»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! | Rakudo Star Released!
Set by diakopter on 6 September 2010.
00:00 orafu joined
ingy star: $_ = 'hello'; .say() 00:00
p6eval star 2010.09: ( no output )
ingy star: 'hi'.say 00:01
p6eval star 2010.09: ( no output )
ingy star: say 'hi' 00:04
p6eval star 2010.09: ( no output )
sjohnson rakudo: say 'hi'
p6eval rakudo : OUTPUT«hi␤»
sjohnson
.oO(???)
ingy rakudo: $_ = 'hi'; .say 00:05
p6eval rakudo : OUTPUT«hi␤»
ingy rakudo: $_ = 'hi'; .= ' world';.say
p6eval rakudo : OUTPUT«===SORRY!===␤Confused at line 22, near ".= ' world"␤»
ingy rakudo: $_ = 'hi'; $_ .= ' world';.say
p6eval rakudo : OUTPUT«===SORRY!===␤Quoted method name requires parenthesized arguments at line 22, near ";.say"␤» 00:06
00:11 risou_ left
ingy sorear: what debian mono package should I install for niecza? 00:12
niecza: say 'sorear' 00:13
p6eval niecza 914f55d: OUTPUT«sorear␤»
00:13 orafu left 00:14 orafu joined 00:15 Chillance left
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....kup_tablet 00:25
diakopter ingy: I didn't know there was one..
00:26 hanekomu left 00:28 orafu left 00:29 orafu joined
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....kup_tablet 00:30
00:33 masonkramer left
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....dex_tablet 00:36
sorear ingy: dunno, I installed from source. I know you need libmono2.0-cil 00:56
moritz would have the full list
you'll also definitely need mono-gmcs and mono-xbuild 00:58
ah! mono-complete 00:59
"Install this package if you want to run software ... not ... from a Debian package"
01:00 redicaps joined 01:25 bluescreen joined, bluescreen is now known as Guest49812 01:26 nymacro joined 01:30 Guest49812 left 01:34 bluescreen_ joined 01:37 bluescreen_ left 01:40 bluescreen_ joined
sorear TimToady: What do <( and )> do in subrules? 01:43
01:48 bluescreen_ left
ingy sorear: where is the name Niecza from? 01:50
sorear masak
ingy any idea what it means
sorear it's a badly misremembered line from a joke dub of a music video on youtime 01:51
ube
ingy cool
sorear: I just showed Damian the irc log 01:53
he thought I said Nietzsche earlier 01:54
and he thinks it should actually be called Nietzsche
colomon "That which does not crash you makes you stronger." 01:55
ingy that which requires mono-complete makes you stronger
colomon ingy: ask Damian how his p6 modules are coming. :) 01:56
sorear ingy: which irc log? 01:58
also, the name Nietzsche is ... taken. 02:01
sorear can't imagine Niecza is very TheDamian-compatible 02:05
02:06 bluescreen joined 02:07 bluescreen is now known as Guest86192
TimToady sorear: all they do is change .from and .to, as usual 02:08
sorear TimToady: consider my regex foo { a )> b }; "abc" ~~ /<&foo>c/; "ab" ~~ /<&foo>b/ 02:12
which one matches?
02:19 TypeNameHere joined
TimToady the second one does 02:20
02:21 whiteknight joined
TimToady what follows )> is like a lookahead 02:21
ingy sorear: /tmp/tmp52a1123b.tmp: 1: viv: not found 02:25
what's viv?
02:30 envi joined 02:35 dd-sphere joined 02:36 gimix joined 02:45 TypeNameHere left 03:06 Guest86192 left 03:15 hercynium joined 03:21 whiteknight left 03:30 kanishka joined 03:36 satyavvd joined 03:37 QinGW joined 03:41 [particle] joined 03:55 plobsing joined 04:00 plobsing left, Chillance joined
sorear ingy: still here? 04:35
04:37 hercynium left
sorear waits for ingy to return. 04:47
05:03 kanishka left 05:26 cjk101010 joined 05:39 plobsing joined 06:06 _kaare joined 06:08 sftp left 06:11 stepnem joined
kthakore Hello, is a perl6ier way of injecting scalar references into a callback's scope? In perl5 I do $foo->set_callback( set{ _local_callback( @_, $local_in_scope_scalar ) } ) a lot. 06:17
or does that just mean a bad design .... prolly the later ...
oops also s/set/sub/g 06:18
06:22 perimosocordiae joined
TimToady my $lexical; $foo.set_callback( -> $param { whatever($param,$foo) } ); 06:22
er
my $lexical; $foo.set_callback( -> $param { whatever($param,$lexical) } ); 06:23
the lexical comes from the lexical scope, the parameter is passed by the caller of the callback
kthakore wow this is magic! 06:24
TimToady just like in Perl 5, except you get to name your parameters
kthakore I really need to get SDL working with Parrot
TimToady seems to me it was long ago, but likely has suffered bitrot 06:25
kthakore TimToady: parrotSDL ? I forked it a few months ago, but parrot was a moving target
it was working for a while though
so let me get this straight ... if I do -> $param { whatever ($param, $lexical ) } 06:26
in whatever( ) I can just us $param and $lexical? 06:27
use*
no need to do 'my ($param, $lexical) = @_' ?
TimToady yes, P6 has real parameters now
kthakore .... -.- 06:28
ok I really need to get parrotSDL caught up now
TimToady well, $lexical isn't a parameter, but yes
kthakore thanks TimToady
TimToady you could write it sub ($param) {...} if you want it to look more p5ish
kthakore ok I am confused ... 06:29
sub ($param) ?
is that like sub { my $param = shift ...
TimToady yes
kthakore ah ok
TimToady you can have multiple parameters
kthakore sorry if I didn't get that right away ... I am pulling an all nighter
What I am really interested in is the perl 6 grammar stuff ... but I am not sure if it is the same as what I am thinking. 06:30
A while ago I was making a context free grammar language in ANTLR for MAgic the Gathering card rules. 06:31
TimToady p6 grammars are something like that
kthakore Magic The Gathering Card rules are essentially ( {Target} {Action} {modifier} ... so on) 06:32
TimToady: are as powerful as context free or similar to regular expressions?
cause using regex for that stuff was really really painful 06:33
TimToady they are as powerful as you want them to be
kthakore hehe ... I should have seen that coming 06:34
TimToady they can be context sensitive if you like
kthakore context sensitive as in @_ or $ context?
TimToady that depends on your language definition
kthakore I meant context free grammars as in defining a way to 'parse' text rules
I am not really good at this stuff bascially
I want to be able to do this
' [Green Cards] {destroy} {turn} ' = Destroy all green cards this turn 06:35
where ' [Green Cards] {destroy} {turn} ' is the card rule
so this would be converted into a subroutine ... 06:36
nah ... that would be crazy ...
or possible?
TimToady anything is possible, and some things are easy
sorear doing NLP with Perl 6 is on the crazy side
but good crazy
kthakore but is it NLP?
because I use notations to tell what is the target
TimToady sounds pretty restricted to me
sorear oh
I misunderstood
thought you'd be parsing the card text 06:37
kthakore no no
that converted manually
hold on let me see if I have the XML we converted
yay! Found it! 06:39
hold on getting it some place public
kthakore dusts off the xml
sorear if you have xml why not use it?
kthakore I will explain 06:41
again you are too fast ;)
sorear: TimToady sdlperl.ath.cx/mtg/ 06:42
C[Self: | Damage:Player,Waiting](Player:Waiting | Discard:1): is pretty much the 'rule' of the card
So it does 'Damage the waiting player and make the player discard a card' 06:43
C means it is only works in a certain phase
... I forgot this ... wow this was sooo long ago 06:44
sorear: I was trying to make a grammar for this in Antlr so I could make it applicable on the game state some how
06:44 _twitch joined
kthakore my plan was to 'parse' this XML into subroutines ... 06:45
06:45 masak joined
masak morning, #perl6! 06:45
kthakore stays quite now while people look at his stupid insane thing ....
hi masak
06:49 justatheory left 06:53 xinming joined
sorear hi masak 06:53
kthakore so ... is that language possible to parse in perl6 grammars? 06:57
or do I need to define my language better?
06:59 xinming left, xinming joined 07:00 __david__ joined 07:02 nymacro left
kthakore hmm I guess you guys are busy I am gonna head to sleep 07:09
masak kthakore: not busy, just low activity right now. 07:10
kthakore: I haven't finished backlogging. when I do, I'll have a look.
sorear kthakore: perl 6 grammars contain a Turing complete language, so they can parse any recursively enumerable language 07:11
your question is not very meaningful
kthakore sorear: you are right I guess... I don't know how to ask a meaningful question here. I guess my real question is 'Given a card game, with rules defined by each card in play. How to generate a subroutine based on it's rule? ' 07:14
I thought grammars was the way to go ... but maybe not
sorear grammars are the way to go for inputting data from adhoc text files 07:15
if you have XML, you should use an XML parser. Not a grammar.
kthakore sorear: I don't think you understant ... 07:18
I am not trying to parse XML
sdlperl.ath.cx/mtg/EasyMTGCardsSyntax.xml
see <Card> .... <Rule> C[Self: | Damage:Player,Waiting](Player:Waiting | Discard:1): </Rule> .... 07:19
sorear ick
kthakore the text I am trying to parse is ' C[Self: | Damage:Player,Waiting](Player:Waiting | Discard:1): '
sorear you combine all the disadvantages of ad-hoc text with all the disadvantages of XML
kthakore sigh ...
this is not about XML sorear 07:20
ignore the xml I am sorry I picked it too
masak sjohnson: what's your criterion for a decent hex editor? would it be very hard to write one?
kthakore sorear: I want to parse something like ' C[Self: | Damage:Player,Waiting](Player:Waiting | Discard:1): '
into a function that I can call like $card->apply_effect() 07:21
sorear You'll confuse yourself less if you use an AST.
masak stands back and lets sorear++ handle this 07:22
sorear Make a few classes, have the grammar produce a tree of objects representing the card, then write a recursive function to apply the effect
kthakore ah ok 07:23
so each node on the tree will have something like $node->target, $node->effect, $node->modifier ? 07:24
and then I will work from the bottom of the tree up?
sorear I'd have to see your code to decide thayt 07:25
kthakore ok
I am in design phase as my first prototype is in C# and ANTLR
... and I am looking for a better way to tokenize, parse, and make an AST
sorear What does C[Self: | Damage:Player,Waiting](Player:Waiting | Discard:1): mean? 07:26
(it's been about 10 years since I last played M:tG)
kthakore it means Damage the waiting player with same health as self ( the caller)
and make tha same player discard a card 07:27
sorear: the rules are very complex .... *sigh*
sorear I know the rules, mostly
but I won't remember them without prompting
How does it say that?
kthakore I had this for a third year design project (2 years ago) and it still haunts me
sorear: in english?
sorear Where does it say "same health"?
kthakore Self: | Damage:Player,Waiting
sorear kthakore: English would be ideal, yes 07:28
kthakore yeah let me get that sorry
sorear What are the other options?
kthakore oops I forget C[ ] means a condition
www.coolstuffinc.com/main_viewCard....ng%20Cards
sorear What does Self: mean by itself?
kthakore the player I guess 07:29
my language making skills are really bad :(
how do I define a good language for this ....
sorear What does Damage:Player,Waiting mean?
kthakore whenever abyssal specter deals damage to a player, that player discards a card.
C[ ] is the condition ....
sorear What is 2bb? 07:30
I remember what 2w, 2r, etc are
kthakore 2 black mana
err
2 other cards
sorear What's the difference between 2b and 2bb?
kthakore blue and black
bb = black
yeah ... it is stupid ...
oh wait no no 07:31
that is 2 any colored card and 2 black cards
2 bb
sorear so bbbb or bbww or bbrg etc?
kthakore bbbb 4 black cards
bbwww 2 black 2 white
yeah
the number means any mana color 07:32
sorear I remember now
kthakore so my real problem I guess is to define a better language for this?
07:32 jummy joined, jummy left
sorear What does 'Waiting' mean in your example language? 07:32
kthakore Player who is waiting to play 07:33
other player
I guess
there is a playing player and waiting player
sorear: I know I have to redesign the language. However I was wondering if going into perl6 grammar was the correct direction for the implementation of such a game? 07:35
moritz_ perl 6 grammars are good for parsing that kind of stuff 07:37
the rest of the implementation should not be done with grammers :-)
kthakore moritz_: indeed ... that will be done with other stuff 07:38
ok I really need to go to sleep 07:39
thank you folks
sorear kthakore: Every time I've played m:tg, there have been 5-10 players, and I've been allowed to attack any of them. That's not standard form? 07:40
ok
masak someone seems disappointed: twitter.com/predominant/status/7947907076460544 07:41
but actually, Inf has some problems that we haven't yet resolved. the biggest one perhaps that it needs to act both as Int & Num & Str 07:42
kthakore sorear: yeah but I have to start some where simple
masak: bah don't listen to them
haha
masak kthakore: I don't let my happiness be contingent on things people say on Twitter, but I do pay attention to them. 07:43
kthakore: Twitter has this wonderful way of letting you hear things you might otherwise turn a blind eye to.
s/blind eye/deaf ear/ # mixing of metaphors 07:44
sorear I think I'd be doing something along the lines of class AbyssalSpecter does Card[power => 2, toughness => 3, type => 'specter', cost => '2bb'] { method onDamage($taget) { $target.discard } }
then use eval to load the card file
kthakore sorear: ... 0o 07:45
I couldn't do that in C# .... that is why I didn't consider eval!
kthakore curses 07:46
omg ... shit ... wow .... I mader this soooooooooooooo hard
kthakore facepalms
sorear: thank you ... so much ... and now I really need to go 07:47
masak some harsh criticism here: www.reddit.com/r/programming/commen..._released/
07:48 Trashlord joined
kthakore masak: oh yeah I saw that ... but you know what that means ... they are disappointed ... because they had expectations !!! 07:50
masak: so it is good hahah
masak also, the criticism that isn't raving lunacy is often spot on.
Rakudo has a high memory footprint and runs slowly.
kthakore indeed but parrot is slimming up
and I think rakudo and parrot feedback loop is gonna get it right soon ... ish 07:51
masak whatever. until it's fixed, the criticism is valid.
those are big problems, and they don't have easy fixes.
yes, people are working on them.
kthakore yeah
please keep doing it!
I can't wait to get SDL onto perl6
masak but I have quite a lot of sympathy for those who write Rakudo off due to it being slow or memory-hungry.
kthakore then have lots of games! 07:52
in perl6
sorear niecza: sub fib($n) { $n == 0 ?? 1 !! $n == 1 ?? 1 !! fib($n-1) + fib($n-2) }; say fib(25)
07:52 am0c joined
p6eval niecza 914f55d: OUTPUT«121393␤» 07:52
masak the main reason for Rakudo not being faster and slimmer already is that developing Perl 6 is non-trivial. 07:53
the past five years have been spent implementing and re-implementing tooling on Parrot just to be able to handle the requirements that Perl 6 poses.
it takes time and effort, and it brought us to this point. that's awesome, but we're not done yet.
most any outsider will look in and see that we're not done yet, but fail to see the massive amount of effort that went into getting us to this point. 07:54
kthakore masak: I am an outsider (if you will grant me that), and you know what I see? Smart People working Real Hard. 07:58
So I cut a lot of slack there
seriously if people can't understand that ... well w/e
but there are some who do (i.e. me ) 07:59
masak indeed. kthakore++
nevertheless, this is a case where deeds speak louder than words. if someone does something cool with Perl 6 (such as the URL dispatcher grammar last week), people will pay attention and get interested. 08:00
and then the speed and memory footprint doesn't matter that much anymore.
kthakore I mean I tried to help with string IO in parrot ... failed sooooo miserably
masak perhaps not the simplest aspect to help with.
kthakore masak: immutable string IO ?
sorear I don't like this notion of Smart People and Non-Smart People
kthakore sorear: it is probably not right, but it is how I see it 08:01
masak sorear: neither do I, but I tend to let it slip when it's a compliment. :)
kthakore sorear: can I put it like this? People who are more competent and capable and productive at compiler design then I am working really hard on a problem 08:02
hence proble not easy so give them slack.
problem*
sorear pmichaud and jnthn leave me utterly awed too, but I don't let that stop me from contributing
kthakore I never said I won't contribute 08:03
I have tried when I have time
I do it in parrot at times
I was just explaining how my last attempt to add a String IO bug feature was a total bust
sorear: still hasn't stopped me :D I am just biding my time till xmas break to get back into parrot 08:04
s/bug//g
sorear in other news, making indexing efficient is... a lot harder than I expected 08:06
08:07 cosimo joined
kthakore sorear++ masak++ thanks guys and bye for real now ... 08:07
sorear more than half of STD's time is being spent in return if @*MEMOS[$¢.pos]<ws>:exists; now
masak pmichaud and jnthn leave me utterly awed too, but I don't let that stop me from using my sledgehammer. :P
sorear (in other other news, niecza-compiled STD is now outperforming viv on the "parse niecza lib/SAFE.setting" benchmark) 08:09
moritz_ \o/
masak sorear: neat! 08:10
08:10 perimosocordiae left
masak I think the most useless entries by far in day-of-year articles on Wikipedia are those that chronicle the forming of US college fraternities. who, besides perhaps those that used to belong to it, cares? 08:12
moritz_ ... with greek TLAs, often without meaning 08:14
masak exactly.
Phi Sigma Sigma was founded today, 97 years ago. FSS! 08:15
08:18 palik joined
jnthn Writing talks the day I'm giving them. I do it for dotnet events too. 08:25
masak :)
jnthn busily slide hacks
masak eyes $¢ warily. again. 08:30
sorear ? 08:32
08:32 QinGW left
jnthn masak: Like to give us your 2 cents on that? 08:32
masak sure. 08:33
moritz_ expected such a *pun*chline
masak I completely missed that :P
anyway, I don't see the rationale for making it a non-ASCII identifier just to make it one character long. 08:34
we used to have a ¥ operator, but we decided against it. it became Z instead. 08:35
moritz_ I think the point is that it shouldn't be colliding with normal user-space variables
rakudo: say '¢' ~~ /\w/ 08:36
p6eval rakudo : OUTPUT«␤»
moritz_ rakudo: say so '¢' ~~ /\w/
p6eval rakudo : OUTPUT«Bool::False␤»
masak right, so it's in the same class as $_ and $/
moritz_ right
masak and those have to be one-char. 08:37
moritz_ maybe you could make it $¢/! but that wouldn't have any advantage over one char 08:41
masak I was going to argue for $MATCH or some such, but I think I buy your argument about it needing to be a one-char non-\w identifier. 08:42
jnthn I guess it'd be $CURSOR rtahre than $MATCH 08:43
moritz_ but that would be a bit much CURSing in the source
masak jnthn: oh, right. thanks.
08:46 Achilles14 joined
jnthn moritz_: yeah, no shit. 08:47
Achilles14 hi all, is there a compiler that can compile perl6 source to .pbc file? I can not find it in perl6 command line options.
moritz_ Achilles14: rakudo can do it 08:48
Achilles14: perlgeek.de/blog-en/perl-6/my-first....writeback
Achilles14 moritz_: thanks, it works. 08:51
moritz_ \o/ 08:52
at some point somebody should write a --compile option or so for rakudo
jnthn --target=pbc would be nais. 08:53
Achilles14 moritz_: I remember early parrot version have this tool.
08:53 wallberg joined
Achilles14 another question: is there anything like java's CLASSPATH, that I can deploy my program and dependencies seperately? 08:55
08:56 [particle] left
moritz_ there's a PERL6LIB environment variable 08:57
rakudo: say ~@*INC
p6eval rakudo : OUTPUT«lib /home/p6eval/.perl6/lib /home/p6eval/p1/lib/parrot/2.10.0-devel/languages/perl6/lib .␤»
08:58 [particle] joined 08:59 rjbs left
Achilles14 can lib be pbc? so any programming language can use it. I think PERL6LIB is not the answer. may be something like PARROT_LIB be. 09:00
09:02 shabble joined, shabble left, shabble joined
Achilles14 "parrot -M./common.pbc main.pbc", something like this. 09:03
moritz_ Perl 6 libraries can be precompiled, but you still need the source around 09:04
there are ways to load foreign libs, but currently they aren't as simple as we'd like them to be
wallberg hi all
moritz_ hi wallberg 09:05
masak wallberg! \o/
Achilles14 wallberg! \o/
09:07 rjbs joined
wallberg masak: i've made up my mind now. i'll have a go at making a phylogenetic program in Perl 6, which is the best way I can contribute to Perl 6. but I will need some help, especially with the grammars to read a newick tree :-) 09:08
masak wallberg: absolutely.
wallberg: pro tip: write tests for your grammar.
that way, you won't regress on silly things.
beyond that, I'm more than willing to help with the nitty gritty stuff. 09:09
09:09 wooden joined, wooden left, wooden joined
masak so, I guess, are many others around here. 09:09
moritz_ wallberg: pro tip: read perlgeek.de/en/article/debug-a-perl-6-grammar :-)
masak yeah, what moritz_++ said. 09:10
understanding the difference between regex/token/rule is also important. but that only comes with using them in practice for a while.
wallberg thanks guys. what editor is everyone writing their P6 code in these days? vim? Padre? 09:11
masak codes in vim 09:12
moritz_ uses vim too
masak my general rule of thumb seems to be 'code in vim, write prose in Emacs' :)
all other editors have left me wildly underwhelmed so far. 09:13
come back when you have something with macros and scriptability, like vim and Emacs.
moritz_ masak: I'm curious, why the distinction?
do you edit code more, while writing prose straight away? 09:14
masak moritz_: it's deeper than that. a lot of subtle things make the whole user experience of the two programs more suitable for that kind of writing... at least for me.
09:15 rgrau_ left
masak moritz_: Emacs seems to "think" in paragraphs. vim has a lot of navigational and manipulative commands that are adjusted for code (or other hierarchical) primitives. 09:15
wallberg well, I need something that tells me "nooooo!! you're doing it wroooong!!" as I type, so I'll start with Padre and see how it goes. or does vim have live syntax checking? 09:16
Achilles14 I suggest emacs. 09:17
masak I emphatically don't need an editor to tell me I'm doing things wrong.
Tene vim 09:18
moritz_ wallberg: it highlits in real time, that's all
wallberg masak: yeah, it would be nice with an editor that says "Good job!" every once in a while. 09:19
Tene masak: {} in vim navigate by paragraphs. also 'p' in object-select. I haven't done much with emacs; what does emacs do differently there? 09:20
Also () and s for sentences.
masak wallberg: reminds me of INTERCAL's insistence of prefixing an appropriate fraction of statements with "PLEASE " :)
Tene: nod. I know that. seems in describing the quality I'm after, I botched it altogether. 09:21
09:21 Achilles14 left
masak Tene: vim is not bad for editing prose text either, it's just that I find Emacs to be more suited for it. YMMV. 09:21
Tene masak: I was curious about your milage, not trying to assert anything about mine. 09:22
masak Tene: and yes, I use {} a lot already. I've actually turned off the arrow navigation keys in order to become more mindful of navigation.
Juerd vim is hard for prose because it doesn't support long lines
I do use it, but with line wrapping and manual re-wrapping
re-flowing even
Tene I haven't used emacs much.
Juerd Well, not really manual. I use gq.
Tene I have vertical arrows mapped to gk and gj 09:23
I rarely use them.
Juerd Ahhh. Thanks 09:24
Tene Oh, I guess if you didn't know those, long lines might be a bit more awkward for you. :) 09:25
masak Emacs is wonderful when it comes to re-flowing emails with those '>'s in them. 09:27
moritz_ #perl6 - learn something new about the editor you've been using for $n years
Tene Every six months or so, I go read through the vim docs and look for something new to try using.
szbalint sometimes I think there is too much to learn about vim 09:28
masak oh, and the vim docs are awesome.
Tene Marks were the biggest thing that stuck with me, since about 1.5 years ago.
szbalint most of the features don't fit in my head.
tadzik hello sixers
szbalint I am happy with those that do, though :)
masak tadzik: \o
Tene I also try to do that with bash, but less-frequently. alt+. is a nice hotkey that is severely underused. ^R even more so. 09:29
mathw o/
Tene It pains me seeing people re-type things that they could call up with like three keystrokes.
moritz_ alt+. is last argument from last line, right?
mathw I've got a friend who can never remember ^R
Tene moritz_: repeat it to recall last arguments from previous commands in the history.
moritz_ right
mathw but since I found out about it I've used it constantly
moritz_ I use Esc. which I guess is the same
Tene Yeah, same. 09:30
mathw Didn't know about alt+.
mathw tries it
AWESOME!
Tene++
Tene You can also use ctrl+alt+y for arbitrary arguments from the previous command
Esc,3,ctrl+alt+y = recall the fourth word from the previous command 09:31
wallberg hmm, I seem to have the ability to turn the whole channel off-topic. This isn't the first time :-)
Tene Or even esc,-,2 for second from the end
I nearly never use it, but sometimes it's useful if there's a long url or something in there. 09:32
moritz_ wallberg: that's fine, as long as no ongoing p6 discussion is disturbed by it
mathw And we're learning things
Tene It's nice to look through: bind -p | grep -Ev 'not bound|self-insert'
I've got a co-worker who is very set in his ways, and very resistant to changing anything, and it's painful to pair with him 09:33
instead of ^Rfoo<enter>, he insists on using !foo:p<enter><mouse copy paste> 09:34
Wow, I never knew about alt-y. I've wanted that many times. 09:36
"Rotate the kill-ring, and yank the new top. You can only do this if the prior command is yank or yank-pop."
www.gnu.org/software/bash/manual/ht...lling.html
moritz_ good news: all those commands are from readline, and every readline app gets the same by default
Tene Yeah, readline++ 09:37
Also, if you're not familiar with it, rlwrap is great.
I use it all over the place to use readline with programs that don't have it built in. 09:38
masak wow! Tene++ # rlwrap 09:44
moritz_ .u phi 09:46
phenny U+03D5 GREEK PHI SYMBOL (ϕ)
09:47 kjeldahl joined
masak getting heisenbug-like behaviour from the .trans thing. 09:51
I have this habit when doing .trans changes to rip the whole bit of code out into a script, do MONKEY_TYPING and augment the Str class with a new 'trans' method. 09:52
when I do it like that, it behaves as expected.
but when I run exactly the same test on the old .trans, it fails. 09:53
moritz_ that's... weird 09:54
mathw that's really quite disturbing 09:55
masak I can't really explain that right now.
mathw I hate that kind of bug
masak anyone want to try the script?
moritz_ did you remove .trans from Str before monkeytyping? 09:56
masak no.
actually, it's in Cool.
moritz_ ah, when it's in Cool you can monkey-type it into Str 09:57
masak here's the code: gist.github.com/716485
first, I monkey-patch Str. then, I run all the Rakudo-enabled spectests. 09:58
09:58 dakkar joined
masak note that I call the experimental variant .xtrans to keep things straight. 09:58
last two tests are the ones that differ. (and really shouldn't)
Tene masak: Perhaps I'm misunderstanding, but afaict, that script doesn't modify .trans in Str, only adds .xtrans 10:02
masak Tene: that's correct.
Tene So, I'm not surprised that they differ
masak Tene: but I just copied the code from Rakudo.
it's the same code. 10:03
Tene Ah.
masak actually, the obvious conclusion is that it's not the same code. that would fit better with the observation. 10:06
but why it wouldn't be, or how to test that it isn't, I don't know.
10:07 timbunce joined
masak there's only ever one definition of .trans in Rakudo, and that's in src/core/Cool-str.pm 10:07
mathw Heisenbug indeed
Tene Well, lemme update rakudo and check. 10:08
masak sure thing. 10:09
suspecting it might have something to do with the naming, I renamed 'xtrans' to 'trans' in my script. the 'trans' test succeeds then. 10:11
which means it's unrelated to naming. 10:12
which is disturbing, because then it's likely the core in itself that's causing this.
oh! could be something related to Cool/Str. 10:13
Tene That sounds to me like 'the version in core and the version augmented here differ'
masak right.
Tene You've recently had trouble with cached copies. Is the compiled version different from the source?
As you mentioned augmenting instead of recompiling.
masak rakudo: say "aaabbbccc".trans( / \s+ / => " " ) 10:14
p6eval rakudo : OUTPUT«aaa ␤»
masak I think that rules out any issues with my local Rakudo.
Tene ... is that right? That looks broken to me. Why does 'b' match \s+ ? 10:15
masak Tene: it's broken. that's the point.
rt.perl.org/rt3/Ticket/Display.html?id=79778
that's what I'm trying to diagnose and fix.
that's what's going all heisen on me. 10:16
Tene Oh, right, this is the part that takes FOREVER now. :(
So, the last two tests differ in behavior. What is the difference? 10:17
masak one goes to the augmented method, the other goes to the method in Cool in the setting.
Tene I'm sorry. What's the difference in behavior? 10:18
I'm still waiting on rakudo to build.
10:19 redicaps left
moritz_ masak: fwiw your example works fine for me on the REPL 10:19
masak Tene: ah. the setting one exposes the bug in the RT.
Tene: the augmented one doesn't.
moritz_: testing.
moritz_ and in a script too
tadzik it semi-works for me
> say "a aab bbccc".trans( / \s+ / => " " )
lambdabot <no location info>: parse error on input `+'
tadzik a aab bbccc
moritz_ This is Rakudo Perl 6, version 2010.11-4-g90cb844 built on parrot 2.10.1 RELEASE_2_10_1-421-g73ceeb1 10:20
Tene I thought we got rid of lambdabot?
moritz_ .u pi
phenny U+A03A YI SYLLABLE PI (ꀺ)
Tene π 10:21
masak moritz_: hm. works on the REPL here too.
Tene .u π
phenny U+03C0 GREEK SMALL LETTER PI (π)
masak I'm starting to suspect I simply have an old installed version of Rakudo. 10:22
moritz_++
Tene Rakudo compile supposed to take so much longer than it used to? :(
masak: As I said. :) 10:23
moritz_ Tene: on a new parrot it's supposed to be a bit faster again
Tene moritz_: Ah, I'm just using whatever --gen-parrot fetches for me
moritz_ Tene: if you rakudo is new enough, you should get a "fast" parrot
masak how do I configure Rakudo to install into /usr/local/bin/perl6 ? 10:24
moritz_ perl Configure.pl --gen-parrot --gen-parrot-prefix=/usr/local
masak thanks.
10:25 leprevost joined
Tene masak: Both tests pass for me. 10:25
masak good. then I seem to have nothing to worry about.
Tene moritz_: I just pulled from git right before building, so latest rakudo master
masak Tene++ # thanks for the help
I'm glad it wasn't a real heisenbug.
moritz_ it seems rakudo rebuild on p6eval is busted 10:26
masak that would explain that part of the mystery.
Tene www.guardian.co.uk/science/2010/jun...ack-hands# -- fascinating
masak now everything works over here. 10:40
Tene :D
masak so there really was no problem with .trans
I'm glad that's the case, because it felt like I fixed it last time :P 10:41
moritz_ $ perl evalbot/build.pl rakudo 10:42
Can't handle revision 'RELEASE_2_10_1-477-gfa56f62
' at /home/p6eval/evalbot/build-scripts//rebuild-rakudo.pl line 43. 10:43
masak I've recently switched to building with --gen-parrot, so I guess that's what confused me.
Tene Ah.
masak now November works (mostly) fine! \o/
Tene \o/ 10:44
mathw wooo!
tadzik wow!
that means Web.pm too?
masak no, but I will turn my attention to Web.pm in the weekend. 10:45
got another reminder email from jhorwitz++ about Web.pm. I'll reply to that one today, and try to get the dreaded Week 20 done in the weekend. 10:46
masak replies now
tadzik does Web.pm have HTTP headers parser? 10:48
masak no.
the Rack paradigm assumes that HTTP headers come in through the usual CGI manner, in environment variables. 10:49
10:49 satyavvd left
tadzik I feel like writing that 10:55
masak please go ahead.
mathw HTTP parsing is fun
Tene tadzik: it's the job of whatever is running the rack-style app to handle parsing the HTTP headers. 10:56
In Web.pm case, usually HTTP::Daemon 10:57
tadzik It also seems simple 11:00
how funny it'd be to have a Plack-alike for Perl 6? 11:03
masak lunch & 11:07
Tene tadzik: Funny? I think it's an admirable goal. 11:13
sorear out
moritz_ didn't mberends++ start a clone? 11:15
tadzik server-simple-psgi, right 11:20
Tene: it doesn't differ much for me :) 11:21
github.com/mberends/http-server-si...-small.pl6 there it is
11:25 nymacro joined 11:36 fcojocaru joined
fcojocaru any way a newbie to perl can help as a dumb tester? 11:37
colomon absolutely! 11:45
masak oh, that would *rock*.
fcojocaru: even going through RT and just running stuff would help us.
fcojocaru any wiki/trac page to get me started? 11:46
what do you use to keep track of the tickets ?
tadzik fcojocaru: newbie to Perl, or newbie to Perl 6?
masak rt.perl.org/rt3/
fcojocaru to perl unfortunately
colomon fcojocaru: masak means the bug tracking software... checking filed bugs to see what they do now.
fcojocaru so I can only help on low level stuff
sorry
colomon but also, you could provide valuable feedback on the book they are working on. 11:47
masak we need all kinds of newbies. :)
fcojocaru one at a time guys
tadzik fcojocaru: go throught the Book (github.com/perl6/book) to learn your way around, it's quite probably you'll find some bugs/broken examples
that'll also help you understand stuff
fcojocaru good point
tadzik Eventually you'll become a warrior in our MMORPG :) 11:48
fcojocaru I send back issues to someone specific or use the rt.perl.org/rt3/?
I'll have to decline the offer with the warrior
I'm more of a bear thing you see 11:49
so maybe a creep
?
tadzik I don't think so
loook at use.perl.org/~masak/journal/39445
masak please don't link to use.perl anymore :P
tadzik fcojocaru: OOC< how did you got interested in Perl6?
masak strangelyconsistent.org/blog/perl-6-is-my-mmorpg
tadzik updates his bookmarks
masak tadzik++ 11:50
fcojocaru quite easy , I'm trying to learn perl5 now and wanted to see the future
since I'm out of ideas on what to write to get me around, testing a bunch of stuff for perl6 gives you lots of insight 11:51
masak the future is here already. it's just unevenly distributed. :)
fcojocaru so it seems a good thing to do , whenever I have the time
11:51 daxim joined
fcojocaru Thank y ou all for the welcome , I'll start going trough the book and hopefully be of some use to you 11:53
Good Luck
masak likewise.
stop by often :)
fcojocaru will do 11:54
12:09 gimix left 12:10 whiteknight joined
tadzik phenny: "さようなら"? 12:30
phenny tadzik: "Bye" (ja to en, translate.google.com)
masak "sayōnara". guess it's the macron over the ō that makes it five characters, even though it's four syllables. 12:32
sbp phenny: "さ よ う な ら"? 12:34
phenny sbp: "That they are" (ja to en, translate.google.com)
12:35 wamba joined
Tene .u さようなら 12:46
phenny Tene: U+3055 U+3088 U+3046 U+306A U+3089
12:52 oha joined
jnthn rakudo: (1, 1, * + * ... Inf)[0..10].join(', ').say 13:10
p6eval rakudo : OUTPUT«1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89␤»
13:15 bluescreen joined, bluescreen is now known as Guest13659
masak rakudo: say (1, 1, * + * ... Inf)[0..10].fmt(:separator(", ")) 13:17
p6eval rakudo : OUTPUT«1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89␤»
masak \o/
13:18 MayDaniel joined 13:24 Patterner left 13:26 sftp joined 13:31 MayDaniel left 13:32 mavrc joined 13:35 takadonet joined, Psyche^ joined, Psyche^ is now known as Patterner
masak *groan* you may take the au++ out of #perl6, but you can never take the #perl6 out of au: twitter.com/audreyt/status/8152126110826497 13:43
punnery worthy of Worthington. :)
13:46 mavrc left 13:49 florz left, buu left, MindosCheng left, _ilbot2 left, estrabd left, robinsmidsrod left 13:50 kjeldahl left, shabble left, cjk101010 left, Chillance left, cotto left
jnthn lol my talk went well 13:51
(yes, I did manage to find a way to get the p6 evalbot into a .Net event talk... :-))
masak nice
13:51 mavrc joined 13:52 kjeldahl joined, shabble joined, cjk101010 joined, Chillance joined, cotto joined, florz joined, buu joined, MindosCheng joined, HarryS joined, simcop2387 joined, f00li5h joined, ashleydev joined, mkramer1 joined, knewt2 joined, pothos joined, lamstyle joined, sunnavy joined, Sarten-X joined, huf joined, sECuRE joined, cookys joined, Bucciarati joined, _ilbot2 joined, estrabd joined, robinsmidsrod joined, Maddingue joined, nperez joined, frodwith joined, leguin.freenode.net sets mode: +v _ilbot2 13:53 HarryS left, kjeldahl left, HarryS joined, kjeldahl joined 13:57 mavrc left 14:00 estrabd left 14:01 estrabd joined 14:05 sftp left 14:06 sftp joined 14:12 mavrc joined
takadonet i cannot wait for smash to do the benchmarking for the newest rakudo star version 14:12
14:16 sftp left, sftp joined, mavrc left
colomon takadonet: why? 14:20
takadonet colomon: should see a huge speed up on startup times 14:21
colomon ah
need to get smash to check in the scripts for running the benchmarks (if he hasn't already) so we can all duplicate them locally.
takadonet ya 14:22
14:23 mavrc joined
colomon I don't see any trace of the harness in bench-scripts. :( 14:23
14:34 cookys left, cookys joined, zby joined 14:35 timbunce left 14:36 `26 left, Eevee joined 14:37 buu left 14:38 buu joined, rurban joined 14:41 timbunce joined 14:52 _kaare left 14:54 Sarten-X left 15:01 Sarten-X joined 15:04 dual left
oha am i missing something, or the parenthesis are mandatory for a method call, but not on a sub call? 15:08
jnthn rakudo: 42.say 15:09
p6eval rakudo : OUTPUT«42␤»
jnthn oha: But if you want args, then yes
masak well, 15:10
jnthn oha: There's also this colon syntax. :)
oha with args, yes
masak right.
jnthn But, well...
oha jnthn: could you explain the colon syntax a bit?
jnthn I think the colon is better left in smileys.
colomon rakudo: $*OUT.say: 42
p6eval rakudo : OUTPUT«42␤»
oha ty
masak rakudo: say 42.fmt: "%05.2f"
p6eval rakudo : OUTPUT«42.00␤»
colomon but yeah, I'm not a big fan of the colon form in general 15:11
jnthn (Some people like the colon syntax a lot. It's not my preference...but I've nothing really against it either.)
colomon still, TMTOWTDI
masak kill it with fire. :)
jnthn masak: -> #parrot
;)
masak jnthn: what's this ':notdef:' thing? twitter.com/uasi/status/8174326008258560 15:13
looks made up.
jnthn .:notdef maybe once woulda worked? 15:15
No, that still loosk wrong
Looks made up to be :)
There was a notdef method once.
MindosCheng pugs: <sleep read>.pick()
p6eval pugs: ( no output )
MindosCheng pugs: say <sleep read>.pick()
p6eval pugs: OUTPUT«read␤»
MindosCheng Okey...
jnthn Perl 6. Solving your important life decisions. 15:16
masak replies to uasi
perl6: say <work rest>.pick
p6eval pugs: OUTPUT«rest␤»
..rakudo : OUTPUT«work␤»
masak oh noes :)
jnthn I like Pugs answer better here. :)
masak the two implementations kinda reveal their current active status, too. 15:17
jnthn :)
MindosCheng masak++
jnthn ooh, almost time for my next talk
mathw ooh 15:18
what's it on
MindosCheng jnthn++
masak oh, there's a blog post too, containing a similar error: d.hatena.ne.jp/uasi/20101126/1290783645 15:19
oh wait! 15:20
:!defined: *does* work.
but it... looks so wrong... :)
the second colon is an invocant colon.
the first is a pair constructor, along the same principles as :e 15:21
masak holds head in hands
how does one designate "invocant undefined" using the :U syntax? 15:22
colomon method blah(Frob:U $self:) 15:23
(he guessed, no idea if this works)
masak right. thanks.
colomon rakudo: sub blah(Str:U $not-a-string) { say $not-a-string.defined }; blah(Str) 15:25
p6eval rakudo : OUTPUT«===SORRY!===␤Invalid typename in parameter declaration at line 22, near " $not-a-st"␤»
colomon rakudo: sub blah(Str $not-a-string) { say $not-a-string.defined }; blah(Str)
p6eval rakudo : OUTPUT«Bool::False␤»
masak that's what I got locally too.
15:28 flussence joined 15:31 MayDaniel joined
masak rakudo: say 42.notdef 15:35
p6eval rakudo : OUTPUT«Bool::False␤»
masak I'm not sure I can submit a rakudobug for this one. :)
for all I know, it's "extra functionality" that Rakudo provides, on top of the spec. 15:36
15:37 whiteknight left 15:38 nymacro left 15:41 yuki_ joined 15:54 clkao joined 15:57 yuki_ left 15:58 ch3ck3r joined 16:02 dd-sphere left
moritz_ std: sub bla(Int:U $x) { $x } 16:10
p6eval std a194beb: OUTPUT«ok 00:01 121m␤»
16:12 __david__ left
tadzik hello again 16:12
masak std: class C { method bla($?CLASS:U $self:) {} } 16:19
p6eval std a194beb: OUTPUT«===SORRY!===␤You may not use the ? twigil in a signature at /tmp/dF09kjh67D line 1:␤------> class C { method bla($⏏?CLASS:U $self:) {} }␤ expecting twigil␤Parse failed␤FAILED 00:01 120m␤»
masak didn't think so.
16:20 jferrero joined
masak wow, JS to "borrow" the Perl 6 way of writing if/while/etc expressions without parens? twitter.com/Smylers2/status/8191234032013312 16:23
takadonet the original post does not even talk about Perl 6 ! 16:25
brendaneich.com/2010/11/paren-free/
16:26 risou joined 16:30 sftp left 16:31 timbunce left 16:38 timbunce joined 16:40 dual joined 16:44 sftp joined 16:47 ch3ck3r left
kthakore masak: morning. 16:49
masak kthakore: hi there.
kthakore I am trying to remember our convo last night (6 hrs ago) 16:50
Where did you say most of perl6's memory gets into?
16:50 mkramer joined, fcojocaru left
kthakore in parrot or perl6 land or a combination of both? 16:51
flussence irclog.perlgeek.de/perl6/2010-11-26#i_3031080 ?
16:52 mkramer1 left, mila_ joined
kthakore flussence: yeah but that makes no sense to me 16:53
tooling?
what is that?
masak kthakore: I was thinking about the tools required to make Rakudo.
kthakore ah ... ok
masak nqp-rx being the most prominent example.
tadzik „Learning Perl 6 has been fun. I already want to stop writing Perl 5.”. Hah, I'm not the only one
kthakore masak: well where does perl6's memory useage mostly lie? Did anyone do a static analysis? 16:54
tadzik: yay from reddit :D
masak nqp-rx is the embodiment of lessons learned from PCT, which in turn profited from lessons learned from PGE and TGE.
kthakore: yes, I'm sure chromatic has done memory footprint analyses.
tadzik kthakore: which thing? :)
masak kthakore: I think "a combination of both" is most fair. 16:55
kthakore ok
masak kthakore: we know of memory inefficiencies in Parrot, and in Rakudo.
tadzik kthakore: yay for Game programming tutorial :)
kthakore is a problem of design or just how it is implemented at the time?
tadzik: :D have you seen SDL_Manual?
16:56 mila__ left
kthakore tadzik: I tried last night to get SDL onto parrot and Perl6 ... but the memory and CPU usage of perl6 is not ready for it yet 16:56
16:58 mkramer1 joined
masak kthakore: maybe you could try to conspire with either jnthn/nqpnet or sorear/niecza, either of which probably has a smaller memory footprint. 16:58
nomz & 16:59
16:59 masak left 17:00 mkramer left
tadzik kthakore: yes, I love it! I wouldn't mind it PDF'd though 17:12
moritz_ wonders if oha's p6l mail should have gone to p6u 17:13
17:13 risou_ joined
tadzik wonders whether he should subscribe p6l 17:14
oha moritz_: if i've written in the wrong place, i'm sorry
tadzik: and sorry if my mail is annoying :( 17:15
moritz_ oha: no problem, I was just surprised because usually p6l posts contain some new suggestions for the language 17:16
17:16 risou left
moritz_ oha: I haven't quite understood what you want do with the signature 17:16
tadzik oha: I don't know, haven't read that yet
moritz_ did you try to get the names of the parameters? 17:17
oha moritz_: it is what i did
moritz_ ah
"TODO: i would like to find a way to use the Signature for table names"
that's the part I didn't understand 17:18
oha but if you consider a "select tab1.field1, tab2.field2"
moritz_ how would a calling example look like?
ah
oha then the signature couldn't be $x.y
moritz_ fully qualified table names
oha yep
(or aliases, whatever)
moritz_ finally understands
moritz_ has no idea how to do that 17:19
17:19 buu left, buubot left
oha moritz_: i was thinking on nested signatures 17:19
but i still can't find a way to name a nested sig
moritz_ of course you could define some substitution rules 17:20
$me-foo becomes me.foo
tadzik oha: what is your realname, so I can find your post?
moritz_ also ' is allowed in identifiers
oha moritz_: yup. anyway i have to confess that i've started thinking how to do something like chagning the perl6 grammar
moritz_ tadzik: sender is Oha [email@hidden.address]
oha tarzik: should be Oha 17:21
moritz_: but then i've realized it could be done without, and i enjoyed it. feeling happy i tout it was good to share this :)
moritz_: nice, the ' is the best candidate for
tadzik doesn't look bad to me 17:22
17:22 M_o_C joined
oha well, time to go home for me. ty for your suggestion moritz_. bye all 17:23
moritz_ cia oha 17:24
oha: fwiw you might like to use placeholders instead of pointies
17:24 oha left
moritz_ oatclip monads: chrisdone.com/posts/2010-11-25-lisk...skell.html 17:32
or rather, fingernails in oatmeal monads 17:35
whatever
moritz_ is out of brainz
17:41 oha joined 17:57 lamstyle left 18:01 daxim left 18:04 timbunce left, lamstyle joined 18:10 dakkar left 18:15 risou__ joined, jferrero left, mkramer joined, mkramer2 joined 18:16 risou_ left, mkramer left 18:17 mkramer2 left 18:18 mkramer1 left, mkramer joined 18:23 takadonet left
tadzik std: my @a; @a[-1] 18:26
p6eval std a194beb: OUTPUT«===SORRY!===␤Unsupported use of [-1] subscript to access from end of array; in Perl 6 please use [*-1] at /tmp/OKV6umIuEc line 1:␤------> my @a; @a[-1]⏏<EOL>␤Parse failed␤FAILED 00:01 120m␤»
18:27 cjk101010 left
tadzik rakudo: my @a = 1, 2, 3; @a[-1]; 18:27
p6eval rakudo : ( no output )
tadzik rakudo: my @a = 1, 2, 3; @a[-1]; say $!
p6eval rakudo : OUTPUT«Any()␤»
tadzik rakudo: my @a = 1, 2, 3; try { @a[-1]; } say $! 18:28
p6eval rakudo : OUTPUT«===SORRY!===␤Confused at line 22, near "try { @a[-"␤»
mkramer rakudo: my @a = 1, 2, 3; say @a[-1];
tadzik oh come on
p6eval rakudo : ( no output )
tadzik ===SORRY!===
Cannot use negative index -1 on Array
so rt.perl.org/rt3/Public/Bug/Display.html?id=63986 can be closed, no?
jnthn *-1
mkramer my @a = 1, 2, 3; say @a[*-1];
blah
tadzik yeah, I meant the bug. It should fail, that's ok
mkramer just make this damn bot print its last expression =P 18:29
tadzik I can fix the error message so it will behave like STD and we can mark that bug fixed
jnthn (from earlier) "how does one designate "invocant undefined" using the :U syntax"?
method (Foo:U $self:) { ... } # works 18:30
But also should
method (::?CLASS:U $self:) { ... }
tadzik I think this one also works now: rt.perl.org/rt3/Public/Bug/Display.html?id=66252 18:32
this too: rt.perl.org/rt3/Public/Bug/Display.html?id=67374 18:33
oha am i wrong or @a[-1] should not be the same error of $x=-1; @a[$x]; ? 18:36
18:36 am0c left 18:37 palik left
tadzik hmm, it can't be in-grammar then 18:37
tadzik feels like fixing
oha tadzik: right now, rakudo exit with the same error to me 18:38
tadzik fail "Cannot use negative index $pos on {self.WHO}" if $pos < 0;
in Any-List.pm. Looks good to me, I think the bug can be closed
18:39 leprevost left 18:40 envi left
tadzik rt.perl.org/rt3/Public/Bug/Display.html?id=69218 -- closable to, methinks 18:40
18:41 MayDaniel left 18:45 PacoLinux joined, ggoebel joined
tadzik rt.perl.org/rt3/Public/Bug/Display.html?id=72968 and this 18:45
18:45 ggoebel left
tadzik ...is there a RT guy around? 18:45
s/a/an/
or this: rt.perl.org/rt3/Public/Bug/Display.html?id=73148 18:47
18:49 wamba left
tadzik std: say 18:52
p6eval std a194beb: OUTPUT«Potential difficulties:␤ Unsupported use of bare 'say'; in Perl 6 please use .say if you meant $_, or use an explicit invocant or argument at /tmp/xyp8lC4b3r line 1:␤------> say⏏<EOL>␤ok 00:01 117m␤» 18:53
flussence ooh, p5.13.7 just made a huge amount of my $dayjob code redundant.
moritz_ flussence: with which feature?
flussence auto-dereferencing array/hashrefs in builtins 18:54
moritz_ ah
tadzik rt.perl.org/rt3/Public/Bug/Display.html?id=74476 -- closable 18:56
moritz_ do we have tests enabled for that?
ah yes, colomon++ unfudged some 18:59
19:02 tylercurtis joined
tadzik rt.perl.org/rt3/Public/Bug/Display.html?id=76238 -- closable 19:03
19:04 patrickas joined
patrickas moritz_ ping 19:05
19:05 bpa joined
moritz_ patrickas: pong 19:06
patrickas hello boss !
did you work on getting rakudo to build after the parrot git migration?
tadzik hey, it works 19:07
moritz_: do you have RT access?
moritz_ patrickas: yes 19:09
tadzik: yes
tadzik: don't you?
patrickas moritz_: there are a few things not working for me (on win32) and I am unable to build rakudo anymore which might be unrelated 19:12
moritz_ patrickas: I don't have a win32 dev machine, so any fixes are very welcome
tadzik moritz_: nah 19:13
I wanted to do some bugfixing today, but everything I check works alredy
19:13 lamstyle left
patrickas The thing is I am building from scratch and I want to make sure that it is not my setup that is messed up ... did anyone else coplain on win32 ? 19:13
moritz_ tadzik: step 1: create an RT account
patrickas or rather ... did anyone build successfully on win32 ?
moritz_ tadzik: step 2: tell pmichaud about your account name, and request edit privs for the perl6 queue
patrickas: jnthn did some testing, fixed a failure, did it RONG, I fixed it differently... I don't know what the final outcome was 19:14
patrickas ok then I'll take it up with him then..t hanks! 19:15
tadzik phenny: tell pmichaud may I get the access to perl 6 RT? I found like 10 bugs today which are alredy working and can be closed 19:16
phenny tadzik: I'll pass that on when pmichaud is around.
19:16 M_o_C left
moritz_ tadzik: go to rt.perl.org/rt3/ and create an account (follow the "Get a Bitcard account" link), and then also phennytell pmichaud your nick 19:18
19:19 lamstyle joined
tadzik moritz_: done that 19:24
bpa I've been trying rakudo on and off, really liking what I see. I'm wondering if I've missed a dropped feature, or if I just imagined it. Did this work before? for 1 .. 3 b $x ... 19:27
Juerd Er...
bpa that is supposed to be the -> unicode character
Juerd Unless you're suggesting blinking text as syntax feature, you may want to send that line again.
bpa I don't think my irc client supports it 19:28
sorry about that
Juerd :)
flussence “→”?
bpa In code it looks like a nice right arrow
Juerd I think it'd be madness to support arrows like that, since there are quite a few in Unicode. 19:29
bpa so I was just imagining it then?
Juerd I guess so, especially as the language operators fit in latin1
gfldex bpa: or you might have visions. with perl6 you cant relly tell the difference :)
bpa I may have used it with pugs, can't really remember 19:30
The hyper operator works though, and that isn't ascii
gfldex perl6 requires unicode
Juerd Still unlikely, bpa
gfldex if you want to you could define operaters like that
Juerd You can 19:31
flussence »« aren't ascii, but they are in iso8859-1
bpa I think I'll stick with the core operators and not confuse people who may come across my code 19:33
19:33 justatheory joined
bpa I do think its amazing that I could extend the grammar do to it if I really wanted though 19:34
(without editing the source that is)
dalek kudo: 82f3de0 | patrickas++ | build/gen_parrot.pl:
extra check needed for gen_parrot on win32
19:35
oha rakudo: "a'b" ~~ /((\w+)\')?(\w+)/; say $/[0][0], "|", $/[0][0][0]; 19:39
p6eval rakudo : OUTPUT«a'|a␤»
oha i expected to get 'a' with the $/[0][0]; why must I use $/[0][0][0]? 19:40
is the ? which make an array here?
moritz_ yes
every quantifier turns that capture into an array 19:41
oha i had to realize it alone
sjohnson hi pals 19:44
19:44 lamstyle left
moritz_ oha: if you use too many nested captures, you should think about using (named) subrules 19:47
TimToady or [] where you don't care to capture 19:51
19:56 _twitch left, patrickas left 19:58 mkramer1 joined, silug joined
oha moritz_: i've used $<foo>=(...) 19:59
19:59 mkramer left 20:00 kjeldahl left
oha can i do something similar to: -> %x<foo> { ... } 20:00
TimToady such binding has been conjectured, though you can't have a subscript as a formal parameter name 20:02
20:09 lamstyle joined 20:10 Raynes joined 20:16 `26 joined 20:18 gbacon joined 20:25 masak joined
masak ahoy! 20:25
jnthn ahoj, masak :) 20:26
tadzik ahoj :)
jnthn rakudo: sub foo($ where (my %x<h> = $_)) { say %x<h> }; foo(42); # curious... 20:27
p6eval rakudo : OUTPUT«===SORRY!===␤Non-declarative sigil is missing its name at line 22, near "$ where (m"␤»
jnthn rakudo: sub foo($x where (my %x<h> = $_)) { say %x<h> }; foo(42); # curious...
p6eval rakudo : OUTPUT«===SORRY!===␤Unable to parse postcircumfix:sym<( )>, couldn't find final ')' at line 22␤»
jnthn oh...
duh :)
rakudo: sub foo($ where (my %x, %x<h> = $_)) { say %x<h> }; foo(42); 20:28
p6eval rakudo : OUTPUT«Constraint type check failed for parameter ''␤ in 'foo' at line 22:/tmp/asBHelsmTw␤ in main program body at line 22:/tmp/asBHelsmTw␤»
jnthn Hmm. :)
silug warren: i'm surprised there aren't any commercial entities taking enough of an interest in spamassassin to put some manpower behind it.
jnthn rakudo: sub foo($x where (my %x, %x<h> = $x)) { say %x<h> }; foo(42);
p6eval rakudo : OUTPUT«Constraint type check failed for parameter '$x'␤ in 'foo' at line 22:/tmp/YkiX0jvns_␤ in main program body at line 22:/tmp/YkiX0jvns_␤»
silug urk
jnthn gets curious what that ends up failing...
It's still pretty nasty looking even if it worked though. 20:29
masak I'm just going to blog a little, and then I'll go to bed. 20:30
don't want to ruin my good sleep streak :)
tadzik++ # digging around in RT 20:32
tadzik masak: I wanted to fix some bugs, but all of them were alredy fixed :(
masak tadzik: how terrible! :) 20:33
jnthn: why was this again? twitter.com/uasi/status/8254037887549440
jnthn: uasi thinks it's "once for (dispatch) search, once for calling". is that the truth? 20:34
also, uasi++ made a brainfuck interpreter in Perl 6: gist.github.com/717036 20:35
tadzik oh, that could be a Parrot language 20:36
dukeleto there is already a br interpreter on Parrot
s/br/bf/
jnthn I think there's (white)space for a br interpreter ;-) 20:37
masak: Yes
20:37 timbunce joined
jnthn masak: It's something that I think can get optimized away but... 20:37
masak ...but not easily.
jnthn Grr. I forget the subtlety
We *do* avoid re-doing the nominal type checks. 20:38
masak ah.
TimToady masak: the Japanese consider the ō to be two syllables (or mora, to be preciser)
jnthn I suspect there's a comment in the binder somewhere hinting at where to stick a future optimization in this area.
masak TimToady: ah, thanks. that makes sense.
uasi++ is very inquisitive. I like replying to his tweets. 20:39
TimToady in fact, they would count it that way in haiku as well
masak TimToady: I was wondering about that.
TimToady english can stuff a lot more into a haiku than japanese can...
tadzik Hmm, brainfuck almost works
20:53 bpa left 21:01 ch3ck joined
lue ohayo wurld o/ 21:11
masak hellue!
21:11 kfo joined
sjohnson hi 21:12
masak sjohnson: \o 21:14
sjohnson :)
lue it's strange. While writing a few tests last night, I started writing tests for things that will fail, and I had the urge to make them pass. But I had to go to bed [curse you rotation of the Earth due to forces I cannot quite name at this moment!!]
masak doesn't sound strange at all. 21:15
you have to learn to pace yourself.
for example, start writing tests earlier in the day :)
then you'll have time to implement them. 21:16
lue it's as if... TDD works, somehow... :)
masak so it's rumoured. 21:17
oha does "a $x b" get compiled as "a "~$x~" b"? 21:18
ch3ck hej masak \o
oha or the expansion does not pass thru ~?
masak oha: more like "a " ~ ~$x ~ " b"
ch3ck: godkväll. 21:19
21:19 timbunce left
moritz_ knows enough bokmål to understand that :-) 21:19
though I'd say "god kveld" :-) 21:20
21:20 rurban left
ch3ck this was swedish, not norwegian 21:20
masak right. the language on which norwegian was based. :P
masak wonders whether frettled or arnsholt will slap him first :) 21:21
moritz_ uhm, I thought it was based on Danish
masak a truly scary thought :) 21:22
hm. seems my hobby today is as a language chauvinist.
frettled hugme: hug masak
hugme hugs masak
masak awww :) 21:23
I'll quote Teratogen. "I'm a failed troll" :)
blog post! strangelyconsistent.org/blog/novemb...-saucepans 21:24
dalek ast: 52c1e70 | moritz++ | S02-builtin_data_types/array.t:
[array.t] unfudge some tests that rakudo now passes
masak tadzik: did all those tickets that you yelled "closable" about ever get closed?
tadzik: would you like an admin bit for RT?
moritz_ masak: I closed the first that tadzik++ yelled about, and claimed the second (because I know of no tests so far) 21:25
then I got distracted
masak oh, good-ish.
moritz_+ # :)
lue today's xkcd is hilarious. 21:27
moritz_ masak: :-)
frettled Zack Weiner's guest comic is nice, yes. 21:28
And SMBC is often entertaining.
masak++ - positively good progress (blog) 21:29
masak er.
masak is happy to have a clown nose: :©)
frettled: yes. and the weekend will be even positiver. 21:30
oha what am i doing wrong here? multi sub infix:<~>(B $l, Any $r --> B) { ... 21:31
lue
.oO(I knew the colon was copyrighted! I knew it!)
moritz_ oha: maybe rakudo doesn't knwo about --> yet?
masak it does.
21:31 tylercurtis left
masak rakudo: class B {}; multi sub infix:<~>(B $l, Any $r --> B) { ... }; say "alive" 21:32
p6eval rakudo : OUTPUT«alive␤»
masak oha: from what I can see, you're doing nothing wrong.
oha: please be more specific, or realize that you're doing it right. :)
oha masak: but when i my $x = B.new; $x ~ 'foo'; the sub isn't called
masak oha: I see.
oha: can you demonstrate that with p6eval?
oha i'm using rakudo here 21:33
masak so is p6eval.
oha oh ok
masak (see above)
oha i've understood what i was missdoing 21:34
the sub was inside class B {
and probably interpreted as a method of B
masak no.
but it makes it lexical inside B.
oha ohoh! got it 21:35
masak oha: subs and methods are separate things, at least as far as declarations are concerned.
oha therefore i must put it outside the class, and if i'm going to include it as a pm i must declare it 'is export' too?
masak well, there's still a difference between the file and the scope of the module. 21:37
but yes, most of the time.
masak sleeps 21:39
21:40 masak left, Guest13659 left
Teratogen I'm a failed troll!? 21:43
21:44 tylercurtis joined, tylercurtis left 21:47 TypeNameHere_ joined 21:54 jferrero joined 22:00 mkramer1 left
dukeleto Teratogen: that is the worst kind of troll. 22:24
Teratogen oh dear =(
lue rakudo: say ":a<b>" ~~ Hash # most likely fails, just curious 22:25
p6eval rakudo : OUTPUT«0␤»
jnthn /win 11 22:31
...fail!
sorear good * #perl6
22:31 `26 left
Tene /fail 11 22:32
jnthn :P
diakopter sorear: good
lue afk 22:38
22:53 jferrero left 22:54 `26 joined, `26 left, `26 joined
tadzik bah, masak's gone 23:00
sorear @part #perl6
aloha: seen kthakore
aloha sorear: kthakore was last seen in #perl6 6 hours 3 mins ago saying "tadzik: I tried last night to get SDL onto parrot and Perl6 ... but the memory and CPU usage of perl6 is not ready for it yet".
kthakore sorear: hi 23:01
23:06 starcoder joined
eternaleye phenny: tell masak Hey! I read HP & the methods of rationality too! I prticularly like how Draco was absolutely floored by muggles being the first humans on the moon. 23:07
phenny eternaleye: I'll pass that on when masak is around.
23:12 wallberg left
kthakore sorear: ?? ok ... 23:13
I have to head out to a class then 23:14
23:21 gbacon left 23:42 dual left 23:44 dual joined 23:56 jferrero joined