»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by moritz on 3 May 2013.
lue r: say (^10)».msb».base(2) 00:07
camelia rakudo cd5ca7: OUTPUT«0 1 1 10 10 10 10 11 11␤» 00:07
lue S32::Numeric details lsb and msb. They return the *position* of the most/least significant set bit. 00:13
diakopter what's this about samba? 00:15
lizmat good * from Austin, TX, #perl6! 01:53
diakopter HOWDY 01:54
grondilu say .WHAT given (my @ = ^10).list; 01:54
rn: say .WHAT given (my @ = ^10).list;
camelia rakudo cd5ca7, niecza v24-55-g7942a09: OUTPUT«(Array)␤»
grondilu is confused by .list returning an Array :/
rn: say .WHAT given list(my @ = ^10); 01:57
camelia rakudo cd5ca7, niecza v24-55-g7942a09: OUTPUT«(List)␤»
colomon lizmat: woah, you're there very early! 01:59
lizmat we like to see a bit of the surroundings before indulging in a conference :-) 02:00
slash hackathon
pmichaud lizmat: Howdy! Welcome to Texas! 02:05
lizmat hi pmichaud! 02:05
pmichaud I'm currently about 700km (ahem, we use *miles* here, so 435 mi) north of you. 02:05
lizmat well, that's more than we drove today… 02:06
pmichaud oh, wait.
lizmat if you'd be in Amsterdam, we'd be in mid France
pmichaud that's round-trip distance.
350km (217 mi)
lizmat ah. that's more like: if we would be at home, you would be in Groningen… about the max distance you can do in the Netherlands :-) 02:07
pmichaud right.
if you go due south, then in about 325 miles you finally hit the Mexico border.
lesson: Everything's Big in Texas. :) 02:08
anyway, glad you arrived safely.
I look forward to seeing you this weekend.
colomon I'll get there Saturday around dinner time, I think. 02:09
pmichaud I don't know exactly when I'll arrive on Saturday, I'd like to get an early start, but I have to make sure everything is well-set at home before departing.
I think I'll try to arrive by 8pm or so, which is when I think jnthn/masak/others will likely be arriving. 02:10
lizmat looking forward to that!
pmichaud I might not be able to get there until midnight, though.
pmichaud I'll /msg folks my cell # so you can call and find out where I'm at. 02:11
or maybe I'll find a way to send my current location via twitter or a location app or something.
pmichaud checks weather 02:13
lizmat ~34±1
pmichaud yeah
looks like the forecast got warmer since I last looked 02:14
but Texas weather is very unpredictable.
as of two days ago, we were forecast for a sunny weekend, now we're forecast for rain+storms on the weekend
btw, if you find nothing to do on Saturday, the Texas Linux Fest is taking place in Austin on Saturday. 2013.texaslinuxfest.org/ 02:16
colomon is planning on going to the pub to play some Irish tunes on Sunday night. 02:30
thesession.org/sessions/269 02:33
sorear lookin forward to seein y'all 02:34
i'm arriving at 1705 in AUS
Sat
diakopter tries to remember the times and who of the 8 people I offered to pick up at the airport Saturday 02:35
diakopter gives up immediateliy
sorear I'm arriving with gdey 02:36
colomon sorear: 1732 for me. 02:40
sorear ponders sticking around and waiting for people
lizmat fwiw, we have a minivan, max 7 people… so we could pick up at least the people staying at the Doubletree 02:42
afk to do some shopping 02:43
sorear I have no idea what gdey's plans for post-arrival are 02:44
dalek ar: 7b615e3 | pmichaud++ | modules/perl6-http-easy:
Update HTTP::Easy module for 2013.05 release.
 02:49
diakopter lizmat: heh, I have a minivan too 02:52
from Alamo
pmichaud The Alamo is 79 mi (127km) from the Doubletree in Austin. :-P 02:55
Remember that. :-P
sorear the city or the theatre? 02:57
pmichaud the monument
the mission in downtown San Antonio
sorear I see 03:01
pmichaud the site of the battle that inspired the "Remember the Alamo" phrase
you know... "The Alamo". :-)
sorear yeah, "remember the alamo" is basically all I remember from that day in grade-school US history :[ wikipedia to the rescue! 03:02
shachaf Remember "Remember the Alamo!"!
pmichaud don't worry about it too much; I had a girlfriend from Sacramento who was visiting San Antonio with me, and she saw a sign with directions to "The Alamo" and asked... "oh, is there a plaque or something there?" :-P 03:03
(note to non-Texans: It's a Spanish Mission with a chapel (the iconic building) and tours. 03:05
diakopter and it's backwards/inside-out from what you'd expect to be fought from behind 03:06
that was the best english sentence evar. 03:07
dalek ar: 6bab91c | pmichaud++ | tools/star/release-guide.pod:
Add myself to release history; update release-guide reminder for that.
 03:23
dalek ecs: 20dd0dc | (Brent Laabs)++ | S32-setting-library/IO.pod:
S32::IO fix layout, bump version
 03:55
pmichaud Rakudo Star 2013.05 released. 04:04
sorear "\o/ " x 20 04:07
bonsaikitten /o\ 04:09
labster \\o// 04:11
dalek ar: a78540d | pmichaud++ | tools/star/release-guide.pod:
Add Wikipedia to the list of places to update.
 04:16
atroxaper WoW WoW!!! 04:18
Even for windows! Thanks a lot, pmichaud! 04:20
lizmat diakopter: great minds think alike :-) ours is also from Alamo :-)
and now off to bed, it was a looooong day
atroxaper Is it right that new R* use Parrot 5.2 like described in rakudo.org/2013/05/31/rakudo-star-2...released/? I thought it will be 5.3. 04:36
sorear 1. R* uses stable parrots 04:39
2. we've had a lot of issues with an IO refactor that parrot did, and for a while (possibly extending to now?) we weren't able to upgrade 04:40
atroxaper I just thought that github.com/rakudo/star/commit/1631...92ce3a1223 make R* is used 5.3 Parrot. Am i wrong? 04:44
pmichaud oh, it is 5.3.0 05:07
the announcement is wrong. 05:08
fixing.
I thought I updated the announcement, but apparently I failed.
tadzik oooh. We should have a "wars" and a "trek" branches in star
so dalek would report that as star/trek
(good morning, #perl6)
dalek ar: 56fe546 | pmichaud++ | docs/announce/2013.05.md:
Correct announcement to read Parrot 5.3.0 .
 05:10
pmichaud we can also have "gate", "crossed", "man", etc. :-) 05:12
tadzik now we only need to find a purpose. Or rather, excuse :) 05:14
pmichaud oh, if I thought there'd be enough of them, we could use them as release nicknames.
like we do .pm cities for the compiler
sorear hey gdey.
pmichaud star/kindustries ;_0 05:15
pmichaud still hasn't seen Iron Man 3 yet :-/ 05:16
nor Star Trek: Into Darkness
tadzik same here 05:17
but I think they're both available in 3d-only
pmichaud oh, I'm sure they're available 2d here.
yeah, 2d here for ST:ID 05:18
and IM3
(3d is available, but 2d is available too)
tadzik ha, ST is available apparently 05:19
in 2D
Iron Man isn't though
gdey hey sorear
tadzik ponders going to see ST today 05:21
sorear looks like DFW has free wifi so we can discuss this later if needed
but diakopter and lizmat are apparently running van pools out of AUS, do you have plans on that front?
pmichaud bedtime -- back tomorrow. 05:27
Teratogen night
atroxaper night 05:28
FROGGS o/ 06:51
atroxaper 0/ 06:52
sorear ō 06:56
labster ó 07:16
Teratogen christ on a bun! 07:35
hoelzro new Rakudo *! 08:10
\o\
/o/
unfortunately building it has killed my VM =( 08:11
at least I wasn't building it on a real machine =) 08:12
bonsaikitten worksforme
hoelzro: ulimit to the rescue ;) 08:13
hoelzro I thought ulimit didn't work for memory on Linux? (I assume that's the issue)
ouch; I even gave the VM 2GB of RAM
moritz why shouldn't ulimit work? 08:15
you just have to be careful which memory you limit (hint: ulimit -v is most often what you want)
bonsaikitten it does
ulimit -m 256000 # 256 metric Mibblibytes 08:16
and if you have multiple processes ... cgroups are fun (although the memory cg is usually disabled)
hoelzro I thought I read somewhere that ulimit doesn't work for memory usage on limit 08:33
but maybe that's out of date
*on Linux
bonsaikitten ??
moritz hoelzro: maybe you've read that, but it's wrong :-) 08:42
hoelzro it's something I'm happy to be wrong about =)
either way, my VM is quite angry.
and I *really* don't want to pull the plug =(
masak morning, #perl6 09:03
atroxaper masak, morning! But in my Siberia is 6am already ^^ 09:05
Not 6, but 4am ) 09:06
I'm not used to using short time-style. 09:07
jnthn
.oO( nor am vs pm )
atroxaper Oh... It's lol! am == pm) 09:08
bonsaikitten 5pm here in China
masak waves to atroxaper-in-the-future 09:15
atroxaper advises masak to be careful with time travel 09:19
moritz it's ALWAYS morning in Universal Greeting Time!
masak moritz++ 09:20
atroxaper :D 09:21
smash pmichaud++ # rakudo release 09:38
tadzik smash! \o/
smash tadzik: o/ 09:39
hoelzro damn 09:42
my package building script for Rakudo * doesn't work with 2013.05 =( 09:43
hoelzro what's the deal with 721f127? 09:59
(on nqp)
that commit threw out code that was necessary for building packages
is jnthn around? 10:01
colomon Looks like three modules (File::Direcotry::Tree, IO::Path::More, and Test::Junkie) started working in the last two days! 10:27
FROGGS O.o
just like that?
colomon FROGGS: well, I fixed Test::Junkie. Dunno what happened to the other two. 10:28
File::Directory::Tree was blocking on IO::Path::More.
Looks like labster++ made changes to IO::Path::More. 10:29
masak wow, www.cs.ucf.edu/~dcm/Teaching/COT48...Backus.pdf is a really interesting observation, considering that it comes all the way from 1978. 12:10
much of what it points out as disadvantages is still very much "unsolved".
well, unless you're a full-time Haskell programmer. but most of us aren't.
there are some silver linings even in the industry: things like LINQ, and Rx, and various non-mainstream languages (Scala, CLojure, F#) nibbling away at how to do things nicely with FP. 12:12
mathw hybrid languages are where it's going to be for a good while I think 12:24
there are some astonishingly powerful ideas floating around in the Haskell world these days, but I'm not convinced Haskell itself is the right language to bring those wonderful things to the world at large with 12:25
I hope to see Microsoft do more with F#, the level of tooling in Visual Studio for it really relegates it to a second-class citizen
mathw They don't bother providing tools for things like ASP.NET MVC projects in F# because their position is that it's not meant for that kind of work. Why, I ask, can't it be? 12:26
It does support all the necessary behaviours after all
masak *nod*
mathw It's still cheating, but it's closer...
I must admit to really missing Haskell typeclasses though 12:27
wonderful, wonderful things
It's like if C# let you add an interface implementation to an object without having the ability to modify its class
and yes there are extension methods, but they're not the same either
I want... roles! 12:28
masak Go seems to do something similar-ish with its interfaces.
it's an idea whose time has come.
inheritance hierarchies are so 20th century. :) 12:29
mathw definitely 12:31
I use inheritance less and less
heck, I use objects less and less
my C# code is now strewn with static classes
objects have their place, but so much can be done with other ways and it seems to work better 12:32
if I have to make an instance of some object in order to do something like a data transform I tend to start turning my nose up
and plotting to write a wrapper function that hides the mess
mathw probably the thing that distresses me most in C# programming is where you're in the shiny new world of LINQ and pseudo-functional programming and you run headlong into a library you can't avoid that's designed like something from Java 1.2 12:36
diakopter holds his breath 14:21
masak calm before the storm... :)
moritz diakopter: breathe again, please! 14:26
masak .oO( moar breathing plz ) 14:29
grondilu DUMP is not documented yet? Will it be a full ASCII serialization? 14:30
r: say <foo bar>.DUMP 14:31
camelia rakudo cd5ca7: OUTPUT«Parcel<1>(:$!storage(RPA<2>(␤ "foo",␤ "bar"␤)))␤»
moritz grondilu: I guess it will mostly be a debugging tool
grondilu ok 14:32
timotimo r: say (1, 2, 3, 4).>>WHAT 14:38
camelia rakudo cd5ca7: OUTPUT«===SORRY!===␤Confused␤at /tmp/dBzLDn0SWI:1␤------> say (1, 2, 3, 4).⏏>>WHAT␤ expecting any of:␤ dotty method or postfix␤»
timotimo :(
r: say (1, 2, 3, 4)>>.WHAT 14:39
camelia rakudo cd5ca7: OUTPUT«(Parcel)␤»
timotimo :(
moritz r: say (1, 2, 3, 4).map: *.WHAT
camelia rakudo cd5ca7: OUTPUT«No such method 'count' for invocant of type 'Whatever'␤ in method reify at src/gen/CORE.setting:5783␤ in method reify at src/gen/CORE.setting:5700␤ in method reify at src/gen/CORE.setting:5700␤ in method gimme at src/gen/CORE.setting:6131␤ in method eager at s…
moritz can't prime macro methods, it seems
grondilu rn: say (^5)».WHAT
camelia niecza v24-55-g7942a09: OUTPUT«(Int) (Int) (Int) (Int) (Int)␤»
..rakudo cd5ca7: OUTPUT«(Range)␤»
[Coke] r: say (1, 2, 3, 4).map: { $_.WHAT } 14:46
camelia rakudo cd5ca7: OUTPUT«use of uninitialized value of type Int in string context in block at /tmp/e1dukKeIVT:1␤␤use of uninitialized value of type Int in string context in block at /tmp/e1dukKeIVT:1␤␤use of uninitialized value of type Int in string context in block at /tmp/e1dukKeIV…
Ulti is there a tutorial for using perl6 to make your own DSL? something a bit like this video of using Python's parsimonious grammar engine www.youtube.com/watch?v=1h1mM7VwNGo 14:51
tadzik I think there was an advent calendar post 14:54
also a masakism task
Ulti perl6advent.wordpress.com/2009/12/2...d-actions/ I guess this 14:59
diakopter passed out from no oxygen 15:00
tadzik Ulti: perl6advent.wordpress.com/2012/12/2...-y-things/ 15:01
that's what google "perl 6 dsl advent" said, at least ;) 15:02
Ulti yeah thats not really what I mean, its not parsing anything you've just made some method calls that look like the language you wanted and its executed as perl6 immediately 15:08
p. neat though
tadzik well, what kind of tutorial would you want with that? 15:10
seems pretty straightforward :)
Ulti like if you defined a custom string with a sub language tied to a grammar that parsed the _ and O to create a tree of execution for all that it would be what I mean.. obviously over kill for that task 15:11
I assume thats how regex is implemented in perl6?
with regex being a sublanguage with the perl5 and perl6 versions having their own grammar classes 15:12
Ulti wanders off to finish his thesis so he can actually play with perl6 properly later ;___; 15:14
masak TimToady: rosettacode.org/wiki/Caesar_cipher#Perl_6 is yours -- I checked. I'm not 100% sure this is an error, but I'm wondering if strdup() really outputs the string, as the output seems to imply. 15:46
ergh, bad link, sorry. meant rosettacode.org/wiki/Call_a_foreign...ion#Perl_6
jnthn masak: I think the puts does the output 15:48
And then returns a result code which is on the next line
masak oh, d'oh! 15:50
yes.
masak "Here we arbitrarily show as many lines of output as there are lines in the program. :-)" -- hah! rosettacode.org/wiki/Count_in_octal#Perl_6 16:24
diakopter wakes up from passed out, and starts holding his breath again 16:27
Ulti masak: this is a cleaner implementation rosettacode.org/wiki/Count_in_octal#LOLCODE 16:29
masak Ulti: :) 16:30
Ulti I can only imagine octocat from github saying that when I read it
<octocat> IM IN YR octalizer 16:31
masak revels in the pure anachronicity of rosettacode.org/wiki/Create_a_file_...ape#Perl_6
pmichaud anyone know how to find the number of RT tickets closed in the perl6 queue since Jan 1? 16:33
Ulti masak: the Racket code below that is even better
geekosaur oh, hm, need an account to do a real search 16:34
masak pmichaud: Advanced Search, "Resolved", "After", "2013-01-01". 16:35
FROGGS pmichaud: 144
I did "resolved after 2012-12-31" though 16:36
pmichaud maybe included rejected, also
masak yeah, also getting 144 resolved.
FROGGS it does, yeah
Ulti rosettacode.org/wiki/Parse_EBNF#Perl_6 wow a grammar of a grammar
FROGGS it maybe includes reopened ticket, though I dont think we did that
masak we did. once. 16:37
FROGGS tickets also*
masak or if there were more than one, we closed them again.
"This example is **incorrect**. It does not accomplish the given task." -- anyone know/see why? rosettacode.org/wiki/Currying#Perl_6 16:39
(I don't.)
geekosaur the "What is the task?" on the discussion page may be relevant 16:43
(argubaly, assuming is *not* currying, it is partial application. Currying might be closer to using a capture to provide arguments? 16:44
masak yeah, the whole task description needs to be clarified. 16:45
colomon masak: I thought we worked out the other day that assuming(*, 0) doesn't actually work? 16:54
at least, doesn't work in Rakudo or Niecza. 16:55
masak I, uh, I'm behind on my backlogging...
lizmat knows the feeling 16:56
[Coke] r: sub f($a, $b) { $a - $b } ; say .(1) for &f.assuming(0), &f.assuming(*, 0) 16:57
camelia rakudo cd5ca7: OUTPUT«-1␤Too many positional parameters passed; got 3 but expected 2␤ in sub f at /tmp/v1K6vRN4Fd:1␤ in sub CURRIED at src/gen/CORE.setting:2069␤ in block at /tmp/v1K6vRN4Fd:1␤␤»
colomon frequently gives up completely on backlogging, for what it's worth. :) 16:59
r: sub f($a, $b) { $a - $b } ; say .(1) for &f.assuming(0) 17:01
camelia rakudo cd5ca7: OUTPUT«-1␤» 17:01
colomon r: sub f($a, $b) { $a - $b } ; say .(1) for &f.assuming(*,0)
camelia rakudo cd5ca7: OUTPUT«Too many positional parameters passed; got 3 but expected 2␤ in sub f at /tmp/hEN7DT0JN6:1␤ in sub CURRIED at src/gen/CORE.setting:2069␤ in block at /tmp/hEN7DT0JN6:1␤␤»
colomon r: sub f($a, $b) { $a - $b } ; say .() for &f.assuming(*,0)
camelia rakudo cd5ca7: OUTPUT«Cannot call 'Numeric'; none of these signatures match:␤:(Mu:U \v: Mu *%_)␤ in method Numeric at src/gen/CORE.setting:865␤ in sub infix:<-> at src/gen/CORE.setting:2987␤ in sub infix:<-> at src/gen/CORE.setting:2985␤ in sub f at /tmp/xFdmbq7FDR:1␤ in sub CURRIE…
lizmat afk for some sight seeing 17:08
masak lizmat: o/ enjoy! 17:09
masak gently suggests jnthn's one-liner solution should be in rosettacode.org/wiki/Execute_Brain****/Perl_6 17:25
[Coke] wonders if next R* will have some jvm. 17:27
[Coke] hasn't kept up with recent commits, and should try to at least build stuff this weekend. 17:27
[Coke] also wants to get rakudo-jvm into the daily roasts sooner than later. 17:28
PerlJam [Coke]: were you going to YAPC this year? 17:35
masak Teratogen: (topic lifted over from #masakism): is Perl 6 becoming an academic language? no. it's not done primarily in academia. nor is it shifting into academia. did you have some other definition of "academic language"? 17:39
Teratogen masak! 17:40
=(
nwc10 [Coke]: jnthn.net/papers/2013-plpw-rakudo-jvm.pdf alledges on page 52: The June compiler release of 17:42
Rakudo
will be the first
gee, thanks firefox
with some level of JVM support
nwc10 one line would have been better 17:42
[Coke] PerlJam: I cannot justify the cost, no. 17:43
colomon [Coke]: :(
PerlJam [Coke]: That's what I thought. I will be sad to miss you :( 17:44
jnthn nwc10: That's mostly just a branch merge at this point, alas I've not really been in a fit state to see to it in the last few days. I'll hopefully manage at YAPC.
nwc10 jnthn: get well soon
I've been out with a bad cold for most of the previous 3 days 17:45
jnthn If only that's all that was with me... :/
But anyway, get well
[Coke] FYI, getting something like mojo6 running on the jvm would certainly help me justify the cost a bit. :) 17:45
hugme: hug jnthn 17:46
hugme hugs jnthn
dalek rl6-roast-data: 502b111 | coke++ | / (4 files):
today (automated commit)
[Coke] all three implementations now dirty. 17:48
pmichaud I don't expect R* to have jvm until july at earliest.
with emphasis on "at earliest"
jnthn pmichaud: I've had R* more like September 17:49
pmichaud jnthn: that works for me, too.
[Coke] once it's at a point it can run *any* spec tests, I'll probably add it to the roast figures.
pmichaud R* wants stability more than anything else, and it's not just the compiler but the modules that would want some working out.
[Coke] then you can get a pretty picture of what happens when you add some bit. 17:50
pmichaud we'll likely spend a good part of midsummer figuring out what modules R* really ought to have.
jnthn [Coke]: Getting into the daily roast, once it can do that, would be great. 17:52
colomon is hoping YAPC::NA hackathon includes some stints of getting the core up and running on jvm...
pmichaud the YAPC::NA hackathon is going to be interesting on many levels. 17:53
jnthn
.oO( The hackathon rooms are on differnet floors? )
 17:54
nwc10 will there be a downloadable Rakudo JAR soon? :-)
jnthn colomon: I expect to spend some tuits on Rakudo JVM, for sure. 17:54
colomon jnthn: didn't specifically mean you. or is it not ready for the rest of us to hack on yet? 17:55
jnthn colomon: Ah. I suspect there'll be LHF-ish things. 17:56
There's plenty of tricky things, though.
colomon not that I've got anything against you working on it, mind you. ;)
PerlJam wonders how long before we have other Rakudo distributions (e.g., biorakudo :)
jnthn colomon: Well, I do prefer it when things are not depending one me, though. :)
[Coke] if hacking on jvm-for-rakudo, which branches do we want? 18:04
jnthn jvm-support in Rakudo, rak-jvm-support in NQP
hoelzro jnthn: I saw that you removed a commit of mine from NQP =( 18:20
er, reverted.
jnthn hoelzro: Does it look intentional or accidental? If intentional, the revert commit should say why. 18:21
hoelzro hang on
digging it up =)
721f127
nwc10 "Toss code that does nothing."
use more tests!
hoelzro reverting that broke packaging =(
dukeleto o/ 18:33
pmichaud o/ 18:34
masak o/
diakopter is very dead from lack of oxygen 18:39
colomon \o 18:40
masak diakopter: dude, we told you to breathe hours ago.
[Coke] wonders if diakopter is waiting for someone to ask what the problem is. 18:42
nwc10 diakopter: have you trained up your replacement yet? If not, you're not permitted to die. 18:44
FROGGS thinks that everyone is silent because of working on the slides 19:15
diakopter prolly not that
FROGGS (for a specific group of ppl called everyone) 19:16
flussence
.oO( they're all holding their breath )
diakopter *gasp*
FROGGS :o)
masak is holding his breath too!
...as if there will be something MOAR happening tonight... 19:17
jnthn blogged 6guts.wordpress.com/2013/05/31/moar...nd-rakudo/
masak jnthn++! \o/
FROGGS O.o
masak ok, it's ON. 19:18
TimToady starts holding his breath now...
diakopter graphs page is awesome 19:20
[Coke] crosses sixparrot off his ever-growing todo list. 19:21
diakopter less awesome graph: i.imgur.com/7ZAFb7U.png NO MOAR PEOPLE PLZ 19:22
masak diakopter: avoid success at all costs. :P
diakopter github.com/MoarVM/MoarVM/contributors
masak yay, the camel has two humps, and so does masak! :P 19:23
FROGGS wow 19:26
FROGGS has just finished reading
masak lol, I blogged! strangelyconsistent.org/blog/a-vm-for-6model
TimToady wow, you write fast :)
masak as a Perl 6 historian, I have to stay one step ahead! :P 19:27
colomon wow
Teratogen I liked when Perl 5 had an option to call the c preprocessor
colomon has not finished reading
Teratogen I think it was perl -P
but they got rid of it =(
Teratogen likes the simplicity and elegance of the C Preprocessor 19:28
masak Teratogen: you've harped on this topic before: irclog.perlgeek.de/perl6/2012-02-19#i_5174077
Teratogen: many times, in fact. 19:29
Teratogen my conscience
=)
masak Teratogen: I think we can take it as established now that you like the simplicity of the bloody C preprocessor.
[Coke] we can also take it as granted that masak is constantly searching the backlog to beat people over the head with it. :)
masak I told you, I'm a Perl 6 historian. 19:30
flussence some people like the simplicity of using regex on arbitrary html...
diakopter argh.
I accidentally it an announcment
masak I find patterns in things.
[Coke] wonders if there is a perl6 program that is beeping at masak as he complains about how many times he's had to rewrite partcl from the ground up. :)
colomon 's mind has been blown
FROGGS .oO( I! *hit* told! *hit* you! *hit* ... )
Teratogen Parsing HTML with Perl Regexes for Fun and Profit
I should write that book
PerlJam Teratogen: tchrist already beat you to it 19:31
Teratogen wow blast from the past
PerlJam (Well, almost anything he writes resembles a book anyway ;)
Teratogen tchrist? that is efnet #perl circa 1996
[Coke] supposes he will have to submit a patch for moarvm shortly.
Teratogen the apostrophic jihad!
where did everyone in efnet #perl go to? 19:32
they all up and left one day
PerlJam Still there.
Teratogen I guess one too many takeovers
PerlJam (I'm still on it in fact)
Teratogen wonder if I am still banned there
diakopter [Coke]: pull request currently I think
PerlJam no traffic really, but every once in a while it perks up
PerlJam Teratogen: I really didn't think you and I had been interacting for so long. 19:33
masak r: loop { say "jnthn++" } 19:34
camelia rakudo cd5ca7: OUTPUT«(timeout)jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤…
FROGGS hehe 19:35
PerlJam masak: btw, I like how every time someone mentions JIT, it's always in the future.
colomon PerlJam: I'm sure we'll get it done just in time.
PerlJam colomon++ :) 19:36
JIT is like flying cars.
masak PerlJam: let me just say that based on what I know about Parrot and Moar, my personal priors are higher that there'll be a useful JIT this time.
quite a bit higher, in fact.
kbenson1 So, I submitted jnthn's blog to HN in case anyone has an account and wants to vote it up or comment: news.ycombinator.com/item?id=5800775 19:38
diakopter r: say 'jnthn++'for^Inf
camelia rakudo cd5ca7: OUTPUT«(timeout)jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤jnthn++␤…
PerlJam masak: There's a select group of people I have in mind that if someone from the list says "I/we will create X", my confidence that X will actually exist at some point is in the area of 90%.
masak upvotes 19:39
kbenson1 getting it past the initial hump of obscurity (WRT HN )might yield a nice discussion there
masak PerlJam: aye.
kbenson1++
PerlJam oddly, most of those people are on #perl6 ;) 19:40
PerlJam upvotes
colomon too 19:42
masak 13th place on the main page.
kbenson1 It was 16th with 4 points, and is 13th with 5. A few more within the next couple of minutes and it will probably reach the top 5, if not #1. 19:43
colomon MoarVM doesn't build on darwin yet. :( 19:45
kbenson1
.oO( I never realized it was so easy to seed a from page event )
[Coke] colomon: working on that.
smace saying hi to everybody (Perl stranger)
PerlJam smace: greetings! 19:46
PerlJam smace: What's your interest in Perl 6? Want to contribute? Or do you just want to lurk and learn? 19:47
smace: or ... are you here by mistake? ;)
smace Lurk and learn, convinced by friend.
kbenson smace: That sounds like a good friend. You should keep 'em. :) 19:48
PerlJam smace: excellent! and what kbenson said too :)
smace Have some background in Python, but he told me about great community out there ;))
GlitchMr Well, relatively small, but active community :-). 19:50
Ok, perhaps not "small".
(but compared to RoR or Python community, it is small)
PerlJam GlitchMr: all of the Perl 5 community is part of the Perl 6 community, only most of them don't know it yet ;) 19:51
Teratogen Perl 5 is under ferocious development right now
what are they up to, Perl 5.12?
colomon 5.16 19:52
Teratogen christ on a bun!
kbenson 5.18
nwc10 higher!
er, higher than 5.16
GlitchMr The problem with Perl 5 is that nearly nobody understands its source code.
colomon apparently needs to upgrade
TimToady well, it has a couple other problems as well...
masak smace: welcome!
kbenson It's a regular release cycle now, so there's less wow features per release, but more steady improvement at a scale you can actually appreciate on a regular basis 19:53
rjbs I am reading 6guts.wordpress.com/2013/05/31/moar...nd-rakudo/
GlitchMr I've 5.18.0 on my computer, but 5.10.1 on my VPS 19:53
rjbs Oh, nevermind. My question has answered itself. 19:53
masak \o/
rjbs (I couldn't find the author, btu now I see that 6guts is *always* jnthn)
flussence 5.18 is the one where they realised most of the backported Perl 6 stuff doesn't make sense and added E_DEPRECA...^W on-by-default warnings to all of them, right?
GlitchMr Yeah
PerlJam rjbs: and here I thought you were like me annoyed that you had to scroll all the way to the bottom to get at the github URL :) 19:54
GlitchMr ~~, given, when, ...
But not say.
smace So, better choice to dive into 6? 19:54
shinobicl rakudo: my @suits = <♣ ♢ ♡ ♠>;my @ranks = 2..10, <J Q K A>;my @deck = @ranks X~ @suits;my %points = @deck Z @( (2..10, 10, 10, 10, 11) Xxx 4 );@deck .= pick(*);my @hand = @deck.splice(0, 5);say ~@hand;say [+] %points{@hand};
camelia rakudo cd5ca7: OUTPUT«5♢ 4♡ A♢ 3♠ 10♡␤33␤»
GlitchMr Oh, and my $_ has E_DEPRECATED now too.
PerlJam smace: "better" along which axes? ;) 19:55
flussence smace: do both. If you get annoyed with one, you can skip it and come back later
rjbs GlitchMr: It's experimental, not deprecated, meaning we're hoping to "fix" rather than remove it.
masak smace: Perl 5 gets you "done" faster. Perl 6 is fun in indescribable ways.
smace PerlJam, just for fun and love ;)
diakopter updated my talk to reflect it: www.yapcna.org/yn2013/talk/4734
GlitchMr Not sure if ~~ even could be fixed. 19:55
masak diakopter++
PerlJam smace: oh, that's *definitely* perl 6
GlitchMr And my $_ simply doesn't make sense. 19:56
rjbs diakopter: Glad to finally know what that talk was/is. 19:56
PerlJam rjbs: Are you hoping to pick up sane smartmatch semantics here? ;)
rjbs PerlJam: No, I'm quite happy with the ones we sorted out last year. 19:56
masak loves guerilla-injecting the 蝶 tag into all the conferences he goes to :>
PerlJam rjbs++ good deal. I stopped paying attention when p5p went crazy over smartmatch semantics. 19:57
[Coke] .u 蝶
yoleaux No characters found 19:57
TimToady wondered briefly what a "leaf bug" was, then said "duh"
GlitchMr my $_ doesn't make currently any sort of sense. I would expect it to either fail, or remove magic.
rjbs GlitchMr: /me will resist the urge to "talk five." 19:58
[Coke] colomon: nearly there.
TimToady well, "flat bug"
colomon [Coke]++
kbenson diakopter: I was unsure as to whether to go to the perl6 vm talk, but I guess it's non-optional now. ;)
GlitchMr But then, somehow my $_ has magic in Perl 6.
masak TimToady: :D 19:59
GlitchMr But $_ doesn't mean the same thing in Perl 5 and Perl 6.
TimToady [Coke]: 蝶 蝶 [die2] /butterfly/
masak TimToady: been doing it for years, but only been attending YAPC::EUs and European workshops.
tadzik good evening #perl6 20:00
masak tadzik! \o/ 20:00
TimToady GlitchMr: $_ is *always* lexical in Perl 6 20:00
masak tadzik: you... need to backlog. :)
colomon tadzik: o/
tadzik what the hell happenend
colomon masak: bit of an understatement there! :)
masak I know :D
GlitchMr The difference is that in Perl 5, you write "say", in Perl 6 ".say".
colomon tadzik: masak has been hiding something from us.
tadzik there are things in my RSS that are sort of puzzling
masak channel is all a-frothy!
masak colomon: I so have! 20:00
GlitchMr The dot before "say" is explicit way of saying "$_.say".
tadzik this guy keeps doing that
GlitchMr But in Perl 5, $_ is explicit. 20:01
tadzik I guess I'll read stuff first
masak it wasn't my sekkrit project this time, though.
GlitchMr implicit*
masak waits for tadzik to get far enough to come back and go "OMG" 20:02
Moar is on 8th place on the HN main page now. 20:03
masak .oO( MOAR UPVOTES )
tadzik oh 20:05
masak 6th place.
tadzik: "...my god"? :) 20:06
moritz just upvoted it
FROGGS masak: done
tadzik the heck did I read
espadrine rn: 0, {+!*} ... * 20:07
camelia rakudo cd5ca7, niecza v24-55-g7942a09: ( no output )
diakopter tadzik: you're making me lolololololol
tadzik I must say I'm very much surprised 20:08
espadrine rn: say 0, {+!*} ... * 20:09
colomon tadzik: me too. in a good way, mind you.
camelia niecza v24-55-g7942a09: OUTPUT«(timeout)»
..rakudo cd5ca7: OUTPUT«0 ...␤»
dalek nqp/rak-jvm-support: c3cdeb6 | (Tobias Leich)++ | src/QRegex/P (2 files):
nqp/rak-jvm-support: enable regex special characters as delimiters
nqp/rak-jvm-support:
nqp/rak-jvm-support: Special chars like quantifiers or control characters can be used
nqp/rak-jvm-support: as delimiters. If they are used as delims they cant be used for
espadrine shouldn't (0, {+!*} ... *) be the same as (0, 1, {$^b;$^a} ... *)? 20:10
moritz espadrine: {+!*} looks wrong to me 20:11
colomon espadrine: what are you trying to do?
masak espadrine: don't think so.
moritz espadrine: * creates a closure, and { } also creates a closure
masak espadrine: it's hard to read when you're writing the stuff so crowded, with '+!*'. what do you want that to mean?
moritz so, you have a closure that, when invoked, returns a closure
colomon espadrine: {^$b;$^a } takes two arguments, { +!* } only one
masak r: say +!5
camelia rakudo cd5ca7: OUTPUT«0␤» 20:12
colomon oh, right, only non.
none
espadrine I thought * served as an argument; I'm trying to convert 0 to 1 and back
moritz nah, {...} always takes an implicit $_
even if you don't use it
masak espadrine: +!$x means "boolify, negate, numify".
espadrine: that is, &prefix:<+>(&prefix:<!>($x))
moritz r: say (0, +|*, ... *)[^10]
camelia rakudo cd5ca7: OUTPUT«===SORRY!===␤Comma found before apparent series operator; please remove comma (or put parens\n around the ... listop, or use 'fail' instead of ...)␤at /tmp/dBJrQLVDyj:1␤------> say (0, +|*,⏏ ... *)[^10]␤»
moritz r: say (0, +|* ... *)[^10] 20:13
camelia rakudo cd5ca7: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&prefix:<|>' called (line 1)␤»
moritz r: say (0, +!* ... *)[^10]
camelia rakudo cd5ca7: OUTPUT«0 1 0 1 0 1 0 1 0 1␤»
moritz r: say (0, 1, {$^b, $^a} ... *)[^10]
camelia rakudo cd5ca7: OUTPUT«0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1␤»
espadrine moritz: that's what I want!
moritz r: say (0, 1, {$^b, $^a} ... *).list[^10]
camelia rakudo cd5ca7: OUTPUT«0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1␤»
moritz r: say (0, 1, {$^b, $^a} ... *).flat[^10]
camelia rakudo cd5ca7: OUTPUT«0 1 1 0 0 1 1 0 0 1␤»
moritz espadrine: I'd write that as ( (0, 1) xx * ).flat 20:14
espadrine: a bit more readable, IMHO
especially more readable than +!*
colomon r: say ( (0, 1) xx * ).flat[^10]
camelia rakudo cd5ca7: OUTPUT«0 1 0 1 0 1 0 1 0 1␤»
espadrine how is {+!*} interpreted then? (purely for understanding) 20:15
moritz espadrine: a closure that returns a closure when invoked
espadrine moritz: oh, I see. thanks!
moritz r: my $cc = { +|* }; my $c = $cc(); say $c(1); say $c(09;
camelia rakudo cd5ca7: OUTPUT«===SORRY!===␤Unable to parse expression in argument list; couldn't find final ')'␤at /tmp/Qq1hExmovY:1␤------> * }; my $c = $cc(); say $c(1); say $c(09⏏;␤ expecting any of:␤ postfix␤Other potential difficulties:␤ Le…
moritz r: my $cc = { +!* }; my $c = $cc(); say $c(1); say $c(09;
camelia rakudo cd5ca7: OUTPUT«===SORRY!===␤Unable to parse expression in argument list; couldn't find final ')'␤at /tmp/ywY1mMhh4j:1␤------> * }; my $c = $cc(); say $c(1); say $c(09⏏;␤ expecting any of:␤ postfix␤Other potential difficulties:␤ Le…
moritz r: my $cc = { +!* }; my $c = $cc(); say $c(1); say $c(0); 20:16
camelia rakudo cd5ca7: OUTPUT«0␤1␤»
moritz isn't very concentrated anymore
espadrine :)
moritz 8 hours work + 3 hours doing taxes and still not finished :( 20:17
jnthn awww :( 20:18
moritz well, have to do the rest tomorrow-ish 20:19
(today is deadline, but they usually take about two weeks to send a demand note, and the first one is free :-)
Teratogen moritz do what I do, always do a 1040EZ 20:20
and claim 1 deduction
or 0 deductions
you will always get money back 20:21
TimToady Teratogen: he's not a USian
Teratogen oh!
well then!
TimToady and he has a family
[Coke] bah. I have to deal with taxes for work all the time, and now on perl6? ENOFUN. ;) 20:22
moritz [Coke]: sorry for brining it up 20:23
Teratogen yeah I guess a bachelor living in an apartment who owns no stocks or land or anything has a simpler tax picture!
masak hugs [Coke]
[Coke] is merely trolling and is in no way sad. 20:25
[Coke] hopes your taxes are easier to file than my company's. :)
moritz [Coke]: I'm quite sure they are 20:26
pmichaud I spent much of yesterday and today filling out insurance applications. Definitely easier than taxes, but still not fun.
moritz it's just annoying and fiddly work, not really complicated 20:27
kbenson TimToady: Darn families. Always stealing time and attention, and only offering unconditional love in return... 20:29
supernovus In the release notes it says postcircumfix:<[ ]> and postcirumfix:<{ }> are becoming subs instead of methods. How does that work? 20:31
sorear why does TimToady have a hat today? 20:32
moritz supernovus: currently the methods dispatch to methods .at_pos and at_key 20:33
supernovus: a sub can do that too
sorear supernovus: I'm told it works in approximately the way it's always worked in niecza
diakopter sorear: are you on [email@hidden.address] or were you one of the multitudes who unsubscribed
moritz (first they do slicing and such, and then redispatch to at_pos)
sorear diakopter: i'm on it. please do not ask for my opinion on moarvm, I will decide that tomorrow evening 20:34
diakopter sorear: I wasn't going to :P 20:35
supernovus How does the sub know what object instance called it? Also, afaik 'self' is not available in a sub... I'm a bit confused. :/ 20:36
jnthn supernovus: It's the first argument. 20:37
PerlJam supernovus: it would be the first arg 20:37
sorear supernovus: how does a postfix:<foo> sub get an argument? 20:38
masak rn: sub foo($object) { say "OH HAI, $object" }; "supernovus".&foo() 20:39
camelia rakudo cd5ca7, niecza v24-55-g7942a09: OUTPUT«OH HAI, supernovus␤»
supernovus Ah, so instead of method postcircumfix:<[ ]> ($offset) { ... } , I'd use sub postcircumfix:<[ ]> ($object, $offset) { ... } ?
moritz multi sub 20:40
TimToady well, subscripts can be more complicated than that, but essentially yes
moritz but yes, that's the general idea
supernovus the biggest difference I guess being that it can no longer access private variables...
TimToady it's likely to be a primitive that can access anything it jolly well pleases 20:41
jnthn supernovus: at_pos and at_key are typically better things to override
(They stay as methods.)
TimToady subscripting is an important optimization target
jnthn Yeah.
TimToady and having it as methods tends to work against that 20:42
supernovus Hmm, well, XML's tests still all pass, so I'm guessing this change hasn't happened quite yet?
jnthn I mean, if postcircumfix:<[ ]> is a sub, the inliner can inline it.
supernovus: No, it's upcoming.
supernovus: I'll see if there can be a decent migration path. 20:43
supernovus I'll try changing XML to use at_pos and at_key and see what happens.
TimToady and for known types, especially compact arrays, it can do much more
jnthn Right.
It was when I was thinking about compact arrays that I hit upon wanting this change.
diakopter sorear: as long as you don't critique my code in moarvm, I won't run away crying. we all know a C novice's code is going to not be superb. ;) 20:44
supernovus Yay, changing to at_pos and at_key worked great. Simpler to remember method names too :-) 20:46
supernovus Is there an underlying method for postcircumfix:<( )> ? 20:50
moritz no 20:53
supernovus Is it changing to a sub too? I have at least one library that uses it.
moritz I don't think so, at least not for now 20:54
tadzik ehh, my nqp now reliably segfaults
moritz invocation is tricky
tadzik gist.github.com/tadzik/5687913 20:55
jnthn tadzik: try nuking any ModuleLoader.pbc in your Parrot install 20:56
tadzik I think I have my parrot install nuked :) will re-check
dalek kudo-star-daily: c9545f7 | coke++ | log/ (5 files):
today (automated commit)
 23:17
[Coke] who controls dalek? 23:18
Teratogen DOCTOR WHO
lue
.oO(davros.feather.nl)
 23:23
.oO(*davros.feather.perl6.nl)
masak 'night, #perl6 23:49
jnthn 'night, #perl6 23:51

Powered by Cro and the Raku® Programming Language.

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

Thank you!