subsig pattern into the signature
04:50 TimToady lue: presumably it oughta do that, unless people think it should work like {} in shell quoting
04:50 lue "it" being «» or <> ?
04:51 TimToady {}
04:51 lue (2nd it)
04:51 TimToady «» is assumed already
04:51 lue assumed when? infix:<X> ? inside {}?
04:52 TimToady you already assumed «» in your question!
04:53 TimToady you asked if «{...}» should work
04:53 TimToady r: say « a { 40 + 2 } z »
04:54 * lue gets a hand on his caffeine before continuing... his brain must be slowing down :)
04:54 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
04:54 camelia ..rakudo-parrot 3cef56: OUTPUT«a42z␤»
04:54 TimToady rp++
04:54 lue r-j: say "Hi TimToady o/"
04:54 TimToady so it looks liek it works in rp anyhoo
04:54 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
04:54 * lue suspects the startup time is too much for camelia
04:55 TimToady rj also has problems from time to time with recursive errors, but you'd think that'd show up in the message if so
04:55 TimToady j: say 42
04:55 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
04:56 lue j: True
04:56 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
04:56 JimmyZ j: say (timeout)
04:56 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
04:57 JimmyZ look, it works
04:57 lue that's not good; it doesn't even parse the syntax!
04:57 lue std: say (timeout)
04:57 TimToady consistency is overrated in this case
04:57 preflex_ joined #perl6
04:57 camelia std 8adbc60: OUTPUT«[31m===[0mSORRY![31m===[0m�Undeclared routine:� 'timeout' used at line 1�Check failed�FAILED 00:01 123m�»
04:57 TimToady j: 1 2
04:57 Mouq rn: sub infix:«\x227c»(\a, \b) { 1 + 2 }; say 1 ≼ 2;
04:58 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
04:58 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
04:58 camelia ..rakudo-parrot 3cef56, niecza v24-98-g473bd20: OUTPUT«3␤»
04:59 TimToady j apparently can't even get the parser going long enough to detect a syntax error
04:59 * TimToady wonders if it's fighting for cpu at the moment
04:59 * lue would like CORE.setting to not b0rk up unicode parsing, so all those ugly infix:["x2A03"]'s can be transformed to infix:<⨃>
04:59 TimToady the limit is wall clock time, isn't it?
05:01 lue You should slow down your clock to give it a better chance then :P
05:02 moritz \o
05:02 lue o/
05:03 lue moritz: I had some spectest failures on ⑩ earlier, if you're backlogging :)
05:10 moritz lue: I am; IMHO you should proceed anyway
05:10 lue did you see that DBIish failed its tests too?
05:10 moritz lue: since the compiler release tested successfully for several people, we can discount the failures as platform specific for now
05:10 moritz lue: no, I did not
05:11 moritz lue: have a link to the failures?
05:11 * lue updates the gist to include broken module
05:12 lue https://gist.github.com/lue/7188705
05:12 lue moritz: ^^^ (I know the noatime is my fault, and I personally think rakudo is correcter than the specific spectest revision it pulls on punctuation. So only the module concerns me.)
05:13 moritz lue: DBIish seems to be caused by the rakudo regression regarding whatever-currying of *[0] etc
05:13 moritz and work-aroundable
05:13 moritz so I'll do that soon
05:14 moritz but first, breakfast for $daughter
05:15 lue for @family { &breakfast; POST { query "proceed to ⑪ or pause, then?" } } # :)
05:17 kaleem joined #perl6
05:27 lue (in any case, I suspect 12-15 at least will happen tomorrow)
05:45 lue ♘ all o/
05:53 dalek DBIish: b0c5d75 | moritz++ | lib/DBDish/Pg.pm6:
05:53 dalek DBIish: work around rakudo regression
05:53 dalek DBIish:
05:53 dalek DBIish: lue++ for noticing it during the R* release process
05:53 dalek DBIish: review: https://github.com/perl6/DBIish/commit/b0c5d75f7f
05:53 dansamo joined #perl6
05:55 dalek rakudo/nom: a7f7522 | benabik++ | tools/build/ (2 files):
05:55 dalek rakudo/nom: Teach j-install about DESTDIR
05:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a7f7522bde
06:14 sqirrel joined #perl6
06:16 FROGGS joined #perl6
06:19 xenoterracide joined #perl6
07:01 denis_boyun joined #perl6
07:02 [Sno] joined #perl6
07:34 dmol joined #perl6
08:01 zakharyas joined #perl6
08:07 FROGGS[mobile] joined #perl6
08:24 masak moarning, #perl6
08:24 moritz \o masak
08:25 sjn o/
08:27 tadzik masak! asdfmovie7 is out :D
08:29 * masak .oO( the latest instalment in the asdf franchise )
08:29 FROGGS[mobile] Morning
08:29 masak good morning, FROGGS in a mobile.
08:30 * masak has a mental image of rubbing a mobile phone with a piece of cloth, and FROGGS pops out all ghost-y and grants him three wishes
08:31 hoelzro tadzik: what.
08:31 * hoelzro is off to Youtube
08:31 tadzik hah hah
08:31 tadzik it has colours
08:33 hoelzro mwahaha
08:33 hoelzro Desmond the Moon Bear reprise!
08:34 FROGGS[mobile] masak: what what do I wear in your phantasy?
08:34 FROGGS[mobile] how was she called, beloved ginny?
08:39 FROGGS[mobile] but maybe I would be more like Myrte
08:40 masak :)
08:40 masak FROGGS[mobile]: you're in a long stylish bathrobe which kinda morphs into mobile phone blue smoke.
08:41 FROGGS[mobile] I am fine with that
08:42 FROGGS[mobile] .oO( of course the smoke is blue, it is digital smoke)
08:47 fhelmberger joined #perl6
08:56 masak of course.
09:19 diakopter masak: without looking at the source, I'm guessing that the current macro implementation has to do hoop-jumping to save ASTs..?
09:22 diakopter masak: I'm actually thinking about wicked things like macros working in string evals after they've been declared.. .oO( does that already work,ish? )
09:24 dalek rakudo/nom: 27fcd70 | (Timo Paulssen)++ | src/ (2 files):
09:24 dalek rakudo/nom: WIP: creating buildplan thingies for native attrs.
09:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/27fcd703b9
09:24 dalek rakudo/nom: 2e5e1e3 | (Timo Paulssen)++ | src/core/Mu.pm:
09:24 dalek rakudo/nom: WIP
09:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2e5e1e3a60
09:24 dalek rakudo/nom: 6c853ea | (Timo Paulssen)++ | src/Perl6/Metamodel/BUILDPLAN.nqp:
09:24 dalek rakudo/nom: make native attribute buildplan work on jvm, too.
09:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6c853eabd4
09:24 dalek rakudo/nom: 17ecfcc | (Tobias Leich)++ | src/ (2 files):
09:24 dalek rakudo/nom: Merge pull request #224 from timo/nativeattr_buildplan
09:24 dalek rakudo/nom:
09:24 dalek rakudo/nom: buildplan operations for native attributes
09:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/17ecfcc334
09:25 SamuraiJack_ joined #perl6
09:27 xinming joined #perl6
09:30 masak diakopter: I promise to backlog yesterday properly and ask followup questions when I have the cycles to process your insight.
09:31 masak diakopter: hoop-jumping -- well, a Quasi needs to save the QAST as a data structure, because it has (unquote) holes in it that need to be filled.
09:31 masak diakopter: I can easily believe that you have found a better way to do that.
09:32 diakopter masak: perhaps.. it's pretty tenuous/shaky/amorphous
09:32 diakopter but it may be more elegant, but then again might not be
09:35 diakopter the "easy" way to do it would be to create a closure for each placeholder that runs a routine generated from the expression passed as the argument, and then for each "invocation" of the macro, it creates another closure with those arguments curried in
09:36 diakopter "easy" in quotes because it's not super-efficient..
09:36 daxim joined #perl6
09:36 diakopter (as it could be if the code were duplicated/inlined
09:36 diakopter )
09:37 masak diakopter: I also want to make sure we don't trade away AST introspection.
09:37 sqirrel joined #perl6
09:37 diakopter how is that used?
09:38 diakopter the only "problem" with the way I described it is it creates intermediate frames that the user might not have imagined
09:41 moritz r: class A { has $!x = 42; method get() { ::('$!x') } }; say A.new.get
09:41 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
09:41 camelia ..rakudo-parrot 3cef56: OUTPUT«No such symbol '$!x'␤  in method gist at gen/parrot/CORE.setting:11916␤  in method gist at gen/parrot/CORE.setting:974␤  in sub say at gen/parrot/CORE.setting:12816␤  in block  at /tmp/RqeHydpurZ:1␤  in any  at /tmp/RqeHydpurZ:1␤  in any  at …»
09:41 dakkar joined #perl6
09:41 spider-mario joined #perl6
09:41 moritz should that work?
09:42 moritz r: class A { has $!x = 42; }; my $obj = A.new; say A.^attributes[0].get_value($obj);
09:42 masak moritz: my first thought is "I don't see why it shouldn't".
09:43 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
09:43 camelia ..rakudo-parrot 3cef56: OUTPUT«42␤»
09:43 moritz masak: my first thought was "rakudobug"
09:43 moritz j: say 1
09:43 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
09:43 diakopter moritz: it's been timing out on nearly everything since you went asleep
09:43 masak omg, it's powered by moritz awakitude!
09:44 masak quick, moritz never go to sleep
09:45 masak p: class A { has $!x = 42; method get() { ::('$!x') } }; say A.new.get
09:45 camelia rakudo-parrot 3cef56: OUTPUT«No such symbol '$!x'␤  in method gist at gen/parrot/CORE.setting:11916␤  in method gist at gen/parrot/CORE.setting:974␤  in sub say at gen/parrot/CORE.setting:12816␤  in block  at /tmp/fkp8grFoPq:1␤  in any  at /tmp/fkp8grFoPq:1␤  in any  at …»
09:46 * masak submits rakudobug
09:48 diakopter masak: but what if it's not private?
09:49 masak diakopter: there's always a $!x container, whether it was declared as $!x or $.x
09:50 masak r: class A { has $.x; method foo { say $!x } }; A.new(:x(42)).foo
09:50 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
09:50 camelia ..rakudo-parrot 3cef56: OUTPUT«42␤»
09:50 masak r: class A { has $.x; method foo { say ::('$!x') } }; A.new(:x(42)).foo
09:50 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
09:50 camelia ..rakudo-parrot 3cef56: OUTPUT«No such symbol '$!x'␤  in method gist at gen/parrot/CORE.setting:11916␤  in method gist at gen/parrot/CORE.setting:974␤  in sub say at gen/parrot/CORE.setting:12816␤  in method foo at /tmp/UATA_w4Lf0:1␤  in block  at /tmp/UATA_w4Lf0:1␤  in an…»
09:50 diakopter hm
09:51 moritz rj: 1
09:52 camelia rakudo-jvm 882e33: OUTPUT«(timeout)»
09:52 moritz rj: 1
09:52 camelia rakudo-jvm 882e33: OUTPUT«Error while reading '/home/p6eval_eval/p6eval-token': Permission denied at /home/p6eval/jvm-rakudo/eval-client.pl line 10.␤»
09:52 masak does anyone want to field finanalyst's FAQ on p6l?
09:53 moritz rj: 1
09:53 camelia rakudo-jvm 882e33: ( no output )
09:53 moritz \o/
09:53 masak it's about the "help, I can't fit a bunch of Xs into an Array of X!" again...
09:53 moritz rj: say 42
09:53 camelia rakudo-jvm 882e33: OUTPUT«42␤»
09:53 masak moritz++
09:53 moritz masak: no, that's not the frequently asked question
09:54 moritz the FAQ is why   sub bla(Int @a) { ... }; my @b = 1, 2, 3; bla(@b) doesn't work
09:54 moritz but he actually declared   my Int @b = ...
09:54 moritz so I don't know off-hand want's wrong
09:55 moritz or did I misread the question?
09:56 masak hmmm
09:56 masak does it have anything to do with the gather, perchance?
09:56 masak r: my Int @b = gather for 1, 2, 3 { take $_ }; say @b
09:56 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«1 2 3␤»
09:56 masak guess not.
09:56 moritz ok, bug_4.p6 is exactly the FAQ
09:57 moritz masak: note that the assignment doesn't bomb, but the call
09:58 masak *nod*
09:58 dalek nqp: 0d34a3e | (Tobias Leich)++ | / (2 files):
09:58 dalek nqp: mapped getlexrelcaller
09:58 dalek nqp: review: https://github.com/perl6/nqp/commit/0d34a3e510
09:58 masak if it were up to me, we should never give the error message "expected 'Array[A]' but got 'Array[A]'" :)
09:59 moritz yes, the error reporting is LTA
09:59 moritz no, LTA doesn't cover it
09:59 moritz it sucks.
09:59 masak LTA does cover it, even when it's an understatement.
09:59 masak LTA was the term I was looking for.
10:11 arnsholt Oh, this is the "class + mixin !~~ class + same mixin" problem, isn't it?
10:11 moritz probably
10:12 masak when it's described like that, I'm not sure it's bug rather than a feature.
10:12 masak r: role R {}; say (Int but R) === (Int but R)
10:12 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«True␤»
10:12 masak r: role R {}; say (Int but R) ~~ (Int but R)
10:12 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«False␤»
10:12 masak hrm.
10:12 masak yeah, that's weird.
10:14 pdurbin left #perl6
10:15 arnsholt IIRC jnthn has said it's a bug, but also that it'll be a bit complicated to fix properly
10:18 * masak can't recall whether it's in RT already
10:19 FROGGS I'd think you would have to check if A consists of the same parts as B, rather than to check if they have the same WHICH
10:19 masak that sounds... risky.
10:19 FROGGS hmmm, why?
10:20 FROGGS the problem is that you create a new thing in a signature, so it is always a new thing and will never match others
10:21 masak by "same parts", do you mean "same mixins"?
10:21 masak because you also have to check the order in which they were applied, and that order *may or may not matter*.
10:22 FROGGS identical mixins
10:22 FROGGS hmmm
10:22 FROGGS well, if the order matters it must be checked, yes :o)
10:23 FROGGS basically you would have to compare how both things where composed
10:23 FROGGS so, do we record that?
10:23 FROGGS like the build-plan
10:24 FROGGS the blueprint
10:24 darutoko joined #perl6
10:31 donaldh joined #perl6
10:38 colomon joined #perl6
10:41 wsri joined #perl6
10:43 moritz the merge base for Arcterus' pull request is from July
10:44 moritz no wonder I get lots of conflicts while merging
10:46 masak maybe rebase it locally?
10:47 moritz problem is, I don't know much about the code he is touching
10:47 moritz that makes it much harder for me to resolve conflicts
10:48 masak *nod*
11:08 denisboyun joined #perl6
11:16 fhelmberger joined #perl6
11:18 lizmat joined #perl6
11:25 diakopter Beyond Less Than Awesome
11:27 lue joined #perl6
11:27 rindolf joined #perl6
11:28 masak well, don't forget that "Less Than Awesome" was always meant as an understatement, implicitly meaning "this sucks".
11:28 masak it's only because it's now become so ingrained that we feel it's not enough.
11:31 pernatiy joined #perl6
11:44 dalek roast: 92e4192 | (Elizabeth Mattijsen)++ | S02-types/whatever.t:
11:44 dalek roast: Unfudge now passing tests, timotimo++
11:44 dalek roast: review: https://github.com/perl6/roast/commit/92e419224d
11:46 lizmat did we recently add § and ¶ as punctuation chars ?
11:46 lizmat sorry, remove as punctuation chars ?
11:47 lizmat t/spec/S05-mass/charsets.t is failing test #10 because of that, at least on parakudo
11:49 lizmat or did lue forget to add a "2" to the todo
11:53 pochi_ joined #perl6
11:53 jnthn_ joined #perl6
11:54 mathw_ joined #perl6
11:54 [particle]1 joined #perl6
11:57 dalek nqp: ffcebf9 | (Tobias Leich)++ | src/vm/ (24 files):
11:57 dalek nqp: make nqp::backendconfig available for all backends
11:57 dalek nqp: review: https://github.com/perl6/nqp/commit/ffcebf9ef3
11:58 jferrero joined #perl6
11:59 dalek roast: 3e12598 | (Elizabeth Mattijsen)++ | S02-types/native.t:
11:59 dalek roast: Added tests for #102416
11:59 dalek roast: review: https://github.com/perl6/roast/commit/3e12598206
11:59 synopsebot Link: https://rt.perl.org/rt3//Publ​ic/Bug/Display.html?id=102416
11:59 FROGGS nqp: say($_.key) for nqp::backendconfig() # this should work in a few(?) minutes
11:59 camelia nqp-moarvm: OUTPUT«ccoptiflags␤uvrule␤tomobjects␤ldshared​␤sharule␤ccinstflags␤ldlibs␤ldimp␤shaobjects␤​hasreadline␤mtclean␤ccdefflags␤obj␤ldflags␤ld​dir␤name␤noreturnattribute␤dllexport␤defs[0]␤​defs[1]␤dcclean␤moar␤ldrpath␤dcrule␤ldusr…»
11:59 camelia ..nqp-parrot: OUTPUT«Error while compiling block : Error while compiling op for (source text: "say($_.key) for nqp::backendconfig() # this should work in a few(?) minutes"): Error while compiling op backendconfig (source text: "nqp::backendconfig()"): No registered operation handler f…»
11:59 camelia ..nqp-jvm: OUTPUT«No registered operation handler for 'backendconfig'␤  in compile_op (gen/jvm/stage2/QAST.nqp:231)␤  in as_jast (gen/jvm/stage2/QAST.nqp:3696)␤  in as_jast (gen/jvm/stage2/QAST.nqp:2971)␤  in  (gen/jvm/stage2/QAST.nqp:1207)␤  in  (gen/jvm/stage2/QAST.nqp:228…»
12:02 lizmat n: my @a = 0..4; @a.=[1, 2];
12:02 camelia niecza v24-98-g473bd20: ( no output )
12:02 lizmat n: my @a = 0..4; @a.=[1, 2]; say @a
12:02 camelia niecza v24-98-g473bd20: OUTPUT«1 2␤»
12:02 lizmat n: my @a = 0..4; @a.=[1, 2]; say @a.perl
12:02 camelia niecza v24-98-g473bd20: OUTPUT«[1, 2].list␤»
12:03 lizmat n: my @a = 0..4; @a=[1, 2]; say @a.perl
12:03 camelia niecza v24-98-g473bd20: OUTPUT«[[1, 2]].list␤»
12:03 lizmat n: my @a = 0..4; @a=(1, 2); say @a.perl
12:03 camelia niecza v24-98-g473bd20: OUTPUT«[1, 2].list␤»
12:03 lizmat n: my @a = 0..4; my $b=[1,2]; @a.=$b; say @a.perl
12:04 camelia niecza v24-98-g473bd20: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in type Array␤  at /tmp/L7VIX4riYw line 1 (mainline @ 5) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4583 (ANON @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4584 (modul…»
12:04 lizmat r: my @a = 0..4; my $b=[1,2]; @a.=$b; say @a.perl
12:04 camelia rakudo-jvm 882e33: OUTPUT«No such method 'postcircumfix:<( )>' for invocant of type 'Array'␤  in any  at gen/jvm/BOOTSTRAP.nqp:1658␤  in method dispatch:<.=> at gen/jvm/CORE.setting:1129␤  in block  at /tmp/qb9GOnVq4z:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in …»
12:04 camelia ..rakudo-parrot 3cef56: OUTPUT«No such method 'postcircumfix:<( )>' for invocant of type 'Array'␤  in any  at gen/parrot/BOOTSTRAP.nqp:1687␤  in any  at gen/parrot/BOOTSTRAP.nqp:1669␤  in method dispatch:<.=> at gen/parrot/CORE.setting:1138␤  in block  at /tmp/_vg0NCwkvs:1␤ …»
12:05 lizmat masak: apart from it being legal in STD, what does the .=[] syntax bring exactly?
12:05 FROGGS r: my @a; my $b = 'perl'; say @a.=$b
12:05 camelia rakudo-parrot 3cef56: OUTPUT«No such method 'postcircumfix:<( )>' for invocant of type 'Str'␤  in any  at gen/parrot/BOOTSTRAP.nqp:1687␤  in any  at gen/parrot/BOOTSTRAP.nqp:1669␤  in method dispatch:<.=> at gen/parrot/CORE.setting:1138␤  in block  at /tmp/eA_xRF628z:1␤  i…»
12:05 camelia ..rakudo-jvm 882e33: OUTPUT«No such method 'postcircumfix:<( )>' for invocant of type 'Str'␤  in any  at gen/jvm/BOOTSTRAP.nqp:1658␤  in method dispatch:<.=> at gen/jvm/CORE.setting:1129␤  in block  at /tmp/j8PRltvuN6:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in an…»
12:05 lizmat this wrt #120383
12:05 synopsebot Link: https://rt.perl.org/rt3//Publ​ic/Bug/Display.html?id=120383
12:06 masak lizmat: it allows me to reassign an array to a slice of that array.
12:06 masak lizmat: @a.=[@slice] is a short form for @a = @a[@slice]
12:06 masak lizmat: beyond that, I don't quite understand the question :)
12:07 lizmat ah, ok, better use different values in examples then, as in this case the indexes where the same as the values
12:07 lizmat which sort of clouds the issue  :-)
12:07 masak ah; sorry, didn't mean to use a confusing example... :)
12:07 * lizmat writes a test
12:07 masak lizmat++
12:10 lizmat this looks indeed as fallout from the [] {} method -> sub migration, wouldn't you say, masak ?
12:14 masak that was my suspicion, but I wasn't sure.
12:14 masak lunch &
12:31 dalek nqp: c91640d | (Tobias Leich)++ | docs/ops.markdown:
12:31 dalek nqp: document getlex_n<[inso]> and bindlex_n<[inso]>
12:31 dalek nqp: review: https://github.com/perl6/nqp/commit/c91640d748
12:31 * [Coke] ~~ and heads off to work.
12:35 arnsholt Does anyone happen to know who created most of the tags in the NQP issues queue?
12:36 FROGGS arnsholt: you mean the seven labels?
12:36 arnsholt Yeah
12:36 FROGGS no, is that important to know?
12:36 arnsholt Or more specifically, the all uppercase ones (TESTNEEDED, PARROT)
12:37 arnsholt Just curious. I'd prefer them to be cased normally (and maybe even a space, "Test needed"), but I'm not sure if it's intended on jnthn's part or some such
12:37 FROGGS http://irclog.perlgeek.de/​perl6/2013-10-21#i_7740656
12:37 arnsholt Oooh, cheers!
12:38 arnsholt I suspect the all uppercase may have happened because initially we had two labels: JVM and LHF
12:38 FROGGS yeah
12:38 FROGGS but I'd prefer "Parrot" too
12:39 * arnsholt renames
12:39 arnsholt There
12:44 dalek roast: cc153df | (Elizabeth Mattijsen)++ | S09-subscript/slice.t:
12:44 dalek roast: Added tests for #120383
12:44 dalek roast: review: https://github.com/perl6/roast/commit/cc153df758
12:44 synopsebot Link: https://rt.perl.org/rt3//Publ​ic/Bug/Display.html?id=120383
12:49 dalek nqp: dcfdfe2 | (Tobias Leich)++ | docs/ops.markdown:
12:49 dalek nqp: fix getlex and bindlex
12:49 dalek nqp: review: https://github.com/perl6/nqp/commit/dcfdfe22f4
13:05 tadzik synopsebot, u so clever
13:10 jnap joined #perl6
13:11 fhelmberger_ joined #perl6
13:11 rurban joined #perl6
13:18 _ilbot joined #perl6
13:18 Topic for #perl6 is now »ö« Welcome to Perl 6! | http://perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:,  niecza:, std:, or /msg camelia perl6: ... | irclog: http://irc.perl6.org | UTF-8 is our friend!
13:21 fhelmberger joined #perl6
13:38 colomon joined #perl6
13:40 dalek nqp: 61e97a9 | (Tobias Leich)++ | docs/ops.markdown:
13:40 dalek nqp: documented getlex[outer|caller|rel|reldyn|...]
13:40 dalek nqp: review: https://github.com/perl6/nqp/commit/61e97a9483
13:41 colomon JSON::Tiny is failing two tests?
13:41 colomon the smoker is completely borked because it cannot rebuild panda right now...
13:42 bluescreen10 joined #perl6
13:42 FROGGS colomon: can you already paste something wrt panda?
13:43 colomon ==> Testing JSON::Tiny
13:43 colomon t/01-parse.t ...... Failed 2/90 subtests
13:43 colomon is the relevant bit, I think
13:43 benabik joined #perl6
13:46 fridim_ joined #perl6
13:46 ajr joined #perl6
13:48 lizmat didn't moritz recently committed some sigwhitespace related fixes for JSON::Tiny ?
13:48 lizmat *commit
13:49 rurban joined #perl6
14:10 denis_boyun joined #perl6
14:15 PacoAir joined #perl6
14:22 kaare__ joined #perl6
14:28 sqirrel joined #perl6
14:30 masak r: my &oct = { :8(~$_) }; my &dec = { :10(~$_) }; say (oct 31) == (dec 25)
14:31 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«True␤»
14:31 masak I couldn't get :8(~*) to work; nor do I know if it should...
14:34 lizmat so does that mean Christmas is 3 days from now?
14:39 PerlJam On #perl6, it's always Christmas  :)
14:40 moritz lizmat: I did
14:43 rurban_ joined #perl6
14:44 lizmat re: #120383, the error is generated in src/Perl6/Actions, line 3753
14:44 synopsebot Link: https://rt.perl.org/rt3//Publ​ic/Bug/Display.html?id=120383
14:44 lizmat but I think the cause is deeper, since the past.op *not* be callmethod at all in the case of .=[]
14:45 lizmat *should
14:45 rurban_ pointme: dalek
14:45 lizmat but rather, it should be 'call'
14:46 masak PerlJam: Christmas is already here, it's just unevenly distributed ;)
14:47 masak rurban_: hm, 'pointme' sounds familiar -- what does/did it do?
14:47 masak r: my @a = <foo bar baz>; say @a.[1]
14:47 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«bar␤»
14:47 masak r: my @a = <foo bar baz>; @a.=[1]; say @a
14:47 PerlJam masak: exactly!
14:47 camelia rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/KPgpK0fADi�Cannot use .= on a non-identifier method call�at /tmp/KPgpK0fADi:1�------> [32mmy @a = <foo bar baz>; @a.=[1][33m�[31m; say @a[0m�»
14:47 camelia ..rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/2UMo_8JCmd�Cannot use .= on a non-identifier method call�at /tmp/2UMo_8JCmd:1�------> [32mmy @a = <foo bar baz>; @a.=[1][33m�[31m; say @a[0m�»
14:48 masak moritz: do the above two outputs differ? oh! the temp files :/
14:48 masak lizmat: seems to me that whatever mechanism @a.[1] uses, @a.=[1] should use, too.
14:51 lizmat moritz: I guess the tempfile names should be normalized in the comparison ?
14:52 rurban_ masak: It expanded the src url for the bots and more
14:52 rurban_ I couldn't find dalek on the web
14:53 FROGGS dalek: help
14:56 diakopter dalek itself just echos lines from a file # sorear++
15:14 lowpor03 joined #perl6
15:14 tipdbmp joined #perl6
15:21 dalek rakudo/moar-support: 753af50 | (Tobias Leich)++ | src/Perl6/ (2 files):
15:21 dalek rakudo/moar-support: added QAST::VM variants for moar idential to jvm
15:21 dalek rakudo/moar-support: review: https://github.com/rakudo/rakudo/commit/753af50813
15:23 dalek rakudo/moar-support: 6e556ae | (Tobias Leich)++ | src/vm/moar/ (2 files):
15:23 dalek rakudo/moar-support: added p6capturelex, which really needs reviewing
15:23 dalek rakudo/moar-support: review: https://github.com/rakudo/rakudo/commit/6e556aea4c
15:23 lizmat seems the storm has died down here a bit, so we're out cycling&
15:23 FROGGS lizmat: keep away from trees please :/
15:24 lizmat we'll trie  :-)
15:24 FROGGS :P
15:25 masak we have a nice fresh wind going here too. it's rattling against the window panes.
15:25 TimToady diakopter++ should also keep away from trees right now
15:30 lowpor03 joined #perl6
15:32 diakopter TimToady++ should also tree away from keeps
15:33 * TimToady might get more writing done in a keep...
15:33 TimToady killing the trees comes later
15:33 raiph joined #perl6
15:33 timotimo moritz: you'll be glad to know, that the workaround you put into DBIish today for rakudo star can already be removed again for rakudo nom :)
15:34 diakopter ..but them your knowledge of Perl 6 will become outdated.. oh wait....
15:35 TimToady diakopter: maybe you should pull down one of those oxygen masks
15:36 diakopter can't reach it from lying flat
15:36 diakopter ..out
15:36 diakopter er, *laying
15:36 diakopter er
15:36 TimToady either way you're hosed in English
15:37 TimToady but I'm less likely to mistake you for a hen than for a liar :)
15:37 bluescreen10 joined #perl6
15:37 TimToady so I'd go with "laying" even if the proscriptivists proscribe it
15:38 dalek roast: 9b8425a | (Timo Paulssen)++ | S12-construction/construction.t:
15:38 dalek roast: tests for native attribute BUILDPLAN.
15:38 dalek roast: review: https://github.com/perl6/roast/commit/9b8425a528
15:42 atroxaper joined #perl6
15:42 atroxaper Hello #perl6
15:43 diakopter pregramming is hard; let's go scribing
15:43 tadzik scrabbling
15:44 masak scrobbling
15:44 TimToady atroxaper: er, hello
15:44 atroxaper Who can tell me how i can get something via network. Some http get request. I have tried to install LWP::Simple but panda said that tests fails.
15:46 FROGGS atroxaper: I'd propose to fix LWP::Simple
15:47 FROGGS (together)
15:47 tadzik that'd be awesome
15:47 tadzik as far as I remember from porting it to JVM, the chunked transfers seem to behave a bit strangely
15:47 ajr joined #perl6
15:47 FROGGS tadzik: is that a known issue?
15:47 tadzik yes :(
15:47 FROGGS :/
15:47 tadzik maybe we need LWP::Simplier
15:47 diakopter "patches compelled.."
15:47 benabik Heh.  My terminal says: ./perl6-p CORE.setting has stages start, parse, syntaxcheck, ast, optimize, jast, classfile, jar
15:48 benabik (The lovely confusion of parallel make.)
15:48 timotimo %)
15:48 FROGGS atroxaper: so, you are on the jvm?
15:48 tadzik or look how panda fetches files over http
15:48 tadzik (prepare a paperbag)
15:49 diakopter .. and a caldera
15:49 tadzik https://github.com/tadzik/panda/blo​b/master/lib/Panda/Ecosystem.pm#L54 ;)
15:49 atroxaper FROGGS: no. I have win R* "This is perl6 version 2013.05 built on parrot 5.3.0 revision 0"
15:49 FROGGS atrodo: ahh, a bit outdated though...
15:49 tadzik I remember hacking Curl binding with bbkr++ on PLPW, but I don't remember if we got it to semi-work
15:50 timotimo yeah, that's quite old
15:50 FROGGS atroxaper: I think I got a 2013.09 on a windows box...
15:50 diakopter FROGGS: ah well, I'll just  transliterate the parrot chain, since I can sorta tell what it does, but don't understand why
15:50 timotimo windows, though ...
15:50 tadzik https://github.com/bbkr/cURL may be worth a try
15:51 * FROGGS starts his windows
15:52 timotimo gentlemen, start your windows
15:53 atroxaper FROGGS: i'm installing 2013.09 R* now
15:53 FROGGS k
15:56 ajr_ joined #perl6
15:57 sqirrel joined #perl6
15:59 _ilbot joined #perl6
15:59 Topic for #perl6 is now »ö« Welcome to Perl 6! | http://perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:,  niecza:, std:, or /msg camelia perl6: ... | irclog: http://irc.perl6.org | UTF-8 is our friend!
15:59 atroxaper R* 2013.09 is even stranger... https://gist.github.com/atroxaper/7199471
16:00 btyler joined #perl6
16:02 Rotwang joined #perl6
16:02 Rotwang joined #perl6
16:02 FROGGS ahh, I know that one...
16:03 FROGGS right after we made the 2013.09, we discovered that unlink does not delete protected files...
16:03 FROGGS it was fixed a short while after the release though
16:03 timotimo that seems to happen often :P
16:03 timotimo that kind of thing
16:03 FROGGS yeah
16:04 timotimo just like all the unexpected fallout from the postcircumfix changes
16:05 atroxaper FROGGS: what i should do with that?)
16:05 FROGGS atroxaper: I am currently thinking about creating star 2013.10
16:06 FROGGS does somebody know if lue got through it?
16:06 timotimo not yet
16:06 atroxaper It would be nice)
16:06 timotimo 06:27 < lue> (in any case, I suspect 12-15 at least will happen tomorrow)
16:07 FROGGS k
16:08 timotimo did we get a fix for .=[] and such in?
16:08 timotimo otherwise i could probably look at that now
16:08 FROGGS timotimo: well, 2013.10 happened like more than a week ago
16:09 timotimo er, i mean into nom
16:09 timotimo today
16:10 bluescreen10 joined #perl6
16:11 FROGGS I'd still would build a star for 2013.10
16:11 atroxaper FROGGS: So, should i wait 2013.10? Ans as i understand LWP:Simple is corrupted now?
16:12 FROGGS atroxaper: I dont know what is wrong with LWP::Simple@rakudo-parrot
16:12 FROGGS you can always build rakudo HEAD, and then fetch panda on your own
16:13 FROGGS if you have git and MSVC and activeperl, and ...
16:14 atroxaper FROGGS: ok. I'l try. It's will a usefull experiens i think )
16:14 FROGGS atroxaper: it is definitely better than waiting a month for a fix :o)
16:14 FROGGS and yes, you only an learn
16:15 atroxaper Do you recomend to use Rakudo.jvm already as well?
16:15 FROGGS depends
16:15 timotimo kind of a sad state, but understandable because our brightest engineers are currently working on moar rather than rakudo itself
16:15 FROGGS if you have long running stuff, then yes
16:16 FROGGS for super-short scripts parrot is better
16:16 xinming joined #perl6
16:18 atroxaper FROGGS: i read that there is a way to call java statements from Rakudo.jvm now. Is there way to do vice versa stuff?
16:19 atroxaper FROGGS: once we started to talk )
16:19 FROGGS atroxaper: I dont think so, but my knowledge about that is rather limited
16:20 atroxaper FROGGS: ok. Thank you. Have a good time
16:20 FROGGS atroxaper: you might want to watch jnthn++'s talk videos about jvm interop
16:20 timotimo perl6 level subs are exposed as something that implements Callable to java, so you can at least have callbacks into perl6 code
16:20 FROGGS :o)
16:21 atroxaper timotimo: thank you too )
16:21 atroxaper FROGGS: where is the video?
16:21 timotimo you can find it on youtube, i believe
16:22 xinming joined #perl6
16:23 atroxaper timotimo: yes. I found i think) Thank you all. Going to learn...
16:23 timotimo enjoy! :)
16:23 timotimo perl6 is cool as long as you can dance around the bugs that inevitably pop up everywhere :P
16:23 * timotimo is just a bit frustrated from yesterday's experience
16:24 FROGGS atroxaper: have fun! :o)
16:24 atroxaper o/
16:24 TimToady \o
16:25 FROGGS timotimo: yeah :/
16:27 FROGGS timotimo: but keep in mind, it is all TimToady's fault :o)
16:27 timotimo at least i got a few good commits out of the experience, so that's good! :)
16:27 iSlug joined #perl6
16:33 dalek nqp: c119b90 | (Tobias Leich)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp:
16:33 dalek nqp: stole set_[core|hll]_op_result_type, attach_result_type from jvm
16:33 dalek nqp: review: https://github.com/perl6/nqp/commit/c119b90f08
16:33 SamuraiJack_ joined #perl6
16:40 TimToady Rule 1: It's all TimToady's fault.  Rule 2: Even if TimToady repents, it's still his fault for not repenting sooner.
16:42 FROGGS well, you could have invented PHP and then we would not have these types of problems :o)
16:42 TimToady no, I ended up with a problem of types instead
16:44 japhb__ That sounds like a good title: "A Problem of Types"
16:44 japhb__ I'm not sure if that should be a paper, a book, or an art film.
16:44 FROGGS TimToady: do you have an opinion about this? http://irclog.perlgeek.de/​perl6/2013-10-28#i_7775529 (10:11 to 10:24)
16:45 * timotimo took the time to read and fill out the CLA
16:45 japhb__ timotimo: Did you find the source of http://irclog.perlgeek.de/​perl6/2013-10-27#i_7773084 ?
16:46 japhb__ timotimo: Does perl6-bench need unbitrotting?
16:47 japhb__ (That's in regards to http://irclog.perlgeek.de/​perl6/2013-10-27#i_7773538 _
16:47 japhb__ s/_/)/
16:47 timotimo japhb__: did not find the source, sorry :(
16:48 japhb__ timotimo: re: http://irclog.perlgeek.de/​perl6/2013-10-27#i_7773747 , there is some sample code, but no one has added it to rakudo-jvm yet, IIRC
16:48 timotimo the second one: run seems to no longer give the changes to the environment to the processes it runs
16:48 japhb__ Gah.
16:49 japhb__ But shell does?
16:49 timotimo i think so
16:49 japhb__ Rather than work around in perl6-bench, I'd like to just fix the regression in Rakudo.
16:50 timotimo that would be good, yes
16:50 timotimo there's one regression we hit in perl6-bench that i've fixed already =)
16:50 japhb__ Oh?
16:50 japhb__ PR?
16:50 timotimo already merged
16:50 timotimo into rakudo nom, that is
16:50 japhb__ Oh, gotcha
16:50 timotimo the *.[] and *.{} thing
16:51 japhb__ And thank you for that.  :-)
16:51 masak r: my @a = <foo bar baz>; @a.=[1]; say @a
16:51 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/zDOY2XLsZ9�Cannot use .= on a non-identifier method call�at /tmp/zDOY2XLsZ9:1�------> [32mmy @a = <foo bar baz>; @a.=[1][33m�[31m; say @a[0m�»
16:51 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/OCC19FtPba�Cannot use .= on a non-identifier method call�at /tmp/OCC19FtPba:1�------> [32mmy @a = <foo bar baz>; @a.=[1][33m�[31m; say @a[0m�»
16:51 timotimo oh wow, i scanned that PDF in super-high-res
16:52 masak timotimo: you're free to look at @a.=[1], though. it's in a recent RT ticket, too.
16:52 masak timotimo: lizmat looked at it in today's backlog.
16:52 timotimo i seem to recall that
16:52 masak moritz: how hard would it be to get camelia to use the same /tmp/* filename, so that rakudo-jvm and rakudo-parrot yield the same output when possible?
16:52 timotimo she didn't attempt a fix?
16:53 * japhb__ contemplates whether direct Rakudo hacking to fix problems, or creating a list of the top most painful Rakudo issues with hopes of others working on them, is a more effective use of my tuits today.
16:54 * japhb__ notes that in camelia's last response, the two rakudos had different sha1s.
16:56 timotimo lizmat said the error comes from dotty:sym<.*>, but the error message doesn't match up with what i get from (my @a = 1, 2, 3).=[1]
16:56 timotimo oh, because that still tries to find a method, righto
16:58 timotimo oh, curious! there's different error cases from .=[1,2] and .=[1]
16:58 japhb__ timotimo: Looked through my notes.  The rp open sketch was at: https://gist.github.com/donaldh/6901992
17:00 [Coke] I actually slightly preferred the all caps tags, but am fine with the change.
17:01 dalek perl6-roast-data: 189a03a | coke++ | / (5 files):
17:01 dalek perl6-roast-data: today (automated commit)
17:01 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/189a03a99d
17:01 dalek perl6-roast-data: 173d2ee | coke++ | p (2 files):
17:01 dalek perl6-roast-data: today (automated commit)
17:01 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/173d2ee55a
17:02 timotimo japhb__: i don't understand why from and to have the same implementation o_O
17:02 [Coke] preflex: seen au?
17:02 preflex au was last seen on #perl6 246 days, 22 hours, 4 minutes and 28 seconds ago, saying: pity that Unicode terminals doesn't support strike-through very well :)
17:02 [Sno] joined #perl6
17:03 [Coke] https://raw.github.com/coke/perl6-​roast-data/master/pugs_summary.out - well, that's disappointing.
17:07 * masak .oO( au has been implementing strike-through in Unicode terminals for 246 days!? )
17:08 dalek nqp: 353787c | (Tobias Leich)++ | src/vm/moar/QAST/QASTCompilerMAST.nqp:
17:08 dalek nqp: added sub want, needed for coercion
17:08 dalek nqp: review: https://github.com/perl6/nqp/commit/353787ce87
17:09 timotimo Use of uninitialized value $thirdpartyjars in join or string at tools/build/create-jvm-runner.pl line 42.
17:09 timotimo what's this all about?
17:10 timotimo probably just need to reConfigure.pl
17:14 [Coke] geekosaur: hey. You're a haskell person!
17:15 geekosaur I wonder if we can get away with upgrading pugs to work with modern stuff instead of complaining about all the deprecated stuff
17:16 geekosaur in fact, some of it we're going to have to fix; as of the next ghc version, manual Typeable instances are forbidden
17:16 geekosaur deriving (Typeable) is the only permitted way to do it
17:16 TimToady someone should just write a Haskell backend for rakudo now :)
17:16 FROGGS ohh noes /o\
17:16 TimToady how hard can it be?
17:17 * masak .oO( GHC-hard )
17:17 [Coke] so easy, even TimToady can do it. :P
17:17 k1lldash9 joined #perl6
17:17 * geekosaur might find that easier than trying to understand au++ --- clearly he is not nearly as smart...
17:17 [Coke] geekosaur: updates to pugs that don't fail more spectests always welcome.
17:18 [Coke] how can I find out why https://gist.github.com/coke/7200634 has the locale error it does? is there a way to see what locales are installed on the box I'm testing?
17:19 geekosaur look in /usr/share/locale usually
17:19 k1lldash9 locale should show you
17:19 BenGoldberg joined #perl6
17:19 geekosaur I was starting to say that over in #haskell when you pinged me over here :)
17:21 * masak .oO( the peoples of #perl6 and #haskell have always had friendly relations with each other, despite their obvious evolutionary differences )
17:21 TimToady .oO(whitespace as syntax?!?)
17:22 geekosaur someone alredy pointed at default(42) vs. default (42) last night, and IIRC there is similar with fucntion calls; we already have whitespace as syntax
17:22 TimToady .oO(indentation as syntax?!?)
17:23 masak TimToady: heredocs.
17:23 TimToady there's a huge difference between "Whitespace, yes or no?" and "Whitespace, how much?"
17:23 kaare_ joined #perl6
17:24 TimToady masak: the exception that proves the rule :P
17:24 masak :P
17:24 * masak .oO( a *true* whitespace Scotsman...! )
17:24 BenGoldberg I understand the difference between with-whiteapce and without-whitespace for function calls.
17:24 BenGoldberg But not default(42) and default (42)
17:25 TimToady again, it's the difference between :default(42) and :default (42)
17:25 TimToady we don't allow ws inside colonpairs either
17:26 BenGoldberg r: my @a is default (42); say @a[5]
17:26 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«True␤»
17:26 BenGoldberg r: my @a is default (42); say @a
17:26 TimToady r: say :default
17:26 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«␤»
17:26 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«␤»
17:26 TimToady r: say (:default)
17:26 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«"default" => Bool::True␤»
17:26 BenGoldberg r: my @a is default; say @a
17:26 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«␤»
17:27 BenGoldberg What is 'is default' with no argument supposed mean, anyway?
17:27 TimToady exactly the same as default(True)
17:27 TimToady I keep sayin'
17:27 BenGoldberg Ahh, ok, I get it
17:27 BenGoldberg r: my @a is default; say @a[1..5]
17:28 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«True True True True True␤»
17:28 BenGoldberg r: my @a isn't default; say @a[1..5]
17:28 camelia rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/C79jY_FJvuâ�¤Confusedâ�¤at /tmp/C79jY_FJvu:1â�¤------> [32mmy @a is[33mâ��[31mn't default; say @a[1..5][0mâ�¤    expecting any of:â�¤        whitespaceâ�¤Â»
17:28 camelia ..rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/s3W1vSBvXqâ�¤Confusedâ�¤at /tmp/s3W1vSBvXq:1â�¤------> [32mmy @a is[33mâ��[31mn't default; say @a[1..5][0mâ�¤    expecting any of:â�¤        whitespaceâ�¤Â»
17:28 BenGoldberg :)
17:28 TimToady thing is, the parser doesn't catch the ttiar because siggies are parsed differently
17:28 TimToady std: my @a is default (42);
17:28 camelia std 8adbc60: OUTPUT«ok 00:01 124m␤»
17:29 BenGoldberg r: say True(42)
17:29 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&True' called (lines 1, 1)␤»
17:29 TimToady r: my @a is default (42); @a = 43;
17:29 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:29 BenGoldberg Identical error messages!  A first! :)
17:29 TimToady is probably just ignoring the subsig
17:30 BenGoldberg r: my @a isn't default(False); say @a[1..5]
17:30 camelia rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/GAZSiNU7VKâ�¤Confusedâ�¤at /tmp/GAZSiNU7VK:1â�¤------> [32mmy @a is[33mâ��[31mn't default(False); say @a[1..5][0mâ�¤    expecting any of:â�¤        whitespaceâ�¤Â»
17:30 camelia ..rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/9JdOxhsRwMâ�¤Confusedâ�¤at /tmp/9JdOxhsRwM:1â�¤------> [32mmy @a is[33mâ��[31mn't default(False); say @a[1..5][0mâ�¤    expecting any of:â�¤        whitespaceâ�¤Â»
17:30 BenGoldberg r: my @a is default(False); say @a[1..5]
17:30 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«False False False False False␤»
17:30 BenGoldberg r: my @a is default(False) (42); say @a[1..5]
17:30 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«False False False False False␤»
17:30 TimToady r: my @a is default ($b,$c); @a = 43;
17:30 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:30 TimToady r: my @a ($b,$c); @a = 43;
17:30 BenGoldberg r: my @a is default(False) (now see here); say @a[1..5]
17:30 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:30 camelia rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/GjuRKuV2I7�Invalid typename 'now' in parameter declaration.�at /tmp/GjuRKuV2I7:1�------> [32mmy @a is default(False) (now[33m�[31m see here); say @a[1..5][0m�»
17:30 camelia ..rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/uV0NSK4UIg�Invalid typename 'now' in parameter declaration.�at /tmp/uV0NSK4UIg:1�------> [32mmy @a is default(False) (now[33m�[31m see here); say @a[1..5][0m�»
17:31 TimToady r: my @a ($b,$c); @a := 43;
17:31 camelia rakudo-parrot 3cef56: OUTPUT«Type check failed in binding; expected 'Positional' but got 'Int'␤  in block  at /tmp/68zqDAlxk7:1␤  in any  at /tmp/68zqDAlxk7:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/stage2/NQPHLL.nqp:1133␤  in any evalfil…»
17:31 camelia ..rakudo-jvm 882e33: OUTPUT«Type check failed in binding; expected 'Positional' but got 'Int'␤  in block  at /tmp/GrkMuoPulC:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/stage2/NQPHLL.nq…»
17:31 TimToady r: my $a ($b,$c); @a := 43;
17:31 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/cEkXHZ80_Yâ�¤Variable '@a' is not declared. Did you mean '$a'?â�¤at /tmp/cEkXHZ80_Y:1â�¤------> [32mmy $a ($b,$c); @a[33mâ��[31m := 43;[0mâ�¤    expecting any of:â�¤        postfixâ�¤Â»
17:31 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/6xKUvOGmMMâ�¤Variable '@a' is not declared. Did you mean '$a'?â�¤at /tmp/6xKUvOGmMM:1â�¤------> [32mmy $a ($b,$c); @a[33mâ��[31m := 43;[0mâ�¤    expecting any of:â�¤        postfixâ�¤Â»
17:31 BenGoldberg r: my @a is default(False) ($do-i-need-to-declare-this); say @a[1..5]
17:31 TimToady r: my $a ($b,$c); $a := 43;
17:31 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«False False False False False␤»
17:31 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:31 BenGoldberg r: my @a is default(False) ($do-i-need-to-declare-this, $how-bizzarre-how-bizzarre); say @a[1..5]; # :)
17:31 TimToady r: my $a ($b,$c) = 43;
17:31 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«False False False False False␤»
17:31 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:32 TimToady yeah, seems to ignore the subsig entirely
17:32 TimToady r: my $a ($b,$c); $b = 43;  say $b
17:32 camelia rakudo-jvm 882e33: OUTPUT«java.lang.NullPointerException␤  in block  at /tmp/GlPGERLxuQ:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/stage2/NQPHLL.nqp:1194␤  in any command_eval at s…»
17:32 camelia ..rakudo-parrot 3cef56: OUTPUT«(signal SEGV)»
17:32 TimToady wahoo
17:33 arnsholt jnthn: I've been pondering a refactor of NQP/QRegex so that HLL::Grammar.O takes actual named arguments, instead of parsing a string. Nay or yay?
17:33 BenGoldberg n: my @a is default(False) ($do-i-need-to-declare-this); say @a[1..5]
17:33 camelia niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m===[0m��Trait default not available on variables at /tmp/2PlGtcgkDI line 1:�------> [32mault(False) ($do-i-need-to-declare-this)[33m�[31m; say @a[1..5][0m��Postconstraints, and shapes on variable declarators NYI …»
17:33 BenGoldberg n: my $a ($b, $c); $a = 42; say $a
17:33 camelia niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m​===[0mâ�¤â�¤Postconstraints, and shapes on variable declarators NYI at /tmp/URPGqNsiwj line 1:â�¤------> [32mmy $a ($b, $c)[33mâ��[31m; $a = 42; say $a[0mâ�¤â�¤Potential difficulties:â�¤  $b is declared but not used at /tmp/…»
17:33 BenGoldberg std: my $a ($b, $c); $a = 42; say $a
17:33 camelia std 8adbc60: OUTPUT«ok 00:01 127m␤»
17:34 BenGoldberg r: my $a ($b, $c); $b = 42; say $b
17:34 camelia rakudo-parrot 3cef56: OUTPUT«(signal SEGV)»
17:34 camelia ..rakudo-jvm 882e33: OUTPUT«java.lang.NullPointerException␤  in block  at /tmp/eAZa2puHK4:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/stage2/NQPHLL.nqp:1194␤  in any command_eval at s…»
17:34 masak whoa.
17:34 * masak submits rakudobug
17:36 TimToady FROGGS: re mixin identities, structural equivalence tends to be a world of hurt for everyone involved, especially implementors, so I'm inclined to go with name equivalence here, i.e. you want the same mixin, then name it.
17:36 ssutch joined #perl6
17:36 TimToady that being said, if the optimizer wants to combine identical mixins to save space, we might want to rethink that a bit
17:37 masak feels weird to rely on though, if it's an optimization.
17:37 masak so conservatism seems to favor "no for now"
17:37 jnthn What on earth does "my $a ($b, $c)" even mean?
17:38 * TimToady is inclined to think that, since classes are mutable, they should not automatically try to share
17:39 TimToady otherwise we might end up with rubyesque difficulties
17:39 benabik jnthn: $b := a[0]; $c := $a[1] # ?
17:40 TimToady well, presumably the same as in sub foo ($a ($b, $c)) {...}, only different :)
17:42 jnthn But it needs an extra pair of aprens if it's getting parsed as a signature, no?
17:42 jnthn my ($a (4b, $c)) # doesn't surprise
17:42 masak r: my $a ($b, $c); say $b
17:42 TimToady 4b or not 4b, that is the question
17:42 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«(Mu)␤»
17:43 masak r: my $a ($b, $c); $b = 42
17:43 camelia rakudo-jvm 882e33: OUTPUT«java.lang.NullPointerException␤  in block  at /tmp/f7mLLTPKnG:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/stage2/NQPHLL.nqp:1194␤  in any command_eval at s…»
17:43 camelia ..rakudo-parrot 3cef56: OUTPUT«(signal SEGV)»
17:43 jnthn Too much lag to spot my typos in time :P
17:43 jnthn I'm surprised it even parses; I must be misremembering something in the grammar...
17:43 TimToady it's just coming in as a postconstraint
17:44 jnthn oh...
17:44 jnthn I suspect we don't try to handle that code path at all :)
17:44 jnthn So heck knows what comes of the $b and $c... :)
17:44 TimToady r: my $a where 42; $a = 43
17:44 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:45 TimToady yeah
17:48 TimToady r: subset Even of Int where * %% 2;  my Even $x; $x = 3;
17:48 camelia rakudo-parrot 3cef56: OUTPUT«Type check failed in assignment to '$x'; expected 'Even' but got 'Int'␤  in block  at /tmp/V229oUIPh3:1␤  in any  at /tmp/V229oUIPh3:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/stage2/NQPHLL.nqp:1133␤  in any ev…»
17:48 camelia ..rakudo-jvm 882e33: OUTPUT«Type check failed in assignment to '$x'; expected 'Even' but got 'Int'␤  in block  at /tmp/NJS1GDNiDG:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/stage2/NQPH…»
17:49 TimToady I guess we're not splitting that test out into post_constraint space as we do in siggies
17:49 TimToady r: subset Even of Int where * %% 2;  my $x where Even; $x = 3;
17:49 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:50 BenGoldberg r: subset Even of Int where * %% 2;  my Even $x; if( rand > 1 ) { $x = 3 }
17:50 * TimToady wonders if the assignment to typed object could be simpler on the nominal end if we did split that out
17:50 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Word 'if' interpreted as 'if()' function call; please use whitespace instead of parensâ�¤at /tmp/d3fhiDyzel:1â�¤------> [32mven of Int where * %% 2;  my Even $x; if[33mâ��[31m( rand > 1 ) { $x = 3 }[0mâ�¤Unexpected block…»
17:50 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Word 'if' interpreted as 'if()' function call; please use whitespace instead of parensâ�¤at /tmp/qNpxAu_Av5:1â�¤------> [32mven of Int where * %% 2;  my Even $x; if[33mâ��[31m( rand > 1 ) { $x = 3 }[0mâ�¤Unexpected bl…»
17:50 BenGoldberg r: subset Even of Int where * %% 2;  my Even $x; if ( rand > 1 ) { $x = 3 }
17:51 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:51 moritz masak: (re camelia patch) no idea
17:51 BenGoldberg r: subset Even of Int where * %% 2;  my Even $x; if ( rand > 1 ) { $x = "string" }
17:51 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: ( no output )
17:51 TimToady BenGoldberg: "instead of"  :P
17:52 BenGoldberg How would I get that to die at compile time, due to "string" not being of type Even?
17:53 moritz by patching the optimizer, somhow...
17:53 jnthn By patching Rakudo to do such things, yeah :)
17:53 BenGoldberg p: subset Even of Int where * %% 2;  my Even $x; if ( rand > 1 ) { $x = "string" }
17:53 camelia rakudo-parrot 3cef56: ( no output )
17:53 BenGoldberg P: subset Even of Int where * %% 2;  my Even $x; if ( rand > 1 ) { $x = "string" }
17:53 camelia pugs: OUTPUT«*** ␤    Unexpected "% 2"␤    expecting "::"␤    at /tmp/wPVmDJINhC line 1, column 29␤»
17:53 TimToady no %% op
17:54 BenGoldberg P: my Int $x; if ( rand > 1 ) { $x = "string" }
17:54 camelia pugs: ( no output )
17:54 TimToady implement type inference?
17:55 TimToady or range inference if want rand > 1 to turn into False :)
17:55 [Coke] diakopter: the pugs tests really want "en_US.ISO-8859-1" available on the testing machine. Any chance we can get that?
17:56 [Coke] -> $dayjob
17:56 TimToady US English doesn't include Latin :P
17:56 BenGoldberg In perl5 (and in Java!), an if( False ) { code } will remove { code } from the parse tree, for appropriate values of False
17:56 BenGoldberg Does perl6's optimizer do that yet?
17:57 TimToady yes, I know--I implement the p5 part of that
17:57 TimToady *mented
17:57 diakopter [Coke]: it has that, you just need to set the env vars
17:57 TimToady these days should probably ISO-8859-15 or thereabouts to include the euro...
17:58 arnsholt BenGoldberg: Good question. If you feel up for it, Rakudo's src/Perl6/Optimizer.nqp is the place to look. It might not be too hard to implement
17:58 TimToady otoh, I would also not mind ignoring locales out of existence
17:59 TimToady if False { use locale; }
18:01 BenGoldberg Shirely, even in perl6, 'use' gets performed the moment the statement it's in is fully evaluated?
18:01 jnthn BenGoldberg: yes
18:02 BenGoldberg So that would need to be BEGIN { if False { require locale } }, or somesuch
18:03 TimToady was a joke
18:03 BenGoldberg :)
18:04 TimToady now, it'd be *interesting* if the compiler could figure out that the block was going away even before it was parsed, and disable any BEGIN-stuff inside it
18:05 TimToady but not even p5 tries to do that
18:05 TimToady (that I know of...)
18:06 BenGoldberg That would be pretty deep magic :)
18:10 sqirrel joined #perl6
18:10 timotimo i can't get rid of Use of uninitialized value $thirdpartyjars in join or string at tools/build/create-jvm-runner.pl line 42.
18:23 dalek nqp: 8ea4967 | dwarring++ | examples/rubyish/ (2 files):
18:23 dalek nqp: general rubyish sigspace tidies. added :s<hs> for horizontal white-space
18:23 dalek nqp: review: https://github.com/perl6/nqp/commit/8ea496793a
18:28 timotimo and also open: No such file or directory at tools/build/create-jvm-runner.pl line 35.
18:29 timotimo Creating '/home/timo/build/rakudo/install/home/​timo/build/rakudo/install/bin/perl6-j'
18:29 timotimo well, this doesn't make sense
18:30 FROGGS ohh
18:30 FROGGS destdir or prefix confusion
18:31 timotimo i've temporarily reverted the recent commit that seems to have touched it
18:41 timotimo that makes it work again
18:43 timotimo who here is Brian Gernhardt?
18:46 retupmoca timotimo: http://irclog.perlgeek.de/​perl6/2013-10-28#i_7774849 ?
18:46 FROGGS timotimo: benabik
18:46 timotimo yes
18:46 benabik timotimo: Hello?
18:47 timotimo hi there :)
18:47 timotimo can you help me with the jvm runner installer?
18:47 benabik Probably.  I tried not to break it too badly.  ;-)
18:47 timotimo Creating '/home/timo/build/rakudo/install/home/​timo/build/rakudo/install/bin/perl6-j' - this is obviously wrong, but i'm not sure how to fix it
18:48 timotimo this is using Configure.pl --backends=jvm,parrot --gen-parrot --gen-nqp=master
18:48 benabik Erm
18:48 benabik I think I forgot quotes in Makefile-JVM.in
18:48 Tene timotimo: Try again with HOME=/home/timo/home/timo/home/tene/home/timo
18:48 timotimo Tene:    :3
18:49 benabik Setting DESTDIR=/ might be a workaround.
18:49 timotimo i'd prefer a fix. this doesn't stop me from working, mind you :)
18:50 benabik Doublequotes around $(DESTDIR) on the create-jvm-runner lines in Makefile-JVM.pl
18:50 benabik I can submit a PR in a moment.
18:50 timotimo ah, for some reason i thouht you had a commitbit
18:50 * timotimo applied for one today :)
18:51 benabik I have one for Moar, NQP, and Parrot but not Rakudo.
18:53 lue hello world o/
18:53 dalek nqp: e13e2f4 | dwarring++ | examples/rubyish/ (2 files):
18:53 dalek nqp: updated rubyish README etc. nqp => nqp-p
18:53 dalek nqp: review: https://github.com/perl6/nqp/commit/e13e2f451c
18:54 * benabik trying to build without DESTDIR before submitting the PR.
18:54 lue so, can I move on from 10? :)
18:56 FROGGS lue: only if unicode chars are involved :o)
18:57 lue :) I ask if it's OK to go on with that DBIish fail and the one/two minor spectest fails. https://gist.github.com/lue/7188705
18:58 FROGGS lue: no idea
18:59 lue moritz: do you know if I should go ahead? (Or did you e.g. fix DBIish and I need to re-do the tarball?)
19:00 ajr joined #perl6
19:03 sqirrel joined #perl6
19:04 dmol joined #perl6
19:04 arnsholt moritz++ # New build system is still awesome =)
19:05 FROGGS it is, yeah
19:05 timotimo moritz++ # what arnsholt said!
19:05 FROGGS moritz++ # :o)
19:05 benabik moritz made it awesome, I just broke it.  :-D
19:06 jeffreykegler joined #perl6
19:07 timotimo fwiw, i don't actually know how to fix the .=[...] stuff
19:07 benabik If someone with a bit could merge https://github.com/rakudo/rakudo/pull/225 that'd be great.
19:07 arnsholt I like the body of the message =D
19:07 benabik :-D
19:10 pecastro joined #perl6
19:14 FROGGS timotimo: the PR solved your problem?
19:15 jaffa4 joined #perl6
19:15 jaffa4 hi all
19:15 jaffa4 I cannot see the module I added to Panda.
19:16 timotimo did not try yet
19:16 jaffa4 Is it not approved yet?
19:16 lue moritz, jnthn, other release pros: if a DBIish fix isn't there/imminent, can I continue to Step 11?
19:18 timotimo DBIish did get the fix
19:19 timotimo or at least a fix
19:20 lue time to go back and spend a few more testing hours then :)
19:25 lue jaffa4: I can see your Path::Util on http://modules.perl6.org/, so perhaps you just need to update your panda's module list (forget the command to do that exactly, though there should be some --help that tells you)
19:27 timotimo i think it's panda update
19:27 jaffa4 yes, that was the problem.
19:28 dalek nqp: 724a9b1 | (Arne Skjærholt)++ | src/NQP/ (2 files):
19:28 dalek nqp: Make 'return' a term taking an optional argument.
19:28 dalek nqp:
19:28 dalek nqp: This way, 'return;' won't be silently ignored in a function. Closes #127.
19:28 dalek nqp: review: https://github.com/perl6/nqp/commit/724a9b12cb
19:29 masak arnsholt++! \o/
19:29 masak arnsholt: you are my hero tonight.
19:29 arnsholt Oh, yay! \o/
19:29 masak arnsholt: next beer is on me :)
19:30 lue ✔ ⑤ Update modules (⑥ and ⑦ still ✔)
19:30 dalek star: e5501b9 | lue++ | modules/DBIish:
19:30 dalek star: Pull DBIish update.
19:30 dalek star: review: https://github.com/rakudo/star/commit/e5501b9321
19:30 lue ✔ ⑧ push un-pushed changes
19:31 sftp joined #perl6
19:32 lue ✔ ⑨ Create tarball
19:32 xdbr joined #perl6
19:32 arnsholt masak: Anyways, happy to be of help. It's been on my hitlist of NQP bugs for a while, but last time I tried to fix it I couldn't quite get there
19:32 Shozan joined #perl6
19:33 arnsholt You coming to Oslo in the near future, BTW?
19:35 masak arnsholt: nothing on the schedule, though I had a Bergen a few days ago, and will have an Ålesund next week.
19:36 FROGGS Å is pronounced like an O, right?
19:37 arnsholt Right. Both are a bit far away for just a beer =)
19:37 arnsholt FROGGS: Yeah
19:39 * lue thinks of å as a "scandinavian o sound" :P
19:39 arnsholt Yeah, the exact phonetic value is probably going to be different from German/English
19:39 arnsholt Close enough though
19:40 FROGGS cool..., we have a software at $work call Blå, which is the color blue as I remember
19:40 dalek rakudo/nom: 9952073 | benabik++ | tools/build/Makefile-JVM.in:
19:40 dalek rakudo/nom: Fix j-install for no DESTDIR
19:40 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/99520735ea
19:40 lue for some reason my mind is saying å ~~ german u ... I'll check the IPAs
19:41 denis_boyun joined #perl6
19:41 geekosaur front-a, isn't it?
19:41 sqirrel joined #perl6
19:41 masak ah, the sensual dance of the Lexical and the Dynamic.
19:41 moritz lue: å is pronounce more like the German o
19:42 moritz oh, FROGGS++ already said that in the backlog :-)
19:42 GlitchMr perl6: .say for (('' xx 2, 'Fizz') xx * Z~ ('' xx 4, 'Buzz') xx *) Z|| 1 .. 100;
19:42 lue yeah.
19:42 camelia niecza v24-98-g473bd20: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤This macro cannot be used as a function at /tmp/1aB2oDIVIb line 1:â�¤------> [32mzz') xx * Z~ ('' xx 4, 'Buzz') xx *) Z||[33mâ��[31m 1 .. 100;[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niec…»
19:42 camelia ..rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«1␤2␤Fizz␤4␤Buzz␤Fizz␤7␤8␤Fizz␤Buzz␤​11␤Fizz␤13␤14␤FizzBuzz␤16␤17␤Fizz␤19␤Buzz␤​Fizz␤22␤23␤Fizz␤Buzz␤26␤Fizz␤28␤29␤FizzBuz​z␤31␤32␤Fizz␤34␤Buzz␤Fizz␤37␤38␤Fizz␤B…»
19:42 arnsholt lue: German u is more similar to Norwegian o
19:42 FROGGS my colleagues tend to say it like Blä >.<
19:42 masak moritz: at least in Swedish, we screwed up the letter 'o' so that it (most of the time) has more of an /u/ sound.
19:42 masak FROGGS: heh :)
19:43 FROGGS masak: and in germany the e gets lost :o)
19:43 Rotwang joined #perl6
19:43 arnsholt Same in Norwegian. It's mostly /u/, but occasionally /o/ just to keep foreigners on their toes =)
19:43 * lue really needs more practice in German :P
19:43 FROGGS for example most ppl say 'nein' like 'nain'
19:44 FROGGS so, at some point we can scratch that e
19:44 lue All I had to do was say "Ostwestfälisch" to myself and I could confirm å == o :)
19:44 arnsholt lue: Which is your first language?
19:44 * FROGGS .oO( Perl )
19:45 lue arnsholt: de & en, though the de part's been lacking since I was 5 :)
19:45 arnsholt Ah, right
19:45 * masak .oO( Leben, das Universum und der ganze Rest )
19:47 * lue is pretty sure some of his ancestors are rolling in their graves because il sait français :)
19:48 moritz masak, arnsholt: which means in the theory there are 4 ways to pronounce "Oslo", though in practice I have heard only three of the variants
19:48 moritz :-)
19:49 lue .oO[ Then again, they should know my opinions on East Prussia and Alsace-Löttrigen :) ]
19:50 masak moritz: only 4? I think you're discounting both vowel length and stress. :)
19:50 masak moritz: not to mention speech tone.
19:51 arnsholt moritz: The inherited pronounciation is probably /ushlu/
19:51 moritz masak: they all sound the same to me :-)
19:51 masak count all those, and you could have 64 different pronunciations of "Oslo".
19:51 arnsholt And not /uslu/
19:51 GlitchMr r: .say for ('' xx 2, 'Fizz') xx * Z~ ('' xx 4, 'Buzz') xx * Z|| 1 .. 100;
19:51 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/ba8RO5tbZZâ�¤Bogus statementâ�¤at /tmp/ba8RO5tbZZ:1â�¤------> [32m Z~ ('' xx 4, 'Buzz') xx * Z|| 1 .. 100;[33mâ��[31m[0mâ�¤    expecting any of:â�¤        postfixâ�¤        prefix or termâ�¤    â€¦Â»
19:51 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/5G5nmaqRBwâ�¤Bogus statementâ�¤at /tmp/5G5nmaqRBw:1â�¤------> [32m Z~ ('' xx 4, 'Buzz') xx * Z|| 1 .. 100;[33mâ��[31m[0mâ�¤    expecting any of:â�¤        postfixâ�¤        prefix or termâ�¤ …»
19:52 GlitchMr r triggers both Rakudos? Why?
19:52 masak ah, yes, the /s/ or /sh/ distinction is another bifurcation. so 128 :)
19:52 masak GlitchMr: because... they're both rakudos?
19:52 moritz GlitchMr: becase rakudo-moarvm doesn't work yet well enough
19:52 masak :)
19:52 Mouq joined #perl6
19:52 lue .oO(Oslo -> Ahzzz-low)
19:53 * masak slaps lue with a doormat labeled "WELCOME TO THE USA"
19:53 moritz .oO( maybe 'rs' should launch all rakudos and 'r' a random rakudo )
19:53 masak rakoulette!
19:54 FROGGS randudo!
19:54 moritz masak: sounds tasty :)
19:54 masak rak-u-tell-me
19:54 FROGGS o-rak-el?
19:54 lue R-A-K-U-D-O-!: say "hello"
19:54 * masak .oO( because what we need from camelia is... more indeterminism...!? )
19:55 moritz and a proper source of entropy, please
19:55 lue sm: -> surprise me, any compiler camelia has available could be used :)
19:56 moritz bdsm: -> be a dear and surprise me :-)
19:56 moritz oh sorry, we are trying to be family friendly to all time zones
19:57 * masak .oO( we crossed a line? where? ) :P
19:57 moritz we didn't? good :-)
19:58 masak far as I'm concerned, unexpanded acronyms are SFW in-jokes.
19:58 moritz woah, rakudo-j setting build just segfaulted the JVM
20:00 arnsholt Whoops!
20:03 * lizmat is back from cycling and was not hurt by any tree parts
20:04 TimToady just whole trees?
20:04 lizmat timotimo: wrt to .=[], I tried to figure out where to apply a fix, but could not find the place, nor how to fix it
20:04 Mouq Uh, re: #120383: There's actually two different rules trying to handle what you're trying to do
20:04 synopsebot Link: https://rt.perl.org/rt3//Publ​ic/Bug/Display.html?id=120383
20:04 lizmat hmmm...  isn;t a whole tree also a tree part ?
20:04 Mouq r: my @a = (4,5); @a.=[1]; say @a
20:04 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/Jy_2TLX7ui�Cannot use .= on a non-identifier method call�at /tmp/Jy_2TLX7ui:1�------> [32mmy @a = (4,5); @a.=[1][33m�[31m; say @a[0m�»
20:04 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/UOqokADIlO�Cannot use .= on a non-identifier method call�at /tmp/UOqokADIlO:1�------> [32mmy @a = (4,5); @a.=[1][33m�[31m; say @a[0m�»
20:04 Mouq r: my @a = (4,5); @a .= [1]; say @a
20:04 camelia rakudo-jvm 882e33: OUTPUT«No such method '&postcircumfix:<[ ]>' for invocant of type 'Array'␤  in method dispatch:<.=> at gen/jvm/CORE.setting:1129␤  in block  at /tmp/gYPsuez15t:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nq…»
20:04 camelia ..rakudo-parrot 3cef56: OUTPUT«No such method '&postcircumfix:<[ ]>' for invocant of type 'Array'␤  in method dispatch:<.=> at gen/parrot/CORE.setting:1138␤  in block  at /tmp/vaHxV07r4l:1␤  in any  at /tmp/vaHxV07r4l:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any…»
20:05 lizmat .oO( feels a set or category theory remark coming up )
20:05 Mouq :P We have an infix <.=> and a term <.*> that includes .=
20:06 * Mouq was messing with this stuff when he made dotty postfixes work
20:06 timotimo lizmat: yup, me either :|
20:06 btyler joined #perl6
20:06 lizmat Mouq: indeed, and that's when I started to lose track
20:06 timotimo at least list and hash access are going to get faster ... i guess
20:07 lue Just to clear this up ahead of time (starting spectests now): if the only failures I get are those two from the spectest (noatime and :Punctuation), can I proceed to 11?
20:07 masak lizmat, Mouq: despite my example being confusing, I think it's still perfectly legitimate even after the postcircumfix method->sub refactor.
20:07 Mouq masak: I agree
20:07 lizmat masak: agree
20:07 masak \o/
20:07 lizmat it's an implementation detail
20:08 masak it's little things like @a.=[@indices] that makes Perl 6 a joy to use.
20:08 lue .oO(Perhaps @a[] is a sub and @a.[] is a method? :P)
20:08 timotimo it's not supposed to be :|
20:08 masak inner dialogue goes "should this work? it *looks* like it should work. yay, it works!"
20:08 masak well, modulo all the bugs. :)
20:08 masak but I'm submitting those :P
20:09 Mouq I think we should "just" make postcircumfixes pretend to be methods everywhere and be done with it :P
20:09 timotimo wouldn't that be a step back?
20:09 lue Now that I said it, I actually like the method/sub thing I mentioned. After all, you have foo(@bar) and @bar.foo, this would be the same thing
20:09 TimToady timotimo: that was a joke
20:10 lue but with funny syntax for the operator's names
20:10 timotimo ah, of course
20:11 timotimo you fave foo(@bar) and @bar.&foo, that's a difference
20:11 timotimo (er, or am i misunderstanding?)
20:11 masak lue: subs and methods are different.
20:11 masak r: class A { method foo { say "method" }; method bar { foo() }; sub foo { say "sub" } }; A.foo; A.bar
20:11 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«method␤sub␤»
20:12 lue sub postcircumfix:<[ ]>(@a, *@b) { @a.[@b] } is what I was thinking, like how you have sub foo($a, $b) { $a.foo($b) }
20:12 iSlug joined #perl6
20:13 masak lue: under the current semantics, that would probably regress infinitely :)
20:13 lue :)
20:13 arnsholt r: grammar Foo { token foo { :temp $foo = 3; foo } }; say "alive"
20:13 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/qoyLzsr3tl�Unrecognized regex modifier :temp�at /tmp/qoyLzsr3tl:1�------> [32mgrammar Foo { token foo { :temp[33m�[31m $foo = 3; foo } }; say "alive"[0m�»
20:13 camelia ..rakudo-parrot 3cef56: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/iZUqBPlpCW�Unrecognized regex modifier :temp�at /tmp/iZUqBPlpCW:1�------> [32mgrammar Foo { token foo { :temp[33m�[31m $foo = 3; foo } }; say "alive"[0m�»
20:14 hoelzro ahoy perl6ers
20:15 timotimo hey rob :)
20:15 jaffa4 panda does not work for me, it cannot delete a file.
20:15 timotimo what are you up to these days?
20:15 timotimo jaffa4: the new release is going to fix that, lue is preparing it at this moment
20:15 arnsholt hoelzro o/
20:16 jaffa4 Are you sure?
20:16 hoelzro timotimo: today, converting a Perl 5 script I wrote to Perl 6
20:16 lue Ooh! I got it!    sub postcircumfix:<.[ ]>(@a, *@b) { @a[@b] }    :D
20:16 lue masak: ^^^
20:16 hoelzro tomorrow, looking at taking advantage of the free time I'm finding in my schedule to work on stuff =)
20:16 Grrrr 26
20:17 lue (I can, in fact, proceed to step 11 if I get just the atime and punct fails, right?)
20:17 hoelzro so I'm having an issue with grammars/actions: https://gist.github.com/hoelzro/7203812
20:18 hoelzro isn't make at the TOP method supposed to determine what parse returns?
20:18 timotimo you need to .ast on that
20:18 hoelzro G.parse(...).ast?
20:18 timotimo i think so
20:18 hoelzro ah ha!
20:18 hoelzro timotimo++
20:18 TimToady lue: there's no bosses here; you'll have to use your judgement :)
20:18 hoelzro thank you!
20:18 timotimo if all things were so easy to solve :)
20:19 hoelzro indeed =)
20:19 lue TimToady: My judgement says yes; my inexperience says "since this is my first time releasing..." :)
20:20 TimToady you could put a "known issues" into the release notes
20:20 lizmat .oO( .ast should really be called .made, as it is the result of the inner make() )
20:21 arnsholt Well, that makes good punny sense, but sometimes practical names make sense too =)
20:21 masak lue: +1 on TimToady's "known issues"
20:21 lue I could; but the atime failure is if you have a noatime drive, and the punct failure is because I can't get 2013.10 rakudo to use the newer tests (which I had fixed because I deemed the tests wrong)
20:21 timotimo hoelzro: github still doesn't use the perl6 highlighter, at least not for nqp files :(
20:21 hoelzro timotimo: I know =( so frustrating!
20:22 lue (I will probably put a "known issues" in the announcement though.)
20:22 hoelzro hmm
20:22 masak lue: will users who download the new R* see test failures?
20:22 lizmat arnsholt: I would think that hoelzro would have thought about a .made sooner than about an .ast
20:22 hoelzro I *may* have forgotten to put it in linguist...
20:22 Mouq lizmat: I may have figured it out
20:22 lue If they run spectest, yes.
20:22 lizmat Mouq++
20:23 lizmat (for persevering)
20:23 timotimo lue: if you put a known issues section there, you can include *<>, *{} and *[] not currying correctly
20:23 lue The specific punct failure, by the way, is :Punctuation delivering § and ¶ (category Po in Unicode), and the old tests not having those two characters in the expected.
20:23 * moritz submitted the JVM bug at http://icedtea.classpath.org/​bugzilla/show_bug.cgi?id=1590
20:23 * lue hopes updating the announcement doesn't require an update of the tarball
20:24 lue it does :/
20:24 arnsholt One more squashed NQP bug and we're down to 25 open tickets
20:24 moritz arnsholt++
20:25 lue Plan: finish spectest, update announcement, new tarball, go through the testing procedure again (but SKIP the spectest; that shouldn't change), 11.
20:25 moritz lue: sounds fine
20:25 lue (I could probably just stop the spectest know, because I don't see how I'll randomly get new failures.)
20:25 lue s/know/now/
20:27 hoelzro why is it that action methods must explicitly take $/ as an argument?
20:28 hoelzro you'd think that would be implied
20:28 timotimo nah, how would you tell an action class apart from a non-action class?
20:28 moritz scoping
20:28 timotimo and you can have non-action methods in an action class, too
20:28 hoelzro I see
20:28 moritz $/ is lexically scoped
20:28 lue method action_rule($not-the-slash) { $not-the-slash<hash> }
20:28 lue ^ works too
20:29 Mouq Oh
20:29 moritz also if you want to do a regex match, you'd better not name the argument $/ :-)
20:29 Mouq ".=" in <.*> was there so that we could do @list».=meth
20:29 lizmat moritz: so would $*/ not work then, as it is marked as dynamic, afaik ?
20:30 lizmat I mean CALLER::$/ ?
20:30 lizmat r: say $/.VAR.dynamic
20:30 camelia rakudo-parrot 3cef56: OUTPUT«True␤»
20:30 camelia ..rakudo-jvm 882e33: OUTPUT«Unhandled exception: java.lang.OutOfMemoryError: GC overhead limit exceeded␤  in  (gen/jvm/stage2/QAST.nqp:288)␤  in map_classlib_core_op (gen/jvm/stage2/QAST.nqp:287)␤  in  (gen/jvm/stage2/QAST.nqp:1889)␤  in  (gen/jvm/stage2/QAST.nqp)␤  in  (src…»
20:30 moritz $CALLER::</> maybe
20:31 moritz what, $/ is dynamic?
20:31 lizmat the thing is, that $/ is usually only needed for error messages
20:31 lizmat yup, as is $_ and $!
20:31 lizmat r: say $_.VAR.dynamic
20:31 * moritz
20:31 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«True␤»
20:31 lizmat r: say $!.VAR.dynamic
20:31 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«True␤»
20:31 * moritz 's world view collides
20:31 moritz erm, collapses
20:31 moritz it collapses so hard, I can'T even type anymore
20:32 lizmat sorry
20:32 lizmat :-)
20:32 * lizmat starts picking up the pieces
20:32 lue r: my $::<###!@#@###!!$$> = 42;
20:32 camelia rakudo-parrot 3cef56: OUTPUT«===SORRY!===␤ResizablePMCArray: index out of bounds!␤»
20:32 camelia ..rakudo-jvm 882e33: OUTPUT«===SORRY!===␤VMArray: Index out of bounds␤»
20:32 hoelzro hmm
20:32 lue std: my $::<###!@#@###!!$$> = 42;
20:32 camelia std 8adbc60: OUTPUT«ok 00:01 122m␤»
20:32 moritz how to make the JVM not segfault, 101: enable core dumps
20:33 * hoelzro feels like there *must* be shorthand for :foo($<foo>)
20:33 moritz hoelzro: $/.hash<foo>:p
20:33 moritz or just $<foo>:p if that works
20:33 hoelzro =O
20:33 lue what does the p adverb do? :P
20:33 moritz if not, $/<foo>:p
20:33 moritz lue: return a Pair, not just the value
20:33 hoelzro bwa
20:34 hoelzro blown away =)
20:34 masak moritz: ISTR $_, $/, $! are "lexical if <mumble>, otherwise dynamic".
20:35 moritz meh, I want intuitive semantics *and* clear, simple rules! :-)
20:36 moritz and more wows than WATs
20:37 moritz and an OOM killer that takes that fat, ever-growing memory eater instead of my window manager
20:38 raiph http://felixge.de/2013/03/11​/the-pull-request-hack.html # the wisdom of liberal commit bits
20:39 masak raiph: au++ would be proud :)
20:39 masak raiph: she was 8 years ahead of her time.
20:39 moritz raiph: we do that for projects under the perl6/ organization
20:39 moritz (at least I do)
20:39 moritz for rakudo, there are some non-trivial legal aspects to consider
20:39 raiph yeah, I've seen it increasingly regularly over the last 2 years
20:40 raiph I've always wanted more of au's thinking applied
20:40 masak and indeed, HN traces the idea back to au++: https://news.ycombinator.com/item?id=5358645
20:40 moritz au++ was even more "agressive", handing out commit bits to people who might possibly want to commit
20:40 masak "trust the anarchy" ♡
20:41 moritz I remember seeing Guido van Rossum in the list of pugs commit bitters
20:41 masak ...because he badmouthed Perl 6? :P
20:42 * masak .oO( "I don't know who this Guido guy is, but he sounds like he wants to improve things" )
20:42 raiph do p5 contributors have to sign a CLA?
20:42 hoelzro hmm
20:42 hoelzro I can do [:r ...] to lexically enable ratcheting, right?
20:42 moritz hoelzro: yes
20:43 hoelzro so how can I disable ratcheting lexically?
20:43 moritz [:!r ... ]
20:43 hoelzro ah ha
20:43 hoelzro that makes sense =)
20:43 hoelzro moritz++
20:44 moritz don't we also have modifiers for quantifiers that enable backtracking?
20:44 * moritz hasn't read through S05 in a long time
20:44 masak moritz: yes.
20:45 moritz is that the * suffix?
20:46 masak the "long forms" are *! (greedy) *? (frugal) and *: (ratchet).
20:47 masak and similarly for all the other quantifiers.
20:47 masak depending on the context you're in, you can use just *
20:47 moritz ah, !
20:47 masak yeah.
20:47 masak which works really well for "greedy" :)
20:47 masak ISTR *+ is a p5-compatible STD-spec'd synonym for *?
20:48 * jnthn back from dinner :)
20:49 masak jnthn! \o/
20:49 hoelzro moritz: I have a goal of summarizing it, someday...
20:50 moritz hoelzro: I have a goal of understanding it, someday
20:50 hoelzro heh
20:50 moritz I think I have come pretty far, but there are still areas that scare me
20:51 masak moritz: pro tip: implement a grammar engine :>
20:52 moritz masak: like, one that does transitive LTM, and bactracking into the NFA?
20:52 * moritz shudders
20:57 masak moritz: well, one must start somewhere.
20:57 masak moritz: those are likely the last few features one implements.
20:57 moritz last, because after that you give up programming due to constant frustration :-)
20:57 masak yeah, but it's *good* frustration!
20:57 |newbie| joined #perl6
20:57 moritz I'm picking up good frustration ♫
20:57 jferrero joined #perl6
20:59 colomon joined #perl6
21:00 masak moritz: now *that* should be the programmer's mantra, instead of "aaaargh don't interrupt me while I'm thinking very complicated thoughts" :)
21:00 masak moritz: programming, at its very heart, is frustration.
21:00 jnthn moritz: You don't "backtrack into the NFA", you just throw away cursors ;)
21:01 lue mantra: I will PERSONALLY torch AMD's factories if you don't work THIS TIME!
21:02 * FROGGS .oO( Please consider the environment before throwing away cursors! )
21:03 lizmat I thought cursors were made of pure recycled electrons only ?
21:03 FROGGS lizmat: used electrons... it is the same like with batteries :|
21:04 moritz jnthn: so ther NFA enerates a passing cursor each time it is in an accepting state?
21:04 moritz FROGGS: the grow weaker over time?
21:04 Rotwang joined #perl6
21:04 FROGGS yeah... positrons would grow weaker back into time though
21:04 jnthn moritz: Not quite. In alternation mode it just queues up the branches on the bstack in reverse order and immediately backtracks to hit the first one.
21:05 jnthn moritz: In protoregex mode it returns an ordering
21:05 jnthn moritz: And we just call the candidate rules in that order
21:06 Mouq r: say (my@=^10).=&postcircumfix:<[ ]>(2)
21:06 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«2␤»
21:06 moritz jnthn: and rakudo's bstack roughly coresponds to a list of cursors in the model advocated by S05, right?
21:07 FROGGS humm, I read "parrot 3cef56" ad "parrot sketch"
21:07 jnthn moritz: Yeah, the bstack is like, "Cursors ain't that cheap, let's go integering" :)
21:08 moritz sounds very integer :-)
21:10 lue is there a windows equivalent of "noatime" ?
21:11 FROGGS lue: http://serverfault.com/questions/3​3932/how-do-you-disable-the-last-a​ccessed-attribute-on-ntfs-windows
21:11 FROGGS (google #1)
21:11 lizmat depends on the file system no?
21:11 lue :)
21:11 lue [this is for the announcement's known issues section]
21:12 moritz lue: then don't mention the windows stuff
21:12 FROGGS yeah, I guess it is not that common to do that on windows
21:13 lue Oh, I just wanted to mention the non-unix equivalent of noatime, just in case.
21:14 lue Should I mention the noatime issue in the first place, in case someone runs the spectest?
21:19 moritz I wouldn't give the whole thing a high priority, since the test is a few months old iirc, and so far we didn't have any complaints
21:23 dalek star: 344b246 | lue++ | docs/announce/2013.10.md:
21:23 dalek star: Update Release Announcement
21:23 dalek star: review: https://github.com/rakudo/star/commit/344b246e5f
21:23 Mouq r: say (my@=^10) .= &postcircumfix:<[ ]>((2,3,4)) #fine
21:23 camelia rakudo-parrot 3cef56, rakudo-jvm 882e33: OUTPUT«2 3 4␤»
21:23 Mouq r: say (my@=^10) .= '&postcircumfix:<[ ]>'((2,3,4)) #less so
21:23 lue I would love some review to tell me what issues should be added/removed/reworded
21:23 camelia rakudo-jvm 882e33: OUTPUT«No such method '' for invocant of type 'Array'␤  in method dispatch:<.=> at gen/jvm/CORE.setting:1129␤  in block  at /tmp/ZUohq1LKAG:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any co…»
21:23 camelia ..rakudo-parrot 3cef56: OUTPUT«No such method '' for invocant of type 'Array'␤  in method dispatch:<.=> at gen/parrot/CORE.setting:1138␤  in block  at /tmp/Ws_a7vFgTl:1␤  in any  at /tmp/Ws_a7vFgTl:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/…»
21:23 Mouq Err, what am I doing?
21:23 Mouq r: say (my@=^10).'&postcircumfix:<[ ]>'((2,3,4)) #less so
21:24 camelia rakudo-jvm 882e33: OUTPUT«No such method '&postcircumfix:<[ ]>' for invocant of type 'Array'␤  in block  at /tmp/0qQfzZIAu9:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/stage2/NQPHLL.n…»
21:24 camelia ..rakudo-parrot 3cef56: OUTPUT«No such method '&postcircumfix:<[ ]>' for invocant of type 'Array'␤  in block  at /tmp/YMtr9_IGhS:1␤  in any  at /tmp/YMtr9_IGhS:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/stage2/NQPHLL.nqp:1133␤  in any evalfi…»
21:27 Mouq lue: The :sigspace thing is more like "the only place whitespace will be significant is after atoms that match something"
21:28 Mouq Although your description is technically correct
21:28 lue Mouq: that's what it already says "leading whitespace NOT" ~~ "non-leadng whitespace IS"
21:29 lue (I didn't write that bit though; it's mostly left as-is. Although I do agree, I did have initial trouble reading it.)
21:29 lue I'd like someone to tell me the notability of L81 of the announcement.
21:30 Mouq Oh, I understand what it's saying now
21:30 * lue & --until="top of the hour"
21:37 logie joined #perl6
21:51 donaldh_ joined #perl6
21:52 lue All the modules pass, so unless anyone has anything else to say about the announcement (I'll wait a few minutes), the current version is what will come in the tarball.
21:55 lizmat argh, "is deprecated" should read "is DEPRECATED"  :-(
21:59 lizmat well, not so important
21:59 lue lizmat: got it.
22:01 dalek star: bec08fa | lue++ | docs/announce/2013.10.md:
22:01 dalek star: Spelling fix in Announcement.
22:01 dalek star: review: https://github.com/rakudo/star/commit/bec08facb1
22:02 * lue will now tarball and test once more, sans spectest
22:02 rurban joined #perl6
22:03 timotimo Greatly reduce object creation during Regex parsing.  <-  hm, but this is only during code parsing, no? well, and anything that uses ident
22:04 japhb__ Which is a lot of things.
22:04 lue timotimo: I assume all regexes are affected, though I could add a "especially for Perl 6 itself" or something to that effect.
22:04 japhb__ timotimo: Do you have local changes to perl6-bench?
22:05 timotimo japhb__: i've been trying to change components.json to use the new configure.pl system
22:05 timotimo and also try to get nqp-moarvm updated
22:06 japhb__ Are you timotimo on github?
22:07 timotimo no, i'm timo
22:07 japhb__ Ah, OK
22:07 timotimo i got in early enough to grab that short name :)
22:08 lue ✔ ⑨ create tarball
22:08 lizmat lue++
22:08 japhb__ timotimo: In honor of au++, you now have a perl6-bench commitbit.
22:08 lue (I didn't add the "esp. Perl 6" though, I don't think it's necessary)
22:08 timotimo \o/
22:09 timotimo japhb__: i wish i could try out the changes to components.json, but i'll have to massively downgrade rakudo or something >_>
22:11 japhb__ If you have any ideas on dealing with components with changing build instructions, I'm all ears.  Right now, I'm imagining specifying multiple sets of instructions per component, each with a "minimum git revision" that set of instructions should apply to.
22:11 japhb__ Downgrade rakudo?  Why?
22:11 timotimo oh
22:11 timotimo to get a working "run" again
22:11 japhb__ Oh, right.
22:11 timotimo i don't think we are at a point where we need that feature, though :)
22:12 japhb__ Yeah, that's more of something to care about when we're watching for regressions.
22:12 timotimo yup
22:12 japhb__ Still, never hurts to vaguely plan ahead.  ;-)
22:14 japhb__ OK, run bugs I'm currently aware of: 1) JVM getting boolean value of run result wrong, 2) Run not passing on updated %*ENV, even though shell does.  Any others?
22:14 timotimo calling run often will cause segfaults
22:14 timotimo or "memory corruption" of any sort
22:15 FROGGS on parrot only, right?
22:15 timotimo yes
22:15 timotimo couldn't test it on moar yet ;)
22:15 FROGGS *g*
22:15 FROGGS we're working on it :o)
22:15 FROGGS btw, if someone would help me, we could fix run()
22:16 FROGGS we only need an implementation (in C) for spawn
22:16 iSlug joined #perl6
22:16 japhb__ OK, so segfaulting run on Parrot makes 3.
22:16 FROGGS and only for linux... since the windows version is done
22:16 japhb__ Are there any known shell() bugs?
22:16 stevan_ joined #perl6
22:16 FROGGS not that I know
22:16 japhb__ OK, how many of the shell fixes can we steal for run?
22:17 FROGGS all
22:17 timotimo oh, JVM doesn't have open pipe yet
22:18 japhb__ Does donaldh's pipe sketch work on JVM
22:18 japhb__ ?
22:18 timotimo didn't try, my rakudo is now going to be too outdated :)
22:19 FROGGS timotimo: btw, I believe we can fix issue #3 by replacing "$state but $state" by Process::State as specced
22:20 donaldh_ japhb_: yes :-)
22:20 donaldh_ timotimo: JVM has nqp::openpipe
22:20 FROGGS this would return the return code, but boolifies to return_code == 0
22:20 japhb__ Then by gosh, let's make it happen!
22:20 timotimo doesn't that require the same sort of thing that we need for IntStr and friends?
22:21 timotimo like, do we have to derive it from Str and Bool?
22:21 japhb__ Ewww.
22:22 timotimo r: class Process::State is Int is Bool { }
22:22 camelia rakudo-jvm 882e33: OUTPUT«===SORRY!===␤java.lang.ClassFormatError: Duplicate method name&signature in class file __P6opaque__95␤»
22:22 camelia ..rakudo-parrot 995207: OUTPUT«===SORRY!===␤Duplicate box_target for native int␤»
22:22 timotimo not Str and Bool, fwiw
22:22 jnthn I sure hope Process::State ain't gonna derive from the two of those :P
22:22 timotimo r: class Process::State is Int is Bool is repr(Int) { }
22:22 stevan_ joined #perl6
22:22 camelia rakudo-jvm 882e33: OUTPUT«===SORRY!===␤Cannot unbox a type object␤»
22:22 camelia ..rakudo-parrot 995207: OUTPUT«===SORRY!===␤Cannot unbox a type object as a native str␤»
22:23 jnthn timotimo: The problem there is that Int and Bool both want to own the slot that unboxes as a native int.
22:23 japhb__ class Process::State is Int { method Bool { self == 0 } }; Process::State $a = 0; say so $a; Process::State $b = 1; say so $b;
22:23 jnthn japhb__: that's more like it :)
22:23 timotimo yeah, and i have no idea how to specify that kind of thing
22:23 japhb__ r: class Process::State is Int { method Bool { self == 0 } }; Process::State $a = 0; say so $a; Process::State $b = 1; say so $b;
22:23 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/FbD5VcuWGZâ�¤Two terms in a rowâ�¤at /tmp/FbD5VcuWGZ:1â�¤------> [32mod Bool { self == 0 } }; Process::State [33mâ��[31m$a = 0; say so $a; Process::State $b = 1[0mâ�¤    expecting any of:â�¤     …»
22:23 camelia ..rakudo-parrot 995207: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/Ga62JEBT0Bâ�¤Two terms in a rowâ�¤at /tmp/Ga62JEBT0B:1â�¤------> [32mod Bool { self == 0 } }; Process::State [33mâ��[31m$a = 0; say so $a; Process::State $b = 1[0mâ�¤    expecting any of:â�¤  â€¦Â»
22:23 FROGGS r: class P::S { method Numeric() { 42 }; method Bool { False } }; my $state = P::S.new; say ?$state
22:23 timotimo needs a my
22:23 camelia rakudo-parrot 995207, rakudo-jvm 882e33: OUTPUT«False␤»
22:23 FROGGS r: class P::S { method Numeric() { 42 }; method Bool { False } }; my $state = P::S.new; say +$state
22:23 camelia rakudo-parrot 995207, rakudo-jvm 882e33: OUTPUT«42␤»
22:24 japhb__ r: class Process::State is Int { method Bool { self == 0 } }; my Process::State $a = 0; say so $a; my Process::State $b = 1; say so $b;
22:24 FROGGS r: class P::S { method Numeric() { 42 }; method Bool { False } }; my $state = P::S.new; say $state unless $state
22:24 japhb__ r: class Process::State is Int { method Bool { self == 0 } }; my Process::State $a = 0; say so $a; my Process::State $b = 1; say so $b;
22:24 camelia rakudo-parrot 995207: OUTPUT«Type check failed in assignment to '$a'; expected 'Process::State' but got 'Int'␤  in block  at /tmp/gg2Ef2SOnK:1␤  in any  at /tmp/gg2Ef2SOnK:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/stage2/NQPHLL.nqp:1133␤ …»
22:24 camelia ..rakudo-jvm 882e33: OUTPUT«Type check failed in assignment to '$a'; expected 'Process::State' but got 'Int'␤  in block  at /tmp/yrUOSGMNhh:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/s…»
22:24 camelia rakudo-parrot 995207, rakudo-jvm 882e33: OUTPUT«P::S.new()␤»
22:24 camelia rakudo-jvm 882e33: OUTPUT«Type check failed in assignment to '$a'; expected 'Process::State' but got 'Int'␤  in block  at /tmp/cc8rDhDBg6:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.nqp:1290␤  in any command_eval at gen/jvm/s…»
22:24 camelia ..rakudo-parrot 995207: OUTPUT«Type check failed in assignment to '$a'; expected 'Process::State' but got 'Int'␤  in block  at /tmp/7BmIjsmxiP:1␤  in any  at /tmp/7BmIjsmxiP:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/stage2/NQPHLL.nqp:1133␤ …»
22:24 spider-mario joined #perl6
22:25 japhb__ r: class Process::State is Int { method Bool { self == 0 } }; my Process::State $a .= new(0); say so $a; my Process::State $b .= new(1); say so $b;
22:25 camelia rakudo-jvm 882e33: OUTPUT«Default constructor for 'Process::State' only takes named arguments␤  in method dispatch:<.=> at gen/jvm/CORE.setting:1171␤  in block  at /tmp/u5rewtmoLz:1␤  in any eval at gen/jvm/stage2/NQPHLL.nqp:1084␤  in any evalfiles at gen/jvm/stage2/NQPHLL.n…»
22:25 camelia ..rakudo-parrot 995207: OUTPUT«Default constructor for 'Process::State' only takes named arguments␤  in method new at gen/parrot/CORE.setting:809␤  in method new at gen/parrot/CORE.setting:804␤  in method dispatch:<.=> at gen/parrot/CORE.setting:1187␤  in block  at /tmp/NkyTRG…»
22:25 FROGGS japhb: it is not specced as being an Int
22:25 timotimo oh?
22:25 FROGGS it can give you the PID too
22:25 FROGGS but numifies/gistifies to state, and boolifies, well, you know
22:26 timotimo does it also do the +> 8 thing?
22:26 FROGGS I dont think so
22:27 japhb__ r: class Process::State is Int { has int $.state is box_target; has int $.pid; method Bool { self == 0 } }; my Process::State $a .= new(:state(0), :pid(42)); say so $a; my Process::State $b .= new(:state(1), :pid(23)); say so $b;
22:27 camelia rakudo-jvm 882e33: OUTPUT«===SORRY!===␤java.lang.ClassFormatError: Duplicate method name&signature in class file __P6opaque__96␤»
22:27 camelia ..rakudo-parrot 995207: OUTPUT«===SORRY!===␤Duplicate box_target for native int␤»
22:27 japhb__ r: class Process::State { has int $.state is box_target; has int $.pid; method Bool { self == 0 } }; my Process::State $a .= new(:state(0), :pid(42)); say so $a; my Process::State $b .= new(:state(1), :pid(23)); say so $b;
22:27 camelia rakudo-jvm 882e33: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤  in any  at gen/jvm/BOOTSTRAP.nqp:1201␤  in sub infix:<==> at gen/jvm/CORE.setting:3978␤  in method Bool at /tmp/peodoN2WvE:1␤  in block  at /tmp/peodoN2WvE:1␤  in any ev…»
22:27 camelia ..rakudo-parrot 995207: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤  in any  at gen/parrot/BOOTSTRAP.nqp:1218␤  in any  at gen/parrot/BOOTSTRAP.nqp:1209␤  in method Numeric at gen/parrot/CORE.setting:994␤  in sub infix:<==> at gen/parrot…»
22:27 japhb__ r: class Process::State is Cool { has int $.state is box_target; has int $.pid; method Bool { self == 0 } }; my Process::State $a .= new(:state(0), :pid(42)); say so $a; my Process::State $b .= new(:state(1), :pid(23)); say so $b;
22:28 camelia rakudo-jvm 882e33: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤  in any  at gen/jvm/BOOTSTRAP.nqp:1201␤  in sub infix:<==> at gen/jvm/CORE.setting:3978␤  in method Bool at /tmp/zJHeuS3ylq:1␤  in block  at /tmp/zJHeuS3ylq:1␤  in any ev…»
22:28 camelia ..rakudo-parrot 995207: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤  in any  at gen/parrot/BOOTSTRAP.nqp:1218␤  in any  at gen/parrot/BOOTSTRAP.nqp:1209␤  in method Numeric at gen/parrot/CORE.setting:994␤  in sub infix:<==> at gen/parrot…»
22:28 japhb__ Harumph.
22:28 FROGGS ETOOMUCHMAGIC
22:29 japhb__ r: class Process::State is Cool { has int $.state is box_target; has int $.pid; method Bool { $.state == 0 }; method Numeric { $.state }; }; (my Process::State $a .= new(:state(0), :pid(42)); say so $a; my Process::State $b .= new(:state(1), :pid(23)); say so $b;
22:29 camelia rakudo-jvm 882e33: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/_iZuWb2oFRâ�¤Unable to parse expression in parenthesized expression; couldn't find final ')' â�¤at /tmp/_iZuWb2oFR:1â�¤------> [32m .= new(:state(1), :pid(23)); say so $b;[33mâ��[31m<EOL>[0mâ�¤  â€¦Â»
22:29 camelia ..rakudo-parrot 995207: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/Y3JivOpK_9�Unable to parse expression in parenthesized expression; couldn't find final ')' �at /tmp/Y3JivOpK_9:1�------> [32m .= new(:state(1), :pid(23)); say so $b;[33m�[31m<EOL>[0m…»
22:29 japhb__ r: class Process::State is Cool { has int $.state is box_target; has int $.pid; method Bool { $.state == 0 }; method Numeric { $.state }; }; my Process::State $a .= new(:state(0), :pid(42)); say so $a; my Process::State $b .= new(:state(1), :pid(23)); say so $b;
22:29 lizmat gnight, #perl6!
22:29 camelia rakudo-parrot 995207, rakudo-jvm 882e33: OUTPUT«True␤False␤»
22:29 japhb__ o/ lizmat
22:29 japhb__ FINALLY.  SHEEESH.
22:30 japhb__ Sorry for the bot spam, folks.
22:30 timotimo japhb__: did you notice how you could use a native int, but not have to make a custom BUILD? :)
22:31 japhb__ You made that work?
22:31 japhb__ timotimo++
22:31 |newbie| timotimo: is it done?
22:31 timotimo yes, just today :D
22:31 timotimo well, between yesterday and today i guess
22:31 |newbie| timotimo: is it done?
22:32 timotimo is what exactly done?
22:32 |newbie| the panda change
22:32 timotimo i'm not sure i follow
22:33 |newbie| panda has a bug, it cannot delete files
22:33 timotimo ah, yes
22:33 FROGGS |newbie|: this was fixed right after the 2013.09 release
22:33 |newbie| so it is not in any relase
22:34 FROGGS |newbie|: not yet, but possible today or tomorrow
22:34 FROGGS well, it is in the 2013.10 compiler release fwiw
22:35 |newbie| because of the new release?
22:35 ajr joined #perl6
22:35 FROGGS lue is working on a new star release, contain this and a lot of other things
22:36 lue I suspect 12 and the rakudo.org part of 13, specifically the availability of its permission-holders, are going to be the only blockers to 2013.10 at this point.
22:36 FROGGS containing*
22:37 |newbie| what is the bug? Permission of folders, right?
22:37 timotimo unlink didn't persist if permissions didn't allow deletion
22:38 timotimo now it deletes files even if it's not permitted to, basically
22:38 timotimo well, that's worded poorly
22:38 |newbie| who is the possible?
22:38 |newbie| how
22:39 FROGGS this is possible on windows :o)
22:40 FROGGS so, unlink behaves now exactly like Perl 5's unlink...
22:40 |newbie| what are the permissiion incorrect, anyway?
22:40 |newbie| SHould not be those be fixed?
22:40 |newbie| what ->why
22:42 FROGGS we decided that P5's behaviour is okay, and it fixed the problem
22:43 |newbie| why did you choose this way?
22:43 FROGGS I said that right now
22:44 |newbie| you found one solution, fixing the unlink
22:44 |newbie| did you consider the other way as well?
22:45 FROGGS yes, but it would be rather annoying if you had to check+change the flags of hundreds of files just to delete them afterwards
22:45 FROGGS so, if you can gain control over a file, you sort of have already control over it
22:46 FROGGS so it does not make much sense to forbid deleting it
22:47 |newbie| ok bye
22:48 xinming joined #perl6
22:49 Mouq lizmat: I need to → bed, but basically the problem is that it ends up being called as $var."&postcircumfix<[ ]>"(|whatev), which, beacuse of the quotes, will currently always call as a method. Adding a dispatch:<quote> method and using it at Actions L3863 should work, but I'm not sure if that's really the best way to do this
22:50 timotimo man, rakudo's ram usage ...
22:51 Mouq Maybe there could be two dispatch methods for <.=>
22:51 * Mouq out
22:51 FROGGS gnight Mouq
22:52 jnthn Wait, what's that trying to fix?
22:53 timotimo the @a.=[1,2,3] thing
22:53 FROGGS I believe some splice mutator thing
22:53 FROGGS yeah
22:53 jnthn hmmm
22:54 jnthn Well, i fit's gonna work, it should be a general solution for postfixes too
22:54 timotimo not all postcircumfix are subs :\
22:57 jnthn yeah, postcircumfix:<( )> is the odd one out now :(
23:01 lue so, unless someone insists that I run the exact same spectest on the exact same rakudo...
23:01 lue ✔ ⑩ Verify tarball
23:02 * timotimo desists
23:02 xinming joined #perl6
23:04 raiph joined #perl6
23:04 BenGoldberg joined #perl6
23:04 lue ✔ ⑪ Tag Star "2013.10"
23:05 lue Unsurprisingly, I have not the permissions for 12.
23:06 BenGoldberg preflex: seen yoleaux
23:06 preflex yoleaux was last seen on #perl6 3 days, 3 hours, 44 minutes and 13 seconds ago, saying: 18:23Z <diakopter> Mouq: awesome work on stdsigspace Mouq++
23:07 lue pmichaud, jnthn, masak, PerlJam, tadzik, moritz : if one of you could scp the R* tarball, that'd be great :) the md5sum of the tarball, so you can verify generation went smoothly, is ef8d1e47d11c5d5ad663405897fb4ee0.
23:07 lue (if you could also put the announcement on rakudo.org that'd be super)
23:08 jnthn lue: Sadly, my key for that is sat on my machine at home, which is a long way away.
23:08 lue :(
23:08 lue you couldn't, say... scp the key, could you? :)
23:09 FROGGS win *cough* dows
23:09 jnthn The machine isn't even running, given I'm away for over a week.
23:10 lue you couldn't, say... scp a power-on command, could you? :P
23:10 * lue knows moritz will be around late tonight from his POV, at least.
23:12 lue "his" being me, that is.
23:39 rurban joined #perl6
23:43 dalek nqp: 1161844 | dwarring++ | examples/rubyish/ (3 files):
23:43 dalek nqp: rubyish - implement compound-statement/multi-line interpolation
23:43 dalek nqp: review: https://github.com/perl6/nqp/commit/1161844b45
23:43 benabik joined #perl6
23:45 Celelibi joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo