»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo: / pugs: / std: , or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by wolfe.freenode.net on 30 October 2009.
00:05 astrojp left 00:06 brrant left 00:07 mberends left 00:12 justatheory joined
cognominal can someone explain me (in private not to pollute the canal) the "not in kansas anymore" meme? 00:26
00:35 colomon_ joined 00:39 dbrock joined
diakopter rakudo: grammar Class is also { rule TOP { x } }; Class.parse("x") # masak 01:05
p6eval rakudo 7ef386: Null PMC access in get_attr_str()␤in Main (file <unknown>, line <unknown>)␤
diakopter (saw yours above)
masak diakopter: hm, yes.
but a difference is that 'Class' is defined in Perl 6. 'Integer' isn't/ 01:06
s[/][.]
diakopter oh 01:07
masak lambda-the-ultimate.org/node/3705 01:08
diakopter: uh, I take that back. I'm not sure there is a Class class any more. I can't find it in the spec. 01:09
this one is kinda interesting, though: 01:11
rakudo: grammar Perl6Array {}
p6eval rakudo 7ef386: ( no output )
masak locally, 'Class Perl6Array already registered!'
cognominal masak : small programs are beautiful but it needs large languages :) 01:12
masak cognominal: maybe so. :)
arnsholt masak: It's an interesting point, but I don't agree with him in everything 01:13
masak I'm sort of hoping Perl 6 will turn out to be a small language hidden inside a large language.
arnsholt For one, shell and friends is ugly and clunky as sin
masak indeed. 01:14
how come shell languages are so ugly?
arnsholt And Lisp, while a very cool language, offends my aesthetic sensibilities (TimToady said something about toenail clippings and oatmeal, if memory serves)
masak arnsholt: I personally am not disturbet by the parentheses in Lisp. 01:15
arnsholt The parens are fine as such. It's just that everything looks the same (which is the point too), which makes it a lot harder to grasp the structure of the program, IMO
masak they are a consession for the compiler to eliminate the concept of precedence entirely, and they basically 'melt away' with good indentation. 01:16
yes, everything looking the same is a point I can understand and sympathise with.
arnsholt OTOH, the syntax (or lack thereof) does give you the possibility of Lisp macros, and they are mega-awesome 01:17
masak :)
arnsholt Returning to the post, I think he's right that bigger languages are in the future 01:18
masak wonders what size of languages artificial intelligences will prefer 01:19
arnsholt I can sympathise with his preference for minimalist languages like C, but when he mentions the search for a small as possible set of types and operations, I can't help but think of Russell and the search for the axiomatisation of mathematics
A wonderful idea, but blown to bits by Gödel
masak I don't know about that. 01:20
arnsholt AIs will probably use something weird
Of course, programming languages being Turing complete won't run into the same problem
cognominal In his paper, Pike says "the art is gone". Is not Perl 6 about art? :)
arnsholt But it's definitely a positivist trait
01:28 gfx joined, jaldhar left 01:29 jaldhar joined 01:31 hcchien left, jaldhar left 01:32 vamped joined 01:33 jaldhar joined 01:37 hcchien joined
colomon_ rakudo: my @a = 1..4; @a <<*=>> 3; say @a.perl 01:46
p6eval rakudo 7ef386: [1, 2, 3, 4]␤
Wolfman2000 Wonder where PerlJam is...oh well 01:47
rgrau :w
ooops, sorry
colomon_ rakudo: my @a 1..4; @a <<*>>= 3; say @a.perl
p6eval rakudo 7ef386: Confused at line 2, near "1..4; @a <"␤in Main (file <unknown>, line <unknown>)␤
01:56 synth left
masak rakudo: class A { method postcircumfix:<{ }>($x) { 42 }; method foo() { my %h; say %h<foo> } }; A.new.foo 01:58
p6eval rakudo 7ef386: ( no output ) 01:59
masak locally, this gives '42'.
masak submits rakudobug
02:01 colomon_ left
Wolfman2000 masak: I don't get the same thing 02:03
Null PMC access in find_method('new') <-- that's what I get
masak that's odd.
maybe we should both update our Rakudos. :)
Wolfman2000 masak: looks like I already updated mine 02:04
masak excellent. I'll update mine.
Wolfman2000 of course, I may not be compiling the same options as you or something... (I just use the normal instructions)
masak oh, me too. 02:05
and in my experience, that has never been the cause of differences before. 02:06
blog post idea: Perl 6 is a result of looking at the corners on which people hurt themselves on Perl 5, and reshaping the language so that it doesn't hurt that way. the PerlMonks meditation at perlmonks.org/?node_id=525392 is a beautiful example, together with the last comment. "It appears you've just [made Perl 6 hurt less by telling us about how you got hurt on Perl 5]". 02:12
I imagine moritz_ might do it well. others are welcome to try it, too. if no-one bites, I might. :) 02:13
02:14 colomon joined
masak I'm also pretty sure that there are tens of other, similar, examples. 02:14
drbean www.urbandictionary.com/define.php?...%20anymore 02:20
masak Wolfman2000: still getting 42 here, after rebuilding newest Rakudo (master) on newest Parrot.
Wolfman2000 masak: I probably copied it wrong then. hold on
masak: ...yeah, copied it wrong last time. I got 42 02:21
masak Lesson 1 of copy-and-paste: Make sure you copy the whole th
Wolfman2000 masak-- for bad joke
masak :P 02:22
02:25 orafu left, orafu joined
colomon I've got a first draft of my Advent day 7 post into wordpress. I'd appreciate it if people gave feedback. 02:28
Wolfman2000 colomon: hang on
...who is supposed to do Day 6 again?
colomon mathw 02:29
mine could go up instead in a pinch.
masak that is an attractive option. it's nice to have a post waiting for people when they wake up. 02:30
Wolfman2000 still previewing: hang on
masak I'll give it a look as well.
Wolfman2000 ...what is this [sourcecode] thing? 02:32
colomon masak: agreed. unfortunately mathw went to bed about fifteen minutes before I could ask him how his post was coming. 02:33
masak yeah, I saw that.
colomon Wolfman2000: look at the preview. [sourcecode] blocks are formatted as source code. It's quite nice. 02:34
TimToady the :action vs :actions was question PM-7
masak colomon: I say go for it. change the title to 'Day 6' and change days with mathw.
TimToady: ah, that's why I remembered it... vaguely.
Wolfman2000 I'm presently in the edit mode: I can do so.
02:35 synth joined
masak I see I made a change that goes against Pm-7. I'll change it back. 02:35
colomon If you guys agree and can give it a quick once over to check my sanity, sure, we can run it for day 6. It actually is a sensible follow-up to day 5 anyway... 02:36
Wolfman2000 colomon: you have about 2 hours before your post goes up.
I'm out of edit mode so you can make your last minute adjustments
pugs_svn r29266 | masak++ | [S05] reverted s/actions/action/ correction 02:39
r29266 |
r29266 | Pm-7 asserts that it should be the plural. Which means Rakudo is lagging
r29266 | behind, not the spec.
colomon afk for moment, wife needs computer and hers just blew up.
masak colomon: I skimmed through it, and didn't find anything to critique.
masak submits rakudo TODO ticket for :actions 02:40
PerlJam colomon: looks good to me too (though the paragraph where you talk about parallelizing hyperops is written in terms of Rakudo, and I would have probably writte it in terms of Perl 6)
masak the ticket is probably of little value, since this change will likely come naturally with the adoption of ng, but you never know. 02:41
02:43 rgrau left
masak rakudo: say :2("0b1") 02:44
p6eval rakudo 7ef386: 23␤
masak o.O
Wolfman2000 rakudo: say 0b1 02:45
p6eval rakudo 7ef386: 1␤
Wolfman2000 rakudo: say 0b2
p6eval rakudo 7ef386: Syntax error at line 2, near "say 0b2"␤in Main (file <unknown>, line <unknown>)␤
Wolfman2000 okay, it has some intelligence...
masak why 23!?
Wolfman2000 rakudo: say :2(ob1)
p6eval rakudo 7ef386: Could not find non-existent sub ob1␤in Main (file src/gen_setting.pm, line 324)␤
Wolfman2000 ...derp
rakudo: say :2(0b1)
p6eval rakudo 7ef386: 1␤
masak oh, I think I get it.
Wolfman2000 rakudo: say :2(0b11) 02:46
p6eval rakudo 7ef386: 3␤
masak 23 = 22 + 1 = 11 * 2 + 1 = 0xB * 2 + 1
PerlJam rakudo: say :2("0b2")
p6eval rakudo 7ef386: 24␤ 02:47
PerlJam rakudo: say :2("0a1")
p6eval rakudo 7ef386: 21␤
PerlJam rakudo: say :2("0aa")
p6eval rakudo 7ef386: 30␤
PerlJam rakudo: say :2("aa") 02:48
p6eval rakudo 7ef386: 30␤
masak rakudo: "1234" ~~ /(23)/; say :4($0) 02:50
p6eval rakudo 7ef386: 11␤
masak I guess it's fine to rely on :4() auto-stringifying here?
02:53 meppl left
pugs_svn r29267 | masak++ | [S05] further improvements to :actions example 02:53
r29267 |
r29267 | Suggested by uasi++ on Twitter
colomon Okay, I'm cool with the post going up as scheduled at 5am UTC. (Is that an hour from now?) I tried to incorporate PerlJam++'s suggestions. 03:02
turning my MacBook back over to my wife so she can chat with Dell tech support.
Wolfman2000 ...that makes little sense, but whatever 03:05
cognominal reading S06, I found lines 92 and 103 in contradiction. line 92 says default scope is 'my' and line 103 says it's 'our'. Do I miss something?
colomon Wolfman2000: notice the previous comment about my wife's computer blowing up.
Wolfman2000 colomon: ah, just found it
colomon I get it back until she's off hold with Dell's on-line tech support. 03:06
That Dell has been a hideous lemon, but their tech support has been awe-inspiring.
I think the hard drive may be the only original component left....
masak cognominal: you probably found a fossil. 03:07
cognominal I am not sure which line is right...
03:09 solarion left
cognominal I guess 'my' is right by reading further 03:09
Also in S12:221 s/;/,/g 03:11
colomon phenny: tell mathw I had my Day 7 post ready to go this evening, so swapped with you. You're up for day 7 now. Hope that is okay. 03:12
phenny colomon: I'll pass that on when mathw is around.
Wolfman2000 ...something's telling me I should get my day 8 post ready 03:13
I may end up with day 7 at this rate
masak cognominal: do you have a pugs commitbit? 03:14
Wolfman2000: it could never hurt to start planning, at least :) 03:15
cognominal I should have even if I have not used it for months. Better use it tomorrow because bed time. 03:16
colomon I meant all along to try to get a few posts done and ready to go in case of unexpected need. But I have been pretty slow, and keep on getting distracted trying to make ng work.... 03:18
masak colomon++ # ng 03:20
diakopter std: <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 03:37
p6eval std 29267: ok 00:01 105m␤
pugs_svn r29268 | masak++ | [S05] further corrections by uasi++ 03:38
diakopter ng: say <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<34>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> * 2
p6eval ng 17a1ec: 68␤
masak that's enough <>'s to last you a year, at least. 03:40
03:43 justatheory left
colomon rakudo: say 4++ 03:45
p6eval rakudo 7ef386: Unable to set lvalue on PAST::Val node␤in Main (file <unknown>, line <unknown>)␤
colomon hmmm... does that mean the (1, 2, 3, 4)>>++ example in the day 6 post is wrong? 03:46
colomon is off to bed with that thought...
masak rakudo: say (1, 2, 3, 4)>>++ 03:47
p6eval rakudo 7ef386: Confused at line 2, near ">>++"␤in Main (file <unknown>, line <unknown>)␤
03:49 am0c joined 03:54 pnate joined 03:57 gfx left 04:06 pnate2 left 04:19 am0c left 04:31 agentzh joined
PerlJam rakudo: for (1;2;3) { } 04:39
p6eval rakudo 7ef386: ( no output )
PerlJam std: for (1;2;3) {}
p6eval std 29268: ===SORRY!===␤Obsolete use of C-style "for (;;)" loop; instead, in Perl 6 please use "loop (;;)" at /tmp/eAuAnNDQ7D line 1:␤------> for (1;2;3)⏏ {}␤FAILED 00:01 106m␤
04:40 jaldhar left 04:42 colomon_ joined, colomon left, colomon_ is now known as colomon 04:52 envi^home joined
PerlJam Hrm... 04:56
I just realized something. If we're casting Perl 6 as another language in the Perl family of languages, then saying "obsolete use of ..." is probably not quite the right thing to say 04:57
masak well, it's obsolete from the timeline of Perl 6. 04:58
PerlJam Ask mst what he thinks about it :) 04:59
masak all that's happened is that we have two timelines nowadays; two versions of Perl history, as it were.
I will ask him.
'obsolete' just means 'we used to do that, and now we try to avoid it for whatever reason'. 05:00
if you accept that the Perl 6 timeline runs back into the Perl 5 one before the fork, it's a valid use of the word. 05:01
...I think. :)
05:02 colomon left
PerlJam wonders what the Perl 5 messages should look like when someone feeds it Perl 6 code 05:02
"It looks like you gave me some Perl 6 code; I'm Perl 5 buddy!" 05:03
masak that's why we start our scripts with 'use v6;'
PerlJam Well, that too breaks the illusion of a family of languages developing in parallel. 05:04
masak yeah.
05:05 Baggio_ joined
masak nobody promised consistency. :) 05:06
PerlJam I guess it doesn't matter much as long as we have sufficient convergence 05:09
zaslon lolperl6adventhazblogged! perl6advent++ 'Day 6: Going Into Hyperspace': perl6advent.wordpress.com/2009/12/0...yperspace/
masak PerlJam: convergence on what? features? community? conferences? :) 05:10
PerlJam all that stuff
PerlJam makes a hand wavey gesture
btw, did someone talk to mathw about his post? 05:12
05:12 synth left
masak I don't think he's woken up yet. 05:12
he has a phennymessage waiting for him.
PerlJam good enough.
pugs_svn r29269 | duff++ | [perl6advent] adjust schedule to put colomon at Day 6 05:13
mdxi speaking of the advent calendar, is there a way to disable the annoying screenshot popups on link text? or am i the only one who despises those? 05:15
Wolfman2000 PerlJam: two of my websites have been moved over to NFS. Unfortunately, it looks like using frameworks such as Catalyst is a no go at this time.
PerlJam mdxi: um ... what are you talking about?
mdxi hover over a link in the article body 05:16
this is the first entry to have them
vamped they were there b4 today
PerlJam Ah ... I wasn't seeing it hovering over the first link.
But there's a link in the lower right corner of the popup that says "turn site preview off" 05:17
mdxi vamped: you're right. i think i read the others in w3m (oops)
masak mdxi: you're not the only one who despises those screenshots, no.
PerlJam isn't a fan of them either 05:18
masak mdxi: they can be turned off per-server on the client side easily.
mdxi yeah, i jsut found it thanks to PerlJam
masak mdxi: don't know how one'd turn them off in Wordpress.
mdxi nevermind :)
masak they *are* annoying, though. :)
mdxi i had been just scribbling my mouse around to make them go away. didn't actually inspect them. 05:19
i am bad programmer
masak no, they are bad widgets.
mdxi i violated Good Bug Report Standards, which i have frequently yelled at people about. karma is inescapable :) 05:20
PerlJam found the option to disable them.
vamped oh please tell! 05:21
masak whatever Good Bug Report Standards are, I'm not sure they hold for applications you didn't ask for. :)
PerlJam I just turned them off. Someone please test (since I already turned them off locally)
vamped i'm on board.
i've disabled them already too 05:22
masak just clear your cookies, then.
what made us peak on 2009-12-02 with 1080 views? was it a wave of people discovering the calendar for the first time and then not coming back? 05:23
or did people just really like learning about C<.fmt>? :P
PerlJam I think we just got a lot of twitter coverage then. 05:24
(more than now anyway)
masak that sounds about right. 05:25
PerlJam plus we're in a weekend and there aren't all those people who would be goofing off at work by reading the advent calendar :)
masak :) 05:26
PerlJam We'll have to see how it goes next week.
masak hugme: tweet rakudoperl Perl 6 Advent Calendar Day 6: Going Into Hyperspace bit.ly/82OMFh 05:27
hugme hugs masak; tweet delivered
masak hugs hugme back; thanks!
05:29 synth joined
masak time to head home. not a very productive night, but quite pleasant anyway. see y'all around! 05:30
05:31 masak left 05:42 jaldhar joined
diakopter www.info.ucl.ac.be/~pvr/paradigmsDI...eng108.jpg 05:57
06:14 envi_home2 joined 06:20 vamped left
diakopter "... talk about meta" 06:24
cute, diakopter; really cute.
zaslon lolperl6adventhazblogged! perl6advent++ 'Failure()<0xb5321970>': Failure()<0xb5319ec8>
Wolfman2000 ...here we go again
diakopter the grass is always greener, but only because of redshift. 06:25
zaslon lolperl6adventhazblogged! perl6advent++ 'Failure()<0xb53b5f8c>': Failure()<0xb53b5500>
diakopter um
06:31 nihiliad left, nihiliad joined 06:32 envi^home left
zaslon lolperl6adventhazblogged! perl6advent++ 'Day 6: Going Into Hyperspace': perl6advent.wordpress.com/2009/12/0...yperspace/ 06:34
06:36 colomon joined
zaslon lolperl6adventhazblogged! perl6advent++ 'Failure()<0xb53464f8>': Failure()<0xb53400a8> 06:36
Wolfman2000 is zaslon on ritalin or something? 06:37
zaslon lolperl6adventhazblogged! perl6advent++ 'Failure()<0xb5383134>': Failure()<0xb5382644>
06:37 zaslon was kicked by diakopter (diakopter))
diakopter I suspect it's reporting that b/c the blog rss is not responding 06:38
pmichaud perl6advent.wordpress.com isn't responding at all for me. 06:55
oh, there it is... taking a long time though.
06:59 Baggio_ left 07:04 Baggio_ joined
carlin Oops, that wasn't good 07:10
07:17 zaslon joined 07:18 zaslon left, colomon left, agentzh left 07:21 zaslon joined
carlin zaslon: add fakitymcfakefake blahblahfake.com/notarealrssfeed/ 07:22
07:22 zaslon left 07:23 zaslon joined
carlin ah well, hopefully I 'fixed' it, we'll know if it happens again :-) 07:23
07:37 r0bby left 07:39 pnate left 07:45 iblechbot joined 08:01 mberends joined 08:08 nihiliad left 08:18 agentzh joined 08:21 Baggio_ left 08:25 xinming_ left 08:31 Baggio_ joined 08:37 iblechbot left, iblechbot_ joined
mberends returns after a Debian hibernate, during which (forgetfully) the laptop re-booted into an Ubuntu partition, updated its Linux kernel, and then returned successfully to the hibernated Debian desktop. That's impressive state management :) 08:39
08:52 mikehh joined 08:59 tann joined 09:01 xinming joined, kaare joined 09:02 kaare is now known as Guest41567 09:04 mikehh_ joined 09:05 xinming left
pugs_svn r29270 | colomon++ | [perl6advent] Choose my second topic. 09:06
09:07 xinming joined
tann rakudo: my @a = 1..5; my @b = 6..10; my @c = @a ››+‹‹ @b; say @c.perl; 09:08
p6eval rakudo 7ef386: Confused at line 2, near "\u203a\u203a+\u2039\u2039 @b; "␤in Main (file <unknown>, line <unknown>)␤
09:13 SirKay left
ng_feed rakudo-ng: colomon++ 09:18
rakudo-ng: Reactivate Int.Complex and Int.unpolar.
rakudo-ng: colomon++
rakudo-ng: Reactivate Num.unpolar and Any.cis.
09:18 JimmyZ joined 09:19 mikehh left, IllvilJa joined
tann rakudo: my @a = 1..5; my @b = 6..10; my @c = @a »+« @b; say @c.perl; 09:21
p6eval rakudo 7ef386: [7, 9, 11, 13, 15]␤
dalek kudo/ng: 2904c6b | (Solomon Foster)++ | src/core/Int.pm:
Reactivate Int.Complex and Int.unpolar.
09:22
kudo/ng: 269b081 | (Solomon Foster)++ | src/core/ (2 files):
Reactivate Num.unpolar and Any.cis.
09:24 JimmyZ left 09:28 am0c joined 09:45 ashizawa left 09:47 ashizawa joined 09:56 colomon joined
pugs_svn r29271 | colomon++ | [t/spec] Split complex.t into complex.t and unpolar.t. Tweak and fudge until unpolar.t passes in ng. 09:56
ng_feed rakudo-ng: colomon++ 09:58
rakudo-ng: CHEAT: Quickly add simple N1 ... N2 operator which generates a list.
rakudo-ng: colomon++
rakudo-ng: Add unpolar.t to spectests.
dalek kudo/ng: 1ffe542 | (Solomon Foster)++ | src/ (2 files):
CHEAT: Quickly add simple N1 ... N2 operator which generates a list.
10:02
kudo/ng: 3968cf1 | (Solomon Foster)++ | t/spectest.data:
Add unpolar.t to spectests.
10:11 colomon left 10:33 Baggio__ joined 10:38 Baggio_ left 10:48 tann left, rgrau joined 10:50 hcchien left 10:52 iblechbot_ left 10:56 hcchien joined 11:09 Baggio__ left 11:11 Baggio_ joined 11:31 pmurias joined 11:42 meppl joined 11:46 agentzh left 11:57 meppl left 12:00 JimmyZ joined 12:02 Su-Shee joined
Su-Shee good morning. 12:02
JimmyZ good morning, Su-Shee 12:03
mberends good St. Nicholas, Su-Shee 12:04
Su-Shee hm, I shouldn't have put my socks into the washing machine.. ;) 12:06
mberends aahh, now your gifts are wet :-( 12:07
Su-Shee but clean! ;) 12:08
12:17 snearch_ joined
pugs_svn r29272 | pmurias++ | [viv] add missing VAST::Tight_and 12:21
r29273 | pmurias++ | [mildew] progress on porting to new STD
mberends oops. irclog.perlgeek.de/perl6/2009-12-06 has just lost the logs from 00:00 to 11:59 UTC. Could have sworn they were there 30 minutes ago :-( 12:24
12:25 Baggio_ left 12:44 Baggio_ joined 12:46 iblechbot joined 13:05 colomon joined 13:10 colomon_ joined, colomon left, colomon_ is now known as colomon 13:22 Baggio_ left 13:24 Baggio_ joined 13:32 jan_ left 13:34 araujo left, meppl joined 13:36 Baggio_ left 13:38 slavik left 13:43 Baggio_ joined 13:47 slavik joined 13:49 JimmyZ left 13:57 Baggio_ left 13:59 Baggio_ joined
pmurias TimToady: the VAST nodes for methodcall contains the method name and argument only in . 14:01
ng_feed rakudo-ng: colomon++ 14:03
rakudo-ng: Move Int.unpolar to Any.unpolar, relax constraints on the type passed as the argument.
14:04 JimmyZ joined 14:05 snearch_ left
dalek kudo/ng: 4d71a27 | (Solomon Foster)++ | src/core/ (3 files):
Move Int.unpolar to Any.unpolar, relax constraints on the type passed as the argument.
14:06
pugs_svn r29274 | colomon++ | [t/spec] Methodically test Int.unpolar, Rat,unpolar, and Num.unpolar.
14:09 colomon left 14:17 SmokeMachine joined 14:20 he_ left, pmurias left, hcchien left, am0c left, xinming left, mberends left, dbrock left, PacoLinux left, pointme left, szabgab left, dalek left, c1sung left, xomas_ left, Wolfman2000 left, renormalist left, obra left, Intensity left, yves left, sjn left, mathw left, krakan left, draxil42 left, yahooooo left, bloonix left, idemal left, wolverian left, avuserow left, awwaiid left, sunnavy left, stepnem left, LionMadeOfLions left, pmichaud left, carlin left, ruz left, mofino left, moritz_ left, omega__ left, meteorswarm left, jrockway left, frew left, zostay left, charsbar left, rhr left, sundar left, hicx174 left, brody left, arnsholt left, gabiruh left, pjcj left, sjohnson left, revdiablo left, PerlJam left, lambdabot left, ilbot2 left, parduncia left, arthur-_ left, Helios left, japhb left, Tene left, jjore_ left, ingy left, ng_feed left, gbacon left, Woodi left, pugs_svn left, spinclad left, hatseflats left, dukeleto left, jiing left 14:22 arnsholt joined, pmurias joined, hcchien joined, am0c joined, xinming joined, mberends joined, dbrock joined, PacoLinux joined, pointme joined, he_ joined, xomas_ joined, szabgab joined, Wolfman2000 joined, renormalist joined, dalek joined, c1sung joined, Intensity joined, meteorswarm joined, Maddingue joined, gabiruh joined, pjcj joined, ilbot2 joined, revdiablo joined, sjohnson joined, lambdabot joined, irc.freenode.net sets mode: +ovvv arnsholt dalek ilbot2 lambdabot, PerlJam joined, IRSeekBot joined, frettled joined, hicx174 joined, rhr joined, brody joined, charsbar joined, zostay joined, sundar joined, frew joined, jrockway joined, ruz joined, obra joined, yves joined, sunnavy joined, krakan joined, wolverian joined, carlin joined, irc.freenode.net sets mode: +ovoo PerlJam IRSeekBot frettled carlin, idemal joined, stepnem joined, avuserow joined, bloonix joined, mofino joined, omega__ joined, awwaiid joined, moritz_ joined, mathw joined, draxil42 joined, LionMadeOfLions joined, sjn joined, yahooooo joined, pmichaud joined, irc.freenode.net sets mode: +ooo moritz_ mathw pmichaud, TimToady joined, parduncia joined, arthur-_ joined, Helios joined, japhb joined, Tene joined, jjore_ joined, ingy joined, ng_feed joined, gbacon joined, Woodi joined, Infinoid joined, cls_bsd joined, dukeleto joined, pugs_svn joined, irc.freenode.net sets mode: +ovov TimToady ng_feed dukeleto pugs_svn, hatseflats joined, jiing joined, phenny joined, spinclad joined, irc.freenode.net sets mode: +vo phenny spinclad 14:24 zaslon left, zaslon joined, carlin sets mode: +vv zaslon pointme 14:38 payload left
Wolfman2000 *yawn* morning 14:43
mberends morning Wolfman2000, almost everyone is away today 14:48
Wolfman2000 mberends: kind of figured
14:54 am0c left 14:55 jaldhar left, JimmyZ left
pmurias mberends: hi 14:55
15:02 vamped joined
vamped mberends: ... so what happened from 0:00 to 12:00 ? 15:04
mberends hi pmurias, I'm beginning to get familiar with mildew-js :) nice work!
15:05 arthur-_ left
mberends vamped: there may have been some recent tweaks to the server code, with unintended side effects. but that's just a guess. 15:06
vamped hopefully nothing interesting was missed
mberends the rows are probably there in some tables, it's probably the SQL queries that are misbehaving. 15:07
carlin For it to lose exactly 12 hours it looks like a config problem 15:08
15:08 arthur-_ joined
mberends yes 15:08
15:13 quantumEd joined 15:26 jaldhar joined
pmichaud good morning #perl6 15:29
15:34 jferrero joined, jan_ joined
mberends good morning pm 15:34
vamped nice advent post, pm 15:36
pmichaud vamped: thanks! 15:37
moritz_ re 15:44
15:49 Psyche^ joined 15:50 Patterner left, Psyche^ is now known as Patterner 15:51 nihiliad joined 15:53 Baggio_ left
diakopter moritz_: you saw the irclog miss? 15:54
15:56 colomon joined 15:57 pmurias left 15:58 vamped left
colomon ng: say (1...10).join 16:06
p6eval ng 4d71a2: 12345678910␤
colomon ng: say (1...10).join(' ');
p6eval ng 4d71a2: 1 2 3 4 5 6 7 8 9 10␤
16:08 SmokeMachine left 16:09 payload joined
colomon realized last night that if we have working basic loops and gather / take, it's easy enough to write an operator that takes two ints and returns the sequence from one to the other. :) 16:09
mberends :) # nqp-rx makes easy things easy and hard things possible 16:10
16:11 envi_home2 left
colomon This was in ng. :) I'm afraid at the moment, ng makes super easy things cool, easy things hard, and hard things beyond impossible. 16:11
pmichaud having seen the many changes to the perl 6 spec over the last month, restarting development with ng was absolutely the right move. I'm not sure we could manage the transitions in the master branch. 16:17
slavik pmichaud: rakudong? 16:18
pmichaud slavik: yes
slavik will it have proper install packages?
colomon pmichaud: oh sure, I agree ng was absolutely the right thing to do. It's just a bugger to program in right now. :) 16:19
pmichaud slavik: how do you mean "proper install packages"?
slavik pmichaud: correct make install that checkinstall can make rpm/deb packages out of :P
16:19 synth left
slavik and place it into /usr/local/bin and a way to change PERL6LIB 16:19
pmichaud slavik: we already have some targets for creating rpms 16:20
PERL6LIB might not be "real" -- at the moment it's a cheat
slavik debian gets no love :(
16:20 synth joined
slavik I see 16:20
pmichaud slavik: anyway, package/install is a goal for rakudo star, not necessarily ng
slavik pmichaud: I know
it just would be easier if I could install rakudo and treat it as "perl6"
16:29 wdsw joined 16:30 wdsw left 16:37 astrojp joined 16:38 araujo joined
moritz_ diakopter: yes 16:55
Wolfman2000 ...and the OS Wars in my household has taken a new turn. I've finally convinced my dad that Mac OS X > Windows 16:57
mberends :)
Wolfman2000 We may be able to salvage a nearly broken laptop and put linux on it too. 16:58
jnthn afternoon, all :-) 17:11
Wolfman2000 afternoon jnthn.
In a few hours I'll prepare my Day 8 post just in case mathw isn't able to do Day 7 17:12
mathw it's done 17:13
phenny mathw: 03:12Z <colomon> tell mathw I had my Day 7 post ready to go this evening, so swapped with you. You're up for day 7 now. Hope that is okay.
mathw scheduled in
Wolfman2000 nice timing mathw
mathw I was rather startled by colomon's swapping
Wolfman2000 what's your time zone again?
mathw GMT
Wolfman2000 the usual goal for us has been to get out the posts within the first hour of the new GMT day
mathw yes I should really have written it and scheduled it last night 17:14
Wolfman2000 it's alright...don't worry about it
I'll make sure to proofread it soon
Wolfman2000 is presently installing Ubuntu on his VirtualBox 17:15
mathw please do, there may be a stupid mistake or two in there
17:15 colomon left, mikehh_ left 17:16 mikehh_ joined, colomon joined 17:23 justatheory joined, brrant joined
Wolfman2000 VirtualBox++ for working rather well 17:28
17:31 colomon left, colomon joined
colomon jnthn, pmichaud, etc: I'm still getting very frequent crashes in ng. 17:31
If someone can clue me in on how to build with GDB debugging info in parrot, I think I can probably produce some useful GDB backtraces and valgrind runs. 17:32
17:33 brrant left
mberends colomon: add -ggdb to the gcc compiler options 17:36
arnsholt colomon: If you get the help info from Parrot's config script there's an option to enable debugging symbols 17:37
jnthn colomon: I think we build an optimized Parrot by default on some platforms, which probably disables debugging info.
arnsholt Then you can pass that to Rakudo's configure
jnthn Well, it may do anyways... 17:38
jnthn knows on his platform that it can be a case of choosing debug info or optimized build.
17:39 colomon left, colomon_ joined, payload left, mikehh_ left 17:40 synth left 17:44 colomon joined, colomon_ left, mikehh_ joined
colomon Thanks, guys, I'll give that a try now. 17:44
Errr, I'll give that a try after I try to add just a couple more sensible unpolar.t tests, and get a bus error again. 17:47
--debugging=0 Disable debugging, default = 1 17:51
Does that mean -- debugging=0 disables debugging, or --debugging=1 disables debuging? 17:52
is parrot build with debugging by default?
17:55 isawallama joined 17:56 mikehh_ left
slavik will it be possible to get thread/heap dumps of memory that parrot has allocated? 17:58
17:58 mikehh_ joined
isawallama Brooke should know. 18:00
colomon Hmmm, definitely looks like the parrot makefile I have does not have debugging: CFLAGS in parrot/Makefile includes -DDISABLE_GC_DEBUG=1 -DNDEBUG -O3 18:01
18:03 isawallama left 18:05 abbe joined
slavik and also limit how much meamory parrot will allocate? 18:05
abbe hi everyone
anyone running rakudo on freebsd ?
18:07 mikehh_ left
mberends colomon: fwiw, -NDEBUG causes all assert() macros to "disappear", ie output no code at all. 18:07
colomon guess I should ask on #parrot, eh? 18:09
abbe gcc is segfaulting for me on 8.0-RELEASE (amd64)
jnthn abbe: That's a bug in gcc. :-( There is a workaround, but I forget it...I'm sure it came up here a month or so back. 18:13
abbe jnthn: yeah, I was the one who mentioned it last month. I've forgot to followup on that :(
18:14 mikehh_ joined
abbe jnthn: Could you please look it in your archives ? 18:14
jnthn abbe: I guess it'll be searchable in the irclogs 18:15
abbe One of my friend was able to build it on his dual-quad core xeon box, but I'm not able to build it on my core 2 duo boxen.
jnthn pointme: irclogs
pointme Sorry, I don't know anything about irclogs
jnthn pointme: irclog 18:16
pointme Sorry, I don't know anything about irclog
jnthn aww
abbe oh okay, i got it from /topic
jnthn oh!
:-)
abbe: I don't have any archive of my own besides that one. :-)
abbe thanks jnthn :) 18:17
moritz_ diakopter: yes
18:21 mikehh_ left
mberends abbe: searching irclog seems not to find the issue, but it was something about freebsd 8.0 having a slightly outdated gcc (4.2x) with a bug. your best option would be to somehow install a newer gcc. 18:25
abbe mberends: yes, I know freebsd has 4.2.1, i'm wondering if there is some workaround patch which I can apply to rakudo source for that 18:26
I want to avoid depending on GCC, if possible at all.
mberends abbe: right, but probably no workaround documented :-( 18:28
abbe It seems some logs are wiped off 18:30
18:30 mikehh_ joined
abbe irclog.perlgeek.de/parrot/2009-11-04 [2001:4860:b006::84]/search?q=cache...=firefox-a 18:30
moritz_ wonders wtf is going on there 18:33
it's still in the DB 18:34
18:38 mikehh_ left
jnthn Maybe the query is a bit queer. :-) 18:44
moritz_ I'll dump the database and try it locally 18:45
abbe lol 18:46
18:48 mikehh_ joined 18:51 tann joined 18:53 krakan left 18:54 krakan joined 18:55 mikehh_ left 19:01 hicx174 left
moritz_ somehow all those rows were marked as spam 19:13
Wolfman2000 ...I think git's make test takes longer than make spectest for rakudo right now 19:15
I don't know if that's an accomplishment or not
mberends it's strange, but why compile git from source? is this your Ubuntu install? 19:16
Wolfman2000 mberends: yes. 19:17
as far as why...well, I'm not in that big of a rush.
lisppaste3 colomon pasted "unpolar.t crash" at paste.lisp.org/display/91689 19:18
colomon now with better debugging info....
arnsholt Hmm. 0x48 looks like a weird memory address 19:20
Perhaps an unitialised variable?
jnthn colomon: Hmm. I think I saw another stack trace not so long ago that also looked like that. :-/ 19:22
colomon also from me, or... ?
jnthn colomon: Maybe.
colomon I've dumped up a bunch over the last week or so. 19:23
jnthn colomon: It's not blindingly obvious from the stack trace where exactly the problem is.
colomon: The interesting bit is where we've been leading up to that point.
arnsholt Good question. I suppose it could be, but my first guess would be that this comes from inside Parrot
colomon I've still got gdb sitting open at the crash, so if there's anything I can run there that would help you...
moritz_ colomon: just read your calendar post - colomon++
arnsholt colomon: How big is the function it fails in?
jnthn arnsholt: It does, right.
colomon moritz_: danke.
moritz_ colomon: just one minor nit: (1, 2, 3)».++ isn't supposed to work 19:24
jnthn arnsholt: The question is, did it go somewhere in Rakudo beforehand.
moritz_ colomon: ++ needs an lvalue, and a literal number is not writable
colomon moritz_: I was wondering about that last night.
no one was on-line who knew.
moritz_ you could write that >>.succ though
jnthn arnsholt: And did something we did in Rakudo break it.
arnsholt: Or is it completely a Parrot bug.
Wolfman2000 and my final results of make testing git... 19:25
colomon jnthn, arnsholt: I think it's a rakudo ng bug.
Wolfman2000 2 fixes, 4656 successes, 11 brokens
arnsholt jnthn: If it's an unitialised variable I think it has to be a Parrot bug
colomon The crashes in the file seem to depend on what is compiled later on in the file, if that makes any sense.
moritz_ rakudo also has some C code, which could be buggy
arnsholt From the stack trace, c isn't an argument to the function, and the return value is c->current_sig
colomon In this case, the file worked, I added a second case, and now it dies before the first case finishes. 19:26
carlin abbe: I also get a segfault compiling rakudo on 8.0
abbe carlin: you know of that workaround...?
arnsholt colomon: Anyways, try to see if c being uninitialised is possible with the flow of the function
lisppaste3 colomon annotated #91689 "code that's failing" at paste.lisp.org/display/91689#1 19:27
abbe carlin: I'm trying to port it to freebsd...
colomon With just the first is_approx there, the code works. With the second one added, the code fails BEFORE the first one has printed out its test message.
And as I said, this sort of thing has been happening to me a *lot* in ng. 19:28
arnsholt: "c"?
arnsholt colomon: From the gdb trace you posted, it failed on this statement: "return c->current_sig;"
19:29 constant left
arnsholt And the error was a protection failure trying to access address 0x00000048 19:29
Which means that c has to point at 0x48, which is an invalid memory location 19:30
abbe A work around I came across is removing all -W switches, and just keeping '-Wall', but wondering if it generates good code ?
colomon The previous line is Parrot_Context const *c = get_context_struct_fast(interp, ctx);
abbe good as in safe, reliable.
colomon I'm pretty much completely ignorant of parrot internals, so I've no idea if that is supposed to return null or not. 19:31
arnsholt: if it would help, I can rerun that code under valgrind and see what it says.
arnsholt colomon: Might be worth a try. Hopefully it'll complain about jumps depending on uninitialised bytes somehwere in get_context_struct_fast() or below it 19:32
colomon firing it up now, usually takes a few minutes.
dalek ok: 1b7ec89 | jnthn++ | src/roles.pod:
[roles] Start to fill out the roles chapter a bit.
19:36
colomon unfortunately, parrot's gc fires off dozens of errors under normal conditions...
arnsholt Heh 19:37
19:37 hicx174 joined
arnsholt Which parrot rev is this? 19:38
19:38 colomon_ joined, colomon left, colomon_ is now known as colomon
lisppaste3 colomon annotated #91689 "end of valgrind on the unpolar error" at paste.lisp.org/display/91689#2 19:39
arnsholt In case you didn't get my last: Which Parrot rev is this?
colomon 42822 19:40
moritz_ colomon: try a newer parrot version
colomon: since that revision a branch was merged...
and the branch had somethiing to do with calling conventions
arnsholt Sounds like a plan 19:41
colomon any advice for a particular version to try?
19:41 r0bby joined, constant joined
moritz_ when in doubt, use HEAD 19:41
jnthn ;-)
A good general principle. :-) 19:42
jnthn afk for a bit.
colomon all right, then, HEAD it is. 19:43
yow, that brought in a *lot* of changes. now I'm scared...
19:44 ShaneC joined
moritz_ our multi method ord() is export { ... } # quiz question: what's the problem with this code? 19:48
19:48 snearch joined
colomon does it need the self argument name to be is export? 19:49
moritz_ A: in the sub version that's generated from that, the first argument is anonymous
right
colomon \o/
of course, probably have that all over core at the moment....
I have a vague memory the syntax to name the first argument doesn't work in ng yet. 19:50
moritz_ ng: class A { method b($foo:) is export { say $foo } }; A.new.b()
p6eval ng 4d71a2: Not enough positional parameters passed; got 1 but expected 2␤current instr.: 'perl6;A;b' pc 255 (EVAL_1:112)␤
colomon and we were certainly bad about doing that in master, too.
arnsholt Does moving to head fix it? 19:51
colomon arnsholt: still compiling. 19:52
arnsholt 'k
If not, I might have something worth checking out
colomon my first build actually failed, so I'm trying to start over again, hoping it was a glitch in my ad hoc build process rather than a parrot / rakudo incompatibility. 19:53
colomon crosses his fingers for rakudo's make... 19:56
we have a build 19:59
we also have a small boy insisting I read board books to him. 20:00
well, that made that version of the crash go away. 20:02
let me play with things a bit and see if it pops back up again or not.
arnsholt 'k 20:03
moritz_ does dalek report from the ng branch facefully? 20:04
colomon facefully?
ng_feed rakudo-ng: moritz++
rakudo-ng: Merge branch 'ng' of [email@hidden.address] into ng
rakudo-ng: moritz++
rakudo-ng: make ord() and chr() work with named args
colomon ooo, now I need to see how you did it. :) 20:05
moritz_ by cheating, ,of course!
colomon :)
dalek kudo/ng: 63caed2 | moritz++ | (14 files):
Merge branch 'ng' of [email@hidden.address] into ng
kudo/ng: 7fc4854 | moritz++ | src/core/Any- (2 files):
make ord() and chr() work with named args
colomon any objections to my checking in the parrot bump? 20:06
moritz_ no objection here
colomon so you got rid of the is export and added a proto instead?
moritz_ right
20:06 ng_feed left
rgrau more or less, how many tests does ng pass? any idea ? 20:06
moritz_ rgrau: I'll tell you after a test run :-) 20:07
colomon it just ran 1669 on mine, but I don't know how many of those are fudged.
rgrau thanks, it's only to know more or less how far is rakudo master branch 20:08
pugs_svn r29275 | colomon++ | [t/spec] Test method forms of cis.
moritz_ 1514 tests pass :-) 20:09
colomon rgrau: it's still pretty far (as you might gather from the count) but there's probably only 20 or so big things to fix and then the rest should fall into place pretty quickly. 20:10
moritz_ like, iterating ranges :-)
carlin abbe: remove -Wmissing-include-dirs 20:11
rgrau yeah, I read something about that in jnthn and masak's posts some time ago. too many yaks to shave at first
20:11 cognominal left
colomon moritz_: don't know if you saw or not, but I implemented a crude Int ... Int overnight, which I suspect will help for a lot of test file uses of Range. 20:12
20:12 cognominal joined
pugs_svn r29276 | moritz++ | [t/spec] unfudge split-simple2.t for rakudo-ng 20:12
r29276 |
r29276 | Somehow it blows up with the fudges, but works fine without them. Huh.
colomon moritz_: hopefully bumping parrot has gotten rid of that issue too. :) 20:13
abbe aha, carlin :-) danke
colomon running tests now...
moritz_ Could not find non-existent sub &fail 20:14
I "love" those
moritz_ hopes Tene++ gives fail() a shot 20:15
Wolfman2000 hey guys! We're famous now! 20:16
developers.slashdot.org/story/09/12...t-Calendar <-- We're on Slashdot now!
colomon no freaking way. 20:17
!!!!!!!!!!!
Wolfman2000 excited now? 20:18
moritz_ woot
moritz_ wonders if the site is still reachable...
Wolfman2000 presently is
colomon it's wordpress, they've got to be able to handle a slashdotting, don't they?
Wolfman2000 colomon: guess we'll find out 20:19
in the meantime, have to wrap up some other things. brb/afk
colomon well, that should help our number of hits for today. 20:20
moritz_ it did already
20:20 explorer__ joined
moritz_ 1.8k hits today, which is the max so far 20:20
dalek kudo/ng: 0a236ac | (Solomon Foster)++ | build/PARROT_REVISION:
Bump parrot revision to head in an attempt to get around the random crashes I have been experiencing.
20:23
carlin It took a whole 4 minutes between the story being posted and a Duke Nukem joke ...
moritz_ not surprised 20:24
colomon you know, if SDL was working on perl 6, it would be cool to program a little game in p6 and dub it Duke Nukem Forever.... 20:26
colomon is now imagining a DNF adventure game in p6 for the Book.... 20:27
moritz_ don't
20:27 jaldhar left
moritz_ adventure games are way too long for a book 20:27
20:30 jaldhar joined
arnsholt moritz_: But it'd be puntastic! ;) 20:30
colomon It could be a very small adventure....
we could explain the delay in release of Duke was due to waiting for perl 6 to be ready... 20:31
moritz_ except that DNF was discontinued 20:32
20:34 tarski joined
tarski where can i get perl6 20:34
rakudo, parrot, pugs, what is the best to use right now in terms of stability?
colomon rakudo. 20:35
moritz_ rakudo is being actively developed, in contrast to pugs
and it uses parrot under the hood
tarski: rakudo.org/how-to-get-rakudo
tarski thanks 20:36
moritz_ you're welcome
colomon Advent calendar appears to be getting about 1800 hits an hour at the moment... 20:41
arnsholt Not bad. Let's hope it gives a bump in interest 20:42
colomon also, pmichaud's metaoperator post has picked up two "isn't this apl?" comments already. 20:43
carlin The important thing is that it lets people know Perl 6 is still alive and kicking 20:57
tarski building perl6 from source now :)
moritz_ colomon: are you aware of any low-hanging fruit in ng? 20:58
(or semi-LHF)
colomon one sec, taking out garbage 21:04
21:08 nihiliad left 21:11 astrojp left
colomon okay. 21:12
off the top of my head, we could probably win back a bunch of tests by systematically doing what you did to ord to other functions. There are fudged named argument tests all over the test suite, on functions that otherwise work. 21:13
Actually, "win back" isn't appropriate for a lot of those, which never worked in rakudo as far as I know. 21:14
need to change boy, will think more while doing it.
21:16 infrared joined
infrared installing rakudo on fbsd, how should i specify the parrot location? 21:17
mberends infrared: --parrot-config=<path>, but by you question you probably haven't installed parrot 21:19
*your 21:20
infrared mberends: i installed it in /usr/local/parrot (non root)
mberends infrared: then confirm you have /usr/local/parrot/bin/parrot_config 21:21
21:21 ShaneC left
infrared i do 21:21
make PARROTCONFIG=/usr/local/parrot/bin/parrot_config install clean 21:22
is what i'm attempting
mberends rather try this:
moritz_ perl Configure --parrot-config=/usr/local/parrot/bin/parrot_config
make install
mberends :)
arnsholt Telepathy =D 21:23
mberends er, Configure.pl
moritz_ right
mberends just checking, the parrot install and install-dev are the same, right? 21:24
moritz_ install = install-dev + install-base or something
mberends :)
moritz_ s/base/bin/ 21:25
infrared thank you
21:26 alester joined
abbe thanks carlin and jnthn :) 21:27
Wolfman2000 Alright, time to prepare my posting...and honestly, I'll need some assistance for part of it. 21:28
colomon Wolfman2000: mathw's post is ready to go for today, it appears. 21:29
Wolfman2000 colomon: right. Well, I'll get mine ready now for tomorrow then
colomon mathw: I just spent day 6 telling people not to do @a>>.say. :)
mathw: also, in the ^$n paragraph, the last character is lost. 21:30
Wolfman2000 ...I may be able to use >>. somehow. At least, if I can figure out how to change my .map version
colomon mathw: other than that, very nice post! methodical and clear, IMO.
carlin abbe: np, it certainly is a weird gcc bug 21:32
moritz_ "This could have been written using a bare <code>say</code> subroutine call, but I prefer the method call syntax." 21:34
not true.
not in Perl 6
abbe carlin: www.freebsd.org/cgi/query-pr.cgi?pr=ports/139010 www.freebsd.org/cgi/query-pr.cgi?pr=ports/139012 the updated PR will appear here in few minutes.
moritz_ mathw: mind if I remove that sentence? 21:36
mathw it's not true? 21:37
moritz_ rakudo: for 1..4 { say }
p6eval rakudo 7ef386: say requires an argument at line 2, near "}"␤in Main (file src/gen_setting.pm, line 2610)␤
mathw cool
I like Perl 6 a little more
moritz_ oh, one more thing
^@array.elems Z @array
can even be written nicer
as @array.kv
mathw I think that requires a new paragraph 21:38
as it's a progression from the original idea
so I'll add one for it when you've removed that sentence
and as for not using >>.say, it should work fine if your I/O system thinks in lines, and you don't care about ordering... 21:39
although >>.say is likely to be rare, ithink
>>.something-else-which-doesnt-do-IO is more likely I think :)
21:40 cspencer joined, Guest41567 left
Wolfman2000 ah, here's what I was looking for... [sourcecode light="true"] 21:40
moritz_ mathw: I'm storing the post right now, but it takes ages :/ 21:41
mathw booooooo
slow bad
but we can cope
moritz_ that's the /. effect
21:41 lmc left
moritz_ still I'm amazed the site is still up :-) 21:41
done 21:42
Wolfman2000 moritz_: preparing my draft as you guys speak
haven't been affected that much
moritz_ mathw: I also changed ^@array.elems Z @array to @array.kv
mathw as I said, iwas going to add another paragraph for that
moritz_ oh, I thought you meant the other thing (empty say) 21:43
mathw heh
I'll sort it out
moritz_ mathw++
21:43 abbe left
moritz_ @karma mathw 21:44
lambdabot mathw has a karma of 14
mathw gosh
how did that happen
moritz_ mathw++ # like this
mathw yes I know that 21:45
21:45 Su-Shee left
moritz_ :-) 21:45
mathw but I'm surprised it's been that much since I changed my nick
21:47 explorer__ left
Wolfman2000 ...I hope I didn't mix up all of the purposes of .comb now... 21:50
rakudo: my @tmp = "Hello".comb; say @tmp[1];
p6eval rakudo 7ef386: e␤
Wolfman2000 ...probably have to recompile rakudo or something
rakudo: say "Hello".comb; 21:51
p6eval rakudo 7ef386: Hello␤
moritz_ rakudo: say "Hello".comb.join('|')
p6eval rakudo 7ef386: H|e|l|l|o␤
Wolfman2000 moritz_: I know of that too
moritz_ rakudo: say "Hello".comb(/<[ao]>/).join('|')
p6eval rakudo 7ef386: o␤ 21:52
Wolfman2000 rakudo: say "Perl 6 Advent".comb(/<alpha>/).join('|'); 21:55
p6eval rakudo 7ef386: P|e|r|l|A|d|v|e|n|t␤
moritz_ rakudo: say "Perl 6 Advent".comb(/<alpha>+/).join('|'); 21:56
p6eval rakudo 7ef386: Perl|Advent␤
Wolfman2000 moritz_++: Another good one.
But I don't want to deal with too many regex patterns yet
That's a gift for another day. :)
21:57 cspencer left
Juerd_ I finally found out why I don't like python's ":".join(list) 21:57
Wolfman2000 Juerd_: counterintuitive?
Juerd_ You can't do foo.split(":").join(":").split(":").join(":")...
So here's a rule: opposite operations should be stackable as methods :) 21:58
Tene moritz_: would a simple &fail that just throws an exception be okay for now? 22:00
could be caught with try { ... }
22:01 synth joined
moritz_ Tene: making it an alias for return Mu; would be a better stopgap solution, I think 22:01
Tene Oh?
Okay, I can do that.
slavik Juerd_: python in general is proud to be dumb
GIL?! 22:02
Tene slavik: rakudo can't do threads *at all*
slavik Tene: not yet :P
Juerd_ slavik: Python is certainly not dumb.
slavik Tene: what's the point of having threads if they aren't concurrent? why not just go back to Solaris ...
Juerd_ slavik: It has a few design bugs. So does Perl 5. And, as I'm sure we'll eventually discover, so will Perl 6.
slavik Juerd_: GIL is there by design 22:03
Juerd_ I don't know what GIL is
Do you know what a design bug is? :)
Tene global interpreter lock
22:04 jan_ left
slavik ie: no thread concurrency 22:04
22:04 pmurias joined
Wolfman2000 rakudo: say "5065726C36".comb(/<xdigit>**2/).map: { chr(:16( $_ ) ) } ).join(''); 22:04
Juerd_ Design bugs are, by definition, by design :)
p6eval rakudo 7ef386: Syntax error at line 2, near ").join('')"␤in Main (file <unknown>, line <unknown>)␤
slavik also ... OpenGL in Python is 50% slower than C/Perl for a very simple reason ...
22:04 araujo left
Wolfman2000 ...BAH 22:04
slavik for every opengl call, there is a call to glError() 22:05
Wolfman2000 rakudo: say "5065726C36".comb(/<xdigit>**2/).map: { chr(:16( $_ ) ) } );
Wolfman2000 copied too much
p6eval rakudo 7ef386: Syntax error at line 2, near ");"␤in Main (file <unknown>, line <unknown>)␤
slavik the red book explicitly says not to do this.
Wolfman2000 rakudo: say "5065726C36".comb(/<xdigit>**2/).map: { chr(:16( $_ ) ) } ;
p6eval rakudo 7ef386: Perl6␤
Juerd_ Wolfman2000: Did you know the bot also responds to private messages? :)
Wolfman2000 Juerd_: ...that may be useful.
slavik I also dislike the almost java like + for string cat 22:06
it should be either explicit like Perl or automatically converted like Java
Juerd_ sighs
Wolfman2000 Python is an interesting beast.
moritz_ slavik: did you mean javascript?
slavik "val = " + str(some_val) + "\n" is messed up IMO
Juerd_ Context is great. Most languages don't have it. 22:07
slavik moritz_: javascript as well. I mean how Java does an auto .toString() for any object that you try to cat to a string
Juerd_ dog
carlin rakudo: say "5065726C36".comb(/<xdigit>**2/).map: { chr '0x' ~ $_ }; # Wolfman2000
p6eval rakudo 7ef386: Perl6␤
Tene slavik: Please try to keep the language-bashing down a bit.
22:08 wayland76 joined
slavik like: int a = 5; System.out.println("" + a); will wrap the 'a' in Integer class and call .toString() on it 22:08
Tene: I am done :P
Wolfman2000 carlin++: ...I think I like that even more.
Tene We all do it a bit, but it's not quite the spirit we want for our community. :)
Wolfman2000 I'm going to borrow that for my post tomorrow.
22:10 iblechbot left 22:11 foudfou joined
Wolfman2000 To all Perl 6 Advent admins and editors: my Day 8 post is ready for critiquing. 22:11
...ah. So THAT'S how I was supposed to do it...? 22:13
alright, which to PM
foudfou hi all, I was wondering: is there a chance to see the statement ending semi-colons as optional ? 22:14
moritz_ foudfou: no
foudfou :-)
moritz_ foudfou: only those preceeding a curly brace at the end of a line
22:14 jonathanturner joined 22:15 zamolxes left
jonathanturner The Perl 6 Advent thing was a cool idea. I'm sure it's getting a lot of people playing with the current Perl6 builds 22:15
Tene foudfou: it's not likely you'll win much support for that as a lgobal language change, but you could certainly write a module that implements that change in a defined lexical scope with macros.
Wolfman2000 ...alright, I need more assistance from the experts. Presently, I'm using .map { method } over >>.method. I'm trying to figure out the right way of implementing it with my upcoming post. 22:16
jonathanturner: If nothing else, we've got a lot of people now due to Slashdot
Tene foudfou: use OptionalSemis; # Semicolons are optional after this, until the end of the scope
Wolfman2000: you need a leading . on method
moritz_ jonathanturner: nearly 3k visitors from slashdot today, YaY
Tene rakudo: (1,2,3).map { .say } 22:17
p6eval rakudo 7ef386: Confused at line 2, near "{ .say }"␤in Main (file <unknown>, line <unknown>)␤
Tene rakudo: (1,2,3).map: { .say }
p6eval rakudo 7ef386: 1␤2␤3␤
jonathanturner moritz_: nice
Wolfman2000 Tene: I'm trying to avoid using map actually
22:17 nihiliad joined
Tene Ah. 22:17
Wolfman2000 I want to use the hyperoperator.
but in this case, I need to use...well, let me PM you
jonathanturner just had a quick question on the zip operator... if I do: ([1, 2, 3] Z [4, 5, 6]).perl.say I get [1, 4, 2, 5, 3, 6] instead of [(1, 4), (2, 5), (3, 6)]. is this flattening normal, or am I just reading it wrong? 22:18
moritz_ jonathanturner: it's a limitation in rakudo right now
22:18 snearch left
moritz_ jonathanturner: it doesn't implement slices yet, which would be needed for the correct semantics 22:19
jonathanturner moritz_: how involved is the implementation side of it? might be a fun project to help with 22:20
22:20 Limbic_Region joined
moritz_ jonathanturner: I don't really know; I guess pmichaud is the right person to ask 22:20
Limbic_Region FYI, there is an unanswered Rakudo question at PerlMonks - perlmonks.org/?node_id=811292
moritz_ Limbic_Region: I'll look at it later 22:21
22:24 zamolxes joined
moritz_ std: module Foo-1.0; 22:24
p6eval std 29276: ===SORRY!===␤Unable to parse module definition at /tmp/jGZLFG1mw9 line 1:␤------> module Foo⏏-1.0;␤ expecting any of:␤ trait␤ whitespace␤FAILED 00:01 106m␤
moritz_ rakudo: module Foo:ver<1.0>; 22:25
p6eval rakudo 7ef386: ( no output ) 22:26
moritz_ rakudo: module Foo:ver<1.0>; say 1
p6eval rakudo 7ef386: 1␤
22:26 mberends left 22:27 foudfou left 22:32 |Jedai| left 22:33 |Jedai| joined 22:35 colomon left
Wolfman2000 Tene++: Thanks for the help and reminding everyont that Perl 6 is still the language of TIMTOWTDI 22:37
22:37 colomon joined
Wolfman2000 colomon: updated my piece. Have a crack at it if you want. 22:37
arnsholt On a somewhat relevant note, does anyone know who submitted the calendar to slashdot? 22:38
arnsholt is curious =)
Wolfman2000 "timothy"
Specifically, www.monkey.org/~timothy/
colomon Wolfman2000: need to make and eat dinner, but I'll try to take a look after. 22:39
22:39 astrojp joined
moritz_ there was a Timothy Nelson on p6l some time ago, iirc 22:40
and maybe even here on irc, not sure
Wolfman2000 *shrugs* doesn't bother me.
moritz_ ah, but he usually uses wayland76 as a nick ;-) 22:41
arnsholt Ah. That name I recognise 22:42
Wolfman2000 that one I don't 22:44
22:44 r0bby left
moritz_ Wolfman2000: you should add the rot13 example that you mentioned at the end 22:45
right now it's quite a cliff hanger
Wolfman2000 moritz_: I was hesistant since I already have a lot of information
moritz_ Wolfman2000: then at least put it in a paste bin and link to it, or so
Wolfman2000 moritz_: Hmm...good points. I'll look into it after I have dinner 22:46
moritz_ (and I also fixed KyleHa's name in the post :-)
wayland76 I didn't submit to Slashdot
But I saw the article, and dropped by in case we needed help explaining things to visitors 22:47
Wolfman2000 then why does it say posted by timothy?
22:47 r0bby joined
moritz_ Wolfman2000: I think there's more than one timothy in the world :-) 22:47
wayland76 Seeing as how the name is in the Bible, it crops up in cultures all around the world where Christianity has been an influence
In this case, it appears to have been "Timothy Lord", who is not to be confused with "Timothy Nelson" (the latter is me :) ) 22:49
Wolfman2000 wayland76: thank you for clearing that up
wayland76 No worries :) 22:51
23:00 r0bby left 23:01 ssm left
jnthn Wow, the calendar made /. :-) 23:02
23:06 astrojp_ joined
Wolfman2000 jnthn: few hours too late 23:07
zamolxes jnthn: meh, I was used with the reddit and hn trolls, completely forgotten about the /. ubertrolls 23:11
jnthn Wolfman2000: Heh, I was afk for a few hours. :-) 23:13
Wolfman2000: Sorry that my comment was a dupe, but it is fitting for a comment on /. ;-) 23:14
Wolfman2000 jnthn: it's alright
23:17 astrojp left 23:19 ssm joined
jnthn checks when his entry is needed for...and is relieved to see he still has a few days to prepare :-) 23:19
23:21 r0bby joined
colomon 6570 hits on the Advent blog so far today. 23:27
moritz_ it's kinda fun watching the stats searing up :-) 23:29
23:30 Limbic_Region left 23:34 jan_ joined
wayland76 Everyone should post at least one comment; it's looking a little sparse. Maybe we should troll slashdot :). 23:41
colomon Wolfman2000: Your post looks pretty solid to me, though it might be fun to add the rot13 code too.
23:43 pmurias left 23:46 payload joined