»ö« | perl6-projects.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot: perl6: say 3;' | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by moritz_ on 27 June 2009.
00:07 frew_ joined 00:09 misssingthepoint joined 00:10 broquaint left, broquaint joined 00:11 ihrd left 00:14 mulephone joined 00:17 hoelzro left 00:19 frew_ left, frew_ joined 00:20 meppl left 00:24 hyfen_ joined, hyfen left 00:33 justatheory left 00:35 hoelzro joined 00:38 hoelzro left 00:43 mulephone left 00:49 REPLeffect left, REPLeffect joined, explorer left
TimToady well, I finally figured out that the problem with / \# foo / is that it thinks it's an unspace; likewise for backslashed space character 00:59
not sure what to do about that...
01:11 orafu left, orafu joined 01:15 frooh joined 01:16 frooh left, Limbic_Region left, japhb left 01:25 frioux joined 01:30 cai joined, frioux left
misssingthepoint pmichaud et al: ->> use.perl.org/~missingthepoint/journal/39359 <<- 01:37
01:43 lumi_ joined 01:45 lumi__ left 01:47 frioux joined 01:58 kidd_` left 02:06 kane_ left 02:09 kane_ joined
TimToady misssingthepoint: have you got a pugs commit bit yet? 02:11
if not, /msg me your email and preferred svn nick, and I can send you a commit bit 02:13
then you can fiddle with anything in the pugs repo, including tests, webpages, specs, etc.
02:14 hyfen_ left 02:15 frioux left 02:17 frioux joined 02:21 colomon left 02:25 alc joined 02:28 hyfen joined
misssingthepoint yes, i have one :) 02:29
02:29 KyleHa joined
KyleHa mtp: Did you read my meditation on how I got involved? 02:29
misssingthepoint no, where?
KyleHa perlmonks.org/?node_id=780001 02:30
That's just how I did it, not how you have to do it. 8-)
misssingthepoint KyleHa: that's very helpful. 02:34
KyleHa I'm glad you found it so.
TimToady one thing you can do right off the bat is monitor this channel when nobody else seems to be active, and be ready to make visitors from perl6-projects feel welcome (since that page points here) 02:37
misssingthepoint ok 02:38
TimToady have you done any webpage hacking?
misssingthepoint yeah
the filtering thingy for moritz_'s irc logs (new version svn'd soon) 02:39
s1n pmichaud: i had an idea for the august dallas.p6m meeting
TimToady someone needs to make the boxes on perl6-projects behave when the window is either too narrow or too wide; don't know if that appeals
misssingthepoint actually that's been my on-and-off job for a few years
TimToady: i'll do it.
TimToady cool. not sure how much refactoring is involved with that though... 02:40
misssingthepoint TimToady: a little... maybe need JS
KyleHa std: sub foo { return 1,2 }; foo() 02:41
p6eval std 27764: OUTPUT«ok 00:02 37m␤»
KyleHa std: sub foo($x, $y) { return $x,$y }; foo();
p6eval std 27764: OUTPUT«ok 00:03 40m␤»
02:42 nihiliad joined
KyleHa rakudo: sub foo { return 1,2 }; foo(); 02:45
p6eval rakudo ea667e: OUTPUT«too many arguments passed (2) - at most 1 param expected␤in sub foo (/tmp/iL0LejZj4w:2)␤called from Main (/tmp/iL0LejZj4w:2)␤»
KyleHa That ought to work, right? 02:46
02:47 dukeleto joined
TimToady rakudo: sub foo() { return 1,2 }; foo() 02:47
p6eval rakudo ea667e: OUTPUT«too many arguments passed (2) - at most 1 param expected␤in sub foo (/tmp/ByU2XymtDT:2)␤called from Main (/tmp/ByU2XymtDT:2)␤»
TimToady rakudo: sub foo() { return 1 }; foo() 02:48
p6eval rakudo ea667e: ( no output )
KyleHa rakudo: sub foo { return (1,2) }; foo();
TimToady rakudo: sub foo() { return 1 }; say foo()
p6eval rakudo ea667e: ( no output )
rakudo ea667e: OUTPUT«1␤»
TimToady rakudo: sub foo() { return 1,2 }; say foo()
p6eval rakudo ea667e: OUTPUT«too many arguments passed (2) - at most 1 param expected␤in sub foo (/tmp/mlxUPanqjl:2)␤called from Main (/tmp/mlxUPanqjl:2)␤»
TimToady looks pretty buggy to me
KyleHa Thanks. 02:49
TimToady rakudo: sub foo() { return 1,2 }; say foo().perl
p6eval rakudo ea667e: OUTPUT«too many arguments passed (2) - at most 1 param expected␤in sub foo (/tmp/5qYLHEjtTx:2)␤called from Main (/tmp/5qYLHEjtTx:2)␤»
TimToady rakudo: sub foo() { return (1,2) }; say foo().perl
p6eval rakudo ea667e: OUTPUT«[1, 2]␤»
TimToady rakudo: sub foo() { return(1,2) }; say foo().perl 02:50
p6eval rakudo ea667e: OUTPUT«too many arguments passed (2) - at most 1 param expected␤in sub foo (/tmp/Y03E4eR82n:2)␤called from Main (/tmp/Y03E4eR82n:2)␤»
TimToady looks like return wants exactly one argument
KyleHa Yes. That's why I'm sticking the test in S06-blah blah/return.t
pugs_svn r27765 | kyle++ | [t/spec] Test for RT #63912 02:55
02:58 hoelzro joined, lucs left
pugs_svn r27766 | bpetering++ | [perl6-projects.org] add a TODO list 03:02
03:03 jauaor joined, hyfen left
pugs_svn r27767 | bpetering++ | [perl6-projects.org] Add note about liberal commit bits 03:05
03:06 kst left
KyleHa rakudo: my $c = { say $_; say @_.perl }; $c(2, 3, 4, 5) 03:06
03:06 kst joined
p6eval rakudo ea667e: OUTPUT«2␤[3, 4, 5]␤» 03:06
03:07 hoelzro left
KyleHa It seems a little odd that '2' would be in both $_ and @_, but that's what the ticket says. 03:07
I guess I ought to gawk at S06. 03:08
03:10 mtp2 joined
mtp2 rakudo: say 'hi' 03:10
p6eval rakudo ea667e: OUTPUT«hi␤»
03:10 mtp2 left 03:12 hyfen joined
pugs_svn r27768 | kyle++ | [t/spec] Test for RT #63974 03:14
03:20 donaldh left, donaldh joined 03:35 KyleHa left 03:41 alester joined 03:51 dukeleto left 03:57 mtp2 joined, mtp2 left 04:00 frew_ left
pugs_svn r27769 | bpetering++ | [perl6-projects.org] Add actual links to IRC channels for friendliness' sake 04:02
04:10 nihiliad left 04:12 frioux left 04:17 kst left 04:18 kst joined
cj TimToady: did you get my note about reading the intro to the camel book while playing in the pit for The Music Man? You coincidentally cleared up a bit of confusion I was having there ;) 04:26
04:29 dukeleto joined
pugs_svn r27770 | lwall++ | [STD] disallow unspace in regex 04:31
r27770 | lwall++ | [STD] don't backtrack out of failed rx// and such
r27770 | lwall++ | [STD] rename rxinfix to regex_infix
r27770 | lwall++ | [STD] handle ; recognition in regex more robustly
r27770 | lwall++ | [STD] fix up messages related to ${...}
TimToady cj: was that an IRC note? my server blew up in the middle of the night... 04:35
04:41 cai left 04:45 japhb joined, frew_ joined, kane_ left 04:49 b2gills left
Tene rakudo: say (gather for 1..3 { take 'a',1,'b',5 }).perl 04:55
p6eval rakudo ea667e: OUTPUT«too many arguments passed (4) - 1 param expected␤in Main (/tmp/ihT29ca5FP:2)␤»
Tene rakudo: say (gather for 1..3 { take ('a',1,'b',5) }).perl
p6eval rakudo ea667e: OUTPUT«["a", 1, "b", 5, "a", 1, "b", 5, "a", 1, "b", 5]␤» 04:56
Tene Is that right? Should it really flatten like that?
That seems wrong to me.
05:06 kane_ joined
pugs_svn r27771 | bpetering++ | [perl6-projects.org] Update TODO - stuff fixed in following commit 05:09
05:09 alc left 05:22 frioux joined 05:28 frioux left
TimToady Tene: that's okay, lists flatten. slice context will preserve structure 05:35
Tene Okay.
TimToady std: rx/ x \# y / 05:36
p6eval std 27771: OUTPUT«===SORRY!===␤No unspace allowed in regex (for literal please quote with single quotes) at /tmp/8MrK8FTway line 1:␤------> rx/ x \⏏# y /␤FAILED 00:02 36m␤»
05:36 sparc joined
TimToady someone will need to fix t/spec/S05-mass/rx.t 05:37
05:42 synth joined 05:44 hyfen left 05:58 schwern joined
schwern Any November developers on? 05:58
05:59 schwern left 06:00 alc joined 06:15 rfordinal left 06:32 frew_ left
pugs_svn r27772 | bpetering++ | [perl6-projects.org] fix box layout + clean up HTML 06:39
06:45 desertm4x joined
moritz_ misssingthepoint: why do you think the width shouldn't be given in units of font size? 06:48
06:49 rfordinal joined 06:51 colomon joined
pugs_svn r27773 | moritz++ | [perl6-projects.org] link to the more canonical vim highlighting repo 06:52
misssingthepoint moritz_: mainly because it's cognitively dissonant. 06:54
moritz_ ok
06:57 Su-Shee joined
Su-Shee good morning. :) 06:57
misssingthepoint good morning Su-Shee, you usually show up smiling :)
Su-Shee well usally I'm in a good mood in the morning :) 06:58
I drink my coffee while staring outside the window and that usally leads to nice ideas. today, I though about a GtkPerl Weather Widget Set, Cairo Easy Shapes, a candy button factory based on Image Magick.. :) 06:59
misssingthepoint Su-Shee: i can so relate to that. Morning + coffee + ideas is a good combination for me too. :) 07:00
07:01 ihrd joined
misssingthepoint moritz_: kinda like having an iterator, but reading all its data into an array, then using it? 07:01
moritz_ that's cognitively dissonant, yes 07:02
07:06 ihrd left
misssingthepoint :-(= 07:07
IE exists to make life difficult for people. 07:08
moritz_ otoh in Perl 6 the arrays are lazy, so 'for $file.lines { ... }' is just fine ;-) 07:09
indeed
IE--
07:09 dukeleto left, jlaire left
misssingthepoint rakudo: say("IE-- " xx 10) 07:09
p6eval rakudo ea667e: OUTPUT«IE-- IE-- IE-- IE-- IE-- IE-- IE-- IE-- IE-- IE-- ␤»
misssingthepoint moritz_: best analogy i had :) 07:10
moritz_ ;-)
07:14 RonOreck left 07:15 RonOreck joined 07:20 donaldh left 07:22 donaldh joined 07:25 synth left 07:31 iblechbot joined 07:36 alester left, dakkar joined
Matt-W Good morning 07:37
moritz_ oh hai 07:38
07:38 colomon left
Matt-W What magic have you been up to 07:43
07:44 colomon joined
moritz_ no magic; 07:44
misssingthepoint use magic; 07:45
fix_everything();
;o)
Matt-W: hi 07:46
Matt-W hi misssingthepoint
07:46 snearch joined 07:49 FurnaceBoy left 07:53 snearch left, snearch joined
cosimo szabgab: hi, i just recovered source for LWP::Simple 07:56
07:58 fridim_ joined 08:02 dalek left, cbk left, awwaiid left, leedo left, jdv79 left, clkao left, jeekobu left, gabiruh left, Ingmar left, fridim_ left, Jedai joined, fridim_ joined, dalek joined, cbk joined, awwaiid joined, leedo joined, jdv79 joined, clkao joined, jeekobu joined, gabiruh joined, Ingmar joined
moritz_ github it! 08:04
improper use of noun where verb was expected, I know :-)
08:04 zamolxes joined
Su-Shee well, googling was made a verb either.. 08:04
I prefer "git it!"
cosimo any quick & dirty instructions? ... 08:05
moritz_ git init . 08:06
git add *
git commit -m 'initial commit'
Su-Shee cosimo: github leads you through it nicely if you click a new repository.
moritz_ then follow... what Su-Shee said :-)
Su-Shee cosimo: you'll need an ssh key.
cosimo ok, np 08:07
08:09 Juerd joined
cosimo it's really simple :) 08:10
08:12 snearch left
Su-Shee having a blog again is nice. a rant a day keeps the doctor away. ;) 08:13
szabgab Su-Shee, in german only ? 08:14
Su-Shee szabgab: no, I switched to english entirely.
szabgab I see only this sushee.schreibsturm.org/ 08:15
Su-Shee szabgab: and that's not a new movable type in light green with english in it? 08:16
szabgab cache
now it is in English 08:17
do you also write about Perl or coding ? 08:18
Su-Shee yes. but not exclusively.
08:18 colomon left
szabgab for now I see 2 out of 2 political rants :-) 08:18
Su-Shee I also post pictures sometimes. ;)
szabgab: I installed MT yesterday. ;)
szabgab so that 1 post per day rate, nice 08:19
Su-Shee let's see how annoying readers get this time. blogging is nice as long as you stay under a certain radar. 08:20
Matt-W get known too much you start getting trolls 08:21
not something I've ever had to worry about
08:21 kst left 08:22 kst joined
szabgab reddit has a few of them, you just need to find a way to ignore them 08:22
Su-Shee Matt-W: yeah. and "interesting" emails of you're female. ;) 08:23
szabgab I for example talk about them on a friendly IRC channel and usuall I get a few good words from my fellow hackers , then it is a lot easier to get over the trolls
moritz_ Su-Shee: you could publish some of the "interesting" mails and publicly ridicule them (but I don't know if that might actually encourage them...) 08:24
Su-Shee moritz_: it does usally. 08:25
moritz_: and you really don't want to fight publicly in a blog. ;)
szabgab Su-Shee, you should certainly talk about those on channels where you know you have friends
Matt-W Here is definitely a channel of friends 08:26
moritz_ Su-Shee: unless you're called fefe and disable comments
Su-Shee he never had any. ;)
well let's see. blogging good is a certain amount of work.. 08:27
moritz_ yeah, but you know what I mean
szabgab this one is really a friendly channel but in case you cannot connect to freenode you can always drop by on #padre on irc.perl.org ;-)
Su-Shee moritz_: I do. ;) 08:28
szabgab: I'm not worried after two postings. :)
and urxvt manages to somehow break my font display.. :/ 08:29
cosimo here it is, github.com/cosimo/perl6-lwp-simple/tree/master
szabgab cosimo, great, let me try to add it to proto
08:31 RonOreck left
moritz_ cosimo++ 08:31
cosimo it's just a quick hack, nothing more 08:32
08:32 RonOreck joined
cosimo my p6 skills are 0.00 08:32
szabgab added to proto project list 08:33
cosimo, do you plan to hand out commits on it or shall I fork it if I have changes ?
cosimo szabgab: of course i can hand out commits :) 08:35
do as you please
szabgab then please add szabgab on github to it
moritz_ and moritz :-) 08:36
I don't have anything to commit yet, but it certainly looks interesting enough to change that soon
szabgab afaik it should use URI module instead of hand processing the given URL
cosimo here you go, 08:37
szabgab ty
misssingthepoint moritz_: i have something to show you 08:38
moritz_ misssingthepoint: go right ahead
szabgab misssingthepoint, 20 hours per week is plenty of time :-) 08:40
misssingthepoint szabgab: that's the idea :)
Su-Shee (now I'm blogging something perl. ;) 08:41
misssingthepoint fts.petering.net/irclog/irclog.html
moritz_: that's what i was aiming for initially, but you've got to start somewhere :)
(and 'special' is broken)
(and there are rough edges, 1 bug i know about) 08:42
moritz_ misssingthepoint: that's way cool
(I don't know if that's the bug you know, but actions have a bogus  in there) 08:43
misssingthepoint (that's just a side effect of taking a snapshot; the encoding gets borked)
the bugs are: 1. special (quits/joins) NYI 08:44
2. turn filtering on, add some, turn off, see what happens :)
but you get the idea :)
szabgab Su-Shee, and the next step down the road would be to get you added to ironman.enlightenedperl.org/
moritz_ misssingthepoint: works here
misssingthepoint oh. 08:45
that's no bug... it's a feature!
dakkar misssingthepoint: the page is encoded in utf-8, but does not declare it
misssingthepoint dakkar: yep, if moritz_ approves i'll commit it and his lovely Perl will send the correct encoding for you at irclog.perlgeek.de 08:46
Su-Shee szabgab: I'm not blogging perl only, I'm really usally a mixed shop of everything from graphics to movies to open source and so on.
szabgab they filter on the perl keyword
misssingthepoint dakkar: (i.e. do the right-encoding-thing)
moritz_ Su-Shee: you can filter by tag or so
dakkar good!
moritz_ Su-Shee: if MT supports an rss feed by tag, just submit that to the ironman 08:50
misssingthepoint: anyway, you have my approval to commit that, even if not perfect yet 08:51
misssingthepoint moritz_: ta 08:52
09:09 ihrd joined, ihrd left
pugs_svn r27774 | bpetering++ | [misc/irclog] Cooler nick filtering 09:17
09:18 Guest44770 is now known as bionoid
misssingthepoint moritz_: little bit to fix, but hope you like 09:19
moritz_: and thanks for the compliment before :)
afk food
moritz_ misssingthepoint: I updated the page, and when I click on "Turn on filtering by nick" I get an empty list of nicks - any idea why? 09:21
09:45 KyleHa joined
pugs_svn r27775 | kyle++ | [t/spec] fix S05-mass/rx.t in light of r27770 09:47
KyleHa Is there anyone awake who can sanity that? 09:48
sjohnson "sanity it" ? 09:50
KyleHa Check for sanity.
A code review, one might say. 09:51
moritz_ will take a look, perhaps after lunch
KyleHa Thanks moritz. 09:52
sjohnson oh, i see
KyleHa Now I can rest easier when I go back to bed.
moritz_ std: rx/ a \# b / 09:53
p6eval std 27774: OUTPUT«===SORRY!===␤No unspace allowed in regex (for literal please quote with single quotes) at /tmp/S9ux4Vshrt line 1:␤------> rx/ a \⏏# b /␤FAILED 00:02 36m␤»
misssingthepoint moritz_: no, gimme5 09:55
moritz_ misssingthepoint: no hurry 09:56
KyleHa: looks good, thanks
KyleHa Thank YOU. 09:57
10:00 KyleHa left
Su-Shee szabgab: now I blogged massive perl. 10:01
szabgab Su-Shee, I just read the first paragraph - I am a slow reader - but we certainly need you in #padre :-) 10:04
Su-Shee I haven't even got to a threaded perl yet. :) 10:05
szabgab one step at a time, we can help you with that but I think the people here would prefer if we discussed this on #padre 10:06
Su-Shee first, I need some - any - breakfast.
szabgab if you are interested I can try to help you through the installation
I already had 3 10:07
Su-Shee I would just have to build a new perl with threads, i plainly didn't build perl with threads, because "wth do I need threaded perl for?!" ;) 10:08
szabgab Su-Shee, oh and I think it is better to link to search.cpan.org/dist/Wx as that link is stable
misssingthepoint moritz_: i suspect it's because i'm emptying a section of the DOM hackily 10:09
jnthn hi all 10:18
moritz_ moin jnthn
10:20 masak joined
masak greetings, futurologists! 10:20
10:23 pmurias joined
masak I have found a glaring bug in Rakudo that I should have found months ago. 10:24
I'm a bit surprised no-onw has caught it in the meantime. 10:25
misssingthepoint do tell.
masak yep -- gist coming up.
gist.github.com/157074 10:27
it's affecting Druid since June 8, and I haven't picked up on it since now. :/
s:2nd/since/until/
an interesting question is whether it affects STD.pm too, since I know Rakudo's mechanism for allowing statement-form declarations was copied more or less verbatim from STD.pm. 10:29
masak tests locally
10:30 |Jedai| joined
masak STD.pm seemingly doesn't have the same problem. 10:34
masak reports rakudobug
10:36 kane_ left 10:41 wayland76 left, wayland76 joined
masak seems you had fun last night. I'll need to backlog. :) 10:43
wayland76 ping bots 10:44
masak is a bot
araujo lwiz,org 10:48
lwiz.org
:D
araujo just got it up
morning all 10:49
masak y0
araujo: one can tell from your blog content that you're one of the old lambdacamel crew. :) 10:50
araujo masak, :)
10:50 Jedai left
wayland76 Lets see if masakbot has a pester command.. :) 10:53
masak: pester pmichaud about wayland76's patches to ins2 branch
:)
Hmm, no response. I guess not :)
masak wayland76: you can pester pmichaud yourself! 10:54
;)
wayland76 Yeah, I will, if I'm ever on at the same time
masak you just need to rise early enough.
wayland76 it's only 9pm here, but I think I'd like to do some reading before I go to bed
Well, the problem is, my computer crashes when it gets cold....
masak failed utterly at going to sleep last night 10:55
wayland76 ...and the earlier I get up, the longer I have to sit around waiting for the room to warm up before my computer will boot
masak: Are you a student or something? :)
Oh, wait, sleep, or failed to go to bed?
masak wayland76: sleep. 10:56
wayland76: and yes, I am a student.
jnthn hates failsleep
wayland76 I found one thing that helps failsleep 10:57
masak failsleep leads to failwake. failwake leads to general unhappiness.
wayland76 Go for a walk during the day
jnthn Me too, but it gives me a hangover the next day.
;-)
wayland76 Oh, mine helps me get up early the next day :)
Only problem is, you usually have to go for a walk a while before you go to bed
masak wayland76: I'll try that. yesterday I swam 2.5 km, so it wasn't that I hadn't got enough physical exercise... 10:58
wayland76 I sometimes to 8pm, and occasionally first thing in the morning
masak: I should think not. It has to be cardio, but swimming is that
10:58 colomon joined
wayland76 I know quite a bit about sleep, but I'm not very good at it :) 10:59
"Did you sleep well?" ... "No, I made a few mistakes" -- Steve Wright :)
masak jnthn: for the first time in a while, I have a blocker bug. rt.perl.org/rt3/Ticket/Display.html?id=67976
wayland76: I _so_ know the feeling.
11:04 alc left
masak szabgab: ping. 11:15
szabgab masak, pong 11:16
did I break proto ?
masak szabgab: reviewing your 'proto help' patch. :) no, I don't think you broke it.
testing it now.
szabgab WORKSFORME :-) 11:17
masak I had a comment or two about the code, that's all.
szabgab wow, 1-2 comments for my 3 lines of code :-)
listening
masak yes, I have exacting standards. :)
but I try to do it as constructive criticism...
ok, first: the `when 'help' { self.help() }` block is redundant. 11:18
you'll note that the method itself is a dispatcher. its purpose is to dispatch to the right method.
none of the other methods are explicitly mentioned, so neither should 'help'. 11:19
wayland76 Construction criticism: You did a bad job of CEMENTing your code together. Were you PLASTERed when you wrote it? You WOOD do better if you tried to NAIL one problem at a time! :)
(excuse the joke :) )
masak wayland76: :P
11:20 donaldh left
masak szabgab: the second tiny comment is about whitespace. try to avoid \s at the end of lines. 11:20
making a commit to fix both those issues now. on balance, though, thanks for the patch; it clearly made proto better.
wayland76 'night all :) 11:21
masak wayland76: 'night
szabgab masak, I wonder if the firts comment really works well
what if the user types ./proto help me
she will get a message
me
masak tries 11:22
szabgab and if the users types ./proto help install it will blow up
masak hm, you're right.
11:22 donaldh joined
szabgab at least the way I wrote help with a single ($msg) argument 11:23
masak 'help' really is a new kind of subcommand.
I'll have to think a bit more about this, it seems.
szabgab and regarding the trailing whitespaces, well I think i have to teach my editor to strip them :-)
masak aye.
they are easy to spot in 'git diff', where they show up as big red blotches. 11:24
rakudo: 0 but True
p6eval rakudo ea667e: OUTPUT«The but operator can only be used with a role or enum value on the right hand side␤in Main (/tmp/1yCN0sNUvg:2)␤»
masak o_O
11:25 meppl joined
masak szabgab: right now, I'm leaning towards removing the $msg parameter, and having the 'help' method simply take a slurpy array and ignore it. 11:26
I think I prefer all subcommands to use the same calling style.
szabgab if I wan to catch any character except closing square bracket, is this the nicest way ? <-[\]]> ?
masak, you are the boss :-) 11:27
masak gotta go offline for lunch now.
see y'all later.
11:27 masak left
pugs_svn r27776 | pmurias++ | [re-smop] change {...}.() to {...} 11:45
11:54 ruoso joined
ruoso Hello! 11:55
pmurias ruoso: hi 11:58
ruoso pmurias, hi
pmurias ruoso: i'm trying to remove all the deviations from perl6 like using {...}.() 12:00
pugs_svn r27777 | bpetering++ | [misc/irclog] Bugfixes in nickfilter.js + 'spoken to' feature 12:01
pmurias ruoso: if i want to experiment with a tracing gc branch how should i check it in? 12:03
moritz_ phenny: tell masak a blocker? in the auld days you have just worked around it :-) 12:04
phenny moritz_: I'll pass that on when masak is around.
jnthn Indeed. Glancing at it, looks like using the block form of classes would work around it... 12:06
pugs_svn r27778 | moritz++ | [irclog] remove left-over debugging alert()
ruoso pmurias, hmm... maybe it would be a good idea to merge re-smop and re-mildew back to smop and mildew before starting other branch 12:07
12:13 skids_ joined
ruoso pmurias, and probably re-sync with current STD 12:13
pmurias merging with re-smop and re-mildew is likely a good idea 12:19
pugs_svn r27779 | pmurias++ | [re-smop] change $code.(...) to $code(...) 12:21
pmurias ruoso: re-syncing with STD, that depends if the current STD emits AST with full info 12:24
ruoso: what would be a intresting thing to try out would be to use viv as our rule engine 12:25
ruoso pmurias, aren't we using it already? 12:42
12:48 abra joined
pmurias ruoso: what i mean using it for rules in the user supplied programs 12:50
pugs_svn r27780 | pmurias++ | [re-smop] change $array.[...] $array[...] 12:53
12:54 iblechbot left
pmurias ruoso: that is compiling rules to perl 5 using viv and then using perl5 interop to call them from perl6 12:59
moritz_ mildew: say 'a' ~~ /a/ 13:00
p6eval mildew: OUTPUT«Could not find variable &infix:~~ in the lexical scope.␤»
pmurias moritz_: mildew doesn't have rules yet 13:02
13:02 Whiteknight joined 13:09 desertm4x_ joined 13:10 desertm4x left
pmurias ruoso: what other work arounds do we use? (fixing $hash.{...} atm) 13:14
13:15 hoelzro joined
pugs_svn r27781 | pmurias++ | [re-smop] change $hash.{...} to $hash{...} 13:17
13:17 SmokeMachine joined 13:21 Front_slash joined 13:22 masak joined 13:26 abra left 13:35 arthur-_ left 13:44 pjcj joined
pmichaud good morning, #perl6 13:45
masak oh hai pmichaud 13:46
phenny masak: 12:04Z <moritz_> tell masak a blocker? in the auld days you have just worked around it :-)
masak moritz_: oh, hush. :P
pmichaud: yes, a blocker. 13:47
13:47 hoelzro_ joined
masak pmichaud: one due to things that you patched two months ago. I'm amazed that no-one else has picked up on this in the meantime. 13:47
pmichaud details? 13:48
masak rt.perl.org/rt3/Ticket/Display.html?id=67976
jnthn masak: The workaround is to use the block form, no? :-P
masak jnthn: yes, I guess it is.
13:48 huf_ joined
masak now, if the fix is complicated, I'll do just that. 13:48
13:49 hoelzro left
jnthn has no idea 13:49
masak and I'll rescue my Pod comments to the end of their respective files.
13:49 hoelzro_ is now known as hoelzro
pmichaud that doesn't seem too hard to fix. 13:49
masak \o/
ruoso pmurias, hmm... maybe that will be better at the moment we have proper implementation of the interopr 13:50
13:53 M_o_C joined 13:59 huf left 14:12 teco000_ joined
misssingthepoint here's a quickie: how would i go about generating graphs of all sort using Perl 6, today? 14:16
I can think of a few things:
1. generate GraphViz DOT manually
masak another question similar to 'what would need to be present in a 1.0-like release?' is 'what would it take for Perl 6 to make the Slashdot frontpage?' 14:17
misssingthepoint 2. generate SVG a la masak++ thingo on use.perl
3. interface to something in Parrot that can do it?
masak misssingthepoint: I've been thinking of creating some sort of graph module on top of SVG. maybe one could use existing cool modules out there as inspiration.
I think I bookmarked one a while ago, let me have a look.
14:18 teco000_ left
masak yep, scruffy.rubyforge.org/ 14:18
misssingthepoint: if you'd like to get started porting that, I'm willing to give you a helping hand.
misssingthepoint masak: sure :) 14:19
masak misssingthepoint: in fact, I might have an incentive to get parts of it done in one week, so that I can use it in my YAPC::EU presentation :P
misssingthepoint: excellent. let me know where I can send patches. 14:20
jnthn YAPC::EU presentation? What's that? ;-)
masak jnthn: I have no idea. :)
jnthn Meh. I'll worry about it next week then.
;-)
masak it's somerhing that's been in the back of my mind in the last few days, that's all.
probably nothing to worry about.
misssingthepoint masak: thanks ;) 14:22
jnthn breathes easily
masak misssingthepoint: thank _you_ for letting me delegate tuit sinks. :P 14:23
I'll be following the work on that port with true interest.
szabgab ppl, how can I debug a grammar ? 14:25
it does not match something that I thought it should, how can I find out why ? 14:26
masak szabgab: Match.perl is a life-saver.
szabgab: other than that, the trick is to not write the whole grammar at once.
szabgab what do you mean by Match.perl ?
masak there's not good introspection tool for "now how come grammar G doesn't match string S?" situations.
szabgab I don't have a match
$/ is empty
masak szabgab: yes, I realize that. so it doesn't help in the exact situation you're in. 14:27
PerlJam szabgab: you could post the input and the grammar and let #perl6 take a look at it. :)
masak but genreally, it's really useful.
szabgab that's nice but I cannot teach that in a class :-)
or I can but, well, you know
jnthn ;-)
szabgab: If it's a grammar made up of various rules, testing the various rules in a bottom-up kinda way can help. 14:28
pmichaud misssingthepoint: I'm also interested in having a CSV parser, if you're looking for things to do :-)
masak szabgab: fwiw, I want to build an Ajax-based tool with two textareas, one for the grammar and one for the string, showing exactly how far things match.
szabgab: mberends++ and I talked about it once.
jnthn Loading the grammar module in at the REPL (use MyGrammar;) and then playing from there could be a quick-ish way to do that.
misssingthepoint pmichaud: sure :) 14:29
masak szabgab: write tests for the grammar, using jnthn's tip.
szabgab: you do have tests, right? :)
szabgab yes yes
pmichaud misssingthepoint: in my "Hacking Rakudo Perl" talk (where I tell people how they can help develop Rakudo Perl), suggestion #1 is to write applications and modules in Perl 6 :-)
jnthn There is that. Tests for a grammar probably would check the various rules.
szabgab and I have my own repl
sort of
masak szabgab: nice! is it better than the Rakudo REPL? :)
misssingthepoint pmichaud: i was just going through that before. it's good :)
szabgab masak, it is is a full IDE :-) 14:30
masak misssingthepoint: so, do you have a repo yet?
jnthn should also get on to writing Grammar::Generative, which would maybe help in terms of letting you see what a grammar could match...
masak szabgab: amazing!
jnthn Or producing examples thereof.
misssingthepoint masak: no, i'm in the class of developers who don't know version control... guess i have to git learning. 14:31
PerlJam having something like YAPE::Regex::Explain for grammars would be nice.
masak jnthn: sounds a bit like that Haskell test module, whatchamagoogle.
14:31 KyleHa joined
masak QuickCheck! 14:32
www.cs.chalmers.se/~rjmh/QuickCheck/ 14:33
hoelzro masak++ for QuickCheck reference =)
misssingthepoint masak: pugs.blogs.com/pugs/2007/10/a-graphical-tra.html might be interesting :)
szabgab can I embed print statements in the grammar to see what does it do ? 14:34
masak misssingthepoint: yes, that's a good part of the way to what I want.
szabgab: maybe you should check misssingthepoint's link out too.
szabgab: yes, but they have to be PIR for the time being.
szabgab: but it's definitely doable. 14:35
szabgab can you give an example ?
misssingthepoint jnthn: i'd love something like Grammar::Generative. :)
jnthn misssingthepoint: I keep planning to write it.
misssingthepoint: But keep getting distracted working on Rakudo. ;-) 14:36
misssingthepoint jnthn: feel free to indulge in that distraction :)
jnthn: is there any way i could help something like G::G happen? 14:37
14:38 abra joined
jnthn misssingthepoint: Well, actually, pmichaud++ is about to work on PGE refactors shortly... 14:38
misssingthepoint: So I was pondering holding off until then. 14:39
misssingthepoint: But really it's not *so* hard to write.
Just multi-dispatching over the PGE AST nodes.
masak szabgab: gist.github.com/157392
jnthn Recursively.
And building up output.
misssingthepoint jnthn: *winces* think a newbie could do it, given some pointers, and help from #perl6? 14:40
jnthn I was figuring the process could be given some "direction" as desribed by supplying a tree of nested captures.
14:40 hudnix left
szabgab masak, ty 14:40
masak szabgab: code immediately becomes Parrot-specific (and probably self-deprecating), of course. 14:41
pmichaud I've been thinking about adding a trace mode to PGE
jnthn misssingthepoint: Oh, another blocker...Regex.perl needs to work also.
pmichaud (based on seeing Damian's debugging output from Parse::Grammar)
jnthn gist.github.com/157394 # this is the tiny bit of code I'd scribbled so far...but more playing than anything serious. 14:42
PerlJam masak: parrot has say these days BTW
masak pmichaud: that would be nice, I think.
PerlJam: thanks for the tip.
and Parrot++
14:42 rfordinal left
szabgab masak, any help is a good help 14:42
moritz_ rakudo: Q:PIR{ say "foo" } 14:43
szabgab you'll see this trick display :-)
p6eval rakudo ea667e: OUTPUT«foo␤»
14:43 rfordinal joined
moritz_ bah, regexes don't seem to set their outer scope properly 14:46
ie a find_lex embedded in a regex will never succeed in finding Perl 6 variables
jnthn moritz_: Think that's known issue. 14:47
moritz_: PGE pre-dates lexicals refactor in Parrot...
:-(
misssingthepoint RFC: Parsing CSV is a (completely?) solved problem. Port Text::CSV? 14:49
masak sounds like a good idea. 14:50
moritz_ sure, why not?
misssingthepoint ok.
pmichaud (regexes don't see lexicals) This is the primary reason that regex { $a } doesn't work 14:52
moritz_ and let me guess, parrot doesn't easily allow something along the lines of $outer_lexpad = $caller.lexpad? 14:53
jnthn Heh, actually you can do a set_outer now 14:54
But it's not really the ideal fix I guess.
pmichaud setting the caller as the outer lexical is actually wrong
(in many cases)
jnthn Aye. Thus not ideal. :-)
moritz_ erm, yes
jnthn pmichaud: This may get fixed in upcoming PGE refactors?
pmichaud on the plus side, I'm working on PGE this week, and making it lexically aware is high on the list 14:55
moritz_ the outer pad where it was compiled, actually
pmichaud++
pmichaud jnthn: sure, it has to be fixed in order for us to get contextuals
jnthn Ah, yes. 14:56
14:56 alester joined 15:03 rfordinal left 15:05 jimmy_ joined
jimmy_ what is »ö« ? 15:07
it's in the topic.
masak jimmy_: it's a character version of a butterfly. 15:08
pmichaud that's the ASCII representation for Camelia
(except it's not ASCII :-)
jimmy_ I thought it was Unrecognizable Code :( 15:09
masak er, a 蝴蝶. :)
jimmy_ masak++, for improving chinese.
masak Wikipedia++
jimmy_ masak: where is the rakudo logo from? 15:10
TimToady in japanese usually just 蝶蝶, which might be fossilized babytalk 15:11
masak jimmy_: I'm not sure who made the Rakudo logo, actually. TimToady++ made the Camelia Perl 6 logo.
pmichaud I had a friend of mine who studies Japanese calligraphy make the logo 15:12
15:12 justatheory joined
jimmy_ It likes chinese character 15:12
pmichaud++ 15:13
masak jimmy_: japanese kanji tend to look like hanzi. :)
TimToady though the tendency is toward traditional over simplified 15:14
jimmy_ masak: yes, that was why i am curious
hoelzro that's because simplified is horrible =P
masak hoelzro: I wish it were that simple.
hoelzro I know that simplified increases literacy, but to me, it's so ugly compared to traditional 15:15
jimmy_ It is traditional Chinese
Simplified it is 乐土
TimToady yes, I was just about to type that :) 15:16
jimmy_ the tone sounds rakudo
masak hoelzro: we all have our own tastes, of course. I also feel some simplified characters are strange choices or ugly simplifications, but mostly I think they did a decent job. I'm definitely impressed that the whole project of simplifying worked as well as it did. 15:17
jimmy_ I think it is transliteration。
TimToady le4 tu3 by my dictionary
pmichaud gist.github.com/157434 <-- masak 15:18
jimmy_ hoelzro: yes, you're right.
masak 'happy soil'?
pmichaud++
pmichaud just need to run a spectest
jimmy_ masak: no, it is transliteration.
pmichaud that took far longer than it should have.
TimToady masak: "paradise"
hoelzro masak: I agree with the "working as well as it did" part =P
masak TimToady: ah, of course!
TimToady: I've heard it before, but doing the understanding with my own brain was much nicer. :) 15:19
jimmy_ reading 乐土， it sounds like rakudo
15:20 donaldh left
TimToady they're historically related 15:20
but chinese and japanese have diverged since the borrowings
15:21 donaldh joined
masak divergence happens. 15:21
TimToady trouble is that nobody bothers to declare it in natural languages :) 15:22
masak here in Europe, we have both that, and simultaneously heave inter-lingual borrowing.
jimmy_ anyway, good logo, i love it.
masak s/heave/heavy/
TimToady Camelia is intended to brainwash all the little children into loving Perl
pmurias ruoso: what do you mean by "a proper implementation" 15:23
?
masak: re slashdot front page, if 1.0 was really bad is propable it would make it to the slashdot frontpage 15:25
masak pmurias: I think a 1.0 release would make the frontpage regardless, actually. 15:26
just by virtue of its version number.
jimmy_ masak: does Camelia mean 蝴蝶，I can't find it in my dictionary
masak jimmy_: no 'butteryfly' means 蝴蝶. 15:27
pmichaud jimmy_: Camelia is the Perl 6 logo. See the upper-right of perl6-projects.org for a picture
masak ugh, 'butterfly'.
jimmy_ but what does Camelia mean? Camellia?
Su-Shee camellia is the tea plant. 15:28
moritz_ jimmy_: I think it's a name.
masak jimmy_: 'Camelia' sounds like a girl's name.
PerlJam Why are butterflies so called?
masak jimmy_: it's a word play on 'camel'.
moritz_ and names don't have to have meanings :-)
PerlJam flying butter doesn't sound descriptive
jimmy_ masak： Wikipedia++ again. 15:29
Su-Shee PerlJam: I've got a visual though :)
15:29 nihiliad joined
Su-Shee well actually "camelia" is a world famous sanitary pad brand. ;) 15:29
mberends Re: "Rakudo 1.0" release name, how about "Preview.0", "Preview.1" etc? 15:30
PerlJam Su-Shee: never heard of it.
masak mberends: gut reaction: I like.
jimmy_ Camelia (born June 20, 1974 as Dayang Noor Camelia Abang Khalid) is a Malaysian singer and model. She was the first Malaysian to be signed to Universal Music :)
15:32 SmokeMachine left, lumi_ left 15:33 r0bby left
PerlJam masak: tell me how you like it when we're at Preview.157 :) 15:33
jnthn Preview might be a name that puts people of playing with it who wouldn't have done so already. 15:34
masak PerlJam: yes, that's an argument for postponing any 'special' release indefinitely.
jnthn "It's not ready yet. Just a preview. Only for the really interested."
That's not really the goal, as I understand it.
masak so, we're looking for something more positive. 15:35
AlmostDone.0
Interesting.0
PerlJam something that doesn't only pique the interest of early adopters.
masak README.0
moritz_ perl-5.12 15:36
moritz_ ducks
PerlJam moritz++
masak :D
pmichaud I'm looking for a term that means "you can really use this for real stuff" while also indicating "it's not everything in Perl 6 yet"
Su-Shee rakudo beta.
moritz_ prakudo, where 'p' stands for 'partial'
PerlJam pmichaud: perl 5i (already been done though :)
Su-Shee moritz_: pre-kudo?
mberends PartridgeInAPearTree.1 # 12 step scale 15:37
masak fracudo, f for 'fractures easily'
moritz_ Su-Shee: "beta" suggests near feture completenes
Su-Shee: I like pre-kudo :-)
masak mberends: :)
pmichaud hopefully by the time of the release we're not fracturing as much
jnthn Perl6-- ;-)
Su-Shee moritz_: won't it be "near feature complete"?
masak jnthn: do you realize what that would do for the karma of Perl6 on this channel? :O
pmichaud Su-Shee: given that the Perl 6 spec would likely not be complete by the time of the release, it's hard to imagine that the implementation would be :-) 15:38
jnthn OMFG I RETRACT MY SUGGESTION!
;-)
moritz_ Su-Shee: depends on what "near" is for you
PerlJam Isn't the name more about branding than advertising? Pick a brand (name) and advertise it well and it doesn't matter so much what the name is (as long as it isn't unfortunate)
pmichaud PerlJam: names matter
TimToady well, the Perl 5 spec isn't complete yet either 15:39
moritz_ I suspect it won't have substantial concurrency, macros, speed and other stuff that we don't have yet (although I'd love to be proven wrong)
pmichaud PerlJam: calling it "Perl 6.0.0" would undoubtedly be a mistake :-)
PerlJam pmichaud: sure they matter, but they don't have to bring any kind of message along with them. They just have to be "nice"
Su-Shee how about one of those cute naming schemas like rakudo riveting ranger? ;)
jimmy_ perl 5++?
moritz_ oh noez
TimToady Perl 6.-1.0
pmichaud PerlJam: the point is that some names bring a message along with them, whether we like it or not 15:40
jnthn It'd probably be good if the name actually included "Rakudo".
Su-Shee well if it is something which one can actually use, I really would call it plainly rakudo beta.
PerlJam what message does "rakduo 1.0" bring?
er, spelt correctly
jnthn "use me"
TimToady rakudo 1.-1.0
pmichaud PerlJam: for some people, it would bring the message "Perl 6 is finished"
masak PerlJam: 'sharp release' 15:41
jimmy_ rakudo 1.0 preview
pmurias if the current release of rakudo is 19 wouln't 1.0 be a very old one?
15:41 lumi_ joined
masak Rakudo 35.0 15:41
moritz_ pmurias: Num !~~ Int
TimToady make it a website and you can keep it beta for five years
jimmy_ rakudo 1.0 alpha1 ?
Su-Shee rakudo "early adopter" ;)
PerlJam pmichaud: how many are "those people"?
pmichaud PerlJam: I suspect a great many (more)
Su-Shee TimToady: exactly my thinking. ;) it works for google. ;)
jimmy_ perl6 preview 15:42
pmichaud PerlJam: in some sense, I don't want to provide too much ammunition for the crowd that wants to say "see, you've been working on Perl 6 for ten years and all you're able to produce is a very incomplete implementation"
jnthn masak: Well in that case, clearly Rakudo 42 is the answer. ;-)
pmichaud I recognize we can't silence that crowd entirely
masak jnthn: that does have some appeal.
jimmy_ Perl 6 preview
moritz_ that'll be in 5 months...
around .... *drummrolls*.... christmas!
PerlJam "preview" doesn't say "use me" at all.
15:43 molaf joined
masak right, 'Preview' says 'wait a while'. 15:43
pmichaud but at the same time, we don't want people thinking "Perl 6 is done!" and being disappointed by the un-done parts
masak "we'll be stable soon"
jimmy_ Perl 6 please use me
PerlJam "rakudo 1.0" says "use me" Whether that's a complete perl 6 or not it doesn't say (IMHO)
pmichaud PerlJam: I agree -- I'm not totally against "Rakudo 1.0" as a name, as long as we can come up with a good way to say what it is/isn't
Su-Shee how about rakudo advent :) that's really, really short before christmas and leads directly to it. :) 15:44
PerlJam Su-Shee++ :)
hoelzro Rakudo 0.99(repeating) titled "Christmas is Here"
Su-Shee: damn, that's even better
PerlJam Su-Shee: and advent is much better than "Rakudo 1.0 - Christmas Eve" too :) 15:45
jimmy_ Rakudo 1.0 the Christms Gift
jnthn It does have the benefit that releases don't get caleld "Advent" normally so people will have to read what it means rather than make assumptions like they might on seeing 1.0 or preview.
jimmy_ s/Christms/Christmas/
Su-Shee well it's a little off if rakudo is actually released around easter the first time. ;)
PerlJam pmichaud: what do you expect to be unimplemented (or mostly so) in Rakudo 1.0 ?
15:46 kst left
pmichaud PerlJam: well, that's also what needs to be decided. But I'd be doubtful of concurrency features 15:46
I also don't know how much of S09 would be present
15:46 kst joined
Su-Shee pmichaud: well there has to be some room for future releases and completing features. look what other programming languages release as their version 1.0 15:46
PerlJam Su-Shee: ruby 1.0 was crap, only no one knew hardly anything about ruby at the time so it didn't matter so much. 15:47
Everybody knows about Perl though 15:48
TimToady heh, "ADVENT" was what you typed on a PDP-11 running RSTS to run adventure, since filenames were limited to 6 letters
masak yes, the problem is the high expectations.
[particle]1 pmichaud: is there a nazareth.pm group?
Su-Shee ah. and all advents have names, btw. so there could be a rakudo gaudete and a rakudo at te levavi and so on :))
pmichaud There is a part of me that says for us to do what's natural for us and not worry too much about expectations.
15:48 [particle]1 is now known as [particle]
PerlJam Then maybe it's time for a rebranding (as much as I hate to say it, stop calling the language "perl") 15:48
pmichaud PerlJam: "Rakudo" is a useful rebranding for this purpose, I think. 15:49
masak PerlJam: but it is Perl.
PerlJam masak: that's beside the point :)
TimToady we're not going to waste a perfectly good four-letter word
KyleHa You could say it's Rakudo 1.0 without saying it's "Perl 6" 1.0
masak PerlJam: that's an interesting opinion. :)
pmichaud I like the "adventure" side of "advent". I'd be a little concerned that calling it "Advent" (with the biblical connotations) might offend non-christians.
masak Rakudo-1.0-but-not-Perl-6.0 15:50
Su-Shee I've always thought of it as a "rakudo perl implementation"
dakkar «Rakudo Advent — it's Perl 6-ε»
PerlJam masak: read something about how GTE figured it would take over 25 years of perfect customer service and marketing for them to overcome the negative impressions that their current customer base had. So what did they do? The changed names.
moritz_ KyleHa: people already confuse Perl 6 and Parrot. I don't think we should give them more room for confusion
15:50 Whiteknight left
Su-Shee pmichaud: well the "before christmas" pun already was christian anyway. 15:50
masak pmichaud: why would 'Advent' offent non-Christians?
PerlJam masak: in some sense Perl has a boat load of negative press to voercome
er, overcome
mberends while TimToady++ is right that specs continue to evolve, it would be a big plus if changes to the Synopses could be restrained somehow after Advent.
hoelzro Rakudo "Winter Solstice"?
[particle] pmichaud: consider naming 'stable' releases differently from 'developer' releases 15:51
masak PerlJam: true. it also has a couple of good connotations.
pmichaud Su-Shee: sure, I recognize that "christmas" is christian already (and I have no problem with that), continuing the same theme might tend to enshrine it
Su-Shee pmichaud: you could wear red and a white beard and do a ho ho ho while releasing rakudo :)
masak mberends: why would it be a big plus to stop improving the specs?
PerlJam Su-Shee++ I like the way you think! 15:52
:-)
masak me too!
Su-Shee PerlJam: you just wanna see pmichaud as santa ;)
pmichaud [particle]: I really want to excise the term "stable release" from anything we do.
masak well, who doesn't?
TimToady how am I supposed to backlog when y'all type so fast?
[particle] yeah, i don't really mean stable, anyway
i mean milestone or something
PerlJam Su-Shee: I still see him as a Willy Wonka mostly, so that would be a change :)
pmurias the point of a 1.0 is to mark a stable release
Su-Shee pmichaud: considering the name rakudo one could develop an international name scheme of hildays of all religions 15:53
pmichaud pmurias: except "stable" means different things to different people. See p5.
PerlJam Su-Shee: The Gene Wilder version of course
TimToady you mean stable like parrot 1.0 is stable :P
awwaiid Rakudo Solstice -- From the depth of winter comes the light
[particle] /kick TimToady
pmichaud although going with the Willy Wonka theme, perhaps Rakudo 1.0 is the "Candy Store" release. :-) 15:54
15:54 TimToady sets mode: +o [particle]
[particle] :) 15:54
pmichaud and everyone who uses it will be like children in a candy store.
hoelzro Rakudo "Golden Ticket"?
Su-Shee rakudo eid (end of ramadan) for example.
[particle] android is using candy names, well, bakery item names
pmichaud (Golden Ticket)++
TimToady I want the world, I want the whole world, I want it now!!!
pmichaud this is your ticket to paradise. :-)
Su-Shee rakudo golden ticket sounds like a drug lord release. ;) 15:55
hoelzro pmichaud: that sounds like something a cult leader would say
pmichaud well, we do hope that Perl 6 is addictive.
TimToady call it Gateway. oh wait...
masak hoelzro: I like "Golden Ticket"!
Su-Shee pmichaud: and being filtered by every parental control on the planet? ;)
pmichaud me, it appears I'm addicted to oxygen.
hoelzro masak: me too, especially Charlie hasn't arrived at the factory just yet =) 15:56
15:56 snearch joined
TimToady every now and then I hold my breath to prove I'm not really addicted to oxygen. 15:56
pmichaud TimToady: and you manage to avoid the withdrawal symptoms?
masak you guys should be careful. oxygen is poisonous.
and very reactive.
[particle] i'm addicted to nitrogen 15:57
pmichaud wonders where he could get a Willy Wonka outfit.
TimToady [particle]: you have Rapture of the Deep?
pmichaud (and yes, the Gene Wilder version)
15:57 donaldh left
Su-Shee the release cycle of rakudo stays with this monthly releases? 15:57
pmichaud Rakudo will likely continue to have monthly releases for a while 15:58
at some point in the future we might move to quarterly releases
Su-Shee pmichaud: let's ask johnny depp if he's doing the release! :)
TimToady speaking of Camelia products...
Su-Shee pmichaud: I was thinking calling it plainly rakudo january, february and so on and later something like easter, vacation, halloween and christmas. 15:59
TimToady I like that
Su-Shee isn't there something in summer..
[particle] turns deep purple 16:00
TimToady apparently nobody has revolutions in August
Su-Shee easter, revolution, halloween, christmas. ;)
TimToady: well napoleon and lawrence of arabia are born in august. ;)
pmurias TimToady: en.wikipedia.org/wiki/August_Revolution 16:01
16:01 dukeleto joined
TimToady maybe we should start with romulus and remus, and then we'll have more names to use up 16:01
masak :)
pmichaud Su-Shee: Rakudo already does month names for its releases
masak it can only go downhill after romulus and remus.
TimToady but it can only do it seven times 16:02
Su-Shee pmichaud: I thought it perl mongers chapters?
pmichaud Su-Shee: that also.
TimToady is there a rome.pm?
[particle] until you get to the sisyphus release...
misssingthepoint rofl @ TimToady 16:03
TimToady "rome wasn't built in a day" and all that...
[particle] too bad the plumbing was lead
pmichaud Rakudo 1.0 -- "Not built in a day"
hoelzro pmichaud: I like that a lot
Su-Shee pmichaud: *hihi* this is pretty funny :) 16:04
[particle] can't wait for the NVTS release
dakkar yes, there is a Rome.pm
16:05 kane_ joined
PerlJam Rakudo 1.0 - vini vidi vici :-) 16:06
TimToady backlogs desperately while everyone is panting
Su-Shee and then we're going to have a cesarian release, a decadence release, a stoic release.. ;)
hoelzro PerlJam: s/vini/veni/?
KyleHa Rakudo one point oh hai!
hoelzro Su-Shee++
PerlJam hoelzro: yeah, htat 16:07
misssingthepoint s/vini/vino/
hoelzro misssingthepoint: nice
16:07 masak left
TimToady s/vino/veritas/ 16:07
Su-Shee
.oO(make it a toga instead of a red dress and a white beard ;)
16:07 SmokeMachine joined
hoelzro TimToady: veritas ~~ truth, right? 16:07
my Latin's terrible 16:08
Su-Shee misssingthepoint: rakudo burgundy, rakudo champagne, rakudo...? ;)
16:08 dakkar left
Su-Shee ("monthly release party!) 16:08
misssingthepoint how about the world's wine regions? I live near Jacob's Creek.
Rakudo Napa Valley?
Su-Shee misssingthepoint: this will lead to many, many french releases. ;) 16:09
misssingthepoint um, yess....
jimmy_ Perl 6 times coming 16:10
16:11 molaf left
misssingthepoint Rakudo Irresistible? 16:11
Su-Shee rakudo seduction ;) 16:12
TimToady now if I said that, it'd be sexist. :) 16:13
moritz_ as long as we don't name it "rakudo porn star" ;-) ...
mberends the Rakudo eXPerience :/
misssingthepoint mberends: sacrilege!
mberends sry ;-)
jnthn moritz_: No, no, we need Rakudo to perform like a porn star first. ;-)
16:13 zamolxes left
TimToady or at least sixist 16:13
Su-Shee moritz_: *haehae* :) 16:14
misssingthepoint Su-Shee: like you were saying, many many French releases.
jnthn <ducks>
Su-Shee jnthn: would you show us..? ;)
TimToady Perl 6 times coming?!?
[particle] we could go the musical route, using words like harmony, staccato, and my favorite for a 1.0 release, cadence
mdxi i'd rather have cheeses than wines. Rakudo Double Gloucester 16:15
Su-Shee TimToady: *ouch* :))
jnthn eww!!
araujo votes for jnthn recommendation
[particle] there's a cheese local to me named camelia
TimToady ducking?
[particle] actually, camellia. 16:16
araujo TimToady, haha, before his duck :D
jimmy_ Rakudo 1.0 Camellia 16:17
[particle] www.redwoodhill.com/cheese.htm :) 16:18
[particle] could provide many, many, many cheese names for releases
Su-Shee cheese will probably make half of asia feel nauseous ;) 16:19
(I vote against a rakudo natto release ;)
[particle] those names might be more fitting for mold/mildew/slime/smop releases
araujo Rakudo 1.0 Anchovies
moritz_ rakudo 1.0 "extra cheezburger" 16:20
TimToady there are a lot of STDes
pmichaud okay, here are a few thoughts
oh wait, brb
Su-Shee [particle]: rakudo rotten release ;)
araujo Rakudo 1.0 Milkshake
[particle] pmichaud needs to be alone with his though
*thought
misssingthepoint Rakudo 1.0: <STD>: Parse it along... 16:21
misssingthepoint shudders at badness of that.
[particle] it's catchy. 16:22
TimToady you aren't allowed to shudder at your own bad puns
pmichaud b
araujo hahaha
pmichaud the names here are all excellent ideas, and we might even use some (more)
the point of this is not to come up with a naming scheme (rakudo already has one -- we use .pm groups), but rather to figure out how we can indicate to the wider world that we're beyond development and into "useful product" 16:23
[particle] well, then with cheese names you start with ricotta 16:24
TimToady then name it "Useful" :)
pmichaud TimToady: yes, I just now thought of that.
But more to the point, the idea I had (possibly bad idea) is that instead of figuring out the name, perhaps we should be drafting the release announcement we'd like to use when we get to that point
[particle] sounds like you could use a lot of smurf/dwarf names like that
pmichaud perhaps the release announcement would then tell us the appropriate terms to use
(I knew I could clear a room quickly, but ...) 16:26
PerlJam pmichaud: sounds good to me :) 16:27
jimmy_ When will come to 1.0?
PerlJam wanders off again to help a student with ldap auth issues
pmichaud jimmy_: is it necessary that we have a 1.0? (honest question)
jimmy_ maybe 16:28
misssingthepoint pmichaud: you just have us all thinking :)
pmichaud I suspect we're all thinking "is that a good idea, or not?" ;-) 16:29
and we're afraid to say either "yes" or "no" :-P
cj TimToady: yeah, I asked someone to @tell yo
Su-Shee pmichaud: if you're concerned with all this "after 10 years" stuff I think it should be an 1.0
pmichaud Su-Shee: that strikes a chord with me, yes. 16:30
jnthn I think part of working this out is working out the priorities for this Bigger Release, whatever we're going to call it.
TimToady call it "Ten" then :)
jnthn And what'll be in it.
[particle] "Perfect Ten"
TimToady 天 16:31
jnthn Or, what must be in it for us to declare "we've reached that point"
pmichaud jnthn: yes, that's one of our primary tasks at yapc::eu
jnthn TimToady: Isn't that sky/heaven/imperial?
16:31 Psyche^ joined
moritz_ if we do release something called "1.0", then please first stress-test rakudo.org if it can stand a slashdot "attack" 16:31
jnthn pmichaud: Right, but I suspect that needs to come before working out names/announcements. Much fun as naming is. ;-) 16:32
pmichaud jnthn: s/come before/as part of/
in many ways, the announcement is "what we want to be announcing"
that tells us what we have to implement
phrased differently, the announcement is our goal 16:33
16:33 rfordinal joined, dukeleto left
jnthn Yes, to some degree I can see that. 16:33
Though I suspect the path towards the goal - something else we'll want to work out a bit - will be more detailed than that.
pmichaud certainly
there will be many features that must be present that won't be mentioned in the announcement 16:34
jnthn Right.
16:34 snearch left
jnthn Thus why I think an announcement would come out of this process, but only be one thing that would come out of it. 16:34
pmichaud right, agreed. 16:35
My feeling is that if we're going to be setting expectations, we have to understand that now and can't wait until next Spring to start addressing it.
jnthn But I do like the notioin that we figure out in advance what we'd like to be announcing.
Right.
16:36 Whiteknight joined
pmichaud going off on another set of thoughts for a bit 16:36
[particle] The Rakudo development team is proud to announce a future release of Rakudo 1.0: a useful, working, subset of Perl 6. Contents and Release Date TBD. 16:37
PerlJam [particle]: ship it!
pmichaud [particle]: I like that.
[particle] :)
wikify it and start adding content.
pmichaud (I'll leave my other thoughts alone for now, actually. They have to do with "stable" versus "predictable")
Su-Shee the "subset" is a little turn-off.
16:38 jimmy_ left
pmichaud Su-Shee: I'm open for other terms, but "subset" is honest. 16:38
[particle] subset is the goal, we should be clear about that
PerlJam Su-Shee: for me, after reading that, I'd just want to know what works and what doesn't
misssingthepoint s/useful/highly useful/ and keep 'subset'? 16:39
Su-Shee pmichaud: subset sounds to me like "it can't do sockets" or something essential like this. main set? I don't know the right english term.
[particle] to the wiki, please!
PerlJam Su-Shee: sockets are essential?!?
Su-Shee misssingthepoint: "highly" usally implies otherwise
misssingthepoint Su-Shee: true :(
Su-Shee PerlJam: uhm, networking..? :)
pmichaud "Rakudo 1.0: a substantial implementation of Perl 6" 16:40
PerlJam Su-Shee: I use perl all the time without needing networking :)
16:40 jan_ left
Su-Shee PerlJam: was just an example. make it "can't do file IO" :) 16:41
[particle] how on earth do i create a new documentation page on rakudo.org?
PerlJam that one would be a showstopper for me :)
pmichaud [particle]: if you want it to be wiki, use github
[particle] ok then 16:42
16:42 cdarroch joined
pmichaud wiki.github.com/rakudo/rakudo 16:42
Su-Shee pmichaud: if I understand you correctly, rakudo release can do the rakudo version of what "today every programming language can do" in terms of features - the fancy stuff like concurrency excluded?
16:42 Patterner left, Psyche^ is now known as Patterner
pmichaud Su-Shee: that seems like a very good approximation, yes. 16:42
(and I like this approach) 16:43
Su-Shee pmichaud: then how about "a base implementation of Perl 6"?
pmichaud or "an implementation of basic Perl 6"
PerlJam Su-Shee: for some people concurrency is essential. 16:44
Su-Shee basic was what I wanted to avoid. :)
pmichaud PerlJam: we can always come up with some feature X that some group will deem "essential"
jnthn Some of the stuff Rakudo can do is not basic. ;-)
Su-Shee PerlJam: yes, but it's in comparison not something which I would call "standard feature"
16:44 jan_ joined
TimToady concurrency means many different things, but if we do laziness, we'll have most of concurrency already 16:44
[particle] wiki.github.com/rakudo/rakudo/rakud...ment-draft 16:45
pmichaud My purpose is not to say "we fit your requirements" for any value of "your". Rather it's to say "we fit the requirements for a lot of people, and you might be in that group"
PerlJam pm: indeed, I think we should just be wary of calling features "fancy" :)
pmichaud ...why is that?
TimToady but then there's also "underpromise and overperform"
[particle] rakudo.org/documentation 16:46
Su-Shee pmichaud: hence my comparison. you probably want to say honestly "you can do with perl 6 what you can do with any other programming language on the market now"
[particle] ok, folks, edit away.
PerlJam it kind of trivializes them to those that think them essential.
[particle] Su-Shee: including segfault occasionally ;) 16:47
PerlJam is wary of generalizations.
pmichaud PerlJam: I think that many features of Perl 5 are both "fancy" and "essential to my purposes".
PDL comes to mind.
PerlJam Su-Shee: lua? erlang? ;)
pmichaud "scripting language" 16:48
Su-Shee on the other hand: how many people really "get" perl 6's feature at a glance and in comparison without taking a detailed, very knowledgable look at it? and if so: wouldn't the very first remark everybody is doing "well, nice language - but there's no environment right now" and not really care so much wether or not it has feature x or y?
PerlJam pmichaud: haskell?
pmichaud Su-Shee: ...no environment?
Su-Shee pmichaud: I mean stuff like DBI, SSL, Gtk, foo, bar..
PerlJam padre!
:) 16:49
Su-Shee: that sounds like "no CPAN"
pmichaud Su-Shee: I'm hoping that by the time of the release we'll actually have some of those things in place as well.
Su-Shee pmichaud: I still got the interesting discussion of reddit's switch from lisp to python in my mind.
16:49 ruoso left
pmichaud We certainly should be able to handle standard modules of various forms. To me that's essential to the release as well -- if we can't do modules, then we're not at "useful" status. 16:49
Su-Shee pmichaud: but it won't be today's CPAN then and I'm sure each and every nitpicker will rub exactly that in. :) 16:50
pmichaud Su-Shee: I'm not aiming at the nitpickers. We have to be aware of them, but we don't have to convince them.
16:51 misssingthepoint left, misssingthepoint joined
pmichaud I'm aiming at the folks who are thinking that maybe they'd like to use Perl 6 but they're not sure it's ready for them yet. 16:51
This release is to say "we think we're about ready for you" 16:52
misssingthepoint ... by which i mean it's late. i would love to stay here all night. bye! D/ :)
16:52 misssingthepoint left
Su-Shee but if more or less everything need is already there - then it really isn't a subset anymore, isn't it? 16:52
needed.
PerlJam wonders where those people are ...
pmichaud PerlJam: they exist. I meet them all the time at conferences.
PerlJam mostly I tend to hear "Perl 6 rocks!" or "Perl 6 sucks!" and never do I hear/see "Perl 6 sounds interesting ... when will it be read" 16:53
16:53 sri_kraih_ left
PerlJam er, ready?" 16:53
16:53 justatheory left, sri_kraih joined
pmichaud "I'd really like to use Perl 6, but I have a difficult time convincing myself (or my managers) to start using it." 16:53
moritz_ PerlJam: I do hear that from fellow perl mongers.
[particle] database bindings will make perl 6 useful for many
pmichaud PerlJam: this is classic "early adopter chasm" stuff (see "Crossing the chasm") 16:54
People are afraid to jump in until they see others like them jumping in
Su-Shee pmichaud: actually, I didn't. :) I just got a job exactly with selling "Perl 6 is the future of Perl and we want smoothly change into it, because we commit to perl for another decade"
pmichaud Su-Shee: excellent!
PerlJam Su-Shee: and they didn't ask "why?" or "why not <some other language> instead?" 16:55
Su-Shee well let's see if they really mean it, but they certainly were more then just "interested".
hoelzro I'd like to write the backend for Try Rakudo! in Perl6, but I want things like chroot and fork =(
pmichaud hoelzro: I think we might be able to get those for you.
hoelzro also, given my certain security ideas, I'd need two versions of Rakudo and Parrot on my machine
pmichaud: yay 16:56
pmichaud Maybe not chroot, but fork we should be able to do.
Su-Shee PerlJam: they've tried a couple of years ago in the era what I call the "perl crisis" ;) but everything else wasn't convincing so they stuck with Perl. Now they like it anyway and don't want to change anymore.
hoelzro pmichaud: binding Rakudo to native functions is still a moving target, yes?
PerlJam Su-Shee: them++
pmichaud hoelzro: yes, but it should be no longer moving by next January
Otherwise Parrot will have missed some major milestones. 16:57
hoelzro pmichaud: is it Parrot we're waiting on?
PerlJam pmichaud: and parrot has never done that </sarcasm>
pmichaud for fork and chroot, I suspect so.
PerlJam is in a weird mood today
hoelzro I was reading up on Parrot opcodes, and it seems like they have dlopen and friends
16:58 kane_ left
pmichaud If parrot already supports fork/chroot at the PIR level, then it should be straightforward to add them to Rakudo. 16:58
hoelzro: yes, dlopen exists.
But the interface to calling the native functions is the moving target.
hoelzro ah I see
PerlJam hoelzro: yes, the NCI is good, but still a tad clunky (at least last time I looked at it)
Su-Shee (what I need desperately with a reakudo for real release is a real book..) 16:59
hoelzro I understand if Parrot needs to do special things for fork, but I'd think chroot would be relatively simple
pmichaud Su-Shee: (stuck with perl) That sounds like a good article in there somewhere.
hoelzro unless you need to find/hang on to certain files
16:59 lucs joined
Su-Shee pmichaud: they also plainly had no reason to change after perl got back on the track again. 17:00
PerlJam Who have perl 6 books in the works and will they be ready by Spring 2010?
pmichaud I don't know that big-publisher perl 6 books will be ready by Spring 2010
Su-Shee I thought, there was a grant for a book?
apress got one announced. 17:01
PerlJam Stonehenge -> Learning Perl 6 and TimToady + Damian -> Programming Perl 6 ? right?
Su-Shee since 2008 ;)
pmichaud I'm wondering if we can put something together that's print-on-demand-ish, though.
moritz_ would love to do such a thing
araujo remember watching a Perl6 book in amazon
moritz_ who else would contribute?
PerlJam I'd volunteer for book writing/editing/etc
Su-Shee me gladly. or at least do some editorial work. 17:02
pmichaud A prominent Perl person (I don't know if he wants to be known yet) already started on one this past weekend. :-)
araujo moritz_, I can help testing code and stuff
:P
[particle] Whiteknight has experience doing this
pmichaud but we were discussing it as an online book, with lots of authors
Su-Shee pmichaud: well we got an ebook reader anyway. ;)
moritz_ pmichaud: if he wants a contributor, feel free to point him/her in my direction 17:03
Su-Shee book == a real documentation with examples.
araujo www.amazon.com/s/ref=nb_ss_gw?url=s...=0&y=0
pmichaud moritz_: I will do that today.
hoelzro if you ask me, any such book should have a "migration/differences from Perl5 chapter"
or there should a separate book entirely
Su-Shee moritz_: at least we know that the german it press is sympathizing. ;)
moritz_ Su-Shee: we do know ;-) 17:04
pmichaud hoelzro: I think that's a separate book, personally.
Su-Shee I think so too.
hoelzro pmichaud: there's definitely enough content for it
also, your target audience would be different
pmichaud The book I'm really interested in is "Perl 6 for new programmers" and "Perl 6 for Python/Ruby/Java/JavaScript programmers"
araujo and Haskell programmers? :P 17:05
hoelzro pmichaud: the latter sounds interesting
araujo can help with that part
moritz_ araujo: I have to admit that I find that book a bit suspicious - to the best of my knowledge, neither of the authors came here or to the mailing lists and asked for clarifications or so
araujo: IMHO that's a bad sign
araujo moritz_, that's what seems weird to me , really
moritz_, I find it weird that even nobody here seems to know the authors or even the books
I don't get it
Su-Shee araujo: that's the apress perl 6 book I'm talking about. 17:06
moritz_ It's hard for me to imagine that there are people who really grok Perl 6 and don't have questions
PerlJam araujo: rdice is known
moritz_ araujo: I know richard dice from his TPF work a bit
17:06 justatheory joined
PerlJam I have no clue who James Lee is though 17:06
araujo ok, but it seems like more people have worked on these books
Su-Shee we could start with a perl 6 pocket reference. that would be a great help. 17:07
araujo weird that none of them even gets connected to this channel
PerlJam Su-Shee: I was contemplating writing a Perl 6 Regex quick ref
araujo considering plenty of development goes on here
hoelzro PerlJam: that'd be welcome to many, myself included
Su-Shee PerlJam: I personally need more. I really love those pocket references.
PerlJam Su-Shee: sure, but that's a niche that should be filled at some point though 17:08
Su-Shee PerlJam: but of course: I'll order one! :)
pmichaud the idea of a "Perl 6 Pocket Reference" also came up at OSCON. That could possibly be done by Spring 2010
(not just regexes
araujo Perl 6 in Nutshell ?
Su-Shee araujo: even better.
araujo would like to see that one too
17:08 fridim_ left
moritz_ araujo: that would bring us dangerously close to an O'Reilly brand, I think 17:08
PerlJam I've come to dislike nutshell books as they have started to seem like "second tier" books to me. 17:09
araujo moritz_, well, O'Reilly is a cool guy right?
:P
moritz_ araujo: I doubt his laywers are as cool ;-)
Su-Shee PerlJam: I think you underestimate how difficult it is to actually get a grasp of what Perl 6 already does and contains.
pmichaud I think the context of "Perl 6 in a Nutshell" was that O'Reilly would publish it, I think.
araujo moritz_, hah, right, cool except when it is about money 17:10
:P
PerlJam Su-Shee: I don't think so. :)
araujo pmichaud, what I meant ....
Su-Shee PerlJam: but you already know all this stuff! :)
hoelzro pmichaud: that reminds me, did you read my thoughts on security for Try Rakudo?
pmichaud hoelzro: where are they? 17:11
araujo thinks other successful name would be "Perl 6 From Moritz"
PerlJam Su-Shee: no, I only know the parts well that intersect useful and interesting *for me*.
pmichaud hoelzro: (no, I don't think I read them)
hoelzro in Sunday's log =(
pmichaud: I can send you an e-mail if you'd like
araujo moritz_, considering you have plenty of doc about it :P
pmichaud hoelzro: I'll go back and read them. I've not had a good chance to backlog yet.
hoelzro still needs to get on the mailing list
moritz_ araujo: thanks :)
hoelzro mkay
PerlJam Su-Shee: there's tons of stuff in perl 6 that I don't know
araujo moritz_, pretty much where I have been learning from :)
hoelzro PerlJam: that's part of the point: it's an adventure! 17:12
Rakudo "Adventure Island"
PerlJam Ah ... James Lee is the guy who took over "Beginning Perl"
hoelzro: I'm all for Rakudo 1.0: Advent (punning is what we do :)
hoelzro PerlJam: I'm a big fan of Advent 17:13
pmichaud Once again, I'm not looking for a name.
(Unless we decide that we not use .pm groups for names)
i.e., I'm not looking for "Rakudo 1.0: Foo", where "Foo" is the name of the Rakudo 1.0 release. I'm looking for "Rakudo Foo", where "Foo" describes the classification of the release (beta, developer, production, etc.) 17:15
moritz_ pmichaud: (unrelated) btw I'm available to do the next rakudo release, but I'd prefer it somebody who never did it volunteered
pmichaud moritz_: Agreed, I want to build up a team of release managers
moritz_: but you'll be the fallback if we FAIL at getting someone else to do it.
I think jnthn++ should do one, though :-)
moritz_
.oO( I'll be your CATCH block if you fail() )
 17:16
Su-Shee ooooh perl 6 romance! ;)
PerlJam Is there music to go with that song?
pmichaud I think PerlJam should do one, too.
moritz_ aye
PerlJam pmichaud: I will. As soon as I get over this "I'm not worthy" thing
or "I'll screw something up" or "I suck" or whatever 17:17
pmichaud PerlJam: I hereby dub thee "worthy".
moritz_ PerlJam: I think you need mst screaming "you're not good enough" at you ;-)
pmichaud You're now the August release unless you convince me otherwise. :-) :-)
(Or find me another vict...volunteer.)
PerlJam pmichaud: give me October. (It's my birthday month so I'm partial to it) 17:18
pmichaud Okay, you get October.
PerlJam Give masak August
pmichaud (on phone)
hoelzro is '...' not yet impl'd, or has that been removed? 17:19
moritz_ PerlJam: did you submit a CLA already?
PerlJam moritz_: long ago.
moritz_ hoelzro: which '...' do you mean? the operator? or the term?
PerlJam moritz_: *that's* the kind of thing that grok should do :) 17:20
grok '...' # Which ... do you mean?
hoelzro moritz_: the yadda yadda yadda operator
moritz_ rakudo: ...
hoelzro eg. fail alias
17:20 xomas_ left
p6eval rakudo ea667e: OUTPUT«Can't return outside a routine␤in Main (/tmp/YzpnxMhUz3:2)␤» 17:20
hoelzro it parses ok when I try, but it doesn't puke
17:20 sparc left
hoelzro when run 17:20
moritz_ hoelzro: fail() is not the same as die()
17:21 kane_ joined
moritz_ hoelzro: ie it will give you an undef which contains an unthrown exception (if all works well) 17:21
lisppaste3 mkelly32 pasted "spectest failures on freebsd" at paste.lisp.org/display/84323
moritz_ rakudo: my $x = { ... }; $x = $x.(); say $x.WHAT
p6eval rakudo ea667e: OUTPUT«Can't return outside a routine␤in Main (/tmp/KvwaOHhG8y:2)␤»
hoelzro needs to re-read that spec
mkelly32 lemme know what other info you need from that ^^
moritz_ rakudo: my $x = sub { ... }; $x = $x.(); say $x.WHAT
p6eval rakudo ea667e: OUTPUT«Failure()␤» 17:22
moritz_ mkelly32: is that latest rakudo?
mkelly32 that's w/ 2009-07-28-gea667e8
hoelzro moritz_: how's that supposed to propagate?
if someone calls a method on that return value?
moritz_ for example, yes
but I must admit that I haven't grokked all the details yet 17:23
mkelly32 moritz_: full build log, if that's helpful: test.pioto.org/rakudo-smoke-2009-07...:22:59.log 17:24
PerlJam hoelzro: !!! is die, ... is fail, and ??? is warn. 17:25
(I don't think rakudo does all of those yet)
17:25 hercynium left
hoelzro PerlJam: that's what I thought 17:26
I implemented them for Pugs a few years ago, but I don't know if they've changed
17:33 kst left 17:34 kst joined 17:37 r0bby joined 17:40 payload joined 17:49 abra left
colomon 22222+ 17:53
whoops. never mind.
18:09 arthur-_ joined 18:10 payload left
hoelzro does Perl 6 have a facility to intercept unknown method calls (like AUTOLOAD in Perl 5)? 18:14
moritz_ it does (but NYI in Rakudo)
grep for CANDO 18:15
hoelzro grep the specs?
18:15 s1n left, molaf joined
moritz_ yes 18:17
hoelzro mkay
PerlJam o/~ I got the horse right here 18:23
The name is Paul Revere
And here's a guy that says that the weather's clear
Can do, can do, this guy says the horse can do
If he says the horse can do, can do, can do.
Oops, that all should have been one line.
moritz_ it's visually much nicer on multiiple lines ;-)
18:23 Chillance joined 18:24 eternaleye left 18:30 payload joined 18:31 eternaleye joined
pmichaud I like it on multiple lines. 18:35
although I thought it was "if the weather's clear" 18:36
pmurias hoelzro: or better ack the specs ;)
PerlJam yeah.
pmichaud buubot: spack CANDO
buubot pmichaud: Couldn't match input.
hoelzro pmurias: I actually did use ack to find it =)
PerlJam Not enough Guys & Dolls material in Perl 6 :) 18:37
pmichaud PerlJam++ # bringing Broadway to #perl6
18:37 fridim_ joined 18:41 arthur-_ left 18:42 arthur-_ joined, arthur-_ left
pmurias what would be even better then ack would be an ack using Perl 6 rules 18:45
alester Rewrites welcome. :-)
18:46 frodwith left
PerlJam I even started an ack-a-like for perl6 a while back. Didn't get very far beyond playing with recursing into dirs though 18:49
alester My first Perl 6 project is going to be writing File::Next in it.
pugs_svn r27782 | pmurias++ | [re-smop] 18:50
r27782 | pmurias++ | implemented until, infix:< <= >, infix:< >= >
r27782 | pmurias++ | added ../../t/spec/S04-statements/until.t to TESTS
moritz_ btw KyleHa has made it into the top 10 test suite committers 18:55
19:00 Front_slash left
KyleHa Woohoo!! 19:02
I been gone for two hours, and I come back to find that. It's like dessert. 19:03
moritz_ acutually you're top 7 ;-) 19:04
KyleHa Is there a list somewhere?
moritz_ I can paste one, hold on a sec... 19:05
lisppaste3 moritz_ pasted "top contributors to the test suite" at paste.lisp.org/display/84340
KyleHa Wow, neat. Thanks! 19:06
moritz_ it's basically the same script as in rakudo's tool/commit-stats.pl 19:07
used on the git-svn clone of t/
KyleHa I'll have a look at that when I get a chance.
I'm in a big-ish meeting right now, almost participating.
moritz_ lol 19:08
jnthn didn't know we have a commit-stats script
19:08 fridim_ left
moritz_ 3 19:08
jnthn aww 19:09
List form of pipe open not implemented at tools\commit-stats.pl line 8.
moritz_ windows :/ 19:10
jnthn: you can try open my $c, 'git log --pretty=format:%ae|%ci |' 19:11
instead
it's less secure, but probably implemented on windows
jnthn moritz_: heh, that just gives me a different error ;-) 19:12
windows :/
moritz_ jnthn: which one now?
jnthn '%ci' is not recognized as an internal or external command, 19:13
operable program or batch file.
Died at tools\commit-stats.pl line 20.
moritz_ ah, that's the close() line
you can comment that out and ignore it, I guess :)
[particle] it's the pipe in the format
jnthn ah 19:14
[particle] need double quotes after pretty=
moritz_ or you use a ':' and change the split in line 13
jnthn payload: bwa, that complains Can't run git log: Invalid argument. ;-)
19:15 sri_kraih left
moritz_ should I just nopaste the output for you? ;-) 19:15
jnthn same with colons
bleh
heh, if you want
I'm not that bothered to know if I'm beating pmichaud yet. ;-)
[particle] you need git installed and in path
jnthn Just hadn't even realized we had such a script. 19:16
[particle] hrmm,
git log --pretty=format:%ae:%ci
works for me
lisppaste3 moritz_ pasted "Rakudo top committers" at paste.lisp.org/display/84341
jnthn [particle]: Works for me too at the momment line. Hmm.
[particle] nice slip. 19:17
jnthn Awww...I'm still WAY behind.
PerlJam hey! I made the top 15. /me is shocked
jnthn Gah! s/moment/command/
[particle] jnthn: you're 70% there
jnthn [particle]: Yeah, but it's a moving target. ;-) 19:19
(Which is a *good* thing!) 19:20
mberends jnthn: if your % score is rising, you can extrapolate to the 100% time and beyond ;) 19:23
19:24 justatheory left 19:33 kst left, kst joined 19:35 hercynium joined
pugs_svn r27783 | pmichaud++ | [t/spec]: Refudge test in S32-list/grep.t . 19:38
jnthn pmichaud: I'm going to fix the WALK tests that I somehow missed yesterday later on today. 19:39
pmichaud okay, good. I was just looking at those.
jnthn Yeah, it was late, I was tired... :-)
pmichaud I know that I fixed the grep.t tests yesterday as well, but apparently the commit got lost somewhere.
jnthn And a little eager to get the various assorted fixes landed.
moritz_ pmichaud: "Refudge" - when had it was it unfudged? 19:40
pmichaud moritz_: the test itself needed some refactoring. Perhaps "refudge" is the wrong word tehre.
For one, it's using $_ as a rw param, and I'm not sure that's correct yet.
jnthn That is dubious, yes. 19:41
19:41 justatheory joined
moritz_ $_ defaults to is context<rw> I thought? 19:41
pmichaud that says it's rw to inner contexts
it doesn't say anything about rw-ness of the outer value it's initially bound to
moritz_ oh.
pmichaud S04 says it's rw 19:42
If you rely on C<$_> as the implicit parameter to a block,
then C<$_> is considered read/write by default. That is,
the construct:
for @foo {...}
is actually short for:
for @foo <-> $_ {...}
so Rakudo seems to get that wrong.
although earlier in the synopsis we have 19:43
$func = { .print if $_ }; # Same as: $func = -> $_ { .print if $_ };
I'm guessing that should be rewritten to be <-> instead of ->
(that last line is what was throwing me off) 19:44
TimToady yes, s/b <->
pmichaud (fixing) 19:45
jnthn Aha.
OK, good to know.
pmichaud fixed.
pugs_svn r27784 | pmichaud++ | Fix incorrect specification of implicit $_ in blocks. 19:46
pmichaud when we fix it in rakudo, the grep.t test should pass (and should be unfudge-able)
defudged
jnthn unfudged
;-)
pmichaud does anyone else besides me dislike the output format of Test::Harness 3?
I find it much harder to read the summary. 19:47
(compared to earlier versions)
moritz_ it took me a while to get used to it
pmichaud what's worse, passing TODOs are visually indistinct from FAILS
jnthn That irks me. 19:48
I've been confused by that before now.
moritz_ same here
(speaking of which, I should clean up the new passing TODOs)
pmichaud I'm doing some now.
So I can get them out of my test output :-)
moritz_ ok, I'll let you do it then 19:49
pugs_svn r27785 | pmichaud++ | [t/spec]: Unfudge passing todo in S02-*/type.t .
r27786 | pmichaud++ | [t/spec]: Unfudge passing todo test in S12-enums/thorough.t . 19:50
jnthn pmichaud: lol, I fudged fails in S12-enums/thorough.t yesterday 19:51
jnthn wonders if we came out equal on balance. 19:52
19:58 donaldh joined
jnthn OK, this is The Cool 19:59
If you look at S12, and exclude the multiple constraints section which is marked as not 6.0.0 anyway
There's only "=head1 Class methods" that we don't really have any coverage of at all in Rakudo. 20:00
And hey, it's in my Hague Grant to do that. :-)
jnthn reads and ponders how to do it
20:01 snearch joined
TimToady moritz_: have answered your p6l 20:02
but hasn't left my machine yet, sigh
pmichaud oh
pmichaud ran to the mailing list and saw.... nothing.
jnthn TimToady: "Alternately, you can associate a class method with the current metaclass instance, 20:05
"
TimToady: Would a reasonable implementation of this be to collect such methods into some anonymous role and mix it in to the metaclass instnace?
(This is from section I just mentioned in S12) 20:06
Or does that cry out "OH NOES" to you?
TimToady nothing OH NOES me, but then few things do offhand. :) 20:08
20:08 desertm4x_ left
jnthn :-) 20:08
OK, I'll try it that way.
std: class Foo { method ^bar() { } } 20:10
p6eval std 27786: OUTPUT«ok 00:02 37m␤»
TimToady moritz_: okay, went thru 20:11
pmichaud TimToady: your explication matches my understanding of things :-) 20:15
pmichaud re-reads to make sure.
moritz_ TimToady: thanks, I'll read it carefully when I'm not on phone anymore ;-)
pmichaud but yes, the basic idea is that @a[0] = ... is a list assignment and not item assignment 20:16
20:18 campuscodi joined
campuscodi hello. is anybody available to answer me a question ? 20:19
pmichaud "Yes."
hoelzro campuscodi: you only get one, and that was it. 20:20
campuscodi i'm a news writer for softpedia.com (webmasters section) I would like to write an article about current Perl 6 development. Can you point me at some articles or pages were I could get some knowledge ?
I'm interested in new features, important changes and so on 20:21
pmichaud sdtimes.com recently ran an article about Rakudo Perl 6
but that was more about availability than Perl 6 features
start with perl6-projects.org 20:22
jnthn didn't see that article
pmichaud also check planetsix.perl.org
jnthn: www.sdtimes.com/link/33566 20:24
campuscodi ok. thank you very much. if I find enough good information around and I'll get to write about this topic I'll be sure to credit you :P @pmichaud 20:25
20:25 kst left
pmichaud campuscodi: that would be great. 20:25
moritz_ campuscodi: if you have more specific questions, also feel free to ask here
pmichaud campuscodi: if you need more information, here's a good place to ask
[particle] you two beat me to it
20:25 kst joined
[particle] pmichaud: are your oscon slides online? 20:26
they should have details on new features etc
jnthn pmichaud: nice article
pmichaud [particle]: my oscon talk wasn't about perl 6 features -- that's normally TheDamian and TimToady's talk 20:27
[particle] ah, right, silly me.
pmichaud (as was the case again this year)
that said, my slides are online in their usual place: pmichaud.com/2009/pres
campuscodi actually this lecture en.oreilly.com/oscon2009/public/sch...etail/8061 made me wanna write something about the topic 20:28
[particle] wasn't sure if you'd uploaded the latest yet, i'll take a look at your revamped pres
pmichaud it was substantially reworked for oscon, yes.
I like the new talk much better than the one I gave at yapc::na.
(it's much more focused)
cbk In perl 6 is there some trailing white space remover like chomp or I think I've seen something like trim...? 20:29
pmichaud cbk: trim exists.
rakudo: say "Hello world ".trim.chars
p6eval rakudo ea667e: OUTPUT«11␤»
moritz_ rakudo: say ' Hello World '.trim.perl
p6eval rakudo ea667e: OUTPUT«"Hello World"␤»
cbk so I can do $myVarHasASpaceAtEnd.trim 20:30
what is that .perl
moritz_ cbk: also .chomp exists, which just removes a trailing newline, and return s the modified string
cbk: .perl is like Data::Dumper
ie it returns perl code that builds the data structure 20:31
cbk is there a consolidated list of all of these .coolFunc (.trim and .perl) out there?
they are like built in functions 20:32
moritz_ cbk: perlcabal.org/syn/ scroll down to S32
[particle] oh, pmichaud, yesterday i built and tested ins2 with msvc. all's well.
cbk thanks :-) 20:33
pmichaud [particle]: \o/
[particle]: that's really good to hear.
I know there's an issue with Solaris builds... have to work on that one a bit. 20:34
20:34 Whiteknight left
[particle] pmichaud: where else do you need coverage? 20:34
pmichaud what is Parrot's list of platforms?
if we cover the Parrot list I'm happy for now. 20:35
campuscodi thank you very much guys. have a good night .... or day :)) 20:38
20:39 szabgab left
pmichaud rakudo: enum Color <white gray black>; say ?(Color.pick == any(Color::white, Color::gray, Color::black)); 20:40
p6eval rakudo ea667e: OUTPUT«Method 'Num' not found for invocant of class ''␤»
pmichaud jnthn: ^^^ # I think this is a related fail to changes made yesterday
(t/spec/S12-enums/basic.t) 20:41
jnthn Oh?
hmm, it fails here too :-S
I was sure that passed :-(
pmichaud I'm not sure if I introduced the fail or you did. 20:42
20:42 szabgab joined
jnthn I remember fuding a different test in that file yesterday. 20:42
I thought I had a clean run on it after that.
Though I could easily be mistaken.
It involves junctions, which would make what I did feel more likely to be to blame.. 20:43
OTOH, all of the junctions tests passed fine (I didn't fudge any of the S03-junctions/*.t)
pmichaud I adjusted Code.ACCEPTS and Regex.ACCEPTS, so that seems less likely to be the culprit. 20:44
jnthn Right.
20:46 campuscodi left
dalek kudo: 7700f40 | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 423 files, 12033 passing, 14 failing

   S12-enums/basic.rakudo aborted 2 test(s)
   S12-introspection/walk.t aborted 6 test(s)
   S32-list/grep.rakudo aborted 6 test(s)		 20:46
20:46 limpmsngr joined
cbk rakudo: my $textLine2 = "mrGreen 01 02 123 "; say "Patern: $textLine2"; if $textLine2 ~~ /(<[a..zA..Z]>+\s)(\d+\s)(\d+\s)(\d+)/ {say "last match vars are: $0,$1,$2,$3 "; $name = $0.trim; } say "Name is:$name."; 20:48
p6eval rakudo ea667e: OUTPUT«Statement not terminated properly at line 2, near "say \"Name "␤in Main (src/gen_setting.pm:3343)␤»
20:49 snearch left
moritz_ cbk: there's a non-printable character in your input, before the 'say' 20:49
probably from copy & pasting
cbk oh ok let me try again...
my $textLine2 = "mrGreen 01 02 123 "; if $textLine2 ~~ /(<[a..zA..Z]>+\s)(\d+\s)(\d+\s)(\d+)/ {say "last match vars are: $0,$1,$2,$3 "; $name = $0.trim; } say "Name is:$name."; 20:50
rakudo: my $textLine2 = "mrGreen 01 02 123 "; if $textLine2 ~~ /(<[a..zA..Z]>+\s)(\d+\s)(\d+\s)(\d+)/ {say "last match vars are: $0,$1,$2,$3 "; $name = $0.trim; } say "Name is:$name."; 20:51
p6eval rakudo ea667e: OUTPUT«Statement not terminated properly at line 2, near "say \"Name "␤in Main (src/gen_setting.pm:3343)␤»
cbk well it works on my perl 6 box
PerlJam missing semicolon
rakudo: if 1 { say "hi" } say "bye"; 20:52
p6eval rakudo ea667e: OUTPUT«Statement not terminated properly at line 2, near "say \"bye\";"␤in Main (src/gen_setting.pm:3343)␤»
PerlJam rakudo: if 1 { say "hi" }; say "bye";
p6eval rakudo ea667e: OUTPUT«hi␤bye␤»
cbk the .trim works just fine. if I do the .trim.perl it put the var in " "'s
dalek kudo: 426942a | pmichaud++ | src/builtins/eval.pir:
Correct problem with statement-style package declarators (RT #67976).
 20:53
20:53 limpmsngr left 20:54 hercynium left 20:57 SmokeMachine left 20:58 kst left 20:59 kst joined 21:02 Su-Shee left
dalek kudo: 5b477c0 | pmichaud++ | docs/release_guide.pod:
Note the release manager for each of the previous releases,
 21:05
PerlJam uhoh ... it's "for real" now :) 21:06
21:07 pmurias left
pmichaud OKAY! WHO ELSE IS GOING TO DO A RELEASE?!? 21:07
VOLUNTEERS NEEDED NOW, OR I'LL START PICKING NICKS AT RANDOM!
I've got a commit bit, and I"M NOT AFRAID TO USE IT!!!
jnthn OMFG 21:08
jnthn runs away from the mad Texan
pmichaud You there! Yes, you with the beers! When's your release?!? ;-) 21:09
jnthn Erm.
Not September because I'm on vacation.
21:10 jauaor left
jnthn Feel free to put me down for Oct/Nov/Dec. 21:10
pmichaud Oct is taken.
jnthn Erm
pmichaud November it is.
although maybe we should give masak++ November :-P
jnthn that is <oct nov dec>.pick, not .all ;-)
Oh yes!
PerlJam yeah, masak should get November
moritz_ wanted to suggest the same thing too
pmichaud Then he can really be in charge of "the November release" :-)
jnthn I think he'll be approving. :-) 21:11
moritz_ and "November.pm" gets a new meaning ;-)
pmichaud I'll reserve November for masak++, if he'll accept.
jnthn: you're on the hook for Dec, unless I can get other volunteers
21:11 iblechbot joined
jnthn pmichaud: That's fine. I'm happy to step aside to let others get involved though. 21:11
PerlJam no one wants August or September. :) 21:12
pmichaud (I'm wanting to increase the number of committers/participants; since you already have plenty of commits, we'll save it for another time)
jnthn pmichaud: Given I'm already plenty involved. ;-)
pmichaud jnthn: indeed.
jnthn But if nobody steps up, I'm happy to do it.
pmichaud well, I think I'll be recruiting folks to do it at yapc::eu :-)
jnthn Yeah!
pmichaud "Immortalize your name in Perl 6. Become a release manager. Half a days work for immortality." 21:13
jnthn just wrote some code in actions.pm, looked at it, thought "pm will kick my ass for this" and ended up re-writing it better :-)
pmichaud jnthn++ # smart move
moritz_ code review++
.oO( immoralize your name )
 21:14
jnthn It gets kinda fun when your code review comes down to, "what would this person say if they looked at my code" :-)
pmichaud it's nice that the mere threat of code review is enough to get someone to rewrite code :)
[particle] pm: list me 21:15
moritz_ [particle]: when? August? September? 21:16
pmichaud [particle]: pick your month. October is taken.
moritz_ so is November (probably)
jnthn pmichaud: Was there a way in PAST to say "I want to look up this PAST::Block"?
pmichaud jnthn: you can give the block a .subid, and then use that .subid in a reference
look for ".subid" in actions.pm for an example 21:17
jnthn Ah, that'll do it.
[particle] i'm doing parrot in september, might as well do rakudo too
jnthn Go for the double!
[particle] and i'd like to name it for seattle.pm
pmichaud particle: done
jnthn: you can also request a .subid from a PAST::Block, which will bind it at that point
(and generate a unique subid for you)
jnthn pmichaud: even beter 21:18
Thanks.
jnthn doesn't find .subid anywhere in actions.pm :-S 21:19
21:19 payload left
pmichaud oh 21:20
hmmm
I know there's a way to get a reference to a blockk, yes.
don't remember what it is, then
21:24 SmokeMachine joined
dalek kudo: 2665eef | pmichaud++ | docs/release_guide.pod:
Add particle++ for the September release.
 21:24
21:27 cj left 21:32 snearch joined 21:33 zamolxes joined
jnthn gah! Parrot's MMD strikes again... :-/ 21:35
21:36 molaf left
pmichaud afk, errands 21:36
21:37 dukeleto joined 21:41 cdarroch left 21:42 maja__ joined 21:45 skids_ left
jnthn TimToady: Hmm. One curiosity of a method ^count { return $count } 21:46
Is that when invoked as $instance.^count
Then it'll be given $instance as a first parameter
I guess that's fine, you just gotta declare it with a signature that's sufficiently permissive. 21:47
(e.g. that accepts that parameter)
21:47 xomas_ joined 21:48 snearch left
jnthn pmichaud: block does have a .subid, we just weren't using it. 21:49
21:53 KyleHa left
jnthn And it worked just how I wanted. Great. Now it works _and_ the generated code doesn't suck. 21:53
21:54 maja_ left
jnthn TimToady: As a minor side-note, I couldn't follow STD.pm in method_def because it doesn't capture the <[ ! ^ ]> in any way, and it needs capturing for the action method. 21:56
rakudo: enum Color <white gray black>; say Color.pick 22:05
p6eval rakudo 2665ee: OUTPUT«Color()␤»
moritz_
.oO( masak! )
 22:07
it seems to treat Color as a 1-item list
rakudo: enum Color <white gray black>; say white; 22:08
PerlJam But it is a 1 item list.
p6eval rakudo 2665ee: OUTPUT«0␤»
moritz_ PerlJam: but .pick is defined for enums too
and that candidate should win
jnthn Yeah
It should.
We already have a test fail at the moemnt from this. 22:09
moritz_ then I'm happy 22:10
I thought it was another testing FAIL
jnthn No
I think I know what the issue is.
But I'm not quite sure how to fix it.
moritz_ with magic. 22:11
jnthn Basically, we want a role type to be narrower than Any
So sigil based dispatch works.
But now the dispatch, when punning, will see the pick method from Any and choose that one.
Rather than punning the role and calling the .pick 22:12
rakudo: enum Color <white gray black>; say Color.new.pick
p6eval rakudo 2665ee: OUTPUT«1␤»
jnthn Aye
I think it's an issue with the way we pun.
moritz_ bad puns? ;-) 22:13
would be nothing new in here ;-)
jnthn Aye, bad puns. 22:14
Well, lack of puns actually.
Which is not a #perl6 problem. ;-0
[particle] seven days without puns makes one week
jnthn I haven't got a quick fix that isn't an epic hack. 22:16
I think the answer is "fix role punning" though.
22:21 masak joined
jnthn decides to sleep on it and look again tomorrow 22:21
lolizmasak
masak jnthn: sleep tight.
yes, who cannot sleep.
jnthn masak: Oh, I'm not going to bed right now.
masak ok, good.
jnthn masak: I'd not be able to sleep this early.
masak because I found a bug for you. :)
jnthn masak: I just don't have a good answer for something tonight and am hoping sleep will magically make things clear to me. ;-) 22:22
masak it sometimes does.
jnthn And your bug, sir?
masak jnthn: well, I'll try and minimize it, but if you're impatient, just do 'perl6 druid'. 22:23
it's right there.
I can paste the result of that one right now and then try to minimize. 22:24
the error looks like something you did. :)
moritz_ masak: we discussed prospective Rakudo release mmanagers earlier... would you do the November release?
jnthn Heh, well, I did stop Rakudo lying to itself in various ways last night.
22:24 donaldh left
jnthn It's entirely possible there were some places the tests didn't catch. 22:25
masak moritz_: that would be very fitting. yes, please.
jnthn: gist.github.com/157719
jnthn: don't know when the bug appeared, because apparently another bug has hidden it for two months. pmichaud++ fixed that one today.
jnthn Hmm
masak "November has been released?" -- "Nono, it's the November release of Rakudo!" -- "Ah, Rakudo November?" -- "No, it's called Rakudo Baghdad." 22:26
jnthn: I'll try to make the bug Druid-independent.
moritz_ in method Druid::Game::Subject[]::attach (/Users/masak/gwork/druid/lib/Druid/Game/Subject.pm:21) 22:28
what does the [] mean?
that it's a role? 22:29
masak yes.
moritz_ or a parameterized role?
masak it's not.
at least not by me.
dalek kudo: e2d6f13 | jnthn++ | src/ (3 files):
Implement declaration of methods on the metaclass (e.g. method ^foo($obj, ... ) { ... }). Also we toss the use of Parrot's MMD in calling !meta_compose; it's less code to not use it anyway.
kudo: 6999e58 | moritz++ | docs/release_guide.pod:
[docs] masak++ is release manager for November
moritz_ ok, I clearly haven't written enough in Perl 6 yet ;-)
masak moritz_: waitwait, I still get to prove myself, right?
I was looking forward to that.
it seemed fun. 22:30
22:30 lumi_ left
moritz_ masak: yes. But you have more than 3 months, I'm sure you can do it in that time ;-) 22:30
masak hopes so
22:30 nihiliad left 22:32 _Jedai_ joined
jnthn sorry, net connection had a fail 22:33
masak argh, this bug is slippery. :/
jnthn what does the [] mean? # that it's a version of the role that takes no parameters
All roles are parametric in a sense.
role Foo { } is just like role Foo[] { } 22:34
Since you'd also want having declared either of those to write role Foo[$bar] { } too
moritz_ I expected something along these lines, thanks
multi role Foo[$bar, $baz] ... :/
jnthn is impressed that scp that is backing up a bunch of files from his server was able to resumre without any input after the connection came back to life. 22:35
moritz_: Well, yeah, but the multi is implicit.
22:36 jferrero joined
jnthn But under the hood, it does use the multi-dispatcher. 22:36
moritz_ so it does ordinary multi dispatch under th... wow
jnthn Yeah, so you have the full power of multiple dispatch available to you when deciding which role you're going to compose.
moritz_ rakudo: role A[*@b] { method x { say @b.perl } }; class B does A[2, 3, 4] { }; B.new.x 22:40
p6eval rakudo 2665ee: OUTPUT«[2, 3, 4]␤»
moritz_ that *is* scary.
masak is scared
moritz_ must remember to write some tests for that
masak :)
moritz_ but first: sleep. 22:41
masak rakudo: class A is B {}; class B {}
p6eval rakudo 2665ee: OUTPUT«No applicable candidates found to dispatch to for 'trait_mod:is'␤in Main (/tmp/iFdZb7lajB:2)␤»
masak that error message, can it be made slightly more awesome?
moritz_ rakudo: class A is B {};␤class B {} 22:42
p6eval rakudo 2665ee: OUTPUT«No applicable candidates found to dispatch to for 'trait_mod:is'␤in Main (/tmp/W987Ujtr2b:2)␤»
jnthn Well, it's telling you the truth. ;-) 22:43
22:43 al23 joined
masak jnthn: that's why I didn't say "o_O" or "wtf?" :) 22:44
jnthn masak: Yeah. I guess
masak just talking about aesthetics.
jnthn ponders
rakudo: multi trait_mod:<is>(Class A, *%foo) { die "Maybe you did stuff in the wrong order" }; class A is B { }; class B { }; 22:45
p6eval rakudo 2665ee: OUTPUT«Unable to parse multisig; couldn't find final ')' at line 2, near "A, *%foo) "␤in Main (src/gen_setting.pm:3343)␤»
jnthn oh 22:46
rakudo: multi trait_mod:<is>(Class A, *%foo) { }
p6eval rakudo 2665ee: OUTPUT«Unable to parse multisig; couldn't find final ')' at line 2, near "A, *%foo) "␤in Main (src/gen_setting.pm:3343)␤»
jnthn oh duh
i fail
rakudo: multi trait_mod:<is>(Class, *%foo) { }
p6eval rakudo 2665ee: ( no output )
jnthn rakudo: multi trait_mod:<is>(Class, *%foo) { die "Maybe you did stuff in the wrong order" }; class A is B { }; class B { }; 22:47
p6eval rakudo 2665ee: OUTPUT«Maybe you did stuff in the wrong order␤in sub trait_mod:is (/tmp/Nax3qB8JnT:2)␤called from Main (/tmp/Nax3qB8JnT:2)␤»
jnthn rakudo: multi trait_mod:<is>(Class, *%) { die "No matching trait; did you forget to forward-delcare a class?" }; class A is B { }; class B { };
22:47 |Jedai| left
p6eval rakudo 2665ee: OUTPUT«No matching trait; did you forget to forward-delcare a class?␤in sub trait_mod:is (/tmp/9otMf9SIEy:2)␤called from Main (/tmp/9otMf9SIEy:2)␤» 22:48
jnthn Guess we could always stick something like that in traits.pm
22:49 hoelzro left
masak jnthn: I think that would be an improvement. 22:49
al23 Hi! Can't build pugs. What's wrong here: paste.lisp.org/display/84353
jnthn rakudo: multi trait_mod:<is>(Class $a, *%f) { die "No matching trait { %f.keys[0] } to compose into $a; did you forget to forward-delcare a class?" }; class A is B { }; class B { };
p6eval rakudo 6999e5: OUTPUT«No matching trait B to compose into ClassHOW()<0xb66c0aa0>; did you forget to forward-delcare a class?␤in sub trait_mod:is (/tmp/2vHgf0zryA:2)␤called from Main (/tmp/2vHgf0zryA:2)␤»
jnthn oh lol
masak al23: cannot configure time-1.1.4 22:50
jnthn rakudo: multi trait_mod:<is>(Class $a, *%f) { die "No matching trait { %f.keys[0] } to compose into { $a.name }; did you forget to forward-delcare a class?" }; class A is B { }; class B { };
masak al23: that's what's wrong.
p6eval rakudo 6999e5: OUTPUT«Method 'name' not found for invocant of class 'ClassHOW'␤»
jnthn heh
well
Close anyway.
masak: We can give it a try.
jnthn afk for a little bit 22:51
al23 masak: so?
masak al23: good question. I haven't tried building pugs in months.
al23: it's kind of a dormant project. 22:52
al23 Me too.
22:52 hercynium joined
masak al23: you do know of Rakudo and all? just checking. 22:52
al23 I thought Pugs still the most comprehensive...
Isn't it? 22:53
masak al23: certainly not the most up-to-date.
al23: I find it hard to compare comprehensiveness.
if it's features you want, Rakudo won't disappoint you, though.
al23 By the amout of implemented stuff?
22:54 cj joined
masak al23: funny guy. :) 22:54
al23 Anyway, I'm just want to play with Perl 6.
jnthn al23: They both implement different subsets of Perl 6, though increasingly Rakudo covers all that Pugs does. Plus Rakudo does a bunch of stuff that Pugs doesn't. Plus it is much more in line with the current spec.
masak al23: then I suggest you try Rakudo.
jnthn Rakudo misses, for example, laziness. 22:55
masak we're putting off implementing it. :P
22:59 lumi_ joined
al23 Hm. I think I've used only pugs... I thought it is really cool since it was written (a) in Haskell and (b) by Audrey Tang :-) 23:04
23:05 M_o_C left
masak al23: nod. 23:05
I thought that was really cool too.
al23 Is it completely stuck? I think Audrey wrote that she plans to resume work. 23:07
23:10 synth joined 23:12 synth left
masak al23: it's been basically motionless for two years. 23:12
23:16 synth joined
masak I'm getting close to catching the bug now. 23:16
23:16 Whiteknight joined 23:18 justatheory left 23:22 justatheory joined
japhb TimToady: could you explain your email statement "I wish we had a way of trapping and testing warnings too so we could see that 3 elements were discarded by the inner list assignment." To whom is this directed? 23:23
masak ok, here it is: 23:24
rakudo: role A { has @!foo; method bar() { @!foo ~~ (1,2); } }; class B does A { has $.baz; }; B.new().bar()
p6eval rakudo 6999e5: OUTPUT«Null PMC access in isa()␤in method A[]::bar (/tmp/XG0DKmk8uZ:2)␤called from Main (/tmp/XG0DKmk8uZ:2)␤»
masak submits rakudobug
rakudo: role A { has @!foo; method bar() { say @!foo } }; class B does A { has $.baz; }; B.new().bar() 23:27
p6eval rakudo 6999e5: OUTPUT«Null PMC access in isa()␤in method A[]::bar (/tmp/CfCtQ08n7K:2)␤called from Main (/tmp/CfCtQ08n7K:2)␤»
TimToady japhb: it's directed at whoever wants to think about how you'd test for proper emission of warnings
23:28 iblechbot left
TimToady which is, presumably, related to the question of how you test for proper production of any exception 23:29
japhb TimToady: Ah. I was wondering if it was a note to yourself, or a gentle reminder to read already-specced behavior, or ....
TimToady it's just a To Whom It May Concern 23:30
japhb gotcha.
al23 \o/ GHC 6.10.x builds Pugs. 23:31
jnthn masak: Hmm. 23:32
Interesting bug...
masak jnthn: packaged and filed, just for you sir.
fix it, and Druid will run again. :)
jnthn masak: Is that minimal?
masak jnthn: very. 23:33
see the bug ticket.
RT 67992.
jnthn rakudo: role A { has @!foo; method bar() { say @!foo } }; class B does A { }; B.new().bar()
p6eval rakudo 6999e5: OUTPUT«␤»
jnthn huh? Having another attr in the class is what breaks it?
:-/
masak yes.
that actually seems to be the problem.
rakudo: class A { has @!foo; method bar() { say @!foo } }; class B is A { has $.baz; }; B.new().bar() 23:34
p6eval rakudo 6999e5: OUTPUT«␤»
masak seems to be specific to role composition.
jnthn Do you know how recently it broke?
masak jnthn: unfortunately not.
jnthn I'm pondering it may be to do with that attribute ordering bug.
That is, role attr composition doesn't contribute to that list. 23:35
And then initialization leaves the attr null.
That'd nicely explain it.
masak aye.
jnthn Yes, looks very much like that. 23:36
Will patch it tomorrow.
masak \o/ 23:37
my work here is done.
see you tomorrow for another glorious day.
23:37 synth left, masak left
jnthn Optimism. He's doing it right. 23:39
al23 Should <<op<< or >>op>> work, besides >>op<<? 23:40
TimToady that one works better in rakudo than in pugs
23:40 nihiliad joined
al23 :-) 23:40
TimToady spec rot 23:41
al23 I see.
TimToady though I think pugs's >>op<< is really the currently specced <<op>>
as in it dwims
pugs: say (1,2,3)>>+<<1 23:42
p6eval pugs: OUTPUT«234␤»
TimToady yes
al23 So, what's "hyper op xs ys = map (\ (x,y) -> x `op` y) $ zip xs ys" in Perl 6?
TimToady rakudo: say (1,2,3)>>+<<1
p6eval rakudo 6999e5: OUTPUT«Non-dwimmy hyperoperator cannot be used on arrays of different sizes or dimensions.␤in Main (/tmp/wOVfrynyA3:2)␤»
TimToady should fail
rakudo: say (1,2,3)<<+>>1
p6eval rakudo 6999e5: OUTPUT«Statement not terminated properly at line 2, near "1"␤in Main (src/gen_setting.pm:3359)␤»
TimToady rakudo: say (1,2,3) <<+>> 1
p6eval rakudo 6999e5: OUTPUT«234␤» 23:43
jnthn std: say (1,2,3)<<+>>1
p6eval std 27786: OUTPUT«===SORRY!===␤Confused (two terms in a row?) at /tmp/fcG3J9PzNo line 1:␤------> say (1,2,3)<<+>>⏏1␤ expecting any of:␤ POST␤ bracketed infix␤ infix stopper␤ postfix␤ postfix_prefix_meta_operator␤ standard stopper␤
..statement modifier loop␤ term…
TimToady that's a subscript
needs ws
jnthn Aha
Phew, Rakudo was right. :-)
TimToady earlier you were using Class as if it were a real one
al23 Where is the actual spec? 23:44
TimToady despite S02 saying there's a Class class, I don't believe it
23:44 Khisanth_ joined
jnthn TimToady: There is no Class class. 23:44
TimToady S02 is actually talking about varieties of packages
23:44 Khisanth_ left
jnthn TimToady: Class is, in liu of you spec'ing what we should have, the "some HOW here" marker. 23:44
TimToady so I think those are really roles for mixing into Package 23:45
jnthn TimToady: Right now it looks like this:
23:45 jferrero left
jnthn subset Class of Object where ClassHOW | RoleHOW; 23:45
TimToady subset Class of Object where { not .defined }
jnthn No, that's not right.
The trait_mod:<is> wants to take a metaclass. 23:46
Which could well be defined.
It's not the proto-object.
TimToady no, it wants the object being declared
it must have the object being declared, or it can't modify it
jnthn Hang on, which are we talking about?
class Foo is Bar { }
dispatches with two parameters: the metaclass of Foo, and then also Bar, which would be the proto-object. 23:47
TimToady somewhere there is a type object named "Foo" that is separate from the metaclass
jnthn It doesn't exist at that point.
We're still in the process of declaring the class.
TimToady it's being created
23:48 jferrero joined
jnthn Yes, in the process of creating it. 23:48
TimToady in STD it already exists, and is called the "declarand"
jnthn Last time we discussed this, you said you'd expect the trait_mod there to receive the metaclass.
TimToady I don't recall that
jnthn Well it's not likely I invented it, given it's made me do a refactor.
It kinda makes sense you'd want to meta-class in there though. 23:49
That's the thing you want to ask to add a parent, for example.
TimToady you can always the the metaclass from the object's .HOW
you know the .HOW based on the keyword "class", "role", etc. 23:50
the name has to be introduced as a valid type name as soon as you see it
even before processing the traits 23:51
jnthn The parser knows it's a valid type name.
But that's an unrelated issue to having to actually create the proto-object at that point.
TimToady the parser knows that because the type object is added to the symbol table right then, in STD
the symbol table entry *is* the protoobject in sTD 23:52
jnthn Sure, it is in Rakudo *after* the class has been defined.
(We create the proto-object after the class has been composed.) 23:53
TimToady what does the symbol table contain, if not the protoobject? 23:54
jnthn The actual namespace entry doesn't get created at that point, it's the parser's symbol table that knows about it, and all it needs to know is "there's something under this name". 23:55
TimToady the parser's namespace *is* the symbol table, for STD 23:56
jnthn OK, but that doesn't really fly for Rakudo. 23:57
Well
I guess it could somehow be made to...
TimToady: I knew I wasn't making crap up. 23:59
irclog.perlgeek.de/perl6/2009-05-14#i_1143515

Powered by Cro and the Raku® Programming Language.

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

Thank you!