»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by moritz on 3 May 2013.
colomon sorear: ping? 00:45
dalek ecza: ba63d9a | (Solomon Foster)++ | FETCH_URL:
Update to latest release.
 00:50
colomon has no idea how cloud.github.com works.
rjbs colomon: It uses the cloud. 01:17
sorear colomon: hi 01:49
adu hi 01:54
colomon sorear: how do I upload stuff to cloud.github.com/sorear/niecza? it's stupid, but it's got me completely stumped. I uploaded the "new release" of niecza to my webpage in the meantime, just to get it working. 02:03
sorear colomon: You can't, github discontinued the file upload feature. 02:07
colomon sorear: glad to know I wasn't just looking in the wrong place for 30 minutes.
I probably should add an MD5 tag or something to make sure no one hacks the niecza-24.zip file at harmonyware.com. 02:09
adu so sad 02:11
colomon sorear: another question: hints for where/how Nil disappears from list comprehensions? The only thing that jumps out at me is lib/CodeGen.cs line 4375, but even that's more of a "well, maybe that could be it, but I don't see how". 02:14
sorear Nil disappears from list comprehensions because it's flagged as iterable and iterates as nothing. 02:19
colomon what flags it as iterable? 02:20
is it that bit in CodeGen? ru.setting.Nil = Kernel.NewRWListVar(nst.typeObj); 02:21
colomon guesses no, because commenting it out did not changes Nil's behavior in this respect 02:22
sorear actually, that is part of what does it 02:23
bonsaikitten colomon: just put a tag in git, then you can get a mostly working zip from github automatically 02:25
sorear bonsaikitten: that would require making a separate branch to hold the contents of the binary packages 02:29
bonsaikitten sorear: oh right. I rarely think about precompiled stuff :) 02:30
colomon sorear: is it something to do with NilP? 02:37
sorear yes 02:38
NilP is iterable, all other type objects aren't
this is handled in multiple places
colomon sorear: I haven't been able to get this working yet, but at least I know where to look now. sorear++ 03:15
colomon afk # bedtime 03:23
kresike hello all you happy perl6 people 07:55
sergot o/ 07:56
diakopter _-_
labster hello 07:57
sorear o/o/ 07:59
dagurval \o/
FROGGS \o 08:01
dagurval r: my @a <== gather for 1..Inf -> $i { print "."; take $i; } 08:06
camelia rakudo 2a04f2: OUTPUT«(timeout)................................................................................................................................................................................................................................................................…
dagurval Why isn't @a lazy?
sorear because it's the last statement in a block 08:07
try adding a ;1 after it
dagurval r: my @a <== gather for 1..10 -> $i { print "."; take $i; }; 1
camelia rakudo 2a04f2: OUTPUT«..........»
sorear erm
r: my @b = (my @a <== gather for 1..10 -> $i { print "."; take $i }) 08:08
camelia rakudo 2a04f2: OUTPUT«..........»
sorear r: my @b := (my @a <== gather for 1..10 -> $i { print "."; take $i })
r: my @b = (my @a <== gather for 1..Inf -> $i { print "."; take $i })
camelia rakudo 2a04f2: OUTPUT«(timeout)..........»
rakudo 2a04f2: OUTPUT«(timeout)................................................................................................................................................................................................................................................................…
tadzik r: my $a <== gather for 1..Inf -> $i { print "."; take $i; } 08:09
camelia rakudo 2a04f2: OUTPUT«(timeout)................................................................................................................................................................................................................................................................…
Juerd That's not annoying at all :) 08:10
dagurval so it's a bug, or are we doing it wrong? 08:13
sorear i suspect bug 08:17
dagurval r: my @a <== 1..Inf; 08:18
camelia rakudo 2a04f2: OUTPUT«(timeout)»
dagurval I'll add a test in roast for it :) 08:19
moritz suspect that feeds just aren't lazy yes 08:21
dalek rl6-bench: 9a14e25 | (Geoffrey Broadwell)++ | timeall:
Factor out some common code in time_command() in timeall
 08:23
rl6-bench: 52b7474 | (Geoffrey Broadwell)++ | timeall:
Significant refactoring of timeall to support time-scaled tests
mathw_ o/ 08:29
FROGGS \o 08:29
labster o/
mathw \o/ 08:30
tadzik \o
FROGGS *g*
labster internet weather sure has been stormy lately. must be climate change.
tadzik there's a storm coming, Mr Wayne
FROGGS if the climat is changing lizmat will tell us surely 08:31
mathw I'll see your storm and raise you a winter
tadzik I see your winter and raise you Dragonstorm! \o/
sorear nuclear winter, asteroid winter, AI winter, or regular old winter? 08:31
tadzik gatherer.wizards.com/Handlers/Image...;type=card of course 08:32
daxim 8r? …… overpriced 08:34
tadzik oh, it's not really 08:35
lizmat The climate is changing
tadzik few dark rituals and seething songs and you get your 8r and more than enough spells on stack to put 4 dragons with haste into play 08:36
instawin :)
lizmat who am I to argue with FROGGS :-) 08:36
FROGGS *g*
dalek : dceb649 | (Tobias Leich)++ | t/test.pl:
better signatures for cmp_ok
: 415b381 | (Tobias Leich)++ | t/test_summary:
parse "print 1..10" too to get the plan
: ec9e394 | (Tobias Leich)++ | STATUS.md:
update after change of parsing plan and cmp_ok
labster casts counterspell
tadzik oh noes 08:37
sorear does counterspell counter the card, or one spell? 08:37
FROGGS like $spell++ ? 08:37
sorear and does the mana cost count sepately for each copy?
tadzik it counters the entire dragonstorm
sorear has been away from m:tg for ~10y 08:38
tadzik I stopped playing around tenth edition, ate up too much monetary resources
and I was not even a student back then
and no, you pay 8r just once
labster lol, I stopped around 4th edition.
tadzik mtg is weird recently 08:39
no mana burn for one thing
and those Planeswalker cards
sorear I stopped around... I wasn't paying attention to the edition numbers :[
tadzik I've been to a couple tournaments, it was about as fun as it was expensiev
dalek ast: 0bb5a4d | (Dagur Valberg Johannsson)++ | S03-feeds/basic.t:
2 lazy tests for <== operator
 08:40
mst can handle collecting RPG supplements but not card games
doesn't quite tweak the "ooh! shiny!" for me
Timbus i thought cards with storm were copied and placed on the stack when played. therefore you would need to play a counter per copy
tadzik hmm 08:41
Timbus i had a deck that used storm and i think thats how it went
labster It led me to people who played D&D and other tabletop RPGs, and then I realized I could have a lot more fun doing group storytelling.
Timbus gatherer.wizards.com/pages/card/det...s%20Desire i used this
and you had better believe it won 08:42
FROGGS p: my @a <== 0..Inf
camelia pugs: OUTPUT«*** Unsafe function '=' called under safe mode␤ at /tmp/XUgGNSdcel line 1, column 1 - line 2, column 1␤»
FROGGS -.-
tadzik oh wow, that's a killer
labster assignment considered unsafe 08:43
tadzik A copy of a spell can be countered, just like any other spell, but each copy has to be countered individually. Countering a storm spell won't counter the copies of it.
right
Timbus i played it with a deck that used twiddle (spell: untap target, cost 1 blue) and a lotus artifact (tap to give 3 mana, pick your colour)
and yeah. i won.
labster but countering the spell that creates the storm... you just have to get far enough up the call stack, correct? 08:44
lizmat wonders if somebody messed with the channel selector
Timbus lol
tadzik hahaha
diakopte1 bzzzzzzzzzt
diakopte1 bzzzzzzzzzzzzzzzzzt 08:44
bzzzt
tadzik Timbus: what lotus artifact are you talking about? 08:44
that sounds like black lotus, but that one is sacficed. And also priced at ~$3000 iirc :) 08:45
FROGGS changes topic to: we're playing card games, not doing peril anymore...
Timbus gilded lotus i think
noooo not that
D:
tadzik oh, this one :)
dalek ast: 5bfc7fd | dagurval++ | S03-feeds/basic.t:
moved nieza fudge
 08:47
mst FROGGS: quick, put on your peril sensitive sunglasses 08:49
sorear tadzik: any luck with CBP? 08:51
tadzik sorear: I'm having my visa appointment on next monday
someone will stare at me and decide if I'm worthy 08:52
lizmat wonders whether a prolific online presence would be helpful for such a staring contest or not 08:54
lizmat also wonders who else US government officials have been staring in the past at and to what purpose 08:55
lizmat is bored doing another spectest 08:56
tadzik oh, you know, Poles are all thieves who fly to the US and stay there forever not paying taxes
arnsholt Yeah, I'm glad I could do the ESTA thing
sorear I thought they were more afraid tadzik would hijack a plane and crash it into the WTC 08:57
tadzik there was an incident when one of our former presidents, a nobel piece prize winner refused to have his fingerprints taken when flying to the US
arnsholt One of my colleagues has some terrible stories of her time with a Serbian passport, getting through passport control here in Norway
bonsaikitten arnsholt: I've heard of horrible things from people with german passportds 08:58
it's quite random and confusing, but apparently germans are expected to have white skin, OR ELSE
arnsholt sorear: Oh, but there's a question for that in the forms, so that's unpossible once they've issued a visa =p
tadzik yeah
lizmat there was a time when I was flying regularly from Amsterdam to Chicago, that I was the only one on the flight having to open my suitcase for inspection :-)
tadzik there's actually a question like this: "if you were a political leader, did you order any military actions which violated freedom of religion?" 08:59
or something like this
arnsholt Some of those questions are a bit hilarious. "Are you a terrorist? Are you a drug smuggler? Were you a nazi between 1936 and 1945?"
tadzik yeah
lizmat is puzzled by the timeframe reference
tadzik "Do you have practical experience with chemical weaponry?"
hoelzro I think on the citizenship exam (or is it the entrance document?) that asks if you're a drukkard 09:00
nwc10 I believe that it's like money laundering regulations. It's easier to prosecute someone for lying on the immigration form, than the actual actions in question.
hoelzro *drunkard
lizmat like there haven't been any nazis post 1945
sorear would Cardinal Ratzinger have been allowed in?
arnsholt lizmat: It's (slightly) more subtle than that
hoelzro I find the Nazi question interesting, considering there's a Nazi party in the US
they're small, but they're there
arnsholt Something about any association with German war crimes during WWII, IIRC
hoelzro that would make sense
sorear would like the countries of the west to shut up and get serious about that "free movement of people" thing... 09:01
arnsholt sorear: Indeed
jnthn +100000 09:02
arnsholt Unfortunately people are all for it when all they see is "no passport control when on holidays", but less than cheerful when they realize -other- people can get into their country as well
jnthn
.oO( wouldn't mind if some countries to the East then also relaxed their visa needs... :/ )
tadzik they probably don't want the british to come here with ships and blackpowder rifles and take away their homeland, right? Hah! Like that's gonna ever happen!
oh, wait
arnsholt (Norway is currently in the middle of what is turning into an annual "OHNOES TEH GYPSIES" scarefest) 09:03
nwc10 yes, exactly. It will be the new aircraft carriers, but black powder rifles because the planes aren't ready yet.
hoelzro imagines a fighter jet armed with muzzle loaded blackpowder rifles 09:04
two shots, back to base.
tadzik 3 shots per minute
if you reload fast
and it's dry
sorear arnsholt: i've never difficulty dealing with passport control, because unlike the US Luxembourg has sensible immigration people 09:05
hoelzro the second to last time I had to go through US customs, I almost missed my connecting flight
and I'm even a citizen
sorear lives a stone's throw from the us/mx border and thinks it should look more like the lu/de border 09:08
(large trebuchet or light-gas gun may be required)
bonsaikitten sorear: same amount of illegal things move over both borders already ;) 09:09
tadzik well, it's sad that people traveling is "illegal things" 09:17
arnsholt sorear: Yeah, I can imagine =) 09:23
sorear dumb americans generalize from "two russian citizens with bombs" to "all foreigners have bombs" 09:24
arnsholt (In other news, Haskell is fun) 09:25
sorear indeed.
let's talk about that instead.
arnsholt Agreed! =D 09:26
sorear haskeelllll
arnsholt I've just a few work-related milestones, so my next project is currently being doodled in Haskell
It's a bit mind-bending to work in, but pretty cool 09:27
moritz uses panda2 for the first 09:28
arnsholt thinks moritz accidentally a word 09:29
sorear time, I guess 09:31
moritz time, yes 09:33
arnsholt Yeah, I figured =) 09:38
Hmm. I've managed to read in and represent my data (it seems). Now for the trickier, less monad-y, bit
Or actually, I'm not representing my data. That's the hard bit 09:39
lizmat can't help liking www.youtube.com/watch?v=KaOC9danxNo very much 10:47
moritz wonders how stable ilbot6 on postgresql will be 10:48
lizmat wrt to optimising away +@ in .nqp files: I'm not seeing any significant difference in CPU usage of the spectest 10:52
*and* since I broke 2 tests with this batch of changes, I'm not going to pursue this any further at the present time
unless someone convinces me it is worthwhile anyway, so I should check out the breakage of these two tests 10:53
hoelzro when did Rakudo start passing S12-methods/qualified.t? 10:54
moritz it was added to t/spectest.data in December last year 10:56
and the test you're interested in was unfudged in Feb
hoelzro wow! 10:57
I guess that's not *too* old =)
I can continue work on my XMPP bot!
\o/
lizmat bad weather again :-( 11:21
lizmat r: my $a; $a{Mu}=1 12:01
camelia rakudo 2a04f2: OUTPUT«Cannot call 'postcircumfix:<{ }>'; none of these signatures match:␤:(: Mu *%_)␤:(: :p(:$p)!, Mu *%_)␤:(: :k(:$k)!, Mu *%_)␤:(: :kv(:$kv)!, Mu *%_)␤:(: :v(:$v)!, Mu *%_)␤:(: :BIND(:$BIND)!, Mu *%_)␤:(\SELF: $key, Mu *%_)␤:(\SELF: $key, Mu \$BIND, Mu *%_)␤:(\SELF: $k…
lizmat n: my $a; $a{Mu}=1
camelia niecza v24-50-gba63d9a: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/lib/CORE.setting line 1355 (warn @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 266 (Mu.Str @ 15) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/1MqAHIY22W line 1 (mainline @ …
lizmat n: my $a; $a{Mu}=1; say $a{''} 12:02
camelia niecza v24-50-gba63d9a: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/lib/CORE.setting line 1355 (warn @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 266 (Mu.Str @ 15) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/JKeMhPmawD line 1 (mainline @ …
lizmat rn: class Mu {}; my $a; $a<Mu>=1 12:04
camelia rakudo 2a04f2: OUTPUT«Nominal type check failed for parameter '$expected'; expected Any but got Mu instead␤ in sub infix:<=> at src/gen/CORE.setting:13414␤ in block at src/gen/CORE.setting:1726␤ in block at /tmp/uEs0gat_pQ:1␤␤»
..niecza v24-50-gba63d9a: ( no output )
lizmat rn: class Mu {}; my $a; $a<Mu>=1; say $a<Mu> 12:05
camelia niecza v24-50-gba63d9a: OUTPUT«1␤»
..rakudo 2a04f2: OUTPUT«Nominal type check failed for parameter '$expected'; expected Any but got Mu instead␤ in sub infix:<=> at src/gen/CORE.setting:13414␤ in block at src/gen/CORE.setting:1726␤ in block at /tmp/mecrqSHDf4:1␤␤»
FROGGS rn: my $a; $a<Mu>=1
camelia rakudo 2a04f2, niecza v24-50-gba63d9a: ( no output )
FROGGS rn: class Mu {};
camelia rakudo 2a04f2, niecza v24-50-gba63d9a: ( no output ) 12:06
FROGGS ?
moritz well
if you redeclare Mu in an inner scope, stuff isn't ~~ Mu anymore
or... something
FROGGS ahh, okay, I feared it does treat <Mu> as a string 12:08
lizmat_ (got disconnected: 2nd attempt) I'm trying to see whether I can use Mu as a key in a hash
FROGGS not*
lizmat_ e.g. when there is no :auth info for a module, it would be logical to use Mu for the key then
FROGGS hmmm
lizmat_ as that would smartmatch with Any from a "use" command 12:09
FROGGS but why do you want to redeclare Mu?
moritz rn: class Mu {}; my $a; $a<bla> = 1;
camelia niecza v24-50-gba63d9a: ( no output )
..rakudo 2a04f2: OUTPUT«Nominal type check failed for parameter '$expected'; expected Any but got Mu instead␤ in sub infix:<=> at src/gen/CORE.setting:13414␤ in block at src/gen/CORE.setting:1726␤ in block at /tmp/Q2Y322EDl7:1␤␤»
lizmat_ I don't
moritz FROGGS: it's not related to the string being Mu
lizmat_ I just wanted to see whether it made a difference
FROGGS rn: my $a; $a{Mu} = 1;
camelia niecza v24-50-gba63d9a: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/lib/CORE.setting line 1355 (warn @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 266 (Mu.Str @ 15) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/ICdEC1vCTf line 1 (mainline @ …
..rakudo 2a04f2: OUTPUT«Cannot call 'postcircumfix:<{ }>'; none of these signatures match:␤:(: Mu *%_)␤:(: :p(:$p)!, Mu *%_)␤:(: :k(:$k)!, Mu *%_)␤:(: :kv(:$kv)!, Mu *%_)␤:(: :v(:$v)!, Mu *%_)␤:(: :BIND(:$BIND)!, Mu *%_)␤:(\SELF: $key, Mu *%_)␤:(\SELF: $key, Mu \$BIND, Mu *%_)␤:(\SELF: $k…
FROGGS r: my $a; $a{Any}=1; say $a 12:10
camelia rakudo 2a04f2: OUTPUT«use of uninitialized value of type Any in string context in block at /tmp/zpowZQI7uT:1␤␤("" => 1).hash␤»
jnthn I'm pretty sure redeclaring Mu will make all kinds of difference. :) 12:11
lizmat_ sorry, please forget about that :-)
FROGGS "all kinds"... yeah, that fits :o)
lizmat_ but being able to use Mu as a key in a hash, does not seem possible at present 12:12
FROGGS strings only atm 12:12
lizmat_ r: say ~Mu
camelia rakudo 2a04f2: OUTPUT«Nominal type check failed for parameter ''; expected Any but got Mu instead␤ in sub prefix:<~> at src/gen/CORE.setting:1271␤ in block at /tmp/hVqTelct4Z:1␤␤»
lizmat_ n: say ~Mu 12:13
camelia niecza v24-50-gba63d9a: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/lib/CORE.setting line 1355 (warn @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 266 (Mu.Str @ 15) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/s9afKbLVLF line 1 (mainline @ …
FROGGS why is the value auf :auth a key in a hash? 12:15
FROGGS what is the value then? 12:15
lizmat_ well, that's a proposal I'm working on, to support module Foo:name<Foo::Bar>:from<perl6>:auth<cpan:JRANDOM>:ver<1.0.0> {} syntax
lizmat_ Mu was supposed to be the value in absence of a specific :auth<cpan:JRANDOM> 12:16
FROGGS so, the value will be the path? or a reference to the module?
I'd vote for 'Any' and '*' beeing these special keys 12:17
(as strings I might need to mention)
lizmat_ rn: class Foo::Bar {}; say OUR::; say OUR::Foo::
camelia niecza v24-50-gba63d9a: OUTPUT«Stash.new(...)␤Stash.new(...)␤»
..rakudo 2a04f2: OUTPUT«("Foo" => Foo).hash␤("Bar" => Foo::Bar).hash␤»
lizmat_ in stashes
FROGGS hmmmm 12:18
lizmat_ from/auth/ver would become part of the longname, as it were
FROGGS if we had hashes as keys, you could do: say OUR::Foo:ver<*>::Bar:: 12:19
jnthn Probably you just need a way to canonicalize the names
FROGGS but there is a bit of black magic needed to get the right key
lizmat_ jnthn: I'm writing a document exploring options in this area 12:20
jnthn It's like infix:<+> is the proper name we install under, even though you may declare it as infix:['+'] or infix:<< + >> or whatever.
lizmat_ well, that would be the local alias, as it were, right>
?
FROGGS jnthn: yeah, but what about wildcards? 12:21
and regex search patterns? >.<
jnthn FROGGS: I'm talking about naming, not searching.
lizmat_ like "module Baz:name<Foo::Bar>" would declare Foo::Bar and make it available as Baz
lizmat_ we need a naming scheme that will allow for easy searching 12:22
FROGGS right, but if you could do that in general, it would be cool: say OUR::Foo:ver{/^1\.<[345]>/}::Bar::
lizmat_ but in the context of namespaces, I guess we can get away with 'Mu' as a string
it's just not as easy to smartmatch
FROGGS: shouldn't that need to be: OUR::Foo::Bar:ver{/^1\.<[345]>/}:: ? 12:23
FROGGS I hope TimToady++ doesnt read what I wrote otherwise we have some NYI's more :o)
lizmat_: both could work 12:24
lizmat_ I think that is an implementation detail, really, so not a TimToady++ concern?
jnthn If that kind of thing should work is a language design concern. 12:24
FROGGS I was thinking about hashes in general
if you would have hashes as keys, you could search for a key using adverbs as search patterns 12:25
jnthn Hashes default to having string keys. Stashes certainly do, and if you start trying to make it otherwise it cascades out all over the place.
lizmat jnthn: I agree
OTOH, something similar is going on with sets of sets 12:26
jnthn Anyway, I need to get back to doing $dayjob stuffs...
FROGGS my %hash = ( :value<Foo>, :auth<cpan:FROGGS> ) => 'Foo/Bar.pm'; say %hash<Foo>:auth({ /FROGGS/ }) 12:27
that looks perl6y to me fwiw 12:28
dalek rl6-roast-data: 3de2c0a | coke++ | / (4 files):
today (automated commit)
 12:29
lizmat is trying to wrap her mind around FROGGS' idea
lizmat hmmm… named parameters on circumfix{} 12:30
it's been done, I guess: 12:31
rn: my %a= ( a => 1 ); say %a<a>:p
camelia rakudo 2a04f2, niecza v24-50-gba63d9a: OUTPUT«"a" => 1␤»
lizmat but this syntax would then not allow you to have any "p" key in the hash of the key 12:32
as it would be ambiguous 12:33
rn: my %a= ( a => 1 ); say %a<a>:auth<FROGGS>
camelia rakudo 2a04f2: OUTPUT«1␤»
..niecza v24-50-gba63d9a: OUTPUT«Unhandled exception: Excess arguments to KERNEL Hash.postcircumfix:<{ }>, unused named auth␤ at <unknown> line 0 (KERNEL Hash.postcircumfix:<{ }> @ 1) ␤ at /home/p6eval/niecza/lib/CORE.setting line 183 (postcircumfix:<{ }> @ 5) ␤ at /tmp/pYzyHhkRgK line…
FROGGS dunno if it really is a good idea 12:35
it is a bit of a question if you need that often or not
lizmat I'm not sure about the current low level implementation of sets 12:36
but TimToady++ seemed to indicate a set of sets should be possible
lizmat which would indicate some magic wrt keys there to me, if the implementation is based on hashes 12:37
jnthn Yes, but those can sanely be object hashes. 12:38
lizmat is not sure what object hashes are 12:39
you mean where the key can be on object? then $a{Mu} would also work, no? as Mu is also an object?
moritz yes 12:41
lizmat but object hashes are NYI, right? 12:42
jnthn They're implemented but using them for Stashes is probably stupid.
lizmat from a performance point of view? 12:43
jnthn That, and also what happens when you need to install the thing lexically?
jnthn Remember that packages in Perl 6 can be lexical. 12:43
lizmat yes, but that just means they live in MY:: , no?
jnthn Well, MY:: is a high-level view into them 12:44
FROGGS I'm not sure at all how it will work out if you want to use several modules that just differ by auth or version...
jnthn Under the hood we don't really look up most lexicals by name. 12:44
lizmat FROGGS: well, that's exactly what I'm trying to figure out
jnthn: it's more for introspection, I guess 12:45
FROGGS .oO( my $dog = Dog:auth({/cpan/}).new( ... ) ) 12:47
lizmat FROGGS: no, you would do something like: use Pooch:name<Dog>:auth({/cpan/}); my $dog= Pooch.new(…) 12:48
lizmat where you could simplify this by saying: use Dog:auth({/cpan/}); my $dog= Dog.new(…) 12:49
FROGGS yeah
lizmat the use would make sure you have an alias (presumably OUR::) for the loaded module
lizmat and it would fail on the use if there was more than one candidate that matched 12:50
FROGGS yeah, you dont come around that alias, since you can't 'use' two modules with the same modulename in the same scope 12:50
lizmat indeed 12:51
see also: gist.github.com/lizmat/5557120 12:52
lizmat actually, I'm not sure that that example is ever going to work, since "use" currently always implies a search in %*INC to check if already loaded 12:53
FROGGS well, maybe its longname (evenlongerlongname?) must be the key of %INC then 12:57
or, hmmmm
FROGGS damn, there is always a 'but' in my mind when writing a sentence here 12:58
lizmat is going to get some fresh air for some hours 12:59
lizmat rn: my %a= (a => 1, b => 2); say %a{*} :delete:p # shouldn't this return pairs ? 20:14
camelia niecza v24-50-gba63d9a: OUTPUT«(Any)␤»
..rakudo 2a04f2: OUTPUT«1 2␤»
jnthn I'm...not sure you can stack 'em... 20:15
colomon rn: my %a= (a => 1, b => 2); say %a{*} :delete
camelia niecza v24-50-gba63d9a: OUTPUT«(Any)␤»
..rakudo 2a04f2: OUTPUT«1 2␤»
lizmat niecza seems wrong here
colomon rn: my %a= (a => 1, b => 2); say %a{'a'} :delete
camelia rakudo 2a04f2, niecza v24-50-gba63d9a: OUTPUT«1␤»
colomon rn: my %a= (a => 1, b => 2); say %a{'a'} 20:16
camelia rakudo 2a04f2, niecza v24-50-gba63d9a: OUTPUT«1␤»
colomon rn: my %a= (a => 1, b => 2); say %a{*}
camelia rakudo 2a04f2, niecza v24-50-gba63d9a: OUTPUT«1 2␤»
lizmat well, some stacking may be nonsensical
colomon indeed
lizmat++ 20:17
lizmat my %a= (a => 1, b => 2, c => 3); my %b= %a<a b> :delete:p; say %a; say %b # use case for :delete:p 20:18
rn: my %a= (a => 1, b => 2, c => 3); my %b= %a<a b> :delete:p; say %a; say %b # use case for :delete:p
camelia niecza v24-50-gba63d9a: OUTPUT«{"c" => 3}␤{"1" => 2}␤»
..rakudo 2a04f2: OUTPUT«("c" => 3).hash␤("1" => 2).hash␤»
lizmat would have expected {c => 3} {a=>1, b=>2}
pmichaud_ Can't stay long -- I just added a note to RT #117935 20:22
(haven't read backscroll either... just read the ticket)
lizmat pmichaud++ even if it is only for a small time 20:23
pmichaud 19:32 <lizmat> r: say True ?^ True # shouldn't this be False ?
Yes, that should probably be False.
But
r: say (True ^ True); 20:24
camelia rakudo 2a04f2: OUTPUT«one(True, True)␤»
pmichaud er
r: say (True ^^ True);
camelia rakudo 2a04f2: OUTPUT«Nil␤»
pmichaud That one is correct.
r: say (True xor True)
camelia rakudo 2a04f2: OUTPUT«Nil␤»
pmichaud That one is correct too.
colomon n: say (True ^^ True);
camelia niecza v24-50-gba63d9a: OUTPUT«Unhandled exception: System.NullReferenceException: Object reference not set to an instance of an object␤ at Niecza.StashCursor.Core (System.String key, Boolean final, Niecza.StashCursor& sc, Niecza.Variable& v, Niecza.Variable bind_to) [0x00000] in <file…
colomon ugh 20:25
pmichaud oops
Nil is no longer correct, it should now be ()
wait
huh? 20:26
pmichaud re-reads the now updated spec.
pmichaud okay, spec says that True ^^ True should return False. 20:28
and True xor True should return False
and 1 ^^ 1 should return False 20:29
except in mythical list context, these are to magically return () somehow.
so I call fossil.
masak o.O
what was the original rationale for returning Nil? I forget.
pmichaud apparently the spec never said Nil, or at least it hasn't said that since 2011 20:30
but the rationale would be that Nil acts like False in item context and it acts like () in list context.
which is what the spec says.
(acts like () in list context before the recent changes to the meaning of Nil)
anyway, I call fossil and think we need a spec issue to get it cleared up. :) 20:31
dalek p/rak-jvm-support: 740258f | jnthn++ | src/vm/jvm/ (2 files):
Assorted code-gen improvements.

With these, we now get a CORE.setting.class generated.		 20:32
pmichaud grrrr... RT just sent another 36-hour-old bug report.
jnthn++
jnthn At first I was a bit shocked it's 9MB. Then I checked how big the PBC is, which weighs in at 16 MB... 20:33
pmichaud oh, and I just got a 48 hour old bug report also.
nwc10 yay 20:35
masak \o/
jnthn bbi10 20:40
[Coke] r: my $c; my $name; BEGIN { say "1"; $c = { say "OH HAI $name" } }; $name = "masak; $c(); 20:42
camelia rakudo 2a04f2: OUTPUT«1␤===SORRY!===␤Unable to parse expression in double quotes; couldn't find final '"'␤at /tmp/vuTcTJgfpt:1␤------> OH HAI $name" } }; $name = "masak; $c();⏏<EOL>␤ expecting any of:␤ postfix␤ argument list␤»…
dalek : 01c9bc7 | (Tobias Leich)++ | lib/Perl5/Actions.nqp:
make $@ always a string
: d495a3c | (Tobias Leich)++ | lib/Perl5/Grammar.nqp:
distingiush between eval BLOCK and eval EXPR
: 16b80bd | (Tobias Leich)++ | lib/Perl5/Terms.pm:
proper value for $/
[Coke] r: my $c; my $name; BEGIN { say "1"; $c = { say "OH HAI $name" } }; $name = "masak"; $c();
camelia rakudo 2a04f2: OUTPUT«1␤use of uninitialized value of type Any in string context in block at /tmp/Gl_wB_rfCf:1␤␤OH HAI ␤»
[Coke] r: my $c; my $name; BEGIN { say "1"; $c = sub { say "OH HAI $name" } }; $name = "masak"; $c();
camelia rakudo 2a04f2: OUTPUT«1␤use of uninitialized value of type Any in string context in sub at /tmp/nZvStnLTwR:1␤␤OH HAI ␤»
[Coke] r: my $c = { say "alive"}; 20:44
camelia rakudo 2a04f2: ( no output )
[Coke] r: my $c = { say "alive"}; $c();
camelia rakudo 2a04f2: OUTPUT«alive␤»
dalek pan style="color: #395be5">perl6-examples: 08f6933 | (David Warring)++ | parsers/CSSGrammar.pm:
added css grammar comments; tweaked newline handling
 20:54
pan style="color: #395be5">perl6-examples: 82fa422 | (David Warring)++ | rosalind/ (4 files):
Merge branch 'master' of github.com:perl6/perl6-examples
lizmat seems like creating a candidate for true values of a named parameter silently doesn't work, is that correct? 21:11
something like "multi method postcircumfix:<{ }>(\SELF: $key, :$exists! where True )"
flussence r: say 1 ~~ True 21:13
camelia rakudo 2a04f2: OUTPUT«Potential difficulties:␤ Smartmatch against True always matches; if you mean to test the topic for truthiness, use :so or *.so or ?* instead␤ at /tmp/djMcaWxYio:1␤ ------> say 1 ~~⏏ True␤1True␤»
flussence that might be your problem
lizmat r: say 0 ~~ True 21:14
camelia rakudo 2a04f2: OUTPUT«Potential difficulties:␤ Smartmatch against True always matches; if you mean to test the topic for truthiness, use :so or *.so or ?* instead␤ at /tmp/y6p9nFOENV:1␤ ------> say 0 ~~⏏ True␤0True␤»
lizmat r: say True ~~ 1
camelia rakudo 2a04f2: OUTPUT«True␤»
lizmat r: say True ~~ 0
camelia rakudo 2a04f2: OUTPUT«False␤»
lizmat aha… ok
flussence (I didn't know there was a :so, that looks like it'll come in useful...) 21:17
sorear smartmatching against Pair does a dynamic method call 21:23
foo ~~ :bar tests like foo.bar
foo ~~ :bar(quux) tests like foo.bar ~~ quux
flussence sorear++ # *That* is useful to know 21:24
[Coke] r: say 3 ~~ :Rat(23) ; say 3.Rat ~~ 23 21:31
camelia rakudo 2a04f2: OUTPUT«True␤False␤»
dalek : 063bd31 | (Tobias Leich)++ | t/test.pl:
added stubs for native_to_latin1 and latin1_to_native
 21:36
: 554f8f3 | (Tobias Leich)++ | lib/Perl5/ (2 files):
treat "local" like "my"
lizmat FROGGS: in Perl 5, local is dynamic like temp 21:38
FROGGS err, what do you mean? 21:41
lizmat perl5: our $a="foo"; do { say $a; local $a= "bar"; say $a }; say $a 21:42
p5eval: our $a="foo"; do { say $a; local $a= "bar"; say $a }; say $a
p5eval lizmat: foobarfoo1
FROGGS r: our $a="foo"; do { say $a; my $a= "bar"; say $a }; say $a 21:43
camelia rakudo 2a04f2: OUTPUT«===SORRY!===␤Lexical symbol '$a' is already bound to an outer symbol;␤the implicit outer binding must be rewritten as OUTER::<$a>␤before you can unambiguously declare a new '$a' in this scope␤at /tmp/e6zyFr8ck_:1␤------> our $a="foo"; do { sa…
FROGGS r: our $a="foo"; do { say $a; temp $a= "bar"; say $a }; say $a
camelia rakudo 2a04f2: OUTPUT«foo␤bar␤foo␤»
FROGGS I see
masak p5eval: our $x = "foo"; sub quux { say $x }; do { say $x; local $x = "bar"; quux }; $x
p5eval masak: foobarfoo
lizmat masak++
masak that shows the dynamic part. 21:44
lizmat indeed
tadzik wikipedia says: "he has more than 8000 likes on facebook"
what the fuck, really. Who cares. What does that even mean.
lizmat also: local $a without assignment sets $a to undef 21:45
tadzik en.wikipedia.org/wiki/Stubbs_(cat)
FROGGS yeah... # lizmat++, masak++
lizmat spectesting :exists & :delete fixes 21:46
will look at it tomorrow and commit tomorrow 21:47
goodnight #perl6!
colomon o/
FROGGS gnight all 21:48
masak 'night, #perl6 22:07
jnthn 'night, #perl6 23:14

Powered by Cro and the Raku® Programming Language.

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

Thank you!