»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! | tinyurl.com/p6contest
Set by moritz_ on 28 December 2010.
flussence How would I go about doing Str.lines but keeping trailing \n? .comb(/^^ .* $$/) doesn't seem to do it, I just get the whole multiline string back. 00:04
00:05 ab5tract left
sorear .comb(/:r ^^ \N* \n?/) maybe 00:07
00:08 GinoMan_ left
flussence that works, thanks 00:09
now the hardest part, figuring out how to deal with \t in .indent... 00:10
00:23 plobsing joined 00:27 felliott left 00:30 masonkramer left, lopnor joined 00:34 felliott joined, lopnor is now known as lopaway 00:37 hacknperl joined
hacknperl We need a port of perl for WebOS. :) 00:38
KatrinaTheLamia blinks 00:40
00:40 impious left
KatrinaTheLamia "WebOS"? 00:40
00:40 ab5tract joined
KatrinaTheLamia will work on Jerl6 the JVM Perl6 with access to Android OS in a bit... just have to deal with a bit of furry drama. 00:40
hacknperl WebOS = HP Palm's mobile operating system
hmmm... I would love to be able to write webos applications in perl rather than using javascript... 00:41
00:41 gbacon left 00:42 QinGW joined
hacknperl I'm so out of touch with the hacking world these days anyway.... 10 years ago I was in love with perl.... that was my first business... selling CGI scripts... now I want to get back into programming and I like WebOS... and I think perl would be awesome on it... 00:42
flussence for a minute there I was thinking that was the same thing as WebTV...
KatrinaTheLamia people still use Palm's Mobile OS? 00:43
hacknperl It's not PalmOS
It is somewhat new...
WebOS is arguably one of the nicest mobile operating systems... way better than IOS
and it is Open... not open source but the source is availaable
00:44 felliott left
KatrinaTheLamia okay... I am more looking at what can be done on Android OS, iPod OS?, and Windows Mobile... I never heard of WebOS... I will look into it. 00:45
hacknperl This year Palm will probably be releasing a nice tablet form factor device and new smart phones... I am just wondering how hard it would be to make perl run on it... there id a plug in development kit avaiable for webos so people can easilly port games written in c++
colomon There was talk of trying to get Rakudo running on the Android at YAPC:NA last year, but I think speed and memory usage need to come down a good bit before it is practical... 00:46
hacknperl You never heard of it because it was so badly marketed... beleive me when you use WebOS you will fall in love... you will probably hate the hardware its on but man what i slick interface with true multitasking... and on and on...
KatrinaTheLamia colomon, well, the even bigger issue is Parrot is written in C, correct? Android applications run on the JVM. 00:47
colomon No, there's some sort of native interface available too. 00:48
Perl 5 runs there, for instance.
KatrinaTheLamia The Android OS, is in actuality a framework of JVM classes and services running ontop of a Linux Kernel.
hacknperl Somone was able to get Open Office running on webos... not very usable with a touch type interface but its definetly possible... so i think they had an xwindows system running on the device ... im not really sure how they did it
KatrinaTheLamia I do not know about a native interface... I am guessing Perl 5 runs on there, coupled with the Linux Kernel functionality--but I doubt you can really use it for anything interesting. 00:49
hacknperl, I may look into WebOS.
colomon I dunno, it seemed fairly interesting to me, but I got frustrated when I had difficulty getting it to talk to my GPS.
flussence POSIX is POSIX, all it might need is a libc.
hacknperl Yes please do... could bring more people to perl and more people to palm
webOS is based on a windows kernel 00:50
colomon But I suspect that may have been because my GPS is dodgy
hacknperl i mean linux kernel
lol sorry
diakopter hee
hacknperl im reading too much stuff here...
diakopter the stuffy stuff, it stuffs you up
KatrinaTheLamia flussence, colomon: yes, but that would not be following the Android OS model of doing stuff. It does have some TIMTOWDI on Android... but that is like doing Perl 4 code in a Perl 5 workspace. 00:51
dalek p-rx/smoke: 1ec882f | dukeleto++ | build/Makefile.in:
Getting closer to smoking nqp-rx
p-rx/smoke: 27f1ff4 | dukeleto++ | build/Makefile.in:
Looks like we need a harness
KatrinaTheLamia it would be easier to just have Perl 6 target the JVM, and do stuff from that point.
and drama is straightened up... time to work on Java Kitty... my JVM+Android dead cat thing. 00:52
flussence I suspect this whole JVM thing is why google are developing Native Client...
sandboxed app, runs in a browser, cross compile and it runs on Android. 00:53
it'd solve the complaints from people not wanting to use Java anyway 00:54
KatrinaTheLamia does a quick summary of Web OS on wikipedia before reading real information of it.
hacknperl I kind of have an idea... WebOS apps are basically javascripts and HTML... why cant perl be compiled on it directly and ran at a lower level and then use the web browser on the device to interact with perl scripts... or just write a native webos app that acts like a web server/client with the html veiw and let it run perl scripts in the backgrpound... kind of wwhat it does now... i dont know... im so stupid with this stuff now a days...
KatrinaTheLamia you know--we have Android and Web OS both running on a Linux Kernel. It is stuff like this, that makes me suggest Windows Next will run on a Linux Kernel as well. 00:55
flussence rofl
KatrinaTheLamia hacknperl, well, Android OS and iPod OS both function with a lot of stuff done in HTML5... I am checking to see if there is more to it than that.
00:55 colomon left
flussence nah, they'd write their own kernel long before they'd give Linux that much credit... 00:55
KatrinaTheLamia flussence, look into Singularity and Miranda. It is not as crazy as it sounds =P
flussence I've heard of Singularity before, not the other one though 00:56
KatrinaTheLamia Singularity 2.x on top of Mono on top of Linux. Now, hand that to the Marketing department."Our new Windows takes the stability and functionality of the Linux kernel, and extends it, with all the user friendly functionality users at home love. We have now marred the easy to use and love operating system Windows, with the robust and stability, many technical experts have rallied behind." 00:57
sbp worked for Apple. except for the whole BSD thing 00:58
KatrinaTheLamia scratches her head, "I kind of have some knowledge in how to do marketing... both proper marketing, and that... which is evil marketing."
00:59 felliott joined
KatrinaTheLamia hacknperl, oh, hey... looking into what they are saying on Wikipedia... WebOS uses libSDL, pulseaudio, Gstreamer, libpurple and a few other standard linux elements. I will now look at the real information on developing for it. 01:00
sorear KatrinaTheLamia: do you have anything actually to show for Jerl yet? 01:01
KatrinaTheLamia sorear, ah... no... I kind of was dealing with being homeless. 01:03
which I was still trying to make commits to it, while staying at the women's shelter.
I will keep shooting stuff out on it, though...
01:03 rgrau left
KatrinaTheLamia seems the universe hates me... I've had to deal with Hypothermia, a maintence issue that resulted in chlorine gas flooding my apartment at regular intervals, having to deal with a crack addicted landlord and homelessness... I think most of the drama is over... and I can get some real work done on it! 01:05
allbery_b suggests sacrificing a young goat or too to appease the demons 01:06
two
dalek p-rx/smoke: 3fc8dc8 | dukeleto++ | / (3 files):
We can now make Smolder shoot 500-smoke
KatrinaTheLamia allbery_b, technically I am a demon... and a boss of specific group of demons. I should not have to appease my employees. 01:07
allbery_b there's always someone above (below? :) you 01:08
hacknperl so KatrinaTheLamia... do you work on perl?
sbp *perl6? 01:09
hacknperl www.webos-internals.org has tons of info on hacking this os
KatrinaTheLamia hacknperl, I have been trying to work on Perl 6. I have to ask diakopter about the model 6 developing group.. as I am in a spot I can join that... I hope. 01:10
hacknperl Once you have command line access to Linux, everything else is a process of installing a community standard library of Linux programs so that users have the same tools and options available to them. (For example, you can install openssh or dropbear enabling shell access to your device.) Once this is process is completed, pretty much anything you can do on a Linux box you can also do on the Pre or the Pre Emulator.
KatrinaTheLamia I think I was suppose to look into that, just before I got evicted by my crack addled landlord...
hacknperl cool
not cool on the eveiction
lol
KatrinaTheLamia hacknperl, that was what I was thinking. You may be get away with compiling Rakudo to ARM Linux, and doing stuff with that. 01:11
it is mostly a matter of doing the tool chain to put Parrot and Rakudo onto WebOS, if that is the case.
sorear I am uninterested in hearing about your demons. Only working code.
hacknperl but the idea is to be able to access the phone functions... so would we have to write some perl modules that worked with Palms PDK? 01:12
KatrinaTheLamia hacknperl, ah... that would probably work once importing C or other language code into Perl6 has been more properly defined. 01:13
diakopter just port a linux kernel to javascript; that'll solve it all. 01:15
KatrinaTheLamia ...
diakopter ;)
KatrinaTheLamia ...
OW! OW! OW! MY BRAINS! diakopter YOU KILLED THEM! YOU BASTARD!
flussence or port Linux to a Z80, then you can run it in that JS gameboy emulator :
);
(note to self: stop missing parentheses 01:16
KatrinaTheLamia flussence, you are a horrible person for even suggesting that notion.
flussence (didn't someone write a haskell thing in JS? That could run pugs...) 01:20
01:20 lichtkind left 01:24 jaldhar joined
coldhead i don't know if it's sufficiently full featured 01:25
also i believe it comes with popup ads 01:26
01:26 envi joined
KatrinaTheLamia either way... looking more into hacknperl's question, it may be easier to just make a tool chain to compile Rakudo/Parrot onto WebOS' Linux arrangement. 01:27
01:31 felliott left 01:34 [particle]1 joined 01:35 [particle] left 01:38 felliott joined 01:39 felliott left 02:00 stkowski left 02:01 noganex joined 02:02 roen left, noganex_ left 02:05 colomon joined
dalek rixel: 9091af5 | diakopter++ | / (9 files):
multidimensional compact arrays:

fixed size multidim array declarations. assignment to individual elements.		 02:10
rixel: bf1002a | diakopter++ | / (9 files):
enable methods to use the trampoline when called from perlesque code,

called from other CLR code (if necessary). Generate an entry point when it's emitting to disk mode. Make return types on methods optional (void).
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....dex_tablet 02:15
rixel: c52246b | diakopter++ | / (4 files):
oops, positional indexers are separated by semis
 02:16
02:21 cdarroch left 02:22 felliott joined
diakopter I could've sworn github enabled searching of source files 02:24
ah well, I have a local clone anyway
02:24 roen joined 02:44 ab5tract left 02:47 flussence_ joined 02:49 flussence left, molaf_ joined 02:52 molaf__ left 02:55 ab5tract joined
dalek rixel: f109344 | diakopter++ | sprixel/codegen/RunSharp/Operands/NewArray.cs:
workaround mono oddness in Module implicit !=
 03:00
03:06 drbean left
dalek rixel: 9d9cc9a | diakopter++ | sprixel/src/ (2 files):
fix
 03:06
diakopter perlesque: my $a=4; my $b=5; my int @blah[$a;$b]; @blah[2;3] = 4; say(@blah[2;3]); sub foo(Array[int;int] @bar) { say(@bar.Length); say(@bar.GetType) }; foo(@blah); 03:07
p6eval perlesque: OUTPUT«4␤20␤System.Int32[,]␤»
diakopter there we go :)
perlesque: my $a=4; my $b=5; my int @blah[$a;$b;4;4;4]; @blah[2;3;3;3;3] = 4; say(@blah[2;3;3;3;3]); sub foo(Array[int;int;int;int;int] @bar) { say(@bar.Length); say(@bar.GetType) }; foo(@blah); 03:09
p6eval perlesque: OUTPUT«4␤1280␤System.Int32[,,,,]␤»
diakopter heh, 1280 ints
03:10 gbacon joined 03:13 drbean joined 03:17 Vlavv left 03:26 satyavvd joined 03:33 qw3rty joined 03:34 qw3rty left 03:43 ashleydev joined, ashleydev left 03:44 ashleydev joined 03:49 ab5tract left 03:59 GinoMan_ joined 04:00 GinoMan_ left 04:03 dwhipp joined
dwhipp rakudo: ($^a for 1..3) 04:04
p6eval rakudo 337f19: ( no output )
dwhipp rakudo: ($^a for 1..3).perl.say
p6eval rakudo 337f19: ( no output )
dwhipp rakudo: say ($^a for 1..3).perl
p6eval rakudo 337f19: ( no output )
dwhipp rakudo: say ($_ for 1..3).perl
p6eval rakudo 337f19: ( no output )
dwhipp rakudo: say ([$_ for 1..3]).perl
p6eval rakudo 337f19: ( no output )
dwhipp rakudo: say ([$_ for 1..3])
p6eval rakudo 337f19: ( no output )
dwhipp perl6: say [$_ for 1..3].perl 04:05
p6eval pugs: OUTPUT«*** ␤ Unexpected "[$"␤ expecting bare or pointy block construct, ":", identifier or operator␤ at /tmp/YWE1a6C3cj line 1, column 5␤»
..rakudo 337f19: ( no output )
..niecza v1-91-ga8291b0: OUTPUT«===SORRY!===␤Unhandled statement modifier for at /tmp/h8e2SxFz71 line 1:␤------> say [$_ for 1..3⏏].perl␤Check failed␤»
04:05 GinoMan_ joined
dwhipp are list comprehensions expected to work? I know I've seen them in a rakudo-star release at some point 04:06
04:06 REPLeffect left
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ics_tablet 04:06
04:07 REPLeffect joined 04:08 GinoMan_ is now known as GinoMan
dwhipp actually, I just tried the same code on try.rakudo.org -- works there! 04:09
The question I really wanted to ask here was about placeholder variables in list comprehensions:
is ($^a for 1..3) legal, or can only $_ be used? (if so, should use of $^a be reported as an error?) 04:11
sorear the answer is in S02, S03, or S04
04:14 shi left
dwhipp Ah, S04, thanks. Most of the examples only use $_, but there is one towards the end: 04:16
rakudo: @names = ({ "$^name.$^num" } for 'a'..'zzz' X 1..100); say @names.perl
p6eval rakudo 337f19: ( no output )
dwhipp perhaps a rakudobug is needed
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ics_tablet 04:17
04:19 GinoMan left
sorear dwhipp: (no output) 04:19
04:20 GinoMan joined
sorear dwhipp: that means that p6eval isn't working 04:20
nothinhg to do with Rakudo bugs
dwhipp Investigated further: seems placeholders can only used if you include the closure. So only $_ is permitted, but is automatically unpacked by the closure
sorear: thanks: I've opened a separate window running try.rakudo.org. So I can see a more reasonable behavior there 04:21
PerlJam dwhipp: if you don't have the closure, how is perl to know the placeholder isn't for the surrounding block? 04:22
dwhipp Though still an LTA error when you omit the closure: 04:23
my @names = ( "$^name.$^num" for 'a'..'c' X 1..2); say @names.perl
Null PMC access in get_string()
PerlJam yuck.
I hate those errors.
04:23 Su-Shee_ joined
dwhipp To answer the "how to you know" question ... not sure. I only came to list comprehension recently, as a side effect of learning python. 04:24
Perl6 seems clumsy in comparison, when probably means either that I'm missing something, or that there's scope for improvement 04:25
PerlJam dwhipp: have you done dict comprehensions yet?
dwhipp: Perl may be a tad "clumsy" compared to python, but the difference is that python relies on a syntactic special case and perl relies on a general principle that has applications elsewhere. 04:26
(in the case of list comprehensions)
04:27 Su-Shee left
dwhipp Yeah. Its just that list comprehensions is the first thing that I found in Python that made me think "wow, this actually is better than perl" 04:28
Perhaps because python community doesn't like multi-line lambdas
not sure about dict comprehension: my python intuition didn't work first time, so I need to google it 04:29
04:31 gbacon left
PerlJam Are you using python 3? 04:31
dwhipp In terms of the general principle in perl, yes the use of closures for placeholders is understandable ... I even realized how to include an "if" modifier that uses those placeholders (by adding it inside the closure) 04:32
No, I'm using python 2.7.
Starting new $job next week, and they told me they use 2.6 04:33
PerlJam ah, dict comprehensions were rejected for python 2.x but got added to python 3
dwhipp: What will you be doing with python? 04:34
dwhipp I do wonder if there are some lessons to by gained from the python3 transition for the perl community
04:36 sftp left
dwhipp What will I be doing? Probably need to be careful when I say. $job is at Google 04:36
PerlJam cool.
congrats on the job then (I assume you *want* to be working at Google rather than have no other options ;-) 04:37
dwhipp It is hard to imagine being qualified for a job there without having other options ;). 04:38
04:39 dylan_ joined
dwhipp Any idea how to refer to an outer $_ when you didn't create a new scope? i.e. ( ($_ + $OUTER::_ for 1..2) for 1..2) -- not that that coding style seems like a good idea 04:41
cotto seen jnthn 04:44
dwhipp (the code I showed there gives a "Null PMC access in type()" error)
aloha jnthn was last seen in #perl6 2 days 19 hours ago joining the channel.
04:44 coldhead left
PerlJam I don't think that works as you would expect even if it didn't die a horrible death. 04:44
dwhipp I'm pretty sure you're right: I think I'd need some curlies there for OUTER to be meaningful. 04:46
PerlJam yep
dwhipp maybe ({[$_ + $OUTER::_ for 1..3]} for 1..2).perl -- though that gives the Null PMC access error. 04:49
sorear dwhipp: without a new scope, there is no outer $_ 04:52
dwhipp Doesn't the closure there provide the scope?
sorear 22:41 < dwhipp> Any idea how to refer to an outer $_ when you didn't create a new scope? i.e. ( ($_ + $OUTER::_ for 1..2) for 1..2) -- not that that coding style seems like a good idea
What closure?
dwhipp My second attempt: ({[$_ + $OUTER::_ for 1..3]} for 1..2).perl 04:53
PerlJam dwhipp: what do you want in the end? A single list containing (2,3,4,3,4,5) ? 04:56
er, ... but done correctly :) 04:57
Hmm. I must be tired 04:58
dwhipp In a way I don't really mind: either one flat list or else a 2d list: the latter seems more useful, and could be flattened to the former
(yes, this is easy if I just use a "map") 04:59
PerlJam Well, I was thinking about X+ actually
dwhipp more than one day to do it. 05:00
PerlJam aye.
dwhipp But if P6 is to have list comprehensions it'd be good if they feel like first class citizens.
plobsing dwhipp: explicit (scope) is better than implicit 05:01
dwhipp What I'm really trying to do is figure out if it is useful to use multiple statement modifiers within a single expression with nested parentheses. 05:02
It's legal syntax, but seems less useful than perhaps it could be 05:03
PerlJam rakudo: (($_ + * for 1..3) for 1..2).perl.say 05:05
p6eval rakudo 337f19: ( no output )
PerlJam hmm 05:06
dwhipp rakudo: say "hello"
p6eval rakudo 337f19: ( no output )
05:06 dylan_ left
PerlJam rakudo: (($_ + * for 1..3) for 1..2).eager.perl.say 05:06
p6eval rakudo 337f19: ( no output )
PerlJam oh
anyhow ...
#!/usr/bin/env perl6
use MONKEY_TYPING;
augment class Cool {
our Str multi method indent ($str: $steps) is export {
my $spaces = ' ' x $steps;
my $copy = ~$str;
return $steps < 0 ?? $copy.subst(/^^ \s ** { 0..4 } /, '|', :g)
!! $copy.subst(/^^/, { $spaces }, :g);
} 05:07
}
my $str1 = "foo
bar
baz";
say $str1;
say $str1.indent(4);
say $str1;
my $str2 = " foo
snarkyboojum errr
PerlJam bar
baz
blat
blah
woot
foo";
say '=' x 50;
say $str2;
say $str2.indent(-4);
say $str2;
say '=' x 50;
05:07 sftp joined
PerlJam my $x = 3.1415; 05:07
say $x.indent(4);
crap
blah ... Those buffers are long-lived ... that was my playing with Str.indent from about 7 hours ago.
dwhipp: on my rakudo here, I get this: 05:08
> ($_+* for 1..3) for 1..2
2 3 4 3 4 5
diakopter heh
PerlJam The use of the Whatever provides an implicit closure
05:10 felliott left
dwhipp Yes, sort of cool, though a little bit of a special case, in that it doesn't scale to a third level of nesting 05:11
diakopter niecza: say (($_+* for 1..3) for 1..2) for 3..5
p6eval niecza v1-91-ga8291b0: OUTPUT«===SORRY!===␤Unhandled statement modifier for at /tmp/HvQsozD06u line 1:␤------> say (($_+* for 1..3⏏) for 1..2) for 3..5␤Unhandled statement modifier for at /tmp/HvQsozD06u line 1:␤------> say (($_+* for 1..3) for
..1..2⏏) for 3..5…
diakopter std: say (($_+* for 1..3) for 1..2) for 3..5
p6eval std 625303c: OUTPUT«ok 00:01 121m␤»
diakopter odd
PerlJam dwhipp: There's more than one way to do it, but some ways are encouraged by making it a tad onerous to try other ways. :-) This is one of those instances where you really want to name your parameters rather use implicit names 05:14
dwhipp Yes, I really want to name my variables ... but no way to do that with statement modifiers 05:15
PerlJam dwhipp: yet more subtle encouragement :) 05:16
cotto Does anyone know nom well enough to explain what KnowHOW is? 05:18
PerlJam cotto: jnthn ;)
05:18 ab5tract joined
cotto correction: 05:18
Does anyone here know nom well enough to explain what KnowHOW is?
;] 05:19
PerlJam I could guess if that would help a little 05:21
cotto sure
05:21 roen left
PerlJam I would guess it's the meta-class for building objects and classes and such 05:22
dwhipp Ah, need to learn to not use the python (or math in general) statement ordering: in perl I can put the domain in front of the expression: ( for 1..2 -> $a {( for 3..4 -> $b { $a*10+$b})}).perl ...
cotto that's plausible
dwhipp And now I realize I can do: ( -> $a { -> $b { $a*10+$b } for 1..2 } for 3..4 ) 05:25
Though that's asking for trouble due to the reversal of the variable<->list association 05:26
PerlJam cotto: Maybe irclog.perlgeek.de/perl6/2010-09-17 helps? (search for KnowHOW) 05:27
colomon rakudo: (*+* for 1..3) for 1..2 05:28
p6eval rakudo 337f19: OUTPUT«Not enough positional parameters passed; got 1 but expected 2␤ in <anon> at line 22:/tmp/R4cLOfqmHl␤ in <anon> at line 1␤ in main program body at line 1␤»
dwhipp As PerlJam's been trying to tell me, there's probably not a readable/maintainable way to do multiple list comprehension in a single expression -- so best use one of the other approaches for anything that gets complex
cotto PerlJam, thanks. that looks helpful 05:29
05:34 [particle]1 left
dwhipp rakudo: ({{ $^OUTER::a ~ $^a } for 1..2 } for 3..4 ).perl.say 05:39
p6eval rakudo 337f19: OUTPUT«("12", "12")␤»
05:49 dwhipp left, f00li5h is now known as TheF00li5h 05:51 yarp joined 05:55 felliott joined
diakopter rakudo's alive again I guess 05:58
PerlJam rakudo: say "alive"
p6eval rakudo 337f19: OUTPUT«alive␤»
05:59 felliott left
dalek ecs: b046dd9 | diakopter++ | S02-bits.pod:
minor grammaro in S02
 06:16
06:32 Vlavv_ joined 06:35 TheF00li5h is now known as f00li5h 06:39 ashleydev left 06:45 apejens joined 06:50 [particle] joined 06:54 siak joined 06:55 wtw joined 06:57 cotto left 07:00 rhr left 07:04 cotto joined 07:07 lopaway is now known as lopnor 07:11 cotto left 07:12 kaare_ joined 07:22 Eevee left 07:23 Eevee joined 07:27 icwiener joined 07:30 mkramer1 left, p6eval left 07:31 mkramer joined, _twitch joined, p6eval joined, ChanServ sets mode: +v p6eval 07:43 zby_home_ joined 07:45 kanishka joined, kanishka left 07:53 redicaps joined, jhuni joined 08:01 daxim joined 08:02 MayDaniel joined 08:08 MayDaniel left 08:13 yarp left 08:17 Su-Shee_ is now known as Su-Shee 08:26 Maksim_ joined 08:33 kanishka joined
dalek ecza: 349ade7 | sorear++ | v6/ (2 files):
[v6] Finish translation of Niecza::Actions
 08:51
TimToady is home 08:55
sorear wb TimToady 08:58
home = #perl6 right? :D
moritz_ "home is where my Perl 6 is" 08:59
09:00 snearch joined
Su-Shee I have a perl 6 in every home. ;) 09:01
09:04 justatheory left 09:12 ab5tract left 09:25 hanekomu joined 09:26 snearch left 09:28 flussence_ is now known as flussence
dalek ecza: e6a7b30 | sorear++ | v6/ (2 files):
[v6] Pass filename/modtime through Unit
 09:31
ecza: bdde84c | sorear++ | / (2 files):
Merge back aggregate improvements from tryfile
ecza: f82d33b | sorear++ | v6/ (2 files):
[v6] Translate Niecza::Grammar
sorear out.
09:31 muixirt joined 09:34 masak joined
masak oh hai, #perl6 09:36
flussence perl6: (1..2 X+ 1..3).perl.say
p6eval niecza v1-92-g349ade7: OUTPUT«===SORRY!===␤Not a HASH reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 1194.␤Check failed␤»
..rakudo 337f19: OUTPUT«(2, 3, 4, 3, 4, 5)␤»
..pugs: OUTPUT«((1, 1), (1, 2), (1, 3), (2, 1), (2, 2), (2, 3))␤»
masak niecza: 1..2 X+ 1..3 09:37
p6eval niecza v1-92-g349ade7: OUTPUT«===SORRY!===␤Not a HASH reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 1194.␤Check failed␤»
masak niecza: 1 X+ 1
p6eval niecza v1-92-g349ade7: OUTPUT«===SORRY!===␤Not a HASH reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 1194.␤Check failed␤»
09:37 dakkar joined
masak submits nieczabug 09:37
flussence niecza: 1 X 1 09:38
p6eval niecza v1-92-g349ade7: OUTPUT«␤Unhandled Exception: System.Exception: Unable to find lexical &infix:<X> in mainline␤ at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤ at
..Niecza.CLRBackend.NamProcessor.RawAccessLex (System…
flussence hm
masak we'll throw that one in as well.
it's probably a TODO.
moritz_ niecza: say 1 Z 1 09:39
p6eval niecza v1-92-g349ade7: OUTPUT«␤Unhandled Exception: System.Exception: Unable to find lexical &infix:<Z> in mainline␤ at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤ at
..Niecza.CLRBackend.NamProcessor.RawAccessLex (System…
moritz_ niecza: say 1 R+ 1
p6eval niecza v1-92-g349ade7: OUTPUT«␤Unhandled Exception: System.Exception: Unable to find lexical &reverseop in mainline␤ at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤ at
..Niecza.CLRBackend.NamProcessor.RawAccessLex (System…
09:45 kensanata joined
muixirt moritz_++ thanks for bisect-parrot.pl, hmm ccache is a prerequisite 09:45
moritz_ muixirt: it could easily be removed 09:47
muixirt does it help? 09:48
moritz_: it assumes "$rakudo_dir/parrot" exists and it builds parrot/rakudo and runs the testfile once and than exits, right? 09:56
moritz_: I'm not sure how to use the script for find what version of rakudo/parrot caused some bug 09:58
something like git bisect run tools/bisect-parrot.pl I guess ... well only for experts :-) 10:00
10:00 cotto joined
moritz_ muixirt: did you read the commit message? 10:01
it can only be used to bisect parrot
and yes, you use git run $script $testfile
sorry, occupied with $other_job atm, and a bit terse
muixirt commit message? oops :-) 10:02
10:02 QinGW left 10:04 BaggioKwok joined
muixirt assuming that i'm not the only person that's clueless it may be useful to put the info of the commit msg into the USAGE msg 10:05
moritz_ aye 10:06
as I wrote (in the commit msg :-) it's a very early draft 10:07
muixirt and is it really git bisect run ../tools/bisect-parrot.pl test-file.t 10:08
not git bisect run tools/bisect-parrot.pl test-file.t
assuming execute bit is set 10:09
(didn't read man git-bisect)
10:10 redicaps left 10:13 BaggioKwok left 10:24 Maksim_ left
apejens are there any good tools for helping write a grammar? Can I for instance dump the syntax tree easily? 10:29
masak apejens: a very good question.
apejens: there is .perl, but it's a bit too informative sometimes. 10:30
apejens: my best tip is to develop your grammar in very small increments, and using tests.
10:31 Chillance joined, siak left
masak apejens: there are also debugging capabilities built into grammars. 10:32
apejens: moritz_++ has a good blog post with all the most common tips and tricks. I'll try to find it for you.
apejens: perlgeek.de/en/article/debug-a-perl-6-grammar 10:34
10:38 kanishka left 10:43 satyavvd left 10:50 satyavvd joined 10:55 Maksim_ joined 10:58 icwiener left
jnthn o/ folks 10:58
jnthn is home
...and looking forward to things being more "normal" again for a while. :) 10:59
lopnor hello, i'm banned from rakudo.org/ 11:00
what's happend?
403 with "Sorry, 122.1.9.166 has been banned." it's my ip. 11:01
11:06 kanishka joined 11:08 wamba joined
colomon jnthn: \o/ 11:09
jnthn o/ colomon 11:12
How's things?
colomon eh, insane-ish
my wife went to the emergency room last night with (it turned out) gallstones. 11:13
she's been referred to a surgeon as a result.
It's at least another four weeks before I'm supposed to be able to pick up our little guy, as a result of my own surgery last week. 11:14
as a result, at the moment, she can't drive (lots of painkillers) and I can't comfortably put on jeans or pick anything up. 11:15
the two of us combined can do almost anything a normal person can! :)
jnthn Ouch!!! 11:16
I wish you both a swift recovery.
colomon thanks. 11:17
muixirt gallstones? Thought that be only a problem for elderly people... 11:18
colomon apparently not.
her dad had his gallbladder out when he was 40, and a grandparent on both sides somewhere in the 30-50 y.o. range. 11:20
11:21 agentzh joined 11:22 drith joined 11:23 lopnor is now known as lopaway
Maksim_ Are there people in p6 community working on a tk or qt modules for p6? 11:24
11:24 lopaway is now known as lopnor
tadzik there was this GUII idea, like DBI is for databases 11:26
I feel like a blogpost, so maybe the idea will get some attention
apejens masak: ahh, thanks 11:29
11:29 mj41 left
masak apejens: the reason I thought yours was a good question is that I could definitely see a use for a tree-of-matches form of output you ask about. someone ought to write one. 11:30
apejens mhm
:p
but thanks for the moritz post, looks like a great place to start :) 11:31
jnthn lolitsmasak!
masak I also think there are far too few tools for giving information about a grammar parsing *failure*. the only options available to the user right now is debugging and scowling.
lolitsjnthn!
jnthn: home already? 11:32
jnthn masak: Yes. :)
masak jnthn: welcome to Lund! :)
Maksim_ tadzik: oh I see.. I thought there is already some work going on ok a GUI module.. I guess it is too early.. thanks :)
jnthn masak: It has snow AND sunshine! \o/
masak :)
Maksim_ tadzik: s/ok//;
11:35 lopnor is now known as lopaway
colomon imgs.xkcd.com/comics/good_code.png 11:37
11:40 Intensity left, redicaps joined
apejens has anyone written a grammar for something like textile or markdown? 11:42
tadzik well volunteered! :) 11:43
I have a grammar for sexpcode
almost complete even
11:43 Chillance left
masak apejens: yes, there is a markdown grammar in parrot. I'm writing one for niecza now. 11:44
flussence rakudo: my @lines = [" quack"]; @lines ~~ s/^ ' ' ** 2//; say @lines.join.perl 11:45
p6eval rakudo 337f19: OUTPUT«" quack"␤»
apejens masak: in the parrot source tree? 11:46
and niecza?
flussence I'm pretty sure this regex's identical to the code I'm running, yet in there it only removes one space :/
oh. *this* is identical: 11:47
apejens masak: ahh, github.com/fperrad/markdown/blob/0...grammar.pg I presume?
flussence rakudo: my $outdent = 2; my @lines = [" quack"]; @lines ~~ s/^ ' ' ** $outdent//; say @lines.join.perl
p6eval rakudo 337f19: OUTPUT«" quack"␤»
masak apejens: yes, in the parrot source tree somewhere.
apejens: my niecza code is not public yet, but OTOH I haven't gotten very far.
11:48 gimix joined
flussence (I guess the {}s around **{$outdent} are required there...) 11:49
masak shouldn't be, I think. 11:50
tadzik oh, that could be moved to nqp
(markdown)
flussence It does behave differently for some reason...
rakudo: my $a = 2; my $b = ' b'; $b ~~ s/' ' ** {$a}//; say $b.perl
p6eval rakudo 337f19: OUTPUT«"b"␤»
flussence without { }, it evaluates $a as 1 every time. 11:51
wonder if that's a bug...
masak a TODO bug, I'd say.
I think we have something like that in RT already.
flussence ok. 11:52
tadzik RT is full of mess :/
there are dozens of fixed, but opened bugs
apejens what does the . mean in <.somerule>?
masak apejens: it means "call, but don't capture" 11:53
tadzik do not call actions for this one, methinks
ah, no. What masak says, yes 11:54
masak no, it has nothing to do with actions.
as far as I know.
apejens ahh, thanks
and {*} ? 11:56
tadzik that's deprecated 11:58
it means "call the action method now"
12:00 mj41 joined 12:11 Maksim_ left
masak nowadays, actions are called upon rule completion. 12:12
(which was where the explicit {*} markers were usually located anyway) 12:20
tadzik there are those tricky {*} #= foobar thingies 12:21
moritz_ they just pass 'foobar' as second argument to the action method 12:23
tadzik yeah 12:24
moritz_ nowadays we typically set a dynamic variable in the regex instead
12:24 GinoMan left
tadzik so $<foobar>=[ blah blah ] 12:24
moritz_ :my $*BRANCH-TAKEN = 'foo'; or so 12:25
tadzik: that's a capture, not a dynamic variable
tadzik oh, right
12:26 kfo_ joined 12:27 kanishka left, gimix left 12:29 kfo left 12:31 agentzh left 12:34 satyavvd left 12:44 kanishka joined 12:48 hanekomu left 12:49 fIorz joined, florz left 12:50 dsp_ left 12:52 felliott joined 12:54 jhuni left 12:58 hanekomu joined 13:05 WonTu joined, WonTu left 13:07 dsp_ joined
takadonet morning all 13:08
colomon o/ 13:12
13:16 roen joined 13:17 noganex left 13:20 MayDaniel joined
masak takadonet: \o 13:20
dalek kudo: 1037fc1 | KodiB++ | src/Perl6/Actions.pm:
[Perl6/Actions] Eliminated $MAX_PERL_VERSION per tadzik++.
13:22 ashleydev joined
felliott hello 13:28
takadonet felliott: hello 13:29
felliott I was looking at some of the xor bugs and hoping I could solve them, but it looks like one of them was fixed already.
thundergnat++ implemented [^^] back in early Decemebr, but I think the patch fell by the wayside.
His patch is here: gist.github.com/731232 13:30
apejens the latest rakudo-star needs a not-yet-released parrot?
felliott and the discussion was here: irclog.perlgeek.de/perl6/2010-12-07#i_3063248
Is there was anything I could do to help get that applied? 13:31
I've run the patch against the spectests and it didn't cause any regressions. 13:32
moritz_ apejens: it usually requires the last released parrot 13:33
apejens last supported, or released?
moritz_ apejens: last released 13:34
felliott: the patch doesn't apply here
felliott: if you supply one that's generated with `git format-patch` I'll happily apply 13:35
felliott ? Hrm, I haven't updated my rakudo in a few days, let me try again.
Okay, shall do. Thanks!
moritz_ just git pull --rebase
and then git format-patch HEAD^
13:36 MayDaniel left 13:37 noganex joined
felliott mortiz_: should I email or nopaste? 13:38
moritz_ felliott: nopaste is fine, as long as you don't wrap any lines 13:39
felliott moritz_: gist.github.com/769459 13:40
moritz_ applies and runs a spectests 13:42
13:42 pmurias joined
pmichaud patch rejected 13:45
the patch removes the short-circuit capabilities of the existing ^^ 13:46
(at least it does as I read it)
also the [^^] operator won't work for an empty list 13:47
moritz_ in what way does ^^ short-circuit now? as a list?
rakudo: 1 ^^ 1 ^^ (say 3)
p6eval rakudo 337f19: ( no output )
pmichaud rakudo: say (1 ^^ 1 ^^ 1 ^^ say('don't see this')
moritz_ rakudo: 1 ^^ (say 3)
p6eval rakudo 337f19: OUTPUT«===SORRY!===␤Unable to parse postcircumfix:sym<( )>, couldn't find final ')' at line 22␤» 13:48
rakudo 337f19: OUTPUT«3␤»
pmichaud (mine was the error)
rakudo: 1 ^^ (say 3)
moritz_ yes, ' in '-quoted string
p6eval rakudo 337f19: OUTPUT«3␤»
pmichaud rakudo: 1 ^^ 1 ^^ (say 3)
p6eval rakudo 337f19: ( no output )
pmichaud ^^ short-circuits after observing the second true value 13:49
moritz_ and the patch changes that. Confirmed. 13:50
takadonet How do you do multiline strings in p6? 13:52
flussence what, like heredocs?
jnthn You can split the contents of a '...' and "..." over multiple lines too, fwiw. 13:53
takadonet p5: perl <<EOF ......... EOF
s perl/print/
apejens can I from Match.perl on the string "" distinguish between a failed match and successful match?
jnthn bbi15 13:54
moritz_ rakudo: say ('' ~~ /^/).perl
p6eval rakudo 337f19: OUTPUT«Match.new(␤ from => 0,␤ orig => "",␤ to => 0,␤)␤»
flussence takadonet: "q:to'EOF'", S02:260
moritz_ rakudo: say ('' ~~ /^foo/).perl
p6eval rakudo 337f19: OUTPUT«Match.new(␤ from => 1,␤ orig => "",␤ to => -3,␤)␤»
masak takadonet: those are called heredocs.
takadonet flussence: thanks
moritz_ apejens: in rakudo, a failed Match has .to < .from 13:55
apejens ok
pmichaud I doubt that's spec, though.
Don't know what the spec would say about it.
moritz_ it just .Bool's to False
flussence
.oO(.indent will make heredocs readable again!)
13:55 dsp_ left 13:57 hanekomu left
colomon aren't heredocs supposed to auto-unindent in p6? 13:57
13:57 dsp_ joined
moritz_ they are 13:58
colomon I've been waiting for that feature to work for five or six years now. :)
13:58 GinoMan joined
colomon sounds stupid, but that was one of the things that excited me most about perl 6 when I first learned of it. 13:59
13:59 GinoMan left, GinoMan joined
masak relying on .to < .from is EVIL, short-sighted, and implementation-dependent. 14:00
use ?$match instead.
moritz_ right
flussence and not perl6ish!
moritz_ if you ?eval($/.perl) you'll get the correct result too
apejens yesyes, I'm not relying
I was just dumping .perl during testing
moritz_ so it's not a .perl bug, strictly speaking
apejens and was curious
moritz_ apejens: it turns out that .perl is only of limited use during debugging... we'll probably have a .pretty method that provides less formal but much more concise output 14:01
apejens mhm
moritz_ I think niecza implements that already
well, .perl *is* useful, but it takes long for a human to parse a .perl of a nested Match object, so it could be better 14:02
apejens and outputing stuff like what matched, not just from and to etc might aid more 14:03
flussence does Match.perl do its own output formatting? 14:04
moritz_ flussence: yes
colomon why are we talking about this instead of just coding something up?
moritz_ because coding is hard.
flussence might be nice to have a yaml-like common formatting function that contains the necessary smarts to make it look nice
moritz_ I think there was a .pretty method nopasted at some point
but wasn't applied back then because not specced 14:05
colomon should have made a module.
shortcircuit rakudo: say {my $a = 7; print $a;}.perl 14:06
p6eval rakudo 337f19: OUTPUT«{ ... }␤»
shortcircuit ...That means nothing to me.
colomon .perl can't do code objects yet 14:07
moritz_ shortcircuit: rakudo can't deserialiaze closures yet
14:07 plobsing left
shortcircuit Aww. 14:07
I was hoping there'd be an interesting Quine solution in there somewhere.
arnsholt There will be =)
moritz_ curses at expired nopastes 14:08
shortcircuit Trying to make a GeSHi file for PIR. The Parrot docs are *not* the greatest-organized for someone looking for a complete list of keywords. 14:10
masak moritz_: expired nopastes are why I went over completely to gists. 14:14
flussence forgotten gist links are why I have a github username :) 14:15
pmurias App::Nopaste gist support is broken :( 14:17
14:17 dsp_ left
moritz_ irclog.perlgeek.de/perl6/2010-08-11#i_2690764 fwiw 14:19
14:19 dsp_ joined
apejens gist.github.com/6765af670a7371835a3a <-- why does the paragraph rule turn into an endless loop? 14:19
moritz_ I'm not entirely happy with the output, but I think it can be tweaked
apejens: pline can match zero chars 14:20
apejens: and if you wantify it, it loops
known limitation in the regex engine
14:20 orafu left
apejens aha 14:21
moritz_ don't quantify it, and instead use \n+ in regex eol
flussence rakduo: my $outdent = 2; my $ws = regex { ^ ' ' ** { $outdent } }; my $str = ' quack'; $str.subst($ws, '').perl;
rakudo: my $outdent = 2; my $ws = regex { ^ ' ' ** { $outdent } }; my $str = ' quack'; $str.subst($ws, '').perl;
p6eval rakudo 337f19: ( no output )
flussence whoops
rakudo: my $outdent = 2; my $ws = regex { ^ ' ' ** { $outdent } }; my $str = ' quack'; $str.subst($ws, '').perl.say;
p6eval rakudo 337f19: ( no output )
flussence ...?
14:21 rhr joined
flussence rakudo: say 1 14:21
apejens dont quantify pline?
p6eval rakudo 337f19: ( no output )
flussence oh.
moritz_ apejens: oh wait, that won't work either 14:22
flussence anyway if I write "{$outdent}" it removes 3 spaces, and "$outdent" removes 1. Neither of which is 2...
moritz_ apejens: better: make pline not match zero chars, and use <pline>* instead
14:22 orafu joined
flussence I'm probably missing something obvious, but I can't see it :) 14:23
colomon gist.github.com/769504 -- oha's Match.dump in a gist. :) 14:25
14:26 thundergnat joined
pmichaud I don't think Rakudo's regex engine knows how to do runtime quantifiers yet. 14:26
moritz_ it doesn't 14:27
14:27 dsp_ left
moritz_ colomon: one improvement I'd like to see is that the matched text should be printed for every rule 14:27
colomon errrr... I'm have a hard time getting oha's Match.dump to do anything interesting?
moritz_: agreed.
flussence the thing is, without the brackets it seems to evaluate $outdent as 1 every time...
pmichaud it's probably mis-parsing something 14:28
moritz_ flussence: it parses it as <token> ** <seperator_token>
flussence: and never matches <separator_token>
flussence argh
moritz_ so it's just <token> once
pmichaud I should probably get the engine to warn/fail about that how
*somehow
moritz_ or maybe we should have a different operator for <token> ** range and <token> ** <sep> 14:29
thundergnat felliot, moritz, Re [^^]: I had uploaded a newer patch to rt.perl.org/rt3/Public/Bug/Display.html?id=65164 that addressed some issues just a few days ago.
apejens moritz_: aha, thanks :) 14:30
colomon ah, that's better
felliott sorry, very afk
colomon afk # "need to make a train"
felliott thundergnat: cool, thanks! I'll try it out here.
14:34 Maksim_ joined
masak flussence: are you using regexes to implement .indent? 14:35
flussence I'm trying a few different things to see what works...
Maksim_ hi, can I use code from p5 and p6 in same script? something like: use v5 {...}, then some where else use v6 {...} ? if yes please provide me with a place of any examples. 14:36
colomon gist.github.com/769515 example of oha's Match.dump in action.
moritz_ Maksim_: it's specced, but not yet implemented
flussence masak: the hard part is figuring out how to do the mixed tabs/spaces stuff :/
Maksim_ moritz_: understood, thanks. 14:37
masak flussence: yes, just like predicted yesterday.
flussence I think I'll have to write a lot more tests...
(I've got 6/8 passing so far!)
masak flussence: I would love to discuss this with you in detail, but I'm at work. maybe we could meet up later today and talk about it? 14:38
flussence yeah, fine :)
masak great! don't know what time zone you're in, but I'll get off in four or five hours.
in the meantime, more than 8 tests sounds like an excellent idea. :) good luck. 14:39
14:39 dsp_ joined
apejens should Grammar.parse($str, :rule<$rule>) work? (the $rule in <>) 14:39
flussence GMT here, so ±3 hours for me
apejens: $rule should usually be the name of the rule there. Try :$rule instead? 14:40
apejens $rule is a string
moritz_ then "yes"
flussence oh, that should work then
takadonet one side note is that I WAS working on porting over Text-Tabs+Wraps a month or so ago but realized I really hated it :( github.com/Takadonet/Text-Tabs-Wrap 14:41
14:41 Maksim_ left
felliott gist.github.com/769524 One of the (currently skipped) tests for [^^] is broken. This patch should fix it. 14:42
masak moritz_, flussence: I'd bet you just misunderstood apejens there. moritz_' suggested fix :$rule should work, though.
felliott According to S03, [^^] () be Bool::False
s/be/should be/ 14:43
rakudo: my $x=0; 1 ^^ 1 ^^ ($x = 5); $x.say; 14:44
p6eval rakudo 1037fc: OUTPUT«0␤»
felliott I'll add a test for that.
masak felliott++ 14:45
felliott Thanks! Testing is fun!
thundergnat felliot: there is an extensive test file included in the RT ticket. You may want to look through it for possible test to go into roast. 14:46
14:48 kanishka left
felliott thundergnat: yep, I've got them in one big xor-tests.t file now. I'm not sure where some of them should go. I'll ask here in a bit. 14:50
gist.github.com/769533 Test infix ^^'s short-circuiting abilties 14:51
14:54 plobsing joined 14:56 Miesco joined
Miesco So perl 6's goals will fix perl 5's shortcomings? Make it faster, make it more readable, make it more consistent 14:57
takadonet Miesco: yes 14:58
moritz_ besides some other goals
PerlJam good morning all 14:59
moritz_ like improved error messages, more flexibility, an improved community etc.
takadonet PerlJam: morning
Miesco takadonet: cool. Is perl 5 considered pretty slow compared to java and python?
I am not aware of how fast it is, i've tried looking at benchmarks but there all different. Is it known to be slower generally? 15:00
PerlJam Miesco: no.
takadonet Miesco: i'm not the best person for an answer on that :(
takadonet looks arounds
flussence it's fast enough for Frozen Bubble, and I was running that on a Pentium 2...
moritz_ I don't think perl 5 is slower than python, generally
15:00 wtw left
Miesco So perl 5 is faster then python?? 15:01
15:01 V1517OR joined
Miesco and how does it compare to java, generally 15:01
moritz_ comparing a dynamic language like perl or python to a compiled language like java isn't that easy
Miesco Is perl 6 going to be a compiled language?
moritz_ java has a higher startup and compilation cost, but often runs faster once it is compiled
Miesco: the specification allows both compilation and interpretation 15:02
PerlJam This language comparison still bothers me. (you can't compare execution speed of languages, only implementations)
moritz_ but most implementations so far do compilation
Miesco moritz_: And that is not the case for perl 5, right? 15:03
felliott thundergnat: the new patch doesn't short circuit infix ^^ properly
moritz_ Miesco: correct
15:03 JimmyZ joined
Miesco moritz_: So is that one of the plans to make it faster? 15:03
moritz_ Miesco: that would be a broad oversimplification 15:04
felliott see gist.github.com/769533 for an example test
PerlJam Miesco: faster for the humans.
thundergnat felliot: looking
moritz_ Miesco: in modern programming languages the boundary between compilation and interpretion are blurred, and not really so interesting anymore
for example java compiles to byte code, which is then interpreted
is it compiled or interpreted?
15:05 Patterner left
moritz_ ... and then maybe JIT-compiled by the VM 15:05
Miesco every program is interpretted at some point
huf perl5 compiles too, and then interprets, but then sometimes goes back to compiling a bit ;)
moritz_ the question of compiled/interpreted doesn't make much sense anymore
huf is it better than cake? YES.
^^ that's an important question to ask
Miesco So does perl 5 have bytecode? 15:06
PerlJam Miesco: nope
moritz_ yes it has 15:07
it's just usually never written to disc
but it has in-memory bytecode
pmurias it's an optree
15:07 Psyche^ joined, Psyche^ is now known as Patterner
Miesco Thats what I thought 15:07
15:07 [particle]1 joined, risou joined
moritz_ and an optree is not a bytecode? 15:07
PerlJam Well, now we're back to definitions again ... 15:08
compiled or interpreted? optree or bytecode? What does it all *mean*?
:-)
15:09 [particle] left
thundergnat felliott: I'm not sure that anything can be done about that at this point. The root problem is the same thing affecting ALL of the short-circuiting operators. 15:09
Miesco You tickle me in a way that if my wife tickled me in that way, I would go ohhhhhh -- Cleveland
pmurias moritz_: it serves the same purpose but calling it that seems a bit misleading 15:10
felliott thundergnat: Oh, okay. What's the root problem? I'd be curious to have a look at it. 15:11
Miesco I dont get how python can make a python python but perl 5 can not make a perl 5 perl 5
thundergnat It almost seems like the &infix syntax caused each parameter to be evaluated before it is passed in. I don't know exactly what is happening.
PerlJam Miesco: it can, but no one wants to do the work to make that happen. There's very little joy in it.
arnsholt Miesco: Because lexical analysis and parsing of Perl 5 is significantly harder than the same for Python 15:12
Miesco ah okay
arnsholt Which is another thing Perl 6 aims to fix
moritz_ making parsing even harder! :-)
felliott Interesting. 15:13
Miesco bye to timy the toad?
thundergnat All of the short circuiting operators have the same problem. namely, they don't.
arnsholt (And there is already a fledling p6-in-p6, masak++'s yapsi)
PerlJam arnsholt: it's even official! :)
15:13 tzhs joined
arnsholt And complete! 15:13
thundergnat short-circuit that is...
arnsholt For some values of official and comple =D
thundergnat felliott: &&. || and // have the same problem. 15:14
arnsholt thundergnat: That sounds like a plausible explanation 15:15
moritz_ they solve this problem by supplying syntactic variants of the operators, and supply the &infix:<...> thing just for the hyper operators to work with
arnsholt If the short-circuits are handled like non-short-circuit operators, the arguments will be evaluated accordingly
thundergnat felliott: rakudo may eventuall need sone kind of special cased infix routine for the s-c operators, I don't know. 15:16
felliott && short-circuiting works for me
15:16 kaare_ left, roen left
felliott rakudo: my $x; 0 && 1 && ($x = 5); $x.say 15:16
p6eval rakudo 1037fc: OUTPUT«Any()␤»
felliott ..or not
pmurias Miesco: and Perl 6 implementations in Perl 6 are being created
thundergnat rakudo: 0 && say 'oops'
p6eval rakudo 1037fc: ( no output )
thundergnat umm... 15:17
PerlJam so ... where's the short-circuit problem? 15:18
moritz_ PerlJam: the ^^ patch removes the short-circuiting from ^^
PerlJam oh.
pmurias arnsholt: niecza will be very soon a p6-in-p6
Miesco How long has perl6 been in development, 6 years now? 15:19
started 2003 right?
moritz_ the idea of creating Perl 6 is about 10 years old
Miesco 2005
PerlJam moritz_: older if you count Topaz as a proto-perl6
jnthn Rakudo has had a large chunk of it written in Perl 6 for a while now too :)
moritz_ don't know at which point "development" starts for you
arnsholt pmurias: Niecza as well? Cool! 15:20
15:20 dsp_ left
Miesco So is will I expect production ready perl6 is 2015? 15:20
Su-Shee moritz_: very simple. when $me started to write the first line. it's a different date for everyone. ;) 15:21
moritz_ some of us use Rakudo in production already
felliott Do you know if this is a problem with the grammar or with the implementation of ^^? Or something else?
arnsholt Miesco: Depends on what you're producing
Su-Shee Miesco: depends on what you're expecting ;)
Miesco Well to be generally stable but not 100% stable 15:22
15:22 [particle]1 is now known as [particle], MayDaniel joined
moritz_ felliott: the patch probably shouldn't remove the :pirop<xor> thing, but supply an infix:<^^> and/or infix:<[^^]> for hyper ops 15:22
PerlJam I don't see how ^^ can short-circuit.
Miesco Like I heard its no where near production ready now
PerlJam Miesco: again, some people use it in production now.
felliott PerlJam: 1 ^^ 1 ^^ ($x = 5) should short-circuit before the assignment 15:23
Miesco Is there a full specification yet?
Su-Shee Miesco: don't believe everything you hear on the internet. install it, try it.
[particle] there is no 100% stable, but rakudo isn't as stable as we'd like
felliott ^^ returns false if more than one argument is true.
15:23 dsp_ joined
Miesco Su-Shee: I heard it from mst I think 15:23
[particle] there is not a full spec.
Su-Shee Miesco: and mst isn't on the internet?
felliott moritz_: I'll give that a try 15:24
[particle] is there a full specification for perl 5?
PerlJam oh. I see. I was looking at it from the wrong end of the universe.
thundergnat felliot: apparently, the short circuiting ops have been fixed sometime in the last few releases. see rt.perl.org/rt3/Public/Bug/Display.html?id=77420 for bug report.
[particle] the answer is no.
Miesco I do want to develop the perl 6 core when im smarter
PerlJam Miesco: Why wait? :) 15:25
moritz_ Miesco: one doesn't need to be very smart to write a compiler
15:25 plobsing left
moritz_ Miesco: it's the simplest kind of program: reads input text, generates some output 15:25
[particle] in some ways you have to be quite stupid.
;)
PerlJam but on purpose, not on accident
Miesco Well okay
I wanna read Design Patterns first 15:26
moritz_ ... and mst says as much (though he yells "You're not good enough") :-)
thundergnat felliott: nope. still broken
PerlJam Miesco: just so you know, that book will be thoroughly unhelpful for coding Perl 6
[particle] Miesco: might i suggest 'pragmatic programming'?
Su-Shee mst certainly didn't yell "design pattern" and he also yells very nicely against this "is the spec ready yet" thing in favor of perl 6. ;) 15:27
pmichaud "is the perl 5 spec ready?"
Miesco [particle]: Is that a c++ book?
PerlJam Su-Shee: he knows not to slit Perl 5's throat when he stabs at Perl 6 ;) 15:28
[particle] Miesco: it's a book about how to approach the topic of programming. it applies to any language.
Su-Shee PerlJam: nicely put. :)
pmichaud The problem with starting a language with a specification seems to be that lots of people assume you have to "finish the specification" before you can have anything usable, stable, or production ready. 15:29
Su-Shee that doesn't even work in companies ;)
Miesco [particle]: Ah I have to get this book
PerlJam pmichaud: only because people are so used to the waterfall model of doing things (even those who should know better)
takadonet It's not like building a house where you need to have the 'finished spec' before you can start
pmichaud even builders know that the 'finished spec' of a house or building isn't the final design. 15:30
in fact, the construction industry lives on change orders.
mathw indeed, they oftne put slightly curvy walls in, just for fun
even if the ones on the plan were straight
Su-Shee realistically nothing is really build that way (I should know, we have a software to manage large construction sites ;)
[particle] there were usable, production-ready c compilers before c89 was an approved standard
jnthn The line between fun and incompetence can be thin. ;-)
[particle] there were usable, production-ready c++ compilers before c99 was an approved standard
pmichaud construction bids are often put in "at cost" (no profit) with large profit to be made on the inevitable change orders during construction. 15:31
jnthn remembers the "features" of his computer science lab. :)
PerlJam wonders how often the design spec is compared to the as-built spec in the construction industry. 15:32
pmichaud depends on the lawsuit.
PerlJam heh, I suppose so
pmichaud has some vicarious experience here. :-)
Su-Shee Imagine someone would have spec'ed the coming down of the Berlin wall first to actually make it happen.. ;)
pmichaud: may I offer you our great product to manage competently large contstruction sites? it's even written in perl! ;) 15:33
cons-truc-tion.
felliott thundergnat: thanks for showing me #77420. Unfortunately, the fix is quite a bit beyond my understanding ATM, but I'll study it. 15:34
takadonet Su-Shee: you mean cons::truc::tion ?
Su-Shee rephrases: build::this::thing::here ;) 15:35
thundergnat felliott: it is still broken in current rakudo
pmichaud I'm sure it was reverted due to an incorrect design. 15:36
let me check the log for the revert
hmmm, *2009*
that's pre-nqp-rx 15:37
that would've been in alpha
so, it wasn't reverted -- it was not transferred to ng.
at any rate, it's wrong because it depends on compile-time detection //= and ||= 15:38
PerlJam pmichaud: btw, can you point me in a good direction to look for making 'aaaab' ~~ / a ** {2} / work properly?
pmichaud PerlJam: are you also looking to do / a ** {$x} / in the process?
PerlJam aye. 15:39
takadonet PerlJam: i had that problem! one sec
PerlJam but actually, I wanted {0..$x} and I'm sort of assuming it's related.
jnthn pmichaud: Is compile time detection actually wrong there?
pmichaud: I mean, we need to thunk the RHS at the very least, I would guess...
pmichaud jnthn: afaic, it is.
I agree with thunking the RHS
jnthn || and // are compile-time detected 15:40
pmichaud but modifying the past tree based on the identification of &infix://= (by name) is really not the right way to be doing things.
jnthn Or at least, we know to compile them to "not just a call to infix:<||>"
Ah, yes.
For one because it doesn't even parse as that these days. :)
pmichaud PerlJam: I think the regex engine itself needs some modification to be able to recognize a closure as a repetition parameter, and then know what to do with it. 15:42
15:42 alester joined
takadonet pmichaud: i have a work around... just trying to get it to work! 15:42
PerlJam pmichaud: does nqp grok closures? 15:44
pmichaud in regexes, yes.
takadonet sorry gtg :(
PerlJam: github.com/Takadonet/Text-Tabs-Wra...rap.pm#L55
pmichaud (nqp groks closures outside of regexes too)
takadonet github.com/Takadonet/Text-Tabs-Wra...rap.pm#L56
that an example how it works 15:45
sorry I cannot golf it... work
pmichaud jnthn: currently we handle || and // by virtue of the :pasttype<...> parameter to the token.
the compiler doesn't specifically recognize them in actions.pm 15:46
jnthn pmichaud: True. I still call that "compile time". :P
But yes, it's in the grammar, not the actions.
felliott Unfortunately, I have to go. thundergnat, thanks for your work on this, and thank you everyone else for your help. I'll work on adding thundergnat's tests when I get back.
It's been fun!
pmichaud felliott++
jnthn I wonder if we can form thunk in a reducecheck...
...or if that's sane. 15:47
thundergnat felliott: thanks, bye
15:47 felliott left
pmurias ggoebel: the concurrency book is fun 15:47
pmichaud well, iirc TimToady++ seems to put || and // and ^^ and && in the realm of macros
...but I'm not quite sure what that implies for implementation yet
STD.pm still parses them as ordinary operators 15:48
jnthn Hmm.
*nod*
I'm hoping that once nqp-rx gains roles, the whole slang/macro thing will become a tad easier to explore.
pmichaud (the synopses mention them in the context of macros in a few places)
jnthn Since we can mix into the current language.
(Also, custom operators can become way less leaky. :)) 15:49
pmichaud it would be very helpful to see a plausible definition of the short-circuiting operators in perl 6
15:49 hanekomu joined
jnthn pmichaud: Yes, very much so. 15:49
pmichaud i.e., if we had a full-fledged p6 implementation, how would they be defined?
then we'd have a much better idea of what we're working towards. 15:50
15:50 Trashlord left
pmichaud or, if they're somehow compiler primitives, let's state so somewhere. 15:50
more generally, if someone wants to write their own short-circuiting operator, (1) is it possible and (2) how do they define it? 15:51
I suppose I could resurrect pm.txt and add those questions to it. :-) 15:52
colomon \o/
15:53 thundergnat left 15:54 kaare_ joined 15:56 mce_ro joined 15:58 spq1 joined
colomon woah, MAIN is really addictive. 15:58
16:00 plobsing joined 16:01 JimmyZ left 16:22 shi joined
PerlJam If you've got: proto token foo { <...> }; token foo:sym<a> { ... }; token foo:sym<b> { ... }; etc. How do you match just the "b" foo? Can I say <foo:b> or something ? 16:25
<foo:sym<b>> maybe? 16:27
arnsholt PerlJam: IIRC, you don't 16:28
At least in nqp-rx
masak what arnsholt said.
16:28 dju left
masak you can always filter on the captured submatch afterwards. 16:28
16:28 dju joined
masak asking for that functionality is like asking "I want to do a multi-dispatch, but only against this multi", if you ask me. 16:29
jnthn PerlJam: Create a rule that the proto candidate calls. 16:30
And call that one directly.
PerlJam Sometimes you might want to do multi-dispatch and sometimes you might want to dispatch directly to a particular multi. I don't see that as a problem. (Other than to do that now, you have to add an extra level of indirection)
16:31 gottreu joined
jnthn I think it's rare enough that the loi isn't so bad. imho. 16:31
PerlJam clearly it's rare. I'm probably the first person to ask for it ever. :)
16:31 [particle] left 16:32 [particle] joined
jnthn :P 16:33
PerlJam Seems like it would be less-rare in a regex setting though. 16:34
jnthn Well, there is a way to do it for code-multis. 16:35
jnthn wonders if you can my &foo-bar = token foo:sym<bar> { ... } and refer to it as <&foo-bar> 16:39
[particle] <&{'foo-bar'}> should work in any case
depending on which 'it' you mean 16:40
jnthn [particle]: - is a valid identifier character in that position, so <&foo-bar> is a fine call, afaik.
[particle]: Just meant, will it work out to call a token that is part of a proto directly.
[particle] ah, i thought you were wondering if the - would perhaps be interpreted as a character set subtraction 16:41
but ltm would make it's parsed as &foo-bar, not &foo - bar
i'm not aware of recent spec changes, so my knowledge is perhaps dated. 16:42
16:42 envi left
jnthn No, I wasn't worried about the parsing of it. Just the semantics. :) 16:43
[particle] yes, - is valid ident.
just like my &foo`bar
make that &foo'bar
colomon Huh. So <foo-bar> doesn't work, but <&foo-bar> does? 16:44
jnthn colomon: Only because I defined it.
colomon: afaik, <foo-bar> doesn't currently fall back to a lexical.
colomon <foo-bar> will get you "regex assertion not terminated by angle bracket" 16:45
jnthn Oh?
std: /<foo-bar>/
p6eval std 625303c: OUTPUT«ok 00:01 120m␤»
jnthn nqp: /<foo-bar>/
p6eval nqp: ( no output )
16:45 hanekomu left
jnthn rakudo: /<foo-bar>/ 16:46
p6eval rakudo 1037fc: OUTPUT«===SORRY!===␤regex assertion not terminated by angle bracket at line 22, near "-bar>/"␤»
jnthn Hmpfh.
I think Rakudo is wrong there.
colomon yes
agreed
I was worrying about what is currently practical. :)
jnthn Ah, OK
16:46 Layla_91 joined
masak rakudo is wrong, it's a known and reported bug. 16:49
Layla_91 hi i am new and have few questions. 16:51
jnthn Layla_91: welcome :) 16:52
tadzik hello zebras
and hello Layla :)
masak hi Layla_91. ask away. 16:53
and hi tadzik :)
PerlJam I'm old and I always have questions too ;) 16:55
Layla_91 jnthn: em.. I usually program in python, I saw p6 site and it seem nice.. but i did not find a lot of full programs or advanced documents. I also need to know if I can embed c or c++ code into p6 because it seems not very complete.
16:56 redicaps left
diakopter is RosettaCode linked from perl6.org ? 16:56
shortcircuit I think so
PerlJam Layla_91: what are you going to use Perl 6 for exactly?
shortcircuit It used to show up in the Documentation bubble. It did when I swung by there within the last week. 16:57
PerlJam shortcircuit: still does
shortcircuit Ah, there, in the Code examples bubble at perl6.org/documentation/
pmichaud diakopter: I don't see rosettacode on perl6.org, and you're absolutely right that it needs to be there, prominently
16:57 am0c left
pmichaud I'd even like to see it on the main page 16:57
oh, there it is (code examples bubble)
I want it on the main page 16:58
masak Layla_91: there are ways to tie together C code and Perl 6 code. and yes, there are fewer libraries written in Perl 6 at the moment than in, say, Perl 5.
jnthn Layla_91: There are ways to call (some) C code from Perl 6.
Though it's line like Perl 5's Inline::C
*like
Layla_91 PerlJam: I started working in new company two weeks ago. I am working on a new website and I thought of using perl6 in some parts. it will have forums and web chat.. 17:00
PerlJam Layla_91: cool. Have you seen modules.perl6.org ?
17:02 cotto left
Layla_91 PerlJam: I seen them but I feel there is very little documentation and no books about them. I only know python and everytime I ask for help they tell me "it is almost like in perl5!" should I learn perl5 first? 17:02
17:04 hercynium joined
dalek href="https://perl6.org:">perl6.org: e6f6f63 | pmichaud++ | source/about/index.html:
Typo fix.
 17:04
PerlJam Layla_91: Perl 6 may not be for you yet. It's mostly still for early adopters who don't mind having to debug strange errors or figure out how things work for themselves. 17:05
Layla_91: though, there is a book: github.com/perl6/book/downloads
Layla_91: it's not complete, but it's a start 17:06
17:06 justatheory joined
Layla_91 PerlJam: but as per your knowledge, can it be used know for web programming.. I have no problem spending few weeks trying it :D 17:07
17:09 daxim left
PerlJam Layla_91: Yes, it can. But I have no first-hand knowledge of doing so however. There are others here who have actually written Perl 6 based web sites who may be able to give you some advice. 17:09
Layla_91 PerlJam: cool! :D Can you tell me how to reach them?
PerlJam: I am bored from python all time :(
tadzik Layla_91: hello :) 17:10
moritz_ masak has written a tool that generates static pages for strangelyconsistent.org/
Layla_91 tadzik: helo tadzik :D
tadzik I've used the grammar engine to do the dispatching
PerlJam Layla_91: just keep asking questions here and you'll reach them :-)
tadzik unusual, but exciting :)
masak Layla_91: my software for that blog is on github nowadays. 17:11
it's delightfully short, and I have plans to make it shorter :)
(through the right means, not through golfing it down) 17:12
Layla_91 masak: I think I am so lucky today! everybody so helpfull! :D
tadzik welcome to Perl 6!
PerlJam Layla_91: #perl6 is always like this! :-)
tadzik (where on earth is hugme)
plobsing does github have a separate perl 6 language category yet? should it?
masak Layla_91: on other channels it might be luck. on #perl6, it's just business as usual :)
tadzik plobsing: we may ask them 17:13
masak plobsing: no. yes.
plobsing I'm also not fond of how PIR gets classified as "Perl". not at all the same things. 17:14
diakopter tadzik: the link to viv from the front page ought not to link to the direct raw source of the viv script in github, as Chrome (and perhaps other browsers) treats it as a downloaded file by default. github.com/perl6/std/raw/master/viv
tadzik: Rather it should point to search.cpan.org/~sorear/STD/bin/viv
tadzik Layla_91: ttjjss.wordpress.com/2010/11/27/gra...p-further/ for I assume you are interested :) 17:15
diakopter as well as link to the std repo's main page on github
Layla_91 tadzik: yay! ^_^
tadzik diakopter: I'm on it 17:16
17:16 cjk101010 joined
tadzik hmm, another two blag toasts await me today 17:18
dalek href="https://perl6.org:">perl6.org: fc84b08 | tadzik++ | source/index.html:
Fixed a viv link, diakopter++
diakopter tadzik: also the STD.pm link from the main page 17:19
Layla_91 Thank you all! I will be back soon after I check docs and Yapsi! now is dinner time bye! :D
tadzik :F
diakopter tadzik: the reason why the cpan link is best is b/c it does formatting
and sorear keeps it updated quite often (when there are changes to STD) 17:20
but like I said both/one of them should also link to the github page :)
probably the STD one
tadzik now where is STD.pm6 on cpan?
17:21 JimmyZ joined
diakopter oh 17:21
good point.
tadzik alright, go-to-library time, bbl
diakopter well, at one time we had an html syntax-highlighted edition, I thought.
17:21 MayDaniel left
diakopter and that should be linked from perlcabal.org/syn 17:22
moritz_ IMHO the best approach would be to put a page on perl6.org that explains what STD and viv are
and link to that
17:22 Layla_91 left
diakopter moritz_: I think that's a good idea 17:22
moritz_ I can start one, and I hope others will chime in and improve it :-) 17:23
should it go under compilers/ or under viv/ ?
erm
compilers/ or specification/ ?
17:23 GinoMan_ joined 17:24 GinoMan left, JimmyZ left
pmichaud does it have to be "or"? 17:30
put the link on both 17:31
flussence is " \t quack".indent(-2), any(' 'x7 ~ 'quack', "\tquack"), ??? # :S
PerlJam flussence: No, it's just the first one. 17:32
er, yes, it's just the first one :) 17:33
flussence oh, that'll save my sanity then :)
17:34 hanekomu joined 17:35 Tedd1 left 17:36 impious joined 17:38 mce_ro left 17:43 plobsing left 17:50 pmurias left 17:51 cdarroch joined, cdarroch left, cdarroch joined 17:53 am0c joined, am0c left, cjk101010 left
pmichaud afk, lunch 17:54
17:58 hugme joined, ChanServ sets mode: +v hugme 18:01 felipe left 18:04 cjk101010 joined 18:08 Trashlord joined 18:09 kjeldahl left 18:10 risou left
dalek href="https://perl6.org:">perl6.org: 9710f83 | moritz++ | source/compilers/std-viv.html:
add dummy page for std and viv
 18:11
18:17 am0c joined 18:21 dakkar left 18:24 _twitch left 18:26 cjk101010 left, tzhs left 18:27 tzhs joined
dalek href="https://perl6.org:">perl6.org: a73081b | moritz++ | source/compilers/std-viv.html:
fill in some more details about STD.pm6 and viv
 18:30
sorear good * #perl6 18:34
diakopter howdy
moritz_ good * * 18:35
if you have anything to add or correct to the last two perl6.org commits, please do so 18:36
diakopter I think the STD.pm link on index.html should link there (sry if that's obvious) 18:37
moritz_ well, that's what I intend to do once it's reviewed 18:38
if you are impatient, feel free to do it right away :-)
diakopter also <a href="search.cpan.org/~sorear/STD/bin/viv...</a> should be on that page (std-viv.html)
I mean
moritz_ then add it
jnthn nom &
PerlJam diakopter: moritz_ didn't ask for commentary, he said "feel free to fix it" :) 18:39
moritz_ well, both if possible :-)
diakopter well, it's sorta ambiguous, but ok
18:40 plobsing joined 18:41 spq1 left
diakopter "it is not suited for general Perl 6 code." - are you sure that's correct? 18:41
18:41 shi left, jerome left
diakopter that was the qualification for "gimme5"... 18:42
I mean, it doesn't handle lots of Perl 6 features, but same for all the implementations 18:43
it *is* suitable for general Perl 6 code, definitively, in that it's the only parser that parses Perl 6 code fully/correctly - it's just that its output is data, not programs, except for the perl5 backend. 18:45
"it doesn't handle lots of Perl 6 features" refers to that perl5 backend 18:46
diakopter waits for more discussion 18:47
18:49 rgrau joined 18:50 jerome joined 18:51 kensanata left 18:55 kjeldahl joined 19:00 shortcircuit left 19:01 shortcircuit joined 19:02 wamba left
moritz_ "it compiles only those features to Perl 5 code that are needed for running STD.pm6" 19:09
diakopter yeah, but switch around "to Perl 5 code" to just after "compiles" 19:10
so that "features" and "that are needed" are right next to each other :)
sorry, I've been helping a cow-orker prep for the GMAT :D
19:11 snearch joined
moritz_ I know I'm not a good technical writer, which is why I ask you guys 19:12
dalek ecza: 0bed1b8 | sorear++ | v6/ (2 files):
[v6] Merge back tryfile -> STD
 19:15
ecza: 75a3966 | sorear++ | src/Metamodel.pm:
Slightly more verbose merge error message
19:19 hugme left 19:21 ashleydev left 19:23 Intensity joined 19:24 risou joined, MayDaniel joined
dalek href="https://perl6.org:">perl6.org: e236dc0 | moritz++ | source/compilers/std-viv.html:
rewording based on feedback by diakopter++
 19:24
href="https://perl6.org:">perl6.org: 287f7ef | moritz++ | source/ (3 files):
update some std and viv links
19:25 ashleydev joined, ggoebel left
diakopter does the link to std-viv need a .html ? 19:27
moritz_ nope 19:28
content negationation
perl6.org/compilers/std-viv
diakopter oh
19:29 VXZ left 19:31 ggoebel joined 19:44 impious left 19:47 am0c left 19:52 kavi joined, Alias__ joined
kavi hi there 19:52
takadonet kavi: hey
kavi I justlike to get a bit more involved in perl 6 19:53
is this the right place to know what is all about ?
or bettere start from a feature list web site ?
19:55 kavi left
takadonet kavi: here is the best place for updated information 19:55
19:55 Alias left
takadonet .... 19:55
[particle] teehee
takadonet [particle]: ? 19:56
[particle] (11:55:36 AM) kavi left the room.
takadonet ya
[particle] not the best way to get answers
19:57 wamba joined, wamba left, wamba joined 19:58 wamba left, VXZ joined
[Coke] . 20:00
20:00 [hercynium] joined, [hercynium] left, [hercynium] joined 20:01 zby_home_ left
diakopter well, kavi did wait 2 minutes 20:03
20:04 hercynium left, [hercynium] is now known as hercynium
KatrinaTheLamia sighs 20:06
looks like a Java Kitty specification will be made, because of my work on Jerl 6. 20:07
diakopter is it open-source?
KatrinaTheLamia The Perl community... we make Java a more intelligent solution.
diakopter, yes... it is also FlOSS.
diakopter where's it hosted
takadonet diakopter: little hard to answer in a timely manner at work... 20:08
KatrinaTheLamia ah... the Java Kitty spec has not been made yet. The working source code is being hosted on github. It is not much to look at it.
20:08 [hercynium] joined, [hercynium] left, [hercynium] joined
KatrinaTheLamia wait... I need to stop being so egotistical to assume people are talking to me, when questioning. 20:09
diakopter, I still have yet to put in the Licensing info into the source tree. I am doing a dual license thing... GNU GPL v3... or new style BSD License...
dalek ecza: 4b3d222 | sorear++ | / (2 files):
Fix class/common merge case
ecza: 212e8c9 | sorear++ | v6/ (3 files):
[v6] First steps towards integrating parser into harness
KatrinaTheLamia but... I will need to read up on how the hell people do the dual license thing... as frankly, it does not make sense to me
20:10 hercynium left, [hercynium] is now known as hercynium
diakopter just license it using the Artistic License 2 20:14
which automatically dual-licenses it
as well as *-licenses it
KatrinaTheLamia ah, okay... thanks diakopter
KatrinaTheLamia has less retarded crap to deal with now... seems diakopter helps remove menial tasks every time I talk to him =3 20:15
you know--one of these days, I am going to make a joke about "getting in the diakopter, to get there better and faster."
diakopter AL2 (listed there:) auto-allows redistro under any of these: opensource.org/licenses/alphabetical
:/
KatrinaTheLamia is... is that just that bad of a pun? I will never make it again. X3 20:16
diakopter er, it used to.
the Artistic License, I mean
KatrinaTheLamia well, I will look into it, then. I might have some "free time" later, that I can do it. By free time, I mean, "needing to stop, so I do not walk into the offices of Sun Microsystems, and strangle the CEO to death." 20:17
diakopter this is so stupid: "Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed."
so no one is allowed to write a license that is worded slightly differently? 20:18
KatrinaTheLamia As I am kind of scared that doing so, will have me become the new CEO of Sun Microsystems, in a standard mythological curse type set up.
diakopter how does one "change a document" anyway
there's a CEO of Sun? 20:19
fith You mean Larry Ellison :)
KatrinaTheLamia diakopter, laywer speak very rarely has anything to do with reality. It is the same world where you can sue, because hot coffee is hot.
moritz_ luckily it's not that crazy in all coutnries 20:20
KatrinaTheLamia fith, yeah... going to try to not strangle him... as I am not certain how that curse is passed on exactly. Could be killing him, could be him biting me during a full moon--just going to play it slow, until I know more.
sorear can we raise the level of this conversation a bit?
KatrinaTheLamia oh, sorry...
fith hehe
KatrinaTheLamia uh, anyways--will work more on Java Kitty... and look into the Artistic License in a bit. 20:21
diakopter I could've sworn a version of the artistic license referenced all the other licenses on opensource.org
a version of the AL2, I mean
it's annoying that it's not versioned (publicly, or at least discoverable (by me)) 20:22
20:24 risou left
dalek rixel: d236cd1 | diakopter++ | / (6 files):
enable lack of signatures, e.g. my $a = sub { }; (same as my Callable[:()] $a = sub () { }; )
 20:27
sorear diakopter: are you planning to do gradual types in this perlesque or will it be 100% static like the last one? 20:29
dalek kudo: 18a9b2c | moritz++ | src/core/Match.pm:
add Match.pretty method
 20:34
colomon moritz_++ 20:37
20:37 lichtkind joined
moritz_ colomon: did you try it? :-) 20:37
colomon nope
just saw it existed now, and was made happy. :) 20:38
moritz_ colomon: gist.github.com/770063 20:39
colomon looks like a lovely start! 20:40
20:40 simon left
colomon (and compiling...) 20:40
20:42 V1517OR left
moritz_ suggestions on how to make it better are very welcome 20:42
I'd love to have the name of the top level rule in the output (here TOP), but that's not available from within the match object
colomon I don't think I'll have a solid clue until I run it one real data. And I'm sure someone else will have better suggestions than me. 20:46
The important point is that it's already going to be very useful, IMO.
20:46 roen joined 20:48 MayDaniel left
colomon yeah, that looks sweet! 20:49
20:49 felipe joined
colomon I wish I'd had that when I was debugging the bits I've added to the ABC Grammar. 20:49
20:50 takadonet left
KatrinaTheLamia btw diakopter... who was I suppose to talk to, on joining model6... I should probably watch for him jumping onto here. 20:50
20:58 wamba joined 21:02 y3llow_ joined, pothos_ joined 21:03 y3llow left, y3llow_ is now known as y3llow 21:04 pothos left, pothos_ is now known as pothos
PerlJam KatrinaTheLamia: probably jnthn 21:07
KatrinaTheLamia PerlJam, ah, that is who I thought. Once he shows up, I will ask him about it. 21:09
KatrinaTheLamia curses a bit...
masak very probably. 21:10
KatrinaTheLamia you know, I will do what I can to get Perl 6 more a reality sooner.. as I am certain it will have its own version/equivalent of SQL Alchemy... which will make some of my "bootstrap" files less messy.
KatrinaTheLamia has determined, that for some things, it is better not to reinvent the wheel SQLite3 has given us.
21:11 Kodi joined, wamba left
PerlJam What little I just read of SQL Alchemy makes it sound like perl 5's DBIC 21:12
21:12 flatwhatson_ left
KatrinaTheLamia PerlJam, ah... then it must have been a while since I have done anything with DBIC. 21:12
21:13 flatwhatson_ joined
sorear Perl 6 is very real 21:13
KatrinaTheLamia Either way... I have learned to find Perl 5 a bit frightening to deal with. Perl 6 is much nicer to look at.
PerlJam sorear: but it needs *more* realism ;)
KatrinaTheLamia sorear, I mean by that, having Perl 6 in a production environment.
PerlJam KatrinaTheLamia: you mean you'll start using it in your production environment? 21:14
KatrinaTheLamia I am willing to bet, most people here, would not suggest using Perl 6 for production uses... well, the more saner of us.
PerlJam KatrinaTheLamia: depends on the production.
KatrinaTheLamia PerlJam, what Perl 6? I hope to get to the point, where it is a good recommendation to make
masak I'm using Perl 6 for production uses.
sorear KatrinaTheLamia: please stop using that word. it means nothing.
KatrinaTheLamia sorear, give me a better one then. 21:15
masak "is it useful for me?"
sbp autousefulness
KatrinaTheLamia masak, PerlJam, sorear look, at this point, Perl 6 is like KDE when it was near (or in) the 4.0 version stages.
I am mostly saying, I really want to see Perl 6, to be at a similar stage as to what KDE4 was, when it was KDE4.2 or KDE4.3 21:16
dalek ecza: d21be50 | sorear++ | lib/Kernel.cs:
Do not NRE on TypeObj.attribute
masak me too, whatever that stage is.
PerlJam KatrinaTheLamia: I don't know what that means, but I'll say that Rakudo is more useful today than Ruby was when I first used it (circa 1999)
sbp yup, nice to watch it gradually building up
the p in perl6 stands for patience, I think 21:17
KatrinaTheLamia PerlJam, I dunno, I still consider Ruby useless... but that is my standard response to something with little to no documentation to it.
sbp e is for eclectic
KatrinaTheLamia giggles at sbp
sbp r is for ruby-beating
KatrinaTheLamia l is for longevity?
sbp nah, l6 is for the Lie group E6 21:18
it symbolises how far perl6's mathematical operators still have to go in complexity before all mathematicians will be happy therewith
sorear hugs KatrinaTheLamia 21:19
PerlJam KatrinaTheLamia: Hmm. If Ruby has "little to no documentation", I can't imagine what you must think of Perl 6.
KatrinaTheLamia I look forward to the AI Duplicate of Larry Wall's brain (which everybody will have loaded on their computer devices) starts work on Perl 7... the language of the Gods... which will be around 2063.
21:19 Trashlord left
KatrinaTheLamia PerlJam, well, Perl 6 has specs on what Perl 6 is suppose to do... to figure out what anything does in Ruby, you need to jump around, and start reading source code. 21:19
sbp Larry once said, "only Perl 7 is True Perl" 21:20
KatrinaTheLamia At least with Perl 6, there are plenty of items to read to say, "no, do not do THAT with that method, that is a very Bad Time(TM Larry Wall)".
sbp Randal said, "perl7 solves the halting problem"
PerlJam I don't know ... in my most recent bout with Ruby alot of the time I just guessed what I would do if it were Perl and Ruby just worked with it. 21:21
KatrinaTheLamia Ruby, it is mostly just going through the source, and guessing how to use it... which often results in some rather horrid looking programs.
sbp fun fact: the Perl6 POD synopsis is written in Perl6 POD
KatrinaTheLamia giggles
sbp so you have to understand it semantically in order to parse it syntactically to understand it semantically
Su-Shee PerlJam: I do the same with Ruby and Python and it works perfectly fine with a little reading up here and there...
sorear KatrinaTheLamia: You're not helping.
masak sbp: but you're OK if I publish my halting-problem-solving Perl 6 module, right? :)
KatrinaTheLamia Katrina Payne said, "I have it on good authority that Eris, "bob", and the real gods use Perl 7." 21:22
sbp masak: as long as it starts the Perl7 bootstrap!
masak sbp: it's not entirely impossible.
tadzik aloha zebras
PerlJam Perl 6 is just a slang of Perl 7 after all
21:23 orafu left
masak or is it the other way around? 21:23
KatrinaTheLamia anyways--I should get back to work on Java Kitty... and once I am happy a Perl 6 DBI will actually run, I will have the Java Kitty Language Map done in that.
21:23 orafu joined
moritz_ KatrinaTheLamia: have you tried MiniDBI? 21:23
PerlJam moritz++ beat me to it
:)
KatrinaTheLamia wait... wait... what colour is Perl 7's bike shed?
moritz_ last I looked it did actually run
KatrinaTheLamia that is the most important question we all need to ask... the colour of Perl 7's bike shed
X3
sbp it's a martian colour, one that only synæsthetes can see
(only some synæsthetes, in fact, probably a minority 21:24
KatrinaTheLamia moritz_, yeah, but I get the feeling that rakudo seems to have something break on each build... that might no longer be the case.
sbp *)
the star releases seem to be pretty good
numbers, I demand numbers!
PerlJam KatrinaTheLamia: I haven't had anything significant break on me with Rakudo in a long while.
moritz_ KatrinaTheLamia: actually most of the time when we discuss rakudo breakage, it's not with the recommended parrot revision, but with a newer one
KatrinaTheLamia moritz_, I will be more than happy once Rakudo, Jerl6, or something is at a "stable" enough to instill confidence. 21:25
moritz_ KatrinaTheLamia: and installing the last release should be a pretty safe choice
KatrinaTheLamia moritz_, PerlJam I recall a pretty big one in 10.9... and another in 10.10
sbp I wonder if that works in two directions differently, like the sorites "paradox"
KatrinaTheLamia moritz_, yeah... once we get to the point where we do not have to have the latest version to not break, I will look into it, mmkay?
moritz_ KatrinaTheLamia: I don't understand... once a version it's installed, it doesn't break anymore 21:26
sbp in other words, consider the progression of perl6 implementations like the adding of grains to a pile, and surveying people as to when it reaches a "usable state" as analogous to when the pile can indeed be called a pile or heap rather than just a group or a bunch of grains
moritz_ "breaking" is implied by changes, and if you don't change, there's no break
KatrinaTheLamia moritz_, sbp, by that I mean, once people can start arguing for a legacy version of Perl 6 and not have people burst out laughing, I will do it.
sbp in cognitive studies, the points reached are different depending on whether you add or take away grains one by one. similarly, the point at which perl6 enters a "usable state" as surveyed descriptively may be different depending on people's prognostications and their retrospectives 21:27
sorear sbp: I think the proper solution is to round up and hang the smart-alecs who ask the questions
PerlJam KatrinaTheLamia: sounds like you'll be in the "late majority" for Perl 6
moritz_ KatrinaTheLamia: if you#re happy with third-hand opinions, #perl6 might not be the proper place for you
KatrinaTheLamia moritz_, ah, well, I am not happy with them. I do however want to put Perl6 to be able to be followed by them.
21:27 estrabd is now known as estrabd_afk
PerlJam sbp: you should make a blog post out of your last few lines. :) 21:28
KatrinaTheLamia PerlJam, not really... there is a chance I may put a bootstrap in Perl6 once I have confirmed that my bootstrap is not the broken bit of the program.
sbp perhaps you should syndicate the #perl6 logs and save me the effort :-)
21:29 MayDaniel joined, snearch left
KatrinaTheLamia moritz_, think of me as the person with the insanely decent marketing idea to take gothic/punk type trends, and make Hot Topic (The Gap with a darker colour scheme)... only somewhat more evil than him. 21:29
X3
sbp if I were a millionaire, I would employ a secretary to summarise #perl6 daily actually
KatrinaTheLamia notes a kick ban should be made for that analogy. 21:30
sbp I think it would be a worthwhile activity, especially for the perl6 mailing list members who miss IRC communications. (and vice versa, perhaps?)
PerlJam sbp: A more productive use of your money would be to fund pmichaud and/or jnthn to work on Perl 6 :)
moritz_ sbp: if I were a millionaire, I'd pay a few devs to work full time on ... what PerlJam said :-)
sbp: would be worthwhile, yes. Feel free to start right away. 21:31
KatrinaTheLamia yeah, going with moritz_ and PerlJam here.
sbp you two don't want funding, eh? :-)
PerlJam If I had a few million $$ to spare, I'd setup a foundation that TimToady, pmichaud, jnthn, moritz, masak, sorear, etc. (sorry, if I listed everybody I'd be here all day ;) could draw funds from as needed to work on Perl 6 related projects. 21:32
21:33 hercynium left
masak quick, someone give PerlJam a few million $$! 21:33
sorear PerlJam: tpf already has a few million (or at least 10^4) $$ to spare and issues grants to people who want to work on Perl related projects - I think jnthn is getting paid by them
sbp there's always that crowdsource finding site that raised over $1m for iPod Nano wristwatch enclosures
sorear PerlJam: it could be scaled up, not built from scratch 21:34
moritz_ sorear: it's not millions, and it's enough for smallish grants, but not enough to pay two peoples full time for two years, for example
PerlJam sorear: aye, something about TPF's public relations leaves me wanting. Perhaps I could fix that aspect of TPF instead.
moritz_ fixing TPF... not sure if that's possible with reasonable amount 21:36
PerlJam maybe it's just the volunteerism that gets in the way. TPF always seems to disappear from my radar for a while and then suddenly pop back up with something interesting for a couple of months.
moritz_ szabgab applied for a part of the Hague grant that was meant for building up fund raising 21:37
he handed in a decent proposal, and since then I haven't seen any public reaction by TPF on that grant
and nobody else applies for that part of the money
Su-Shee putting up a flattr button might at least give better funding a start..
PerlJam a perfect example :(
21:38 orafu left, orafu joined
dalek ecza: 6e32c05 | sorear++ | / (3 files):
[v6] Misc bug fixes; Hello world test works
 21:41
KatrinaTheLamia Either way... going to work on turning Perl 6 into trendy crap... mostly via me doing crap with MUMPs, JVM and a bit of Android OS stuff. You will see that I am more evil than the Hot Topic guy.
[Coke] note that szabgab's request is for $$ that doesn't go through the grant committee. Find the person in charge of the Hague grant. 21:42
KatrinaTheLamia notes that having Perl 6 on the JVM allows Perl into the Enterprise and Android Circles. Then putting MUMPs onto the JVM will get Enterprise and Androids into Hospitals...
PerlJam KatrinaTheLamia: I raise my can of Dr Pepper in salute to you :)
KatrinaTheLamia which means, your next heard transplant could be powered by Perl 6.
X3
thanks PerlJam
heart transplant*
damn, malpractice by typo X3 21:43
moritz_ [Coke]: that's another poblem... I thought quite a bit about Hague grants, and read all posts on that topic, and I still don't know who that might be
I might have read it and forgotten it
or I might have never read it
21:43 MayDaniel left
PerlJam what moritz_ said. I have no clue. 21:44
moritz_ from the outside, TPF is pretty opaque
[Coke] When in doubt, go to TPF, look at the who's who, and ask someone on the board.
moritz_: s/from the outside/
but they are very responsive to emails, IME, so ping them if there's a topic that is unresolved that you care about.
PerlJam I guess Karen Pauley administers the Hague money? 21:45
dalek kudo: 35c9071 | moritz++ | / (2 files):
"fix" rakudo on newest parrot, and bump PARROT_REVISION

Arguably this is more of a workaround than a fix, but it does work, and my limited debugging skills didn't find the right calling location to fix
  (PIR line numbers being nonsenical doesn't help either)		 21:46
tadzik sniff sniff, I smell testing
"Your branch is behind 'origin/master' by 9 commits, and can be fast-forwarded." 21:48
is there a way to check it manually?
[Coke] PerlJam: if not, she's on the board, and is an excellent resource.
dalek ast: 039f3b1 | (Kodi Arfer)++ | S02-names_and_variables/names.t:
[names.t] Added tests for RT #77528.
PerlJam [Coke]: well, www.perlfoundation.org/ian_hague_pe...ent_grants says "Evaluation of Ian Hague Perl 6 Development grants will be done by the president of TPF in consultation with other TPF members" 21:49
Kodi Hm, I pushed to Rakudo first, but dalek noted my change to roast first.
tadzik yeah. Just pulled and your change is there 21:50
moritz_ Kodi: the repos in the perl6/ organization on github push the changes to dalek, whereas it pulls from rakudo/rakudo
jnthn moritz_: Looks kosher.
moritz_: If incomplete.
21:51 Cyrus joined
Kodi moritz_: I see. 21:51
jnthn moritz_++
dalek kudo: 71328bd | KodiB++ | src/Perl6/Grammar.pm:
[Perl6/Grammar] Added many instances of "<!before '-'>" to permit calling subs named, e.g., "my-foo".
moritz_ Kodi: I think that's incomplete 21:52
Kodi: for one there's also ' that's allowed in identifiers
Kodi: and I think it should be a separate rule... I think STD.pm has <.nofun> for that, or so 21:53
tadzik parrot-nqp.c:1280: error: expected ',' or ';' at end of input 21:54
jnthn I guess this patch diverges us further from STD too?
sorear <!apostrophe>
tadzik ...does the new PARROT_REVISION compile for you?
sorear nofun is something slightly different
Kodi moritz_: Aha, I just saw the name "nofun" and assumed it only forbid parentheses.
moritz_ Kodi: listen to sorear, he knows STD better than me 21:55
tadzik: you might need at really hard clean in the parrot repo (at least I needed it recently) 21:56
sorear token nofun { <!before '(' | '.(' | '\\' | '\'' | '-' | "'" | \w > }
Kodi Vs. Rakudo's token nofun { <![ ( \\ ' \- ]> » } 21:57
21:59 nat2610_ left
tadzik moritz_: yeah, that's what I did and it helped 21:59
21:59 lestrrat left
diakopter $1M doesn't go very far in California 22:00
Kodi sorear: moritz_ was right; apostrophes remain a problem. If nofun works, think should I use that, or add a not-a-hyphen-or-apostrophe rule? Are there other punctuation characters allowed in identifiers? 22:01
diakopter at least compared to my previous homes (Texas & Kansas). The real cost of living is tremendously higher
22:02 masak left
sorear Kodi: the rule in STD for matching - or ' is called <apostrophe> 22:02
Kodi: so not - or ' would be <!apostrophe>
tadzik github.com/masak/proto/blob/master/PIONEER -- how about moving this to perl6/ecosystem? 22:03
moritz_ tadzik: +1 22:05
colomon +1 22:06
dalek osystem: 8620ed9 | tadzik++ | PIONEER:
Added PIONEER from the proto repo
 22:07
tadzik wham!
moritz_ tadzik++ # good ideas, and carrying them out 22:10
22:11 gbacon joined
tadzik moritz_++ # knowing how to motivate, and how to reward the effort 22:12
sbp "Proto cannot change your PERL6LIB environment variable" 22:13
why can't it?
tadzik NYI in Parrot
sbp thanks
tadzik oh, this document is proto oriented, mind me changing it to be more general?
As proto is not working now anyway
moritz_ would be nice, yes 22:14
(to be more general, that is)
22:15 masak joined
tadzik plenty of code to change after the done_testing => done change 22:15
masak oh, and remember, contestants: only a week remains of the contest. so send in your solutions!
submissions after the deadline will be considered to be after the deadline! :)
'night, #perl6. 22:16
22:16 masak left
tadzik :) 22:16
22:17 coldhead joined, Kodi left
tadzik hmm 22:18
should I resist the temptation to, uhm, standarize PIONEER a bit?
I know masak hates that word, but "we allow everything" leads to a stupid situations imho 22:19
colomon tadzik: you mean like multiple test files in every module I have?
tadzik colomon: hm? 22:20
colomon when temptation comes you should give right in.
"plenty of code to change after the done_testing => done chang"
tadzik colomon: I mean the lib6 situation a while ago
ah, yeah
btw, I think about implementing a bit of S22 in neutro and blogging about it 22:21
colomon Ah, I think getting rid of lib6 for now works for me.
tadzik I think the field in META.info telling the module installer where are the modules and where are the tests are a bit lacking sense in a broader view 22:22
22:29 kaare_ left, lestrrat joined
tadzik wklej.org/id/452447/ -- how do you like a META.info like this? 22:30
json, S22-alike, bla bla
moritz_ looks like a nice start 22:31
tadzik S22 proposes YAML, but even CPAN is moving to json iirc, plus we don't have a YAML parses :)
I'll try implementing it in neutro and blogging about it 22:32
it should have something that deps.proto doesn't have... oh wait
moritz_ S22 can be changed :-)
tadzik what do you think of something like this:
META.info will contain the repo url, or a tarball url, whatever 22:33
moritz_ or possibly both
tadzik our perl6/ecosystem will be a bunch of dirs, like json, neutro, etc
with its META.info in each one
moritz_ if you tell me what the META.info should contain, I could make a release of JSON::Tiny, create a tarball and provide that for testing purposes 22:34
tadzik now, we're fetching ecosystem with a module installer, we have names, descriptions, urls, deps, everything w/o even fetching the separate repos
moritz_: my humble hope is that the important and popular modules will be ported to META.info, so people will get interested 22:35
moritz_ which is why I offer to make a start 22:36
tadzik and look, we can get rid of this ugly perl6-Module-Tools thingy. It will only be a repo url
moritz_++
give me a few minutes to implement all this :)
moritz_ well, I'm going to bed now, so I'll have to do it tomorrow
tadzik oh, sure
can I make some mess in ecosystem in the meantime? :) 22:37
I mean, add some files and dirs
that's a big change after all
maybe I'll just make a fork
moritz_ do you want to add one directory per project?
tadzik si
moritz_ that doesn't seem right
tadzik why not?
moritz_ just have a list with URLs
because it's dependant information 22:38
that shouldn't need to be in a repo
tadzik well, one of the points was to keep META.info's in one place
hmm
moritz_ just make a list of URLs
and a perl script which fetches those URLs
tadzik so URLs to META.infos?
moritz_ and summarizes them into a collected file
tadzik I'll just make a fork and make a mess there
moritz_ which you can then download from modules.perl6.org
tadzik hmm, right 22:39
wklej.org/id/452456/ -- new info
moritz_ that's basically what is done now, except that the github API is used
instead of a META.info
tadzik I'll mangle a bit, and maybe post the effects to perl6-users@ later 22:40
thanks for support moritz_
moritz_ you're very welcome
bed &
22:43 [Coke] left 22:46 newbie joined, newbie is now known as Guest81110
Guest81110 hi, can someone tell me what is piranha? and how is it related to perl6? 22:48
sorear never heard of it 22:49
sbp where did you gear of piranha, Guest81110?
*hear of
22:52 [Coke] joined 22:56 MayDaniel joined
Guest81110 sorear: emm.. oh sorry I mean in rakudo source there are files with extension of .pir , what is that extension? I thought it was short for piranha which is a linux tool.. 22:59
pumphaus parrot intermediate representation, if I'm not mistaken
low-level parrot code, effectively 23:00
plobsing pumphaus: something to that effect
Guest81110 low level parrot code.. but is it written in perl6? 23:01
plobsing Guest81110: no. PIR is the language.
sbp Guest81110: en.wikipedia.org/wiki/Parrot_interm...esentation 23:03
that's the language
Guest81110 oh i see .. so there is pir and PASM.. but in rakudo they used pir.. 23:06
23:10 Guest81110 left 23:14 GinoMan_ left 23:21 MayDaniel left 23:22 GinoMan joined 23:59 hanekomu 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!