»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! | tinyurl.com/p6contest
Set by moritz_ on 28 December 2010.
dalek ecza: a058b70 | sorear++ | lib/ (2 files):
Remove some redundant Perl 6 definitions
 01:07
snarkyboojum dobar dan #perl6 03:36
colomon \o
snarkyboojum loving the p1 solutions 03:37
snarkyboojum phenny: tell tadzik, when is the metainfo branch of neutro going to replace master? :) 03:46
phenny snarkyboojum: I'll pass that on when tadzik is around.
dalek ecza: f5d2a4a | sorear++ | / (5 files):
Make item, list, hash primitive contexts
 03:51
thundergnat rakudo: warn 'wtf!'; 03:58
p6eval rakudo 388eed: OUTPUT«wtf! in <anon> at line 22:/tmp/NmI0nXWG70␤»
thundergnat rakudo: try { die 'omg!'; CATCH { say 'wtf!'; } }; 03:59
p6eval rakudo 388eed: OUTPUT«wtf!␤»
thundergnat rakudo: try { die 'omg!'; CATCH { warn 'wtf!'; } };
p6eval rakudo 388eed: OUTPUT«(timeout)»
thundergnat That last one dumps core locally. :(
Parrot VM: PANIC: Out of mem! 04:00
lue phenny: tell masak s/custum/custom/ on strangelyconsistent.org/p6cc2010/p1-util/ (under "Clarity of Intent") 06:23
phenny lue: I'll pass that on when masak is around.
sorear hello lue 06:30
moritz_ good *, #perl6 06:56
dalek ecza: af8228c | sorear++ | / (3 files):
Implement mostly eager map and [] primitives
 07:08
dalek odel: 3ced6a4 | (Martin Berends)++ | java/compiler/Makefile:
[java/compiler/Makefile] add a test target for unixy systems - thanks to debugging by jnthn++, 6model/java passes 45 of the 299 nqp tests that 6model/dotnet passes!
 07:41
dalek ecza: bf31133 | sorear++ | lib/ (2 files):
Add mostly-eager grep primitive
 07:51
dalek ecza: bb55e22 | sorear++ | / (2 files):
Rewrite map, grep, [] calls to use the new primitives
 08:08
rod_ msg\ rakudo: say time (3:4) 08:23
msg/ rakudo: say time (3:4)
diakopter ?
rod_ just nothing..... 08:24
my apology..
diakopter :) 08:25
rod_ im waiting for a friend..
jnthn Morning, #perl6 08:27
rod_ ;)
jnthn mberends++ # yay 08:28
mberends hi jnthn, back in NL today, a bit of time for hacking :) 08:29
rod_ NL? 08:30
mberends Netherlands 08:31
rod_ ok.. im getting educated here.. :)
huching...hmm.. a hucker... you are?
mberends not a hooker, if that's what you were thinking ;) 08:32
jnthn lol :)
rod_ just jowking......... :) 08:33
i mean hucker...?
are you..??
mberends probably not, since the word hucker is unknown to me in any language
rod_ hmmm. why using it.. since...your dictionary denied it.. 08:34
so whats the word hucking does mean to you....in anyway....! 08:35
mberends rod_: I said hacking with an A. This off topic conversation is closed.
jnthn: for 6model/java I am considering a emit switch with possible values java/jasmin/bytecode. 08:36
jnthn: does that align with your possible plans for backend options?
rod_ ok..... i just offened an offender.................Xp 08:37
rod_ rakudo: say 08:40
p6eval rakudo 388eed: OUTPUT«␤»
jnthn mberends: I see no problem with that - I guess it's just picking a different JST2X
rod_ rakudo: say 1+1 08:41
p6eval rakudo 388eed: OUTPUT«2␤» 08:41
rod_ rakudo: say night1+night2
p6eval rakudo 388eed: OUTPUT«Could not find sub &night1␤ in main program body at line 22:/tmp/UuVZZ6hkga␤»
rod_ rakudo: say 45%(86+46+56){465-400} 08:42
p6eval rakudo 388eed: OUTPUT«Method 'Bridge' not found for invocant of class 'Failure'␤ in 'infix:</>' at line 3706:CORE.setting␤ in 'infix:<%>' at line 3711:CORE.setting␤ in main program body at line 22:/tmp/_KAQT288Dz␤»
rod_ rakudo: say 45%(86+46+56){deluge} 08:43
p6eval rakudo 388eed: OUTPUT«Could not find sub &deluge␤ in main program body at line 22:/tmp/Fclo7aPZGq␤»
rod_ rakudo: say 45%(86+46+56){deluge}/anode-24 08:44
p6eval rakudo 388eed: OUTPUT«Could not find sub &deluge␤ in main program body at line 22:/tmp/cUONgDxWgf␤»
dalek odel: 1e7f3cb | (Martin Berends)++ | java/runtime/Rakudo/ (3 files):
[java/runtime] sync with dotnet/ and clean out some debugging code
 09:05
IllvilJa colomon: REPENT! I SAW YOUR BLOG! YOU MENTIONED THE "Real.sin" METHOD IN IT!! HOW COULD YOU? :-) 09:27
Oh, BTW, hello everyone!
colomon you don't believe in sin?
IllvilJa Perl 5 and 6 are strange languages, you can bump into bless($self) as well as Real.sin... 09:28
flussence rakudo: (1i).sin
p6eval rakudo 388eed: ( no output )
flussence there's that too...
colomon rakudo: say (1i).sin
p6eval rakudo 388eed: OUTPUT«0 + 1.1752011936438i␤»
IllvilJa I think we need to implement Unreal.sin!
colomon yes, there is also imaginary sin in Perl 6. 09:29
rod_ i'd like to give sympathy to the sinner (colomon)
IllvilJa You mean as in "dirty thoughts"?
I assume fractional sin is an option too...
colomon rational sin is an option, yes. 09:31
colomon is going back to bed. afk
rod_ gnu:assert 10:44
mica_ella_tuazon gnu:deluge 10:45
raduko: say 1414-143 {assert6}
raduko: say 1414-143 {assert6} 10:46
raduko: say1414-143 {assert6}
png: assert 10:47
jnthn rakudo, not raduko :) 10:48
rod_ example: 2456150 last three time 2 = price
mica_ella_tuazon total 300 con. by 200 10:49
you have no idea... so shut up... NOBS 10:50
masak oh hai, #perl6! 10:55
phenny masak: 06:23Z <lue> tell masak s/custum/custom/ on strangelyconsistent.org/p6cc2010/p1-util/ (under "Clarity of Intent")
jnthn masak: Getting up earlier FAIL? :)
masak lue: thanks :)
lue++
jnthn: aye :/
jnthn :/
jnthn grudgingly managed
masak at ~2:00 am, I felt it was too warm indoors. turned off the radiators. now it's... freezing-ish inside. :) 10:57
jnthn D'oh.
The temperature has decidedly dropped overnight too.
arnsholt_ Ah, the joys of Scandinavian living =) 11:01
arnsholt (I have the same problem, incidentally) 11:02
Especially since I live in an old building and my room has single glazed windows
masak having tried settings 5 (too warm) and 0 (goosebumps), I'm now exploring 3.
arnsholt: I think people here in Sweden are silently proud of almost everything being triple-glassed. 11:03
masak czechs how many layers his apartment has
mathw I was impressed when I was in Sweden with the glazing and insulation 11:06
I was there in a very hot summer, and the buildings somehow managed to stay bearable
while still being obviously constructed to handle a horribly cold winter 11:07
in the UK, we build buildings which are suitable for neither winter nor summer
huf insulation is insulation, goes either way
mathw the only thing they are good at is keeping the rain outside
and some of them struggle with that
it really is quite shameful 11:08
huf you dont get extreme summers/winters though?
mathw they do happen
but they're not predictable
but if you build here to, say, Swedish norms, you can heat the place by lighting a single candle (almost) 11:09
huf :)
mathw so shouldn't we be doing that
masak yes, you should. 11:09
mathw modern houses are better insulated, there are rules
but they're still not as good
masak I say this having stayed in some UK buildings.
flussence my house is so waterproof that I just wrote "hose" and didn't notice until I was trying to think of a suitable analogy. 11:10
huf i still prefer the older houses, although upgrading the windows is a must
mathw I'm trying to persuade my landlord to upgrade the windows
but this seems unlikely to happen
since it's kind of expensive
huf they built proper walls back in the 1910-1930 timespan 11:11
arnsholt masak: So are most Norwegians. Unfortunately I don't think my building has seen much renovation the last three or four decades :/
huf (at least here)
arnsholt But at least the rent is cheap =)
masak as far as I can see, my windows are double-glassed. 11:12
huf yes, but are the two layers stuck together or each in their separate frame? 11:13
mathw mine are double-glazed... with about a 4mm gap between the panes
my parents have rather more modern windows, with a 20mm gap - and perhaps more importantly, frames that don't admit draughts 11:14
huf i used to have the old-style wooden frame, double window thing. that was horrible 11:15
now it's nice plastic frame, thermo glass
flussence I have one of those :(
the rope's broken on it.
huf now the only place my room vents is above the windows, through the casing of the blinds
rope? on a window? how? 11:16
mathw sash window? ow 11:17
those are expensive to get fixed
huf the kind that dont open but slide up? ow
i hated those when we had them (in the us)
flussence yeah, my house is pretty terrible for insulation. The wooden frame on our back door is rotting away too, but it's a nonstandard size so it's impossible to get fixed. 11:19
(and I keep banging my head on the top of it.)
mathw doh 11:20
masak Rakudo release day today? 11:45
jnthn Believe so. 11:52
masak and ++tadzik is the one to do the releasing. \o/ 11:53
frettled masak: why are you using prefix ++? :)
frettled seems to have missed a memo.
masak frettled: pre-incrementing people who haven't performed the kudosworthy act yet? 12:32
frettled masak: ahaha, clever. 12:34
++masak for that upcoming blog post :D
masak hah
frettled It's almost fiendishly mst-ish.
takadonet morning all 12:56
colomon o/ 12:58
masak \o 13:06
flussence evil code time! 13:16
rakudo: say "LOLOLOL!1"~!!111 but not so ....fail
p6eval rakudo 388eed: ( no output )
rod_ hahahah.. evil... exist??? 13:17
evil.. rakudo.......self ternatination...... 13:18
rod_ im begging for maple syrrup... 13:20
colomon mmmmm.... maple syrup. sorry, don't have enough to share. 13:23
tadzik o/ 14:11
phenny tadzik: 03:46Z <snarkyboojum> tell tadzik when is the metainfo branch of neutro going to replace master? :)
tadzik snarkyboojum: one day, yes :)
tadzik ah, when? Dunno, maybe when masak's modules will arive 14:12
it's release time!
any last words?
PerlJam tadzik: good luck! :)
tadzik :)
takadonet tadzik: how many modules have meta.info file? 14:13
tadzik takadonet: like 60% so far 14:13
takadonet nice 14:14
jnthn Go tadzik! :)
tadzik :} 14:14
jnthn: how about a one-sentence review of the caching that speeds up everything? For the NEWS 14:15
it's a hit, people like performance
oh, what PM group?
ok, can be BristolBath
jnthn tadzik: Hmm
Implemented nominal type check caching in signature binding, giving a large performance improvement, especially for programs that are call-heavy. 14:17
...it's a long sentence. :)
tadzik does `perl tools/update-tai-utc.pl src/core/tai-utc.pm` takes ages to run for you too? 14:19
flussence it's doing ftp-ey stuff in there, I've always found the speed of such things is very hit-and-miss. 14:20
tadzik bah, ETOORESTRICTIVEFIREWALL
can someone run this for me?
PerlJam tadzik: I would if my local clone wasn't behaving strangely. 14:21
a "git pull" is taking forever for no apparent reason.
tadzik why BristolBath btw? 14:23
PerlJam no new leap seconds. 14:24
masak tadzik: I've also asked that once. got no reply.
tadzik alright
PerlJam BristolBath is the name that gets used when no one can think of a better name? :) 14:25
masak it's currently the only suggestion in the list of suggestions in the release guide.
so, in a sense, yes.
tadzik it'd be nice to have some nice words about it 14:26
PerlJam perl.bristolbath.org has some Perl 6 stuff on it, but it all looks incomplete or old 14:27
tadzik any idea who is "quihw"? 14:30
masak tadzik: sounds familiar. 14:32
tadzik: you can always write just the nick. 14:33
tadzik that's what I'll do
dalek kudo: a7d5f19 | tadzik++ | unknown:
Added announcement for 2011.01
 14:38
tadzik any last-minute additions/changes? 14:39
masak reads 14:39
PerlJam tadzik: I think quihw is a misspell for Hongwen Qui's nick 14:39
er, I did it too! Qiu 14:40
PerlJam qiuhw == Hongwen Qiu 14:40
tadzik PerlJam: it came from "quihw++" in moritz's commit
masak was thinking the same thing
should be qiuhw++, then. 14:41
but could be a typo.
PerlJam According to google, that commit is the only place that nick appears in the #perl6 logs
tadzik fixed
masak is surprised that he hasn't contributed this month 14:42
oh well, something to do better till next release :)
tadzik oh
PerlJam I wonder if we should include commit stats (and authors) for roast in this announcement too 14:43
Those are indirect contributions to Rakudo, but maybe more appropriate for R* 14:44
masak they are indirect contributions to all implementations...
PerlJam indeed. 14:45
dalek kudo: fc2ffd7 | tadzik++ | VERSION:
[release] bump VERSION
kudo: 0289d97 | tadzik++ | docs/announce/2011.01:
Small fix in release announcement
PerlJam I'd rather every implementation mention them than none mention them because they aren't implementation specific.
JimmyZ So release roast? 14:49
masak interesting notion.
question is how to motivate the extra work/administration that entails. 14:50
JimmyZ I think no implementations pass all roast 14:52
masak correct. that'd be the day.
JimmyZ most whenever 14:53
masak going with the strict interpretation of "pass[ing] all of roast", I would guess that no implementation ever will.
if there ever is one, so much the better. 14:54
but usually there are a couple of TODOs or SKIPs in every project, even very complete ones.
JimmyZ yes
tadzik release done! 14:55
flussence postincrement!
PerlJam If an implementation passes all of roast, I think that's a good indication that roast needs more tests :)
JimmyZ I'd like define roast version, So implemetation can say 'we pass roast 1.1', or 'we pass roast 1.5', something like C89 and C99 14:56
or like java? 14:58
PerlJam masak: sounds like JimmyZ has the requisite motivation already ;)
masak :)
tadzik++
JimmyZ PerlJam: just got inspiration from you :) 15:00
tadzik (check if it runs)
PerlJam JimmyZ: good deal! I'm like that sometimes. I make sparks here and there and sometimes they create fire :) 15:01
JimmyZ PerlJam: hehe
something like "火上浇油' 15:02
masak fuel? 15:03
JimmyZ yep
masak "pouring oil on fire" :P
JimmyZ most means "you're unhappy and I make you more unhappy", it's not good words in china 15:04
哈哈 15:05
masak JimmyZ: you make us happy \o/
JimmyZ thanks :) 15:05
masak today's mini-challenge: write a multi join that special-cases the separator between the last two elements (if any): "A, B, C, D, E, and F" (the separator being ', and' between E and F. 15:12
this was brought up during Advent 2009, but I think that nicer solutions can be wrought than the one I did then.
pmichaud good morning #perl6 15:18
jnthn multi method join(*@ [$cur, *@rest]) { self[0..*-2].join(|@rest) ~ $cur ~ self[*-1] }
masak pmichaud! \o/
jnthn o/ pmichaud
masak jnthn: yes, that was what I had in mind. nice with the use of a nested signature. wouldn't that one collide with the regular &join, though? 15:20
jnthn Why? 15:21
Oh, if it gets exported...
masak because it also has that signature. 15:21
I think the special last separator needs to be a required named parameter.
jnthn Well, I had it that way so you could have as many separators as you want :)
masak ooh :)
that's over-solving the problem, though :) 15:22
jnthn A, B, C, D or E or even F
:P
I know but I wanted to show off my 1337 sig skillz, you see. :)
But yes, it is overkill perhaps. :)
masak jnthn: you have 1337 sig skillz. now try to solve the problem so that your new &join multi can co-exist with the regular one. that's part of the problem. :) 15:23
jnthn I'll think about it on the way to systemet. :) 15:24
jnthn bbiab :)
flussence meep. 15:25
ok 2 - Fancy .join works
pmichaud jnthn: ping 15:27
masak flussence++ # TDD
jnthn pmichaud: argh... :)
pmichaud: I didn't leave yet. :)
jnthn can go in a little bit ;)
pmichaud: pong ;) 15:28
pmichaud I may be a bit delayed on looking at code -- we're back in the emergency room again this morning :-|
jnthn pmichaud: Oh no. :-( Very sorry to hear that. :(
pmichaud: And of course, the code can wait. Always. 15:29
pmichaud I'm sure this morning will be spent waiting for test results and the like 15:31
tadzik++ # rakudo release #37 15:32
do we have release managers in place for the next few months?
tadzik has anyone checked yet if it works? 15:34
jnthn pmichaud: No, all slots are vacant at the moment. 15:35
pmichaud: I'm sure somebody will step up for them though.
pmichaud tadzik: I'll set it up for checking 15:36
just a sec
jnthn really afk for 15 mins or so 15:37
mberends pmichaud: will you permit me give one of your Rakudo talks to FOSDEM (Brussels 5-6 February) 15:48
masak mberends++ # going to FOSDEM 15:51
colomon I'm planning on giving a p6 talk at Penguicon in April. Hadn't really considered giving one of pmichaud's talks, but I might borrow heavily if that's okay. 15:53
mberends if szabgab++ gets hold of you there is no escape ;)
pmichaud it's okay 15:55
mberends: please feel free to steal liberally
I wish I was going to FOSDEM. I had so much fun when I went in 2008
mberends thank you, it will be the appropriate amount of fun :) 15:56
PerlJam github seems to be having problems today. I keep getting "Something went wrong" while browsing repos and my "git pull" of rakudo took like 3 minutes for 7 objects
pmichaud mberends / colomon : let me know if you need any additional materials for the talks
pmichaud PerlJam: yes, I got an angry unicorn a bit earlier when trying to download #37 15:56
colomon pmichaud: thanks! 15:57
PerlJam and, yeah, I get the unicorn occasionally too
pmichaud idly wondered if angry unicorns had anything to do with angry birds. 15:58
jnthn back 16:05
sorear good * #perl6 16:45
masak \o
pmurias sorear: hi
masak hm. consider a code analysis tool. it sees a call to a multi method. it wants to try to figure out *which* multi will bind, if any. 16:50
...there's probably a Halting Problem involved, since types are involved...
...but it's essentially the same as the runtime will have to do, except with possibly slightly less information available. 16:51
pmurias the runtime has more information available
much more
masak that's what I said.
minus the "much" :)
the multi call could be late-bound by intent. 16:52
pmurias but we have to do that to get even vaguely resonable performance on numerics
masak in that case, we won't ever get *one* candidate to fall out of the analysis. but it's not necessarily a conflict.
it's more of a case-by-case analysis. 16:53
pmurias you have to do type inference
but of a different kind that you do during typechecking
masak aye. 16:54
sorear yesterday's changes were a substantial improvement, but not quite substantial enough
pmurias it should be possible to do that vi
sorry
sorear: what changes?
masak I'm thinking in the context of code path analysis.
pmurias dataflow analysis should do that 16:55
sorear pmurias: primitive @(), [], mostly-eager grep and map
pmurias sorear: i thought you were refering to infering multis
masak: with a framework like hoopl it should be even possible to combine them together 16:56
so that my $foo = 123; if 0 {$foo = "hi"}; say $foo + 12 # the correct multi for addition is determined at compile time
masak cool. 16:57
slightly useless example, but I get it.
'if some-real-condition() {$foo = "hi"}' is probably vastly more common. 16:58
masak hm, if the correct multi for addition is determined at compile time, the addition could be *made* (and persisted) at compile-time. 16:59
basically, constant folding across statements.
pmurias yes that would be done
too
what i meant is that it should be possible to combine type-inference, constant folding and dead branch elimination and other stuff together 17:00
masak aye. 17:05
masak I can see how that would be useful/awesome. 17:05
jnthn masak: There's a big difference in the amount of analysis we can do in multi subs vs multi methods. 17:07
masak: With multi subs, in a given lexical scope we know the candidate list statically. 17:08
masak aye.
jnthn With multi-methods, we don't know the candidate list.
pmurias unless we know the type of the invocant 17:08
masak it strongly parallels the case with only subs and only methods.
pmurias: even then...
pmurias and the class is closed
jnthn pmurias: Doesn't help though
masak right, when the class is closed, it's a different story. 17:09
jnthn pmurias: Right, the closed is needed.
masak but that's not the default in Perl 6.
jnthn But you can't close a class individually. So it's left for full program optimization to worry about.
Which is tricky because it's then cross-module optimization so you may have to re-compile modules specifially for the way they're used in a given whole program. 17:10
masak if we can somehow rule out that anon subclassing has taken place, we might also be able to do some more analysis.
jnthn Well, yeah
masak like, if the object was created in the same scope.
(and nothing suspicious happened since then)
jnthn Closedness only gives you more room for analysis because it forbids the anonymous subclassing.
But what if the program relies on it? Well, then it's a trickier issue. :) 17:11
masak right.
you can't have the dynamism cake and eat it, too.
jnthn Decidedly not. It causes terrible indigestion if you try.
masak jnthn, the destroyer of metaphors. :P 17:12
jnthn: you could put that on your business card!
PerlJam wonders what's the point of having a cake but not being able to eat it.
jnthn masak: :P 17:13
masak PerlJam: it produces cake envy in all people nearby.
jnthn masak: I'm still sad I couldn't get "Implementationalist Elder" :P
masak "Elder" is so much better than "Senior" :)
frettled mm
masak "Elder Being of Implementation"
frettled jnthn: Have you tried getting «Elder Horror»? 17:14
masak ooh
jnthn :P
frettled Oh, you young people today. :D
Add «OYYPT» to the calling card. 17:15
masak fails to get that one 17:16
jnthn too :)
Well, need to find myself a job title by next week. :) 17:17
masak too 17:17
jnthn Heh, then I can't be an Elder. You're elder than me. :P 17:18
masak maybe something with "craftman" in it?
jnthn Hmm...yeah 17:19
Want something non-boring. :)
masak troo 17:20
masak well, we've got a week :) 17:20
TimToady Master Brewer 17:21
masak that's jnthn, no doubt.
"Master" is not too bad.
perigrin I should change my title from CEO to "Responsible Party"
masak Master $Something, more exactly.
TimToady Master Cylinder 17:22
perigrin Master Mold
sorear no, that's pmurias 17:22
perigrin Master of Puppets?
Pastor of Muppets? 17:23
jnthn
.oO( I wonder if I can find a pun job title... )
TimToady that would be me
perigrin TimToady: good point.
perigrin jnthn: "Remover of Vowels" ? 17:24
jnthn :)
Why'd y sggst tht? 17:25
TimToady Vwlrmvr
masak Hh
sbp otherwise known as the Hebraist
perigrin though then people might confuse you for a vowel mover ... and that migth imply the job was crappy
masak sbp: I think you've got causation turned around there :)
TimToady if you have the right font you can see the little chickenscratches around "jnthn"
masak perigrin: people might not want to think about Vowel Movements when they see you :) 17:26
perigrin masak: I can't help that though ... I've got such an impact upon people ... viscerally.
TimToady Phase Shifter 17:27
perigrin Metaclass Janitor 17:27
TimToady +1
perigrin "Got a MOP and a set of Hash Buckets" 17:28
jnthn :D
PerlJam perigrin: you're in good form today :)
masak :) 17:29
perigrin PerlJam: have pun, will travel.
PerlJam Does that make #perl6 a punnery? 17:30
masak more like a punastery, in that case.
PerlJam and jnthn would be Priest of Pun (but maybe that applies more to TimToady than jnthn :)
jnthn No, no, he'd be the Puntiff.
PerlJam Our very own Pun See 17:31
perigrin jnthn: that would make you the BishMOP
if this were lisp you'd get to be an Arc BishMOP
PerlJam He's certainly a punter in any case 17:32
perigrin but I think Paul Grahm got there first
TimToady he's just a pundit
diakopter with spunk 17:33
PerlJam jnthn: why do you need a title? 17:35
perigrin he already has a job?
er jot?
[Coke] I assume business cards are being printed! 17:36
jnthn PerlJam: Getting some business cards done. Turns out they should have a job title on or something. ;)
[Coke] . o O (Nun of the Above)
PerlJam jnthn: can you get hexagonal business cards? 17:37
jnthn ...
PerlJam jnthn: or just put a blank line on the card for you to write-in your title du jour.
diakopter suggests mobius strips
jnthn PerlJam: True. :) 17:38
TimToady someone should have a job title of IR Clogger
moritz_++ I suppose
PerlJam waits for a plumbing pun 17:39
diakopter if IR meant intermediate repr, then Perl 6 could be... er
PerlJam for some reason this whole conversation has me thinking of "An Exaltation of Larks" 17:41
perigrin PerlJam: possibly because some of us are just having a lark. 17:43
diakopter PerlJam: it's not a coincidence; we're all Pointers of Veneries, apparently
diakopter heh. meta-punning on a term about punning on pairs of terms with multiple meanings. 17:45
mica_ella_tuazon k 17:47
bored me.....
-----;{@ 17:48
sorear niecza: say 2 + 2 17:52
p6eval niecza v1-149-g56c80b0: OUTPUT«4␤»
sorear looks like a clean rebuild is all it took 17:53
diakopter sorear: !! wow. 17:56
diakopter goes to commit the evalbot.pl changes 17:57
pmichaud tadzik: rakudo release built fine and passed all spectests on my system tadzik++ 18:02
dalek albot: cf1752c | (Perl 6 Evalbot)++ | / (2 files):
fixes for niecza rebuild & p6eval
 18:03
albot: 06afa15 | (Perl 6 Evalbot)++ | /:
Merge branch 'master' of github.com:perl6/evalbot

Conflicts: evalbot.pl
masak mica_ella_tuazon: I'm sorry to hear you are bored. maybe a bit of Perl 6 will cheer you up? 18:06
Cyrus It really does work. Even for other things like cuts and scrapes. Scraped knee? Rub some Perl 6 on it and it feels better. :) 18:07
mica_ella_tuazon :) 18:09
a better cheer from u...:)
masak rakudo: role Times[$type] { method Str { "$type times" } }; say "mica_ella_tuazon, may you live in {Times["interesting"].new}!"
p6eval rakudo 388eed: OUTPUT«mica_ella_tuazon, may you live in interesting times!␤»
masak there, how about that?
masak hides
mica_ella_tuazon tnx.... 18:10
i hav no reason to get bored.. 18:11
tadzik is back
mica_ella_tuazon may be im to young to this scripting thing..
mica_ella_tuazon hahaha... 18:11
but tnx for the cheer..
masak it was sincere. 18:12
PerlJam that's one thing #perl6 probably has in excess as compared with most other IRC channels: sincerity. 18:13
pmichaud and sometimes we're even sincerely humorous. 18:14
TimToady mica_ella_tuazon: no one is too *young* for scripting; if Camelia scares you, it means you're too *old* for scripting 18:32
masak :)
PerlJam I have a doodle of Camelia with red eyes and fangs that looks pretty scary. 18:33
TimToady and quitefantatic did: wall.org/~larry/camelia-angry.pdf 18:35
*quietfanatic
TimToady rakudo: say 1i.sin 19:01
p6eval rakudo 388eed: OUTPUT«0 + 1.1752011936438i␤»
TimToady rakudo: say i.sin
p6eval rakudo 388eed: OUTPUT«0 + 1.1752011936438i␤» 19:02
masak .oO( imaginary sin is a complex thing... )
PerlJam rakudo: say i.virtue 19:03
p6eval rakudo 388eed: OUTPUT«Method 'virtue' not found for invocant of class 'Complex'␤ in main program body at line 22:/tmp/JrJI8Y4Spt␤»
TimToady we need a .robot method 19:04
me wonders if roman numerals are imaginary
PerlJam you'll invoke the law of unintended consequences for sure. 19:05
TimToady I prefer unindented consequences 19:07
masak I like inundated consequences. 19:08
TimToady
.oO(flooding algorithms)
frettled Using the waterfall method? 19:16
masak .oO( they mentioned logging, but never this... ) 19:17
PerlJam masak: you need to listen to the timber of their voices 19:19
masak >< 19:21
[Coke] PerlJam: Do you really think that wood help?
rindolf Hi all. 19:23
PerlJam [Coke]: only if he carved enough. 19:30
masak hola, rindolf.
rindolf masak: what's up? 19:34
colomon /me wishes he could write 1, 1.5 ... 3 that easily in C++... 19:35
masak rindolf: pancakes! \o/
rindolf masak: nice.
masak yep. life's good.
moritz_ hello zebras! 20:02
tadzik hello moritz_!
masak hello koala. 20:05
rindolf Hi moritz_
moritz_ grunts
being a father can be quite exhausting, even if the small one(s) aren't at home 20:06
[Coke] aye.
PerlJam Usually for the father, it's not the little one that's the source of the exhaustion ;)
masak tries to compute whether that counts as sexism by Swedish norms 20:09
moritz_ if you call somebody with an insulting but true name, is that an insult? 20:10
tadzik it's just impolite then I thik 20:13
but some would say that's disputable
Tene moritz_: If you call somebody with an insulting but false name, is that an insult? To me, it's just nonsense. 20:14
TimToady ooh, we now have 2**8 - 1 Perl 6 entries on rosettacode 20:15
Tene Although there's also the question of whether the potential-insulter believes it or not.
TimToady I like the French approach of getting rid of all the guesswork by just saying "I insult you." 20:24
takadonet the french Canadian way would to say "I insult you, but I'm sorry" 20:25
masak takadonet: are they?
takadonet masak: yes :)
tadzik that reminded me of the joke with 3 cowboys entering the bar
TimToady wouldn't that have to be a saloon? 20:26
tadzik oh, probably
masak has a feeling we didn't get to the punchline yet 20:28
TimToady they were really 3 cowgirls walking into a salon
tadzik oh well, it's not That good
masak tadzik: hey! you can't do this to hundreds of people! :) 20:29
Su-Shee aren't the girls usally already waiting _in_ the salon?
TimToady not cowgirls
masak they're waiting on cows. 20:29
Su-Shee in germany, you insult girls by calling them cows. 20:30
tadzik so the first cowboys gets into the bar, spins his fellow Colt on his finger, shots the middle of the dart target and says "Hey, I'm Lucky Luke!". Then the second gets into the bar, spins a Colt on his finger, shots the middle of the bottle on the bar and says "Hey, I'm Billy Kid!"
TimToady fershure there's no waiters in a saloon
TimToady waits for it
tadzik then the last one enters the bar, spins his fellow Colt, then shots the bartender in the middle of the head, and says "Hey! I'm sorry!" 20:31
tadzik (toldya it's not so good) 20:31
Su-Shee well back to the cows then. ;) 20:32
masak tadzik: it was clearly relevant :) 20:32
tadzik it now reminded me of the joke with Inf mathematicians walking into the bar... ah 20:33
masak crowded place.
TimToady is that the bar at the Hilbert Hotel 20:34
tadzik so the infinite number of mathematicians walk into the bar, the first orders a beer, the second orders half a beer, the third orders quater and so on... The bartender says "you're all stupid" and pours two beers
TimToady you sure it wasn't 1.99999... beers?
tadzik hmm 20:35
rakudo: say [+] (1, 0.5, 0.25 ... 0)
p6eval rakudo 388eed: OUTPUT«2␤» 20:36
tadzik see? Rakudo makes a good bartender
or bad, depends on how you look at it
pugs: say [+] (1, 0.5, 0.25 ... 0)
p6eval pugs: OUTPUT«*** ␤ Unexpected "0"␤ expecting operator or ")"␤ at /tmp/w0PhKzTjjD line 1, column 27␤»
Su-Shee you usally look at the looks of a bartender ;)
flussence the bartender should just give them the ingredients for 2 beers and tell them to go integrate them themselves 20:37
tadzik integrating two beers should give many, many beers 20:38
for when you derive beer, you get the amount of beer you can buy from selling the empty bottles 20:39
and a good party is when the second deriv... something is > 0
Su-Shee no tadzik, that's not a good party. ;)
tadzik well, they say so ;)
TimToady that doesn't rule out the jerks 20:40
"Your third derivative is > 0!!!" 20:41
moritz_ at least 0!!! == 0!
TimToady that's a 3-dimensional factorial 20:43
masak brane hertz
TimToady that's because you've been cowed 20:45
moritz_ Mu 20:46
masak ow.
TimToady job title for a cow that doesn't give milk:
Milk Dud
Udder Failure
colomon is very impressed that [+] (1, 0.5, 0.25 ... 0) actually worked 20:48
TimToady relies on FP underflow, I suspect 20:48
masak rakudo: say (1, 0.5, 0.25 ... 0).elems 20:49
TimToady and perhaps printf rounding
p6eval rakudo 388eed: OUTPUT«1076␤»
masak :)
takadonet rakudo: say (1, 0.5, 0.25 ... 0)
p6eval rakudo 388eed:
..OUTPUT«10.50.250.1250.06250.031250.0156250.00781250.003906250.0019531250.00097656250.000488281250.0002441406250.00012207031256.103515625e-053.0517578125e-051.52587890625e-057.62939453125e-063.814697265625e-061.9073486328125e-069.5367431640625e-074.76837158203125e-072.38418579101562e-071.1…
TimToady I'd rather have 1075, thank you 20:50
1076 is phosgene
oddly, milk doesn't have a UN number 20:52
just before Christmas I passed a truck full of 1970: Krypton, refrigerated liquid (cryogenic liquid) 20:54
[particle] it powers his sleigh. 20:55
[particle] you know who. 20:55
TimToady Superman?
Su-Shee now I understand the red cape. 20:56
TimToady they're really the same person, y'see 20:57
moritz_ thinks that Krypton is execeptionally bad at powering anything
only topped by He, Ne and Ar
TimToady the whole kryptonite thing is a ruse to get his enemies to mail him more krypton
TimToady and the movies that show Supes flying by himself are really to divert attention from the reindeer droppings 20:58
[particle] "pay no attention to the reindeer behind the cape."
tadzik did the announcement email hit the perl6-compiler ML? 21:00
TimToady wonders what it means that Kr is bottomed by Xe, Rn, and Uuo...
jnthn tadzik: I sore it 21:01
er
TimToady also wonders what they'll do with all the Xe when they stop filling light bulbs with it
jnthn *saw
moritz_ TimToady: XePtF6 -- what else could you make of it? :-) 21:02
diakopter well that's sad. 21:03
vmware esxi comes with python on its host but not perl 21:04
colomon rakudo: say [+] (1, 0.5, 0.25 ... 0)[^40]
diakopter but .. a lot of their remote api stuff is in perl
p6eval rakudo 388eed: OUTPUT«1.99999999999818␤»
colomon rakudo: say [+] (1, 0.5, 0.25 ... 0)[^45]
p6eval rakudo 388eed: OUTPUT«1.99999999999994␤» 21:05
colomon rakudo: say [+] (1, 0.5, 0.25 ... Rat)
p6eval rakudo 388eed: OUTPUT«1.5␤»
colomon rakudo: say [+] (1, 0.5, 0.25 ... Num)
moritz_ huh?
p6eval rakudo 388eed: OUTPUT«1.99999999953434␤»
moritz_ ah
colomon first is Int, second one is Rat, so 1 + .5 = 1.5 21:06
masak ooh! '... Num' -- that's... emergent. 21:07
TimToady we noticed it a month or two ago, so it's re-emergent 21:08
the ... 0 is also emergent, or convergent, or something
rakudo: say [+] 1, 2, 4 ... Inf; 21:10
p6eval rakudo 388eed: OUTPUT«Inf␤»
TimToady wow, rakudo is FAST!!
masak ...optimized for infinite loops...
rakudo: say [+] 0, 0, 0 ... Inf; 21:11
p6eval rakudo 388eed: OUTPUT«(timeout)»
masak heh.
TimToady rakudo: say [+] 0, 0, 0 ... 0;
diakopter masak solved the halting problem
p6eval rakudo 388eed: OUTPUT«0␤»
masak diakopter: I knew I would acheive greatness some day. :P 21:12
colomon rakudo: say ~(1, 2, 4 ... Inf);
p6eval rakudo 388eed: 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 137438953472
..274877906944…
masak or is that "achieve"? I keep mixing those up.
colomon that sequence ends 4.49423283715579e+307 8.98846567431158e+307 Inf on my MBP.
TimToady didn't you know? that's Inf/2 21:13
masak talk about taking a shortcut.
colomon but that one will not work once we have arbitrary precision Ints, alas.
diakopter quick, write a test
TimToady better, write a quick test 21:14
masak or at least one that doesn't loop forever...
TimToady 1, 10**1000, 10**2000 ... Inf will fail much faster, I suspect 21:22
TimToady now imagines a machine that uses the cloud for swapspace... 21:24
colomon rakudo: say ~(1, 10**1000, 10**2000 ... Inf)
p6eval rakudo 388eed: OUTPUT«1 Inf␤»
colomon that's about what I expected there.
diakopter cloud 9
TimToady that's in outer space
masak from outer swap space.
TimToady no, it's *from* outer space
masak++ beat me to it 21:25
diakopter thought it was from outta space
TimToady no, that's where it's to
diakopter OUTATIME 21:26
TimToady of course, the compiler should just realize what's going on there and turn it into 10 X** (0,1000,2000...Inf)
masak TimToady: that'll make it... fail slower...? 21:27
[particle] depends on hardware. how fast is **?
TimToady rakudo: say ~( 10 X** (0,1000,2000...Inf) ) 21:28
p6eval rakudo 388eed: OUTPUT«(timeout)»
TimToady rakudo: .say for 10 X** (0,1000,2000...Inf)
colomon rakudo: say ~( (10, 10 ... *) Z** (0,1000,2000...Inf) ) 21:29
p6eval rakudo 388eed: 21:29
..OUTPUT«(timeout)␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf␤Inf􏿽xE2
rakudo 388eed: OUTPUT«(timeout)»
colomon oh, much slower fail 21:29
I'm actually impressed it got that far 21:30
TimToady mine at least gave some answers
colomon oh, I missed that you'd gotten one in there as well.
TimToady pity the (timeout) covered the first one
colomon huh. are you sure that's yours there with the printout? There's a strange lack of newlines... 21:31
TimToady why does that happen, anyway?
colomon why does what happen?
TimToady mine has newlines
why does the (timeout) block the early answer
diakopter sometimes it does; sometimes it doesn't 21:32
TimToady rakudo: .say for 1..10000000
p6eval rakudo 388eed:
..OUTPUT«(timeout)␤6␤7␤8␤9␤10␤11␤12␤13␤14␤15␤16␤17␤18␤19␤20␤21␤22␤23␤24␤25␤26␤27␤28␤29␤30␤31␤32␤33␤34␤35␤36␤37␤38␤39␤40␤41␤42␤43␤44␤45␤46␤47␤48␤49␤50␤51␤52␤53␤54␤55␤56␤57␤58␤59␤60␤61␤62␤63␤64␤65␤66␤67␤68␤69␤70␤71␤72␤73␤74␤75␤76␤77␤78␤79␤80␤81␤82␤83␤84␤85␤86␤87␤
TimToady like that 21:32
two different file pointers into the same disk file, I guess 21:33
colomon very weird.
TimToady wonders if stderr should be opened for append 21:34
really, should be sent to a different file and combined by the bot 21:35
TimToady rakudo: .say for 1..10; note "oops" 21:35
p6eval rakudo 388eed: OUTPUT«1␤2␤3␤4␤5␤6␤7␤8␤9␤10␤oops␤»
TimToady strange 21:36
rakudo: .say for 1..10; warn "oops"
p6eval rakudo 388eed: OUTPUT«1␤2␤3␤4␤5␤6␤7␤8␤9␤10␤oops in <anon> at line 22:/tmp/PSWkg8UR_B␤»
TimToady maybe the (timeout) is supplied by the bot
if so, I think it's a bug that it overrides the output 21:37
it should append the message on the end instead
and color it, or make it flash, or something... 21:38
diakopter the (timeout) is supplied by the bot 21:49
sometimes it overrides any output; sometimes it doesn't
TimToady whether it overrides is probably just buffering 21:54
dalek ast: 8447ca8 | (Kodi Arfer)++ | S03-operators/inplace.t:
[inplace.t] Added tests for RT #70676.
 22:04
masak 'night, #perl6 22:27
dalek p-rx/nom: 4d75922 | jonathan++ | src/metamodel/knowhow_bootstrapper.c:
The find_method op throws method not found exceptions, and the API is for find_method v-table not to; make the KnowHOW.^find_method play that way.
 22:42
p-rx/nom: 26ac097 | jonathan++ | src/metamodel/reprs/P6opaque.c:
Get P6opaque most of the way to consuming the storage spec provided by a REPR. In theory, this means it can allocate storage for natively typed attributes. In practice, there's no way to test this just yet.
lichtkind_ tschuldige hatte noch revert zu checken 22:59
sorry wrong win
jnthn sleep & 23:00
lichtkind_ good night
snarkyboojum full names in the rakudo release announcements now eh? :D 23:43
guten moaning all
lichtkind_ snarkyboojum: haha 23:44
snarkyboojum lichtkind_: :)
snarkyboojum I don't have commit access to the rakudo repo, but I'd be happy to do a release if people are looking for future release boffins :) 23:45
though I suppose commit access is a prerequisite :D
lichtkind_ i think so too :) 23:46
[Coke] supposes he should volunteer for another one. 23:47
maybe after dayjob stops being nightjob also.
snarkyboojum is between dayjobs atm 23:49

Powered by Cro and the Raku® Programming Language.

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

Thank you!