»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
dalek blets: fbba0e7 | (Herbert Breunung)++ | docs/ (3 files):
more coherent formating of code snippets
 00:02
00:03 vlixes left 00:08 whiteknight joined, japhb_ left
lichtkind good night 00:38
00:38 lichtkind left 00:40 mspaulding left 00:41 pmurias_ left 00:51 buubot_backup joined 00:59 replore left 01:05 replore_ joined 01:07 jaldhar joined
dalek blets: 5042d4d | (Herbert Breunung)++ | docs/appendix-a-index.txt:
typo in double caret op anchor
 01:07
01:15 Quann_Quan joined, Quann_Quan left 01:28 FACEFOX joined 01:37 FACEFOX-DOT-COM joined 01:38 replore_ left 01:42 edwinorc joined
edwinorc Hi, I just ran the perl6 rosetta code for Levenshtein distance ( rosettacode.org/wiki/Levenshtein_distance#Perl_6 ) in rakudo 01:45
But I get the error Cannot assign to a non-container in sub levenshtein_distance at assn1-Q4.p6:6
is that a known thing? 01:46
assn1-Q4.p6 is the name of the file I was running it in 01:47
rakudo: my @a; @a[0][1] = 'X'; say @a.perl; 01:53
p6eval rakudo 45ef4e: OUTPUT«Array.new([Any, "X"])␤»
edwinorc ah maybe it's just my build :)
geekosaur I think that may have just been fixed (but not quite in time for the latest release)? autovivification had been broken for sometime
edwinorc yeah I was just looking at logs from march I think it has been fixed 01:54
just in the last few days
dalek blets: f55c003 | (Herbert Breunung)++ | docs/appendix-a-index.txt:
fixing special var links (have to have same case as item they link to now)
02:05 whiteknight left 02:07 machine1 joined 02:14 PacoAir_ joined, PacoAir left, PacoAir_ is now known as PacoAir
dalek blets: 541c06a | (Herbert Breunung)++ | docs/appendix-a-index.txt:
small format tidy
 02:27
blets: 459e572 | (Herbert Breunung)++ | docs/appendix- (2 files):
more small fixes
blets: d130d6b | (Herbert Breunung)++ | docs/appendix-a-index.txt:
still fixing format typos and stting links
 02:35
02:37 JimmyZ_ joined, Chillance left, JimmyZ_ left 02:46 skids joined 02:51 PacoAir left 02:55 orafu left, orafu joined 02:56 benabik left 02:59 benabik joined 03:06 daemon joined 03:07 daemon left 03:08 daemon joined 03:37 edwinorc left 03:43 tokuhirom joined 03:52 Khisanth joined 04:04 thelazydeveloper left 04:32 nif left 04:48 mucker joined 05:22 kaare_ joined 06:07 GlitchMr joined
moritz \o 06:17
phenny moritz: 20 Apr 23:34Z <lichtkind> tell moritz i rewrote the html generator in perl and it does more now, please let the cron start this
dalek kudo/nom: 498508b | moritz++ | docs/ROADMAP:
[roadmap] :exhaustive is done
06:19 mucker left
moritz phenny: tell lichtkind the cronjob now runs build-html.pl 06:22
phenny moritz: I'll pass that on when lichtkind is around.
06:26 plobsing left 06:37 tokuhirom left 06:39 tokuhiro_ joined
pmichaud good morning, #perl6 06:41
dalek atures: 89540de | pmichaud++ | features.json:
Add autovivification as a Rakudo feature.
 06:44
06:45 tokuhiro_ left
dalek href="https://perl6.org:">perl6.org: e411374 | pmichaud++ | source/compilers/index.html:
Rakudo's setting library is now in the 'nom' branch, not 'master'.
 06:50
07:07 jaffa4 joined, derrazo_ joined
moritz good am, pm 07:14
07:27 plobsing joined 07:44 tokuhiro4 left 07:46 cognominal left 07:50 nwc10 left 07:59 daniel-s_ joined 08:02 daniel-s left 08:03 cognominal joined 08:06 tokuhiro4 joined 08:07 cognominal_ joined 08:10 cognominal left 08:19 agentzh joined
dalek p: 655049e | jonathan++ | src/QRegex/Cursor.nqp:
Make !LITERAL support getting a flag to match cAsE InSenSitiVE.
 08:21
kudo/nom: 5ec18cd | jonathan++ | / (3 files):
Make /:i $foo/ match the interpolated variable case-insensitively.
 08:23
jnthn pmichaud: rt.perl.org/rt3/Ticket/Display.html?id=76500 08:24
moritz b: my $x = 'Foo'; say 'foo' ~~ /:i $x/ 08:27
p6eval b 922500: OUTPUT«␤»
08:31 fsergot joined, nif joined 08:34 nif left
dalek atures: 6af5339 | jonathan++ | features.json:
Now have CORE, SETTING and UNIT.
 08:52
pmichaud jnthn: locally (with icu): 08:55
> my $x = 'XYZ'; say 'xyz' ~~ /:i $x/;
=> <xyz>
jnthn pmichaud: yay :)
pmichaud trying without icu now
jnthn pmichaud: OK. Will tag the ticket testneeded 08:56
pmichaud n: say (1|2|3).Str
p6eval niecza v16-23-gaa61ed5: OUTPUT«any(1, 2, 3)␤»
pmichaud r: say (1|2|3).Str 08:57
p6eval rakudo 498508: OUTPUT«any(1, 2, 3)␤»
pmichaud r: print 1|2|3
p6eval rakudo 498508: OUTPUT«This type cannot unbox to a native string␤ in method print at src/gen/CORE.setting:6666␤ in sub print at src/gen/CORE.setting:6537␤ in block <anon> at /tmp/ZhORcYdOry:1␤␤»
moritz r: say (1|2|3).Str.WHAT
p6eval rakudo 498508: OUTPUT«Junction()␤»
pmichaud n: say (1|2|3).Str.WHAT 08:58
p6eval niecza v16-23-gaa61ed5: OUTPUT«Str()␤»
moritz r: say Mu.Str 08:59
p6eval rakudo 498508: OUTPUT«use of uninitialized value of type Mu in string context in method Str at src/gen/CORE.setting:667␤␤␤»
moritz n: say Mu.Str
p6eval niecza v16-23-gaa61ed5: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/lib/CORE.setting line 1224 (warn @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 235 (Mu.Str @ 10) ␤ at /tmp/ilVnXiP2ya line 1 (mainline @ 2) ␤ at /home/p6eval/niecza/lib/CORE.set…
dalek kudo/nom: 507e7f6 | jonathan++ | src/core/Mu.pm:
Dispatchers should not decontainerize the return value.
 09:02
jnthn pmichaud: auto-viv ticket: rt.perl.org/rt3/Ticket/Display.html?id=84000 09:12
pmichaud: Doesn't seem quite closable yet...
09:13 NamelessTee joined 09:16 jferrero joined 09:20 tokuhirom joined 09:25 plobsing left
moritz r: role A[::X] { method a(::?CLASS:D: ) { }; class B does A[A] { }; B.new.a 09:28
p6eval rakudo 507e7f: OUTPUT«===SORRY!===␤None of the parametric role variants for 'A' matched the arguments supplied.␤Cannot call ''; none of these signatures match:␤␤»
moritz r: role A[::X] { method a(::?CLASS:D: ) { }; class B does A[Int] { }; B.new.a 09:29
p6eval rakudo 507e7f: OUTPUT«===SORRY!===␤None of the parametric role variants for 'A' matched the arguments supplied.␤Cannot call ''; none of these signatures match:␤␤»
moritz r: role A[$] { method a(::?CLASS:D: ) { }; class B does A[Int] { }; B.new.a
p6eval rakudo 507e7f: OUTPUT«===SORRY!===␤None of the parametric role variants for 'A' matched the arguments supplied.␤Cannot call ''; none of these signatures match:␤␤»
moritz erm, what?
jnthn r: role A[$] { method a(::?CLASS:D: ) { }; class B does A[Int] { } 09:32
p6eval rakudo 507e7f: OUTPUT«===SORRY!===␤None of the parametric role variants for 'A' matched the arguments supplied.␤Cannot call ''; none of these signatures match:␤␤»
jnthn oh
missing 09:33
}
moritz r: role A[::X] { method a(::?CLASS:D: ) { }}; class B does A[Int] { }; B.new.a
p6eval rakudo 507e7f: ( no output )
moritz r: role A[::X] { method a(::?CLASS:D: ) { }}; class B does A[Int] { }; B.a
p6eval rakudo 507e7f: OUTPUT«Parameter '' requires an instance, but a type object was passed␤ in method a at /tmp/2XHBWAsUQK:1␤ in block <anon> at /tmp/2XHBWAsUQK:1␤␤»
moritz \o/
dalek kudo/nom: 48c8394 | jonathan++ | src/core/control.pm:
Don't explode when a :lang passed to eval doesn't locate a compiler.
 09:41
10:00 libertyprime joined
moritz r: sub f is rw { [1, 2, 3 ] }; .say for f 10:01
p6eval rakudo 507e7f: OUTPUT«1 2 3␤»
moritz r: sub f { [1, 2, 3 ] }; .say for f
p6eval rakudo 507e7f: OUTPUT«1 2 3␤»
pmichaud > my $x = 'XYZ'; say 'xyz' ~~ /:i $x/ 10:03
=> <xyz>
jnthn: running w/o icu:
> my $x = 'XYZ'; say 'xyz' ~~ /:i $x/
=> <xyz>
jnthn pmichaud++
pmichaud: Works for the non-variable case too?
10:04 JimmyZ joined
dalek ast: 50b3dbc | moritz++ | S05-match/capturing-contexts.t:
untodo a now passing test
 10:04
p: 232518c | pmichaud++ | src/QRegex/Cursor.nqp:
Enable case-insensitive literal interpolation for systems w/o ICU.
 10:10
10:10 infosophy joined
masak is at the hackathon, having a great time 10:13
looking out the Window over Oslo, I feel on top of the world ;)
10:16 whiteknight joined
dalek p: a88d0b8 | jonathan++ | src/6model/reprs/CArray.c:
Serialize/deserialize CArray representation's REPR data. Should fix pre-compilation of things that use CArray based types.
 10:16
10:17 spider-mario joined
jnthn moritz: ^^ should help 10:17
Note you'll need to re-compile
10:22 xenu left, xenu joined
dalek p: b83d81f | pmichaud++ | src/QRegex/Cursor.nqp:
Add the parts of the previous commit that apparently weren't commited.
 10:23
10:25 aindilis joined 10:27 plobsing joined 10:32 birdwindupbird joined
dalek kudo/nom: 1917a7a | jonathan++ | t (2 files):
Set PERL6LIB in the harness, which unbusts running spectest on Windows.
 10:35
10:36 araujo joined, araujo left, araujo joined 10:41 mdcw` joined 10:42 localhost left
mdcw` hello 10:42
moritz hi
dalek kudo/nom: 5511d03 | moritz++ | tools/autounfudge.pl:
autounfudge needs PERL6LIB=lib too
10:43 cotto joined 10:45 NamelessTee left 10:47 localhost joined
dalek p/qbootstrap: d6e4019 | pmichaud++ | Configure.pl:
Add --parrot-option to Configure.pl.
 10:49
p: 05b65e9 | pmichaud++ | Configure.pl:
Add --parrot-option flags to Configure.pl.
10:50 nebuchadnezzar joined 10:54 brrt joined
tadzik dalek didn't report it, but perl6 --doc=HTML etc now DWYM 10:54
jnthn oh
*ooh
tadzik++
moritz tadzik++, dalek-- 10:55
10:56 machine1 left
masak r: say Match.^methods 11:05
p6eval rakudo 48c839: OUTPUT«prematch postmatch caps chunks make Str Numeric Bool ACCEPTS perl gist <anon> <anon> <anon> <anon> <anon> at_key at_pos hash list elems Capture FLATTENABLE_LIST FLATTENABLE_HASH pairs values keys kv gist Str Numeric perl BUILD␤»
11:06 machine1 joined
masak r: "hohohoooo!" ~~ /[<!before '<'> .]+/; say ~$/ 11:06
p6eval rakudo 48c839: OUTPUT«hohohoooo!␤»
imarcusthis waves 11:08
in a slow sine. 11:09
moritz _/-\_ 11:10
my wave has a bit of high frequency components too :-) 11:11
maybe I should just wave after FFT
___|___
there, perfect sine wave :-)
pmichaud ...rod logic? 11:12
imarcusthis pmichaud++
11:12 eiro joined
imarcusthis cloning parrot takes a while. 11:13
11:14 brrt left
jnthn They have a lot of history... 11:14
moritz parrot speaks your language, where "your" changes a lot over time :-)
masak it's a Parrot. it has a lot to say.
imarcusthis masak: I thought they mostly repeat what you say. 11:15
moritz well, they repeat your words, but they can communicate with them 11:16
11:17 harukomoto joined
imarcusthis I see rakudo star is still lacking non-blocking io and parallelism? 11:17
did you consider libdispatch / gcd for threading?
moritz well, the problem with threading is that it needs to come from the VM
jnthn I thought tadzik had some stuff on non-blocking IO
moritz yes, a bit 11:18
imarcusthis jnthn: I was only referring to rakudo.org/2012/02/28/rakudo-star-2...-released/ - things may have changed since february.
well, almost march :)
tadzik imarcusthis: there is tiny bit of infrastructure for async io
imarcusthis tadzik: neat
tadzik imarcusthis: see ttjjss.wordpress.com/2012/04/02/asy...in-perl-6/ for example 11:19
imarcusthis is building nom atm
dalek blets: e145278 | (Herbert Breunung)++ | docs/appendix-a-index.txt:
false set braces in flat()
imarcusthis let's hope my single-fan mbp won't overheat :)
tadzik: thanks for the url
masak .oO( fans of Perl 6 )
jnthn imarcusthis: yeah, I suspect we haven't included the stuff tadzik++ has been doing on that in a nom relesae yet 11:20
tadzik imarcusthis: you can also look at github.com/tadzik/MuEvent
jnthn: you mean Star release?
imarcusthis I mean github branch, btw
jnthn tadzik: yes :)
imarcusthis (assumed nom was master, as it was checked out by default in the clone)
jnthn imarcusthis: yes, it is 11:21
11:21 benabik left, noam_ left, fgomez left, lumi_ left, cosimo left, broquaint left 11:22 lichtkind joined
lichtkind moritz: HAI 11:24
phenny lichtkind: 06:22Z <moritz> tell lichtkind the cronjob now runs build-html.pl
lichtkind moritz: thanks
11:28 benabik joined, noam_ joined, fgomez joined, lumi_ joined, cosimo joined, broquaint joined
masak nom: say Cool.can("comb").^name 11:29
p6eval rakudo 626ef2: OUTPUT«Parcel␤»
masak nom: say Cool.can("comb")[0].^name
p6eval rakudo 626ef2: OUTPUT«Method␤»
masak nom: say Cool.can("comb")[0].signature
p6eval rakudo 626ef2: OUTPUT«:(Cool , Any , Mu %_!)␤»
masak well, thanks for nuttin' :P
masak goes looking at the source 11:30
11:30 derrazo_ left
dalek kudo/nom: 271d219 | jonathan++ | src/core/Pair.pm:
Correct a bogus contextualization.
 11:33
kudo/nom: 94ae635 | jonathan++ | src/core/EnumMap.pm:
%h.hash should not make a new hash, just return identify.
kudo/nom: fa1ca28 | jonathan++ | src/core/Mu.pm:
The identify .item should not decontainerize.
kudo/nom: 6ef66c9 | jonathan++ | src/Perl6/Actions.pm:
$.foo, @.foo and %.foo should contextualize appropriately.
blets: 5168fd8 | (Herbert Breunung)++ | docs/appendix- (2 files):
added . as prefix for nonecapturing subrules
 11:34
11:39 NamelessTee joined
tadzik phenny: ask sorear any ideas why dalek is not reporting my rakudo commits? 11:58
phenny tadzik: I'll pass that on when sorear is around.
masak r: "foo" ~~ /foo/; say $/.text 12:00
p6eval rakudo 626ef2: OUTPUT«Method 'text' not found for invocant of class 'Match'␤ in block <anon> at /tmp/u18UxXtvTe:1␤␤»
masak r: grammar G { regex TOP { { "foo" ~~ /foo/; say $/.text } } }; G.parse("")
p6eval rakudo 626ef2: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
masak moritz: ^
imarcusthis ponders improving perl6 --help 12:08
tadzik well volunteered! 12:09
imarcusthis seems to me it should emit something more similar to man perl6
tadzik: indeed, I even made a branch
tadzik awesome
moritz r: 'a' ~~ /a/; $/.text 12:12
p6eval rakudo 626ef2: OUTPUT«Method 'text' not found for invocant of class 'Match'␤ in block <anon> at /tmp/0s7Be9x65G:1␤␤»
moritz r: "foo" ~~ /foo/; say $/.text
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
moritz r: "foo" ~~ /foo/; $/.text
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
12:13 fgomez left
moritz huh. Why does it need "foo" to fail at erroring? 12:13
jnthn r: "bar" ~~ /bar/; say $/.text 12:14
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
jnthn A bar will do too
masak wtf 12:15
r: "bar" ~~ /bar/; $/.text
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
moritz r: 'ab' ~~ /ab/; $/.text
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
moritz r: 'ab' ~~ /a/; $/.text 12:16
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
moritz r: 'a' ~~ /a/; $/.text
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
moritz oh, it recompiled rakudo inbetween
masak r: my $s = "it escapes things like & for you! &, &, &!"; $s ~~ s:g['&'] = '&amp;'; say $s
p6eval rakudo 6ef66c: OUTPUT«it escapes things like &amp; for you! &amp;, &amp;, &amp;!␤»
masak that's just *so cool*.
I can't really get over how cool that is.
moritz my first execution used rakudo 626ef2, the later ones 6ef66c 12:17
masak moritz++
sjn pmichaud: I've dumped some of my thoughts on yesterday's crazy discussion on code.foo.no/ :)
12:24 plobsing left
imarcusthis has been hacking on perl6 for 10 minutes, and is already ready for a drink. 12:26
no heredocs? wat
tadzik not yet :)
imarcusthis what about __DATA__ ? 12:27
JimmyZ no 12:28
imarcusthis how do you output multiple lines of text then? a bunch of say lines? :)
tadzik or just stick \n in a string :)
maybe just 'say "' <newline newline newline> '";' 12:29
imarcusthis starting to$!usage := $!usage ~ " $_\n"; 12:30
tadzik naah, don't
imarcusthis is this because you don't have anything like $foo .= ?
tadzik imarcusthis: how about just gist.github.com/2436884 ?
imarcusthis: there's ~= for that
imarcusthis tadzik: that's great actually
JimmyZ .= ? it's not php 12:31
tadzik it's not Perl 5 either :)
12:31 Teratogen joined
imarcusthis JimmyZ: perl -lE'$_="foo";$_ .=" bar"; say $_' 12:31
JimmyZ did not know perl much ..... 12:33
perl 5
imarcusthis Alas, poor Yorick. 12:34
btw, how come this doesn't throw an error: "perl6 -Ilib example.pl" 12:35
it doesn't seem to modify the include path either 12:36
tadzik imarcusthis: it's just that -I is NYI
imarcusthis: you can use PERL6LIB=lib perl6 example.pl
should be quite implementable, I think 12:37
imarcusthis tadzik: just -I throws an error about "No such option"
tadzik correct
there is no such option :)
imarcusthis tadzik: so why doesn't -Ilib throw the same error_
anyways, MuEvent runs here now. yay 12:39
It does seem like it is bored too :)
dalek kudo/nom: 7c97944 | jonathan++ | src/Perl6/ (2 files):
Fix combination of state declarator and list assignment.
 12:42
tadzik imarcusthis: hmm, no eye deer :)
I guess -Ilib doesn't play well with HLL::Compiler, which accepts either -I or --Ilib
imarcusthis a blind deer!
tadzik that's why it's different 12:43
it should be quite possible to implement -I=lib, if we have a way to setenv() from nqp
and then it probably won't be windows-friendly again :)
imarcusthis gist.github.com/2436933 12:44
masak r: my $value = "foo"; my $string = "I have a foo under my bed, foo!"; $string ~~ s:g/<$value>/ = "bar"; say $string 12:45
p6eval rakudo 6ef66c: OUTPUT«===SORRY!===␤Preceding context expects a term, but found infix = instead at line 1, near " \"bar\"; sa"␤»
masak r: my $value = "foo"; my $string = "I have a foo under my bed, foo!"; $string ~~ s:g[<$value>] = "bar"; say $string
p6eval rakudo 6ef66c: OUTPUT«I have a bar under my bed, bar!␤»
masak \o/
JimmyZ how to explain it? 12:47
imarcusthis JimmyZ: masak has a bar under his bed.
JimmyZ hmm, How it works 12:48
tadzik imarcusthis: cool! 12:50
imarcusthis: however, that's nqp, not perl6, right?
(the file)
JimmyZ r: my $value = "foo"; $_ = "I have a foo under my bed, foo!"; s:g[<$value>] = "bar"; say $_
p6eval rakudo 6ef66c: OUTPUT«I have a bar under my bed, bar!␤»
imarcusthis tadzik: yes. It was the only place I found the output in the perl6 source dir, so I assumed it got it from nqp 12:51
tadzik hm 12:52
imarcusthis: let me take a look at this
imarcusthis: okay, I got this. In nqp, see src/HLL/Compiler.pm, line 299 12:53
it says: self.usage($program-name) if %adverbs<help> || %adverbs<h>;
so what you need to do is to redefine method usage() in Perl6/Compiler.nqp (I think) 12:54
imarcusthis tadzik: yeah, that is the file I'm working on.
12:54 thelazydeveloper joined
dalek kudo/nom: 754e4a0 | moritz++ | src/core/Backtrace.pm:
succeed in creating error message much more often
 12:54
imarcusthis tadzik: can't they just use the same text?
tadzik imarcusthis: well, now you made everything that uses HLL::Compiler print help for Rakudo Perl 6
12:55 fglock joined
imarcusthis tadzik: I see. 12:55
tadzik and what you really want is just to make Rakudo act differently than the default
moritz note that you can add a method to src/Perl6/Compiler.nqp
which inherits from HLL::Compiler 12:56
imarcusthis yupp
I'm doing that now
tadzik great, imarcusthis++
12:56 Patterner left
masak r: my $s = "siht si yrev gnitseretni"; $s ~~ s:g[(\w+)] = (my $c = $0; $c.flip.uc); say $s 12:57
p6eval rakudo 6ef66c: OUTPUT«THIS IS VERY INTERESTING␤»
masak TimToady: BEGIN inside a quasi -- runs at quasi-parse, or at quasi-apply? 13:02
TimToady: there are arguments for either. quasi-parse is "as early as possible". quasi-apply is where the BEGIN block would fire if we assume that the quasi isn't "concrete code" until it gets applied in some actual code. 13:03
13:04 Psyche^ joined, Psyche^ is now known as Patterner
tadzik r: %*ENV<FOO> = "bar"; shell 'echo foo $FOO' 13:06
p6eval rakudo 6ef66c: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
tadzik evalbot rebuild nom
p6eval OK (started asynchronously)
masak evalbot rebuild my entire secret robot army
13:07 sftp left
moritz tadzik: it says it does that, but it doesn't actually :( 13:08
tadzik I see 13:09
Teratogen good morning
masak Teratogen! \o/ 13:10
Teratogen masak!
masak we're in Oslo, doing a hackathon! 13:13
tadzik \o/ 13:14
if anyone's looking for an LHF: github.com/tadzik/perl6-File-Tools...and.pm#L25 13:15
as in: implementing rm_rf would be nice and easy and helpful to me :)
Teratogen I hear Norwegian chicks are hawt!
13:15 lichtkind_ joined
tadzik I see no chicks :( 13:16
masak Teratogen: we're currently discussing whether to kickban you. please behave.
Teratogen now what did I do?
13:18 lichtkind left, lichtkind_ is now known as lichtkind
imarcusthis Teratogen: you implied perl6 hackers care about girls. (ick) 13:19
>D
13:19 bjarneh joined
tadzik well, I think what masak meant is that what you say may offend some women, which we would really not want 13:19
(correct me if I'm wrong) 13:20
imarcusthis Isn't that a compliment to norwegian women? 13:23
moritz some women might take it as such; others might take it as reducing their value to sexuality 13:24
imarcusthis has realized he has to be a feminist, with a second daughter on the way.
tadzik imarcusthis: I would read it as such, but it's not me who gets offended eventually
masak r: sub foo($t) { $t ~~ s:g[foo] = "bar" }; grammar G { regex TOP { { my $s = "OH HAI, foo!"; foo($s) } } }; G.parse("") 13:26
p6eval rakudo 754e4a: OUTPUT«Error while creating error string: Method 'message' not found for invocant of class 'Any'␤»
masak moritz: ^
13:26 plobsing joined
masak Teratogen's utterance was offensive against Norwegian poultry. 13:28
13:28 isBEKaml joined
isBEKaml Hello, #perl6! What's going on Hakkathon? :) 13:30
masak isBEKaml! \o/ 13:31
isBEKaml yo, masak! you at hackathon too?
masak aye
imarcusthis masak: yes. our poultry is very cold, and lacks creature comforts.
jnthn Wait, what, now I'm confused...is Norwegian poultry hot or cold? 13:32
masak imarcusthis: I think this is a problem in many places.
jnthn
.oO( Guess it depends if you cook it )
isBEKaml jnthn: I think you're in for a frigid meal. ;)
dalek kudo/nom: 096cf7d | (Marcus Ramberg)++ | src/Perl6/Compiler.nqp:
Better help for perl6 binary
 13:33
kudo/nom: b1f5835 | moritz++ | src/Perl6/Compiler.nqp:
Merge pull request #61 from marcusramberg/improve_help

Better help for perl6 binary
13:37 NamelessTee left 13:39 BigBear joined
TimToady masak: maybe means we need two phasers there 13:49
masak nnnnnnnaaaaarrrghh
jnthn QBEGIN
pmichaud QBEGIN! 13:50
moritz QBEGIN!
masak explodes
TimToady has admittedly had no coffee yet
pmichaud hacking has now ceased in Oslo due to uncontrollable fits of laughter.
and also concern that TimToady++ might be serious :)
moritz TimToady: I had the same answer to masak++ when he asked his question in RL :-) 13:51
TimToady <-- glad to be of entertainment value at least
masak I think that QBEGIN is a bad idea until we have use cases for both possible semantics. 13:52
13:53 icwiener joined
masak until then, BEGIN will mean whatever is most convenient for whoever implements macros in Rakudo ;) 13:53
TimToady in a sense {{{}}} is running QBEGINishly already
13:53 grondilu joined
TimToady goes to get caffiene to see whether that's a stupid notion 13:54
dalek kudo/nom: 501e401 | jonathan++ | src/Perl6/ (2 files):
Fix constant initializers that include a block.
p: b6a9303 | pmichaud++ | / (3 files):
Bump PARROT_REVISION to get latest Parrot upcase/downcase improvements for non-ICU environments.
 13:56
moritz hugme: add torehaug to mberends
hugme moritz: You need to register with freenode first
moritz hugme: add torehaug to MiniDBI 13:57
hugme hugs torehaug. Welcome to MiniDBI!
13:59 whiteknight left 14:01 isBEKaml left, isBEKaml joined
imarcusthis tadzik: I have another improvement suggestion, for --doc - pastie.textmate.org/private/tcklbs6...2at8u9yqsa 14:02
14:02 PacoAir joined
imarcusthis tadzik: unless it's crack-fueled, could you point me towards a good place to start looking at implementing this? 14:02
tadzik imarcusthis: looks cool 14:03
pmichaud ...is that valid Perl 6?
tadzik pmichaud: what do you mean? 14:04
pmichaud the use of the colons to denote docs (I think)
14:05 icwiener left
imarcusthis pmichaud: you mean #= ? 14:05
14:05 icwiener joined
tadzik imarcusthis: look at github.com/rakudo/rakudo/blob/nom/...ext.pm#L73 14:05
imarcusthis: add a few \n there and you're done :) 14:06
pmichaud nm... I was looking at the output and thinking it was source
tadzik++ # for setting me straight
tadzik imarcusthis: you may need to do s/\s/\n/ in $_.signature somewhat
pmichaud and yes, it looks very cool
jnthn cd ..
oop
tadzik (that's where the subroutine parameters stringification comes from)
imarcusthis tadzik: great, thanks. :)
tadzik imarcusthis: I really like it, it makes output much more readable
14:07 NamelessTee joined
tadzik ohh, you mean to remove Any as well? 14:07
and a bit more mangling as I see
imarcusthis yes
moritz the signature printing improvements would need to go into src/core/Paremeter.pm method perl 14:08
tadzik can't say I disapprove :) although you may need to reimplement .signature more than just regex-modify it after that
indeed
imarcusthis remove Any, remove the ':', rewrite :foo(:$foo)' as :$foo
dalek kudo/nom: 59b31b6 | pmichaud++ | tools/build/NQP_REVISION:
Bump NQP_REVISION to get latest case-insensitive improvements for non-icu environments.
moritz and other parts (error messages, for example) would benefit from an improved Parameter.perl too
imarcusthis cool 14:09
tadzik imarcusthis: erm, remove which ':'? The very leading one?>
imarcusthis tadzik: no, the one I replace with \n
tadzik oh, sure
but that's just Pod::To::Text, no Signature.perl itself 14:10
but it's a good idea indeed
dalek p: 7cd7557 | pmichaud++ | tools/build/PARROT_REVISION:
Revert PARROT_REVISION to released version; I was a little too eager with the previous bump.

We can bump PARROT_REVISION later (e.g., after the R* release). Using the older PARROT_REVISION means that non-ICU users won't get the Parrot improvements, but we'll still be able to make releases based on latest Parrot.		 14:13
14:14 harukomoto left, isBEKaml left
dalek p: ac3e430 | jonathan++ | src/6model/reprs/CArray.c:
Need to be lazier about calling fill_repr_data in CArray; last patch made things a bit too eager.
 14:19
tadzik generated using 2 small scripts: tjs.azalayah.net/new.html 14:21
masak: you may happen to like this :)
jnthn сырsly awesome! tadzik++ 14:22
tadzik :)
masak tadzik++
14:22 Chillance joined
dalek pan style="color: #395be5">perl6-examples: ab8c861 | (Filip Sergot)++ | best-of-rosettacode/hailstone-sequence.pl:
best-of-rosettacode/hailstone-sequence.pl
 14:26
pan style="color: #395be5">perl6-examples: bef4145 | (Filip Sergot)++ | best-of-rosettacode/hailstone-sequence.pl:
best-of-rosettacode/hailstone-sequence.pl
 14:30
14:33 grondilu left
pmichaud I'm wondering if we should make --stagestats a permanent part of the core.setting compilation step 14:33
It would help to show progress as that (very long) step is taking place 14:34
imarcusthis pmichaud: sounds useful to me. 14:35
either that or make it really fast ;-) 14:36
pmichaud well, the second would be better, yes. But it's taking us a long time to make it really fast. :)
jnthn It's not really fast now, but it's faster than it was. :) 14:37
I'm curious about the proportions.
imarcusthis has been waiting for it to complete for this entire conversation =)
jnthn (how much time we spend in each stage)
pmichaud most of it is still in "parse" (which is really "parse" + "create ast" + "BEGIN" + ... ) 14:38
jnthn pmichaud: Well, a lot of it was in the later stages, before we got bs. 14:39
(Because we used to make twice as many PAST nodes and thus POST nodes)
dalek kudo/nom: 0d3220f | pmichaud++ | tools/build/Makefile.in:
Make --stagestats a permanent part of the setting compilation, to provide a progress indicator of sorts.
p: 09bf632 | jonathan++ | src/6model/reprs/CStruct.c:
Do just enough to make sure serializing modules that declare something with the CStruct REPR should not go awry.
 14:40
14:40 icwiener left 14:41 icwiener joined, mucker joined
mucker o/ #perl6 14:41
tadzik hello
dalek kudo/nom: 06d40a0 | moritz++ | src/core/ (2 files):
Make backtrace printing more robust
 14:42
icwiener Hi.
JimmyZ aloha 14:43
masak r: my $m; $m max= 42 => 5
p6eval rakudo 501e40: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \$v, Mu %_!)␤␤ in method Numeric at src/gen/CORE.setting:648␤ in sub infix:<==> at src/gen/CORE.setting:2384␤ in sub infix:<cmp> at src/gen/CORE.setting:1256␤ in method max at src/gen/CORE.setting:10…
fsergot Is there any solution masak? :) 14:44
masak r: +(42 => 5) 14:45
p6eval rakudo 501e40: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \$v, Mu %_!)␤␤ in method Numeric at src/gen/CORE.setting:648␤ in sub prefix:<+> at src/gen/CORE.setting:2192␤ in block <anon> at /tmp/7uPTpOZ4vF:1␤␤»
moritz r: my $m = (-Inf) => (-Inf); $m max= 42 => 5; say $m
p6eval rakudo 501e40: OUTPUT«42 => 5␤»
masak oh, it's even simpler.
moritz: right, there it doesn't numify :)
but... why would max= numify? 14:46
moritz yes. "Doctor, it hurts when I numify my pairs" :-)
masak yes.
fsergot++
masak submits rakudobug
icwiener I want to include a small test file into one of my test cases. I found "Q :to 'EOT'" for that but when I try to run the test, it complains that "Colons may not be used to delimit quoting constructs". Is the info on www.perlfoundation.org/perl6/index....t#heredocs wrong or do I miss something here?
moritz icwiener: heredocs simply aren't implemented in Rakudo 14:47
pmichaud r: my $m; say $m max (42 => 5)
p6eval rakudo 501e40: OUTPUT«42 => 5␤»
moritz icwiener: it tries to parse it as Q<delimiter>...<delimiter>, and then complains that : isn't a valid <delimiter> 14:48
icwiener moritz: Oh, that explains it. :)
pmichaud r: my $m; $m max= (42 => 5)
p6eval rakudo 501e40: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \$v, Mu %_!)␤␤ in method Numeric at src/gen/CORE.setting:648␤ in sub infix:<==> at src/gen/CORE.setting:2384␤ in sub infix:<cmp> at src/gen/CORE.setting:1256␤ in method max at src/gen/CORE.setting:10…
icwiener icwiener: Yep, makes sense.
icwiener: So I guess I fill in the content line-wise, then? 14:49
imarcusthis wow, make spits out a lot of warnings now :)
pmichaud you can always do Q{ }
imarcusthis feels much faster!
pmichaud or qq{ }
icwiener pmichaud: reading...
pmichaud r: say qq{{ this is a string with curlies } that can span multiple lines }} 14:50
p6eval rakudo 501e40: OUTPUT« this is a string with curlies } that can span multiple lines ␤»
dalek pan style="color: #395be5">perl6-examples: 7f7a7c5 | (Filip Sergot)++ | best-of-rosettacode/hailstone-sequence.pl:
hailstone-sequence.pl works.
 14:51
icwiener pmichaud: But I have to account for stuff that looks like variable names, right? 14:59
pmichaud if you use qq, yes.
but you can use q{{ }} if you want to avoid that
14:59 nif joined
pmichaud r: say q{{ This has $var names in it }} 14:59
p6eval rakudo 501e40: OUTPUT« This has $var names in it ␤»
JimmyZ how Perlito/src6 is executed?
by using rakudo?
masak I believe it's bootstrapping. 15:00
icwiener pmichaud: Cool, thanks. :)
JimmyZ I don't know how to run Perlito/src6 .. 15:01
15:05 nif left
dalek kudo/nom: c89ce89 | moritz++ | src/core/Backtrace.pm:
finally fix masak++'s example of failed error reporting

also use .DEFINITE instead of .defined		 15:07
15:08 nif joined 15:09 fsergot left
tadzik r: "Everything's fine \o/" 15:09
p6eval rakudo 501e40: OUTPUT«===SORRY!===␤Unrecognized backslash sequence: '\o' at line 1, near "/\""␤»
tadzik hehe
15:09 nif left 15:10 nif joined
moritz r: say -Inf max 'foo' 15:12
p6eval rakudo 501e40: OUTPUT«foo␤»
moritz r: say -Inf max 'ZZZ'
p6eval rakudo 501e40: OUTPUT«ZZZ␤»
moritz r: say -Inf max 5 => 5
p6eval rakudo 501e40: OUTPUT«5 => 5␤»
masak r: say (-Inf) but (+Inf) # hee hee
p6eval rakudo 501e40: OUTPUT«-Inf␤»
moritz masak: that just installs a .Num method that return +Inf, or so 15:13
15:13 agentzh left
moritz r: say ((-Inf) but Inf).Num 15:13
p6eval rakudo 501e40: OUTPUT«Inf␤»
pmichaud r: say -Inf cmp 42 => 5;
infosophy what is an idiomatic way of creating a temporary file for testing file statistics? (modification time etc.)
p6eval rakudo 501e40: OUTPUT«-1 => 5␤»
pmichaud r: say -Inf cmp (42 => 5);
p6eval rakudo 501e40: OUTPUT«-1␤»
masak infosophy: moritz will know more, but... 15:14
infosophy: I think there are a few cases like this in the spectests already.
pmichaud say &infix:<cmp>(-Inf, (42=>5))
masak there should be a bit of prior art.
pmichaud r: say &infix:<cmp>(-Inf, (42=>5))
p6eval rakudo 501e40: OUTPUT«-1␤»
15:14 nif left
moritz infosophy: currently we just attach $*PID to a temp file name which is specific to that test file 15:14
pmichaud say (&infix:<max>(), (42=>5)).max
r: say (&infix:<max>(), (42=>5)).max
p6eval rakudo 501e40: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \$v, Mu %_!)␤␤ in method Numeric at src/gen/CORE.setting:648␤ in sub infix:<==> at src/gen/CORE.setting:2384␤ in sub infix:<cmp> at src/gen/CORE.setting:1256␤ in method max at src/gen/CORE.setting:10…
pmichaud r: say &infix:<max>() 15:15
p6eval rakudo 501e40: OUTPUT«-Inf␤»
moritz ie a t/03-stats.t would create a temp file 't-03-stats-tmp-' ~ $*PID ~ '.temp' or so
wait, we call them temp-03-stats-$*PID.temp
but the name doesn't matter much
infosophy moritz: gotcha. I'll look for it in the spectests.
moritz it's just a good idea to have some hint about which test file it creates 15:16
so if they are left behind, you know where to investigate
infosophy yup. I'll go by the prior art to see where the files are placed (and what cleanup is performed) 15:17
moritz the cleanup is just try unlink $temp-filename;
pmichaud r: say &infix:<cmp>(-Inf, (42 => 5)) 15:22
p6eval rakudo c89ce8: OUTPUT«-1␤»
pmichaud r: say &infix:<cmp>((42 => 5), -Inf)
p6eval rakudo c89ce8: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \$v, Mu %_!)␤␤ in method Numeric at src/gen/CORE.setting:648␤ in sub infix:<==> at src/gen/CORE.setting:2384␤ in sub infix:<cmp> at src/gen/CORE.setting:1256␤ in block <anon> at /tmp/cljmG63PB1:1␤␤»…
imarcusthis re doc, this is what I have now : gist.github.com/2437709
mucker JimmyZ: I think, this is how bootstrapping works. First you use rakudo to compile src6. But within src6 itself there is a translater for perl6 to perl5. So after the initial rakudo bit, we no longer need rakudo. After getting the first compiler using rakudo, I beleive, fglock generated used it to generate the perl6 to perl5 compiler. It's get's really hairy after that because if new perl5 compiler can be used to parse perl6 too ! You can see the util-perl5 15:23
directory to see the bootstraping procedure
tadzik imarcusthis: awesome!
moritz imarcusthis++
there's still a tiny problem
tadzik imarcusthis: not doing \n inside parameter list after all?
imarcusthis not sure how I can reformat the whitespace in the method signatures, maybe just use a subst in the pod.
moritz Callable &cb actually means a Callable (implied by the &) which returns a Callable
tadzik imarcusthis: well, how did you modify them?
15:24 sergot joined
fglock hi 15:24
phenny fglock: 20 Apr 23:51Z <japhb_> tell fglock rc-forest-fire now works with perlito5.pl, and quite quickly too. However, for perlito5.js, it gets through the compile but dies at runtime (after printing the starting map) with this error: Warning: TypeError: Object [object Object] has no method 'next'
fglock: 20 Apr 23:54Z <japhb_> tell fglock And you might have thought it succeeded too fast to see because you supplied no arguments telling it forest size and iteration count. The tested command is: node perlito5.js -Isrc5/lib /path/to/perl6-bench/perl5/rc-forest-fire 30 20 100
fglock: 20 Apr 23:55Z <japhb_> tell fglock er ... "30 20 10"
imarcusthis moritz: you mean it should drop Callable as well because it's already implied by '&'?
tadzik: I'm not modifying the list yet, just the individual elements. 15:25
mucker fglock :)
moritz imarcusthis yes
imarcusthis: as well as 'Positional' for @ and 'Assocative' for %
15:26 xinming_ left
dalek p/toqast: 29da48d | jonathan++ | / (3 files):
Stub in QAST::Block and QAST::Stmts.
 15:27
fglock re bootstrap, the initial bootstrap was done by translating p6 to p5 by hand
tadzik imarcusthis: I see
imarcusthis: so digging in Parameter.pm, right?
imarcusthis tadzik: yes 15:28
tadzik cool
fglock: Knuth style ;)
mucker fglock: wicked
masak github.com/petdance/vim-perl 15:29
dalek pan style="color: #395be5">perl6-examples: 80e1d29 | (Filip Sergot)++ | other/combinations.pl:
masak++ \'s combinations solution.
fglock the original file was called iterator_engine, there is a copy in feather at /home/fglock/iterator_engine 15:31
there is a long README and a TODO list :) 15:32
dalek pan style="color: #395be5">perl6-examples: 9a1be5b | (Filip Sergot)++ | README:
Added other dir.
 15:33
fglock "2006-02-26: - working Perl 6 compiler - about 100 lines of code are now written in perl6" - that's 6 years ago 15:34
imarcusthis re hot norwegian chicks - www.youtube.com/watch?v=d-N9daqANcw...AAAAAAAAAg 15:36
fglock re running src6 - README-perlito6 has instructions on how to transform src6 into runnable code (p5, python, or js) 15:37
src6 should eventually be runnable by rakudo or niecza, I think 15:38
jaffa4 fglock: do you accept some advice? 15:40
fglock yes, please 15:41
re no method 'next': probably because next LABEL is not implemented yet in js
tadzik new, more descriptive contents on tjs.azalayah.net/new.html Does anyone have any ideas which boxes should end up being yellow rather than red? 15:46
15:48 Psyche^ joined 15:51 Patterner left, Psyche^ is now known as Patterner
fglock jaffa4: what is the advice? 15:51
jaffa4 private window?
imarcusthis Hmm, seems there is no $SIG_ELEM_CALLABLE_SIGIL 15:52
jaffa4 fglock: can you see what?
what -> that 15:53
fglock no, let me login using irssi 15:54
15:55 fglock left, fglock_ joined
imarcusthis moritz: seems you cannot check for '&' sigil in Parameter.pm at the moment. 15:55
dalek osystem: cc1b79f | (Filip Sergot)++ | META.list:
Text::Markdown
 15:56
moritz imarcusthis: wow, indeed
15:56 zhangkaizhao joined
imarcusthis moritz: guess it would have to be added to World.pm 15:56
15:56 zhangkaizhao left
moritz imarcusthis: maybe you can use self.name.substr(0, 1) eq '&' 15:57
15:57 zhangkaizhao joined
moritz imarcusthis: though it's something we should discuss with jnthn++ at some point 15:57
15:57 fglock_ left, icwiener left
moritz it feels oddly asymmetric, but maybe there's a good reason ($, % and @ muck with itemness, & does not) 15:57
15:58 zhangkaizhao left, fglock joined
fglock jaffa4: can you try again 15:58
15:59 zhangkaizhao joined
dalek pan style="color: #395be5">perl6-examples: bd9bf92 | (Salve J. Nilsen)++ | / (4 files):
Two new niecza-only examples from fsergot++
 16:07
pan style="color: #395be5">perl6-examples: af51ac7 | (Salve J. Nilsen)++ | / (3 files):
Merge remote-tracking branch 'origin/master'
16:20 machine1 left 16:23 jaffa4 left 16:26 machine1 joined 16:27 sirhc joined 16:28 tokuhirom left, tokuhirom joined 16:33 tokuhirom left
krunen Pizza is served! 16:34
tadzik \o/ 16:35
cognominal_ bon apétit 16:36
G grand. a petit # pun 16:37
16:40 vlixes joined
cognominal_ what is a OwnedHash? 16:42
16:43 birdwindupbird left 16:47 icwiener joined 16:48 fod joined 16:52 JimmyZ left 16:53 fod left
tadzik haha 16:54
3rd time this question is asked :)
irclog.perlgeek.de/perl6/2012-04-20#i_5470326
cognominal_: ^
cognominal_ sorry, I failed to backlog even if I watch the patch comming by 16:55
tadzik seen colomon?
aloha colomon was last seen in #perl6 1 days ago saying "Ulti: I think so. (Not my area, but if no one else is answering...)".
cognominal_ tadzik++ 16:56
tadzik phenny: tell colomon Math::Polynomial hangs on one of its tests
phenny tadzik: I'll pass that on when colomon is around.
colomon ?
phenny colomon: 16:56Z <tadzik> tell colomon Math::Polynomial hangs on one of its tests
tadzik hah
colomon which compiler? 16:57
masak .oO( Yapsi )
tadzik colomon: Rakudo 16:58
colomon k 16:59
masak pizza++ 17:01
17:02 icwiener left, icwiener joined 17:03 icwiener left, icwiener joined 17:09 jaffa4 joined 17:10 brrt joined
pmichaud > my $m; $m max= 42 => 5; say $m 17:12
42 => 5
masak: ^^ 17:13
moritz \o/
masak \o/ 17:15
jnthn ETOOMUCHPIZZA 17:19
jnthn feels like a leaning tower of pizza now
jaffa4 too much pizza no good 17:20
jnthn oh, it was good ;)
jaffa4 to your health
jnthn I'll live :) 17:21
.oO( famous last wrods )
moritz you'll die, like all of us :-)
tadzik PerlJam: in Grammar-Profiler-Simple META.info source-url is broken
jaffa4 with stomach burn
jnthn moritz: :P
cognominal_ jnthn will die, but thanks to moritz with a clean stacktrace. 17:23
masak X::IO::Pizza
moritz masak: I hope it's only X::I::Pizza :-) 17:24
masak it's IO even when you're only doing I :P
dalek ast: d9d5118 | pmichaud++ | S32-list/sort.t:
[S32]: Re-fudge some sort tests with latest infix:<cmp> fixes.
kudo/nom: 120e95d | pmichaud++ | src/core/Any.pm:
Fix &infix:<cmp>(Any,Any) to use value identity for Inf/-Inf instead of numeric identity.
jnthn r: say 647 - 83 17:25
p6eval rakudo c89ce8: OUTPUT«564␤»
masak "QBEGIN" reminded me of en.wikipedia.org/wiki/Q*bert 17:27
tadzik seen supernovus? 17:28
aloha supernovus was last seen in #perl6 80 days 17 hours ago saying "No code link yet. I just finished abandoning my Garden template language, which was a monstrosity. So, I'm starting from scratch for this project (currently using the very unoriginal name of Template6)".
masak seen Memento? 17:30
aloha Sorry, I haven't seen Memento.
tadzik aloha: shame on you, it's a cool movie
masak .oO( but we saw it yesterday! )
moritz sjn: got the confirmation mail. Thanks! 17:31
jaffa4 I have not seen Memonto either 17:32
masak jaffa4: yes, but... have you seen Memento? 17:34
jaffa4 same answer I did not spell it correctly
masak oh, I didn't even see the typo :)
17:36 sftp joined
sjn moritz: yay! 17:41
jaffa4 how to upload stuff to panda? 17:46
17:46 bjarneh left
tadzik jaffa4: you mean the module repository? 17:47
jaffa4 exactly
tadzik oh, we have docs for that. Hold on...
wiki.perl6.org/Create%20and%20Distr...%20Modules
moritz they are hopefully linked on modules.perl6.org :-)
tadzik jaffa4: basically, you need a META.info, a git repo, and someone with write access to perl6/ecosystem
jaffa4 How do I know if I have access to it? 17:51
tadzik jaffa4: github.com/perl6/ecosystem where it says HTTP and Git-read-only, does it also say SSH?
if it does, then you have it
imarcusthis: I think there's a bug in your branch
sub _block1685(): Event-driven programming in Perl 6
I think it used to say "module MuEvent: bla bla bla"
infosophy am I using the wrong syntax for chmod? chmod 0o000, 'temp-16-filehandles-filestat-39884.temp' 17:52
imarcusthis tadzik: sweet :)
tadzik oh wait
infosophy Undefined routine '&chmod' called (line 1)
imarcusthis tadzik: it does say module Muevent here
still
tadzik it looks like I'm still using the old branch or something
moritz infosophy: currently only available as a method
tadzik must be wrong on mi side, hold on
infosophy On IO?
moritz infosophy: so $filename.IO.chmod(0o000)
yes
infosophy thx
masak right. I only see the method form. 17:53
tadzik argh. I did git fetch, but not git checkout :)
imarcusthis :D
masak maybe the sub form should exist by spec, but I don't see it in Rakudo.
17:53 pernatiy joined
infosophy got fooled by S16-filehandles/chmod.t 17:53
tadzik anyway, let me review it on your machine :)
frettled got fooled as well
moritz infosophy: we don't run that test file at the moment
frettled …which is not quite so much to brag about.
masak infosophy: t/spectest.data contains the files Rakudo runs. 17:54
jaffa4 tadzik: ok
krunen sjn: 3160.38 17:55
18:09 brrt left
masak oh wow -- myusuf3.github.com/octogit/ 18:10
I'm a user of this from now on.
sjn: perlcabal.org/syn/ 18:11
Ulti masak can you accept pull requests to the github repo with that? 18:18
dalek pan style="color: #395be5">perl6-examples: 0c9beca | (Filip Sergot)++ | best-of-rosettacode/TODO:
3 new examples.
 18:20
sergot sjn: ping :)
18:21 birdwindupbird joined 18:22 mucker left
masak Ulti: doesn't look it. 18:23
sorear good * #perlt6 18:27
phenny sorear: 11:58Z <tadzik> ask sorear any ideas why dalek is not reporting my rakudo commits?
masak sorear! \o/ 18:28
pmichaud r: say (:a<42> cmp Inf) 18:30
p6eval rakudo 120e95: OUTPUT«-1␤»
pmichaud r: say (:a<42> cmp -Inf)
p6eval rakudo 120e95: OUTPUT«1␤»
pmichaud r: say (-Inf cmp :a<42>)
p6eval rakudo 120e95: OUTPUT«-1␤»
pmichaud r: say (Inf cmp :a<42>)
p6eval rakudo 120e95: OUTPUT«1␤»
sorear tadzik: ping 18:31
frettled r: my $x="foo"; say "What it is: " ~ $x.WHAT; 18:40
p6eval rakudo 120e95: OUTPUT«use of uninitialized value of type Str in string context in method Str at src/gen/CORE.setting:670␤␤What it is: ␤»
frettled but:
r: my $x="foo"; my $y=$x.WHAT; say $y; 18:41
p6eval rakudo 120e95: OUTPUT«Str()␤»
frettled r: my $x="foo"; my $y=$x.WHAT; say $y ~ " is what it is";
p6eval rakudo 120e95: OUTPUT«use of uninitialized value of type Str in string context in method Str at src/gen/CORE.setting:670␤␤ is what it is␤»
moritz precompiled MiniDBD::mysql works \o/ 18:45
jnthn OMG FINALLY \O/ 18:48
dalek p/toqast: 3ccdf9e | jonathan++ | / (11 files):
Get the various QAST nodes we have stubbed in and start to fill some of them out.
tadzik sorear: pong 18:49
sorear tadzik: how do you commit to rakudo? 18:50
tadzik sorear: git push
well, commit 18:51
frettled hm
sorear to what repo?
tadzik rakudo/rakudo, nom branch
frettled r: my $x="foo"; say "What it is: " ~ $x.WHAT.perl;
p6eval rakudo 120e95: OUTPUT«What it is: Str␤»
frettled Now that works.
sorear I would guess that maybe nobody updated the push address for rakudo
tadzik for example github.com/rakudo/rakudo/commit/626ef2a8fe6
sorear can you use the Admin tab?
tadzik sorear: it reports everyone else, just not me
nope
moritz sorear: but other rakudo commits are reported fine
sorear wha 18:52
tadzik aye
the only special thing about me is the "ś" in my name, maybe that affects something
sorear
.oO( You could try removing that and see what happens )
 18:53
tadzik yeah, well
masak dubs tadzik "Teddy" 18:54
did it help? :)
tadzik tail amputation
sorear tadzik: when did you push your last comitt?
colomon r: say Str.WHAT
p6eval rakudo 120e95: OUTPUT«Str()␤»
tadzik sorear: 8 hours ago
colomon r: say ~Str
p6eval rakudo 120e95: OUTPUT«use of uninitialized value of type Str in string context in method Str at src/gen/CORE.setting:670␤␤␤»
colomon r: say Str.WHAT.perl 18:55
p6eval rakudo 120e95: OUTPUT«Str␤»
colomon r: class X { }; say X.new.WHAT.perl
p6eval rakudo 120e95: OUTPUT«X␤»
colomon r: class Y::X { }; say Y::X.new.WHAT.perl
p6eval rakudo 120e95: OUTPUT«Y::X␤»
colomon :\
sorear tadzik: github doesn't seem to have sent dalek *any* push packets containing the string doc=HTML 18:59
I haven't a clue
tadzik huh 19:00
so maybe github's failing with my name
hold on, I'll be pushing imarcusthis' commits in a minute
dalek kudo/nom: 9b16150 | (Marcus Ramberg)++ | / (2 files):
Reformat --doc output
 19:02
kudo/nom: 43445e5 | (Marcus Ramberg)++ | / (2 files):
Fix method stringification in Pod::To::Text, fix slurpy star in Parameter.perl
kudo/nom: 21325bb | (Marcus Ramberg)++ | src/core/Parameter.pm:
Remove typo in paramter and cleanup
tadzik yeah, must be my name
masak hugs tadzik's "ś"
ś: don't ever change. you're beautiful just the way you are. 19:03
colomon r: say Str.WHAT.WHAT
p6eval rakudo 120e95: OUTPUT«Str()␤»
colomon r: say Str
p6eval rakudo 120e95: OUTPUT«Str()␤»
tadzik r: say .WHAT.WHAT.WHAT for "what was that word young man!?"
p6eval rakudo 120e95: OUTPUT«Str()␤»
masak r: sub the { @_ }; class Butt {}; .WHAT.WHAT given the Butt 19:04
p6eval rakudo 120e95: ( no output )
colomon r: class X { method perl() { self.WHAT; }; }; say X.perl 19:05
p6eval rakudo 120e95: OUTPUT«X()␤»
19:05 geistteufel joined
colomon r: class X { method perl() { self.WHAT.perl; }; }; say X.perl 19:05
n: class X { method perl() { self.WHAT.perl; }; }; say X.perl
guess it's consistent.
p6eval rakudo 120e95: OUTPUT«(timeout)» 19:06
niecza v16-23-gaa61ed5: OUTPUT«(timeout)»
tadzik myusuf3.github.com/octogit/ looks cool
frettled colomon: fun, eh? 19:07
colomon frettled: that used to work fine in Rakudo. Now it's an infinite loop.
tadzik: Math::Polynomial should work fine now. 19:08
tadzik colomon++, great
we can now automagically smoketest ALL THE MODULES :)
colomon \o/ 19:09
what are you using? 19:10
tadzik colomon: tjs.azalayah.net/new.html is generated using a single perl 6 script :)
masak tadzik++!
tadzik I'll push it to the panda repo right away
colomon OMG, the graphic is fantastic!
tadzik oh, those are outdated a bit
(not the graphics) 19:11
frettled is that shopped? ;)
tadzik no, not at all 19:12
list updated
masak no, he actually has wings on his head.
dalek kudo/nom: cf19ce4 | moritz++ | / (2 files):
update docs/running.pod and perl6 --help output

  --ll-backtrace is now --ll-exception. Add --profile
tadzik like, where do I get photoshop for linux? :)
moritz gimpshopped!
tadzik shush!
it's original!
(genuine?)
colomon tadzik: now you need to have it smoke under both Rakudo and Niecza... 19:13
dalek kudo/nom: edf2e0d | (Geir Amdal)++ | src/core/IO.pm:
added stat times (modified/changed/accessed) for IO
kudo/nom: 631d8ab | moritz++ | src/core/IO.pm:
Merge pull request #63 from gam/nom

added stat times (modified/changed/accessed) for IO
tadzik colomon: colomon that's a single-line change if Panda can run on niecz
-a
colomon ooooo 19:14
sorear Whose head is that?
tadzik Camelia's
:)
frettled Tadzelia
tadzik spunky little brother
masak but still carrying the high ideals. 19:15
sorear My super-sensitive decision procedures think that head is more likely to belong to a human than a butterfly.
frettled What's the spectest for that?
masak tadzik: now where was that picture with the man with the expression? :)
dalek kudo/nom: d7b76d9 | moritz++ | tools/build/NQP_REVISION:
bump NQP revision

now MiniDBD::mysql finally works with precompiled modules. Time to celabrate!		 19:16
masak tadzik: "you don't say?"
:)
tadzik :)
sorear colomon: my gut told me grammars should be anchored on both ends. I guess spec compatibility requires removing the right anchor? 19:17
frettled n: my $x="foo"; say "What it is: " ~ $x.WHAT;
p6eval niecza v16-23-gaa61ed5: OUTPUT«What it is: ␤»
colomon sorear: I don't really know. Both ways make above equal amounts of sense to me. 19:18
sorear: well, I might even prefer your way.
frettled n: my $x="foo"; say $x.WHAT;
p6eval niecza v16-23-gaa61ed5: OUTPUT«Str()␤»
colomon sorear: but some of the old ABC tests assumed the Rakudo way.
frettled n: my $x="foo"; say $x.WHAT.perl ~ " what"; 19:19
p6eval niecza v16-23-gaa61ed5: OUTPUT«Str what␤»
dalek p/toqast: fdba789 | jonathan++ | src/QAST/Op.nqp:
We'll just spell pasttype as op now. Now that'll save some keystrokes...
 19:23
p/toqast: 6b5c556 | jonathan++ | src/QAST/ (2 files):
Fill out QAST::Var and QAST::Block a little.
masak today sergot++ and I published two modules: Text::Markdown, which does simple Markdown parsing (italics, bold, inline code, paragraphs), and Sambal, which builds a PDF presentation using a simple DSL-y syntax. Sambal has Text::Markdown as a dependency, so you can format your text with Markdown. 19:25
jnthn najs
masak++, sergot++
moritz masak++, sergot++
jnthn Are they on modules.perl6.org>
masak here's the general idea for Sambal: gist.github.com/2173720
Text::Markdown is on there. Sambal will be soon. 19:26
sorear What happened to fsergot?
jnthn
.oO( Decided to "f" off )
 19:27
tadzik oh, he changed his name for the 3rd time :)
masak sergot reports that the nick change was due to pronunciation considerations.
tadzik it's like with my ś
moritz decided to dynamically typecast? :-)
masak but ś is so easy to pronounce...
tadzik heh. Hearing people saying tad-zik makes me see why people have issues with nie-cza
masak yeah! 19:28
both are dead easy to pronounce.
moritz so easy that you drop dead if you try? :-)
tadzik it's that people literally say 'dz' rather than the sound-y thing, like in "dź"
moritz r: say $*VM.elems 19:30
p6eval rakudo 21325b: OUTPUT«2␤»
moritz r: say $*VM.keys
p6eval rakudo 21325b: OUTPUT«name config␤»
19:31 PacoAir left 19:33 PacoAir joined
dalek pan style="color: #395be5">perl6-examples: fdeb067 | (Filip Sergot)++ | best-of-rosettacode/TODO:
New examples in TODO list.
 19:34
frettled tadzik: and now you know why my nick is "frettled" and not "jingvoldstad" or somesuch ;)
tadzik heh, right :) 19:35
masak jingvoldwhat?
frettled Mnemonic: Ing(en) vold i stad(en), which works in several Scandinavian languages. 19:36
…got English outnumbered by at least three to one!
masak :D
sergot sjn: ping
sjn ping 19:37
oops.
pong
19:37 PacoAir left
frettled Oh dear, you know the example code you're looking at is slightly old when it sayd: 19:37
says
./pugs -e "say 1+1"; # 2
19:37 PacoAir joined
masak p: say 1+1 19:37
p6eval pugs: OUTPUT«2␤»
jnthn r: say 1 + 1
p6eval rakudo 21325b: OUTPUT«2␤»
masak frettled: no, that's still so. :)
jnthn Still works :) 19:38
frettled I was thinking more that it naturally assumed ./pugs over ./perl6
colomon The following step can take a long time, please be patient. 19:39
./perl6 --setting=NULL --optimize=3 --target=pir --stagestats --output=src/gen/CORE.setting.pir src/gen/CORE.setting
make: *** [CORE.setting.pbc] Segmentation fault
:(
moritz phenny: "ingen vold i staden"?
phenny moritz: "no vold in the city" (sv to en, translate.google.com)
moritz phenny: no en "vold"?
phenny moritz: "violence" (no to en, translate.google.com)
tadzik colomon: sooner that you'd expect! :)
masak phenny: no en "ingen vold i staden"? 19:40
phenny masak: "no violence instead" (no to en, translate.google.com)
frettled ouch
masak Google Translate needs to learn Norwegian properly.
moritz too 19:43
krunen phenny: no en "ingen vold i stedet"? 19:46
phenny krunen: "no violence instead" (no to en, translate.google.com)
frettled Hmm. 19:48
Am I correct in guessing that in order to get the output of .WHAT, I need to use .WHAT.gist?
masak phenny: sv en "inget våld i städet"?
phenny masak: "no violence in the anvil" (sv to en, translate.google.com)
masak :D
the anvil appreciates not being beaten up, for once. 19:49
19:58 daemon left 20:00 GlitchMr left
frettled n: my $x="foo"; say $x.WHAT.gist ~ ' what what'; 20:03
p6eval niecza v16-23-gaa61ed5: OUTPUT«Str() what what␤»
frettled r: my $x="foo"; say $x.WHAT.gist ~ ' what what';
p6eval rakudo 21325b: OUTPUT«Str() what what␤»
frettled So I suppose I could substitute .WHAT.gist for nearly every use of .WHAT in github.com/perl6/perl6-examples/bl...duction.pl 20:04
pugs: my $x="foo"; say $x.WHAT.gist ~ ' what what'; 20:05
p6eval pugs: OUTPUT«Str what what␤»
frettled pugs: my $x="foo"; say $x.WHAT ~ ' what what';
p6eval pugs: OUTPUT«Str what what␤»
frettled hrm.
not quite, should probably use .WHAT.perl, but that feels wrong sortof
masak p6: say Str.^name 20:06
p6eval pugs, rakudo 21325b, niecza v16-23-gaa61ed5: OUTPUT«Str␤»
masak frettled: ^^
frettled aha
masak simpler :)
frettled p6: my $x="foo"; say $x.^name ~ ' what what';
p6eval pugs, rakudo 21325b, niecza v16-23-gaa61ed5: OUTPUT«Str what what␤»
frettled excellentium, masak++
20:06 harukomoto joined
moritz r: say 'WHAT?'.^name 20:07
p6eval rakudo 21325b: OUTPUT«Str␤»
masak also, don't concatenate arguments of a say :) we premature optimizers get fidgety :)
20:07 jaffa4 left
dalek pan style="color: #395be5">perl6-examples: e6f1005 | (Salve J. Nilsen)++ | best-of-rosettacode/ (6 files):
Update pot and add v6 requirements
 20:08
pan style="color: #395be5">perl6-examples: 38cded4 | (Salve J. Nilsen)++ | best-of-rosettacode/TODO:
Merge remote-tracking branch 'origin/master'
20:09 daemon joined
sergot sjn++ 20:10
frettled masak: why not?
masak frettled: because `say "foo", "bar", "baz"` and `say "foo" ~ "bar" ~ "baz"` look the same on the screen, but the latter allocates less memory and is faster. 20:11
er, the former, I mean.
frettled aha 20:12
I'll amend the examples for that as well, then.
masak \o/ ++frettled
jnthn i'D HOPE that the second gets constant folded soon though :) 20:13
dalek pan style="color: #395be5">perl6-examples: 441c07f | (Filip Sergot)++ | best-of-rosettacode/TODO:
bug...
 20:15
jnthn multi postfix:<++>("foo") { "bar" }
masak :P
tadzik r: multi postfix:<++>("foo") { "bar" } say "foo"++
masak nom: multi postfix:<++>("foo") { "bar" }; say "foo"++
p6eval rakudo d7b76d: OUTPUT«===SORRY!===␤Confused␤at /tmp/bOXeactxs8:1␤»
rakudo d7b76d: OUTPUT«bar␤»
masak \o/
moritz r: class FooStr is Str { method succ() { self.index('foo').defined ?? self.subst(:g, 'foo', 'bar') !! nextsame } }; my $s = nqp::box('somefoo', FooStr); $s++; say $s 20:16
p6eval rakudo d7b76d: OUTPUT«===SORRY!===␤Unrecognized nqp:: opcode 'nqp::box' at line 1, near "; $s++; sa"␤»
moritz r: class FooStr is Str { method succ() { self.index('foo').defined ?? self.subst(:g, 'foo', 'bar') !! nextsame } }; my $s = nqp::box_s('somefoo', FooStr); $s++; say $s
p6eval rakudo d7b76d: OUTPUT«somebar␤»
frettled gazonk.foo
moritz r: class FooStr is Str { method succ() { self.index('foo').defined ?? self.subst(:g, 'foo', 'bar') !! nextsame } }; my $s = FooStr.new('foobar'); $s++; say $s 20:17
p6eval rakudo d7b76d: OUTPUT«Default constructor only takes named arguments␤ in method new at src/gen/CORE.setting:560␤ in block <anon> at /tmp/lwGylcl_Sy:1␤␤»
20:23 kaare_ left
krunen masak: say "foo"~"bar" may be faster if using unbuffered io over RFC1149 20:25
masak krunen: :D 20:27
20:27 bruges left
masak I had a feeling which RFC that would be even before I looked... 20:27
tadzik oh, this one 20:28
...I suppose
20:28 birdwindupbird left 20:29 bruges joined
moritz perlgeek.de/blog-en/perl-6/2012-osl...eport.html # I blug 20:31
sergot moritz++ 20:34
dalek ast: 6e66bc7 | moritz++ | S04-declarations/state.t:
unfudge passing TODO tests
 20:35
masak moritz++
20:35 cggoebel left 20:36 cggoebel joined
dalek pan style="color: #395be5">perl6-examples: 99e14ad | (Salve J. Nilsen)++ | best-of-rosettacode/ (6 files):
Update titles, add use v6, fix links, add pod
 20:36
pan style="color: #395be5">perl6-examples: 6f30b8e | (Salve J. Nilsen)++ | best-of-rosettacode/TODO:
Merge remote-tracking branch 'origin/master'
pmichaud What version of Perl should we indicate that Rakudo implements (i.e., in $*PERL<ver>) ? 20:39
20:39 vlixes left
tadzik r: say "6.0.0".prev 20:40
p6eval rakudo d7b76d: OUTPUT«Method 'prev' not found for invocant of class 'Str'␤ in block <anon> at /tmp/vEtkBcc9Nk:1␤␤»
jnthn 6.0ish
;)
20:41 geistteufel left
pmichaud r: say "6.0.0".pred 20:41
p6eval rakudo d7b76d: OUTPUT«5.0.0␤»
pmichaud Ummm..... *no*. 20:42
20:42 geistteufel joined
dalek pan style="color: #395be5">perl6-examples: 48cd5de | (Salve J. Nilsen)++ | best-of-rosettacode/README:
Say few words about what this directory is about
 20:43
masak pmichaud++ # :D
sjn r: say v6.0.0.pred; # :)
p6eval rakudo d7b76d: OUTPUT«===SORRY!===␤Confused␤at /tmp/t2DL0RdGWe:1␤»
sjn r: say "v6.0.0".pred; # :-o 20:44
p6eval rakudo d7b76d: OUTPUT«v5.0.0␤»
sjn meh
pmichaud yes, seems like we need to implement the Version type (src/core/Version.pm) 20:45
dalek pan style="color: #395be5">perl6-examples: 889f983 | (Filip Sergot)++ | best-of-rosettacode/balanced-brackets.pl:
balanced-brakets.pl
 20:47
pan style="color: #395be5">perl6-examples: f8abe2c | (Filip Sergot)++ | best-of-rosettacode/ (7 files):
Merge branch 'master' of github.com:perl6/perl6-examples
20:48 simcop2387 joined
dalek pan style="color: #395be5">perl6-examples: 4d613ec | (Jan Ingvoldstad)++ | cookbook/01strings/01-00introduction.pl:
Replaced "pugs" with "perl6"
 20:51
pan style="color: #395be5">perl6-examples: 38621c4 | (Jan Ingvoldstad)++ | cookbook/01strings/01-00introduction.pl:
Replaced erroneous use of .WHAT with .^name, also replaced concatenation of say's arguments with comma-separated list (currently better practice due to performance). Cleaned up minor whitespace thingies in the same examples.
pan style="color: #395be5">perl6-examples: c7658f1 | (Jan Ingvoldstad)++ | cookbook/01strings/01-00introduction.pl:
Rearranged things a bit, changed examples for concatenation and display, and tried to re-categorize a bit
pan style="color: #395be5">perl6-examples: 86ca520 | (Jan Ingvoldstad)++ | best-of-rosettacode/ (10 files):
Merge branch 'master' of github.com:perl6/perl6-examples
tadzik can you think of any way how I could print $?USAGE from actions.pm? 20:52
20:52 nif joined
tadzik ($*W.symbol_lookup(['$?USAGE'], $/) gives me a Capture[0xfoo 20:52
dalek kudo/nom: 5d7626b | pmichaud++ | / (3 files):
Revise $*PERL variable a bit.
 20:53
20:53 nif is now known as manners
pmichaud 5d7626b isn't exactly what we discussed here in Oslo; I left the compiler information as $*PERL<compiler> for now instead of creating a $*COMPILER variable. 20:53
jnthn tadzik: You're confusing symbol_lookup and find_symbol, I think.
frettled jnthn: 6.0ish would be nice, except that version compare would show it as always greater than 6.0.$whatever :) 20:54
.oO(Would 6.0.-ish work?)
 20:55
tadzik jnthn: mebbe :) I just saw that Actions.pm uses symbol_lookup in lots of places when it looks up $?CLASS 20:56
jnthn tadzik: That means "generate code to do the lookup" 20:57
tadzik okay
so I could past::call that, si?
jnthn find_symbol gives you the actual looked up value
tadzik: call it?
tadzik: Isn't $?USAGE the tring message? 20:58
*string
tadzik it's a String, yes
but it's in Main.pm as my $?USAGE=gen-usage()
jnthn ah
tadzik so it's lazy-ish?
pmichaud actually, 6.0ish would be treated like 6.0.ish. Perhaps 6.0.0ish would work, though.
jnthn What are you trying to do with it?
pmichaud 6.0.0 would end up sorting after 6.0.0ish
tadzik jnthn: say it
pmichaud (at least if I'm reading S02 correctly)
jnthn tadzik: When?
tadzik like, Past::Op(:pasttype<call> :name<&say>)
jnthn: in DOC INIT 20:59
jnthn oh
frettled pmichaud: hmm, that breaks with version parsing in other thingies than Perl 6, heh :)
20:59 harukomoto left
jnthn PAST::Op.new(:pasttype('call'), :name('&say'), ...the symbol lookup...) 20:59
tadzik right
jnthn should do it
pmichaud > say $*PERL<compiler>.perl 21:00
{"name" => "rakudo", "ver" => "2012.04-42-g5d7626b", "release-number" => "51", "build-date" => "2012-04-21T20:47:05Z", "codename" => "Brazos Valley"}
dalek p/toqast: d16bf96 | jonathan++ | src/QAST/Compiler.nqp:
Start sketching out compilation of some more of the nodes in QAST::Compiler.
tadzik pmichaud: cool
jnthn: bah, that gives me Could not find symbol '$?USAGE' 21:01
whereas it works then I do say($*W.find...(...)) in Actions.pm
jnthn :S
tadzik my face exactly
jnthn tadzik: are you doing it too late?
frettled pmichaud++ - nice!
tadzik hrm, no eye deer
jnthn Like, after $*UNIT was popped off in comp_unit
tadzik it's in a new lexpad, but not much else I suppose 21:02
okay, paste coming
jnthn: gist.github.com/2439577 21:03
pmichaud r: sub xyz( :f($foo) ) { ... }; say &xyz.signature.perl 21:04
p6eval rakudo d7b76d: OUTPUT«:($foo)␤»
pmichaud ^^^ that's a regression
21:04 harukomoto joined
sorear pmichaud: so I'm thinking that at least for consistency's sake it would be best to remove Niecza's automatic end anchor in Grammar.parse 21:11
masak huh? why? 21:12
I feel I've missed the beginning of this discussion.
masak backlogs
pmichaud or to see if we can get Rakudo to add it, if that feature is still spec (and deserving to be spec)
tadzik pmichaud: github.com/rakudo/rakudo/commit/9b...98d#L1L153 21:13
sorear masak: there isn't really a beginning, it's been going on forever ;) 21:15
I think the spec is silent
to me, it's self-evident that a grammar should match the entire input, not some prefix
to other people, including ones who write tests for roast and ones who implement Rakudo, it's the other way 21:16
pmichaud the spec is not completely silent
masak sorear: IMHU Rakudo simply doesn't do it yet.
sorear imh..u?
pmichaud smart matching against a regex, token, or rule (directly) causes both ends to be anchored
geekosaur understanding?
pmichaud i.e., 'abc' ~~ regex { \w+ } is supposed to have implicit anchors, per spec 21:17
masak yes, "understanding".
to me, the question has always been "where does this extra anchoring magic reside?". I guess the answer is "in .parse" 21:18
pmichaud I think the spec is somewhat silent on Grammar.parse
21:18 ens joined 21:19 fglock left, manners left
pmichaud certainly I think there needs to be a way to do .parse that doesn't force anchoring 21:20
otherwise you simply end up with "failed match", or you have to make sure your regex includes something like .*
21:21 manners joined
masak what's the value of a s/// expression? 21:22
p6: $_ = '[[[][]][]]'; say s/'[]'// 21:23
p6eval rakudo 5d7626: OUTPUT«[[[]][]]␤»
..pugs: OUTPUT«<Subst>␤»
..niecza v16-23-gaa61ed5: OUTPUT«True␤»
masak *lol*
fwiw, I'm with Niecza until I get more of the backstory here. 21:24
pmichaud after reading S05, I'm with Niecza also 21:32
21:33 Trashlord left
masak TimToady: does a s/// expression evaluate to a boolean representing True iff it matched? 21:33
pmichaud afk here -- time to close down hackathon for the night 21:35
more hacking tomorrow!
TimToady a s/// should return the number of times it matched, I think, which without :g is 0 or 1 21:36
21:36 Trashlord joined
TimToady but it doesn't need to be a Bool 21:36
21:37 manners left, vlixes joined
masak ooh 21:38
yes, that's probably even better.
TimToady and there's prior art :)
masak p5? :) 21:39
TimToady and 4, and 3...
jnthn pmichaud says that .subst, which s/// uses, doesn't give us the number of matches, or even if anything matched 21:41
TimToady well, he should stop saying that :)
masak and there are recent rumors that pmichaud is not susceptible to torment any more... :)
jnthn "see, I told you!"
Pm wants to know if there's anything else he should stop saying :P 21:42
masak the room just got colder...
masak sleep & 21:43
jnthn indeed
'nighty night o/
TimToady hypothermia?
sergot o/
21:43 sergot left 21:45 NamelessTee left 21:52 NamelessTee joined 22:20 zhangkaizhao left 22:22 PacoAir left 22:30 brrt joined 22:31 datums_nb joined, NamelessTee left
pmichaud back at hotel :) 22:40
22:41 datums_nb left 22:49 brrt left 22:57 harukomoto left 23:03 harukomoto joined, fgomez joined, fgomez left, fgomez joined
sorear yay. 23:24
23:51 icwiener left

Powered by Cro and the Raku® Programming Language.

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

Thank you!