»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo:, alpha:, pugs:, std:, or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by lichtkind on 5 March 2010.
00:04 azert0x left, Sarten-X joined
lue afk 00:05
00:13 Sarten-X left 00:16 ggoebel left 00:23 lest_away is now known as lestrrat 00:24 Sarten-X joined 00:28 cygx left 00:33 Chillance joined, Sarten-X left 00:38 Fozo joined
Fozo I was wondering how would I sign up to contribute to the devolpmet oof Perl 6? 00:38
devolpment*
sjn Fozo: have you looked at rakudo.org/how-to-help ? 00:39
Fozo No I have not semm that yet I beleive here let me check it outr
out* 00:40
00:40 yinyin joined, masonkramer left, masonkramer joined
Fozo Hmm I can not acess the mailing lists 00:42
wait can anyone explain how all of this would work? Do I just submit bug reports and nothing else or do I say give away my email and then perhaps send code contributing to the complier? Or do people request fetures and then people volunenteer to make these desires a reality, I really want to help make perl 6 a reality and hopefully I am not too late to join or any thing. Answers will be appreacaited 00:45
00:46 kensanata left
Fozo everyone here is soo quite, 00:47
colomon well, submitting bug reports is useful and easy.
It's also easy to get a commit bit to the spectest svn repo.
Fozo okay but say I wanted to contribute to the code, or sytenx of the launguage is there I way I "sign-up" for this?
colomon Working on tests is a very useful way to get started, that's what I did. 00:48
submitting short patches to the code is certainly doable.
Fozo hmmm what do I mean? excuse if I am a complete noob
colomon you don't really sign up, in general. you just do it.
Fozo yes that sounds like something I would be intertsed 00:49
in*
so where do I get the source code?
colomon github
github.com/rakudo/rakudo 00:50
instructions are on that page.
everyone is quiet because they're logged in but elsewhere -- in many cases, asleep.
(say, the Europeans) or at least having a quiet Sunday away from their computers. 00:51
Fozo now is this just to contribute to the complier or the whole project?
colomon I imagine a lot of the Americans are traveling to YAPC::NA, that was my day as well.
rakudo's source is just that. there's also the Parrot project for the VM underlying Rakudo. 00:52
and the spectest suite, which is for all versions of Perl 6.
Fozo hmm where is that website, I was confused on what parrot auctally was 00:53
colomon parrot.org
but unless you explicitly want to mess with parrot, it's easier to just let Rakudo build it for you and not worry about it.
Fozo no for the suite
colomon svn.pugscode.org/pugs/t/spec 00:54
Fozo so I installed parrot and now I need a plug in? 00:55
colomon how did you install parrot? 00:56
Fozo I used the sourceforge main install
LOoL! I'm sooo confused 00:57
colomon github.com/rakudo/rakudo has install instructions about halfway down the page... 00:58
Fozo so I need a c ide and a perl ide? 01:03
I have strawberry perl
wait so I have installed Parrot and all I see in my task bar is unistall parrot 01:05
colomon you need a C compiler and Perl 5.8 or better, yes. 01:09
Fozo okay then do I run parrot or something the instructions are rather confusing, so please excuse my most likely stupid questions
okay then do I run parrot or something the instructions are rather confusing, so please excuse my most likely stupid questions 01:10
01:12 Mowah left
colomon do you have git? 01:14
Fozo I made an account
colomon errr... backing up -- what platform are you on? 01:15
Fozo I made an account I think
Windows
colomon okay, do you have a C/C++ compiler? 01:17
dalek kudo: b98d7fa | (Solomon Foster)++ | src/core/ (2 files):
Add Real.ACCEPTS.
 01:18
01:20 mikwss left
Fozo yes dev c++ 01:21
will that work
will it have any licinse issues?
colomon Rakudo or Parrot don't have licenses that might cause issues. 01:22
colomon wishes jnthn was here, he builds Rakudo on Windows every day... 01:23
Fozo no I mean do contribute to perl do I need a spefic liscense or can free downloads work, in other words I am wondering if Dev c++ has a license that will not working for distributing perl, does perl make any money off of its disbutions that I do not know about?
to* 01:24
colomon The Perl Foundation doesn't make any money off of it, but there's nothing stopping someone else from making money from it -- like ActivePerl does for Perl 5. 01:25
Fozo I can always use eclipise
okay hold on let me check something 01:26
is the GNU GPL liscense for my complier work 01:27
01:27 meppl left
Fozo "Status : Free Software (under the GNU General Public License) " 01:27
according to my complier's website 01:28
????????????? 01:30
01:32 ashleydev joined
Fozo hmm 01:33
hmm
colomon your compiler is GNU?
Fozo I believe so
unless the main site is lieing to me 01:34
sorear you're repeating yourself...why?
Fozo what OS are you on, I assume LInux?
colomon I'm on OS X, though I build Rakudo on Linux a couple of times a month as well. 01:35
Fozo sorry bad habbit excuse that
so will the GNU license work in order so I can devolp
sorear GPL does not restrict how you can use the software *in any way* 01:37
colomon Probably at least half of us (including me) are using GNU-based C compilers for this work.
sorear you can use a GPLed compiler to write commercial software
the software which runs macos and the iphone is all compiled with GNU CC 01:38
Fozo okay thank you
01:38 xinming joined
sorear also, if you want to contribute code to Rakudo you'll need to mail a signed copyright sharing agreement to TPF 01:39
Fozo okay now colomon what elese will I need, thank you soo much for all of your help
sorear so it's probably better to start somewhere else
than the compiler core, I mean
Fozo hmm TPF? 01:40
Fozo feels very stupid 01:41
Fozo feels very very stupid
colomon The Perl Foundation
01:42 xinming left
colomon you can contribute small bits without, but for bigger stuff, yes, you need to sign their agreement. 01:42
Fozo well could I email' it?
well could I email' it? to them
what do you mean by "small" and "bigger"? 01:44
what do you mean by "small" and "bigger"? 01:45
sorry
sorear do you realize that every time you type a line twice, it gets forwarded to 172 people who then have to look at both copies? 01:46
colomon I'm not sure what the dividing line is, but I know I contributed several small patches before sending my form to them.
Fozo Sorry, I will stop doing that
01:47 yinyin left, xinming joined
Fozo Okay so I can just add like one small feture at a time to see if it is right for me? and then sign this contract and mail it? 01:47
01:48 TiMBuS joined, molaf_ joined
sorear Fozo: yes 01:49
the best thing you can do is just lurk here for a while
when you see something interesting, JUMP ON IT
Fozo okay then I have a C complier and parrot now what in theory would be my next step
01:49 whiteknight left
Fozo hmmm okay 01:49
colomon you need git 01:50
not just github, but actual git software for your machine
Fozo oh okay where would one install such a sosftwear 01:51
???
colomon help.github.com/
look at install git link there.
01:51 yinyin joined 01:52 molaf left
Fozo okay now I have that installing 01:55
all done 01:56
what is the next step? 01:57
colomon you have a perl 5? 01:58
Fozo I installed strawberry perl 01:59
so I think I do
01:59 patspam joined
colomon strawberry should be fine. 02:00
then you need to do the equivalent of
git clone git://github.com/rakudo/rakudo.git
Not sure how to do that with Windows git. 02:01
might just be a matter of typing that at the command line.
02:02 yinyin left, yinyin joined
Fozo wait I have the git gui but the problem is do I have to install a certain file? 02:03
what do I clone?
I c that you contributed just recently 02:04
02:04 [mark] joined
colomon while we were talking 02:07
what you are cloning is git://github.com/rakudo/rakudo.git 02:08
Fozo wait what I put that in a browswer?
colomon no
on the command line, you just say 02:09
git clone git://github.com/rakudo/rakudo.git
Fozo okay 02:10
who uses git on windows here? 02:14
rrgh why is this so hard 02:15
it did not wotk 02:17
work*
can anyone help me with this 02:19
??????
hmmm I wish I could contribute to perl, but whatever thank you everyone for all of your help but I think I'm gonna lave 02:22
leave*\
02:24 Sarten-X joined 02:25 Sarten-X is now known as SartenX
TiMBuS Fozo, I can try to help if you want 02:26
i know git for windows is a bit annoying
02:26 yinyin left 02:27 agentzh joined
Fozo thanks 02:27
TImBus how do I clone the repsoitory
TiMBuS well that depends on what git you are running
TortoiseGit? 02:28
Fozo umm
I think mysgit
TiMBuS ah, so you need to run it from the command line
Fozo I have a GIt Gui thing
but I will install what ever is the easiest to teach me
TiMBuS sorry im just looking for what gui thing you could mean so i can see it 02:29
Fozo SO TImBus it is up to you
I will get tortiseGIt if you wish 02:30
02:30 yinyin joined
TiMBuS it might be easiest, unless you're familiar with using the dos command line 02:31
colomon TiMBuS++ 02:32
Fozo not too much only a bit so let me get it 02:34
TiMBuS sure. don't uninstall msysgit though! its needed by tortoisegit 02:35
Fozo TImBus thank you for your help 02:36
TiMBuS no problem
Fozo colomon thank you too
okay TimBus what now ( I Say your name so you will get a notice if you swap windows) 02:37
sorry
I no it is very annoying I apoligize 02:38
TiMBuS thats fine
Fozo umm what do I do now that it is installed? 02:39
TiMBuS you want to make tortoise clone the rakudo repository. 02:40
its relatively simple
Fozo yes I want to join the devolpment of perl 6 and I was told I needed to do that as my next step
TiMBuS i think tortoise has installed a right-click menu when you click on a directory in explorer 02:41
Fozo okay go ahead TIMBus
What do you mean?
TiMBuS when you right-click a dirctory, or your desktop, there should be options for tortoisegit 02:42
Fozo I don't have the directory to clone
I don't know what to clone! 02:43
TiMBuS thats ok. the first thing you need is to find tortoisegit's 'clone' option. it might be in the main gui 02:44
vimeo.com/9984118 according to this video, you merely need to right click in any directory (you can make your own) and the 'clone' option is there 02:45
Fozo Colomon what is the thing I want to clone? 02:46
TiMBuS git://github.com/rakudo/rakudo.git is the url
Fozo ty 02:47
sucuss
now what is the next thing that I must do
02:47 alester joined
TiMBuS well for now, you should have rakudos source code 02:48
Fozo right I think I do I also have parrot and I also have a complier for C and I have strawberry Perl 02:49
TiMBuS thats good, then you'll need to configure and build rakudo.
to do that, go into rakudo's directory (the one you just cloned) 02:50
you'll need to run Configure.pl first. it should work just by double clicking it (i hope)
itll pop up a window for a second and then close, hopefully because it worked 02:51
and now we hit unknown territory because i have no idea how the windows 'make' works =/. where is fperrad when i need him 02:53
Fozo no the Pl file just gives me a txt file 02:56
TiMBuS oh i see
well, youll need to use the command line to run it. hold 'shift' and then right click a blank spot in the folder so you get an extended right click menu 02:57
Fozo okay what do I do there? 02:58
TiMBuS and click 'open command prompt here'
i'm assuming you are using vista or windows 7 02:59
Fozo nope XP ftw 03:00
03:00 lestrrat is now known as lest_away
TiMBuS ah damn. you'll need to goto start->run and type cmd then 03:00
Fozo yah you should of said open comand prompt 03:02
let me guess
let me guess I need to cd it to that folder?
TiMBuS yep thats it
Fozo k now what? 03:03
TiMBuS type 'perl Configure.pl'
Fozo it came up with an error 03:04
TiMBuS perl isnt a valid command?
Fozo no it said 'parrot install' was not 03:05
TiMBuS oh
but you installed parrot? thats not good :/ 03:06
Fozo I think I did
TiMBuS on linux you can install parrot for using rakudo, and another parrot for making rakudo. 03:07
is there another option in the parrot install to install the 'developer' parrot?
I'm just guessing here, I wouldnt know 03:08
Fozo hmm I'm confused I dunno 03:09
TiMBuS this would go a bit easier if i still used windows
i can't really solve this one on my own, but i'm sure there are others in here who develop rakudo on windows! sorry i could only get this far for you 03:10
but at least you have rakudo's source code, so you can at least look at it, if that's what your aim was 03:11
Fozo LOoL! well ty all of you 03:12
where is the source any way 03:13
03:13 justatheory left
TiMBuS you can also probably write perl6 code since I think parrot comes with rakudo. 03:14
the source should be in src
Fozo how do I view .pm files 03:15
TiMBuS just in a text editor
i hear notepad++ is good
03:19 Fozo left 03:26 Fozo joined
Fozo I'm back TimBus 03:26
how do I read the src files 03:27
03:32 Fozo left 03:34 Fozo joined
TiMBuS Fozo, all of them should be text files. .pm files are written in perl 6 03:35
Fozo wait I don't see really any text files only .pm 03:37
sorear .pm is a kind of text
Fozo how do I open them?
sorear use a text editor
22:15 < TiMBuS> i hear notepad++ is good 03:38
having not used windows in seven years I can't comment myself
TiMBuS i remember i used to write c code in plain old notepad. i don't recommend it 03:39
Fozo hmmm so now what is the point of doing configure.perl? 03:40
sorear the rakudo code won't open in notepad anyways unless microsoft seriously improved it in the last seven years
configure.pl sets up rakudo so you don't have to 03:41
Fozo an I set it up manually it won't work for me
sorear you'll need to be more specific
"it won't work" is fine if I'm standing over your shoulder
but I'm not
Fozo it opens up in plain notepad 03:42
it does nothing just pops open a txt file
TiMBuS configure.pl is a text file that perl runs 03:43
Fozo oh then how do I run it in perl
03:45 plobsing left 03:46 plobsing joined
sorear open a command prompt 03:48
type "perl configure.pl --gen-parrot"
Fozo it did not work it said retry using the same command or the other one I tried 03:50
03:51 tedv left
sorear Did you follow the directions it gave? 03:51
Fozo what instrutions I did follow the one the comand prompt said 03:52
sorear There isn't enough punctuation in that sentance for me to figure out what words are connected to what. 03:53
Fozo what instrutions? I did follow the ones that were given by the comand prompt. 03:55
**
sorear And what happened when you followed them? 03:57
Fozo nothing same message 03:59
TiMBuS whats the message exactly? that parrot can't be found? 04:00
Fozo okay hold on it is rather long 04:02
sorear paste.lisp.org/new/perl6 04:04
use this
Fozo "to automaticly checkout(svn) and build a copy of parrot r47723 try re-running configure.pl with the '--gen-parrot' option. or use '--parrot-config' option to specify the location of parrot_config to be used to build Rakudo Perl." 04:06
04:08 patspam left 04:12 supernovus joined
sorear Fozo: what was the command you typed to get that output? 04:13
perl6: say int.^parents
p6eval elf 31394: OUTPUT«Unknown rule: dotty:.^␤It needs to be added to ast_handlers.␤ at ./elf_h line 2850␤»
..pugs: OUTPUT«*** No such method in class Class: "&parents"␤ at /tmp/kv2lx3k5tq line 1, column 5 - line 2, column 1␤»
..rakudo b98d7f: OUTPUT«Could not find sub &int␤ in main program body at line 11:/tmp/NSTT0GKKIc␤»
Fozo perl configure.pl 04:14
and then what you recomend
sorear 22:48 <@sorear> type "perl configure.pl --gen-parrot"
ah
Fozo and then
perl configure.pl --parrot-config
sorear "or" does not mean "do both" 04:15
you should only have run perl configure.pl --gen-parrot
04:15 s1n joined
sorear what happened when you ran that? 04:15
TiMBuS im going to guess it cant find svn ._. 04:16
Fozo what is svn I don't think I got that
TiMBuS fperrad should bundle his toolchain as a zip 04:17
Fozo what can I have a link? 04:18
sorear google "svn"
it's the first link
Fozo apache thing? 04:20
sorear yes
Fozo where is download on the site? 04:24
04:26 Fozo_ joined 04:28 JimmyZ joined
Fozo_ so where is download I did not see one? 04:28
JimmyZ rakudo builds failed on Windows ...
04:29 Fozo left
supernovus subversion.apache.org/packages.html 04:29
04:29 Fozo joined 04:30 Fozo__ joined
Fozo__ sorry missed it agian 04:30
sorry missed it agian because I failed twice on my internet 04:31
supernovus Fozo: subversion.apache.org/packages.html
04:33 Fozo_ left, Fozo left
Fozo__ which package? 04:36
supernovus Fozo__: There are several different versions. Here is the CollabNet client-only release for Windows: www.open.collab.net/servlets/OCNDir...1.6.11WINC 04:38
Fozo__ is this one the only one I can use? 04:40
04:40 JimmyZ left 04:43 yinyin left
supernovus There are many different command-line subversion clients for Windows. I picked the CollabNet one as it was on the top of the list on the Apache page. I haven't used subversion on Windows, but I do know CollabNet was the company that originally created subversion (svn) so I would assume their package would be a good choice :-) 04:45
Fozo__ okay it is just that they have a whole regrestaration process thatr I do not want to go thru 04:46
sorear Please, out of respect for the other 174 people here, be concise.
supernovus Fozo__: Here's another subversion client from "slik". It does not require registration. www.sliksvn.com/pub/Slik-Subversion...-win32.msi 04:48
Fozo__ thx but I have one 04:52
now what did I just install?
04:52 yinyin joined
Fozo__ okay now I have a new error 04:54
sorear Good! That means you've made prgress. 04:55
supernovus In the current rakudo is 'constant' implemented? 04:56
04:58 synth left
sorear rakudo: constant $pi = 3.14; say $pi; 05:08
p6eval rakudo b98d7f: OUTPUT«===SORRY!===␤Symbol '$pi' not predeclared in <anonymous> (/tmp/7Vyod4hBg_:11)␤»
sorear rakudo: constant pi = 3.14; say pi;
p6eval rakudo b98d7f: OUTPUT«Could not find sub &constant␤ in main program body at line 11:/tmp/DIvtnFvGi4␤»
supernovus well, that's the same results I'm getting
I tried different scopes too
Fozo__ for me it does not run lke 194
supernovus rakudo: our constant $pi = 3.14; 05:09
p6eval rakudo b98d7f: OUTPUT«===SORRY!===␤Malformed our at line 11, near "constant $"␤»
supernovus rakudo: my constant pi = 3.14;
p6eval rakudo b98d7f: OUTPUT«===SORRY!===␤Malformed my at line 11, near "constant p"␤»
Fozo__ well thanks every one bed But I am going to log, and I guess I just won't devolope for perl 6 you guys have enough staff I presume allready :( 05:12
lue We could always use more! Feel free to help in any way you can! :) 05:13
Fozo__ I would if I could but I can't
I really don't even have any experence with making launguages and perl in general 05:14
but alas 05:15
supernovus Fozo__: Not all contributors need to be core language developers. There are many different ways you could help out. I'm sure we can help you get Rakudo running. Then you'll be able to experiment with the language itself, which is the best way to start with Perl 6. 05:16
Fozo__ well i dunno cya 05:17
thank you all!
Fozo__ goes to close the window 05:18
05:18 Fozo__ left, lest_away is now known as lestrrat 05:21 alester left
supernovus no paste bot atm? 05:22
sorear nope 05:34
it was turned into a spam vector, so we removed it 05:35
supernovus Ah, okay. Well then: paste.lisp.org/+2E5R
05:37 orafu left, orafu joined, Su-Shee joined
supernovus oh nice, I'm getting a lot of "Method 'key' not found for invocant of class String" errors in all sorts of places... 05:46
rakudo: say %*ENV.perl
p6eval rakudo b98d7f: OUTPUT«Method 'key' not found for invocant of class 'String'␤ in 'Any::join' at line 4315:CORE.setting␤ in 'EnumMap::perl' at line 4321:CORE.setting␤ in main program body at line 11:/tmp/HlpvZrMFUM␤»
supernovus It seems some of the magic variables such as %*ENV and %*VM have a few issues yet... 05:53
rakudo: for %*ENV.kv -> $k, $v { say "$k = $v" } 05:54
p6eval rakudo b98d7f: OUTPUT«Method 'key' not found for invocant of class 'String'␤ in main program body at line 4315:CORE.setting␤»
supernovus Unrelated, but augmenting Hash appears to have some issues as well. 06:13
rakudo: use MONKEY_TYPING; augment class Hash { method has { say "CAN HAS" } }; my %h; %h.has(); 06:14
p6eval rakudo b98d7f: OUTPUT«Method 'has' not found for invocant of class ''␤ in main program body at line 11:/tmp/DAiougOILe␤»
sorear TimToady: ping 06:15
06:18 Chillance left 06:24 PZt left 06:32 rgrau left
dalek kudo: c6a829d | moritz++ | docs/ChangeLog:
add "handles" to ChangeLog
 06:41
06:47 casan joined
Tene supernovus: I committed some fixes for %*VM today. Can you confirm whether there are still issues with %*VM in the latest rakudo? 06:52
rakudo: say %*VM.perl
p6eval rakudo b98d7f: OUTPUT«Cannot substr on a null string␤ in 'Pair::perl' at line 1␤ in <anon> at line 4321:CORE.setting␤ in 'Any::join' at line 1␤ in 'EnumMap::perl' at line 4321:CORE.setting␤ in 'Pair::perl' at line 4078:CORE.setting␤ in <anon> at line 4321:CORE.setting␤ in 'Any::join' at line
..1␤ …
Tene Huh.
rakudo: say %*VM<config>.perl
p6eval rakudo b98d7f: OUTPUT«Cannot substr on a null string␤ in 'Pair::perl' at line 1␤ in <anon> at line 4321:CORE.setting␤ in 'Any::join' at line 1␤ in 'EnumMap::perl' at line 4321:CORE.setting␤ in main program body at line 11:/tmp/C1xUhPMs7k␤» 06:53
Tene rakudo: say %*VM<config><build_dir>
p6eval rakudo b98d7f: OUTPUT«/home/p6eval/rakudo/parrot␤»
Tene rakudo: say %*VM<config>.WHAT
p6eval rakudo b98d7f: OUTPUT«Hash()␤»
moritz_ rakudo: say %*VM<config>.keys.join(' ') 06:55
p6eval rakudo b98d7f: OUTPUT«mandir has_sigaction icu_dir HAS_GETTEXT osname i_signal encoding libparrot_ldflags cc_o_out has_static_linking runcore_tests HASATTRIBUTE_COLD gccversion has_extra_nci_thunks ld_debug rpath_blib clock_best make msvcversion installed pmc has_opengl i_sysexit a
..ccwarn::src/spf_rende…
06:57 PZt joined 07:06 tadzik joined 07:07 casan left
pugssvn r31395 | sorear++ | [STDeco] Tie up a loose end from the pad->lex rename 07:34
07:34 supernovus left
sorear How much rule of least suprise violation would it cause if BEGIN time code ran slower than runtime code? 07:36
moritz_ you just have to sell it better 07:37
"run time code is now much faster than BEGIN"
hugme: add qiuhw to book 07:41
hugme hugs qiuhw. Welcome to book!
mathw (footnote: accomplished by making BEGIN slower) 07:42
07:43 pnate left 07:47 Mowah joined
mberends just eated a kanelbulle irl :) 07:49
mathw kanelbulle? 07:50
moritz_ sounds like some swedish stuff
mberends it's a cinnamon bun, shaped like an @
moritz_ doughnutpretzel? :-) 07:51
mathw oooh 07:53
we have those here sometimes 07:54
lovely things
07:54 tadzik left
dalek ok: 1126727 | (Hongwen Qiu)++ | src/subs-n-sigs.pod:
[subs] fix typo
 07:54
07:54 Mowah left 08:02 pnate joined 08:14 kensanata joined 08:16 dakkar joined 08:17 cosimo left 08:22 yinyin left
pugssvn r31396 | sorear++ | [Cursor] A more platform neutral way to recurse the parse 08:27
08:27 cygx joined
cygx moritz_: hi 08:28
08:30 thebird joined
moritz_ cygx: hi 08:31
cygx: I saw your nopaste in the backlog... do you have a github ID?
cygx no, just a mercurial account at intuxication.org 08:33
that can be remedied, though
08:33 plobsing left
moritz_ great 08:33
sorear mildew: say int.^methods
p6eval mildew: OUTPUT«Can't open perl script "mildew": No such file or directory␤»
08:33 Woody2143 left
moritz_ how can perl know it's a perl script if it doesn't exist? :-) 08:34
mathw Because it assumes that you won't be asking it to run something written in Lisp, I guess
cygx it's quantuum ;) 08:35
mathw Very narrow-minded of it
cygx ok - that's me: github.com/cygx 08:36
moritz_ hugme: add cygx to try.rakudo.org
hugme hugs cygx. Welcome to try.rakudo.org!
moritz_ cygx: you can now git clone github.com/moritz/try.rakudo.org 08:37
arnsholt mathw: Well, that's part of the problem Perl 6 aims to solve, no? ^^
moritz_ cygx: incorporate your changes, commit and push
and profit!
mathw arnsholt: of course, there'll be no need for Lisp anymore :) 08:38
cygx moritz_: thanks
moritz_ cygx: thank you for contributing!
08:42 sdo_ joined
jnthn morning 08:42
mathw I've been here too long 08:43
moritz_ morning jnthn
mathw I'm trying to write some Perl 5 and I keep getting frustrated that I don't have gather/take
moritz_ of course there's Perl6::GatherTake
mathw It feels so *clumsy*
moritz_ (but it sucks. I'm the author, so I'm entitled to that opinion :-)
08:44 sdo_ left 08:48 |jedai| joined, jedai left 08:49 sdo joined 08:52 azawawi joined, azawawi left, azawawi joined
azawawi moritz_:ping 08:52
sorear azawawi!
moritz_ azawawi: pong
azawawi moritz_: re perlgeek.de/blog-en/perl-6/contribu....org.html, i think we can reuse a lot of the already-working runpugs console javascript code 08:53
sorear azawawi: What should I do with a 31KB binary data file that STD needs to be able to find?
azawawi sorear: the utf8 data file?
moritz_ azawawi: wfm
sorear it's a Unicode data file 08:54
not specifically utf8
but yeah
08:54 Mowah joined
azawawi sorear: include it via PERL6LIB ? 08:54
sorear: one of things that i hate the most about STD is the use of the / namespace 08:55
sorear: it should be inside a good Perl6::STD or something
sorear: $PERL6LIB/data ? 08:56
sorear yeah, sounds good
azawawi moritz_: cool i'll try it
sorear it's in the root namespace because they are specced parts of Perl 6 itself 08:57
moritz_ azawawi: maybe coordinate with cygx, who also had some running js code
sorear any full Perl 6 compiler will support use STD;
and Cursor is the implicit base class of all grammars
08:58 dju joined 09:00 dju left, dju joined 09:01 globalmixmaster_ joined
globalmixmaster_ hi all 09:01
can i optimize this pastebin.com/UKJAhBHz 09:02
with perl6?
moritz_ globalmixmaster_: it won't do you any good, because perl 5 is currenlty faster than all Perl 6 compilers
globalmixmaster_: I'd use C if were you 09:03
globalmixmaster_ ok ty
09:03 globalmixmaster_ left
sorear haha 09:04
TiMBuS what,
sorear hmm. 09:05
azawawi sorear: sounds good then... one thing to watch for if we go for $PERL6LIB/data is accidentally polluting other namespaces... 09:06
sorear foo; means foo(); means &foo.(); means postcircumfix:<( )>(&foo); means &postcircumfix:<( )>.(&foo)
where does the circularity saw stop?
azawawi: I'm just putting it in $PERL6LIB directly
azawawi sorear: great thx 09:07
colomon TiMBuS: thanks for helping that guy last night.
TiMBuS thats ok. i'm still unsure as to how he wanted to help, but i couldnt turn him away
pugssvn r31397 | sorear++ | [Cursor] Look for Unicode bits in $PERL6LIB. azawawi++
09:14 azert0x joined 09:16 mac joined, mac left 09:17 azawawi left 09:22 foodoo joined
pugssvn r31398 | sorear++ | [STDeco] Add 'make dist' target for the benefit of Syntax::Highlight::Perl6 09:23
colomon is S::H::p6 still getting updated? 09:25
sorear I should hope so 09:26
azawawi all but asked me to do that
colomon sweet 09:27
09:30 mmcleric joined
moritz_ yesterday's blog post about physical modelling in Perl 6 attracked about 3k visitors 09:40
jnthn Wow! 09:41
arnsholt A grammar (NQP-rx) conundrum for your amusement:
moritz_: Awesome!
jnthn moritz_++
colomon moritz_++
jnthn
.oO( I hope not too many of them came expecting to look at pictures of the other sort of models... :-) )
09:41 moritz_ sets mode: +oo colomon arnsholt
arnsholt :- can be three things: an atomic term, a prefix operator and an infix operator 09:41
moritz_ jnthn: :-) 09:42
arnsholt How on earth do I get "a :- b", ":- a" and "X = :-" to parse correctly, all at the same time
?
moritz_ sub infix:<:-> { <sym> } 09:43
arnsholt If I kill the prefix variant I get the other two, but I really need the prefix as well
moritz_ erm, multi sub infix:sym<:-> { <sym> }
multi sub term:sym<:-> { <sym> <EXPR>? }
(maybe needs to constrain the precdence of what <EXPR> is allowed to match) 09:44
sorear arnsholt: you'll need to decide how :- :- :- is supposed to parse
jnthn Maybe a bit like * * * in Rakudo. ;-)
sorear * in rakudo isn't a prefix op
arnsholt sorear: I can't remember how it's supposed to go off-hand (it's Prolog FWIW) 09:45
sorear yeah, I recognize the samples
arnsholt Yeah, it's the prefix op that really makes trips me up
moritz_ it kinda messes up the predicitve parsing 09:46
jnthn No, true. Rakudo has examples of things that are prefix and infix, as well as infix and term, but not all three.
moritz_ because after a :- either a term or an infix can occur
jnthn Well
[+] can show up as any of
[+] @thingy
[+] # with nothing after it
$a [+] $b
moritz_ rakudo: say [+] 09:47
p6eval rakudo c6a829: OUTPUT«===SORRY!===␤Unable to parse postcircumfix:sym<[ ]>, couldn't find final ']' at line 11␤»
jnthn Though we don't get the middle case right in Rakudo yet.
moritz_ jnthn: but don't parse it :-)
09:47 am0c left
jnthn moritz_: Yeah, STD looks like it lies to the OPP about being a term or something. ;-) 09:47
Or something tricky.
Anyway, may be worth looking at how STD does it.
moritz_ arnsholt: I'd try parsing it as infix or term, where the term can optionally swallow up another term
arnsholt sorear: ":- :- :- ." gives me "Operator priority clash" after the first :- in SWI Prolog 09:48
moritz_: Won't the optional EXPR after the term mess up the rest of the parse?
moritz_ arnsholt: why should it? 09:49
you have to go looking for *something* after a :-
arnsholt Heh, true
moritz_ you can make that either a term or an infix
term with <EXPR>?, infix with <EXPR>?? 09:50
arnsholt How do those limit to term and infix? 09:51
moritz_ with <EXPR>? the first attempt will be to match an EXPR (ie new term, with optional prefixes) 09:52
with <EXPR>?? it first tries to match nothing, so the OPP will try an infix after it, no?
arnsholt Oh, right 09:53
09:57 masak joined
masak oh hai, #perl6! 09:57
phenny masak: 20 Jun 17:45Z <sorear> tell masak declarations? class, function, variable, you name it. .pm6 -> .pod converter
masak: 20 Jun 18:39Z <moritz_> tell masak nopaste.snit.ch/21416 feedback welcome
masak: 20 Jun 19:03Z <moritz_> ask masak how t/hitomi/01-xml-parsing.t is supposed to work. It has a line 'use Hitomi;', but there's no Hitomi.pm to be found anywhere near...
masak moritz_: the Web.pm tests generally need some love. this is one instance of that. I don't remember the exact backstory. 09:58
moritz_ ok
Tene++ and I also found that most routes test die with the same error (on alpha) 09:59
I've patched them up to produce the same failures with master (in the 'ng' branch)
masak moritz_: I think t/hitomi/01-xml-parsing.t is simply bogus. an attempt to start coding with tests, and then Hitomi itself (which works to a large extent) took off in another direction.
hm, or there's supposed to be a lib/Hitomi.pm, but that was never committed. 10:00
that looks more probable, actually.
moritz_ [t/hitomi/01-xml-parsing.t] added more tests 10:01
The last one fails. Don't know why yet.
so it seems they ran at some point
masak yes.
I'll have a look at it soonish and see if I can get it to work again.
jnthn o/ masak 10:02
masak moritz_: re nopaste.snit.ch/21416 -- looks good. let's hope the web-challenge effect brings in many contributions. :)
jnthn: \o
moritz_ masak: actually it's already running, and cygx++ has contributed something already
masak oh wow. 10:03
moritz_ though not committed yet
masak I should avoid sleeping altogether, there's so much happening here.
masak backlogs
moritz_ :-)
10:05 lestrrat is now known as lest_away
mathw masak: but just think how much joy you get out of knowing that when you wake up and come to #perl6, something awesome will have happened 10:15
cygx moritz_: speaking about committing - I'm new to git/github, so I'm not sure ho to go on
moritz_: should I fork and commit to a clone? 10:16
moritz_ cygx: nope, commit to the repo directly
that saves me the trouble of merging
the forking/push to fork model is often used, but I rather hand out commit bits 10:17
cygx ok; will do after some cleanup
masak mathw: when you put it like that, is does sound rather nice... 10:24
phenny: tell sorear -- still not 100% sure what you're asking. perhaps best that you follow S26 to the letter until I get around to updating it to the latest discussions. in the best case, you'll also become convinced that the current way isn't optimal. :) 10:26
phenny masak: I'll pass that on when sorear is around.
10:27 |jedai| left 10:29 azert0x left
bbkr any idea why "Determine what C compiler and linker to use" test on --gen-parrot returns "Permission denied" for parrot/test_X ? there is only snprintf in *.c file, nothing suspicious 10:30
10:41 swaroop joined
mmcleric hi 10:42
just sent patch about WhateverCode type
10:42 swaroop left
mmcleric it doesn't fix 2 + * + 3 yet, it turned out to be complicated (for me at least :) ) 10:42
and i've got a lot of stupid questions
first, i can tag any PAST::Node object simply by using it as hash, right? 10:44
jnthn mmcleric: You can but...
...we try and avoid doing so.
I figured that .returns('WhateverCode') would be enough. 10:45
(e.g. enough tagging)
mmcleric isn't .returns only Op's property?
jnthn I think it can go on anything, but you will only ever have Op anyway, I think. 10:46
mmcleric or even PAST::Var? i'm confused, where can i see whole list?
10:46 colomon left
bbkr (parrot bug above solved. disk was mounted with nodev) 10:46
jnthn Since all blocks are wrapped.
PAST docs are at...moment..
mmcleric i'm reading pdd26_ast, it doesn't say anything about other types
ok :)
jnthn docs.parrot.org/parrot/latest/html/...t.pod.html
oh heh, that only claims it works on PAST::Val, afaict 10:47
It's true that the code generator ignores it on everything but PAST::Val.
But I believe you can apply it to anything just for "tagging" purposes. 10:48
mmcleric oh, ok :) but it's still there, so docs are lying
jnthn Yeah
docs--
well
mmcleric ok, another question
jnthn (lying docs)--
mmcleric is it really necessary that () in expressions create new Stmts node?
because 2 + ((((3 + *)))) will have 4 Stmts nodes, one in another
and () doesn't cause whatever_curry to be called :) 10:49
jnthn eww
I think it may be needed though
mmcleric so i'll have to put it in circumfix:sym<( )>($/), right?
jnthn For @other-thing
Erm
That feels...weird. 10:50
But maybe we'll need to
I think maybe put something in to "pass the tag"
10:50 lest_away is now known as lestrrat
jnthn e.g. the .returns of the PAST::Stmts is set to whatever the .returns of the (last) expression in it is. 10:50
mmcleric yes, i thought so too (but didn't tried it yet) 10:51
jnthn But I'd prefer that whatever_curry should only be called from the one place.
Propagating .returns around the tree is just generally healthy, anyway.
We'll want to do more of that in the future, I suspect.
(In the future, when we get into doing type analysis and so on.)
mmcleric ok, good 10:52
another thing that worries me is performance 10:53
i understand it's only compile time hit, but still, checking for types on every operator seems wasteful
should i worry about it? 10:54
jnthn Well, what else can we do, really?
I mean, any operator could potentially be whatever-curried.
10:54 dakkar left
cognominal rakudo: class A { method TOP { self.a(&A::b) }; method a() { self."$b"() }; method b { say "hi" } }; A.TOP 10:54
jnthn We could try and do epic evil with a contextual that we flag if we see a Whatever
p6eval rakudo c6a829: OUTPUT«===SORRY!===␤Symbol '$b' not predeclared in a (/tmp/8Ol0nAYmcn:11)␤»
mmcleric contextual is something from Grammar.pm, right? 10:55
jnthn Yeah but...let's not do that. Not now, anyway.
mmcleric ok :)
jnthn It's too...evil.
10:55 dakkar joined
cognominal rakudo: class A { method TOP { self.a(&A::b) }; method a($b) { self."$b"() }; method b { say "hi" } }; A.TOP 10:56
p6eval rakudo c6a829: OUTPUT«Null PMC access in can()␤ in 'A::a' at line 11:/tmp/mZUyIIQ2h5␤ in 'A::a' at line 11:/tmp/mZUyIIQ2h5␤ in 'A::TOP' at line 11:/tmp/mZUyIIQ2h5␤ in main program body at line 11:/tmp/mZUyIIQ2h5␤»
mmcleric i didn't know about contextual and thought about simply copypasting all expr-grammar into two versions, with and without Whatever (yes, it's even more evil)
ok, i'll try not to think about it
jnthn: ok, thanks. there are some things which stop me from getting working implementation, but i'll just try to read more docs, answers are probably there 11:00
cognominal I am not sure of the syntax to pass a method as a parameter and then call it :(
It seems that I need to prefix the method name by the class name to refer to it, even within the class itself 11:01
jnthn cognominal: It appears you expect the method to be in the namespace.
Or the lexpad. 11:02
It won't be unless you declare "my method ..." or "our method ..."
11:02 synth joined
jnthn cognominal: methods are has-scoped by default 11:02
mmcleric: OK, let me know if you hit any more really sticky points. :-)
mmcleric i will. it's going to busy week at my job, but maybe at nights i'll find some free time :) 11:03
11:03 mmcleric left
cognominal jnthn, what would be the correct way to write that? I test with genuine class, but I want to do it with a grammar. 11:04
cygx moritz_: I have 2 commits ready, but my git-FU is weak (more)
moritz_: with my freshly installed cygwin-git, all I'm getting is `error: Cannot access URL github.com/moritz/try.rakudo.org.git/, return code 22`
11:09 synth left
jnthn cognominal: In above, maybe try "our method b" rather than just "method b" 11:09
cognominal: Or do "my method b" and then &b rather than &A::b
cognominal rakudo: class A { method TOP { self.a(&b) }; method a($b) { self."$b"() }; our method b { say "hi" } }; A.TOP 11:11
p6eval rakudo c6a829: OUTPUT«hi␤»
cognominal jnthn++
11:12 egor joined
cognominal I should reread S12 to unserstand the merits of has scope for methods 11:14
jnthn cognominal: In short, it means "this method only gets entered into the methods table in the metaclass"
11:14 egor left 11:15 \shade\ left 11:16 shade_ joined
TiMBuS cygx, are you trying to clone the repository? 11:16
cygx TiMBuS: I cloned the repository and did 2 local commits; now I'm trying to push them upstream 11:17
TiMBuS do you have access to the github repo? 11:18
cygx TiMBuS: github tells me 'This URL has Read+Write access', so I guess I have? 11:19
moritz_ cygx: the trailing slash is wrong 11:20
edit your .git/config to remove it
TiMBuS also, that
cygx ;)
I'll first look up how to fix the author information, though, so it might be a minute 11:21
moritz_ should be
git config --global author.email [email@hidden.address] 11:22
git config --global author.name 'Your Name'
cygx moritz_: I'm not confident in the unicode abilities of my setup - let's see how it goes ;)
moritz_ git itself on a clean UTF-8 environment works fine 11:23
_sri kraih.com/rakudo-perl6-sri.png # refined 11:25
11:28 wrac joined
TiMBuS thats pretty nice 11:29
11:29 ejs joined
_sri :) 11:29
11:30 Guest23195 joined
arnsholt Yeah, I like it too 11:33
moritz_ it's just that rakudo will likely never have a version number "1"
TiMBuS the '0.' is implicit 11:34
_sri hah
moritz_ as it looks now, the compiler releases will be numbered as they are now (#30, 2010.06, 'Kiev')
and the distribution releases will have the 'Star' in the name
arnsholt IIRC pmichaud wants to avoid traditional version numbering altogether 11:35
TiMBuS can we name each one after a different star
marcus__ rakudo galaxy
11:35 marcus__ is now known as imarcusthis
TiMBuS super rakudo galaxy! 2 11:35
moritz_ star -> galaxy -> galxy cluster -> universe -> multiverse -> ?
TiMBuS well then 'rakudo ?' will be the rakudo 1.0 11:36
moritz_ THERE WILL BE NO RAKUDO 1.0 11:37
TiMBuS ahaha
11:37 thebird left 11:38 agentzh left 11:39 FardadJalili joined 11:41 azawawi joined 11:42 cognominal left
masak has anyone ever used the `given True { ... }` idiom? it's mentioned (in an earlier form) in A04. 11:44
azawawi phenny: tell colomon Yes, S:H:P6 gets updated regularly on my spare time. I try to do that once I see enough commits on STD.pm6, see cpansearch.perl.org/src/AZAWAWI/Syn...81/Changes
phenny azawawi: I'll pass that on when colomon is around.
azawawi masak: hi there
masak azawawi! \o/
azawawi: long time no see. 11:45
11:45 cognominal joined
masonkramer (use IO::Logging).assuming(logfile => ".log"); is going to make for some really neat APIs 11:46
azawawi masak: I always to read backlogs but lately I have been busy a bit at work
11:52 azawawi left
moritz_ rakudo: for 1..Inf { .say; last }; say "alive" 11:57
p6eval rakudo c6a829: OUTPUT«1␤alive␤» 11:58
moritz_ wow
masak \o/
moritz_ that's new, right?
masak think so.
moritz_ rakudo used to evaluate the argument eagerly
masak rakudo: for 1..* { .say; last }; say "for loops are lazy"
p6eval rakudo c6a829: OUTPUT«1␤for loops are lazy␤»
moritz_ just found it in autounfudge output
12:01 ejs left 12:02 dakkar left
takadonet morning all 12:02
masak \o 12:06
12:07 envi^home joined
cygx moritz_: git and/or cygwin positively hate me: they ignore settings, completely hang in the password prompt and don't like the github SSL certificate 12:13
I'll try to setup my environment properly later taday, but for now my changes can be found at www.informatik.uni-frankfurt.de/~cg...end.tar.gz
moritz_ if somebody wants a quasi-LHF to work on: write a constructor for class Match that accepts the values of $match.perl 12:14
cygx check www.informatik.uni-frankfurt.de/~cg...end.tar.gz for an online version
masak rakudo: use MONKEY_TYPING; class A {}; my $a = A.new; augment class A { has $.x }; say $a.x
p6eval rakudo c6a829: OUTPUT«Method 'x' not found for invocant of class 'A'␤ in main program body at line 11:/tmp/nmU9D4bNkC␤»
moritz_ I don't have much time atm, but the link is now in the logs, and I'll get back to it if you don't manage to push the stuff 12:15
jnthn cygx: I found the msys git (outside of cygwin) works very nicely, fwiw. 12:16
12:17 colomon joined
colomon at yapc;:na! 12:17
phenny colomon: 11:44Z <azawawi> tell colomon Yes, S:H:P6 gets updated regularly on my spare time. I try to do that once I see enough commits on STD.pm6, see cpansearch.perl.org/src/AZAWAWI/Syn...81/Changes
colomon azawawi++
cygx jnthn: I've got a somewhat Frankenstein-like mix of Cygwin and MinGW going for my C development and I'm not really keen on adding MSYS to the mix 12:18
but we'll see, I might reconsider
12:19 mmcleric joined
cygx or perhaps take alook at hg-git... 12:19
anyway, I'm off for now 12:20
12:20 cygx left
pmichaud good morning #perl6 (from yapc::na) 12:20
takadonet pmichaud: morning
masak morning! 12:21
colomon o/
moritz_ heya; how is yapc going?
colomon keynote okay so far
pmichaud just started
colomon I got kind of lost on the way here.
pmichaud not enough room in the room
moritz_ wil there be a 6y hackathon?
*will
colomon hope so
masak not enough space in this space...
pmichaud I plan to do six-y hack-y things on Thursday
jnthn parsed that as "6 year hackathon" :-) 12:22
moritz_ :-)
masak jnthn: that would be AWESOME.
jnthn masak: That's what I thought!
masak checks his calendar
how about... 2015 to 2020?
colomon I was planing on doing six-y hack-y things Mon - Wed. ;) 12:23
pmichaud a "6 year hackathon" is basically what I've done thus far, I guess. :)
I'll do six-y hack-y things Mon-Wed too, but I have presentations to prepare and give also :-|
jnthn pmichaud: Yes, but did you eat pizza every day? :-)
colomon pmichaud: right, and I have your presentations to go to. ;)
pmichaud jnthn: aha! Now we know why it's taken six years!
colomon also Damian's, etc.
masak pmichaud: that could be a nice presentation title: "Rakudo Perl: The Six-Year Hackathon" :)
pmichaud changes his title slide for tomorrow's talk :) 12:24
s/changes/creates/ :-)
masak :P
slavik "How to beat the Guinness world record for the longest hackathon." 12:26
cognominal jnthn++ # gave me the way to pass a rule as an argument to another 12:27
jnthn cognominal: I guess you're then using regex interpolation? 12:28
araujo slavik, 5 years? 12:29
slavik araujo: suggesting title for pmichaud's presentation 12:30
pmichaud: what's it about anyway?
araujo :P
well, still a 5 years hackathon is good idea
cognominal token begin-end(Str $str, Regex $content) { .... <content> ... } # jnthn, here what I do
oops
token begin-end(Str $str, Regex $content) { .... <$content> ... } # jnthn, here what I do
I am doing some stuff to generate slides that could be embedded in Pod6 12:32
masak cognominal++ # release early, release often! 12:34
pugssvn r31399 | moritz++ | [t/spec] some Rakudo unfudges
cognominal masak: I am more the anal retentive guy
masak cognominal: there's no real contradiction between the two :) 12:35
12:36 pjcj left
colomon TimToady: you seem to have found a power supply? 12:37
12:37 moritz_ sets mode: +oooo colomon masak cognominal araujo, moritz_ sets mode: +o slavik
TimToady I'm using an old computer that is actually the quiz machine for my wife's quiz team 12:38
it runs Winders
colomon oh noez!
mathw aaargh no
jnthn ...it's not *that* bad. :P 12:39
Unless it runs Windows ME, in which case it *is* that bad.
moritz_ right, you can always open a ssh connection to a unixish server and access your screen session there :-) 12:40
12:40 thebird joined, PacoLinux joined, masonkramer left, masonkramer joined 12:42 pjcj joined 12:43 bphillips joined 12:47 perlygatekeeper joined
masak it's fun reading the Apocalypses. did you know that &proceed was once known as &skip, before it was &continue ? :) 12:49
colomon did not
12:49 Intensity joined
colomon have you seen the old Perl 6 book? 12:49
12:50 rgrau joined, dakkar joined
masak only the cover. 12:50
jnthn I think I have a copy of it, but back in England.
moritz_ the perl 6 parts are in the pugs repo 12:51
I've updated them partially, but then lost interest
pugssvn r31400 | moritz++ | [t/spec] remove an outdated range test, and unfudge another that rakudo now passes 12:52
TimToady coffee # I sincerely hope... & 12:53
masak rakudo: given 42 { say "Int" when Int }
p6eval rakudo c6a829: OUTPUT«===SORRY!===␤Missing block at line 11, near "}"␤»
masak alpha: given 42 { say "Int" when Int }
p6eval alpha 30e0ed: OUTPUT«Int␤»
masak submits rakudobug
alpha: given 42 { say "Int" when Int; say "Num" when Num }
p6eval alpha 30e0ed: OUTPUT«Int␤Num␤»
masak statement-modifier 'when' (by spec) doesn't break.
12:54 colomon left
moritz_ right 12:54
isn't that proceed now?
12:54 colomon joined
colomon probably no point in giving me op today, I imagine this machine will be on and off on a regular basis. :) 12:55
masak moritz_: no, it's "succeed". as in, "now we're done with the block that bound $_". 12:56
colomon then I think I will head to the Perl 6 BOF for the rest of the morning.
masak rakudo: class MyException is Exception {}; throw MyException.new("OH ERROR") 13:03
p6eval rakudo c6a829: OUTPUT«Could not find sub &throw␤ in main program body at line 11:/tmp/2Ed568LNIm␤»
jnthn masak: die 13:04
masak rakudo: class MyException is Exception {}; die MyException.new("OH ERROR")
p6eval rakudo c6a829: OUTPUT«O␤ in main program body at line 11:/tmp/n1AH4xBz1X␤»
masak submits rakudobug
jnthn heh 13:05
what on earth is it meant to do there
stringify it I gues
*guess
13:05 plainhao joined
jnthn class MyException is Exception {}; die MyException.new("ERROR") 13:07
masak plainhao: 你好!
jnthn rakudo: class MyException is Exception {}; die MyException.new("ERROR")
p6eval rakudo c6a829: OUTPUT«E␤ in main program body at line 11:/tmp/fX0pY0wDN3␤»
masak :P
jnthn masak: wtf, it outputs...the first character?!
masak jnthn: somewhat LTA, no? :)
jnthn I...can totally wait to debug that. :-S 13:08
pmichaud I wonder if something is indexing into a string when it thinks it has an array.
jnthn pmichaud: oh 13:09
pmichaud: That's a good guess.
plainhao 你好 masak
masak guessed right :)
jnthn pmichaud: That'd make it less weird.
Apart from, what is doing that. :-)
plainhao masak, i had to google that, i’m an illiterate fugee 13:10
cognominal jnthn, I think you could close perl #74110
masak plainhao: oh -- I'm just an illiterate foreigner :P
plainhao haha, it is my name, though
masak plainhao: what brings you to #perl6? 13:11
cognominal I thought I had filed a bug for lack of support for pairs as positionnals in captures, but I can't find it :(
plainhao i’m in phl.pm, currently not active 13:12
they’re active, i’m not
jnthn cognominal: You did, I'm sure 13:13
13:13 metaperl joined
masak rakudo: class MyException is Exception {}; throw MyException.new("你好 ERROR") 13:13
p6eval rakudo c6a829: OUTPUT«Could not find sub &throw␤ in main program body at line 11:/tmp/PHAO3YGVkv␤»
masak rakudo: class MyException is Exception {}; 爹 MyException.new("你好 ERROR")
er.
p6eval rakudo c6a829: OUTPUT«Could not find sub &爹␤ in main program body at line 11:/tmp/054uY96Fp0␤»
masak rakudo: class MyException is Exception {}; die MyException.new("你好 ERROR")
p6eval rakudo c6a829: OUTPUT«你␤ in main program body at line 11:/tmp/YXZO3VoMVX␤»
jnthn masak: You're not good at dying, are you? :-)
masak jnthn: third time's a charm.
pmichaud masak's die seems to fail
we should warn him about that :) 13:14
pmichaud throws out puns as quickly as he can
jnthn hands masak more dies to try
masak dies irae, maybe.
pmichaud if you have multiple dies, do we call that "dice"? ;-)
plainhao is anyone using psgi or plack with rakudo? 13:15
masak plainhao: there are plans to port those to Perl 6/Rakudo, but nothing concrete yet.
plainhao nice, looking forward to it 13:16
masak plainhao: for now, there's Web.pm, in an early stage of development.
plainhao thanks for the info, masak
are anyone here on forrst? i’ll follow you
*is
jnthn pmichaud: Hmmm...throwing dice ;-) 13:17
colomon huh. did I somehow break my local proto? :\ 13:20
masak or is it just broken in general? :/ 13:21
colomon I was trying to fix it....
13:22 gabiruh joined
masak does a 'try { ... }' block have any semantics over a '{ ... }' block? 13:24
13:24 JimmyZ joined
moritz_ an implicit CATCH { when True } ? 13:25
masak 卓明亮! \o/
moritz_: ...which does what?
JimmyZ 麦高！ 晚上好
masak just absorbs all exceptions?
moritz_ right 13:26
dunno if that's really spec
masak JimmyZ: 晚上好, 朋友.
JimmyZ 呵呵
masak moritz_: not sure that's desired behaviour.
mathw regrets his linguistic non-awesomeness
JimmyZ BTW，rakudo builds failed on windows XP here 13:27
jnthn moritz_: I think it's spec.
JimmyZ: If it's the ops thing, just run make again.
colomon huh. now its working again. 13:28
moritz_ heisenproto?
JimmyZ nope, it's another thing 13:29
masak JimmyZ: if you nopaste, I can submit a rakudobug for you.
JimmyZ compile core.pm failed. 13:30
colomon moritz_: I did trivially change two unrelated lines...
So, is there a portable way to create projects.state if it doesn't exist?
JimmyZ masak: I will nopsate it tomorrow
masak JimmyZ: 好了
colomon presuming shelling to touch not portable...
oooo, [particle] is in the room... 13:33
colomon is a bit freaked to have other irc-perl-people around in real life...
plainhao you at yapc, colomon? 13:34
colomon yes. 13:35
13:36 wrac left
colomon note to the world: if you're trying to make sure a file exists, checking to see if it exists and overwriting it if it *does* is a bad plan. 13:36
13:36 gbacon joined
colomon if $filename !~~ :e { say "Hello, $filename!"; my $fh = open($filename, :w); $fh.close; } 13:36
masak, moritz_, does that look reasonable to you guys? 13:37
(debug say now removed)
seems to work, so pushed. ;) 13:40
so, I installed Math-Model, and proto totally didn't install the deps. 13:42
:\
masak colomon: I'm lacking context to your question. 13:43
colomon but all my changes from yesterday are now pushed
masak what file?
colomon projects.state
but in general, that block of code should work for any file, no?
13:43 kensanata left
colomon going to go looking for the Perl6 room in a moment. 13:44
masak colomon: it's syntactically correct, if that's what you're asking.
colomon does it make sense? 13:45
I mean, I don't think run "touch $filename" is portable...
It seems to work on OS X
masak I don't see the connection between 'touch' and your code.
colomon my $fh = open($filename, :w); $fh.close; 13:46
masak oh.
colomon that's the same as touch on an empty file.
or so I hope.
masak I guess.
it should be, yes.
jnthn I forget if :a creates a file if it's not there or expects it to already exist 13:47
13:47 FardadJalili left
jnthn If the first, then you can save the :e check 13:47
colomon gotta move, back soon
13:48 shade_ is now known as \shade\
arnsholt jnthn: fopen(3) says both "a+" and "a" creates the file if it doesn't exist 13:48
So I'd guess that :a should create it 13:49
masak is :a implemented, though?
arnsholt That's an entirely different question =)
masak seems it is \o/
jnthn rakudo: say &open.signature 13:50
p6eval rakudo c6a829: OUTPUT«Signature()<0x3b19fa0>␤»
masak I tried locally, and it worked.
jnthn rakudo: say &open.signature.perl
p6eval rakudo c6a829: OUTPUT«:(Any $filename, Any :r($r), Any :w($w), Any :a($a))␤»
jnthn \o/
rakudo++
13:51 colomon left 13:53 wallberg joined
masak wallberg: \o 13:54
wallberg hi :-)
13:56 colomon joined
colomon back 13:56
and now at the Parrot/Perl 6 BOF 13:58
masak: so, how soon will pls be up?
should I be mucking with further proto debugging? 13:59
13:59 patspam joined
masak colomon: I'm one proof-of-concept away from showing how to Actually Install things. 13:59
13:59 FardadJalili joined
masak colomon: but my tuits are tied up in strange and unpredictable ways. 13:59
pmichaud is there anyone at the Parrot/P6 BOF? 14:00
14:00 am0c joined
colomon yes 14:01
me, Coke, and another
bbkr rakudo: say "1"~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 14:03
p6eval rakudo c6a829: OUTPUT«1␤»
masak bbkr: parses as "1" ~~ ~(~(~(1))) -- but with more prefix:<~> 14:04
moritz_ std: say "1"~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1
p6eval std 31400: OUTPUT«Use of uninitialized value $bad in concatenation (.) or string at STD.pm line 62712.␤===SORRY!===␤Preceding context expects a term, but found infix instead at /tmp/Q3ZxIjnFZG line 1:␤------> say
.."1"~~~~⏏~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~␤…
masak stdbug!
[Coke] is in the yapc BOF room.
bbkr masak: see, STD not parses it 14:05
masak it should, IMO.
jnthn I'm surprised STD doesn't parse it too.
It's evil but...
:-)
bbkr however IMO it should be parsed as you said: ~~ ~(~(...))
report?
reported 14:07
colomon "Math-Model has no dependencies. " :\
bbkr++
masak bbkr: it's not a rakudobug, but fine :)
colomon it's a STDbug!
moritz_ it has some depencies 14:08
colomon gotta keep TimToady honest. ;)
14:09 mmcleric left
jnthn reported as...rakudobug? :/ 14:10
We have enough actual ones.
:-)
colomon I suspect we have a somewhat lazy infinite list of actual ones. ;) 14:11
jnthn True ;-) 14:12
14:14 clintongormley joined
pmichaud (from schwern++ 's "How to write bug reports" talk) Why does "report perl 6 bug" on Google not go anywhere useful? 14:14
(and is there anything we could do about it?) 14:15
[Coke] because no one has any good links? add one to rakudo. =-)
pmichaud "rakudo bugs" goes somewhere useful
"report rakudo bugs" doesn't
right now "perl 6 bug" goes to camelia 14:16
jnthn lol
[Coke] is rakudo.org in git or a cms, or...? 14:17
pmichaud cms
and the page is misnamed, imo
right now it says "developer's guide" but describes how to submit a bug
we should probably rename it to "how to report a rakudo / perl 6 bug"
jnthn fwiw, "perl 6 bug" for me when googled has rt.perl.org as the top hit 14:18
pmichaud for me it's www.perl6.org
oh, wait
I see it
yes, RT is the top thing
jnthn the rt.perl.org page is not exactly helpful though.
doesn't mention rakudobug at all 14:19
[Coke] that we can ping ask and robrt on.
14:19 am0c left
jnthn schwern++ for noticing, anyways. 14:20
pmichaud but if I were looking to submit a bug, the RT hit from Google isn't immediately obvious (i.e., I didnt even see it). That might be a good thing though, since RT *isn't* the interface to use to submit a bug report
jnthn Trouble is that "I want to report a bug in Perl 6" is...a tad unspecific. :-)
pmichaud schwern++ is talking about bug tracking in general, I just said "gee, I wonder how people would find our bug tracker"
colomon it's a good thought. 14:21
pmichaud it's okay if it's unspecific, that's kinda the point
jnthn We could do with a page that ranks well for "report Perl 6 bug" that explains things.
pmichaud right
we should probably switch the page on rakudo.org to rank better for "Perl 6 bug"
jnthn That also.
pmichaud (and rt.perl.org isn't solely for rakudo bugs :) 14:22
bbkr I miss also short info how to commit tests to pugs repo. there is no PUGS repo URL on Rakudo's "Developer's Guide", and no command line example how to run test with fudging
moritz_ that example is in the README only
pmichaud bbkr: could you draft a document (or a patch to an existing document)? That could probably go on perl6.org, too.
bbkr: (or otherwise suggest where we can improve this) 14:24
moritz_ an overview document for the test suite would be awesome
pmichaud we probably need to recognize that people filing bug reports are often not rakudo developers
(i.e, they're people using rakudo, not people develpoing rakudo)
[Coke] q 14:27
colomon does ack skip files in directories named lib?
[Coke] %^$#@ mini keyboard!
moritz_ colomon: nope
colomon yes, but think how easy it is to carry around your... errr... is netbook the proper term?
moritz_ colomon: but unless -a is given, only files with known extensions are searched 14:28
colomon ah, so the .pm6 is the problem.
moritz_ aye
need to convince alester to include .pm6 for the next release 14:29
14:29 macdaddy joined, skids left
pmichaud submits rakudobug about rakudobugs 14:29
masak .oO( [meta #72593] )
bbkr pmichaud: i'm not familiar with maintainig test suite. but my idea for improvment is simple: there can be a switch '--get-pugs-tests' that would checkout test suite from Pugs SVN in place of those delivered with monthly build. this way developers could commit right to rakudo-version/t/spec without manual fetching tests from Pugs repo. 14:30
pmichaud (looking)
(no ticket 72593)
[Coke] (hai, netbook)
masak pmichaud: sorry, didn't mean to send you off looking. just imagined a meta-RT.
pmichaud ahhhh 14:31
masak ours is named 'perl', IIRC.
pmichaud is lost
masak every RT instance has a name. the name of the one on rt.perl.org is 'perl'.
pmichaud yes, I know that
masak :) 14:32
pmichaud I'm not sure what you're getting at
(I'm probably distracted)
moritz_ just forget it :-)
14:32 mmcleric joined
masak yes. never mind. 14:32
masak tweaks his sillyometer a bit
14:33 cygx joined
cygx moritz_: just successfully pushed my changes 14:33
moritz_ cygx++
14:34 kensanata joined, am0c joined
cygx btw: the link to the online version was borked - that should have read www.informatik.uni-frankfurt.de/~cg.../frontend/ 14:35
masak moritz_: no, I think it takes a CATCH block to clear the exception. that block might be empty though -- but a try block seems to have the exact same semantics as a normal block. 14:36
I'm willing to be proven wrong with a spec reference, though.
colomon hey, the parrot/perl6 BOF room has a microphone and presentation screen. very nice
so, ack's source is easy enough to hack to add .pm6!
masak moritz_: oh wait. I had not read the Apocalypse far enough. it even mentions it here try { ... } =:= try { ... CATCH { default {} } }. so, never mind. 14:37
14:38 Mowah left
mmcleric is it a known issue that email to [email@hidden.address] can get lost? 14:40
14:40 tadzik joined
mmcleric i've sent patch several hours ago and still don't see it on rt.perl.org 14:40
colomon mmcleric: it's certainly a known issue that it is slow. :) 14:41
what's your patch for?
masak but that's p6c that's slow, right? not RT.
mmcleric Subject: [PATCH] WhateverCode type implemented
colomon masak: errr... you're the expert.
masak usually my RT tickets go in in less than a minute.
colomon mmcleric: you got it already?! 14:42
masak and are visible on rt.perl.org
colomon \o/
mmcleric no
it was just WhateverCode type
colomon ah
mmcleric but it fixes two skipped tests
colomon cool.
masak: you've got RT-fu, can you see his patch in there? 14:43
masak colomon: I'll search for it.
nope. if that's the name of it, it's not in there. 14:44
mmcleric i can simply send pull request
just did it
colomon mmcleric++
jnthn mmcleric: Спасибо, I'll glance over the patch a bit later on. :-) 14:45
mmcleric ok :) back to work 14:46
14:47 macdaddy left
colomon jnthn: it's in his whatever code branch. :) 14:48
colomon would pull it, but is not particularly qualified in that area....
14:48 FardadJalili left
colomon is alone in the Perl 6 BOF room now... 14:49
jnthn colomon: Yeah, I'd like to give it a look over first. 14:50
colomon jnthn: right, I will certainly defer to you on this one. 14:51
ooo, someone just peaked in the window. 14:52
14:52 Woody2143 joined, molaf_ left
colomon okay, enough fruitless mucking about with proto. 14:52
time to destroy 25% of Perl 6's test suite. 14:53
pmichaud since we're currently passing 83% of the suite, if you can destroy only tests that we aren't currently passing then we'll be passing 108% of the suite :)
colomon unfortunately, I plan on only destroying tests we already pass. :( 14:55
pmichaud I think that's called a "regression" :)
colomon oh bother, my script for generating the trig tests gets a Null PMC in master. :( 14:56
14:58 Kodi joined, macdaddy joined
arnsholt Bah. Term, prefix op, infix op; pick two 14:58
Kodi Do you think "2, -4, 6 ...^ -32" should yield "2, -4, 6, -16" or "2, -4, 6, -16, 32"? S03 seems to imply the former, but I'd think the latter would be more useful. 15:00
And by "6", I mean "8". Heh. 15:01
pmichaud iiuc, it should yield the latter, even per spec
perhaps I'm missing something.
15:02 rv2733 joined
masak Kodi: I think it's the former. but ...^ isn't implemented yet. 15:02
pmichaud note that "up to" will mean something different here 15:03
Kodi Well, S03 says that ...^ "excludes the limit if it happens to match exactly". So it depends on how you interpret "exactly".
pmichaud 32 doesn't match exactly, so it'll be in the list.
masak oh.
colomon is mystified by that series
pmichaud and the next value generated will be the one that is beyond the series
masak retracts his statement
colomon oh, 8
pmichaud++ is right 15:04
masak aye.
colomon hopes he can resist being distracted by ...^ long enough to finish his spectest 2.5 x decimation.
Kodi Hm, yes. I mixed up which seemed to specified and which was my opinion. 15:05
masak :) 15:06
colomon I almost implemented ...^ the other day by doing ... and then grepping to filter out the limit value.
but I decided that was kind of evil.
pmichaud colomon: it's actually pretty easy, afaict 15:07
you return an element if it's not beyond the end
otherwise, you return an element if it is the end and if the ^ wasn't specified 15:08
otherwise, the series ends
(with no element)
colomon pmichaud: but that requires a bit of a refactor.
pmichaud I suggest the refactor, then :-)
colomon not saying it's not the right way to do it, just the the grep version was a one-liner.
15:08 TiMBuS left 15:09 alester joined
colomon and the grep version would probably work well enough that no one was likely to notice until I got around to doing the refactor. ;) 15:09
literal does code like this compile in Perl 6? paste.scsys.co.uk/45057 15:11
pmichaud as long as it doesn't eventually lead down a blind development trail (i.e., encouraging bad code practice), I'm okay with anything that works
literal I had always thought thought the $f would be local to the first if condition in Perl 5, but I was wrong
pmichaud in Perl 6, the my is local to the curlies (which are sometimes implied)
15:12 Kodi left
pmichaud in this case, the scope of $f is from the if statement to the end of the code 15:12
masak literal: it works in Perl 6.
what pmichaud said.
literal so the elsif ($f) would not compile?
masak if you want to make $f local to the if block, use -> $f
literal: here it would.
literal ok
pmichaud to the end of the code, not the end of the if block (or if statement) 15:13
masak literal: the variable ends up belonging to its innermost surrounding block. which in this case is the whole program.
pmichaud what masak++ said. :)
literal well, not in Perl 5
$f is not accessible after these conditionals 15:14
pmichaud in Perl 6 we try to not have magic boundaries for declared symbols
masak Perl 6 has fewer exceptions of that kind.
pugssvn r31401 | colomon++ | [t/spec] Get the trig test generation script working under current Rakudo. 15:16
colomon hey, very easy to get the trig test generator working again. Rakudo++
pmichaud so, since I still have 100 minutes to go.... anything in particular people want to see from my NQP talk? 15:17
PerlJam pmichaud: live streaming or video online ;) 15:18
jnthn ;)
pmichaud I could turn on the webcam on my laptop, you'd probably see me come in and out of the video at times :)
I really ought to find a way to capture my presentations locally onto my computer, though. 15:19
colomon I guess I could video skype it, if I can find a power outlet.
pmichaud even if I'm not in the picture, audio capture + video capture while I'm giving the presentation would work out.
PerlJam if you had another webcam, you could capture audience response too
pmichaud I do have another webcam, but left it at home.
colomon though that would definitely cut down on my ability to hack during it. ;)
pmichaud I'm thinking I might include my additional webcam in my travel supplies. There have been two times now where it might've been useful. 15:20
15:23 ash__ joined
pmichaud and since conferences have been historically unreliable about posting videos of my talks, maybe I should just treat it as a live screencast and record it locally myself. 15:23
colomon someone just peeked in the door's window again
PerlJam pmichaud++
now convince everyone else to do that too (or make it easy for them to do so) 15:24
pmichaud otoh, perhaps there's a good reason the conferences haven't posted my talks :-)
PerlJam especially at a YAPC
pmichaud yeah, I may work on that a bit tonight. It would be very useful for my Rakudo talk tomorrow
colomon ah, but they're here to see the pictures on the wall.
ash__ i have been hearing a lot of buzz about perl6 recently, keep up the good work guys 15:27
colomon rakudo: say ((1, 2, 3) xx *).munch(10)
p6eval rakudo c6a829: OUTPUT«123123123123123123123123123123␤»
colomon rakudo: say ((1, 2, 3) xx *).flat.munch(10) 15:28
p6eval rakudo c6a829: OUTPUT«1231231231␤»
colomon oooh, groovy
thanks, ash__
rakudo: my $a = ((1, 2, 3) xx *).flat; say $a.munch(10)
p6eval rakudo c6a829: OUTPUT«1231231231␤»
colomon rakudo: my @a = 1, 2, 3; my $a = (@a xx *).flat; say $a.munch(10) 15:29
p6eval rakudo c6a829: OUTPUT«1231231231␤»
15:31 patspam left 15:33 Ross joined 15:35 JimmyZ left
pmichaud hmmm, I might have something that will work. I'll run it when I give my presentation and see how it goes. 15:35
rakudo: (1..28).pick 15:37
p6eval rakudo c6a829: ( no output )
pmichaud rakudo: say (1..28).pick
p6eval rakudo c6a829: OUTPUT«21␤»
15:37 snearch joined
m6locks rakudo: say (1..28).pick 15:39
p6eval rakudo c6a829: OUTPUT«8␤»
ash__ rakudo: say ((1, 2, 3) xx *).pick; 15:42
15:42 wallberg left
p6eval rakudo c6a829: ( no output ) 15:42
ash__ rakudo: say ((1, 2, 3) xx *).munch(10).pick 15:43
p6eval rakudo c6a829: OUTPUT«1␤»
15:43 JimmyZ joined
colomon hmm, so pick flattens. 15:43
ash__ does .pick flatten?
ah, thought so, just wondering
colomon rakudo: say ((1, 2, 3) xx *).munch(10)[0] 15:44
p6eval rakudo c6a829: OUTPUT«1␤»
colomon rakudo: say ((1, 2, 3) xx *).munch(10)
p6eval rakudo c6a829: OUTPUT«123123123123123123123123123123␤»
JimmyZ rakudo: say (1..*).pick
ash__ rakudo: say ((1, 2), (3, 4)).pick
p6eval rakudo c6a829: ( no output )
rakudo c6a829: OUTPUT«2␤»
colomon JimmyZ: (1..*).pick definitely does not work.
15:45 justatheory joined
JimmyZ Will it work in the further? 15:45
colomon JimmyZ: doubt it greatly 15:46
ash__ going to infinity is hard
colomon what would it even mean?
ash__ pick a number between 1 and infinity?
colomon ash__: yes, but how? you can't make the list of things to pick from so you get an even distribution. 15:47
and the average result is larger than your computer's main memory.
ash__ yeah, i don't think it's possible 15:48
is there a way to check the upper bound of a range?
maybe .pick (and some other functions too?) should check the upper bound of the range to see if its valid
and if its not, throw a runtime error
saying its out of range or out of bounds or is infinity, or something? 15:49
arnsholt NQP-rx doesn't support look-behind yet?
colomon ash__: that would require making a special case of all those methods for Range. 15:50
15:50 ashleydev left
colomon right now, pick doesn't know or care whether it's getting 1..*, 1...*, or some array you've been carefully seeding. 15:50
15:51 jaffa4 joined, tadzik left
ash__ colomon: is that bad though? it could help prevent things like (1..*).pick; since if you do that currently it just sits in an infinite loop and eventually runs out of memory 15:52
i'd rather have stuff in the core libraries throw an exception than go into infinite loops, but maybe thats me personally
15:52 am0c left
masak JimmyZ: it's not possible to randomly pick a natural number, giving all the natural numbers equal probability. 15:53
JimmyZ: loosely put, the objection can always be "that's a *really* small number, why didn't you pick a bigger one?"
colomon ash__: I agree about the exceptions, but the problem is that in general, it's very difficult to know when a loop is infinite. Range is just a special case where it's obvious, and it's the only such case I can think of at the moment. 15:55
15:56 JimmyZ left
colomon is now dreaming of a .finite-list method that throws an exception if it knows it is infinite.... 15:57
masak colomon: just check whether .elems == Inf :P 15:58
PerlJam colomon: or doesn't know that it's finite? (i.e. it's finite-ness is indeterminant)
15:58 sdo left
colomon PerlJam: probably the majority of non-array lists won't know if they are finite. 15:59
16:00 skangas` joined
colomon should have been thinking about lunch rather than trig hacking.... 16:00
16:01 skangas` left
ash__ rakudo: my $a = method { fail 'Bad Max Range' if @.max == Inf; my @l = @.list.Seq; @l[floor(@l.elems.rand)]; }; say (1..*).$a(); 16:02
p6eval rakudo c6a829: ( no output )
ash__ hm, well, that works locally
its a relatively minor change to .pick (the one with no arguments)
16:04 tadzik joined, colomon left 16:05 skids joined 16:06 masak left 16:11 meppl joined
PerlJam Two logos from sri: kraih.com/perl5-camelia-sri.png kraih.com/perl6-camelia-sri.png 16:19
melding p5 with p6 in the logo department. 16:20
16:22 Mowah joined
ash__ PerlJam neat logos 16:26
PerlJam yeah, I thought so too 16:27
(they're sebastian's, not mine)
ash__ still neat, wasn't perl5's logo the camel? 16:28
Su-Shee *shiver* yes. ;) or onions.
_sri the camel is oreillys logo 16:29
avar Perl 5 never really had a logo, just the camel book logo
PerlJam Padre is already using a pretty blue butterfly. We might be able to switch "Perl's logo" to that easily
(to *some* sort of butterfly anyway) 16:30
16:30 colomon joined
ash__ why was it an onion? 16:30
colomon now has a Rakudo * t-shirt!
PerlJam ash__: Because ogres are like onions :) 16:31
ash__ lol
they have many layers?
tadzik I thought networks are like onions
ash__ i like cakes better, they have layers, and frosting
i think PerlJam is quoting Shrek 16:32
tadzik yeah, I know that
16:32 ashleydev joined
ash__ there are a number of � in the perl site, i wonder if anyone else ever notices them 16:33
diakopter I do
16:34 ashleydev_ joined
PerlJam ash__: where? 16:34
16:34 diakopter sets mode: +oooo colomon ash__ avar Su-Shee, diakopter sets mode: +o p6eval
ash__ www.perlfoundation.org/perl_trademark the footer on that page for one 16:34
PerlJam ah, indeed 16:35
diakopter both firefox & Chrome on windows have the ?
PerlJam I'd never noticed before
16:36 ashleydev_ left 16:37 ashleydev left, ashleydev joined, ashleydev left 16:38 ashleydev joined 16:41 envi^home left 16:42 SmokeMachine joined
colomon is in the room with pmichaud, but is letting him prep for his talk rather than introducing himself. 16:47
16:47 b_jonas joined 16:48 b_jonas left 16:53 azert0x joined, azert0x left 16:54 azert0x joined
colomon the daisy chain of power strips in front of me probably breaks some fire codes.... 16:56
ash__ colomon: ssshh, its not an issue if nothing happens :P 16:57
16:58 dakkar left
colomon but I'm closest to the wall! if it goes up in flames, I'm going to have to vault over tables to escape! 17:01
the talk is starting... 17:02
Su-Shee PerlJam: you're texan, right? 17:03
PerlJam Su-Shee: yes-ish 17:05
Su-Shee PerlJam: how much heat and spice is considered a good level in chili con carne? as hot as you can or more balanced? 17:06
17:07 cdarroch joined, cdarroch left, cdarroch joined, thebird left
PerlJam Su-Shee: I can only speak for myself. I like it just spicy enough to warm you up, but not so spicy that it has you reaching for something cold to drink every few minutes. 17:07
That's how I like my cajun food too. 17:08
17:08 patspam joined
PerlJam (when I eat spicy cajun food) 17:08
Su-Shee cooked the first chili con carne (after research) yesterday and I'm not sure about the hotness.
PerlJam: is that above or below a good thai hotness level? ;) 17:09
PerlJam Su-Shee: slightly below :) 17:10
Su-Shee: what did you put in it?
Su-Shee PerlJam: I followed roughly this recipe: tasteinvader.com/2010/02/01/texas-five-chili/
PerlJam: but I took fresh tomatoes. 17:11
17:11 meteorjay left
lue ohai 17:11
PerlJam heh. I like Rule 3. That's hotly debated in Texas
Su-Shee: you didn't roast the tomatoes? 17:12
Su-Shee PerlJam: I usally made it with beans, fresh peppers, fresh corn and ground meat, but I've wanted to try this variation
PerlJam: no, they were italian overly ripe tomatoes, perfectly for cooking. 17:13
the beef was from a german cow. ;)
PerlJam Su-Shee: If yours looked anything like those pictures it must have been good whether hot or not-so-hot 17:15
colomon this is a great talk so far!
Su-Shee PerlJam: it was very good - but what can go wrong if you're slowly cooking nice things in the oven? :)
PerlJam indeed
Su-Shee colomon: anything you don't know yet or plenty of fans cheering :) 17:16
PerlJam: I was very pleasantly surprised by the taste of the different chilis I got.
17:17 meteorjay joined, adey joined
PerlJam wonders if he can convince his wife to make this dish for supper tonight 17:18
colomon Su-Shee: current NQP is more powerful than I knew.
also, pmichaud is a great speaker, which I had somehow not expected
sorear hello #perl6 17:19
phenny sorear: 10:26Z <masak> tell sorear -- still not 100% sure what you're asking. perhaps best that you follow S26 to the letter until I get around to updating it to the latest discussions. in the best case, you'll also become convinced that the current way isn't optimal. :)
PerlJam colomon: what did you expect?
kensanata All this talk about chili is making me hungry again and I just ate Marillenknödel.
colomon PerlJam: kind of more academic, I guess.
ash__ colomon: pmichaud was a professor, wasn't he? I think he probably did a bit of lecturing in his time
colomon ash__: he's probably better than 80% of the professors I had. 17:20
PerlJam colomon: ah, so the knowledge that he was a professor colored your imagination in a negative way
colomon PerlJam: yes.
ash__ colomon: true, i have only 2 or 3 professors that can really put together a good presentation, the rest are average
17:20 Trashlord left, adey left 17:21 meteorjay left, meteorjay joined, jaffa4 left
lue I don't hear anything [ then again, I'm not actually there :) ] 17:21
colomon also, I had no idea protoregexes were developed that late in the p6 process. (2008) 17:22
Su-Shee protoregexes veloceraptor. 17:23
kensanata: are you the emacs-wiki guy?
17:23 dual left 17:24 rv2733 left, hejki left 17:25 huf joined, hejki joined, dual joined
colomon also have just learned I want to avoid hacking NQP if possible. ;) 17:26
Su-Shee colomon: if that really was the intension of the talk.. :)
PerlJam sometimes hacking NQP is way easier than hacking Rakudo
*and* hacking NQP might just make whatever it is available to Rakudo :) 17:27
colomon Su-Shee: now I know more about programming *in* NQP, and want to avoid changing NQP itself. :)
Su-Shee knows nothing about nqp. 17:28
PerlJam Su-Shee: sure you do! You know it's "not quite perl" :)
Su-Shee well besides that. :)
I totally lost track what is what. I remember pugs is haskell, rakudo is parrot, parrot is beneath, diakopter made sprixel which is now c#.. ;) 17:29
arnsholt NQP is pretty much a convenience interface for Parrot's assembly language 17:30
(And oh, how it's more convenient than PIR =) 17:31
Su-Shee I looked into pugs a week ago and besides that I don't understand anything, it looked very readable. 17:32
arnsholt It's Haskell, I'm not sure there's more than a handful of people understand it =) 17:34
17:35 cygx left
PerlJam I don't know ... Perl 6 seems to borrow many haskellian ideas, so if you grok Perl 6, you can grok Haskell modulo the syntactic weirdness. 17:36
Su-Shee I just looked into the code; I liked pugs. 17:37
arnsholt PerlJam: Perl (both of 'em) have a lot of Lisp in them as well 17:43
Just with nicer syntax =D
Su-Shee actually I found haskell quite nice to look at. 17:44
PerlJam Haskell is certainly nicer to look at than scheme or lisp :)
Su-Shee well those are at least easy to type :) 17:45
PerlJam I'll probably get the quote wrong, but Larry once said something like "Lisp has all of the visual appeal of oatmeal with fingernail clippings" and that about sums it up :-) 17:47
ash__ do other multi-dispatch languages (like lisp) have similar dispatch mechanisms as perl6's?
cognominal Also, with may national keyboards you need to shift your way to parentheses
*many 17:49
ash__ oh god... github.com/richhickey/clojure/blob/....java#L162 to line 300 ish, i hate when i find that in code 17:50
PerlJam wow 17:51
17:51 patspam left
ash__ ruby does that too 17:51
but they only do 15 params
not 20
PerlJam that's something for thedailywtf.com though 17:52
17:52 ashleydev left
ash__ i can understand why they do it, but at the same time, you'd think they could use a stack to push params between functions 17:53
(like parrot does)
17:53 ashleydev joined, colomon left
arnsholt ash__: Well, if they want to support Java 1.4 (IIRC) they have to do it that way 17:54
17:54 ash__ left
arnsholt Varargs are 1.5 I think 17:54
17:54 ash__ joined 17:55 patspam joined
ash__ arnsholt: but you could have an object that holds the params, like an array instead of doing it that way 17:57
if its a speed thing, you could still use the array for >20 items, and have the hand crafted versions for < 20 17:58
arnsholt That's true =) 17:59
ash__ granted, if you have that many parameters you might want to check if the function is still doing something useful, but still... 18:01
18:03 colomon joined
colomon back in BOF room 18:03
awwaiid will the REPL in Rakudo* keep lexicals accross lines? 18:10
18:11 ash__ left 18:12 Ross left
colomon awwaiid: someday it will 18:12
hopefully someday soon. 18:13
18:13 Ross joined
awwaiid lovely! 18:14
colomon I guess what I should say is that it is a goal for R* to have better lexicals in the REPL. 18:19
18:20 ShaneC joined
sorear lexicals in the REPL are very much within reach 18:20
NQP-rx has them
Tene re bug reporting, whiteknight (iirc) had an awkward time trying to figure out how to submit a patch to rakudo recently; he did find rt, but gave up on figuring out the RT interface to submit a ticket. 18:34
Tene still backlogging
lue Last I checked, you can only email the bug. 18:35
there is no web interface.
Tene masonkramer: try {} has a default CATCH { $!.handled = True } oslt, but ONLY if the block passed to try has zero other handlers already in it. 18:36
oh, ENOMASAK
backlogging is weird. :)
18:37 dual left 18:39 kensanata left 18:42 patspam left 18:47 ejs joined 18:51 patspam joined
colomon My trig test thinning war is getting a bit crazy. 18:55
I may end up with the same number of tests but twice as many different things tested. 18:56
18:58 ashleydev left, mmcleric left
Tene And I see he figured it out anyway. 18:59
18:59 mmcleric joined 19:01 ashleydev joined 19:04 cono left
pmichaud well, my screen capture of my talk seems to have worked great, but audio capture looks like it was a fail 19:05
19:05 Mowah left 19:06 cono joined
colomon pmichaud: with any luck the conference's video's audio will be better. 19:07
pmichaud colomon: agreed, but my experience at yapc's past is that the videos never get posted.
colomon we'll just need to cleverly liberate it from them, then! ;) 19:08
TimToady \o/ my new power supply worketh!
pmichaud \o/
19:08 SmokeMachine left
colomon \o/ 19:08
crap, got so busy hacking on trig tests I forgot that chromatic's talk was beginning soon. :( 19:09
19:10 patrickas joined
colomon that's as in, ten minutes ago. 19:11
19:14 ash__ joined, mmcleric left
_sri TimToady: please take a look at my interpretation of camelia :) kraih.com/perl6-camelia-sri.png 19:15
colomon lovely!
19:15 ispeak_ joined
_sri there's also a matching perl 5 one (transitioning to a 6), kraih.com/perl5-camelia-sri.png 19:16
19:16 ispeak_ left
pugssvn r31402 | colomon++ | [t/spec] Partial update to trig test generating code to be smarter. 19:16
19:17 ShaneC left
_sri TimToady: just poke me if you like them, i'll finish and release them under an open license then 19:19
19:20 rokoteko left
pmichaud _sri: just a note that we don't have a "Rakudo 1.0" in the works. 19:27
_sri pmichaud: that part was just an example, i also have a * variant :) 19:28
colomon TimToady, pmichaud: where are you guys? I should probably come be social rather than continuing to hack away at trig tests.
pmichaud I'm in chromatic++'s talk, back row.
TimToady ~~ s/back/front/ 19:29
Su-Shee I'm getting a rock star vibe. ;)
TimToady _sri: I've seen much worse. :)
colomon man, all the cool kids are in chromatic's talk! 19:30
patrickas TimToady: I hear _sri++'s camelia even has stylized P and 6 on her wings !
19:31 plainhao left
colomon I just need to add two more types to the trig tests... 19:31
Su-Shee colomon: once geek, always geek? ;) 19:32
colomon Su-Shee: these sorts of things really bring out my inner introvert.
_sri kraih.com/tpf-camelia-sri.png # draft 19:33
Su-Shee "inner introvert". I'll have to remember that. :)
colomon Put me in the right situation and I'm an extravert. 19:34
I usually do fine at traditional Celtic music gatherings, for example. 19:35
*extrovert ?
ash__ "bring out my inner introvert" is somewhat ironic
_sri++ nice logos 19:36
colomon hey, I'm having a one-person hackathon here.
_sri \o/
19:36 snearch left
ash__ colomon: thats not a bad thing, i am hacking on parrot now 19:37
TimToady I was up there earlier today, and no one was there, but I'm not allowed to bring out my inner introvert at yapc 19:38
Su-Shee TimToady: wouldn't you bring your inner introvert in and not out? 19:39
lue is currently typing a "book" for people who wish to get their tenses and such right when time travelling. 19:40
lue also thought it was a good idea to start by describing how the temporal bits of language work normally. 19:41
pmichaud I've yet to actually make it to the room. :) 19:42
19:44 patspam left
TimToady coffee & 19:44
19:45 colomon left 20:03 cygx joined 20:04 foodoo left
lue afk 20:04
20:05 Psyche^ joined 20:06 Patterner left, Psyche^ is now known as Patterner
cygx I've got a question about the Rakudo-part of the design of try.rakudo.org (see moritz.faui2k3.org/tmp/tryrakudo.svg ) 20:11
is there a way for Rakudo to tell the Controller that no more output will be produced as long as no new input is provided?
20:23 SmokeMachine joined
ash__ cygx: do you mean a way to tell if there are any tasks running? 20:27
cygx that's not exactly equivalent, but it could still be useful (more) 20:33
it would be nice for the frontend to know that it nedd not poll for new output as long as it doesn't send new input 20:34
*need
this affects the design of the client-server-communication and the uiser interface 20:35
pmichaud btw, I'm giving a talk about Rakudo Star tomorrow at yapc::na. If anyone has a point they want me to be sureto hit (or to avoid?), please let me know :)
jnthn pmichaud: Having done similarish recently, my only suggestion is to be very clear on the difference between compiler release vs distribution release. 20:39
pmichaud jnthn: that's really kinda the whole topic of my talk :)
jnthn pmichaud: Sure, but to instill the difference before digging into what's in the distribution. 20:40
pmichaud I'm also hoping to have stuff in place to record my talk. I tried it with my NQP talk, and managed to capture the video fine but apparently lost the audio. 20:41
(I suspect hardware configuration on my laptop.) I'll see if I can improve things tonight.
jnthn I'll be interested to see/listen to what you say, given I've got an R* talk on Saturday in Brussels. :)
pmichaud obra's keynote this morning was excellent, a good lead-in to some of what I plan to cover in my talk 20:42
20:43 SmokeMachine left
obra heh. "Perl 6 is our kid sister" 20:43
pmichaud I liked that 20:45
ash__ is any of this going to be online? 20:47
moritz_ did anybody review github.com/berekuk/rakudo/commit/e7...ac77315291 yet?
pmichaud reviewing now 20:48
the line defining WhateverCode needs to go into WhateverCode.pm, not Block.pm 20:49
moritz_ agreed
pmichaud Other than that, +1 from me
(i.e., I approve the patch)
jnthn++ can request revert if it doesn't work
(but it looks pretty normalish to me) 20:50
jnthn oh, let me quickly glance
moritz_ it's the first step to get chained Whatever-currying
jnthn (slipped my mine, partly thanks to watching some world cup) 20:51
Tene moritz_: why the extra empty stmts node?
pmichaud normally the first node is reserved for a place to stick parameters and other initializations
jnthn hmm
pmichaud we might not need it for WhateverCode nodes
Tene ah, right 20:52
jnthn I'd have preferred that make_block_from got an extra parameter
That defualts to Block
But lets you pick the wrapper type
pmichaud jnthn: I agree, but I'm thinking that may refactor a bit when I do the closures update anyway.
jnthn Than copy-paste of the code
pmichaud I'm fine with rejecting in favor of that refactor, if we want to do that.
jnthn And yes, separate file
pmichaud (i.e., to add a parameter rather than copy-pasta)
jnthn Well, I think easiest is to suggest to mmcleric++ those changes. :-) 20:53
pmichaud agreed
so let's do that.
20:54 whiteknight joined
moritz_ I'll reply to his pull request and include a link to this discussion 20:55
jnthn phenny: tell mmcleric Looked over your patch. Please add an optional parameter to make_block_from that's like $wrapper_type = 'Block' and then pass 'WhateverCode' in as that parameter, so as to save a little code duplication. Also, would be good to add a WhateverCode.pm rather than putting it into the Block.pm (will just need a little addition in build/Makefile.in - a line below Block.pm would be fine.)
phenny jnthn: I'll pass that on when mmcleric is around.
jnthn phenny: tell mmcleric +1 on overall direction/intent though. :-)
phenny jnthn: I'll pass that on when mmcleric is around.
moritz_ -> bed now
20:56 jrtayloriv joined
jrtayloriv Where would I start looking if I wanted to understand how to access functions from C libraries in Perl 6 code? What are some terms I could Google for? What are the major limitations in Parrot/Rakudo that prevent people from doing this, if any? 21:02
moritz_ jrtayloriv: github.com/jnthn/zavolaj 21:03
(really going to bed now :)
21:03 mmcleric joined
jrtayloriv Night night. 21:03
(and thanks)
21:07 bphillips left 21:08 ejs left 21:11 Su-Shee left, jrtayloriv left 21:12 tadzik left 21:19 cygx left, kensanata joined 21:23 xinming_ joined 21:24 patrickas left 21:26 xinming left 21:29 clintongormley left 21:33 Guest23195 left 21:35 ashleydev_ joined, ashleydev left, ashleydev_ is now known as ashleydev 21:44 jrtayloriv joined
mmcleric jnthn: i just sent one more pull request with updated patch 21:47
phenny mmcleric: 20:55Z <jnthn> tell mmcleric Looked over your patch. Please add an optional parameter to make_block_from that's like $wrapper_type = 'Block' and then pass 'WhateverCode' in as that parameter, so as to save a little code duplication. Also, would be good to add a WhateverCode.pm rather than putting it into the Block.pm (will just need a little addition in build/Makefile.in - a line below Block.pm would be fine.)
mmcleric: 20:55Z <jnthn> tell mmcleric +1 on overall direction/intent though. :-)
21:47 donaldh joined, donaldh left
mmcleric off to bed now... 24 hours are suddenly not enough :) 21:48
21:50 mmcleric left 21:54 donaldh joined, donaldh left, donaldh joined 21:55 donaldh left 22:01 Transformer joined, betterworld left, betterworld joined 22:02 Ross left 22:03 jrtayloriv left, Transformer left
dalek kudo: 00efa15 | (Vyacheslav Matjukhin)++ | (3 files):
WhateverCode type implemented
 22:06
22:09 tedv joined
jnthn sleep, EEARLYMEETING 22:11
22:12 dual joined 22:13 gbacon left, kthakore joined
kthakore hola 22:13
sorear Hello. 22:14
kthakore I have a perliminary version of parrotSDL working with parrot 2.5
22:14 eternaleye_ left
kthakore how do I get rakudo to see it? 22:14
github.com/kthakore/parrotSDL
22:15 Trashlord joined
kthakore hrm ... 22:16
sorear: any clue?
oh welll 22:17
ash__ rakudo: Q:PIR { say "PIR works in here" }
p6eval rakudo c6a829: OUTPUT«PIR works in here␤»
kthakore ash__: ??
ash__ in rakudo, if you do Q:PIR you can drop down to pir code directly, so you can load/unload any pir stuff you want, or call other PIR functions or w/e 22:18
22:18 colomon joined
kthakore ash__: but then so perl6 will have to have a wrapper ? 22:18
Tene I had "use SDL:from<parrot>;" working in the past. 22:19
kthakore Tene: then SDL::init() will work?
colomon rakudo bof tonight @ yapc question
Tene All you had to do on the parrot side was use the 'parrot' compiler to set up the exports, and that only if you didn't want to export everything in the SDL namespace.
But that's been long since broken; I gather there's something working again these days, but I don't really know what it is. 22:20
You might look to see what zavaloj does, maybe.
sorear phenny: tell cygx to check out search.cpan.org/perldoc?IO::Pty::HalfDuplex 22:21
phenny sorear: I'll pass that on when cygx is around.
kthakore Tene: what and where does zavaloj do?
Tene or ask sorear, iirc
zavaloj is what's supporting the perl5 interop
and I think I misremember the name. :)
sorear Rakudo doesn't currently support PDD21 HLL interop
Tene zavolaj
sorear blizkost 22:22
Blizkost uses the PDD31 interface, which is different
pugssvn r31403 | mmcleric++ | remove skip marker from first WhateverCode tests
Tene I'm still grumpy about hll interop being broken repeatedly, after every time I tried to fix it. I need to get over that. 22:23
sorear I'll try and get PDD21 going again tonight 22:24
kthakore is totally lost 22:25
there really shouldn't be so many interfaces to go through to plug a c library in to perl6
this is ridiculous 22:26
*sigh* oh well
Tene: trhanks I will try it 22:29
gtg home
22:30 micahjohnston joined 22:33 synth joined
ash__ hmmm i know what i could try working on (later?), something like ruby's RubyInline module for perl6, to facilitate C code in perl6 22:35
22:36 micahjohnston left
Tene kthakore: You are absolutely right. There's infrastructure in place and work (slowly) in progress to fix it, but it hasn't been a priority for anyone involved with the project, for the most part. 22:36
ash__ i think everyones focused on getting the core of rakudo working right, then they will worry about other features, like interfacing other languages 22:37
Tene I've been a little less burned out lately. I guess I'll try to look at interop again tonight. 22:39
22:39 cdarroch left, Limbic_Region joined 22:43 colomon_ joined 22:47 colomon left
kthakore Tene: what is this interface that is coming in place 22:48
22:48 ab5tract joined 22:51 synth left, synth joined, BinGOs left, BinGOs joined
ash__ kthakore: perlcabal.org/syn/S21.html is the current purposed way of doing it 22:52
kthakore ash__: thx
22:52 eternaleye joined
ash__ parrot would probably look like what Tene wrote up above, "use SDL:from<Parrot>" 22:53
kthakore ash__: is any one doing the fromc stuff? 22:55
Tene rakudo: use Curses:from<parrot>; 22:57
p6eval rakudo 00efa1: OUTPUT«===SORRY!===␤Parent isn't a Class.␤»
Tene So... it's at least doing *something* with that. 22:58
Sorry, I can't investigate further until I get home from work. That'll be about another two hours.
kthakore no worries 22:59
23:02 kensanata left
kthakore Tene: I think it needs .namespace 23:02
hmm my perl6 compile is stuck at perl6/Grmar 23:03
src/Perl6/Grammar.pm 23:04
it is stuck there
Tene kthakore: you think that what needs .namespace? 23:05
Curses.pir does have .namespace ['Curses']
23:06 lestrrat is now known as lest_away
colomon_ [Coke]: ping? 23:08
23:09 \shade\ left, shade_ joined
kthakore Tene: i dunno 23:10
sorear Everyone stop. All that's needed is to implement PDD-21 importation in Perl6::Module::Loader near line 90 23:11
Parrot has two different and incompatible cross-HLL import APIs 23:12
Rakudo only implements one of them at the moment
Tene Which is the other, these days?
sorear PDD21 is the unimplemented one 23:13
Tene Right, and which is the other that rakudo is implementing?
sorear PDD31
Tene sorry, AFK now, work emergency. 23:15
23:29 mj41_ joined, masonkramer left, PacoLinux left, mberends left, jaldhar left, p6eval left, ajs left, sykes_ left, solarion left, kcwu left, mj41 left, cls_bsd left, c9s left, kloeri left, mathw left, obra left, IRSeekBot left, mj41_ is now known as mj41, ajs joined 23:30 jaldhar joined 23:31 sykes_ joined, cls_bsd joined 23:32 kloeri joined, solarion joined
kthakore sorear: what? 23:33
23:33 mathw joined 23:36 p6eval joined
kthakore jeezus 23:36
how long dose src/Perl6/Grammar.pm take to compile?
arnsholt A while =) 23:38
23:38 masonkramer joined 23:39 rgrau left
kthakore arnsholt: it is like it is compiling a language or something 23:40
:p
23:43 mberends joined
sorear also, NQP is really slow 23:43
Parrot in general
23:44 dju left 23:49 IRSeekBot joined, PacoLinux joined 23:54 solarion left, solarion joined 23:55 colomon_ left 23:59 skids left

Powered by Cro and the Raku® Programming Language.

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

Thank you!