»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by sorear on 25 June 2013.
00:02 lizmat_ joined
BenGoldberg While I realize that .infinite is part of the spec, it seems to me that a more perl6-ish way to identify (in)finiteness would be to have to separate roles, one called List::Finite (or perhaps finte-list) and the other called List::Infinite (or infintite-list)... to mark a gather/take as finite or infinite, one would then use 'gather ... but infinite-list'. This would also allow multi method 00:02
dispatch based on whether or not a list is finite.
00:03 woolfy left
BenGoldberg r: say (^Inf)[^10].infinite 00:03
camelia rakudo 5f4601: OUTPUT«Nil␤»
BenGoldberg r: say (^Inf).infinite
camelia rakudo 5f4601: OUTPUT«True␤»
BenGoldberg r: say (^10).infinite 00:04
camelia rakudo 5f4601: OUTPUT«False␤»
BenGoldberg n: say (^Inf)[^10].infinite
00:04 lizmat left
camelia niecza v24-88-g1f87209: OUTPUT«Unhandled exception: Unable to resolve method infinite in type Parcel␤ at /tmp/4g37q1oGA3 line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4579 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4580 (module-CORE @ 576) ␤ at … 00:04
BenGoldberg Also... if an infinite list is sliced by a finite list, the result should be finite, not Nil/unknown 00:05
00:08 bearman joined 00:10 Mjbmr joined 00:11 nyuszika7h left, Mjbmr left 00:12 nyuszika7h joined 00:18 woolfy joined 00:25 ajr_ left 00:26 Pleiades` left 00:30 Pleiades` joined 00:49 bearman left 01:04 kbaker joined 01:08 fgomez joined 01:11 aindilis joined 01:26 benabik left, benabik joined 01:40 FROGGS_ joined 01:44 Mouq joined, FROGGS left
slavik what does '-->' mean? (example: int --> int, CArray[int] --> int), is that the lazy iterator? 02:12
benabik slavik: sub (--> int) is a sub that returns an int. 02:14
slavik ahh
benabik Parameters before, return type after.
slavik oh, I see
ok
like haskel :D
ty
JimmyZ or sub() returns int { } 02:25
02:44 EvanTeitelman left 02:45 daniel-s left 02:52 EvanTeitelman joined, EvanTeitelman left 02:53 EvanTeitelman joined 02:58 EvanTeitelman left
colomon is having problems building jakudo at the moment. :\ 03:08
03:11 abnorman left 03:12 wtw joined 03:14 wtw_ left 03:29 preflex_ joined, ChanServ sets mode: +v preflex_ 03:30 preflex left 03:31 preflex_ is now known as preflex 03:40 xinming left 03:43 xinming joined 03:44 kbaker left 04:04 FROGGS_ left 04:12 fridim_ joined 04:14 abnorman joined 04:18 BenGoldberg left 04:22 FROGGS joined 04:50 birdwindupbird joined 04:58 jaldhar joined
Mouq S05 (Match Objects): “In boolean context it evaluates as true or false (i.e. did the match succeed?): 04:58
if /pattern/ {...}
# or:
/pattern/; if $/ {...}” 04:59
Shouldn't that be m/pattern/ ? I'm only reading S05 now, so I'm not sure
r: $_ = 'abcd'; if /qwerty/ { say "Hi there" } 05:00
camelia rakudo 144a3f: ( no output )
Mouq r: $_ = 'abcd'; if m/qwerty/ { say "Hi there" } 05:01
camelia rakudo 144a3f: ( no output )
Mouq Hm
n: $_ = 'abcd'; if /qwerty/ { say "Hi there" }
camelia niecza v24-88-g1f87209: ( no output )
Mouq Oh! “Specifically, a /.../ matches immediately in a value context” :p 05:03
05:05 SamuraiJack joined 05:26 rindolf joined, rindolf left 05:30 thou left 05:38 FROGGS left
moritz good morning 05:49
tirania.org/blog/archive/2013/Aug-15.html # this reminded me about jnthn++'s talk about threads and async in Perl 6
(at yapc::eu) 05:50
05:51 FROGGS joined, bearman joined
lizmat_ good *, #perl6! 05:53
packing up in Kiev 05:55
moritz \o lizmat_
05:56 lizmat_ is now known as lizmat
moritz
.oO( don't forget to pack your underscore )
 05:56
lizmat my underscore just got worn out :-)
moritz yesterday the flight from Kiev to Vienna got delayed by an hour 05:57
luckily we had enough time to catch the connection
lizmat is lucky to have a direct flight 05:58
06:04 FROGGS left 06:09 kingbeast left 06:11 FROGGS joined 06:27 dmol joined 06:36 dmol left
dalek kudo/nom: c20be1f | (Elizabeth Mattijsen)++ | src/core/ (2 files):
Reliably handle mixins with is_default and is_dynamic
 06:42
06:45 bearman left 07:03 Mouq left 07:05 SmokeMachine left
timotimo o/ 07:21
FROGGS moin timotimo 07:30
07:33 daxim joined
timotimo for some reason, riding trains across the country always makes me super excited 07:36
in a mixed-good-and-bad way 07:37
tadzik hello #perl6 07:51
07:55 Seganku joined
masak morning, #perl6 07:56
moritz morning, masak | tadzik | * 08:00
erm, that should have been & :-)
masak eww, a junction of a whatever... :P
timotimo if you junction in a *, it should take every object that it can find in the running program and cram it in there 08:01
masak like a black hole. 08:03
tadzik: want to meet up for more tourism?
timotimo i really need to get playing with the async stuff
08:05 fhelmberger joined
dalek kudo/nom: 10b9ac4 | (Elizabeth Mattijsen)++ | src/core/ (2 files):
Hmmm... seems I broke it, so reverting

This reverts commit c20be1f21b69c3eb61c1795fbe04ecab6cd2a0fc.		 08:05
tadzik masak: sure thing :)
08:06 shachaf left
tadzik I'm ready (almost) when you are 08:06
masak tadzik: I suggest metro to avoid sunburn.
tadzik I have to checkout within an hour or so
08:06 shachaf joined
masak tadzik: we're heading down to lobby now, I think. we can wait for you. 08:06
dalek rlito: 267ef7e | (Flavio S. Glock)++ | / (5 files):
Perlito5 - js - fix string conversion, string increment and decrement
 08:07
tadzik masak: ok, I will slowly progress in your direction
lizmat decommuting&
08:07 lizmat left, crab2313 joined, shachaf left
masak tadzik: sounds good. 08:07
08:09 woolfy left
timotimo dumps lots of todos for rakudo into questhub 08:11
08:16 rindolf joined 08:17 Seganku left 08:21 crab2313 left 08:26 denis_boyun joined 08:29 Seganku joined
timotimo i'm getting lots of errors like this: symbol: class LexoticInstance 08:31
location: package org.perl6.nqp.sixmodel.reprs
src/vm/jvm/runtime/org/perl6/rakudo/RakOps.java:59: error: cannot find symbol
what do i have to do to fix this? or does somebody have a known good jakudo commit?
i wanted to play with the threading stuff :( 08:38
oh, that looks like not my rakudo, but my nqp version is wrong 08:39
08:39 PacoAir joined
timotimo still can't get it to work :( 08:47
08:57 dayangkun joined 08:58 dmol joined, dayangkun left
moritz timotimo: are you trying with nqp/master? 08:58
timotimo yes 08:59
moritz tries too 09:00
09:07 PacoAir left 09:09 dmol1 joined 09:11 dmol left
moritz huh, I get 09:25
./perl6 --target=jar --output=blib/Test.jar lib/Test.pm
===SORRY!===
org.perl6.nqp.runtime.UnwindException
mathw o/ 09:39
09:43 berekuk joined 09:44 bearman joined
timotimo huh 09:46
09:50 NIco_ joined 09:51 NIco_ left 10:06 homesteak joined 10:14 PacoAir joined 10:15 sromanov joined
colomon moritz: That's what I got when I tried to build jakudo last night. 10:30
timotimo in my new tablebgrammar it is allowed to have more than one space in your columns if they are below an explicit header 10:31
10:35 mtk left 10:42 mtk joined 10:57 homesteak left 11:09 abnorman left 11:12 not_gerd joined
not_gerd r: my %h = <a 1 b 1 c 2>; %().push(%h.invert).perl.say 11:14
camelia rakudo 10b9ac: OUTPUT«("1" => ["a", "b"], "2" => "c").hash␤»
not_gerd is there a less obscure way to do that?
moritz isn't aware of one 11:16
11:18 rindolf left
not_gerd perhaps classify could take an optional argument that transforms the value is stores? 11:19
%h.classify(*.value, :map(*.key))
11:21 bearman left
moritz is Hash.classify still specced? 11:21
r: say Hash.^can('classify') 11:22
camelia rakudo 10b9ac: OUTPUT«classify␤»
moritz hm
not_gerd: seems like it should be easy to map yourself before classifying
or after, depending on what you want
remember that map allows funny stuff with closures that expect multiple arguments, so that's non-trivial to replicate inside another function 11:23
11:26 rhr joined
not_gerd r: my %h = <a 1 b 1 c 2>; %h.classify(*.value).pairs.map({ ; .key => .value>>.key }).hash.perl.say 11:30
camelia rakudo 10b9ac: OUTPUT«("1" => $("a", "b"), "2" => $("c",)).hash␤»
not_gerd I wouldn't call that easy exactly
ot do I miss something obvious?
^or 11:31
r: my %h = <a 1 b 1 c 2>; %h.classify(*.value).pairs.map({ ; .key => .value.map(*.key).item }).hash.perl.say
camelia rakudo 10b9ac: OUTPUT«("1" => ().list.item, "2" => ().list.item).hash␤»
not_gerd ^bug?
moritz I don't know Hash.classify enough to know 11:32
not_gerd moritz: I suspect it's the map within a map that causes the problem 11:33
see that it works when replacing .map(*.key) with >>.key
r: my @a = [<1 2 3>], [<4 5 6>]; @a.map({ .map(* + 1).item }).perl.say 11:38
camelia rakudo 10b9ac: OUTPUT«((2, 3, 4).list.item, (5, 6, 7).list.item).list␤»
not_gerd ^ works, so it might indeed be classify-related
moritz r: my %h = <a 1 b 1 c 2>; %h.classify(*.value).perl.say 11:41
camelia rakudo 10b9ac: OUTPUT«("1" => ["a" => "1", "b" => "1"], "2" => ["c" => "2"]).hash␤»
11:41 slavius joined
slavius perl6: say "Hello slavi" 11:42
camelia rakudo 10b9ac, niecza v24-88-g1f87209: OUTPUT«Hello slavi␤»
11:42 bearman joined
slavius perl6:   say "Nice :-)" 11:43
camelia rakudo 10b9ac, niecza v24-88-g1f87209: OUTPUT«Nice :-)␤»
11:44 slavius left 11:59 skids left
not_gerd rn: printf '%1$i', 1, 2, 3 12:00
camelia rakudo 10b9ac: OUTPUT«'$' is not valid in sprintf format sequence '%1$'␤ in sub sprintf at src/gen/CORE.setting:3154␤ in sub printf at src/gen/CORE.setting:3159␤ in block at /tmp/4N73w6DT_m:1␤␤»
..niecza v24-88-g1f87209: OUTPUT«1»
12:02 iSlug joined
FROGGS not_ge[12:35] *** ilogger2 joined perl6 12:04
12:35 ChanServ sets mode: +v ilogger2
FROGGS grondilu: did you mean 'once' ? 12:37
rn: for <a b c> { .print; once { print '!' } }
camelia niecza v24-88-g1f87209: OUTPUT«===SORRY!===␤␤Undeclared routine:␤ 'once' used at line 1␤␤Unhandled exception: Check failed␤␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 1502 (die @ 5) ␤ at /home/p6eval/niecza/src/STD.pm6 line 1147 (P6.comp_unit @ 36) ␤ at /hom…
..rakudo 10b9ac: OUTPUT«a!bc»
grondilu FROGGS: I did not know this one. (I guess I need to read S04 again) 12:38
ty
FROGGS you're welcome 12:39
grondilu r: for <3 1 4 1 5> { .print; once print '.' }
camelia rakudo 10b9ac: OUTPUT«3.1415»
grondilu was just checking if he had to use {}
FROGGS yeah 12:40
colomon has anyone gotten jakudo to build today, or is it just me and moritz having issues? 12:49
FROGGS I can try... 12:52
12:58 bluescreen10 joined
FROGGS I get the following when making rakudo: src/vm/jvm/runtime/org/perl6/rakudo/Binder.java:5: error: package org.perl6.nqp.runtime does not exist 13:01
13:02 denis_boyun joined 13:07 kbaker joined 13:09 ajr joined 13:10 ajr is now known as Guest10032, Guest10032 is now known as ajr_ 13:27 skids joined, denis_boyun left
FROGGS nqp-jvm: say("1.01".WHAT) 13:29
camelia nqp-jvm: OUTPUT«␤»
masak oh hai from l'airport!
FROGGS hi masak 13:30
TimToady oh hai from l'ome
FROGGS o/
TimToady or would that be oh 'ai? :)
TimToady was never particularly noted for consistency though... 13:31
masak moritz: I think Hash.classify got un-spec'd, but it may not have got un-impl'd.
FROGGS o' 'ai froḿ l'ome 13:32
TimToady when I try to make rakudo/jvm, I get: make: *** No rule to make target `src/Perl6/ConstantFolder.nqp', needed by `blib/Perl6/ConstantFolder.jar'. Stop. 13:34
huf accent on an 'm'?! that's going too far!
FROGGS :o) I just wanted to "say" it nasal 13:36
is "nasal" the right word?
TimToady well, English speakers should probably not be making fun of other people's orthographic systems, glass houses and all that...
huf m is already pretty nasal :)
TimToady yes, "nasal" is a perfectly good linguistics word :)
FROGGS k
huf TimToady: i found interspel the other week. it's pretty clever and would let english speakers throw those glass houses around 13:37
TimToady that 'ouses to you :P
huf you mean "huss", surely
[Coke] TimToady: when you say rakudo/jvm do you mean the jvm branch?
13:38 sqirrel joined
TimToady yes 13:38
[Coke] (should be using nom)
it's all nom these days.
TimToady my nom has defaulted to jvm for a few weeks now
but maybe that came unstuck somehow
attempting to reconf nqp to see if it helps 13:39
masak was just gonna suggest that. 13:44
FROGGS my nqp/head+rakudo/nom is broken too
err, nqp/master... both HEAD 13:45
I might clean up nqp just to maske sure
TimToady nqp rebuilt fine, now for rakudo
FROGGS no, mine still dies with: src/vm/jvm/runtime/org/perl6/rakudo/Binder.java:5: error: package org.perl6.nqp.runtime does not exist 13:49
TimToady now compiling lib/Test.pm I get: org.perl6.nqp.runtime.UnwindException 13:51
did you re-install the nqp?
FROGGS I did
TimToady and reconfig rakudo with that nqp?
FROGGS hmmm, no, I did that before reinstalling... 13:52
TimToady did yours die before or after compiling the setting? mine was after
FROGGS k, did git clean -dxf, configured, made, but it explodes the same
mine earlier 13:53
perl -MExtUtils::Command -e mkpath bin
javac -source 1.7 -cp /nqp-runtime.jar:/asm-4.1.jar:/asm-tree-4.1.jar:/jline-1.0.jar -g -d bin src/vm/jvm/runtime/org/perl6/rakudo/*.java
src/vm/jvm/runtime/org/perl6/rakudo/Binder.java:5: error: package org.perl6.nqp.runtime does not exist
pretty early actually
TimToady yeah, mine got further for some reason
32 or 64-bit?
mine's 64
FROGGS mine too (ubuntu) 13:54
TimToady did pull right before too, so we should have the same codebase
FROGGS Today
13:51 UTC We are currently investigating problems with GitHub.com. We'll provide an update once we have more information.
-.-
TimToady never did the clean though
so maybe something got cleaned out too much 13:55
FROGGS yeah, and I wanted to have a look at the latest commits, because I think grondilu said he has problems for two days now
TimToady tries again with 'make clean' 13:56
13:56 jeff_s1 joined
tadzik wow, Borispol has a working free wifi 13:56
TimToady FROGGS: are you up-to-date on your java version? 13:57
javac 1.7.0_25 13:58
FROGGS java version "1.7.0_25"
OpenJDK Runtime Environment (IcedTea 2.3.10) (7u25-2.3.10-1ubuntu0.13.04.2)
OpenJDK 64-Bit Server VM (build 23.7-b01, mixed mode)
yeah
TimToady 4gig memory here, if it makes a difference 13:59
masak there's some trouble building rakudo-jvm in the backlog. not exactly the same as yours, but build problems. 14:00
the solution there was rebuilding the nqp dependency, I think.
TimToady strange, the parse went from 83 to 98 seconds
masak: we both rebuilt our nqp 14:01
FROGGS I've got 8gig
TimToady same failure in Test.pm for me
masak ok, then I think we're in bisect land.
14:03 thou joined
FROGGS masak: how do I get a gommit id from a week ago? 14:04
(without a working github website)
masak FROGGS: git show master@{one week ago} 14:05
FROGGS that sounds a bit like a joke, problem is, I trust you
fatal: ambiguous argument 'master@{one': unknown revision or path not in the working tree. 14:06
masak FROGGS: is that Rakudo? then it's nom, not master
FROGGS nqp
I'm on master fwiw 14:07
masak ah, I needed to do `git show 'master@{one week ago}'`
FROGGS awesome # masak++
tadzik bah, I can't confinve NQP::Configure to like my --gen-nqp on JVM 14:10
convince, even
14:11 rindolf joined
TimToady did you mean --gen-parrot? 14:11
oh, nm
if you mean on nom, I use --with-nqp
FROGGS yeah, --with-nqp ~/dev/nqp-jvm/install/nqp # or so 14:12
TimToady goes to get more coffee, at this point more a necessity than a luxury...
FROGGS yeah, --with-nqp=~/dev/nqp-jvm/install/nqp # <---
tadzik yeah, I wanted it to DTRT for --gen-nqp
but it seems to hang on git-fetch somehow
I eventually gave up ;)
I blame airport wifi
FROGGS tadzik: github has problems
tadzik ah
maybe I shouldn't have git checkout -f'd my changes 14:13
FROGGS AHHHHHHHHHH 14:15
it fails so early when I do: perl ConfigureJVM.pl --with-nqp=~/..., I really have to supply an absolute path 14:16
stage parse 56s, I love it 14:18
(that is both nqp and rakudo from a week ago)
14:19 sqirrel left 14:20 sqirrel joined 14:27 ivanoff joined
alexghacker [Coke]: yesterday you said that NFG / graphemes are not currently the default (which is viewed as a bug, or at least a missing feature), but it occurs to me that maybe there is a way to get NFG semantics if I ask perl6 nicely 14:32
[Coke] if by that you mean "write your own code", sure. 14:38
but SFAIK, there's nothing baked into the language at this point, in any impl. 14:39
14:39 Psyche^_ joined 14:41 colomon joined
TimToady FROGGS: so, did your build get past Test.pm? 14:41
FROGGS TimToady: yes, the versions from a week ago made and installed fine 14:42
14:42 nyuszika7h joined
FROGGS but bisecting is hard because ops got added 14:42
TimToady ah, you were doing from a week ago 14:43
FROGGS now I am on nqp HEAD, and try to bisect rakudo, starting at HEAD 14:44
TimToady doesn't know whether to hope it fails :)
FROGGS well, I hope rakudo/HEAD fails here :o) 14:45
TimToady FROGGS++ for bisecting this
FROGGS np 14:46
TimToady or at least monosecting it so far :)
FROGGS hehe
14:46 benabik joined
FROGGS monosecting sounds more like niecza 14:46
TimToady looked through the git logs, but didn't see anything obvious
FROGGS ./perl6 --target=jar --output=blib/Test.jar lib/Test.pm 14:48
===SORRY!===
org.perl6.nqp.runtime.UnwindException
so far so good
TimToady that looks familiar :) 14:50
TimToady is looking at the palindrom detector containing .flip eq $_ and thinking, wouldn't it be nice if we could write that .flip eq ._ or some such 14:57
just to keep the parallelism 14:58
*literary parallelism
FROGGS what about .flip eq .Str ? 14:59
geekosaur ._ looks kinda off conceptually, though. seems like * would be a better fit somehow?
hm, .id
TimToady well, it's the identi...yeah
geekosaur or maybe that's what ._ is
TimToady and ._ would basically just a different name for $_ :) 15:00
as a term
I dunno if it's really an improvement, but we should probably have an identity method in any case 15:01
rjbs Another day answering encoding questions on p5p, another day longing for Buf.
TimToady nodnod
or maybe even nodnodnod 15:03
rjbs The good part is that I felt really really good about my answer and providing it has improved my day already. 15:04
TimToady rjbs++
rjbs I may reward myself by turning my ZSCII p6 library into a proper module and putting it in the ecosystem.
You know, for all those p6 Z-Machine emulators we're expecting. 15:05
TimToady now where have I heard this before...
rjbs Yow, looks like I haven't backported 90% of ZMachine::ZSCII to my write-z5.p6 code. So there are two parts to do. 15:06
FROGGS what is a Z-Machine?
rjbs perl6++ # hyphen-minus in identifiers!
FROGGS: It's the VM from the mid-80's on which Zork and friends ran. 15:07
TimToady geekosaur: but we can't afford to use up the .id method, since that's too useful in userland
rjbs FROGGS: ZSCII is its bizarre charset and encoding.
("pronounced to rhyme with xyzzy")
FROGGS rjbs: I've heard about a Z80 chip... is it that one?
rjbs No, totally unrelated.
FROGGS ahh, okay
rjbs The Z-Machine is more like the JVM. :-)
FROGGS ahh, I see
15:07 kbaker left
rjbs It's actually a fun little machine, read the spec someday if you're bored. 15:07
FROGGS ahh, now I read the VM bit.... :o) 15:08
rjbs: I have two kids, sooo........ :P
TimToady you are in a maze of twisty little adventure games, all the same...
15:08 ajr_ left
rjbs FROGGS: So you'll need to show them Zork anyway... :) 15:08
FROGGS *g*
someday, yes
TimToady well, first adventure, since that predated everything else :)
rjbs TimToady: Ovid has just written a p5 'terp for the Scott Adams games, too...
TimToady
.oO(That adventure game, it's so derivative. :)
 15:09
15:09 ajr joined
rjbs At OSCON, I tried to sit at any of the "topic tables" that weren't what I do. 15:09
TimToady I have a radio in my brain!
15:09 ajr is now known as Guest24973
rjbs So I sat down with the "UX Designers" table and found out the guy next to me was trying to learn programming and had just written a text adventure in Python (or was it Ruby?). 15:10
I totally geeked out and showed him the state of the art for writing them. Fun!
dalek rlito: b3566a1 | (Flavio S. Glock)++ | / (3 files):
Perlito5 - grammar - simplify the grammar compiler
rjbs TimToady: Um... I hope that is an allusion that was lost on me! 15:11
TimToady well, then there was DDL "Dungeon Definition Language", which was kind of a Lisp variant, as I recall
"You're in the COSMIC ALL! How did you get here?!?"
(sez the room contain all the other rooms)
15:11 Guest24973 is now known as ajr_
TimToady *taining 15:11
FROGGS when rakudo gets a bit faster this year I'm going to port my games to Perl 6 too (because I really don't wanna write Perl 5 anymore for some reason) 15:13
rjbs Right, I know a little about DDL.
It was an MDL variant, of course.
Later it became ZIL, the Zork Implementation Language, which is mostly lost to time, but we have some fragments. 15:14
TimToady the allusion was to Bureaucracy, btw
rjbs Ah!
I have not played that one yet.
TimToady it was an Amiga game
rjbs About a year ago I promised myself I would play through all of them in order.
Amiga was an Infocom game, so it plays on any Z-Machine!
Including the one that will be written for p6. :)
Er, s/Amiga/Bureaucracy
TimToady it was a Douglas Adams game 15:15
rjbs Yeah. I've heard it stank. :(
TimToady I never solved it, so maybe it did. :) 15:16
rjbs rjbs.manxome.org/rubric/~rjbs/int-fiction <-- haven't made progress on "The Great Infocom Replay" since February!
I'll surely make some on the flights to/from YAPC::Asia, anyway.
FROGGS TimToady: this one is the bad commit: github.com/rakudo/rakudo/commit/14...4f2bdc1e86 15:29
lizmat reverted a commit, but it looks like she reverted the wrong one
so, this one is the last that is okay: github.com/rakudo/rakudo/commit/5f...2b5efdbc60
but btw, my changes to nqp::sprintf seems to work (specifying parameter index using 1$, 3$, ...) 15:31
15:32 sqirrel left 15:41 slavik joined
TimToady wonders if the right way to make p5 interoperate with NFG is to give p5 an NFG implementation in parallel with its utf8 implementation... 15:48
then p6 could call into p5 without translating to utf8 and back 15:49
slavik nfg? 15:51
TimToady one step beyond NFC, where we synthesize codepoints for missing precomposed chars
15:56 colomon left 16:05 pmurias joined
pmurias TimToady: re give p5 and NFG, can it be done without a slowdown? 16:05
s/slowdown/paying the price of a check on every string operation/? 16:06
16:07 ivanoff left 16:08 wtw_ joined, wtw_ left, wtw joined 16:16 diakopter joined
diakopter r: [say $] 16:16
camelia rakudo 10b9ac: OUTPUT«===SORRY!=== Error while compiling /tmp/cNQK4oKRMG␤Non-declarative sigil is missing its name␤at /tmp/cNQK4oKRMG:1␤------> [say ⏏$]␤ expecting any of:␤ argument list␤ prefix or term␤ prefix or meta-prefix␤…
diakopter r: [say $$]
camelia rakudo 10b9ac: OUTPUT«===SORRY!=== Error while compiling /tmp/VtkPbhWkNy␤Non-declarative sigil is missing its name␤at /tmp/VtkPbhWkNy:1␤------> [say $⏏$]␤ expecting any of:␤ argument list␤ prefix or term␤ prefix or meta-prefix…
diakopter r: [say my $] 16:17
camelia rakudo 10b9ac: OUTPUT«(Any)␤»
diakopter std: [say my $]
camelia std c2215f0: OUTPUT«===SORRY!===␤Unsupported use of $] variable; in Perl 6 please use $*PERL_VERSION at /tmp/6oNqrbC5vP line 1:␤------> [say my $]⏏<EOL>␤Parse failed␤FAILED 00:00 44m␤»
diakopter mwahaha
16:18 colomon joined 16:19 pmurias left 16:20 kaare_ joined
gfldex is there video evidence for jnthn's concurrency YAPC talk? 16:21
16:23 dmol joined 16:27 japhb joined
grondilu rn: say max :by(*.elems), ().list 16:27
camelia rakudo 10b9ac, niecza v24-88-g1f87209: OUTPUT«-Inf␤»
grondilu does not understand where that -Inf comes from :/ 16:28
16:28 ajr_ left
diakopter probably that's the default the algorithm starts with 16:28
grondilu still, it's supposed to sort by length (so ok the default is -Inf), but the result should be a member of the list, not a length. 16:29
so I would have expected (), Nil or something. 16:30
not a number anyway, even infintely small. 16:31
in case you're wondering, I come up with this issue while trying to write code for rosettacode.org/wiki/Longest_increa...ubsequence 16:32
colomon -Inf isn't infinitely small, it's infinitely big and negative. 16:36
itz ==> Building Rakudo::Debugger 16:37
error:imcc:loadlib directive could not find library `perl6_group' in file 'perl6-debug.pir' line 1
colomon rn: say ().list.max(:by(*.elems))
camelia niecza v24-88-g1f87209: OUTPUT«Unhandled exception: Excess arguments to List.max, unused named by␤ at /home/p6eval/niecza/lib/CORE.setting line 0 (List.max @ 1) ␤ at /tmp/j9ZUzQRNqn line 1 (mainline @ 4) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4579 (ANON @ 3) ␤ at /home/p6eva…
..rakudo 10b9ac: OUTPUT«-Inf␤»
colomon rn: say ().list.max()
camelia rakudo 10b9ac, niecza v24-88-g1f87209: OUTPUT«-Inf␤»
colomon Yeah, that's just the default "nothing to see here" value 16:40
rn: say (().list).max() 16:41
camelia rakudo 10b9ac, niecza v24-88-g1f87209: OUTPUT«-Inf␤»
colomon rn: say [().list].max()
camelia rakudo 10b9ac, niecza v24-88-g1f87209: OUTPUT«-Inf␤»
16:45 spider-mario joined 16:48 isBEKaml joined
TimToady it's just the maximum of an an empty list in each case 16:48
r: say [min]() 16:50
camelia rakudo 10b9ac: OUTPUT«Inf␤»
TimToady r: say [max]()
camelia rakudo 10b9ac: OUTPUT«-Inf␤»
TimToady that one
you've just discovered there are many ways to write something that is a list of 0 elements :) 16:51
16:58 EvanTeitelman joined, colomon left, isBEKaml` joined
dmol r: say [+]() 16:58
camelia rakudo 10b9ac: OUTPUT«0␤»
16:58 isBEKaml` left
dmol r: say: [=]() 16:59
camelia rakudo 10b9ac: OUTPUT«===SORRY!=== Error while compiling /tmp/G9wWm1kYyh␤Two terms in a row␤at /tmp/G9wWm1kYyh:1␤------> say⏏: [=]()␤ expecting any of:␤ argument list␤ postfix␤ infix stopper␤ infix or meta-infix␤ …
TimToady std: [=]()
camelia std c2215f0: OUTPUT«===SORRY!===␤Cannot reduce with = because list assignment operators are too fiddly at /tmp/bhaci_efat line 1:␤------> [=]⏏()␤Check failed␤FAILED 00:00 42m␤»
TimToady that's less LTA
r: say [||]()
camelia rakudo 10b9ac: OUTPUT«False␤»
TimToady r: say [||]
camelia rakudo 10b9ac: OUTPUT«False␤»
TimToady okay, don't need the () anymore, good 17:00
17:02 isBEKaml left 17:03 EvanTeitelman left
tadzik WAW! \o/ 17:04
17:05 araujo joined
geekosaur ן 17:06
17:06 GlitchMr joined 17:09 Celelibi joined 17:17 kaare_ left 17:21 colomon joined 17:23 xinming joined 17:32 colomon left 17:44 abnorman joined
PerlJam when building rakudo-jvm, what's the solution to an "Could not reserve enough space for object heap" error? 17:46
[Coke] looks like the rakudo.jvm build failure is tanking the spec today.
arglebargle. my local copy of the daily run repo has a screwed up ysterday, looks like I never dealt with it. 17:52
diakopter oopsie 17:53
[Coke] and now removing it is a PITA because I have a valid today after it.
17:54 vk_ joined 17:55 crab2313 joined
dalek rl6-roast-data: c4d33cd | coke++ | / (5 files):
today (automated commit)
 17:57
[Coke] saved the summary for yesterday, pretty sure we didn't have the .out files today.
jnthn: rakudo.jvm now at 0% - update slides immediately!
diakopter heh 17:58
17:59 SamuraiJack joined
[Coke] diakopter: now is your chance for moar's come-from-behind victory 17:59
diakopter hah.
[Coke] seen lizmat 18:09
.seen lizmat
yoleaux I saw lizmat 08:07Z in #perl6: <lizmat> decommuting&
preflex yoleaux: you have 2 new messages. '/msg preflex messages' to read them.
[Coke] .to preflex DIE DIE DIE
yoleaux [Coke]: I'll pass your message to preflex.
geekosaur is there actually consensus that preflex should not be here? if so I can talk to mauke about it 18:11
PerlJam geekosaur: preflex only seems to pipe up when yoleaux (another bot) speaks. 18:13
geekosaur: it's annoying at the very least.
geekosaur has no idea who/what put messages for one bot in the other. preflex is *supposed* to be silent most of the time 18:14
but it has no idea that someone is trying to send messages to another bot via it
PerlJam aye.
(I rarely notice it, but in the last few days either I've been paying better attention, or it's been happening more often) 18:15
geekosaur I mean, if that's the only issue, I can ask mauke to clear those messages (but if we don't know where they came from, what prevents it happening again?P
otoh if there's no actual purpose being served by the bot, no reason for it to be here
[Coke] IWBNI preflex could keep track of bots and not bother passing along things for them." clearing out those messages would be a fine start, though. 18:17
looking at the help for preflex, I see nothing in there we need that yoleaux isn't doing. 18:18
can mauke tell us who asked for it to be in here and why? that'll also help.
tadzik after-landing push :) 18:20
dalek p/jvm-sockets: a81a89a | tadzik++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Don't leave VMArray in a bogus state
[Coke] tadzik: I'll fix star tonight - which modules are replacing file tools? 18:23
tadzik [Coke]: Shell::Command and File::Find
thank you, [Coke]++
I can't grok submodules
[Coke] me neither, but I have a night off. ;)
tadzik hah :)
[Coke]: S:C depends on F:F
I need to read up on read() and recv() on sockets 18:24
oh, not specced. I can do what I want >:>
18:32 woolfy joined 18:36 Mouq joined
dalek p/jvm-sockets: 0a65414 | tadzik++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Fix a copy-pasto
 18:37
p/jvm-sockets: 46d2162 | tadzik++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Make nqp::accept() less wrong, and thus make it actually work
 18:43
18:58 donaldh joined
tadzik fun fact: rakudo-jvm for $*OS returns 'Linux', which confuses roast, which expects 'linux', as Parrot does 18:58
I don't see a spec on this, I'll just make roast less stric
labster $*OS isn't specced, it's supposed to be $*KERNEL. 18:59
FROGGS r: say $*KERNEL 19:01
camelia rakudo 10b9ac: OUTPUT«Dynamic variable $*KERNEL not found␤ in method gist at src/gen/CORE.setting:11391␤ in method gist at src/gen/CORE.setting:990␤ in sub say at src/gen/CORE.setting:12287␤ in block at /tmp/TE9yOT7gq_:1␤␤»
labster Yeah, I'm well aware that no one actually implements it as $*KERNEL. I get the feeling that changing it now would break a lot in DarkPAN6. :P 19:03
FROGGS r: say " " ~~ m:P5/[ ]/ 19:11
camelia rakudo 10b9ac: OUTPUT«===SORRY!=== Error while compiling /tmp/rL87SVXu5v␤failed to parse character class; unescaped ']'?␤at /tmp/rL87SVXu5v:1␤------> say " " ~~ m:P5/[⏏ ]/␤ expecting any of:␤ colon pair (restricted)␤»
FROGGS r: say "1" ~~ m:P5/[1]/ 19:12
camelia rakudo 10b9ac: OUTPUT«｢1｣␤␤»
FROGGS r: say "1" ~~ m:P5/[1 ]/
camelia rakudo 10b9ac: OUTPUT«｢1｣␤␤»
FROGGS r: say " " ~~ m:P5/[1 ]/
camelia rakudo 10b9ac: OUTPUT«False␤»
FROGGS I call it a bug
dalek rlito: c3449a4 | (Simon Bertrang)++ | t5-xs/ (41 files):
Perlito5 - XS - organize tests a bit
 19:19
19:20 donaldh left, berekuk joined 19:21 SamuraiJack left, benabik left
tadzik oh gosh, IO-Socket-INET.t is tricky 19:24
19:27 abnorman left 19:30 rindolf left
dalek p: 10db957 | (Tobias Leich)++ | src/QRegex/P5Regex/Grammar.nqp:
whitespace is literal in P5 regexes
 19:30
kudo/jvm-sockets: e727960 | tadzik++ | / (3 files):
Add a crude IO::Socket(::INET) port to JVM, add to build
 19:32
tadzik hi-ha
FROGGS cool :o) 19:34
tadzik if someone could actually make rakudo-jvm work outside the source directory (and load modules and stuff), that'd be super awesome :) 19:41
I expect that requires some CLASSPATH-fu or such
for now, when testing LWP::Simple I have to copy Test.pm to its lib/, otherwise -Ilib overwrites the path to the regular Test.pm 19:42
PerlJam tadzik: setting PERL6LIB doesn't work? 19:43
Mouq excuse me, but when are submethods useful?
PerlJam Mouq: for BUILD at least. 19:45
moritz when you understand them :-)
19:45 shinobicl_ joined
Mouq Hm. Alright thx :) 19:46
19:46 abnorman joined, awwaiid joined
tadzik PerlJam: hmmm, didn't try :P 19:47
rebuilding now, will try in a sec
19:47 crab2313 left, crab2313 joined
TimToady Mouq: more generally, submethods are in charge of the current class's part of the current object, and not in charge of other parts of the current object. 19:49
submethods are also restricted from (easily) calling virtual attribute names, to prevent accidental interference with other class's attributes before they are initialized (or after they're destroyed) 19:51
you still *can* call virtual methods, but the bare $.foo form is disallowed since it makes it too easy to initialize or finalize out of order 19:52
a submethod should generally only mess directly with $!foo storage
19:52 stevan_ joined 19:54 kbaker joined
Mouq Ahhh, okay; I begin to understand 20:00
TimToady maybe Promise.break should be renamed to Promise.bust 20:04
20:05 stevan_ left
[Coke] only if we rename Promise to A-Move 20:06
dalek kudo/jvm-sockets: fb42021 | tadzik++ | src/vm/jvm/core/IO/Socket.pm:
Unbust recv() and read(), a bit
 20:08
diakopter [Coke]: *groan* 20:09
[Coke]: what about Your-Chops
[Coke] if it's any consolation, that song snippet is now on infiloop in my brain 20:10
TimToady wonders how this mutated into what kind of a bikeshed to put under the paint...
.oO(Bustoleum Paint)
 20:11
diakopter I sorta wonder if the bikeshed should be hurricane and tornado proof, or at least infinitely heavy
TimToady It's really hard to paint an infinitely heavy bikeshed any color other than black... 20:12
[Coke] wonders what color hawking radiation is. 20:13
geekosaur whatever color it wants to be 20:14
TimToady for anything other than a microscopic black hole, the color is still going to be very, very, very dark
dalek p: 68ce715 | (Tobias Leich)++ | src/HLL/sprintf.nqp:
added precision for s and d, also parameter index option
TimToady stellar-sized black holes might emit a photon a year, or less, iirc 20:15
geekosaur ...and a microscopic one you still won;t be able to tell unless you're so far away you only see broad spectrum
diakopter EHEATDEATH 20:17
TimToady well, I bet the radiation is mostly gamma colored
[Coke] ah. My marvel history tells me gamma radiation is green. awesome. 20:18
TimToady glad we got that settled 20:19
diakopter +1up You Gained 1Googol J/K 20:20
TimToady but then, why aren't black lights green too?
diakopter EATTEMPTEDFREEASSOCIATIONWITHNEGATIVEDEGREESOFFREEDOM 20:21
TimToady I hate it when that doesn't happen. 20:22
the nice thing about jet lag is that you can believe impossible things after breakfast as well 20:23
diakopter also, for breakfast.
TimToady free association is much more idealistic than open source association 20:25
diakopter EDETERMINISTICRANDDETECTED
TimToady EOVERUSEOFSNOWCLONEDETECTED 20:26
diakopter ^autopun 20:27
I'm finding it difficult to remove the image of a snowglobe from my mind 20:29
TimToady that's because it's deterministically random
you should have stored it in a variable instead of in a constant 20:30
20:31 bluescreen10 left
diakopter my gc is really lazy 20:31
like, on strike. 20:32
20:37 skids left, BenGoldberg joined 20:39 Ben_Goldberg joined
Ben_Goldberg r: say (^10).infinite, (^Inf).infinite, (^Inf)[^10].infinite; 20:39
camelia rakudo 10b9ac: OUTPUT«FalseTrueNil␤»
20:41 BenGoldberg left
Ben_Goldberg r: ((map { $_ }, ^Inf) but role { method infinite() { True } }).say 20:44
camelia rakudo 10b9ac: OUTPUT«This type does not support elems␤ in method REIFY at src/gen/CORE.setting:7493␤ in method reify at src/gen/CORE.setting:6719␤ in method reify at src/gen/CORE.setting:6706␤ in method reify at src/gen/CORE.setting:6706␤ in method reify at src/gen/CORE.setting:68…
Ben_Goldberg n: ((map { $_ }, ^Inf) but role { method infinite() { True } }).say
camelia niecza v24-88-g1f87209: OUTPUT«(timeout)»
Ben_Goldberg r: ((map { $_ }, ^Inf) but role { method infinite() { False } }).say 20:45
camelia rakudo 10b9ac: OUTPUT«(timeout)»
Ben_Goldberg What's the proper way to tell perl that a particular list is finite or infinite? 20:46
r: ((map { $_ }, ^Inf) but role { method infinite() { Mu } }).say
camelia rakudo 10b9ac: OUTPUT«(timeout)» 20:47
Ben_Goldberg r: ((map { $_ }, ^Inf)).print 20:50
camelia rakudo 10b9ac: OUTPUT«(timeout)0123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121…
Ben_Goldberg r: ((map { $_ }, ^Inf) but role {}).print
camelia rakudo 10b9ac: OUTPUT«This type does not support elems␤ in method REIFY at src/gen/CORE.setting:7493␤ in method reify at src/gen/CORE.setting:6719␤ in method reify at src/gen/CORE.setting:6706␤ in method gimme at src/gen/CORE.setting:7139␤ in method print at src/gen/CORE.setting:12…
Ben_Goldberg n: ((map { $_ }, ^Inf) but role {}).print 20:51
camelia niecza v24-88-g1f87209: OUTPUT«(timeout)»
Ben_Goldberg is confused 20:52
Why does mixing in an empty anonymous role cause rakudo to not like the list?
TimToady something relying on introspection that isn't properly heritable, maybe 20:53
Ben_Goldberg Like testing whether a list's class is exactly equal to some particular class? 20:58
TimToady something inspecting $!foo when it should inspect $.foo, is another way to say it
Ben_Goldberg Oh
TimToady anyway, it just feels like something of that nature, dunno for sure
Ben_Goldberg Perl6's guts might be cleaner than perl5's, but I still don't understand them fully, yet :) 20:59
21:07 donaldh joined, kbaker left 21:19 donaldh left 21:20 EvanTeitelman joined 21:21 bruges_ joined 21:23 vk_ left 21:34 Ben_Goldberg left 21:36 iSlug joined 21:53 BenGoldberg joined
timotimo hi friends. what have i missed all day? 21:54
21:55 spider-mario left 22:01 shinobicl_ left 22:05 donaldh joined
timotimo dunno if anybody answered it yet, but itz: you can make Rakudo::Compiler compile by removing "perl6_group" from the Build.pm and running panda folder_it_is_in/ 22:06
er, i mean panda install folder_it_is_in
22:10 sivoais joined 22:13 berekuk left
timotimo okay, backlogged 22:15
FROGGS++ # sprintf work, finding the commit that broke rakudotjvm 22:16
tadzik++ for work on sockets for the jvm! :D
rest++ for the inverse bikeshed painting %)
22:16 donaldh left, berekuk joined 22:19 crab2313 left 22:26 Mouq left 22:27 rjt_pl joined 22:28 araujo left, araujo joined 22:31 skids joined 22:35 FROGGS joined
jnthn made it home safely from YAPC 22:39
yoleaux 02:15Z <JimmyZ> jnthn: I didn't see which one clones allocate object ...
02:24Z <JimmyZ> jnthn: re: dded to libuv structs probably makes teasing it apart a bit harder, it' won't be harder, just copy uv__fs_* part on *nix and fs__* part on windows to MoarVM
02:24Z <JimmyZ> jnthn: I think I can do it
15:28Z <JimmyZ> jnthn: I re-do libuv in libuv1 branch, without importing(patching) libuv
jnthn Hm, messages ain't sorted by channel :)
22:51 iSlug left 22:52 rjt_pl left 23:00 shinobicl_ joined 23:05 shinobicl_ left 23:13 census joined 23:27 EvanTeitelman left 23:29 berekuk left
[Coke] yoleaux-- 23:30
23:32 berekuk joined
[Coke] do we want to do a star release this month? 23:32
dalek kudo-star-daily: a0b5b81 | coke++ | log/ (5 files):
today (automated commit)
 23:33
23:34 cono joined 23:48 berekuk left 23:50 berekuk joined
[Coke] do we have anyone in the house who understands submodules? 23:51
23:59 cognominal joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!