»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.
TimToady it's only a special case where we already have a special case to deal with name arguments 00:04
*named
frooh hey guys, I was hoping to use the build tools from github.com/perl6/book to create a commercial work; does anyone know if I can *use* the build tools for theat, given that they themselves won't be distributed or sold? 00:13
sorear frooh: where do you live? 00:14
frooh US
(TX, to be specific)
sorear AFAIK, unless you clicked "I agree" at some point, there are no rules on usage 00:15
TimToady well, it my fall under implicit copyright under the Berne Convention, which does potentially have some restrictions 00:17
PerlJam frooh: ask chromatic if you can use the tools he used for his book.
frooh PerlJam: yeah, that's really what I should do
anyone know the best way to get in touch with him? 00:18
TimToady say something stupid on reddit? 00:19
frooh me?
ohhh
I get it
PerlJam frooh: or just email. 00:20
frooh [email@hidden.address]
frooh guesses 00:20
TimToady looks about right 00:21
jdhore1 TimToady++ 00:27
tadzik insomnia-- 00:36
sorear hello jdhore1! 00:37
jdhore1 morning sorear
tadzik wonders if 02:38 AM is morning 00:38
jdhore1 It's actually 8:41PM here so... :P 00:41
PerlJam jdhore1: you're an hour ahead of me ... how's the future look so far? :) 00:42
jdhore1 PerlJam, Dark. It's past sundown, so... 00:43
sorear jdhore1: I hear you like writing release announcements 00:49
jdhore1 sorear, So I put a release announcement in your release announcement so you can release while you release? 00:50
sorear, Seriously though, I just really wanted release #42 so I could make about 5 Douglas Adams/HHGTG jokes in the announcement 00:51
dalek kudo: 3696656 | perlpilot++ | docs/release_guide.pod:
Claim my traditional birthday-month release :-)
01:01
tadzik :) 01:16
colomon phenny: tell jnthn You were supposed to see gist.github.com/987868, but I forgot to tag it. That's the stack trace from the rakudo-nom segfault. It didn't crash under valgrind, alas. 01:32
phenny colomon: I'll pass that on when jnthn is around.
sorear jdhore1: anyways, I was going to tell you I have a releaser opening. 02:23
jdhore1 sorear, what've you got? 02:24
sorear ...a compiler? 02:26
jdhore1 oh, I didn't read that correctly 02:27
Well the thing is, I don't have any Windows boxes and i'm rather glad I don't have Mono (or Java, for that matter) on my system.
sorear did you use Java before it was associated with Oracle and Larry Ellison? 02:29
jdhore1 not really
but between when they open-sourced it and the Oracle transition was complete, I thought Java was DECENT. 02:30
dalek ecza: 7acdc4a | sorear++ | lib/ (2 files):
More streamlining of the metasub construction process
05:09
tadzik jdhore1: fwiw, I keep mono only for niecza 05:10
jdhore1 heh
tattochan Hello 05:32
jdhore1 sorear, still on? 05:32
sorear jdhore1: yes 05:33
jdhore1 sorear, I'll take the job :)
sorear jdhore1: only if you want it; I offered because I interpreted a prior statement of yours as "I want to do whatever I can" 05:54
jdhore1 sorear, sure, why the hell not?
sorear currently niecza generates 1.85 MB for STD.dll; 7.5% is CLR metadata, 20% is the general initializer blob, the rest is code 05:55
niecza could use an operation turning large initializers into data 05:56
jdhore1 sorear, Well...I've never written C# so i'm not sure how much I can help 05:57
sorear the hash literal for %STD::open2close compiles to about 20kb of code, the second larges single method 05:57
daniel-s perl6: say $*PERL_VERSION 10:11
p6eval rakudo 369665: ( no output )
..niecza v5-62-g7acdc4a: OUTPUT«Any()␤»
..pugs: OUTPUT«␤»
tadzik perl6: say $*PERL<version>
p6eval rakudo 369665: OUTPUT«2011.05-5-g3696656␤» 10:12
..niecza v5-62-g7acdc4a: OUTPUT«Any()␤»
..pugs: OUTPUT«␤»
daniel-s only rakudo answers?
tadzik seems so, but it's wrong anyway, there's an RT ticket for that
daniel-s when you pass a variable into a subroutine, it's passed by reference? 11:01
tadzik everything is a reference in Perl 6. Do you mean whether it's modifyable then? 11:02
gfldex rakudo: my $a = 5; sub foo($a is copy){ $a = 10 }; sub bar($a is rw){ $a = 11 }; foo($a); say $a; bar($a); say $a; 11:58
p6eval rakudo 369665: OUTPUT«5␤11␤»
takadonet morning all 12:05
tadzik morning takadonet 12:13
takadonet tadzik: how are u sir? 12:14
tadzik takadonet: quite well, thank you
how are things?
takadonet busy but good
tadzik heh, the next weeks are my new definition of "busy". I never thought I'll look forward to an exam session 12:15
exam session is three exams a week. Phew! Compare that to two exams a day
and programming projects to finish 12:16
moritz phenny: tell frooh that the you can use the build tools of perl6/book for anything you like - the non-commercial license only covers *redistribution* - if you don't re-distribute the tools for commercial purpose, you are fine 14:02
phenny moritz: I'll pass that on when frooh is around.
moritz I guess frooh only wants to commercially redistribute the *output* from the tools, which is not covered by the license at all 14:03
tadzik I read that as "hey, your book build system is awesome" :) 14:42
jnthn oh hai 15:42
phenny jnthn: 01:32Z <colomon> tell jnthn You were supposed to see gist.github.com/987868, but I forgot to tag it. That's the stack trace from the rakudo-nom segfault. It didn't crash under valgrind, alas.
masak o/, #perl6! 18:01
phenny masak: 22 May 16:47Z <sbp> ask masak what's up with proto. is it deprecated so that panda can replace it? or will you be getting back to it? if someone installs perl6, which one would you advise them to use? if proto is out of date, can you put a little message to that effect at the top of proto/README please?
masak: 23 May 20:34Z <Tene> ask masak whether he ever released his slides from yapc::ru
masak: 23 May 20:39Z <Tene> tell masak nm; moritz mentioned that you already posted a link
sbp welcome, masak 18:02
masak sbp: yes. no. panda. sure, I'll try. :)
sbp thanks! 18:02
masak #phasers in 58?
mberends o/ masak. Aye.
masak mberends: \o! 18:03
masak "Go is a language that combines the safety and performance of statically compiiled languages, with the expressiveness, the convenience and the fun of dynamic languages." -- Rob Pike in itc.conversationsnetwork.org/shows/...l4764.html 18:55
sound familiar? ;)
Util masak: manifesto published anywhere? 18:59
masak: familiar, modulo performance :( 19:00
masak gist.github.com/984783
Util: I'm hopeful wrt performance. I've spent a week in jnthn's company :)
jnthn I performed well? :)
masak looks distraught o.O 19:01
jnthn :P
masak I mean, I've learned a lot about what 6model and nom will do for Rakudo.
optimizations++ 19:02
jnthn I know, I know... :)
I even tricked you into offering to implement a chunk of it :P
masak \o/
masak <3 ASTs
Util masak: thanks for pushing through that S06 change! I started (almost) writing a talk on APIs last month, and it occured to me that Perl6 would be over-obligating future API designers.
jnthn tossed that chunk of the binder last night :)
Patch removed 100ish lines of code from a hot path. :) 19:03
masak now we only have to toss the part of S06 that says that many nameds with the same name maps to a parameter value that's an array... 19:04
jnthn Oh, I ignored them
:) 19:05
masak yeah, well.
I think the S06 should follow suit.
tylercurtis tadzik: In my two quarters (discounting the current) at university thus far, the week of final exams has been rather relaxing compared to the rest of the quarter. 19:15
Util mberends: repo for Perlduino ? 19:24
mberends Util: not yet, but will be created soon :) 19:25
Util thx! mberends++
masak .oO( release early... ) 19:26
sorear good * #perl6 19:33
masak o/ 19:34
masak sorear: what does 'making niecza avoid appdomains' mean? 19:37
sorear masak: are we leaving #phasers for this on purpose? 19:38
masak sorear: yes, unless it makes more sense to discuss it there. 19:39
sorear finishes backlogging #phasers 19:40
masak: how familiar are you with .net/CLR stuff? 19:41
masak oh, it's CLR-specific?
sorear yes
masak I was worried it was about Perl 6 applications :P
I un-ask my question, then.
sorear ok. 19:42
sorear next week I may make #phasers ON TIME 19:44
wouldn't that be great? 19:45
masak ;)
dalek kudo/nom: 5360e04 | jnthn++ | src/Perl6/ (3 files):
Get term:name fixed up a bit.
19:57
sorear perl6: sub foo($xyz, :abc($def)) { say "$xyz $def" }; foo(1,2) 20:15
p6eval pugs: OUTPUT«*** ␤ Unexpected "abc"␤ expecting space or ")"␤ at /tmp/ePfcXNqqyX line 1, column 16␤»
..rakudo 369665: OUTPUT«Too many positional parameters passed; got 2 but expected 1␤ in 'foo' at line 1:/tmp/ehx_qiqrgm␤ in main program body at line 22:/tmp/ehx_qiqrgm␤»
..niecza v5-62-g7acdc4a: OUTPUT«Unhandled exception: Excess arguments to MAIN foo, used 1 of 2 positionals␤ at /tmp/3rlH7rt7fz line 0 (MAIN foo @ 0)␤ at /tmp/3rlH7rt7fz line 1 (MAIN mainline @ 1)␤ at /home/p6eval/niecza/lib/CORE.setting line 1414 (CORE C590_ANON @ 2)␤ at
../home/p6eval/niecza/lib/CORE.setting l…
cognominal_ jthn I don't want to be pushy but if you still want to come, we are longing to see your talks proposals in fpw 2011 site. Too bad you can't drag them from an ACT site to another :) 20:24
jnthn cognominal_: I'm not entirely sure I can...it's a week after Nordic Perl Workshop and a week before the BJPW. 20:26
cognominal_ thx, I just to konw the odds. 20:27
jnthn cognominal_: Will try and decide for sure soon.
Like, next couple of days.
pmichaud TimToady++ # best way to contact chromatic: say something stupid on reddit lolololololol
phenny pmichaud: 23 May 23:48Z <sorear> tell pmichaud How hard do you intend to fight 1 => $foo ? I'd rather not be implementing it if it will go away soon.
cognominal_ jnthn: ok, thx
pmichaud sorear: I don't intend to fight hard.
I disagree with it, even strongly, and will support others who oppose it, but I don't plan to lead the fight. :) 20:28
I'm also open-minded enough to accept that perhaps it's useful and just haven't seen the example yet.
masak 'night, #perl6 20:30
dalek p: 2ca4db0 | jonathan++ | src/6model/reprs/P6 (3 files):
Update native repr handling to fit latest thinking.
21:27
colomon jnthn: ping? 21:35
jnthn colomon: pong 21:41
colomon was the stack trace of any use to you? is there something else I can do to help along that front?
jnthn colomon: It was useful of you to provide it. Sadly, it doesn't immediately pinpoint where the problem lies - it's more action-at-a-distance-y. :( 21:42
colomon :(
sorear I plan to do nothing wrt. positionals and names until the spec is next changed 21:42
the change I am hoping for is 1. revert last change 2. make named-positional conversion "opt in" (syntax TBD) 21:43
but any solution that pmichaud is happy with, I'd also be fine with
moritz fwiw rakudo/nom also fails to build for me 21:44
"Method 'compose' not found for invocant of class 'SixModelObject'"
while compiling the setting
sorear how much memory does rakudo take to build these days?
jnthn moritz: Ouch. 21:45
sorear if it's back under 80 Mwords or so, I'd like to get an up to date rakudo installed
moritz I think the aggressiveness of the GC depends on available memory these days
jnthn moritz: Can I have full output?
moritz jnthn: forthcoming...
jnthn std: Submethod 21:47
p6eval std 4b1b100: OUTPUT«ok 00:01 111m␤»
jnthn std: SubMethod
p6eval std 4b1b100: OUTPUT«===SORRY!===␤Undeclared name:␤ 'SubMethod' used at line 1␤Check failed␤FAILED 00:01 111m␤»
sorear I put very little stock in that, fwiw
moritz jnthn: gist.github.com/989813
sorear perl6/std:CORE.setting is pretty much at the bottom of the spec hierarchy 21:48
jnthn moritz: oddness. 21:52
moritz fwiw nom never built for me 21:54
jnthn It only started building for me yesterday :)
moritz I know :-) 21:55
moritz jnthn: sooner or later you need to produce better error messages than Method <methodname> not found for invocant of class 'SixModelObject' -- maybe "sooner" helps debugging 21:57
colomon jnthn: oh, hey, I'm back to the OS X bug, I think!
> ld: can't open output file for writing: perl6, errno=21
sorear colomon: perl -E '$! = 21; say $!' 21:58
moritz "is a directory"
we've had that before
colomon moritz: yes, thus "I'm back to".
moritz stupid case insensitivity
sorear ...do you have a directory with the same name as the file?
colomon yesterday I couldn't even get that far. :)
jnthn moritz: Yeah, I knows. :)
colomon sorear: yes
sorear right, makes total sense.
shouldn't it be src/Perl6/... ? 21:59
jnthn sorear: That's the uncompiled source.
sorear does nom do away with src?
jnthn sorear: the Perl6 directory holds the compiled output.
sorear: Maybe should be lib :)
or blib 22:00
:)
moritz or src/gen/
jnthn has no idea :)
moritz: Well, not really (more)
These are actual PBCs that we will want to install
benabik HFS is case insensitive. On most OS X machines, "Perl6" and "perl6" are the same file.
jnthn benabik: Yeah, I was OS X ignorant. :) 22:01
moritz thinks that "gen" stands generally for generated files, indepently of whether they should be installed or not
benabik Oh, I think someone mentioned it already. Timer went off in kitchen and delayed advice. :-/
moritz oh, bold fail :-)
jnthn moritz: Well, it's more about where a "use" statement will look. 22:02
benabik jnthn: I've spent the last few years making fixes in git's source where someone didn't think about it, so I'm far more used to it than anyone should. 22:03
moritz -> sleep 22:05
dalek kudo/nom: 6918b7c | jnthn++ | src/Perl6/SymbolTable.pm:
Fix up setting loading.
22:13
kudo/nom: ca8a1d2 | jnthn++ | src/Perl6/Metamodel/EXPORTHOW.pm:
Declare type of attribute to use with class.
kudo/nom: 590e586 | jnthn++ | src/Perl6/ (3 files):
Start to get attribute declaration in place.
kudo/nom: 69d2767 | jnthn++ | src/Perl6/Grammar.pm:
A little STD alignment.
kudo/nom: bd6373a | jnthn++ | src/Perl6/ (2 files):
First very basic and incomplete implementation of methods.
pmichaud yes, the gen/ directory is meant to be for files that are generated files. 22:21
It's unrelated to install.
it's okay if gen/ gets subdirectories
jnthn pmichaud: How do you propose use knows to look in gen pre-install and not post-install? 22:22
pmichaud jnthn: PERL6LIB?
jnthn pmichaud: That'd probably work.
pmichaud seems like the most obvious answer
jnthn If we make NQP pay attention to it as well, anyways. :) 22:23
pmichaud for NQP, we needa -L option anyway
so that's a possibility also
jnthn Already added something like that, I think.
pmichaud so, -L then
pmichaud for NQP I'd either suggest having an NQPLIB or using -L 22:24
jnthn Will try and fix those bits up soon...right now I'm mostly focused on getting to the point where @other can give in.
er
join in :)
(Quite hard to do that at the moment... :))
Still putting back primitives. 22:25
pmichaud maybe I can pitch in on that tomorrow. today I've managed to eliminate a lot of @other_tasks
jnthn Should have attributes and methods mostly in place by tomorrow.
Need to sort out multi-dispatch soonish.
pmichaud being called to dinner -- bbl (or maybe bbt)
jnthn k 22:26
I gotta sleep soon anyways :)
dalek kudo/nom: 2a6e69a | jnthn++ | NOMMAP.markdown:
Update NOMMAP with more details of most immediate next tasks.
22:29
jnthn sleep & 22:33
jdhore1 I found the perfect release name for Rakudo 2011.06 23:05