»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by sorear on 25 June 2013.
lizmat has arrived in Orlando 00:51
colomon \o/ 00:56
perigrin \o/
but I did beat you here by a few years.
:D
colomon is not going to be anywhere near Orlando, alas.
perigrin :(
lizmat perigrin colomon o/
perigrin wonders if lizmat can hear the cat complaining about her bath from there. 00:57
colomon sorry, perigrin. I'd love to be there, but big family vacation in April, big family vacation in August, and I need to $work as much as possible in between.
lizmat I did hear something squeak :-)
perigrin colomon: understood, next time I'll try to have it in a more family friendly location than near Disney. :) 01:00
colomon :p
perigrin (seriously though, you'll be missed ... ) 01:01
colomon awww
I'll certainly miss seeing everyone. 01:02
dalek ast: 0ea4d75 | Mouq++ | S26-documentation/08-formattingcodes.t:
Add test for RT #114510
 05:01
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=114510
Mouq r: my &infix:<c> = {$^a+$^b}; say 1 c 2; 05:13
camelia rakudo-jvm bcd186: OUTPUT«(timeout)»
..rakudo-{parrot,moar} bcd186: OUTPUT«3␤»
dalek p: f050c36 | (Tobias Leich)++ | tools/build/MOAR_REVISION:
bump moar rev for various spesh fixes
 05:36
dalek kudo/nom: 5fdfc78 | (Tobias Leich)++ | tools/build/NQP_REVISION:
bump nqp/moar rev for various spesh fixes
 05:37
dalek c: eeb0dd2 | Mouq++ | lib/Perl6/TypeGraph/Viz.pm:
Make SVG urls absolute per issue #17
 05:45
c: b33c6c0 | Mouq++ | lib/Language/Sets_Bags_and_Mixes.pod:
Add intro to Sets, Bags, and Mixes
 05:46
masak morning, #perl6 06:18
moritz \o masak 06:19
atroxaper masak: o/ 06:22
jnthn o/ masak 06:29
FROGGS_ masak: o/ 06:32
Mouq /o/o/ \o\o\ 06:36
|o|o| masak |o|o|
\o\o\ /o/o/
masak :) 06:47
jnthn airport & 06:51
masak leaves for airport
dalek c: 0216cbd | Mouq++ | / (2 files):
Lower headings so that they're uniform

Classes and routines, especially when defined in language docs, are not necessicarily going to written be at the right =headN for consistent formatting within sub-pages		 07:43
c: 5aeb785 | Mouq++ | htmlify.p6:
Inline &add-new in &find-definitions

There used to be a reason for it to be a sub; no longer
Mouq "necessicarily"? 07:45
Mouq is disappointed in himself
FROGGS *g* 07:48
Mouq: you also take care that its test pass in the next days? so we will have a clean rakudo-star?
Mouq FROGGS: doc? Sure, done :9 07:57
FROGGS thanks :o) 07:58
I've just seen fails yesterday
Mouq bed& 07:59
FROGGS gnight 08:00
FROGGS arnsholt: ping 08:58
ChoHag If I have a sub foo (:$bar) {...} and a variable $baz which may or may not be defined, how can I call foo so that bar is only set if $baz if defined? 09:46
ChoHag And conversely, how do I change the signature so that something which isn't a Bar can get in (it's actually 'sub foo (Bar :$bar, Bar :$other_bar)') 09:48
I forget the error message because it was yesterday, but if $baz is not set then calling foo complains because it's not a Bar. 09:49
Timbus declare $baz as a Bar, or assign it a Bar type object 10:03
Timbus or.. foo(bar => $baz||Bar) 10:05
or, my $baz is default(Bar) 10:06
sergot o/ 10:54
hi
FROGGS m: my $baz = 42; sub foo(:$bar) { say $bar }; foo( :bar($baz.defined) ) 11:08
camelia rakudo-moar 5fdfc7: OUTPUT«True␤»
FROGGS ChoHag: ^^ 11:09
ahh, wait
FROGGS I don't see a $baz in your example, so I don't know what you want to achieve 11:10
timotimo if $baz is defined, :bar should exist in the signature and be .defined, if it's not defined, it shall not be defined in the signature 11:12
m: sub frob(:$bar) { say $bar.defined }; my $baz; frob(bar => $baz); $baz = "hi"; frob(bar => $baz)
camelia rakudo-moar 5fdfc7: OUTPUT«False␤True␤»
timotimo ChoHag: ^ that should be enough
sergot Can I get a commit permission to nqp just for today? :) 11:25
FROGGS sergot: nqp would be okay I think 11:26
moritz: can you hand that out?
psch hi #perl6 o/ 11:33
FROGGS hi psch
psch FROGGS: you read the clog i assume 11:34
FROGGS I did
psch platform worries aren't over though! NativeCall fails tests 10..12 in 05-arrays.t for me
FROGGS psch: you have a 32bit machine? 11:35
psch FROGGS: a 32bit vm, yes
i can rerun on a real 32bit too if that does anything, although i wouldn't assume so
FROGGS no, it should fail on 32bit hardware also 11:36
FROGGS we have that bug since a while now, and I don't know how to fix it 11:36
psch: see github.com/jnthn/zavolaj/issues/37
psch it works if include stdint.h and pass specific bit sizes
but that's not really a solution i guess, seeing as perl6's int behaves as 32bit 11:37
and sizeof(long) is 4 too, so it should just work..
right, it's all in the ticket already :) 11:38
FROGGS yeah, and I still don't know what the solution might be 11:38
psch i don't think having perl6 int vary in size is sensible 11:39
FROGGS but int is native
moritz sergot++ added to team nqp
sergot moritz++ thanks! :)
moritz that one doesn't require a CLA, so you can keep that bit afterwards :-) 11:40
FROGGS so it could/can/should? be platform specific
perhaps I should read the spec :o)
psch oh right
yeah S09 #Sized types says "int32 (aka int on 32-bit machines)"
sergot moritz: ok :) 11:41
psch so it should (and does, 'my int $x = 2 ** 32; say $x; #prints 0 here) be platform specific
s/be/behave/
smls m: my $a := 'aa'; my $b := <bb cc dd>; sub say_two_things ($x, $y is rw) { say "thing #{++$_}: $_" for $x, $y }; say_two_things $a, $b 11:42
camelia rakudo-moar 5fdfc7: OUTPUT«thing #1: aa␤thing #2: bb␤thing #3: cc␤thing #4: dd␤»
smls ^^ Perl 6 binding + Perl 6 flattening = sigils become unreliable :( 11:42
without the "is rw", $y does not flatten - but that only makes it *more* confusing...
moritz agrees 11:43
smls I really like flattening in Perl 5, where its behavior is static. But Perl 6's dynamic flattening, especially in the presence of binding etc, may end up causing more WAT than DWIM... 11:45
moritz note that Perl 5 flattening isn't totally static either 11:46
push @a, mysub();
will the return value of mysub() flatten?
smls yes 11:48
of course if it returns a single scalar, it "flattens" to that item (like any other list that only contains a single item)
moritz sub mysub() { [1, 2,3 ] } # that array ref doesn't flatten 11:50
TimToady p6item == p5ref more or less
smls moritz: because array refs *never* flatten; they are scalars. 11:51
dalek p: 816447f | sergot++ | VERSION:
bump VERSION to 2014.06
 11:52
sergot moritz: could you give me access to [email@hidden.address] ? 11:54
FROGGS sergot: wait wait wait... 11:59
are the tests clean?
smls TimToady: Don't you consider it a problem that one can't rely on a dollar-sigiled variable to behave as a single item? 12:01
For me, this calls the entire concept of sigils into question 12:02
TimToady binding is cheating, and cheaters get caught sometimes :)
smls but you need it to work with lazy lists 12:03
which is one of the heavily advetised "wow" features
to people (and even newbies) *will* use binding
TimToady so don't bind to $, bind to @
FROGGS .oO( Unsupported use of $, variable; in Perl 6 please use $*OUT.output_field_separator() ) 12:04
sergot FROGGS: yes 12:05
tests are clean
FROGGS also spectests for all backends?
smls TimToady: that's easy to mistype though, and can cause nasty action-at-adistance (as in the example i showed) 12:06
FROGGS TimToady: can't we "just" fix it? 12:06
sergot FROGGS: for nqp, yes
FROGGS sergot: I'd recommend to test rakudo too, before uploading any tarball 12:07
sergot FROGGS: ok :)
TimToady we can probably detect more such bad bindings at compile time and warn 12:08
*ost
psch gist.github.com/peschwa/acbe01ccc45d80c2c5d1 # only in the REPL, is this 32bit specific? 12:09
smls Maybe $item variable binding could be restricted to itemized things? 12:11
TimToady would rather not make that a run-time check unless it can be optimized out most of the time 12:12
smls Like @array variable binding is restricted to Positional things.
TimToady well, signature binding does ensure $ is itemized, I think 12:14
smls Besides for lazy lists, binding is also advertised to newbies with a Perl 5 background as a replacement for references (or rather, all uses for references except the building of nested data structures) 12:15
TimToady but that's ::= not :=
smls and P5 people use references a lot :)
So I don't think its fair to argue that binding is such an "advanced" feature that it's ok if it's dangerous 12:16
TimToady epcot & :)
smls TimToady: Apparently with "is rw", it does not (on Rakudo Moar). 12:19
FROGGS psch: not sure why it blows up 12:22
psch FROGGS: something about different native types for the same lexical...? my int $x = 0; my int64 $x = 0; # doesn't blow up at all, although it also doesn't throw a X::Redeclaration either 12:35
moar defines the int repr as always 64bit, which is why i tried the second case
to clarify, the default for a P6Int repr without a nativesize argument is 64bit 12:38
which it probably shouldn't be, seeing as that's what behind int and int should be platform specific
FROGGS I think the problem is in getattr_i 12:42
ahh, hmmm, perhaps not 12:44
sergot FROGGS: gist.github.com/sergot/07e27d01984b20e3c1f6 12:51
rakudo's make test fails on jvm
FROGGS sergot: the evalserver might has crashed on your box... 12:52
can you re-run it?
sergot Yes, in a sec 12:53
dalek kudo-star-daily: 448b391 | coke++ | log/ (14 files):
today (automated commit)
 12:56
[Coke] modules/perl6-lwp-simple started failing on the daily star run for moar. (t/get-perl6-org.t) 12:57
using commit d52b3bd76a5f9354030b475c1268ea7d582f359f 12:58
which looks like the most recent merge commit.
parrot is failing modules/jsonrpc, modules/doc, modules/perl6-lwp-simple 12:59
colomon oooo, release day
[Coke] java still failing "lots"
(parrot -and- moar are failing that updated test in lwp-simple; jvm is still throwing an NPE on every test) 13:00
FROGGS damn, I have tests that fetch web pages -.- 13:01
hate*
psch FROGGS: i'll poke at that int32/int thing a bit more later, $work came up 13:03
that is, if you or someone else doesn't beat me to it :)
FROGGS psch: k, and good luck :o)
have to do $work too
sergot FROGGS: clear now :) 13:07
FROGGS moritz: can you apply please? github.com/cosimo/perl6-lwp-simple/pull/34 13:09
moritz pulls the trigger 13:10
FROGGS sergot: for me also
[Coke]: the test will be fixed in a minute
perhaps I should look at modules/jsonrpc for parrot before the release is done 13:11
sergot FROGGS: stresstesting now 13:12
FROGGS sergot++
[Coke] FROGGS: why not create a file in the github repo that is of known length and pull that? 13:13
(if firing up our own server is not going to fly)
FROGGS github does a redirect to the https site... that would be a problem 13:14
moritz hey 13:15
[Coke] k
moritz aren't there some jquery hosting sites?
like, from google etc.?
those should be quite reliable
and if we fetch a specific version (not jquery.js but jquery-1.8.45.js or so), it should be immutable
FROGGS good idea
[Coke] (we could also setup a page on perl6.org that is just for our testing) 13:16
FROGGS because, I do not want to adjust that test anymore :o)
moritz [Coke]: also works, but feather is less reliable than google :-)
FROGGS yeah
FROGGS t/spec/S32-io/IO-Socket-Async.t (Wstat: 134 Tests: 0 Failed: 0) 13:40
Non-zero wait status: 134
I wonder why that fails...
timotimo does it fail consistently? 13:41
FROGGS timotimo: yes 13:43
ohh, hmmm, no :/
1..6
Unhandled exception in code scheduled on thread 140234788333312
No such method 'payload' for invocant of type 'Str'
moritz somehow my general impression is that rakudo had a month or two of feature grows, and should now focus a bit more on quality again 14:02
PerlJam speeed! 14:04
:)
timotimo if we add more features, we can focus on even more quality! 15:00
hoelzro you can compile a script to a JAR via perl6-j --target=jar -o hello.jar hello.p6, right? 15:05
I'm getting a NullPointerException when I try =/
lizmat good *, #perl6!
wrt quality: we still seem to have some rather serious race conditions in the async stuff 15:06
or regular features (like EVAL, matching) perhaps not being entirely threadsafe 15:08
anyways, today I'm not going to commit anything S11 related
as it seems more or less stable right now (apart from maybe performance) 15:09
FROGGS hoelzro: perhaps use --output= instead -o 15:11
hoelzro tries
hoelzro huh, that did it... 15:12
how odd
FROGGS hehe
I guess -o= does it too
but I've not yet seen -o in action 15:13
hoelzro weird
hoelzro if I wanted better arguments processing than what MAIN offers, I take it I should use an external module? or is there a way to enhance how MAIN handles arguments? 15:45
masak o/ 15:51
...from London :) 15:52
ChoHag was forced to visit London today. 15:52
FROGGS hoelzro: rakudo is open source... so you could enhance it... (or copy the logic and modify it) 15:58
dalek ast: 1a0d506 | (Elizabeth Mattijsen)++ | S22-package-format/parse-spec.t:
Initial S22 tests
 15:59
lizmat hoelzro: a few days ago this question was raised as well 16:00
I seem to recall TimToady not being against expanding MAIN features
hoelzro FROGGS: right, I'm just wondering if such a change would be welcome, and if it would be in accordance with the spec
lizmat: oh?
hmm
I'll have to check my logs
lizmat hoelzro: as long as it could be generalized :-) 16:01
(is what I recall)
sightseeing now &
masak release day today, no? 16:04
oh, sergot++ is already bumping stuff.
sergot masak: yes, stresstesting now :D 16:06
Everything else if ready.
is
masak najs. 16:09
hoelzro lizmat: of course =) 16:15
FROGGS hoelzro: make a prototype, if it is convincing... it will be considered 17:02
hoelzro: otherwise it can stay a module of course
ChoHag If a foo has a private attribute, can I access that from a role which is mixed into it? 17:03
s/foo/class/
masak no. 17:04
vice versa works, though.
hoelzro FROGGS: will do! 17:05
ChoHag What about an inherited class? 17:06
dalek kudo/nom: d45fccd | sergot++ | tools/build/NQP_REVISION:
[release] bump NQP revision
 17:33
kudo/nom: 15df0f5 | sergot++ | VERSION:
[release] bump VERSION
kudo/nom: a4e6d90 | sergot++ | docs/announce/2014.06.md:
2014.06 release announcement added
sergot FROGGS: could you give me the access to [email@hidden.address] ? 17:36
or someone else? moritz masak ?
:)
FROGGS sergot: if you could provide the tarballs, I'd like to test them and then I should be able to upload the files 17:38
sergot ok 17:39
:)
FROGGS uhh wait, I only have UMTS here and a very limited limit :/
(normal internet is broken atm) 17:40
dalek kudo/nom: 8812ca4 | sergot++ | docs/release_guide.pod:
release done, moved to past releases, new plan added
 17:43
masak sergot++
sergot Who can upload the tarball for me? :) 17:44
sergot tarballs 17:44
moritz
sergot moritz++
moritz sergot: where are they?
FROGGS on C:\
:P
sergot moritz: where do you want them to be? :) 17:45
moritz sergot: somewhere on the internet, where I can download them 17:46
sergot: or give me your pubkey by /msg 17:47
then you can scp them yourself
sergot moritz: done 17:49
FROGGS sergot: have you fudged failing spectests of perl6-m and perl6-p? 17:52
sergot FROGGS: no, I haven't 17:53
FROGGS eww
raiph m: my @foo = [1,2]; say @foo.perl # Would it be worthwhile changing .perl of @foo to be just `[1, 2]` ? 17:54
camelia rakudo-moar 5fdfc7: OUTPUT«Array.new([1, 2])␤»
sergot doing it now 18:03
moritz raiph: [1, 2] doesn't flatten, @foo does 18:06
vendethiel it flattens to [1, 2] though, really 18:07
moritz vendethiel: that statement doesn't add any information 18:08
FROGGS not again >.< 18:08
moritz it just confuses everybody, because it redefines what "flattens" means
Mouq m: say Array.new([1, 2]).push([6, 7]).perl 18:09
Mouq Oh, we lost camelia in the netsplit 18:11
m: say Array.new([1, 2]).push([6, 7]).perl
camelia rakudo-moar 5fdfc7: OUTPUT«Array.new([1, 2], [6, 7])␤»
psch had her on the other side, but not the person i wanted to indirectly answer with a bit of code heh
FROGGS p: say $*VM.perl 18:28
camelia rakudo-parrot 5fdfc7: OUTPUT«VM.new(config => {"sha1" => "fed40d8394df2f64681ca6651664daf095b42091", "PARROT_HAS_DLINFO" => "1", "has_sigaction" => "define", "memalign" => "posix_memalign", "HAS_GMP" => "1", "hugeintvalsize" => "8", "HASATTRIBUTE_WARN_UNUSED_RESULT" => "1", "ccwarn"…» 18:29
FROGGS p: say $*VM.perl ~~ m/\w/
camelia rakudo-parrot 5fdfc7: OUTPUT«｢V｣␤␤» 18:30
FROGGS p: use Test; ok $*VM.perl ~~ m/\w/, 'We can do a $*VM.perl';
camelia rakudo-parrot 5fdfc7: OUTPUT«ok 1 - We can do a $*VM.perl␤»
FROGGS why does that fail when running the spectest?? 18:31
FROGGS p: use Test; todo('no VM.signature yet'); ok $*VM.signature, "Signature is '{$*VM.signature}'"; ok $*VM.perl ~~ m/\w/, 'We can do a $*VM.perl'; 18:33
camelia rakudo-parrot 5fdfc7: OUTPUT«use of uninitialized value of type Blob in string context in any at gen/parrot/Metamodel.nqp:1480␤␤not ok 1 - Signature is ''# TODO no VM.signature yet␤ok 2 - We can do a $*VM.perl␤»
dalek ast: f7b42fb | sergot++ | S (2 files):
fudge failing tests, only for release purposes
 18:44
raiph m: say @(1,2); # m: my @foo = [1,2]; say @foo.perl # Would it be worthwhile changing .perl of @foo to be `[1, 2].list` ? 18:45
camelia rakudo-moar 5fdfc7: OUTPUT«1 2␤»
FROGGS raiph: no, because you would strip off the itemness of that thing
m: say [1, 2]; say [1, 2].list 18:46
camelia rakudo-moar 8812ca: OUTPUT«1 2␤1 2␤»
FROGGS m: say [1, 2].perl; say [1, 2].list.perl
camelia rakudo-moar 8812ca: OUTPUT«[1, 2]␤Array.new(1, 2)␤»
raiph (Urgh. That was supposed to go to camelia)
FROGGS m: say "a" for [1, 2]; say "b" for [1, 2].list
camelia rakudo-moar 8812ca: OUTPUT«a␤b␤b␤»
raiph m: my @foo = [1,2]; say @foo.perl # Would it be worthwhile changing .perl of @foo to be `[1, 2].list` ? 18:47
camelia rakudo-moar 8812ca: OUTPUT«Array.new([1, 2])␤»
raiph (Now *that* is intentionally to the channel !) 18:48
m: say Array.new(1,2) == [1,2].list 18:49
camelia rakudo-moar 8812ca: OUTPUT«True␤»
FROGGS right
m: my @foo = [1,2]; say @foo.elems 18:50
camelia rakudo-moar 8812ca: OUTPUT«1␤»
FROGGS that is why we want to keep the Array container
(also in future when it is a typed or shaped array)
dalek rl6-roast-data: 42cb179 | coke++ | / (6 files):
today (automated commit)
 18:56
[Coke] moar is clean today. jvm has 2 files with errors. parrot has a abort. 19:06
if someone could fix/fudge those, \o/
FROGGS [Coke]: the parrot fail is (temporarily) fudged 19:07
[Coke] FROGGS++ 19:12
jvm failures: github.com/coke/perl6-roast-data/b....out#L1720
FROGGS I can take a look but I doubt that I can do anything about it 19:13
[Coke] no worries. 19:13
Mouq m: say Array.new(1,2) == [1,2].list 19:14
camelia rakudo-moar 8812ca: OUTPUT«True␤»
Mouq m: say Array.new(1,2) == [1,2].list
camelia rakudo-moar 8812ca: OUTPUT«True␤»
Mouq oops, sorry
FROGGS you can say that again :o)
Mouq m: say Array.new(1,2) eqv [1,2].list 19:15
camelia rakudo-moar 8812ca: OUTPUT«True␤»
Mouq m: say Array.new(1,2) eqv [1,2]
camelia rakudo-moar 8812ca: OUTPUT«True␤»
Mouq Hm
FROGGS j: for ^2 { ENTER { say "hurz" } } 19:20
camelia rakudo-jvm 8812ca: OUTPUT«(timeout)» 19:20
Mouq FROGGS: Is "hurz" a nonsense word like "foo" or "baz"? 19:22
FROGGS this explodes: for ^2 { last; ENTER { say "hurz" } } 19:22
Mouq: there is a german comedian who was singing a song about Hurz :o) 19:23
Mouq FROGGS: Ahh, ok :)
[Coke] Doctor, it Hurz when I do this. 19:24
dalek ast: d09bffe | (Tobias Leich)++ | S04-phasers/in-loop.t:
RT #122134 last+ENTER in loop (skipped for jvm)
 19:34
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=122134
PerlJam Are any of the "cut" regex ops implemented? (:, ::, :::, <cut>) 19:37
dalek ast: 1dbb9de | (Tobias Leich)++ | S04-phasers/in-loop.t:
RT #122134 last+ENTER in loop (skipped for jvm)
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=122134
FROGGS PerlJam: no
PerlJam bummer. That's what I thought, but I was hoping to be surprised :)
pippo o/ #perl6 19:43
PerlJam: what the cut operator is supposed to do?
PerlJam pippo: fail if the regex engine backtracks across it. 19:46
pippo PerlJam: Ho. Thanks! 19:47
dalek c: f7705ba | Mouq++ | htmlify.p6:
Process =DESCRIPTIONs in language docs and use as summary
 20:05
c: 5cc202c | Mouq++ | / (10 files):
Add =DESCRIPTIONs to the language docs, normalize =TITLEs some
vendethiel p6: for ^2 { last; ENTER { say "hurz" } } # where ? 20:06
camelia rakudo-{parrot,jvm,moar} 8812ca, niecza v24-109-g48a8de3: OUTPUT«hurz␤»
FROGGS huh
vendethiel FROGGS: I fixed perl 6.
implems bent over camelia
FROGGS jnthn: please merge github.com/jnthn/zavolaj/pull/39 20:07
dalek c: acbc68b | Mouq++ | html/css/style.css:
Center titles and descriptions so that they look better together

Left-aligned descriptions looked weird		 20:14
Mouq actual review: doc.perl6.org/language.html 20:15
(may require Ctrl-R) 20:16
vendethiel did need ctrl-r 20:17
ah, too bad we don't have on-site highlighting yet 20:18
maybe that's something I should tackle
moritz might be as easy as piping through Text::VimColor 20:19
dalek c: 1dec2d6 | Mouq++ | lib/Language/classtut.pod:
[classtut.pod] flush code with margin
 20:21
FROGGS or somebody would port the Kate syntax highlighting module from P5 20:35
could be a fun task
timotimo the kate module isn't very thorough so far 20:37
ma ybe better to shell out to pygmentize 20:38
FROGGS hmmm, I use it at one place to colorize XML, works well enough 20:39
timotimo er, i meant to say: the kate syntax highlighter fro p6 20:40
FROGGS ahh
I'm usually fine with P5 highlighters :o) 20:41
Mouq I think for the most part, I'm done with adding features myself. IMO doc is pretty usable and extensible now, if slow, so I'd rather work on making the documents themselves better 20:42
But if someone wants to add syntax highlighting, go for it :) 20:43
raiph moritz++, FROGGS++ # for earlier explanation and patience 20:53
sergot good night o/ 21:43
Mouq \o sergot 21:47
FROGGS gnight sergot
timotimo m: say lines() 21:49
camelia rakudo-moar 8812ca: OUTPUT«Céad slán ag sléibhte maorga Chontae Dhún na nGall Agus dhá chéad slán ag an Eireagal ard ina stua os cionn caor is coll; Nuair a ghluais mise thart le Loch Dhún Lúich’ go ciúin sa ghleann ina luí I mo dhiaidh bhí gleanntáin ghlas’ Ghaoth …»
timotimo m: say $*ARGFILES
camelia rakudo-moar 8812ca: OUTPUT«use of uninitialized value of type Any in string context␤IO::Handle<>(closed)␤»
timotimo m: say $*ARGFILES[0].perl 21:50
camelia rakudo-moar 8812ca: OUTPUT«IO::Handle.new(path => Any, ins => 0, chomp => Bool::True)␤»
dalek c: 9534e90 | Mouq++ | lib/Language/Sets_Bags_and_Mixes.pod:
Add Baggy operators to Sets, Bags, and Mixes
 21:51
timotimo why is "closed"?
m: say $*ARGFILES[0] 21:52
camelia rakudo-moar 8812ca: OUTPUT«use of uninitialized value of type Any in string context␤IO::Handle<>(closed)␤»
timotimo m: say $*ARGFILES[0].DUMP
camelia rakudo-moar 8812ca: OUTPUT«Default constructor for 'IO::Path' only takes named arguments␤ in method new at src/gen/m-CORE.setting:827␤ in method path at src/gen/m-CORE.setting:14088␤ in method DUMP at src/gen/m-CORE.setting:1065␤ in sub DUMP at src/gen/m-CORE.setting:139…»
Mouq m: say $*ARGFILES[0] for lines 21:53
camelia rakudo-moar 8812ca: OUTPUT«use of uninitialized value of type Any in string context␤IO::Handle<>(closed)␤use of uninitialized value of type Any in string context␤IO::Handle<>(closed)␤use of uninitialized value of type Any in string context␤IO::Handle<>(closed)␤use of uni…»
FROGGS $*ARGFILES would only contain something if you invoke: perl6 -e '...' < foo.txt 21:55
timotimo well, in this case we're getting something piped into stdin 21:57
Mouq Is this readable? doc.perl6.org/language/Sets_Bags_and_Mixes
Still need to add an explaination for *Hash 21:58
timotimo perl6-m -e 'say $*ARGFILES' doesn't seem to print two files when i supply two files 21:59
in fact, it only prints one entry at all; is that how it works? 22:00
ah, ARGFILES is just a single handle that distributes commands over multiple files 22:01
pmichaud note about moarvm #105 22:04
pmichaud files created in programs should generally be coded with 0666 permissions. 22:05
oh, maybe this belongs in #moarvm 22:06
moving there
dalek c: 4798c80 | Mouq++ | lib/ (4 files):
Move *Hash into Sets, Bags, and Mixes
 22:08
c: 73a5f1c | Mouq++ | lib/Language/Sets_Bags_and_Mixes.pod:
Remove pointless headings
 22:11
timotimo moarvm doesn't report file paths when failing to open files ... 22:15
raiph Has the potential-surprise-due-to-last-statement-in-block-not-being-sunk somehow been eliminated? 22:21
Mouq m: 42; 43 22:23
camelia rakudo-moar 8812ca: OUTPUT«WARNINGS:␤Useless use of constant integer 42 in sink context (line 1)␤»
raiph Mouq: S06:715 "Note that the default return policy assumes functional semantics, with the result that a loop as the final statement would be evaluated as a map, which may surprise some people." (plus lots of discussion of this and spec changes in last few years -- but I haven't been able to see it in my private camelia experiments) 22:28
synopsebot Link: perlcabal.org/syn/S06.html#line_715
lizmat do we have a release already? should I commit an S22 tweak? 22:29
timotimo lizmat: branches are super cheap in any case 22:30
lizmat I'll take that as a "wait until released"
it's not that important
timotimo hm, ok 22:32
raiph timotimo: at one point you were trying to understand the issue I just mentioned; do you now understand it and can demonstrate the "problem"? 22:33
timotimo the thing with for loops at the end of a block? 22:36
for loops are now always sunk 22:37
dalek volaj: 53317e5 | tadzik++ | / (2 files):
Get rid of deprecation messages
 22:45
volaj: 265e504 | jonathan++ | / (2 files):
Merge pull request #39 from jnthn/tadzik/deprecations

Get rid of deprecation messages
jnthn There we go 22:49
timotimo i'm having some trouble with await and start :o 22:54
ooooh 22:55
i know why %)
raiph .tell TimToady S06:715 says "a loop as the final statement would be evaluated as a map"; this is now a fossil, right?
synopsebot Link: perlcabal.org/syn/S06.html#line_715
yoleaux raiph: I'll pass your message to TimToady.
lizmat r: my @a; my $c = Int; @a.push: $($c,1); $c=Str; @a.push: $($c,2); say @a.perl 23:00
camelia rakudo-{parrot,jvm,moar} 8812ca: OUTPUT«Array.new($(Str, 1), $(Str, 2))␤»
lizmat jnthn: ^^^ this feels wrong to me
feels like the push binds to the variable, rather than just copy ? 23:02
am I missing something?
hmm... it appears jnthn is already asleep, gnight jnthn! 23:03
Mouq r: my @a; my $c = Int; @a.push: item Array.new($c,1); $c=Str; say @a.perl 23:04
camelia rakudo-{parrot,jvm,moar} 8812ca: OUTPUT«Array.new([Int, 1])␤»
timotimo what are other common names for our Promises?
lizmat timotimo: you mean things that return Promises ? 23:05
timotimo no
BenGoldberg Vocabulary words?
timotimo i know at least one language calls it "futures"
Mouq timotimo: en.wikipedia.org/wiki/Promise_%28programming%29
BenGoldberg oath, vow, agreement, etc?
timotimo mhh 23:06
Mouq Oh look, we can add Perl 6 to "List of implementations" on that page :)
timotimo i think promise and future will be familiar to enough attendees
timotimo what kinds of things would return Channels by themselves? 23:12
i think asynchronous lines does that
lizmat they would be more like Supplies, I would guess
lizmat n: my @a; my $c = Int; @a.push: $($c,1); $c=Str; @a.push: $($c,2); say @a.perl 23:13
camelia niecza v24-109-g48a8de3: OUTPUT«[$(Str, 1), $(Str, 2)].list␤»
timotimo aye, though i see a reason for having them be channels instead, too
on the other hand, turning a supply into a channel is just .tap({ $my_chan.send($_) }) 23:14
lizmat .Channel, even
timotimo actually
yes
lizmat m my @a; my $c = Int; @a.push: $($c,1); $c=Str; @a.push: $($c,2); say @a.perl 23:15
m: my @a; my $c = Int; @a.push: $($c,1); $c=Str; @a.push: $($c,2); say @a.perl 23:16
camelia rakudo-moar 8812ca: OUTPUT«Array.new($(Str, 1), $(Str, 2))␤»
lizmat I would expect to see:
Array.new($(Int, 1), $(Str, 2))
am I wrong? Is there something about type objects that make them bind in a .push? 23:17
timotimo i'm looking for some acceptable examples to show promises and channels with :|
timotimo should we get Channel.Supply, too? 23:19
damnit, i was going to present "perl 6 without rocket surgery" and now i'm about to write a slide on reactive programming 23:20
lizmat That would be trickier, as we cannot be sure there are no other threads picking values off the Channel
so you could never be sure the resulting Supply would see all values of the Channel
timotimo this won't do at all!
ah, that's a good point. 23:21
lizmat submits rakudobug 23:23
timotimo m: my @a; my $c = 1; @a.push: $($c,1); $c = "hi"; @a.push: $($c,2); say @a.perl 23:24
camelia rakudo-moar 8812ca: OUTPUT«Array.new($("hi", 1), $("hi", 2))␤»
timotimo it would appear the container of the $c is being pushed, rather than the value? 23:25
lizmat #122137 23:26
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=122137
Mouq It's not the fact that it's being pushed, it's the building of $($c, 1), isn't it? 23:26
timotimo could very well be 23:27
is that being done at compile-time perhaps?
Mouq m: my $c = 1; my \b = $($c, 1); $c = Int; say b.perl
camelia rakudo-moar 8812ca: OUTPUT«$(Int, 1)␤»
Mouq m: my $c = 1; my $b = $($c, 1); $c = Int; say $b.perl
camelia rakudo-moar 8812ca: OUTPUT«$(Int, 1)␤»
timotimo no, the optimizer leaves that alone. 23:28
timotimo i really should have started making the slides much earlier :S 23:29
at least the "what do i talk about?" part
Mouq m: my $c = 1; my $b = ($c, 1); $c = Int; say $b.perl
camelia rakudo-moar 8812ca: OUTPUT«$(Int, 1)␤»
Mouq m: my $c = 1; my $b = [$c, 1]; $c = Int; say $b.perl
camelia rakudo-moar 8812ca: OUTPUT«[1, 1]␤»
Mouq m: my $c = 1; my $b = Array.new($c, 1); $c = Int; say $b.perl 23:30
camelia rakudo-moar 8812ca: OUTPUT«[Int, 1]␤»
Mouq Many WTFs to be had
p6: my $c = 1; my $b = Array.new($c, 1); $c = Int; say $b.perl 23:31
camelia niecza v24-109-g48a8de3: OUTPUT«Unhandled exception: Excess arguments to List.new, used 1 of 3 positionals␤ at /home/p6eval/niecza/lib/CORE.setting line 0 (List.new @ 1) ␤ at /tmp/tmpfile line 1 (mainline @ 4) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4595 (ANON @ 3) …» 23:31
..rakudo-{parrot,jvm,moar} 8812ca: OUTPUT«[Int, 1]␤»
Mouq n: my $c = 1; my $b = ($c, 1); $c = Int; say $b.perl
camelia niecza v24-109-g48a8de3: OUTPUT«$(Int, 1)␤»
Mouq OTOH, ($c,) having the container is the only way ($c, *) = * could work 23:33
Other than special-casing
timotimo that's true 23:34
m: my $c; say $($c, 1).WHAT 23:35
camelia rakudo-moar 8812ca: OUTPUT«(Parcel)␤»
timotimo ah, it's a parcel
those are supposed to be internal-ish magic-ish any way
to be honest, i'd rather use [ ] for nested structures
dalek ast: 2913cd1 | (Elizabeth Mattijsen)++ | S22-package-format/parse-spec.t:
Some more parse-spec tests
 23:54
Mouq lizmat: Could those for blocks be re-written as map &parse-ok, ...? 23:58
Mouq or rather &parse_ok 23:59
lizmat I guess for the single test cases, yes, good point
feels it is still early and very late already
I wonder why :-)

Powered by Cro and the Raku® Programming Language.

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

Thank you!