»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, std:, or /msg camelia p6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by moritz on 25 December 2014.
00:09 Ugator left 00:16 Peter_R left 00:17 gfldex left 00:26 Tux__ joined 00:28 rurban joined 00:30 raiph left 00:31 Mouq joined 00:32 raiph joined 00:35 [Sno] left
dalek kudo-star-daily: 04f7943 | coke++ | log/ (9 files):
today (automated commit)
00:35
rl6-roast-data: ff5c980 | coke++ | / (5 files):
today (automated commit)
00:46 Tux__ left
[Coke] colomon: mojo6 is basically a skeleton, nothing usable, unless hoelzro is even more amazing than I thought. :) 00:49
00:50 spider-mario left 00:53 lizmat joined
dalek ast: b354a93 | skids++ | S04-declarations/my.t:
Add tests for RT#102414

Current status:
Original error fixed on moar and parrot, NullPointer on jvm. One of two errors added later to the ticket seem to be fixed. Spec clarification still needed as to one of the two syntaxes. However since that syntax is the one that now works, test it.
00:56
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=102414
00:57 andreoss left 01:01 [Sno] joined 01:08 kjs_ left
skids m: loop (my $x = 1, my $y = 2; $x > 0; $x--) { $y.say }; # Gives "Useless use of , warning" 01:08
camelia rakudo-moar ffbb70: OUTPUT«WARNINGS:␤Useless use of "," in expression "my $x = 1, my $y = 2" in sink context (line 1)␤2␤»
masak m: my $x = 1, my $y = 2; say "$x $y" 01:09
camelia rakudo-moar ffbb70: OUTPUT«WARNINGS:␤Useless use of "," in expression "my $x = 1, my $y = 2" in sink context (line 1)␤1 2␤»
masak I contest that it's useless.
01:09 espadrine_ left
skids m: loop (my $x = 1; my $y = 2; $x > 0; $x--) { $y.say }; # Gives "Useless use of , warning" 01:10
camelia rakudo-moar ffbb70: OUTPUT«===SORRY!=== Error while compiling /tmp/TdiW3990MA␤Missing block␤at /tmp/TdiW3990MA:1␤------> loop (my $x = 1; my $y = 2; $x > 0⏏; $x--) { $y.say }; # Gives "Useless use␤»
masak may not be idiomatic, but skids++ just showed that it can be useful.
skids Well, to C coders that is idiomatic :-). 01:11
Kristien generating javascript code is difficult 01:19
I wish JavaScript had goto.
timotimo just make craptons of functions :P 01:20
like you did for your SSA
Kristien my new SSA doesn't have that!
it also supports more than one basic block per function :P
and there are conditional jumps that actually compile to if statements
skids RT#69492 might be closeable? 01:21
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...l?id=69492
Kristien I bet φ instructions will be fun to implement. 01:22
masak they usually are.
Kristien the obvious implementation is having each jump set a variable that indicates the block it comes from, and have the phi instruction look it up in a table :P 01:23
01:27 colomon left 01:29 colomon joined
vendethiel- m: say ^10 01:33
uh? camelia?
01:36 raiph left, ilbot3 left, Diederich left, Timbus_ left, jdv79 left, amkrankr1leuen left 01:44 ilogger2 joined, ChanServ sets mode: +v ilogger2, timo joined, alinbsp_ joined, alinbsp_ is now known as alinbsp 01:45 timo is now known as Guest76060 01:46 Mouq joined 01:47 adu joined 01:48 woshty joined, jfried joined, avar joined, jantore joined, colomon_ joined 01:50 colomon_ is now known as colomon 01:51 mtj_ joined 02:04 Guest76060 is now known as timotimo 02:15 Peter_R joined
Kristien goes to sleep 02:16
bye
02:20 bjz joined 02:35 asdf12z_ joined, andreoss joined 02:36 Peter_R left 02:42 Peter_R joined 02:43 Peter_R left, Peter_R joined, Peter_R left, Peter_R joined 02:47 ilbot3 joined 02:53 alinbsp left 03:09 muraiki_ joined 03:15 andreoss left
Mouq I'm thinking the "my $x = 1, my $y = 2;" warning is to help people who don't yet know the difference between item and list assignment 03:24
(still, LTA)
m: my $y; my $x = 1, $y =2;
camelia ( no output )
03:24 colomon left
Mouq m: my $x; $x = 1, my $y = 2; 03:24
camelia ( no output )
Mouq m: my $x = 1, my $y = 2; 03:25
camelia ( no output )
Mouq Oh, last expression
m: my $x = 1, my $y = 2; 1
camelia rakudo-moar ffbb70: OUTPUT«WARNINGS:␤Useless use of "," in expression "my $x = 1, my $y = 2" in sink context (line 1)␤»
Mouq m: my $y; my $x = 1, $y = 2; 1
camelia rakudo-moar ffbb70: OUTPUT«WARNINGS:␤Useless use of "," in expression "my $x = 1, $y =" in sink context (line 1)␤»
Mouq m: my $y; my $x = 1, say "side effects!"; 1
camelia rakudo-moar ffbb70: OUTPUT«WARNINGS:␤Useless use of "," in expression "my $x = 1, say \"side effects!\"" in sink context (line 1)␤side effects!␤»
03:25 colomon joined
Mouq std: my $y; my $x = 1, say "side effects!"; 1 03:26
camelia std f9b7f55: OUTPUT«ok 00:00 139m␤»
03:30 muraiki_ left
skids So in RT, I notice sometimes there are automated updates to tickets when roast updates are made. But not for the ones I did. 03:30
03:32 noganex joined
Mouq skids: I've noticed that too, I'm not sure if that feature still works though. Or maybe there's some special incantation niether of us know for commit messages or something 03:33
m: my @a; (0,1,2).map(-> $k {(2,3,4).map(->$k2 {@a[$k][$k2]})}).eager.Parcel = 4..12; say @a.perl 03:34
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, Any, Any, Any, 6], [Any, Any, Any, Any, 9], [Any, Any, Any, Any, 12])␤»
Mouq ^^ Notice, it autovivifies @a's elements, but the assignment is wrong
m: my @a; (0,1,2).map(-> $k {(2,3,4).map(->$k2 {@a[$k][$k2]})}).eager.Parcel = 4..12; (0,1,2).map(-> $k {(2,3,4).map(->$k2 {@a[$k][$k2]})}).eager.Parcel = 4..12; say @a.perl
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, Any, 4, 5, 6], [Any, Any, 7, 8, 9], [Any, Any, 10, 11, 12])␤»
Mouq ^^ But doing it twice, it works fine 03:35
The past few days I've been trying to get LoL assignment to a) do autoviv and b) do assignment correctly 03:36
I'm not sure where the above bug comes from, but I'm guessing it's a bug in Parce.STORE 03:37
skids (RT#64990 is an example of a ticket with the automatic roast notice.) 03:38
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...l?id=64990
Mouq Golfed:
m: my @a; (@a[0][1], @a[0][2], @a[1][1], @a[1][2]) = 1,2,3,4; say @a.perl
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, Any, 2], [Any, Any, 4])␤»
Mouq skids: Right, but that was back in 2009 :/ 03:39
skids Yeah.
The other thing is I mailed two followup emails to RT for tickets I had no comment in, and they did not seem to take.
psch std: my $x = 1, my $y = 2; 1; 03:40
camelia std f9b7f55: OUTPUT«ok 00:00 138m␤»
psch aren't we just warning too much then for that assignment?
m: my $x = 1, my $y = 2; 1;
camelia rakudo-moar ffbb70: OUTPUT«WARNINGS:␤Useless use of "," in expression "my $x = 1, my $y = 2" in sink context (line 1)␤»
psch or is that not a warning std would catch? 03:41
skids Yes I think just the warning is the problem. I noticed that because roast files were emitting them.
psch m: my @a; @a[0][1] = 1; say @a.perl 03:42
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, 1])␤»
psch m: my @a; (@a[0][1],) = (1,); say @a.perl
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, 1])␤»
psch m: my @a; (@a[0][1], @a[0][2]) = (1, 2); say @a.perl 03:43
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, Any, 2])␤»
03:43 raiph joined
skids m: my @a; (@a[0][2], @a[1][2], @a[0][1], @a[1][1]) = 1,2,3,4; say @a.perl 03:44
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, 3], [Any, 4])␤»
psch oh but there's an NPE on jvm
i'll check that out
skids m: my @a; my @b := (@a[0][1], @a[1][1], @a[0][2], @a[1][2]); @b = (1,2,3,4); @a.perl.say; # same behavior this way. 03:47
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, Any, 3], [Any, Any, 4])␤»
psch std: my (int $a) 03:53
camelia std f9b7f55: OUTPUT«ok 00:00 137m␤»
psch hm
are those parens superstituous?
oh, found it, S03:Declarators 03:54
synopsebot Link: perlcabal.org/syn/S03.html#Declarators
psch m: my :(int $a = 5)
camelia rakudo-moar ffbb70: OUTPUT«===SORRY!=== Error while compiling /tmp/xE2UOAcy1a␤Malformed my␤at /tmp/xE2UOAcy1a:1␤------> my ⏏:(int $a = 5)␤»
psch m: my :(Str $x = "foo")
camelia rakudo-moar ffbb70: OUTPUT«===SORRY!=== Error while compiling /tmp/lXvt9KuXrV␤Malformed my␤at /tmp/lXvt9KuXrV:1␤------> my ⏏:(Str $x = "foo")␤»
psch p: my :(Str $x = "foo") 03:55
camelia rakudo-parrot ffbb70: OUTPUT«===SORRY!=== Error while compiling /tmp/UpOxwmm3le␤Malformed my␤at /tmp/UpOxwmm3le:1␤------> my ⏏:(Str $x = "foo")␤»
skids Mouq: it looks to me like on that first list assignment, each autoviv nukes the existing array. 03:56
Mouq skids: Ohh, nice! So @a[0] = [Any, 1]; and then @a[0] = [Any, Any, 2] 03:59
03:59 |Tux| joined
psch std: my :(Dog $foo is woof = 42) 03:59
camelia std f9b7f55: OUTPUT«===SORRY!===␤Malformed my at /tmp/xONOC85Ix0 line 1:␤------> my ⏏:(Dog $foo is woof = 42)␤ expecting any of:␤ name␤ scoped declarator␤Parse failed␤FAILED 00:00 134m␤»
psch huh, std and specs don't agree, apparently
std: my (Dog $foo is woof = 42) 04:00
camelia std f9b7f55: OUTPUT«===SORRY!===␤In parameter declaration, typename 'Dog' must be predeclared (or marked as declarative with :: prefix) at /tmp/ZPkSmA9Aac line 1:␤------> my (Dog⏏ $foo is woof = 42)␤Parse failed␤FAILED 00:00 136m␤»
Mouq m: my @a; my \p := @a[0]; (p[1], p[2]) = 1,2; say @a.perl 04:04
camelia rakudo-moar ffbb70: OUTPUT«Array.new([Any, 1, 2])␤»
04:04 telex joined
asdf12z_ any good books, guides, or tours whatever you want to call them to perl6? 04:19
psch asdf12z_: learnxinyminutes.com/docs/perl6/ is a good introductory tour 04:20
asdf12z_ is supplies core to perl6 or is it part of it's standard lib? 04:21
Mouq perl6 -e'my @a; @a[0,1;2,3] = 1,2,3,4; say @a.perl'
Array.new([Any, Any, 1, 2], [Any, Any, 3, 4])
\o/
psch m: Supply.interval(1).tap(-> $s { say "{$s}th tap }); sleep 5
camelia rakudo-moar ffbb70: OUTPUT«===SORRY!=== Error while compiling /tmp/9tdJhXEzOZ␤Unable to parse expression in double quotes; couldn't find final '"' ␤at /tmp/9tdJhXEzOZ:1␤------> .tap(-> $s { say "{$s}th tap }); sleep 5⏏<EOL>␤ expectin…»
psch m: Supply.interval(1).tap(-> $s { say "{$s}th tap" }); sleep 5
camelia rakudo-moar ffbb70: OUTPUT«0th tap␤1th tap␤2th tap␤3th tap␤4th tap␤5th tap␤»
asdf12z_ :(
skids asdf12z_: Core, but only recently implemented, so you need the moarvm backend.
asdf12z_ seems like a lot of 'overloading' with sigils and things packaged into core 04:22
wouldve been nice to see things like that be more explicitly imported in
Mouq asdf12z_: Can you give an example? I'm not sure what you mean
skids For the most part, nothing gets in the way.
psch why? async is hard. having tools to make it easy is useful 04:23
asdf12z_ but what if i don't need it?
it's for visibility purposes
mostly
skids Don't use it, then.
asdf12z_ still for visibility purposes
psch what does visibility mean in this context?
asdf12z_ i should be able to understand by looking at some module or code what its using it its module/code 04:24
skids I'm having a hard time imagining where visibility gets impaired.
asdf12z_ just by its imports
psch that sounds rather C-ish to me
asdf12z_ it is, which i like, but not the way C does it 04:25
but it's true in mostly every other modern/popular language too
psch yeah, i find that confusing. why is exponentiation a basic operation, but logarithms aren't? - looking at python here 04:27
i.e. i have to 'from math import log' to log()
asdf12z_ visibility in that if i look at a piece of code, i have to remember all of perl6 core vs things that are not part of core and having to process that
psch maybe log() is special because people (probably?) want to log-to-file more often than log-base-e 04:28
asdf12z_ if you want a great example, i like the way golang does it
very very little to remember that are part of golang's core, very little keywords 04:29
skids really hates includes, personally.
asdf12z_ everything else is part of its standard lib
psch i haven't looked at golang at all
asdf12z_ makes reading code so much easier
psch FWIW, i don't see any benefit in having a «list of stuff to expect in this file» at the top 04:30
asdf12z_ with perl6 i feel like i have to memorize a lot BEFORE i even touch any modules, libs
psch i think i understand the argument for not having everything implicitely imported, but i don't understand the argument for having it in the stdlib (which is always there) but not imported
because runtime exports mean that maybe we only import something later on, and that gets more confusing than having everything there from the start 04:31
s/exports/imports/
*the posibility of runtime...
asdf12z_ oh i guess thats true for runtime imports 04:32
but i guess that can be solved by common coding practice?
psch m: module Foo { constant TAU = 2 * pi }; BEGIN try EVAL "say TAU"; try EVAL "say TAU"; END { import Foo; say Foo::TAU }
camelia rakudo-moar ffbb70: OUTPUT«6.28318530717959␤»
skids Perhaps we'll eventually have a lot of "no" pragmas and then if you like, you can turn off parts of the language for visibility. 04:33
psch m: module Foo { constant TAU = 2 * pi }; { say TAU; CATCH { default { say "no TAU :(" } } }; END { import Foo; say Foo::TAU }
camelia rakudo-moar ffbb70: OUTPUT«===SORRY!=== Error while compiling /tmp/tcNNRvSi6a␤Undeclared name:␤ TAU used at line 1␤␤»
psch grml
can't push the constant to runtime like that
m: module Foo { our \TAU = 2 * pi }; { say Foo::TAU; CATCH { default { say "no TAU :(" } } }; END { import Foo; say Foo::TAU } 04:34
camelia rakudo-moar ffbb70: OUTPUT«===SORRY!=== Error while compiling /tmp/Ag9PltD_yY␤our scoped term definitions (only 'my' is supported at the moment) not yet implemented. Sorry. ␤at /tmp/Ag9PltD_yY:1␤------> module Foo { our \TAU = 2 * pi ⏏}; { sa…»
psch ...and an NYI :)
psch stops noise-ing about
asdf12z_: there might be an argument made for "don't do runtime imports", but i doubt the practice would achieve more than (potentially very) big BEGIN blocks 04:35
or maybe INIT, i'm a bit fuzzy on phasers 04:36
asdf12z_ i mean to a newcomer that doesn't know perl, it just seems overwhelming in that there's a lot to remember for just using perl6 minus any libraries, you can say if you don't want it, don't use it, but i'm still forced to memorize it exists just in case later on i see it
so i know Supplies isn't pulled in by some import 04:37
Mouq m: module Foo { constant TAU is export = 2 * pi }; { say Foo::TAU; CATCH { default { say "no TAU :(" } } }; { import Foo; say TAU }
camelia rakudo-moar ffbb70: OUTPUT«6.28318530717959␤6.28318530717959␤»
asdf12z_ or some kind of magical trickery some lib is doing
skids But you also have to memorize what to exmplicitly include on the flip side, so...
asdf12z_ that's not true, i can just look up the standard lib in golang and find what i want to use 04:38
so i don't have to memorize every lib's name
psch asdf12z_: p6doc exists, albeit incomplete
asdf12z_ besides the naming of supplies is ambigious 04:39
psch asdf12z_: the plan is to have it available in the REPL as well, for the self-documenting introspection
skids But you can just look up identifiers you find in sources in the docs, when you don't know them, so I'm not seeing the difference.
asdf12z_ but i'd have to memorize it to know where to look is what im saying
skids Right or through introspection.
psch m: #| does the magicks␤sub foo() { warn "NYI" }; say &foo.WHY 04:40
camelia rakudo-moar ffbb70: OUTPUT«does the magicks␤»
skids why? You see someone using class "fnord" and you look up "fnord"?
psch m: say "Life, the Universe, and Everything".WHY
camelia rakudo-moar ffbb70: OUTPUT«42␤»
asdf12z_ skids: and then if fnord doesn't exist, i'd have to look at the libs its importing to see which one creates that namespace ?
skids Wait what who put that in there. 04:41
psch skids: lizmat++ iirc
asdf12z_ skids: unless i memorized every identifier to know it doesn't come from perl6 core, then i'd go look at the imports, otherwise i'd have to check in 2 places
skids I... really don't see your point. 04:42
psch m: module Foo { sub baz is export { } }; import Foo; say &baz.package 04:43
camelia rakudo-moar ffbb70: OUTPUT«(Foo)␤»
psch asdf12z_: ^^^ would that help?
asdf12z_ no 04:44
because i can look at perl6 code and see something like `convert;` 04:45
and i won't know (unless i memorized every identifier perl6 has) if convert is some macro or part of perl6
or if* 04:46
its a non-issue if it was small enough to memorize
thats my point 04:47
psch but it works the other way around as well. if there's no «use» or «import» statement at the top you know nothing's been imported 04:48
well, except for runtime imports, as mentioned
if those exists you can check what they export, which is probably a smaller list that golang's stdlib
i understand your point, but to me that's not a language design issue, it's a tooling issue 04:49
and yes, the tools are lacking at the moment
but thanks to p6doc/POD6 and introspection we can get that from the language itself 04:50
i.e. it's probably not very hard to write a tool that you point at a file and ask it about an identifier and it gets you the package it comes from with the documentation that exists 04:51
nobody's gotten around to that yet though
asdf12z_ in golang its very different, and i'm not trying to make perl6 like golang of course, it's just an example
gobyexample.com/range-over-channels for instance 04:52
anything that doesn't begin with fmt namespace is part of golang
Mouq asdf12z_: And you may want to check out doc.perl6.org too. The search interface isn't perfect, and it doesn't know everying, but it should be able to identify the majority Perl 6 builtins
asdf12z_ anything that does, is part of the fmt lib
so if i didn't know what `make` was, i do _know_ it comes from golang
its very easy to reason about 04:53
skids How does it benefit you to know whether it comes from golang or a lib, exactly, other than telling you weher eto look for docs?
asdf12z_ knowing where to look is really really important man 04:54
skids But if you had an search tool that could just find it across docs, that would solve the problem, no?
asdf12z_ especially if this code wasn't just some short example and 1,000+ lines of code
depends on how good the search tool is 04:55
psch that makes me wonder if our "did you mean" could extend to installed modules 04:58
it probably could but would be expensive or need an index
asdf12z_ im just giving my impressions as an outsider of anything perl 05:00
skids Well, expense really isn't as much of a concern in the error path.
asdf12z_ i've just been casually looking at perl6 talks on youtube whenever i have the time 05:01
and sometimes it seems perl6 was made for perl users
skids Another good way to cut your teeth is try the one-liners github.com/dnmfarrell/Perl6-One-Liners 05:02
psch skids: right, but depending on the available CURs that could still be too expensive
skids: wouldn't want my LWP one-liner to crawl all modules for "geptrint" :) 05:03
skids Yeah, it sounds like something you'd want to turn on with an --ide flag or something. 05:04
psch asdf12z_: well, perl6 is still perl, nobody ever denied that 05:10
hrm, no idea how to fix the NPE in «my (int $x)», but the null-check that should've thrown a more specific error only created an Exception but didn't throw it 05:21
that's at least something i suppose... :)
05:36 fil_ joined
fil_ hellos 05:36
yoleaux 30 Jan 2015 17:27Z <FROGGS> fil_: soon meant 7 hours later :o) (I had no idea when I would get around to it, and what it would involve... happily it is just about setting a flag to true)
skids o/
fil_ is there a convention about putting classes in files like there is in java? 05:37
any1? 05:38
skids Very often the easiest way to provide a single class module is to name it after the class, but files can contain multiple classes.
fil_ are the best practices about how to split your work up into multiple files documented anywhere? 05:39
it's not clear to me how I should be splitting up my work...
05:39 adu left
fil_ in Java it's easy--the language forces you to have 1-to-1 05:39
skids Well, not quite yet, but if you look at modules.perl6.org you can see what others are doing.
fil_ i spent the last half hour doing that 05:40
didn't see much consistency
ok, sounds like it's premature to be asking...
thx
05:40 fil_ left 05:44 adu joined 05:50 konsolebox joined 06:03 adu left 06:05 jack_rabbit joined 06:09 Mouq left 06:28 adu joined 06:35 khisanth_ joined
moritz good morning 06:36
.tell fil_ IMHO it's best practice to put each class that is publicly accessible into a separate class. As for rest, use your best judgement :-) 06:38
yoleaux moritz: I'll pass your message to fil_.
asdf12z_ niecza is no longer in development right? 07:01
:( perl6 on the clr would be cool
07:03 darutoko joined
moritz asdf12z_: aye 07:04
asdf12z_: I still hope somebody starts a CLR backend for rakudo 07:05
07:15 adu left 07:18 raiph left 07:24 [Sno] joined 07:34 xfix joined 07:40 FROGGS[mobile] joined
FROGGS[mobile] o/ 07:41
07:42 Rounin joined 07:43 yeahnoob joined
tadzik o/ 07:51
07:54 Oe_eden joined 07:55 Mouq joined
FROGGS[mobile] tadzik: breakfast? 07:56
08:01 nige joined
Oe_eden is there a perl6 oreilly book in the making yet or will they update programming perl to accomidate perl6? 08:07
tadzik FROGGS[mobile]: yep 08:14
FROGGS[mobile] Oe_eden: I fear such an oreally book will take a fear years from now 08:15
I dont think anybody is working at it atm, given that we want to get 6.0.0 out
that said, I hope the Perl 6 book gets revived, and I'd like to write a chapter or two about topic I know well 08:17
Oe_eden hmm also I own version 2 op PP but I noticed only yesterday at Fosdem that v3 is almost twice the size... what is that about?
08:19 lizmat joined
FROGGS[mobile] ask wendy :o) 08:21
I honestly dont know
08:24 sqirrel joined
dalek kudo/nom: 0959c68 | Mouq++ | src/core/ (2 files):
Try once more to do LoL assignment correctly

There shouldn't be any need to muck about with special cases in the case of LoL assignment; that handling should be the same as in normal LoL access. This patch makes sure we're returning the proper container after an LoL access of any sort; this way, assignment is trivial.
08:26
kudo/nom: 84ca774 | Mouq++ | src/core/ (2 files):
Merge branch 'nom' of github.com/rakudo/rakudo into nom
Mouq Far too many hours went into that tiny patch
Lists are hard :P
08:27 denis_boyun joined
lizmat Mouq++ 08:27
lizmat is going to try that now
Mouq lizmat++ for alerting me in the first place 08:30
08:31 denis_boyun__ joined, denis_boyun left 08:36 labster_ joined 08:37 mr-fooba_ joined, denis_boyun__ left, jdv79_ joined 08:38 jschulz_ joined, atta_ joined, sven_123_ joined 08:39 __rnddim__ joined, __rnddim__ is now known as lue, xinming_ joined, xiaomiao joined 08:40 petercom1and joined, pochi_ joined, avuserow_ joined 08:41 petercom1and left, petercom1and joined, FROGGS[mobile] left 08:42 colomon_ joined, rmgk_ joined, rmgk_ is now known as rmgk, gfldex joined 08:44 masak__ joined, esaym joined
dalek ast: 4954d43 | lizmat++ | integration/weird-errors.t:
Unfudge passing test
08:45
08:45 colomon left, labster_ is now known as labster, colomon_ is now known as colomon, petercom1and is now known as petercommand 08:46 mst joined 08:47 _sri joined 08:48 esaym left 08:49 esaym153 joined
pmichaud bonjour, #prl6 08:49
*#perl6
08:50 awwaiid joined, lestrrat joined 08:54 rurban joined 08:58 rurban1 joined 09:02 alinbsp joined 09:03 haroldwu left, haroldwu joined, kshannon_ joined 09:04 rindolf joined 09:06 rindolf left 09:07 FROGGS joined, FROGGS[mobile] joined 09:09 sqirrel left, simcop2387 joined 09:13 [Sno] left 09:20 kjs_ joined, FROGGS[mobile] left 09:21 FROGGS left 09:23 atroxaper joined 09:29 wilx left 09:38 brrt joined 09:39 anaeem1_ joined, nige left
dalek kudo/nom: 27ebd7e | lizmat++ | src/ (3 files):
Implement the "aka" trait on methods

This basically aliases a method to other names in the same package. This is different (apparently) to the "handles" trait on method.
method foo aka <bar baz> { # can do $o.foo, or $o.bar or $o.baz for same effect
09:42
09:50 anaeem1_ left 09:51 anaeem1_ joined 09:52 FROGGS joined
FROGGS lizmat: that's my version: multi trait_mod:<is>(Routine $r, :$aka!) is export { $r.package.^add_method($aka, $r) }; 09:53
@all: I want to improve the Perl 6 book, and write one or two chapters (regexes/grammars and perhaps something else I have knowledge of).... somebody also picks topics?
@all: would be nice if one or two ppl would subscribe to 1-2 topics 09:54
09:55 [TuxCM] joined, anaeem1_ left
tadzik [Tux]: how's it going? :) 09:57
I managed to get internets
I'll also update rakudo and see if it breaks my stuff too
moritz FROGGS: there are already regexes and grammar chapters in book 10:02
*the book 10:03
FROGGS: no objections to improvements though
do the live streams work today? 10:04
10:04 brrt left 10:06 rurban left
tadzik in the room I'm in (graphics), there was a conversation like "we don't know, we only see this bar going up and down" 10:06
FROGGS sits in the Rust talk and sees that their development is identical to ours... just is put in different words 10:07
i.e. the speaker calls it "Empiric Iteration", we call it Whirlpool
tadzik heh
is it also in K? 10:08
FROGGS in the Janson
tadzik ah
nine dj_goku: thanks! 10:11
10:12 jack_rabbit left 10:13 Kristien joined
dalek kudo/nom: 4dd6c91 | lizmat++ | src/core/traits.pm:
Better implementation of "aka" trait on methods
10:14
10:16 dagurval joined 10:18 masak__ is now known as masak
masak good antenoon, #perl6 10:18
big day today, huh?
10:18 masak is now known as Guest26461, rurban joined
FROGGS masak: aye, Rust will be released this year :P 10:18
10:18 Guest26461 is now known as masak_grr
masak_grr \o/ 10:19
10:19 rurban2 joined
moritz but that's no news, I thought? 10:19
I mean, they've had that timeline for several weeks now 10:20
or even months
FROGGS moritz: to somebody like me that does not care about other languages it is :o)
masak_grr hah! we've had our deadline for *years*, and it's always been "around 18 months from now" :P 10:21
10:23 rurban2 left 10:24 synopsebot joined, spider-mario joined
moritz masak_grr: what's _grr today? :-) 10:25
10:30 _dolmen_ joined
FROGGS use std::thread::Thread; -- Rust 10:30
10:30 [TuxCM] left, yeahnoob left
dalek kudo/nom: 6c957ba | lizmat++ | src/core/traits.pm:
Unbust the Method.handles trait
10:32
masak_grr moritz: just the same old nickserv business.
moritz: it passes after a while. 10:33
tadzik FROGGS: all I can think of is "use std::badger::Badger::BADGER" :P
FROGGS *g* 10:34
10:34 yeahnoob joined
cdc moritz: rtmp://stream-a.fosdem.org:1935/play/janson_a 10:36
10:37 jack_rabbit joined
alpha- is there a page for Perl 6 which is equivalent to perlfunc in Perl 5 ? 10:38
FROGGS alpha-: perhaps this? doc.perl6.org/routine.html 10:39
[ptc] is the sound dropping out all the time for anyone else?
alpha- thanks
FROGGS you might want to highlight 'sub' using your browser in case you are searching for subs only...
alpha- hmm 10:40
FROGGS though, many stuff exists as methods only where it makes sense
alpha- FROGGS I see that there is no 'async' keyword/function/operator/whatever 10:41
perl6 has no 'async' ?
10:41 lizmat left
FROGGS it is a keyword 10:41
Kristien green thread master race
TimToady async is callled 'start' now, and returns a promise
alpha- oh ok 10:42
moritz cdc: thanks
TimToady and it's really a statement prefix operator
moritz [ptc]: sound is stable here
TimToady so it might or might not take a block
[ptc] moritz: thanks; must be just my settings 10:43
FROGGS alpha-: here are slides about that topic: www.jnthn.net/articles.shtml look out for async and concurrency
alpha- thanks FROGGS
FROGGS alpha-: I can also find talks (videos) in case you are interested 10:44
cdc moritz: this stream doesn't work anymore here :'(
[ptc] are the slides at least available? The link doesn't appear on live.fosdem.org 10:45
cdc it works again \o/ 10:46
alpha- FROGGS not atm, I will read the slides and try writing examples, thanks
cdc [ptc]: I don't see them either
[ptc] cdc: it's nice to know it's not just me :-)
10:47 gfldex left
alpha- what is the alternative to perlbrew for Perl 6? e.g. a non-root install of perl6 contained in some directory 10:47
10:47 jack_rabbit left, gfldex joined
FROGGS there is rakudobrew 10:47
alpha-: but for rakudo we always do a non-root install by default
so, cloning rakudo will also get you set up, though rakudobrew also installs useful modules and the module installer 10:48
moritz alpha-: github.com/tadzik/rakudobrew/
10:49 raiph joined 10:50 yeahnoob left, kaare_ joined
alpha- No suitable nqp executables found! 10:51
I think I am missing something
tadzik is that what rakudobrew said?
alpha- no, it's from rakudo Configure.pl
FROGGS alpha-: perl Configure.pl --gen-moar
alpha- ok
FROGGS alpha-: though I thought you get hints printed O.o 10:52
gfldex try perl Configure.pl --backends=moar --gen-nqp --gen-moar
[ptc] INSTALL.txt in the rakudo repo has good installation instructions 10:53
moritz gfldex: --gen-nqp is implied by --gen-moar
gfldex: and --backend=moar too (unless you have more NQPs available)
10:54 konsolebox left
alpha- I can't figure out what NQP is even after reading the description page 10:54
is it like llvm ?
FROGGS alpha-: it is like a very simple version of Perl 6, that lets us write the very core Perl 6 compiler parts also in a Perl 6-ish language 10:55
nqp-m: say("hello world")
camelia nqp-moarvm: OUTPUT«hello world␤»
alpha- interesting
FROGGS like in my example, the parenthesis are mandatory in nqp code 10:56
timotimo o/
[ptc] timotimo: \o 10:57
alpha- are moarvm and nqp strictly perl6 stuff or there are any other projects using this ?
moritz alpha-: there aren't
timotimo we have toy interpreters for ruby and python
moritz alpha-: well, there's a python compiler work-in-progress on nqp 10:58
timotimo and there was a tcl implementation but that was more or less relying on many parrot internals
moritz github.com/arnsholt/snake/
timotimo the stream is giving me audio, no audio, audio, no audio etc etc
actually, video is cutting out as well it seems
but only sometimes
moritz back now here 10:59
[ptc] timotimo: same here
moritz audio seems to be far ahead of video
alpha- ok, looks like it built and runs, I think I have something to play with now, thanks
[ptc] but there's no stream for the slides...
timotimo this is going to be designing a perl number theory module?
[ptc] yup
timotimo OK
FROGGS alpha-: have fun doing so :o) 11:00
TimToady dana has started 11:01
11:02 jack_rabbit joined
dalek kudo/nom: 92211ce | Mouq++ | INSTALL.txt:
Update INSTALL.txt; Parrot now needs ICU
11:02
kudo/nom: 6e182d3 | Mouq++ | src/ (3 files):
Merge branch 'nom' of github.com/rakudo/rakudo into nom
alpha- is there a perl6 compiler that can output native code? eg. create .exe or elf ?
FROGGS alpha-: that's in the works and is a GSoC project proposal for this year
alpha- I see
Kristien So many Perl 6 implementations. 11:03
11:03 kaare_ left, _dolmen_ left
timotimo wow, i cannot at all understand anything the presenter's saying 11:03
jdv79_ where? 11:04
timotimo "i think ... perl ... m ... interesti ... int fo ... envy ... ha ... for lo ... only ... progra ... lang ... island ... grea ... for now ..."
11:04 jdv79_ is now known as jdv79
alpha- timotimo link to stream ? 11:05
timotimo rtmp://stream-a.fosdem.org:1935/play/janson_a
alpha- looks fine here.
timotimo i'm jealous
alpha- 1080p too. nice 11:06
timotimo i have a quite beefy internet connection, so ...
what's wrong? :(
alpha- maybe their connectivity is not great
timotimo that'd surprise me 11:07
alpha- stream is only using 3-4 megabit per second
try different player maybe?
[ptc] video's coming through fine, just the audio keeps dropping out... frustrating 11:08
alpha- everything 100% fine here,
timotimo huh
alpha- video and audio perfect 1080p
timotimo mpalyer is working much better than vlc for this stream
alpha- vlc is known to have warts in some places
timotimo well, what do i know, apparently :) 11:09
11:09 jack_rabbit left
alpha- also mplayer is afaik abandoned 11:09
jdv79 yeah, smplayer is ok for me
alpha- and so is mplayer2
jdv79 vls is choppy
alpha- I recommend mpv.io/
timotimo well, the mplayer2 fork spread a whole lot of FUD
alpha- see link it's a fork of a fork
seems to be alive 11:10
tadzik wow, dat lazer 11:11
11:11 kaare_ joined
timotimo our is-prime is broken for very large numbers? =o 11:12
alpha- perl6 has builtin is-prime?
timotimo yes
alpha- wow
FROGGS he'll send a patch over :o)
timotimo m: say 100121201203010120.is-prime
camelia rakudo-moar 27ebd7: OUTPUT«False␤»
timotimo m: say 100121201203010121.is-prime
camelia rakudo-moar 27ebd7: OUTPUT«False␤»
timotimo m: say 100121201203010123.is-prime 11:13
camelia rakudo-moar 27ebd7: OUTPUT«False␤»
timotimo m: say 100121201203010127.is-prime
camelia rakudo-moar 27ebd7: OUTPUT«False␤»
timotimo i don't know how to prime
alpha- is it an approximate or definitive algorithm ?
[ptc] m say 1.is-prime
m: say 1.is-prime
camelia rakudo-moar 27ebd7: OUTPUT«False␤»
moritz alpha-: probailistisc
[ptc] m: say 2.is-prime
camelia rakudo-moar 27ebd7: OUTPUT«True␤»
timotimo m: say (1001212012030101..1001212012040101).grep(*.is-prime)
camelia rakudo-moar 27ebd7: OUTPUT«1001212012030103 1001212012030187 1001212012030189 1001212012030273 1001212012030283 1001212012030319 1001212012030339 1001212012030367 1001212012030403 1001212012030421 1001212012030463 1001212012030471 1001212012030519 1001212012030537 1001212012030631 1…»
alpha- moritz that's... 11:14
FROGGS m: my $i = 10012120120301012787439875938749587398475; until $i.is-prime { $i++ }; say $i
camelia rakudo-moar 27ebd7: OUTPUT«10012120120301012787439875938749587398503␤»
alpha- brave
arnsholt is-prime is approximate, but IIRC the probability of it being wrong is on the order of a bit in RAM being flipped by a cosmic ray, or some such
pmichaud bon après midi, #perl6
FROGGS pmichaud: o/
timotimo wolframalpha agrees it's prime
11:14 konsolebox joined
tadzik m: 9.is-prime.say 11:14
camelia rakudo-moar 27ebd7: OUTPUT«False␤»
tadzik uff :) 11:15
FROGGS pmichaud: we are in the Janson already, sitting in the middle first row
alpha- /usr/bin/factor too
pmichaud FROGGS: I'm seven seats to your left, one row back.
Kristien m: sub sqrt($x) { PRE { $x >= 0 }; ??? }; say sqrt(9); say sqrt(-1);
camelia rakudo-moar 27ebd7: OUTPUT«Stub code executed in sub sqrt at /tmp/JO6oh8Ow8G:1␤␤0␤Precondition '{ $x >= 0 }' failed␤ in sub sqrt at /tmp/JO6oh8Ow8G:1␤ in block <unit> at /tmp/JO6oh8Ow8G:1␤␤»
tadzik oh, *that*'s where you hid
timotimo greetings, pmichaud
Kristien nice.
preconditions
timotimo i'm glad a really number-theory-savy is interested in perl6, too, so that our stuff can be put through its paces 11:17
[ptc] yay, audio is working with mplayer2
FROGGS timotimo: aye
11:17 espadrine_ joined
FROGGS dana++ # and he really know its stuff 11:17
11:18 vendethiel joined
FROGGS his* 11:18
jdv79 theres no slide feed?
[ptc] jdv79: doesn't seem to be
11:19 kaare_ left
Kristien Can one access in a POST phaser the return value in a different way than assigning it to a variable first? 11:19
11:19 kaare_ joined
moritz Kristien: iirc it's passed in as $_ 11:19
Kristien m: sub abs($x) { return $x; POST { $_ >= 0 } }; say abs(-10) 11:20
camelia rakudo-moar 27ebd7: OUTPUT«Postcondition '{ $_ >= 0 }' failed␤ in sub abs at /tmp/BniuO2z3tg:1␤ in block <unit> at /tmp/BniuO2z3tg:1␤␤»
Kristien nice
m: sub abs($x) { $x; POST { $_ >= 0 } }; say abs(-10)
camelia rakudo-moar 27ebd7: OUTPUT«use of uninitialized value of type Nil in numeric context in sub abs at /tmp/bp6TIjyFhS:1␤␤Nil␤»
Kristien m: sub abs($x) { POST { $_ >= 0 }; $x }; say abs(-10)
camelia rakudo-moar 27ebd7: OUTPUT«Postcondition '{ $_ >= 0 }' failed␤ in sub abs at /tmp/fc2dyQdXRF:1␤ in block <unit> at /tmp/fc2dyQdXRF:1␤␤»
Kristien m: sub abs($x) { POST { $_ >= 0 }; $x }; say abs(10) 11:21
camelia rakudo-moar 27ebd7: OUTPUT«10␤»
Kristien most shiny
FROGGS that makes sense I guess
Kristien I love contract programming.
moritz m: sub abs($x) { POST { $_ >= 0 }; $x }; try abs($x); say $!.perl 11:22
camelia rakudo-moar 27ebd7: OUTPUT«===SORRY!=== Error while compiling /tmp/_BY72tB2j8␤Variable '$x' is not declared␤at /tmp/_BY72tB2j8:1␤------> $x) { POST { $_ >= 0 }; $x }; try abs($x⏏); say $!.perl␤ expecting any of:␤ postfix␤»
moritz m: sub abs($x) { POST { $_ >= 0 }; $x }; try abs(-1); say $!.perl
camelia rakudo-moar 27ebd7: OUTPUT«X::Phaser::PrePost.new(phaser => "POST", condition => " \{ \$_ >= 0 }")␤»
alpha- does perl6 have arbitrary precision math ? 11:23
11:24 denis_boyun joined
TimToady m: say [\*] 1..* 11:24
camelia rakudo-moar 27ebd7: OUTPUT«1 2 6 24 120 720 5040 40320 362880 3628800 39916800 479001600 6227020800 87178291200 1307674368000 20922789888000 355687428096000 6402373705728000 121645100408832000 2432902008176640000 51090942171709440000 1124000727777607680000 25852016738884976640000 62…»
TimToady m: say 1,2,4 ... * 11:25
camelia rakudo-moar 27ebd7: OUTPUT«1 2 4 8 16 32 64 128 256 512 1024 2048 4096 8192 16384 32768 65536 131072 262144 524288 1048576 2097152 4194304 8388608 16777216 33554432 67108864 134217728 268435456 536870912 1073741824 2147483648 4294967296 8589934592 17179869184 34359738368 68719476736…»
pmichaud I'm not sure our precision is arbitrary. :)
TimToady m: say 2 ** 1000
camelia rakudo-moar 27ebd7: OUTPUT«1071508607186267320948425049060001810561404811705533607443750388370351051124936122493198378815695858127594672917553146825187145285692314043598457757469857480393456777482423098542107460506237114187795418215304647498358194126739876755916554394607706291457119…»
alpha- cool so integers are there
and fractions?
Kristien m: say 2.6 ** 1000
camelia rakudo-moar 27ebd7: OUTPUT«Inf␤»
Kristien m: say (2+6/10) ** 1000
camelia rakudo-moar 27ebd7: OUTPUT«Inf␤» 11:26
Kristien lol ** fails on rationals
pmichaud m: say 1 / 2**1000
camelia rakudo-moar 27ebd7: OUTPUT«9.33263618503219e-302␤»
Kristien m: (2+6/10).WHAT.say
camelia rakudo-moar 27ebd7: OUTPUT«(Rat)␤»
pmichaud m: say 1 / 2**100
camelia rakudo-moar 27ebd7: OUTPUT«7.88860905221012e-31␤»
TimToady we do rational arithmetic up to a denominator of 2**64
Kristien oh OK
pmichaud m: say (1 / 2**50)
camelia rakudo-moar 27ebd7: OUTPUT«0.00000000000000089␤»
alpha- m: 22/7
camelia ( no output )
pmichaud m: say (1 / 2**50).WHAT
camelia rakudo-moar 27ebd7: OUTPUT«(Rat)␤»
TimToady m: say .1 + .2 - .3
camelia rakudo-moar 27ebd7: OUTPUT«0␤»
alpha- m: say 22/7
camelia rakudo-moar 27ebd7: OUTPUT«3.142857␤»
alpha- how to get to 50 digits after . ? 11:27
pmichaud m: say 355/113
camelia rakudo-moar 27ebd7: OUTPUT«3.141593␤»
gfldex m: say '0.', '0' x 50, '1';
moritz say qx/pi/
camelia rakudo-moar 27ebd7: OUTPUT«0.000000000000000000000000000000000000000000000000001␤»
gfldex easy! 11:28
moritz say qx/pi 50/ # actually
Kristien m: say (22/7).perl
camelia rakudo-moar 27ebd7: OUTPUT«<22/7>␤»
jdv79 did the video just freezw?
gfldex jdv79: works for me
jdv79 maybe my conn
alpha- once it froze, yes
a bit earlier
cdc same for me 11:29
Kristien m: say (22/7).Num.perl
camelia rakudo-moar 27ebd7: OUTPUT«3.14285714285714e0␤»
Kristien m: say (22/7).Num
camelia rakudo-moar 27ebd7: OUTPUT«3.14285714285714␤»
11:29 denis_boyun left
moritz m: say pi.Rat.perl 11:30
camelia rakudo-moar 27ebd7: OUTPUT«<355/113>␤»
11:30 manab joined
tadzik :) 11:30
pmichaud wow, pi.Rat gives back the 355/113 fraction.
that's... neat.
manab hello everyone
alpha- wow, perl has builtin pi
haha
nice
moritz hi manab
tadzik is pi in settings ratified, or Numd?
manab new to this community 11:31
moritz tadzik: Num
Kristien m: pi.WHAT.say
moritz m: say pi.^name
camelia rakudo-moar 27ebd7: OUTPUT«(Num)␤»
rakudo-moar 27ebd7: OUTPUT«Num␤»
tadzik awesome that it comes up with the Rat representation
manab can someone please help me out so i can start contributing for tpf projects
Kristien m: sin(pi).say
camelia rakudo-moar 27ebd7: OUTPUT«1.22464679914735e-16␤»
FROGGS manab: how can I help you?
moritz manab: find one that interests you
manab: rakudo.org/how-to-help/ for example 11:32
jdv79 my video has diverged from audio now
Kristien m: class PI { }; method sin(PI $x) { 0 }; sin(PI).say
camelia rakudo-moar 27ebd7: OUTPUT«Useless declaration of a has-scoped method in mainline (did you mean 'my method sin'?)␤Cannot call 'sin'; none of these signatures match:␤:(Numeric \x)␤:(Cool \x)␤:(num $x)␤ in block <unit> at /tmp/5lGzJ8PtX0:1␤␤»
jdv79 neat
manab thanks a lot
Kristien m: class PI { }; multi sub sin(PI $x) { 0 }; sin(PI).say
camelia rakudo-moar 27ebd7: OUTPUT«0␤»
11:32 denis_boyun joined
Kristien :D 11:32
m: class PI { }; multi sub sin(PI $x) { 0 }; sin(pi).say
camelia rakudo-moar 27ebd7: OUTPUT«1.22464679914735e-16␤» 11:33
tadzik m: my constant pi = 4; say sin(pi)
camelia rakudo-moar 27ebd7: OUTPUT«-0.756802495307928␤»
tadzik :-D-/-<
Kristien write a CAS in Perl 6
moritz manab: in general, you look for TODO files, or simply use the project, find something that could be improved, and do it
Kristien: iirc somebody started one. Was it raydiak?
manab ohh okay
thank u :)
11:34 sqirrel joined
manab i actually wanted to contribute for tpf gsoc15 11:34
FROGGS look a sqirrel!
11:34 anaeem1 joined
moritz manab: why TPF? 11:34
manab: I'm just curious
cdc jdv79: you can use "-" or "+" with mplayer to resync manually
sqirrel :P
jdv79 oh. i restarted 11:35
FROGGS manab: look at this, and (let) add ideas as needed: github.com/perl6/mu/blob/master/mi...5/ideas.md
Kristien moritz: github.com/raydiak/Math-Symbolic
manab moritz: i want to learn perl actually, curiosity
moritz manab: that's a good motivation :-)
manab i have done many projects on java , html, apps 11:36
jdv79 slides are now working
[ptc] yay!
moritz jdv79: stream URL?
manab FROGGS: thanks a lot :)
colomon are there ongoing p6 talks?
FROGGS manab: you're welcome
[ptc] mplayer rtmp://stream-a.fosdem.org:1935/play/janson_a
jdv79 ..._b
[ptc] ups ... _b
FROGGS colomon: larry's talk will come next
cdc jdv79: thanks for this good news !
colomon do we have audio?
tadzik colomon: currently mostly Perl 5, but Larry comes in 25 minutes from now 11:37
jdv79 yes
colomon \o/
jdv79 seems to all work atm in mplayer
Kristien I learned a new programming language yesterday.
I want to learn more soon.
FROGGS Kristien++
moritz \o/
11:38 anaeem1 left
moritz working streams 11:38
colomon hmmm…. I’m getting audio with janson_a but not janson_b ?
Kristien Such as Perl 6.
11:38 alinbsp left
[ptc] colomon: I don't think there is audio on janson_b 11:38
life.fosdem.org says that the presentation stream is without audio 11:39
moritz does a mplayer -vo null jason_a and an mplayer json_b
cdc same as moritz
well "-ao null" for json_b because it has some sound glitch 11:40
[ptc] cdc: thanks, that was a good idea 11:41
11:41 ggoebel111111112 joined 11:42 ggoebel111111111 left 11:43 denis_boyun left, kshannon_ left, kshannon joined
pmichaud Throwback sunday: picasaweb.google.com/lh/photo/o5ix...directlink # Pm presenting in Jansen at FOSDEM 2008 11:43
colomon TimToady!
FROGGS pmichaud: nice picture :o) 11:44
11:45 rurban1 left
masak_grr so, TimToady in 15 minutes? 11:45
11:45 masak_grr is now known as masak
pmichaud masak: that's the schedule 11:45
Kristien :O 11:46
11:46 djanatyn joined
pmichaud There are butterflies on the main screen :) 11:46
11:46 diana_olhovik joined
masak \o/ 11:46
11:46 masak is now known as Guest34903, Guest34903 is now known as masak_grr
jdv79 unless he's at the podium for another reason 11:46
Kristien Is it Get ready to party!?
masak_grr yes!
pmichaud The main screen says we should get ready to party, yes.
Well, it did.
Kristien nice haircut
pmichaud Now it has some scary fibonacci code. 11:47
FROGGS now it shows linenoise instead :P
tadzik haha, we're hiring :)
FROGGS err, what pmichaud++ said
pmichaud Room is filling nicely
masak_grr this is exciting
FROGGS aye
Kristien if only the stream worked
masak_grr I don't have a feed, so I'm grateful for people reporting here on IRC 11:48
pmichaud "Don't cross the streams."
colomon I seem to have lost audio?
masak_grr if only
tadzik all aboard the hype train!
Kristien it shows just one frame
jdv79 is the a feed broke?
11:48 denis_boyun joined
FROGGS colomon: the mic has been muted 11:48
11:48 torbjorn joined
cdc the video is OK, but no audio here 11:49
pmichaud I don't know how long I'll be able to report; my laptop battery is degraded and I don't know how much charge i'll have
colomon I’m getting live TimToady but no audio
FROGGS++
sergot hiho \o
jdv79 huh
rurban was there audio with dana?
masak_grr if TimToady announces Perl 7 today, I will switch to Ruby.
tadzik :D
sergot :D
FROGGS rurban: there was
pmichaud runs powertop to try to save some power
colomon doesn’t know how long he’ll be able to watch …
FROGGS rurban: only on one of the streams of course
colomon needs to start shoveling snow sometime in the next hour 11:50
rurban maybe they'll turn it on later
11:50 diana_olhovik left
manab FROGGS: can i talk to u on personal chat? 11:50
Kristien eh how does mpv.io work
FROGGS manab: sure
cdc for mplayer users, on janson_b: -vf crop=844:698:216:21
timotimo hum
currently i'm seeing a still frame of lwall setting up
Kristien let's try vlc instead 11:51
timotimo ah, better now
pmichaud it's not a still... TimToady is actually frozen. :) :) :)
Woodi_ hallo #perl6 :)
11:51 _dolmen_ joined
timotimo twitter says "i'm hoping for dance moves" 11:51
sergot hi Woodi_
Kristien oh nice it works
timotimo now i can't really tell if there's going to be sound or not
11:51 Ugator joined
masak_grr Dance Dance Perlvolution 11:52
pmichaud lots of people taking pictures of TimToady at the podium with the cover slide behind him
rurban vlc is good for me, just the audio is missing still
pmichaud it's almost as if they expect this to be a historic announcement or something.
jdv79 is he ok?
rurban NLPW?
pmichaud I suspect audio is muted. Wouldn't expect it to be on for another 5 minutes.
jdv79 frozen indeed 11:53
timotimo you know shit just got real when there's two computers on the speaker's desk
one balanced on a trolley, no less :)
masak_grr does anyone have a URL directly to a stream? I'm so bad at this.
colomon hopes he successfully got VLC recording the stream
Woodi_ I switched vlc to mpv and voice is good. also I managed to discover that jason_a and _b are both one presentation :)
rurban live.fosdem.org/watch.php?room=Janson
cdc masak_grr: rtmp://stream-a.fosdem.org:1935/play/janson_a
alpha- no audio on janson_a stream 11:54
timotimo is this about ergonomic placement of laptops?
rurban you can use HTML5, RTMP (the best) or FLV
cdc alpha-: mic is muted
timotimo or trying to get better wifi reception? %)
masak_grr cdc: muchos gracias!
it's working \o/
alpha- is that a thinkpad :) 11:55
rurban yes
alpha- cool
timotimo can you really call it thinkpad if it doesn't have the red light-up dot in the corner any more?
11:56 Mouq left
dylanwh thinkpads don't even have three mouse buttons any more. :( 11:56
rurban Stands this for "No Limit Pro Wrestling" or "Nederlandse Perl Workshop"?
masak_grr my stream went black
jdv79 craigslist?
FROGGS the hall is getting quite full it seems 11:57
alpha- I think he works there
for some time now
jdv79 o
Kristien yay audio
timotimo dylanwh: the x1 carbon seems to
masak_grr ah, got it back by restarting it
cdc sound !
jdv79 audio?
timotimo SOUND! 11:58
masak_grr AUDIO?
cdc not any more
FROGGS NOISE!
moritz wow, two streams in parallel seem to really straini my laptop's battery
it says 1h remaining, before I started them it said 3h 11:59
FROGGS moritz: you might better plug in your a/
a/c*
masak_grr .oO( <laptop> that was before I know you were gonna do THAT )
timotimo air conditioning?
[ptc] what was the command in mplayer to refresh? was it + or something?
11:59 Alina-malina joined
FROGGS timotimo: well, it can keep you warm 11:59
cdc [ptc]: do you mean to resync a!v ? 12:00
a/v
it's "-" and "+"
[ptc] cdc: probably....
cdc: cool, thanks :-)
colomon stream keeps freezing on me. :\
cdc audio is back
tadzik all these people kill the wifi
masak_grr colomon: same
Kristien only the slide stream is broken for me 12:01
timotimo i don't think they are brave enough to be running their streaming setups over wifi
masak_grr got audio o/
colomon arrr! 12:02
masak_grr you too, huh?
alpha- much lag
masak_grr the suspense is killing me!
jdv79 man
timotimo noooo why is the stream now lagging? >:(
colomon yes, audio and/or video cutting out every five seconds or so
12:03 denis_boyun left
cdc hint: use "-vo null" agains jason_a with mplayer 12:03
*against
timotimo god damn it, stream :(
pmichaud room continues to fill up
[ptc] come on stream! you can do it!
tadzik you know the talk is popular when wifi ping is measured in minutes 12:04
(almost)
timotimo is there any way to get a look at the slides, too?
cdc timotimo: rtmp://stream-a.fosdem.org:1935/play/janson_b
timotimo oh
nice 12:05
transparent windows <3
colomon gives up on the stream of TimToady and looks at the stream of the slides
alpha- . o O ( whats this all about )
timotimo the slide stream doesn't do audio, right?
[ptc] timotimo: no
colomon right
masak_grr sticks to TimToady 12:06
12:06 espadrine_ left
timotimo now i have audio and a still frame 12:06
[ptc] you have to do -vo null on the video stream
cdc colomon: use can use "mplayer -vo null rtmp://stream-a.fosdem.org:1935/play/janson_a"
timotimo and it seems the slide stream is kind of waaayyyy delayed from the other stream
colomon I don’t have mplayer here.
mst timotimo: sycnrhonisation is hard
cdc timotimo: use vlc against janson_b, it works better than with mplayer
mst ... apparently also hard to spell
cdc vlc resync automatically 12:07
to sumup a good setup:
mplayer -vo null rtmp://stream-a.fosdem.org:1935/play/janson_a
vlc rtmp://stream-a.fosdem.org:1935/play/janson_b
enjoy :D
12:07 _dolmen_ left
alpha- listening to audio only is a good idea 12:07
no slides anyway
12:08 brrt joined
cdc and restarts mplayer if it says: "Too many buffered pts" 12:08
brrt TimToady++'s talks is nearly watchable
at least on vlc
colomon _b way ahead of _a, yes?
cdc not here 12:09
timotimo i have no clue actually
in what direction it goes
Kristien I should give Catalyst a try.
brrt well, i'm in EU, so that may help 12:10
timotimo me, too
i don't seem to be benefiting greatly from that, though
[ptc] me neither...
gfldex Larry's popularity exceeds the capacity of my intartubes :( 12:12
timotimo i'm already seeing things about perl6 itself on the slide stream, but still hearing talk about tolkiens stories on the other stream >_< 12:13
tadzik it's stll about tolkien
or is it :) 12:14
Kristien Oh, it assumes a diredtory layout
nvm then
12:14 nige joined
cdc Perl 6 makes you travel in time 12:14
masak_grr tadzik: the whole Tolkien universe is all just a metaphor for the Perl 6 project.
timotimo what is a FNORK 12:16
Kristien let's try PSGI
dg mobile.twitter.com/davidgl/status/...2276941824 -- taken from the very back row
masak_grr timotimo: it's what you get when you combine "fnord" with a spork?
timotimo oh, that's a nice big hall 12:17
masak_grr with a nice Wall
timotimo damn you, stream
still frame on wall since a minute or so
but the audio is stable 12:18
when i restart mplayer, the video is animated again, but audio stutters
and i'm pretty sure the slide stream is hanging, too
masak_grr well, we know one thing as of today: when TimToady says "I've got something to say", people come to listen.
masak_grr .oO( schedule it, and they will come )
[ptc] timotimo: have found mplayer -vo null on the video stream helpful
timotimo: and vlc for the slide stream
timotimo: at least it works atm ... 12:19
12:21 rindolf joined 12:22 brrt left
pmichaud I'm suddenly reminded of the Higgs boson announcements at CERN :) 12:23
masak_grr pmichaud: TimToady is using Comic Sans!?
that's it, it's Ruby for me.
:P 12:24
pmichaud currently not comic sans. :)
cdc to stream viewers: _b is about 1 minute ahead of _a
juste pause _b for 1 minute :)
masak_grr looks forward to watching this again in a few weeks without the sound going all i-a-i-i-a-y. 12:25
colomon has to give up and get ready. Somebody post a summary he can read it after rehearsasl? 12:26
[ptc] is looking forward to seeing ovid's and jnthn's talks with audio...
tadzik colomon: I'll write you a tl;dr :)
12:26 sqirrel left
masak_grr colomon: people will probably mention this in the blogosphere :P 12:27
masak_grr checks what twitter is saying currently
timotimo i can easily adjust audio/video delay, but not video/reality delay
moritz pause reality? 12:28
masak_grr DON'T DO IT
moritz well, who would resume it? :-) 12:29
cdc TimToady has started to talk about Perl 6 :) 12:30
timotimo i remember a show on tv when i was a kid about a girl whose father was an alien and that meant she could freeze and thaw time for everyone except her by clapping or touching the tips of her fingertips
masak_grr no!
[ptc] only seeing "companions" the whole time... *sigh*
12:31 rindolf left
moritz [ptc]: restart your player 12:31
[ptc] moritz: been there, done that :0/
12:31 sivoais joined
pmichaud the "companions" slide was up for a long time 12:31
[ptc] now I have "eagles"
pmichaud "moonlight"
I can give the slide words here :) 12:32
"the nick of time"
masak_grr yes, please
[ptc] ah, ok, now I think I've caught up :-)
timotimo %)
pmichaud "where it goes"
cdc pmichaud: with 1 minute of delay, please :)
kjs_ I keep expecting to see some p5/p6 source code
timotimo thank you, pmichaud
[ptc] pmichaud: and we need all the code ;-)
timotimo but those are undelayed apparently? 12:33
[ptc] pmichaud++
pmichaud "when it goes"
masak_grr lost audio
timotimo "skew me"
pmichaud "BEGIN END"
"CHECK INIT" 12:34
timotimo i seem to be exactly synchronized with pmichaud
pmichaud "ENTER LEAVE"
timotimo with my slide stream
pmichaud "KEEP UNDO"
"FIRST NEXT LAST CATCH"
gfldex pmichaud is about 5 minutes ahead of my intartubes :) 12:35
masak_grr yes, that sounds right.
my streamed TimToady just said "eagles".
12:35 Kristien left
masak_grr looking forward to hearing a lot about phasers in 5 minutes :) 12:35
pmichaud constant fib = 0, 1, * + * ... *; 12:36
timotimo "meeting gollum as he just lost his ring"
12:36 alinbsp joined
timotimo save the dwarves from burning trees 12:36
rescue gandalf at least twice
masak_grr bloody eagles
timotimo "phases of activity"
pmichaud (switching to live demo stream)
vendethiel oooh, it works!
timotimo it's a console! 12:37
El_Che do you guys have video?
timotimo yes
El_Che fantastic!
timotimo well, for some value of "have"
masak_grr it's a bit timey-wimey wibbly-wobbly, but it's definitely video
vendethiel yes, video + sound. somewaht 12:38
masak_grr there's some kind of native American on my screen, talking about phasers. I'm confused.
vendethiel wow, that's a big crowd to talk before
timotimo the slide stream is reconnecting every few seconds it seems
12:38 rmgk_ joined
vendethiel how do I get audio, video *and* slides? :P 12:38
El_Che I am *so* glad we move Larry from the devroom to the main track
timotimo you don't
12:38 rmgk left, rmgk_ is now known as rmgk
vendethiel there can only be one vlc :( 12:38
pmichaud "lazy context" 12:39
masak_grr El_Che: yes, that was clearly a wise thing to do.
pmichaud "lazy exceptions"
vendethiel I want to see the slides, but then I don't hear the sound. dang!
timotimo hyper operators <3
kjs_ vendethiel: use mplayer for one stream, vlc for another
timotimo vendethiel: well, you have to receive both streams at the same time, duh
pmichaud "lazy lexers"
vendethiel kjs_: currently on windows
vendethiel lost the slides stream
pmichaud "ANNOUNCEMENT!" 12:40
masak_grr don't shake the camera!
timotimo still frame on larry again ;(
cdc ANNOUCEMENT!
pmichaud "Christmas"
El_Che Christmas
masak_grr o.O
five minutes into the future seems awesome...
tadzik I though he's going to say "I disappear", like Bilbo :P
masak_grr haha
timotimo christmas is going to be actual christmas this time?
tadzik in september:P 12:41
pmichaud "2015"
masak_grr o.O
[ptc] yay, Christmas in September :-)
El_Che big applause
timotimo well, we'll have a bit more time to work on stuff then
vendethiel did someone say september?
vendethiel only sees 2015
pmichaud Yes, September.
tadzik Larry said about beta
masak_grr quick, grab him and drive him away in a black van!
El_Che ok, guys, stop chatting and go back to work ;)
[ptc] TimToady mentioned September earlier in the talk
pmichaud September is consistent with hallway discussions here
"buses"
"volunteers" 12:42
12:42 konsolebox left
pmichaud "snapshot" 12:42
[ptc] hearing about Fib numbers still...
El_Che hey, guys, yoi have the right to die
according to larry about delays :)
masak_grr he's too kind
timotimo "I ha ... a ... a ... ve ... an ... an ... nn .. nn .. ounce ... ent ... lbo ... ys"
pmichaud "there can be only one" 12:43
[ptc] timotimo: are you using vlc for video?
timotimo no
tadzik ooooh
this is big
pmichaud moarvm: first among equals :)
masak_grr come on, stream!
timotimo tadzik: i hope you're talking about a cake
tadzik timotimo: no, big declarations
timotimo oh
like MY $foo? 12:44
masak_grr caught the tail end of the applause
tadzik tl;dr it's okay to leave other VMs behind if they slow down progress
pmichaud "part way there on performance"
timotimo or SSUUBB foobar?
vendethiel the slides stream is totally dying on me
timotimo same here
pmichaud "6.0"
masak_grr o.O 12:45
pmichaud "-Ofun"
vendethiel alright, I guess I'll pass on being so happy with that amazing announcement, and watch the video at some point in the future :-à
s/à/)/
pmichaud "coming of age"
timotimo i completely missed jnthn's talk, didn't i? it was yesterday, corrects
vendethiel timotimo: yes 12:46
did miss it too, because the stream was really bad quality
tadzik: I just got to that part, so -- how's the future?
timotimo did jnthn put up his slides yet?
pmichaud "GLR"
tadzik vendethiel: looking good, more details 12:47
timotimo yes GLR
masak_grr on twitter: "Larry Wall says Perl 6 will be released by Christmas - THIS Christmas. Taking bets."
pmichaud "NSA"
There should be an under/over on that :)
"NFG"
"I FAIL GOOD" 12:49
masak_grr :)
timotimo oh, huh
i've always pronounced "preface" as "pre-face"
12:49 FROGGS left
masak_grr timotimo: English is wonky. 12:49
[ptc] masak_grr: oh yes
timotimo i'm pretty hype for the GLR still 12:50
[ptc] man, the audio is waaay behind the slides
my audio just got to NSA...
[ptc] restarts
timotimo restarting many times didn't help me sync up slides and video at all :( 12:51
cdc we are still stuck 3 minutes in Larry's past :)
*3 minutes back
El_Che He just announced that the september date was a joke and the release will be in april 12:52
(just kidding ;) )
timotimo oh che you so silly
[ptc] maybe the release should be at Easter - then it could be an Easter Egg
timotimo %) 12:53
i personally am not very interested in a fully perl6-bootstrapped rakudo
but since it'd have to mean our perl6 compilation has to be at least as efficient as the nqp compilation we have now ... that would be nice to me
pmichaud applause applause applause some standing o's 12:54
cdc "»ö« perl6.org" 12:55
"NO SIGNAL"
:D
timotimo »ö«
pmichaud yeah, I skipped the last slide. :)
EOT
cdc Thanks pmichaud++ !
masak_grr pmichaud++
[ptc] audio still coming through...
pmichaud++
12:56 masak_grr is now known as masak
Ugator when will it be on youtube? 12:56
timotimo huh, it's suddenly over?
i just restarted mplayer so that i could listen better
turns out the hour is already over :o
signs of a good speaker, i suppose :3 12:57
btyler just ended a minute ago
timotimo there's a pretty person right in front of the camera
rurban I'm still at 52:40 in vlc
masak Ugator: that's a good question but not usually possible to answer with any certainty.
12:57 KPTN joined
btyler great talk, the message was just right. TimToady++ 12:58
Ugator sure:) so no one here has the original recording?
rurban a fully perl6-bootstrapped rakudo would have a much easier to debug compiler
cdc Ugator: last year it took 5 to 10 days to get the recordings on video.fosdem.org 12:59
12:59 manab left, clement joined, KPTN left
jdv79 just ended here 13:00
nice
13:02 diana_olhovik joined
masak we might need full Perl 6 bootstrapping for everything we want to do with macros. the jury is still out on that one. 13:03
arnsholt Would definitely make things easier, yeah 13:04
timotimo OK 13:05
masak right. same reasons 6model bootstraps, basic'ly.
arnsholt But I'd hope access to the compiler pipe-line will make most things possible
13:06 rindolf joined
masak that, too. 13:06
13:06 diana_olhovik left 13:10 diana_olhovik joined, Kristien joined 13:11 clement left
Woodi_ it is a bit silly to stream slides :) but presenter uploading slides up fron is no-no too (ruined announcements...) some html pages with PUSH button in front of presenter would be nice... 13:12
timotimo let's invent Yet Another Solution! 13:13
Woodi_ timotimo: yea ! :)
13:14 diana_olhovik left
Kristien I like what InfoQ does with slides. 13:14
but those aren't livestreams 13:15
vendethiel Kristien: agreed. 13:17
moritz aye 13:22
masak I realize I haven't been to SlashDot for ages. but now I'm visiting the place just to see if the Perl 6 announcement will show up there.
timotimo do you really want to go to that place?
moritz it's just a bit rusty, otherwise not too bad 13:23
13:23 diana_olhovik joined
moritz ... until you read the comments :-) 13:23
timotimo i thought those sites were mostly about the comments ...
masak yes. I have enjoyed SlashDot quite a bit over the years. I seem to have fallen out of the habit, but I don't really see how it's worse then, say, reddit or HN.
even the comments are OK if you just stay above a certain score threshold.
timotimo HN is terrible, too :) 13:24
moritz science-beta.slashdot.org/story/15/...s-of-light was done a by good friend of mine, btw
timotimo reddit is a very mixed bag, depending on where you end up
moritz I mean the science, not the article
masak moritz: next up: Klein bottle from beams of light!
timotimo it took me quite a bit of restraint to not make that comment, moritz :) 13:25
masak moritz: (just take two Möbius strips made of light and glue them together)
moritz :-)
timotimo "light glue" is glue with reduced fat and sugar contents? 13:28
masak no, it's the cheerful version of dark glue 13:29
13:29 denis_boyun joined 13:30 raiph left, diana_olhovik left 13:31 raiph joined 13:32 rindolf left, FROGGS joined 13:34 denis_boyun left
nine I'm trying to embed libmoar, but I then get "Unhandled exception: failed to load library '/home/nine/install/rakudo/install/languages/perl6/runtime/dynext/libperl6_ops_moar.so'". Any idea what I'm doing wrong? 13:38
masak m: class Foo { has Hash of Hash $.foo = {} }; Foo.new
camelia rakudo-moar 6e182d: OUTPUT«Type check failed in assignment to '$!foo'; expected 'Hash[Hash]' but got 'Hash'␤ in block at src/gen/m-CORE.setting:1023␤ in method BUILDALL at src/gen/m-CORE.setting:978␤ in method bless at src/gen/m-CORE.setting:966␤ in method new at src/ge…»
masak resolves #116212 13:39
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=116212
FROGGS nine: perhaps you have to preload that perl6_ops library 13:40
13:44 Colomon_ joined
Colomon_ So the tl;dr is September release, maybe just of moar? 13:45
FROGGS Colomon_: that seems to be the plan 13:46
tadzik it's possible
13:46 manav joined
nine September beta release, Christmas 6.0.0 13:46
13:46 Colomon_ left
masak what will 6.0.0 mean in practice? how will the world be different after that? 13:48
13:48 jschulz_ is now known as jschulz
FROGGS masak: that depends on the reaction of the world I suppose 13:49
vendethiel masak: does the world need to be different?
I'm fairly happy with the world in its current state (ok, I'm kidding)
FROGGS but, for us it means that we perhaps have to be more about backwards compatibility
masak vendethiel: it's an allowed answer to say "it won't be different"
vendethiel masak: I'm Mu-ing your question, asking why that question :P
13:50 brrt joined
masak yes, got it. 13:50
13:50 FROGGS left
masak I'm pretty sure the difference for *me* will be fairly minimal, but only because I became an avid Perl 6 user long ago. 13:50
13:53 FROGGS[mobile] joined
tadzik hmm, "Too few/many positionals passed" is an X::AdHoc. Unexpected 13:54
any reason why it doesn't have its type? moritz?
masak the answer is probably "patches welcome" 13:55
tadzik time to get to work then :)
13:55 grondilu joined
tadzik oh, it's at Moar level 13:57
grondilu jeez why such high resolution for the FOSDEM livestream. I can't watch that properly on my machine :(
tadzik that probably makes it trickier 13:58
grondilu I mean 1920x1088 come on, there is no need for that many pixels. 13:59
14:00 _dolmen_ joined
sjn \o 14:00
tadzik oh hai sjn 14:01
sjn tadzik: where's everyone hanging out, btw? :)
tadzik sjn: I'm in Jonsomething hall, on the typesetting talk 14:02
sjn aah
Simon Cozen's thing
tadzik I thinkmost of the 6folks are here
yes
14:03 manav left 14:06 jschulz left 14:08 rindolf joined
vendethiel what's the current talk? 14:14
arnsholt nine: Regarding the so loading, basically what FROGGS said. You need to make sure the so file is in the search path of the dynamic linker when your app with the embedded moar runs 14:15
tadzik vendethiel: fosdem.org/2015/schedule/event/int...cing_sile/ 14:16
vendethiel thanks 14:18
14:19 sqirrel joined
nine One step further: /home/nine/install/rakudo/install/languages/perl6/runtime/dynext/libperl6_ops_moar.so: undefined symbol: MVM_frame_inc_ref 14:19
grondilu ok it seems to work not too bad with a proper filter: mplayer -vf scale=800:-2 stream-a.fosdem.org:80/janson_a.flv 14:21
arnsholt nine: When does that happen_ 14:23
s/_/?
grondilu spoke too fast, increases cache
14:24 muraiki_ joined, brrt left
nine Ok, If I LD_PRELOAD libmoar.so, it works. So I guess I have not made it sufficiently clear to Inline::C that I need to link libmoar.so 14:26
arnsholt Yeah. That's what the error message made me think as well 14:27
Definitely looked like the moar lib wasn't loaded
14:33 zebardy joined, _dolmen_ left
nine According to ldd, my .so is actually linked to libmoar.so: 14:39
ldd /home/nine/Inline-Perl6/_Inline/lib/auto/Inline/Perl6_a71e/Perl6_a71e.so linux-vdso.so.1 (0x00007fffa751d000) libmoar.so => /home/nine/install/rakudo/install/lib/libmoar.so (0x00007f4b036c5000)
14:46 telex left
tadzik sjn: plans for the next talk? 14:47
FROGGS[mobile] what are we going to do now?
tadzik hallway track? :) 14:48
14:48 telex joined
tadzik K building is always fun 14:48
FROGGS[mobile] the booth, aye 14:49
14:50 raiph left 14:53 kjs_ left, FROGGS[mobile] left
nine jnthn: you around? 14:55
Does anyone know where I might find jnthn? 14:56
vendethiel nine: around FOSDEM? *g*
14:57 rurban1 joined, dj_goku joined, dj_goku left, dj_goku joined
Kristien does moarvm use a stock GC or a custom one? 14:58
gfldex custom
masak there are stock GCs? o.O 15:04
pmichaud nine: jnthn is in the main talks room (sitting next to me)
15:05 rurban1 left, rurban1 joined, FROGGS[mobile] joined 15:07 rindolf left
Kristien masak: there is Boehm 15:08
masak ok, TIL 15:09
Woodi_ in September we can expect beta probably. but look what Perl6 did to spec and developmen time! imagine what could happen to the deadline date! ;) 15:12
masak I'm imaginig a buncha git commits
Woodi_ it's technical detail :) 15:13
but moar is embaddable ??
15:15 kjs_ joined 15:16 skids joined 15:17 Kristien left
grondilu ok, it's official: the FOSDEM livestream sucks. 15:20
15:21 Oe_eden left 15:23 FROGGS[mobile] left, Arma joined 15:24 Arma left, rindolf joined 15:25 raiph joined 15:26 rurban left
vendethiel :( 15:26
15:27 zakharyas joined, zebardy left 15:30 BenGoldberg joined 15:32 FROGGS[mobile] joined 15:36 FROGGS[mobile] left 15:42 kjs_ left
[Tux] is back home, leont slept the complete journey :) 15:46
15:52 gfldex left
rindolf [Tux]: meow! 15:54
15:55 kaare_ left 15:56 sqirrel left
skids wonders how long it takes a 3-liter bottle of cola to melt at room temperature. And wants caffeine. 15:57
15:58 grondilu left
skids But if I open that bottle now, it will be nothing but foam and misery. 15:58
16:00 spider-mario left
pierrot Good afternoon. Has TimToady's talk been uploaded somewhere ? I couldn't watch it this morning. 16:07
vendethiel pierrot: not yet. wait a few weeks 16:11
16:25 molaf joined
masak still nuthin' on Slashdot. 16:25
and here I was looking forward a bit to reading lame comments about DNF and the Hurd... 16:31
ugexe_ can i iterate over a parse-tree (of type NQPMatch) without parsing the entire thing as a string? For instance, if i wanted to look for every instance of (in its dumped form) `'sym: use' 'module_name: (.*?)'` (*realizes how helpful perl6 object introspection is*) 16:32
16:32 LLamaRider joined
masak ugexe: sounds like a job for a well-placed multi sub. 16:34
(a recursive one, that knows how to descend an NQPMatch) 16:35
ugexe hmm maybe im going about my goal the wrong way. for trying to extract dependencies of a script, does looking at the QAST make sense (so I dont have to rely on my own regex to parse source code)? 16:44
16:45 alinbsp left 16:47 fil_ joined
fil_ hey 16:47
yoleaux 06:38Z <moritz> fil_: IMHO it's best practice to put each class that is publicly accessible into a separate class. As for rest, use your best judgement :-)
fil_ I'm using Padre to run my perl6 script. But it's not finding my own modules in ./lib. What's best practice around this stuff? 16:48
in Eclipse, there's a concept of a project that has a classpath. Does Padre have anything like this? 16:49
ugexe use lib 'lib', or add -Ilib to its perl6 cmd line invocation
ugexe doesnt use padre, i just assume you can change the perl6 target command
fil_ all the online modules stick their submodules in "lib." I assumed this meant there is a convention on perl6 that lib is automatically added to @INC 16:50
16:50 espadrine_ joined
fil_ what IDE do you use? 16:50
ugexe no, what if you want to use the installed module?
i use sublime text
also what if you want to test blib and not lib 16:51
thats why you have to explicity include them
pierrot vendethiel: thanks
ugexe i.e. perl6 -Ilib t/00-basic.t
itz_ hmm warning to anyone just googling "La Bécasse" .. there are two and the correct one is by ULB 16:52
ugexe perl6 t/00basic.t would make any modules loaded by 00basic.t be from installed versions
16:54 anaeem1 joined 16:55 konsolebox joined 17:03 Kristien joined 17:05 anaeem1 left 17:06 anaeem1_ joined 17:10 anaeem1_ left 17:12 konsolebox left 17:13 espadrine_ left
skids r: role A does Positional { has @!children handles 'postcircumfix:<[ ]>'; }; my $tree = A.new; $tree[0] = 1; # RT#68138 code now golfs to this so it is no longer a parametric type problem. 17:13
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...l?id=68138
camelia rakudo-parrot 6e182d: OUTPUT«Cannot modify an immutable value␤ in method assign_pos at gen/parrot/CORE.setting:2093␤ in method assign_pos at gen/parrot/CORE.setting:2084␤ in sub postcircumfix:<[ ]> at gen/parrot/CORE.setting:3298␤ in sub postcircumfix:<[ ]> at gen/parrot…»
..rakudo-moar 6e182d: OUTPUT«Cannot modify an immutable A␤ in method assign_pos at src/gen/m-CORE.setting:2083␤ in sub postcircumfix:<[ ]> at src/gen/m-CORE.setting:3287␤ in block <unit> at /tmp/tmpfile:1␤␤»
17:13 anaeem1 joined 17:15 FROGGS[mobile] joined
hzhou321 Is there any one who do not like perl6's regex syntax as much as perl5's? 17:15
moritz hzhou321: with the number of people out there, I would be surprised if *everyone* liked the Perl 6 syntax better 17:16
skids hzhou321: not sure it is implemented yet, but yes.
FROGGS[mobile] hzhou321: unlikely... I mean, compare non-capturing groups, positive and negative lookahead of both 17:17
hzhou321 moritz: but I need to know whether it is my problem to conquer.
17:17 Bucciarati joined
FROGGS[mobile] m: say "foo" ~~ m:p5/foo/ 17:17
camelia rakudo-moar 6e182d: OUTPUT«===SORRY!=== Error while compiling /tmp/_AgZtd5KvI␤Adverb p5 not allowed on m␤at /tmp/_AgZtd5KvI:1␤------> say "foo" ~~ m:p5/foo/⏏<EOL>␤»
FROGGS[mobile] grrr 17:18
moritz m: say 'foo' ~~ rx:Perl5/foo/
camelia rakudo-moar 6e182d: OUTPUT«「foo」␤»
moritz m: say 'foo' ~~ m:Perl5/foo/
camelia rakudo-moar 6e182d: OUTPUT«「foo」␤»
skids m: say "foo" ~~ m:P5/foo/ 17:19
camelia rakudo-moar 6e182d: OUTPUT«「foo」␤»
moritz hzhou321: if there is a problem, it's most likely yours to conquer, yes
skids That's what the spec says.
17:19 pmurias joined, FROGGS[mobile]2 joined
hzhou321 moritz: taking that answer, I wouldn't really have any self-esteem :) 17:20
pmurias is Perl 6 really planned for official release this christmas?
17:21 LLamaRider left
moritz pmurias: that's what TimToady said, yes 17:21
FROGGS[mobile]2 pmurias: don't has to be christmas, but yeah, this year
hzhou321 FROGGS: non-capturing groups, positive and negative lookahead is a bit awkard in perl 5, but on the other hand, they are not used often. And when they are used, they are often logically complex and it requires some comments anyway. 17:22
FROGGS[mobile]2 hzhou321: not so in P6
17:22 FROGGS[mobile] left
FROGGS[mobile]2 I use them more often, because I can easier do so 17:23
geekosaur I find I use non-capturing groups more often than captiring groups in p5
17:23 FROGGS[mobile]2 left
geekosaur but the syntax is awkward, which I am glad is fixed in p6 17:23
pmurias is it just an estimate when it will be decently usable or are there any more specific plan for achiving that?
hzhou321 So I gather that every one here all prefers perl6's regex syntax to perl5's?
geekosaur pretty much, yes 17:24
moritz hzhou321: it takes a while to get used to if you've done a lot of perl/pcre regex
hzhou321: but it's totally worth it
geekosaur p5's regex syntax is based on the original regex syntax, which has been twisted well beyond what it was originally intended to do
p6 reimagined regex syntax to be easier to understand and more convenient
moritz hzhou321: not just the syntax, but also the possibilities for modularization
geekosaur (perl isnt even the only thing that twistred the original regex syntax; traditional unix had a bunch of regex variants which POSIX boiled down to two main ones) 17:25
hzhou321 geekosaur: the original regex syntax is logically simple and I never need to re-learn even before Perl. Perl6's is like an new realm.
geekosaur um, you've never had BRE vs. ERE confusion? 17:26
pmurias hzhou321: yes, but it offer a bunch of possibilities
geekosaur never had to deal with variant regex implementations like p5 vs. PCRE?
hzhou321 geekosaur: I avoid them, and it works for me.
geekosaur you're very lucky if you can avoid them
masak mdk++ blogged: blogs.perl.org/users/shadowcat_mdk/...stmas.html 17:27
geekosaur I end up providing a lot of regex support, with BRE vs. ERE and PCRE vs. ERE vs. other variants, on some other channels
hzhou321 I use m/.../gcx a lot for complicated cases when full parsing is required
masak and we have HN comments (currently empty): news.ycombinator.com/item?id=8980387
geekosaur maybe you can just decree that everything shall be done in perl5 past some particular version that guarantees the existence of certain features. most people don't have that luxury 17:28
17:28 FROGGS[mobile] joined
geekosaur and, frankly, old style regex is a cesspool 17:28
long past time someone dumped chlorine in it
jdv79 Perl6 v1.0 - nice. how about 6.0 or 6.0.0 like in reality. 17:29
FROGGS[mobile] that's what I'd like to call it too 17:30
fil_ I'm just about ready to give up on perl6 17:31
Soooo disappointed
jdv79 you might be in the wrong chan then;)
fil_ For me the nail in the coffin is no IDE that lets me run tests
That's a showstopper
jdv79 that's a lame reason but ok 17:32
geekosaur generally thinks need for an IDE is a language smell
hzhou321 I agree with geekosaur. But with perl, you can whip up quick automation easily anyway. 17:33
fil_ I'm a hardcore perl fanboy. I helped put one of the first newspapers of record in the world on the Internet using perl. perl's recent assension on the TIOBE index got my hopes up. But the tooling just isn't there.
Kristien geekosaur: youtu.be/LH75sJAR0hc?t=10m12s
jdv79 fil_: maybe you could pen a ticket or an article or a blog post 17:34
someone might be able to look into it eventually
hzhou321 Kristien: I like this guy :)
geekosaur I will also note that if your criteria for a successful language include maturw tooling out of the box, then you will never be able to adopt any new language or technology
Kristien yeah he's incredibly funny
fil_ Mature tooling is a sign of language adoption. 17:35
geekosaur and the world will leave you behind, safely boxed in by your ancient tooling
that's still an impossible loop
fil_ Lack of tooling tells me that nobody except a few old vim hardcores is actually using perl6 for meaningful work.
geekosaur it has to be adopted in order for it to ever be adopted?
Kristien there are still languages that are widely adopted and still have no decent tooling whatsoever 17:36
pmurias Kristien: listening to that guy for me is more annoying than tracking down segfaults in generated code
jdv79 that might be true cause its not mature yet
geekosaur again, this leads to never being able to adopt anything new
jdv79 but you could help by enumerating exactly what you might like to have
fil_ Maybe it's just too early. I was under the impression that perl6 has been out for a while.
geekosaur uh, no
17:37 FROGGS[mobile] left
pmurias fil_: not in a useful form 17:37
geekosaur there is no formal release yet, we're hoping for one sometime this year
fil_ oh ok
well my criticism is premature then
I'll check back in a couple of years.
I need a tool I can use today.
geekosaur there are prereleases of various kinds, which one can typically expect to be played with by some folks but nobody's going to do anything serious with them
fil_ The Grammar / Actions stuff is amazing!
jdv79 yeah, didn't you hear? the attempt is for 6.0 release on 2015-12-25:)
pmurias but currently in terms of tooling we are not yet on 5.0.0 level
17:38 xinming joined
pmurias as writing Makefiles by hand is required :( 17:38
muraiki_ elixir has some incredible tooling for a new languge
fil_ I'm used to my IDE managing build paths, running tests etc.
Does perl6 even have a concept of a project? Is there an equivalent of a "pom" file for perl6? 17:39
geekosaur indeed, a formal release will make better tooling *possible* because there will finally be something that isn't likely to change radically at the drop of a hat
(we still have the GLR pending, right?)
jdv79 yes
pmichaud said he'll shoot for march
fil_ While I'm waiting for an IDE, how should I run my tests?
moritz ufo && make test
fil_ Does everyone just run their tests commandline?
Kristien pmurias: I don't have that problem, since I don't have segfaults.
moritz github.com/masak/ufo/ 17:40
ugexe prove -e 'perl6 -Ilib' t/ or just 'perl6 -Ilib t/file.t'
masak confesses that he does not use ufo much these days
17:40 xinming_ left
moritz PERL6LIB=lib prove -e perl6 17:41
jdv79 masak: what do you use?
moritz PERL6LIB=lib prove -e perl6 t/
masak I used it when compilation times were a big problem, and figuring out the compilation order mattered to bring the time down.
jdv79: I use `prove -e 'perl6 -Ilib' -r`
17:42 FROGGS[mobile] joined
masak should make a one-letter alias for that 17:42
pmurias masak: ufo figures out in which order it should do precompilation? 17:43
masak yes, that was its core feature once upon a time.
moritz it doesn't anymore, though
it merely extracts dependencies
and lets 'make' figure out the order 17:44
pmurias hopefull automatic precompilation is a part of 6.0.0 requirements
fil_ t\FormatQFX.t .. ''perl6' is not recognized as an internal or external command, operable program or batch file.
when I run prove -e 'perl6 -Ilib' t/
17:44 anaeem1 left
masak works here. 17:44
pmurias do you have perl6 installed?
moritz fil_: seems your perl6 is either not installed, or not in $PATH/%PATH 17:45
fil_ perl6 works commandline 17:46
it's installed 17:47
damn gtg
thanks for your help!
17:47 FROGGS[mobile] left
ugexe try double quotes depending on your OS 17:47
moritz or if you don't run it through the shell, make sure that your launcher has the same environment as the shell
tadzik: most binder errors are X::AdHoc because the binders are backend specific, and thus hard / much work to port to typed exception 17:49
17:52 FROGGS[mobile] joined 17:55 atroxaper left, rurban joined 17:57 kurahaupo1 joined
vendethiel hzhou321: coming in a bit late, if I had to choose, the simple fact that grammars exist would sell me on p6's syntax 17:57
I admit I don't know p5's regexes, I only know "PCRE" (the subset that every language, or almost, has, but often means no named captures (though ruby has them, yay))
vendethiel saw a talk at fpw14 about those and was impressed, but scared at the same time
Kristien PCRE is good. 17:58
rurban no named captures and unicode names, but it is jitted 17:59
jdv79 reaaly? i've generally found the recent p5 re engine very fast 18:02
18:02 alinbsp joined
jdv79 just mean i'm not sure jitting that would make much difference; to the cases i've seen 18:02
18:03 FROGGS[mobile] left, Vlavv_ joined
Kristien But PCRE has support for Unicode right? 18:03
I will most likely use it soon.
18:04 andreoss joined 18:10 FROGGS[mobile] joined 18:17 prime joined
rurban Sure 18:18
18:18 FROGGS[mobile] left
rurban UTF-8/16/32 encodings and Unicode general category properties, However not our name aliases 18:19
It also contains a faster non-backref matcher (as in re2)
masak inspired by Russ Cox's article? 18:21
fil_ ugexe: thanks you were right, it was the quotes
masak swtch.com/~rsc/regexp/regexp1.html 18:22
rurban probably. he was not the first, but the most prominent
and perl5 still has no simple backref detector to fallback to the faster version 18:23
18:25 raiph left, FROGGS[mobile] joined 18:28 FROGGS[mobile] left
Kristien rurban: nice 18:28
fil_ How do I get Grammar.parsefile to use relative paths? It only ever works with absolute paths on filenames 18:36
Kristien make the path absolute 18:40
fil_ If i publish a module, I can't use absolute paths 18:43
all the online examples use relative paths
there must be some way to get perl6 to know where you are
moritz fil_: it works on relative paths, relative to the current working directory
m: say $*CWD
camelia rakudo-moar 6e182d: OUTPUT«"/home/camelia".IO␤»
flussence std: sub foo(Pair (:key($a), :value($b))) { say $a + $b } # Are any of these parens removable? Code works but it looks icky... 18:44
camelia std f9b7f55: OUTPUT«ok 00:01 142m␤»
moritz m: grammar A { token TOP { (. ** 10) .*? } }; say ~A.parsefile('evalbot.log')[0] 18:45
camelia rakudo-moar 6e182d: OUTPUT«use of uninitialized value $v of type Any in string context in block <unit> at /tmp/OrS_EGaU1L:1␤␤␤»
fil_ got it. just realized that cwd is parent of t/ when running tests
vendethiel fil_: cwd is "where you ran the command", usually :o)
moritz ... unless somebody chdir()ed or assigned to $*CWD 18:46
flussence fil_: if you want to know where your module is, use $?FILE
moritz m: say $?FILE
camelia rakudo-moar 6e182d: OUTPUT«/tmp/8Z_xUrlJMN␤»
18:49 _dolmen_ joined
fil_ nice. I like $?FILE much better. thanks! 18:53
Kristien m: say say.WHAT 18:56
camelia rakudo-moar 6e182d: OUTPUT«===SORRY!=== Error while compiling /tmp/0sSacchkfa␤Unsupported use of bare 'say'; in Perl 6 please use .say if you meant $_, or use an explicit invocant or argument␤at /tmp/0sSacchkfa:1␤------> say say⏏.WHAT␤ …»
raydiak std: sub foo(Pair (:$key, :$value)) { say $key + $value } # flussence: if you can rename $a and $b, this is somewhat more readable 18:58
camelia std f9b7f55: OUTPUT«ok 00:01 142m␤»
raydiak good morning #perl6
fil_ can perl6 classes have private methods? 19:00
raydiak fil_: yes, just start the method name with a !
fil_ thanks. that should probably be mentioned on doc.perl6.org/language/classtut 19:01
raydiak it's at github.com/perl6/doc/blob/master/l...asstut.pod if you want to help us improve it, or even just file issues 19:05
m: say say('').WHAT; say &say.WHAT 19:10
camelia rakudo-moar 6e182d: OUTPUT«␤(Bool)␤(Sub)␤»
19:11 zakharyas left
Kristien :O 19:11
19:11 colomon left
masak "From the bottom of my heart, thank you. I have been keenly looking forward to this for a long, long time." -- blogs.perl.org/users/shadowcat_mdk/...nt-1536111 19:12
19:13 colomon joined
masak <3 .classify 19:18
19:18 Mouq joined
colomon masak wants to classify things less than 3? 19:18
masak 😻 .classify 19:19
19:20 anaeem1_ joined
flussence
.oO( my terminal emulator doesn't support cat :( )
19:20
19:21 Rounin left
vendethiel
.oO( but does it support 😻 ? )
19:21
Kristien m: '😻'.classify.say 19:24
camelia rakudo-moar 6e182d: OUTPUT«Cannot call 'classify'; none of these signatures match:␤:(Any $: Any $test, *%_)␤:(Any $: Any $test, Any :into($into)!, *%_)␤ in block <unit> at /tmp/Jtb5GTs13y:1␤␤»
masak m: say {}.perl
camelia rakudo-moar 6e182d: OUTPUT«{}␤»
Kristien m: '😻'.perl.say
camelia rakudo-moar 6e182d: OUTPUT«"😻"␤»
masak m: say (1, 2, 3).classify(+*)
camelia rakudo-moar 6e182d: OUTPUT«1 => 1, 2 => 2, 3 => 3␤»
Kristien m: say {}.WHAT
camelia rakudo-moar 6e182d: OUTPUT«(Hash)␤»
masak m: say (1, 2, 3).classify(+*).perl
camelia rakudo-moar 6e182d: OUTPUT«Hash[Any,Any].new(1 => [1], 3 => [3], 2 => [2])␤»
Kristien How do you make the empty block?
m: say ->{}.WHAT
camelia rakudo-moar 6e182d: OUTPUT«(Block)␤»
raydiak m: say {;}.WHAT # probably works too 19:25
camelia rakudo-moar 6e182d: OUTPUT«(Block)␤»
19:26 colomon left
Kristien m: say { Nil }.WHAT 19:27
camelia rakudo-moar 6e182d: OUTPUT«(Block)␤»
masak why do we prefer `Hash[Any,Any].new(1 => [1], 3 => [3], 2 => [2])` in .perl over `{ 1 => [1], 3 => [3], 2 => [2] }` ?
Kristien m: say (->{})().perl; say ({ Nil })().perl
masak something to do with item context?
camelia rakudo-moar 6e182d: OUTPUT«Nil␤Nil␤»
raydiak well, a hash literal from { ... } isn't a Hash[Any,Any], is it? Hash[Str,Any], I thought 19:29
masak hm, wait. Hash[Any,Any] isn't the default. Hash[Str,Any] is.
raydiak: right.
just realized.
raydiak maybe it should do :{ ... } instead?
in the specific case of [Any,Any], of course
Kristien maybe it complicates nesting in other .perl methods from other types
raydiak m: say :{ 1 => 2 }.perl 19:31
camelia rakudo-moar 6e182d: OUTPUT«:{1 => 2}␤»
raydiak hm
19:32 kjs_ joined
masak m: sub &foo() {}; say "alive" 19:32
camelia rakudo-moar 6e182d: OUTPUT«===SORRY!=== Error while compiling /tmp/GE7JBiJGWP␤Missing block␤at /tmp/GE7JBiJGWP:1␤------> sub ⏏&foo() {}; say "alive"␤ expecting any of:␤ new name to be defined␤»
masak (a) I'm wondering if this shouldn't be allowed, and mean the same as `sub foo`
(b) the error message is LTA, but I don't have any good ideas for what it ought to be 19:33
19:33 gfldex joined, colomon joined 19:39 _mg_ joined
raydiak wrt (b), maybe "invalid sub name" or "invalid character in sub name" or along those lines? perhaps additionally a specific mention of no leading '&' if that's the first char it saw in the name? 19:40
I guess it'll be the same message we see when botching the body of an anonymous sub too, though...
raydiak wouldn't mind allowing a leading &, either, but the message will still be LTA in other cases 19:41
19:44 _dolmen_ left, colomon left, colomon joined
masak troo 19:45
masak submits rakudobug
Kristien I never understood typeglobs and file handles in Perl 5. 19:46
They seem obscure.
geekosaur they are obscure. worse, they're kinda retrospectively imposing a story on top of what was originally just syntax 19:47
masak there's a reason nobody suggested having typeglobs in Perl 6 ;) 19:48
Kristien masak: let's change that!
geekosaur (perl3.000 had filehandles, but typeglobs were grafted in later)
Kristien How about we add typeglobs to Perl 6?
geekosaur let's not and say we did (p6 has more principled ways of doing it instead of abusing existing dubious syntax) 19:49
masak doesn't feel particularly compelled to repeat the typeglobs lesson in Perl 6 19:50
fil_ how do I get the type of an object?
masak m: say 42.WHAT
camelia rakudo-moar 6e182d: OUTPUT«(Int)␤»
masak m: 42.^name
camelia ( no output )
masak m: say 42.^name
camelia rakudo-moar 6e182d: OUTPUT«Int␤»
masak fil_: either of those ways. 19:51
fil_ I tried WHAT and got Cannot look up attributes in a type object
masak fil_: the .WHAT is actually not a string, but an object representing the type which stringifies to "(<type>)"
fil_: then you did something other than print it
_mg_ Hello all! I'm new here. Heard Larry today on FOSDEM, and Perl6 talks yesterday. Now I need to find my way to Perl6. :D I'll start by updating the homebrew formula (Mac OS) to the latest rakudo release (github.com/Homebrew/homebrew/pull/36437).
fil_ ^name worked. thanks
masak _mg_: welcome ;) 19:52
_mg_: that sounds awesome \o/
_mg_ masak: thanks!
geekosaur didn't someone say the other day that the homebrew formula was out of date?
masak probably, but also probably no-one did anything about it. 19:53
_mg_++
19:54 colomon left
_mg_ With that a "brew install rakudo-star" is all you need to get it running on Mac OS. 19:54
raydiak methinks we just pointed the person the other day at rakudobrew 19:55
Mouq _mg_++: It looks like that only knows about jvm and parrot, and builds parrot by default 19:56
masak ugh.
build moar by default, please!
19:58 rurban1 left, darutoko left 19:59 colomon joined
masak an I the only one who finds it confusing to talk about "Perl 6 v1.0" ? 20:00
vendethiel masak: no, I agree 20:01
_mg_ masak: ok I'll try
raydiak agreed, 6.0 or 6.0.0 is much much better imo 20:03
b2gills There has been talk of calling the latest release of Perl 5 "Perl 5 version 20"
masak I mean, I can see where those people are coming from, making the "6" just part of the name. 20:04
but it'll be even more confusing as we get to Perl 6.1
what'd that be, version 1.1 of Perl 6?
20:04 Kristien left
moritz version 0.1 of Perl 6 :-) 20:05
masak moritz: so we go from v1.0 to v0.1? no thanks... 20:06
(I'm speaking purely hypothetically, by the way. I don't believe we should encourage this versioning at all.)
b2gills It's about like talking about the first annual ___.
masak I think we should be very clear that the version being released is called Perl 6.0.0 20:07
huf b2gills: it's not just talk, perl -v reports itself like that
i dont think that's a new thing
20:07 bcode joined
raydiak braces for more rounds of "you shouldn't have called it Perl 6" from outside the echo chamber as the year passes and release gets closer :P 20:07
b2gills huf: I didn't realize they had actually made that change
masak m: say "A B" ~~ /A B/
camelia rakudo-moar 6e182d: OUTPUT«Nil␤»
masak m: say "A B" ~~ /:s A B/
camelia rakudo-moar 6e182d: OUTPUT«「A B」␤»
20:09 kjs_ left, rindolf left
pmurias Perl6 v1.1 seems both descriptive and rather silly 20:10
20:10 kjs_ joined
masak m: "A, B, C" ~~ /(\w)* % ', '/; say $0.list.perl 20:10
camelia rakudo-moar 6e182d: OUTPUT«Array.new(Match.new(orig => "A, B, C", from => 0, to => 1, ast => Any, list => ().list, hash => EnumMap.new()), Match.new(orig => "A, B, C", from => 3, to => 4, ast => Any, list => ().list, hash => EnumMap.new()), Match.new(orig => "A, B, C", from => 6, to…»
masak m: "A, B, C" ~~ /(\w)* % ', '/; say ~>>$0
camelia rakudo-moar 6e182d: OUTPUT«===SORRY!=== Error while compiling /tmp/f4_tR8JOQM␤Preceding context expects a term, but found infix ~> instead␤at /tmp/f4_tR8JOQM:1␤------> "A, B, C" ~~ /(\w)* % ', '/; say ~>⏏>$0␤»
masak m: "A, B, C" ~~ /(\w)* % ', '/; say ~<<$0 20:11
camelia rakudo-moar 6e182d: OUTPUT«A B C␤»
masak m: "A, B, C" ~~ /(\w*) % ', '/; say ~<<$0
camelia rakudo-moar 6e182d: OUTPUT«===SORRY!=== Error while compiling /tmp/htQFolDBz4␤Missing quantifier on the left argument of %␤at /tmp/htQFolDBz4:1␤------> "A, B, C" ~~ /(\w*) %⏏ ', '/; say ~<<$0␤»
masak I expected those last two to be identical.
usually it doesn't matter if I put the quantifier inside the parens, or outside.
m: "ABC" ~~ /(\w)*/; say ~<<$0 20:13
camelia rakudo-moar 6e182d: OUTPUT«A B C␤»
masak m: "ABC" ~~ /(\w*)/; say ~<<$0
camelia rakudo-moar 6e182d: OUTPUT«ABC␤»
masak hrm :)
...never mind :) 20:14
b2gills I think of % as being a modifier on * and + 20:15
20:16 bjz left
vendethiel the same 20:19
Mouq b2gills: So does the parser :9
20:21 beastd joined
pmurias masak: did TimToady call planned "big release" Perl 6.0.0 v1.0? 20:21
20:21 andreoss left 20:22 Kristien joined
Kristien hi 20:22
pmurias masak: a big problem with Perl 6.0.0 is what is the one before that?
masak: Perl 5.9.0 already exists
masak: with Perl6 v1.0 we could have Perl6 v0.9 20:23
moritz fwiw debian version numbers use ~ to mean "slightly less than"
so a 6.0.0~0.1 would be smaller than 6.0.0 20:24
though I'd propose we call it "Perl 6 2014.12"
20:24 jack_rabbit joined
moritz and avoid haggling about the exact digits and semantics 20:25
colomon perhaps 2015.12 ? 20:27
;)
moritz erm, yes. Still getting used to the 2015 thingy.
20:28 araujo joined
masak pmurias: TimToady hasn't been using v1.0 to my knowledge. I think it was mdk's wording in his blog post. 20:29
pmurias: I have never and will never spell Perl 6 "Perl6". 20:30
20:30 Kristien left, raiph joined, Kristien joined
b2gills pmurias: with the way Perl5 handles versions the version before 6 would be 5.999.999 20:31
Kristien I find * vs $_ confusing. 20:33
20:34 bjz joined
Mouq .oO( PerlAmpersandEnBeEsPeeSix ) 20:35
masak Kristien: matter of habit, I believe. 20:37
pmurias masak: were you at FOSDEM?
dalek c: 54d8707 | moritz++ | lib/Language/functions.pod:
Write introduction to functions.pod
20:37 anaeem1_ left
masak Kristien: I use both, frequently, and without confusion. 20:37
pmurias: no :/
Kristien masak: yeah, but you are competent! 20:38
moritz well, it took me a few years to avoid too much confusion between 5 and 6
masak pmurias: I am carrying around a currently inconsolable 3wo boy.
Kristien: I believe competence, at least most of the time, is a prerequisite for programming.
Kristien Me too.
moritz just managed his one commit per day to perl6/doc 20:39
pmurias anyway wasn't Perl 6.0.0 supposed to refer to the spec version? 20:40
so it should be Rakudo 1.0
moritz pmurias: I'm pretty sure rakudo will keep up its monthly release cycle and version scheme 20:41
1.0 < 2015.01
pmurias: also, larry mentioned that the version will primarily refer to the test suite, and the design documents count as accompanying material
20:42 kjs_ left
moritz pmurias: and that the idea is to snapshot the passing tests, and declare them (plus the compiler passing them) as 6.0 20:42
20:43 anaeem1 joined 20:44 bjz left 20:45 jack_rabbit left 20:49 nlfly joined
nlfly Hi guys 20:49
raydiak \o nlfly 20:50
nlfly can you help to understand the 'winner' keyword?
winner * {more @channels ... }
how exactly does it work? where do the threads come in?
raiph first, it's been renamed to "earliest" 20:51
second, does that clear it up? :)
20:52 zakharyas joined
_mg_ masak, Mouq: github.com/Homebrew/homebrew/pull/36439. This builds moar by default, and parrot is now optional. 20:52
raiph moritz: would it be reasonable for me to do `sudo apt-get install libperl-dev` logged in as me on hack? 20:53
masak \o/
20:53 bjz joined
nlfly it is winner in the test suite, and earliest does not compile 20:53
20:53 [Sno] joined
raydiak then you have an old rakudo 20:53
nlfly anyway: if i've got two channels, 20:54
i have latest rakudo star
anyway,
two channels. how can i write independently in two threads and read in another one?
20:54 bjz left
raiph .ask moritz: I'd like to use Inline::Perl5 and modules from CPAN on hack; seems I need to do `sudo apt-get install libperl-dev`; would that be reasonable? 20:55
yoleaux raiph: What kind of a name is "moritz:"?!
raiph .ask moritz I'd like to use Inline::Perl5 and modules from CPAN on hack; seems I need to do `sudo apt-get install libperl-dev`; would that be reasonable?
yoleaux raiph: I'll pass your message to moritz.
20:55 kaare_ joined 20:58 Kristien left, Kristien joined
raiph .tell nine www.reddit.com/r/perl/comments/2ubs...he/co7sa9y 20:59
yoleaux raiph: I'll pass your message to nine.
dalek ast: 4d0d32b | usev6++ | S02-types/built-in.t:
Add test for RT #122094
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=122094
raydiak nlfly: Channel is thread-safe, so .send and .receive/.poll should do it. for reading from more than one channel easily, you'll likely want earliest (which is called "winner" in your version, but will be documented as "earliest" on the web, which is current) 21:00
Mouq _mg_++ :D 21:01
21:02 anaeem1 left
_mg_ I still have problems with the shebang line, for example when I run panda 21:02
dalek ast: 6a639ce | usev6++ | S02-types/built-in.t:
Fix syntax for ticket number
21:04
pmurias raiph: and the first comment is wrong ;)
raydiak nlfly: btw have you already looked over S17? 21:05
raiph pmurias: audioen's comment?
21:06 spollei joined 21:09 _mg_ left 21:10 bjz joined 21:12 davercc joined
raydiak after working with nativecall a bit, seems like I should have wrapped libpng and bundled a dll with it, instead of sloppily re-implementing a minscule portion of it in P6 for Image::PNG::Portable :P 21:13
*miniscule
nlfly how to create threads then? :-) 21:14
masak raydiak: not too late for that ;)
pmurias raiph: yes, that's simply not true
vendethiel we need a C2P6 21:15
:P
Kristien I just found this: search.cpan.org/~ether/Task-Kensho-.../Kensho.pm
Mouq nlfly: What's your perl6 -v ? That might be why earliest isn't working 21:16
pmurias C2P6?
nlfly This is perl6 version 2014.09 built on MoarVM version 2014.09
masak ooh, Task::Kensho is actually a pretty interesting target for Perl 6 compatibility.
now there's a talk all in itself.
nlfly aha, not the latest star as i thought
Mouq nlfly: And you might want to look at the "start" keywork
*keyword
nlfly Mouq: cool, thanks, i forgot about that word 21:17
raiph pmurias: I wanted to post working code that demonstrated (s)he was wrong but I'm waiting for an answer from moritz to my question ^^ a few minutes ago 21:18
pmurias: and I'd much rather see an answer by nine 21:19
21:19 xfix left
raydiak masak: true enough :) guess I'll chalk that one up to practice...just wondering if there is anything useful I could still squeeze out of it, but I really can't think of much, unless I wanted to make it pure perl by reimplementing zlib too heh 21:19
nlfly yes, with 2014.12 earliest works
21:20 bjz left
colomon has somehow lost his p6 script to convert a git log —status file to a rough changelog. 21:22
colomon searched for five minutes and finally found it 10 seconds after writing that. 21:24
masak colomon: remember, it's always in the *last* place you look! :P 21:25
colomon doesn’t actually remember how to use it, either 21:26
raydiak used to use a trick when waiting for rakudo to compile, where whining about it manipulates it into finishing right then and proving him wrong :) 21:28
21:29 nlfly left
colomon “Saw 2 calls to deprecated code during execution.” apparently it needs some patching, too 21:30
21:37 kjs_ joined 21:38 gr33n7007h joined, kurahaupo1 left 21:44 kurahaupo1 joined 21:45 kjs_ left 21:52 zakharyas left 21:53 spider-mario joined, nige left
moritz raiph: libperl-dev installed 22:00
yoleaux 20:55Z <raiph> moritz: I'd like to use Inline::Perl5 and modules from CPAN on hack; seems I need to do `sudo apt-get install libperl-dev`; would that be reasonable?
raydiak perlbrew doesn't allow that? 22:01
22:03 Ugator left
skids m: say Grammar.^methods>>.name 22:04
camelia rakudo-moar 6e182d: OUTPUT«parse subparse parsefile MATCH MATCH_SAVE INTERPOLATE DYNQUANT_LIMITS OTHERGRAMMAR INDRULE RECURSE orig target from pos prune CAPHASH !cursor_init !cursor_start_all !cursor_start !cursor_start_cur !cursor_start_fail !cursor_start_subcapture !cursor_capture…»
skids m: say Grammar.^methods
camelia rakudo-moar 6e182d: OUTPUT«No such method 'gist' for invocant of type 'NQPRoutine'␤ in method gist at src/gen/m-CORE.setting:8557␤ in sub say at src/gen/m-CORE.setting:17341␤ in block <unit> at /tmp/KV8yXwsB5h:1␤␤»
moritz raydiak: I'm pretty sure it does
skids I think that's just a matter of adding a .Str to NQPRoutine knowhow? 22:05
moritz gist
though a more complete solution would fake the complete type Mu
22:06 rurban left 22:10 kjs_ joined 22:13 Kristien left
raiph moritz++ # :) gracious hosting 22:14
nine Oh the horrors, the abuse, the hackery. But I made it. I can embed MoarVM in Perl 5 and call into Perl 6 code :) 22:18
yoleaux 20:59Z <raiph> nine: www.reddit.com/r/perl/comments/2ubs...he/co7sa9y
colomon nine++ 22:19
nine raiph: that's true. But it's also quite possible to not copy hashes and arrays. Copying was just quicker to implement and there are probably not many useful APIs that need modifications of arguments. 22:20
dalek Heuristic branch merge: pushed 17 commits to rakudo/birdless by moritz 22:23
masak surprisingly positive comments at news.ycombinator.com/item?id=8980387
22:23 brrt joined
brrt masak: yes, except for chromatic for some reason 22:24
masak a lot of people are already getting entangled in the whole "v1.0" thing, though.
brrt: "for some reason"?
brrt: you must be new here. ;)
brrt i'm new enough not to know the reason
masak right.
brrt and honestly i don't care
colomon rakudo/birdless ?
masak it'd be a mistake to dismiss chromatic entirely.
brrt that's not what i'm saying 22:25
masak please read him for one of the most insightful critiques of the Perl 6 community.
at least if you can stomach his way of expressing it without burning out.
brrt what i'm saying is that i don't care about the reason of the bitterness
masak (which I can on some days)
moritz colomon: an experimental branch to see what happened if we dropped parrot support from rakudo
masak brrt: I understand.
moritz $ git diff --shortstat nom...birdless 73 files changed, 87 insertions(+), 6926 deletions(-)
colomon !! 22:26
masak moritz: is that... Rakudo sans Parrot?
brrt :-) 22:27
moritz masak: yes, though quite some docs still aren't adapted in that branch
pmurias masak: re Perl6 v1.0, that's the price for doing the annoucement only in talk form
masak moritz: we waste 7kLOC on Parrot support?
dalek p-js: f71a04a | (Pawel Murias)++ | src/vm/js/bin/run_tests:
Add passing test 47 to run_tests.
masak pmurias: I believe you're right.
pmurias: mdk basically got to pick the terms as he saw fit. 22:28
pmurias I didn't see that TimToady talk 22:29
moritz can somebody please reply to news.ycombinator.com/item?id=8981931 and say they should join #moarvm? 22:30
colomon on it
moritz I've reached my limit of allowed replies per time unit on hackernews, it seems
colomon that’s on freenode?
moritz yes
masak yes.
moritz++ # replying to the hilt!
pmurias maybe you need moritz_sockpuppet? 22:31
colomon moritz: just replied, let me know if you’d like me to tweak it.
hoelzro o/ #perl6 22:33
22:34 brrt left 22:36 Ugator joined, telex left 22:38 telex joined 22:47 beastd left 22:48 raiph left
moritz colomon++ 22:51
time for sleep here& 22:54
22:57 skids left, raiph joined
El_Che fosdem perl pictures: imgur.com/a/1ynV7#0 22:58
dalek osystem: e1cd2fd | Mouq++ | META.list:
Add TOML to the ecosystem
23:00
23:02 skids joined 23:03 pmurias left 23:06 kjs_ left, espadrine joined 23:12 jack_rabbit joined 23:16 skids left 23:21 skids joined
raydiak notices that rakudo and moar are in their own organizations but any of us is allowed to break nqp :) 23:29
japhb Besides TimToady++'s talk, any other cool announcements and/or "Just One More Thing"s out of FOSDEM? I saw the link to nine++'s talk slides (I *love* that he included complete notes); any others posted? 23:34
timotimo jnthn didn't update his "talks" page yet; it still says "i'll be talking at the apw2014, topic TBA" 23:35
i was hoping i could find his slides soon-ish :)
japhb Ditto. I become the biggest fanboy nudge after his talks, asking for posted slides. :-) 23:36
23:38 skids left
timotimo o/ 23:42
23:42 gfldex left 23:48 gr33n7007h left, skids joined 23:53 gr33n7007h joined
masak jnthn's slides are of consistently high quality. and they stand quite well on their own, in most cases. 23:59