»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! | Rakudo Star Released!
Set by diakopter on 6 September 2010.
00:05 Italian_Plumber joined 00:08 xinming left 00:25 Drizzt_ joined, risou joined 00:29 Drizzt_ left 00:31 justatheory joined 00:34 TomDLux left 00:54 QinGW joined 00:55 justatheory left 00:58 QinGW2 joined, QinGW left 01:00 lamstyle joined 01:07 risou left 01:25 masonkramer left 01:26 masonkramer joined 01:27 masonkramer left, aloha left 01:30 lamstyle left, Italian_Plumber left 01:31 aloha joined 01:33 lamstyle joined 01:38 ggoebel left 02:26 bluescreen left, justatheory joined 02:33 felipe__ joined 02:35 felipe__ left, QinGW2 left, leprevost joined 02:36 QinGW joined 02:38 QinGW left 02:47 am0c^ joined 02:48 Raynes joined 03:05 Entonian left, jedai_ left, jedai joined 03:06 Woody2143 left 03:07 Woody2143 joined 03:19 eternaleye joined, QinGW joined 03:22 leprevost left 03:23 eternaleye left, eternaleye joined 03:25 QinGW1 joined 03:27 ruoso left, QinGW left 03:36 QinGW1 left 03:37 meppl left, QinGW joined 03:39 meppl joined 03:42 QinGW left 03:43 rgrau_ left 03:44 QinGW joined 03:49 QinGW1 joined 03:53 QinGW left 03:54 QinGW1 left, QinGW joined 03:58 QinGW1 joined 03:59 QinGW left 04:10 satyavvd joined 04:15 QinGW1 left, sftp joined 04:22 am0c^ left 04:23 am0c^ joined
sorear good * #perl6 04:24
diakopter good 04:27
plobsing Could not find sub &good 04:28
sorear diakopter: another part of why I want to hack apart STD.pm6 is that it doesn't have a fully reusable frontend 04:29
STD.pm6 is a static checker, not a parser; it's possible to lift parses out of it, but you're still paying for the cost (and occasional false positive errors) of the static checker part 04:30
04:32 atomicstack joined 05:01 justatheory left 05:12 justatheory joined, justatheory left 05:13 justatheory joined 05:20 fpants joined 05:27 mberends left 05:30 QinGW joined
sorear std: class q { }; say(q) 05:32
p6eval std 237d266: OUTPUT«===SORRY!===␤Use of a closing delimiter for an opener is reserved at /tmp/izQhovQXzl line 1:␤------> class q { }; say(q⏏)␤ expecting any of:␤ argument list␤ prefix or term␤Parse failed␤FAILED 00:01 115m␤»
sorear std: class q { }; say( q ) 05:33
p6eval std 237d266: OUTPUT«===SORRY!===␤Use of a closing delimiter for an opener is reserved at /tmp/E9dsxlBMmt line 1:␤------> class q { }; say( q ⏏)␤Parse failed␤FAILED 00:01 115m␤»
05:37 QinGW left 05:38 kaare joined, fpants left, kaare is now known as Guest81868 05:40 QinGW joined 05:41 pmichaud left 05:43 QinGW1 joined, Juerd left 05:47 QinGW left 05:51 hatseflats left 05:57 pmichaud joined 06:00 hatseflats joined
sorear finds a memory leak in STD.pmc 06:03
oh, misread 06:05
06:07 uniejo joined 06:22 justatheory left 06:23 QinGW1 left, QinGW1 joined
sorear grumbles at EXPR using a *lot* of internal CursorBase calls 06:25
06:25 pythonian4000 is now known as pythonian4000afk 06:33 envi joined 06:38 am0c^ left 06:39 sjn left 06:45 Juerd joined 06:49 sjn joined 06:51 AyStupidMuddaFuk joined
AyStupidMuddaFuk r 06:51
06:51 AyStupidMuddaFuk left
TiMBuS so my perl5 irc bot has gone up from about 130mb ram usage, to ~430mb. 06:55
perl6* 06:56
had parrot changed its gc
moritz_ yes 06:57
phenny moritz_: 11 Oct 20:55Z <PhatEddy> tell moritz_ his missing rakudo_test_run.tar.gz file could be caused by not having TAP::Harness::Archive installed. Without the module installed the tests are still run but you get the LTA error message.
moritz_ phenny: tell PhatEddy thanks, I had reinstalled my operating system, so that TAP::Harness::Archive was missing indeed. Will try again now. Thanks for the hint! 06:58
phenny moritz_: I'll pass that on when PhatEddy is around.
TiMBuS moritz_, oh i see then, that explains it 07:02
in that case, how is the new gc overall? does it still segfault like the old one?
07:02 azawawi joined
azawawi hi 07:02
moritz_: ping :)
moritz_ azawawi: pong
azawawi moritz_: i noticed while reading backlogs what you said about ~azawawi... will fix it soon :) 07:03
moritz_ azawawi: it was about the syntax hilighted test snippets in the specs
azawawi: they seemed to be unmaintained, so I removed the links
azawawi moritz_: they depended on svn and now they need to work with git... 07:04
moritz_ I also replaced the inline test snippets to links to the github repo - makes searching in the HTML much faster
azawawi: exactly what I meant with "unmaintained" 07:05
azawawi moritz_: yup i know :)
azawawi fixing... :)
moritz_ azawawi: feel free to add links back in, but please make them ordinary links, not inline 07:06
azawawi: (you have again commit access to the "mu" repo, which used to be the pugs repo) 07:07
afk
azawawi moritz_: thx
07:12 azawawi left
sorear azawawi needs to drop by for more than 10 minutes at a time 07:19
I find it *quite* annoying
std: rx:moo { }
p6eval std 237d266: OUTPUT«===SORRY!===␤Unrecognized adverb :moo(1) at /tmp/ARkHYCRYUV line 1:␤------> rx⏏:moo { }␤Null pattern not allowed at /tmp/ARkHYCRYUV line 1:␤------> rx:moo { ⏏}␤ expecting quantifier␤Parse failed␤FAILED 00:01 115m␤» 07:20
sorear std: token:moo { }
p6eval std 237d266: OUTPUT«===SORRY!===␤Undeclared routine:␤ 'token:moo' used at line 1␤Check failed␤FAILED 00:01 115m␤»
07:26 azawawi joined
azawawi sorear: sorry :) 07:26
sorear: but im using a browser for IRC... 07:27
sorear Oh, you're one of those people who can hear when not present.
Anyway. Is S:H:P6 broken again yet?
azawawi sorear: didnt try it recently 07:29
sorear: will check
sorear: it installs correctly 07:32
sorear: I backlog a lot :) 07:33
moritz_ sorear: I just /ignore joins and leaves. Works out quite well 07:34
07:37 fwa joined 07:38 daxim joined 07:42 fwa left
moritz_ is there a #cpan6sketch today? 07:42
07:50 Mowah joined 07:58 sukria joined 08:03 am0c^ joined 08:15 azawawi left 08:17 [particle] left 08:18 xinming joined 08:19 araujo joined, araujo left, araujo joined 08:22 sunnavy_ left 08:23 icwiener joined, Raynes left 08:25 sunnavy joined 08:26 tadzik joined
sorear I have gone over STD.pm6 and identified 63 items where Niecza is not currently up to the task 08:29
dalek ecza/master: 8f7545b | sorear++ | / (6 files):
Start review/port of STD.pm6
ecza/master: 56625fd | sorear++ | v6/ (2 files):
Finish first scan
sorear (and fixed or commented out a few places where STD gratuituosly used advanced features like sprintf)
08:40 pythonian4000afk is now known as pythonian4000 08:49 [particle] joined 08:50 tadzik left 08:51 Guest81868 is now known as kaare_ 09:00 cjk101010 joined
moritz_ sorear: are the # end role comments significant? 09:11
09:33 adv_ joined 09:36 icwiener left 09:38 PhatEddy joined 09:40 timbunce joined 09:41 masak joined
masak oh hai 09:41
moritz_ oh hai masak 09:42
are you back in .se?
masak aye. 09:43
just woke up from a good night's sleep.
moritz_ envies masak for long/late sleeping
masak if it's any consolation, I would have liked to sleep a bit shorter :) 09:44
09:44 jrockway joined, am0c^ left
moritz_ that said, the last 4 night have been quite peacefully, compared to those of the previous weeks 09:44
:-)
masak relative measurement is relative :P 09:46
moritz_ indeed :-)
but since humans are mostly relative, that's OK
masak oh, very OK. 09:47
moritz_ just read an announcement for a workshop: Small Talk in English: Do´s and Dont´s 09:48
one Don't: using wrong apostrophe in workshop titles :-)
09:49 QinGW joined
PhatEddy moritz_: spectest_smolder runs t/harness with --archive option. Would be easy to warn if harness sees --archive option but no TAP Archive module. Interested in patch? 09:50
phenny PhatEddy: 06:58Z <moritz_> tell PhatEddy thanks, I had reinstalled my operating system, so that TAP::Harness::Archive was missing indeed. Will try again now. Thanks for the hint!
moritz_ PhatEddy: yes; with s/warn/die/
warnings are too easy to overlook 09:51
and there's really no point in running _smolder if the submission won't work
PhatEddy ok - will try in next day or two ... 09:52
09:52 QinGW3 joined
moritz_ btw if you have a patch that affects multiple files, it's easier for the reviewer if you submit it as a single patch 09:53
09:53 QinGW1 left 09:54 QinGW left 09:57 PhatEddy left
masak unless it's actually several unrelated changed. 10:08
s/changed/changes/
moritz_ then it belongs in separate tickets. 10:10
(or commits, in the case of pull requests)
masak *nod*
10:13 QinGW3 left 10:14 araujo left 10:24 Axius joined 10:25 araujo joined, envi left 10:35 pythonian4000 is now known as pythonian4000afk
moritz_ rakudo: say ' - - '.trim.perl 10:35
p6eval rakudo 5d23e3: OUTPUT«"- -"␤» 10:36
10:51 Alias joined
jnthn o/ 10:51
masak: Glad you made it back safely too :-) 10:54
10:56 Axius left 10:57 Trashlord left 11:07 araujo left 11:13 araujo joined 11:20 araujo left, araujo joined 11:27 araujo left 11:28 araujo joined 11:31 araujo left
masak jnthn: I had a pleasant day in Utrecht yesterday. it also has canals (grachten). 11:34
jnthn masak: nice :-) 11:35
colomon back home we had canal (singular). I canoed and cross-country skied on it when I was younger. 11:37
masak skiing, not skating? 11:38
colomon more skiing than skating. I had a bad tendency to outgrow my skates very quickly.
(which maybe means the skiing was a bit later, because I don't remember outgrowing them so much. phases.) 11:39
jnthn masak: I had currywurst in Hamburg while changing train. :-)
'twas nommy. 11:40
11:40 satyavvd left
masak sounds nice. :) I tried some of the local fries-with-mayonnaise in Utrecht. odd but nice. 11:41
11:41 baux joined
jnthn Fries + mayo is a nice combination :-) 11:43
11:44 rhr left 11:51 rhr joined 11:56 araujo joined, araujo left, araujo joined 12:04 takadonet joined
takadonet morning all 12:05
masak takadonet: \o 12:07
nymacro lo
takadonet masak: little surprised that you have not commented on : www.reddit.com/r/programming/commen..._by_regex/ 12:09
masak takadonet: been traveling yesterday. :)
moritz_ takadonet++ # sending me ~6k visitors from reddit 12:11
takadonet masak: ic
moritz_: np anytime
was nice to see people discussion the issue rather then ' Perl 6 sucks!' 12:12
masak indeed.
let's have more of that.
rakudo: sub j { :2($^n.fmt("%b").comb.rotate.join) }; say j($_) for 1 .. 20 # Josephus function; see secure.wikimedia.org/wikipedia/en/...us_problem 12:26
p6eval rakudo 5d23e3: OUTPUT«1␤1␤3␤1␤3␤5␤7␤1␤3␤5␤7␤9␤11␤13␤15␤1␤3␤5␤7␤9␤»
masak \o/
12:27 Italian_Plumber joined 12:28 envi joined
moritz_ \o/ 12:28
daxim masak, do the ⌹dyalog creature⌹ next 12:36
masak good idea. 12:37
dalek ecs: 00572dd | masak++ | S06-routines.pod:
[S06] s/tail-recursion/recursion/

Because you can use &?BLOCK calls in a non-tail position (and execution still carries on afterwards), it's really just normal recursion.
Omission found by (Ben Goldberg)++.		 12:38
12:38 meraxes joined
daxim but as a port from the apl, else it doesn't count 12:38
masak here's the original apl, for someone who wants to tackle it: www.dyalog.com/dfnsdws/n_life.htm 12:39
I really like the division into the spaces <plane cylinder torus möbius klein>. 12:40
never thought about it that way before.
I have not yet convinced myself that that's an excaustive list of such spaces, though. 12:41
12:43 ruoso joined
moritz_ waits for a topologic, exact definition of "those spaces" :-) 12:44
krunen ~. 12:45
argh. sorry, frozen ssh :( 12:46
12:46 [particle] left
moritz_ hands krunen a hair drier to thaw his ssh session :-) 12:46
jnthn Winter is coming... ;-) 12:47
12:48 [particle] joined
masak (Stark family motto.) 12:50
moritz_: things that can be described by that square thing they have, putting optional arrows in some direction on each of the sides. 12:51
moritz_: seems to me they don't include the one where the arrows all point clockwise, for example. 12:52
moritz_ old (2003), but still interesting: www.perlmonks.org/?node_id=261631 12:54
"I am scared that someone will write a Python parser as a perl 6 regex." 12:55
masak aren't we all? :P 12:56
moritz_ if you s/regex/grammar/, it's not scary to me at all
masak Perl 6 has been "not yet released" since at least 2003. 12:57
I wonder if it'd have been possible to counteract that meme early, or if no matter which path one chose, Perl 6 would get the "vapor" label... 12:58
moritz_ we could start another next-version-of-a-popular-language and try to find out :-) 12:59
... but only when Perl 6 is "done" :-))
masak oh yeah. only then.
wouldn't want to "detract from" Perl 6. :P 13:00
13:03 ambs joined 13:11 Woody2143 left 13:12 ambs left 13:13 Woody2143 joined 13:19 Patterner left 13:20 bluescreen joined, Psyche^ joined, Psyche^ is now known as Patterner 13:21 Dunearhp left 13:25 Dunearhp joined 13:32 sftp left 13:35 sftp joined 13:36 kaare_ left 13:49 uniejo left 13:53 whatashui joined 13:57 whatashui left, araujo left
shortcircuit A P6 example got tagged incorrect, if someone wants to fix it: rosettacode.org/wiki/Sort_stability#Perl_6 13:58
It looks like the issue is that the example data set isn't meant to be taken as a list of strings, but as a list of records, and the sort is supposed to be applied to the second column.
moritz_ the issue is just .[0] vs .[1] 13:59
easy fix 14:00
14:06 justatheory joined 14:07 spq1 joined 14:17 Trashlord joined 14:18 MayDaniel joined 14:20 Trashlord left 14:21 rgrau_ joined 14:23 araujo joined, araujo left, araujo joined 14:25 slavik left 14:30 mavrc joined 14:31 justatheory left 14:32 Trashlord joined 14:39 Patterner left 14:41 eternaleye left 14:56 Psyche^ joined, Psyche^ is now known as Patterner 14:57 am0c^ joined 15:02 daxim left 15:03 Guest5383 joined 15:06 meteorjay left, cjk101010 left 15:08 Guest5383 left 15:09 burning_aces joined, burning_aces left 15:13 timbunce_ joined 15:15 nymacro left 15:17 timbunce left, timbunce_ is now known as timbunce
moritz_ std: [1, 2]«-»[3, 4] 15:18
p6eval std 263c207: OUTPUT«ok 00:01 117m␤»
moritz_ shouldn't that barf because « starts a postcircumfix?
masak noö why_ 15:21
no; why?
array indexing of hash indexing of array. 15:22
moritz_ rakudo: say [1, 2]«-»[0, 1].perl
p6eval rakudo 5d23e3: OUTPUT«12␤»
moritz_ wrong? 15:23
masak IMO yes.
colomon maybe I'm being dense, but how could it possibly be right?
15:23 slavik joined
masak right. it looks like one of those "even if it were right, it'd be wrong" things. 15:24
moritz_ rakudo: print ([1..3]«-»[1..3]).perl
p6eval rakudo 5d23e3: OUTPUT«[0, 0, 0]»
colomon rakudo: say [1, 2] «-» [0, 1].per
p6eval rakudo 5d23e3: OUTPUT«Method 'per' not found for invocant of class 'Array'␤ in main program body at line 22:/tmp/FZgLeatNgw␤»
colomon rakudo: say [1, 2] «-» [0, 1].perl
p6eval rakudo 5d23e3: OUTPUT«12␤»
colomon rakudo: say [1, 2] «-» [0, 1] 15:25
moritz_ rakudo: say ([1, 2]<<->>[0,1]).perl
p6eval rakudo 5d23e3: OUTPUT«11␤» 15:26
rakudo 5d23e3: OUTPUT«(Any, Any)␤»
moritz_ that looks better
flussence perl6: our multi sub postcircumfix:<« »>(Array $a) { ... }; say ([1,2]«*»[2,5]).perl;
moritz_ maybe « » is just NYI?
p6eval pugs: OUTPUT«*** Cannot cast into Hash: VRef <Array:0x2acb6f227d91>␤ at /tmp/U4BOAX1EB1 line 1, column 59-72␤»
..rakudo 5d23e3: OUTPUT«===SORRY!===␤Null PMC access in get_string()␤»
colomon rakudo: say ([1, 2] «-» [0, 1])
p6eval rakudo 5d23e3: OUTPUT«11␤»
flussence std: our multi sub postcircumfix:<« »>(Array $a) { ... }; say ([1,2]«*»[2,5]).perl;
p6eval std 263c207: OUTPUT«===SORRY!===␤Unable to parse postcircumfix expression at /tmp/GfO3dtDnjL line 1:␤------> fix:<« »>(Array $a) { ... }; say ([1,2]«⏏*»[2,5]).perl;␤Couldn't find final '»'; gave up at /tmp/GfO3dtDnjL line 1:␤------> »>(Array $a) { ... };
..say (…
moritz_ rakudo: say « 1 2 3 » 15:27
p6eval rakudo 5d23e3: OUTPUT«123␤»
flussence rakudo doesn't like that sub declaration for some reason...
moritz_ right, «» as postcircumfix is simply NYI 15:28
15:28 zulon joined
flussence std: say ([1,2]«*»[2,5]).perl; 15:31
p6eval std 263c207: OUTPUT«ok 00:01 119m␤» 15:32
flussence huh, I never realised std allowed custom operators 15:33
15:33 risou joined
pmichaud good morning, #perl6 15:43
takadonet pmichaud: morning
moritz_ \o 15:44
masak pmichaud: morning! 15:46
pmichaud how was paris?
15:48 zulon left
moritz_ github has a scary bug 15:54
I go to one of my repos
delete it
successfully
colomon pmichaud: o/
moritz_: oh noez 15:55
moritz_ and then it redirects me to the perl 6 organization account
masak o.O
moritz_ and says "your repo $Blah has been deleted successfully"
so I think I removed perl6/$Blah, even though I just deleted moritz/$Blah
masak please submit githubbug.
moritz_ (which gave me quite a heart attack when I thought I had deleted perl6/modules.perl6.org, not moritz/modules.perl6.org) 15:57
masak :) 16:01
16:03 cdarroch joined, cdarroch left, cdarroch joined 16:05 araujo left
moritz_ uhm, no feedback if that message to the github team has been sent successfully :( 16:05
it just shows me the blank form again 16:06
pmichaud another githubbug? ;-)
moritz_ it's also unclear from their UI if I should use 'contact' or 'support' 16:07
16:08 mberends joined
masak mberends! \o/ 16:09
mberends yo!
slavik will there be any sort of spec for what modules constitute "standard" distribution?
pmichaud there is no "standard distribution". 16:11
(so no, there's no such spec)
mberends: o/ 16:12
16:14 Schwern joined
mberends oh hai pmichaud! 16:14
16:15 envi left
pmichaud omg I posted a response to p6l 16:16
moritz_ OH NOEZ! 16:17
masak that reminds me...
masak starts writing a summary in the attr introspection thread 16:18
pmichaud: *gasp* -- you used "real-world use cases" on p6l!
(sorry for jokingly linguifying an important issue) :) 16:19
16:19 envi joined
masak pmichaud++ # nice reply 16:19
pmichaud I recognize that talking about real implementation details is against p6l standards.... but sometimes a few mugs need to be broken 16:20
masak can't make a language without breaking a few mugs...
pmichaud well, you *can*.... but you might not like the language you end up with :)
masak I'd rather use Perl 6 than Consensus 6, yes. 16:21
moritz_ :-)
pmichaud did we agree on "Consensus 6" as a name? I thought we had decided on "Group Think 6".
moritz_ wants Plan 6
pmichaud or even "Bikeshed 110"
masak no, I distinctly remember it as "Committee 6".
[particle] isn't C for Cabal? 16:22
masak ok, so we've all agreed about the need to agree. now we just need to agree what to agree on.
pmichaud [particle]: o/
[particle] howdy, pmichaud! 16:23
[particle] has been buried under a $work-shaped rock
i finally freed one hand to type
pmichaud I've been buried under a <cold work family> shaped rock 16:24
[particle] i just don't understand how perl 6 will torture the implementors properly if they're the ones defining the spec
(re: threads)
pmichaud [particle]: counter-example: lists 16:25
[particle] :)
moritz_ now here is pmichaud again, arguing with real-world examples 16:27
how can we successfully build a vapor language if we start to look at the real world?
pmichaud the real world gives us lots of vapor to work with :) 16:30
16:42 risou_ joined 16:45 risou left 16:51 wamba joined
masak p6l email sent. 16:52
16:55 wamba left, wamba joined
colomon masak: believe you got former and latter backwards in the .perl section 16:56
masak colomon: indeed -- correcting. 16:57
colomon++
colomon masak++ # for being brave enough to post to p6l
pmichaud I like the analogy very much. My comment would be that unlike MONKEY_TYPING and .HOW, ".get_value" doesn't visually appear as though "you're working outside the rules." 16:59
s/outside/above/
masak indeed.
and that was TheDamian's point too, IIRC.
you can pick up an Attribute meta-object and run with it to the other end of your program. 17:00
I must confess I'm not too worried about that, though.
because *some* part of your program is in the tower, above the rules.
pmichaud well, you don't necessarily know that. It could be a module. 17:01
masak and if that part decides to let Attribute objects out of the tower, then you do indeed have two problems. :P
moritz_ masak++ # good summary
17:01 alester left
masak thought you'd like it. :) 17:01
nom &
17:01 masak left, alester joined
pmichaud lunch here, also. 17:01
moritz_ Damian mentioned he disliked .get_value, because the call that involved the meta class could be in a different scope 17:02
and an Attribute is just being passed around
I don't buy that argument
because if you use get_value, your code explicitly works with an Attribute, so it knows about this breaking-of-the-rules 17:03
pmichaud I'm also not too keen on "get_" and "set_".... in general that's not very perlish. (Perhaps the meta-ness of attributes requires breaking that pattern, though)
17:03 bluescreen left
moritz_ (he is right in that it's not visually obvious though) 17:03
colomon should they be upper case?
moritz_ pmichaud: it should probably be bind_ instead of set_, right
colomon I mean, it would look obnoxious, but that might be the point....
pmichaud fwiw, we already have things like FETCH and STORE.... any reason not to use those?
moritz_ pmichaud: they act on containers 17:04
17:04 MayDaniel left
pmichaud attributes aren't containers? 17:04
oh, nm
moritz_ if you have an Attribute object, it's not a container.
pmichaud right
it's a way of getting at a container... 17:05
moritz_ FETCH and STORE would work for me too, even if I'm not entirely convinced
pmichaud it might be interesting if Attribute were actually some sort of method (more) 17:06
then one could do something like $obj.$attr = ...
moritz_ interesting approach
pmichaud instead of doing $attr.set_value($obj, $value)
17:08 am0c^ left
pmichaud that feels a lot more perlish to me, somehow. 17:08
especially since "has $.abc" tends to create a "method abc" that is used to access the $!abc container. 17:09
so perhaps an Attribute can really act like one of those autogenerated methods
or the autogenerated methods are really just instances of Attribute 17:10
lunch here
17:13 contempt joined 17:18 bluescreen joined 17:22 pyrimidine joined 17:23 plainhao joined 17:34 jferrero joined 17:35 araujo joined 17:39 slavik left 17:44 risou joined 17:45 risou_ left 17:46 mberends left 17:57 envi left 18:01 slavik joined 18:05 xinming left, xinming joined 18:06 Entonian joined 18:11 baux left 18:13 contempt left 18:18 timbunce left 18:24 timbunce joined 18:30 icwiener joined 18:31 molaf joined
sorear moritz_: # end role and company are to help out the "parser" in gimme5. They're no longer relevant in the viv era. 18:39
Util How can I return 2 distinct arrays from a sub? 18:44
18:44 risou_ joined
moritz_ return [1, 2], [3, 4] 18:44
Util And if they are already in arrays, is this the best syntax? return [@q], [@r]; 18:46
18:46 risou left
moritz_ if you want avoid an extra copy, return @q.item, @r.item 18:47
or $(...) isntead of .item 18:48
Util moritz_: thanks 18:51
dalek ecza/master: a6766cd | sorear++ | / (5 files):
Starts on the TODO...

Add &chars, &substr, &join, &item, &not, &defined, &reverse, &note,
  &push, &pop, &shift, &unshift, Mu.item, Bool.Numeric; work around
absense of //=, ||=, named arguments to Mu.new.		 18:56
18:57 masak joined, meppl left
masak ahoj! 18:57
#phasers in -2? 19:02
moritz_, jnthn, pmichaud? 19:04
19:05 molaf left, adv_ left
pmichaud present 19:05
19:07 Raynes joined 19:12 cliu joined 19:13 cliu left
moritz_ raises hand 19:13
19:14 timbunce left, timbunce joined 19:15 plainhao left 19:17 coldboot joined 19:21 araujo left 19:22 edon joined
dalek ecza/master: 4cf4ea5 | sorear++ | v6/ (2 files):
Work around $_ scoping bugs
 19:23
ecza/master: dab9893 | sorear++ | v6/ (2 files):
Work around lack of given/when
19:33 Entonian left 19:44 risou joined 19:46 mberends joined
sorear seen tylercurtis 19:46
aloha tylercurtis was last seen in #perl6 4 days 6 hours ago joining the channel.
19:47 risou_ left 19:48 timbunce left 19:55 timbunce joined 20:06 Italian_Plumber left 20:09 andrei_alexandre joined
andrei_alexandre hello everyone 20:10
I have a question about Perl6's handling of static type annotations.
I've used google for a while, no avail
Question is, how does Perl handle pass by reference for values that are statically anotated with e.g. int? 20:11
masak andrei_alexandre: hi!
PerlJam andrei_alexandre: #perl6 >> google for Perl 6 questions :)
masak andrei_alexandre: hm. good question. 20:12
andrei_alexandre I am googling but human touch is always best :D
masak something like this? my int $a = 42; foo($a); sub foo(int $a is rw) { ... }
20:13 molaf joined
andrei_alexandre no, foo() would be old-style, i.e. it won't specify int on its parameter 20:13
it would take a reference nonetheless
masak like this? my int $a = 42; foo($a); sub foo($a is rw) { ... } 20:14
andrei_alexandre in brief what I'm asking is whether "my int $a;" can actually use a plain 64-bit value for $a instead of a variant 20:15
yah
@masak: yah
20:15 wamba left
sorear andrei_alexandre: I don't think it's legal to call foo($a) like that 20:15
PerlJam sorear: why not? 20:16
andrei_alexandre sorear: so that would be a compile-time error I reckon
PerlJam int isn't in Any's tree?
sorear int isn't a type
masak what is it?
PerlJam Hmm.
sorear it's ... something weird
andrei_alexandre PerlJam: if it is then the representation of $a would still be a Variant with the associated overhead
sorear my int $x; # $x is of type int, but it also has a representation trait applied 20:17
andrei_alexandre sorear: what's a representation trait?
sorear andrei_alexandre: a trait that controls data representation
I'm making this up as I go along, if you can't tell.
PerlJam andrei_alexandre: an "implementation detail" of the storage for the var. :)
sorear: #perl6 is where dreams are made real anyway. 20:18
sorear No definite answers have been reached on my int $foo
I have my thoughts on how it has to work
andrei_alexandre is there an article somewhere discussing the design in good detail?
problem is I have even trouble finding keywords to search for
sorear There isn't a design at all
andrei_alexandre I tried "perl6" "static typing"
20:19 molaf left
sorear let alone a document describing it 20:19
andrei_alexandre "perl6" "explicit typing"
etc.
no avail
sorear see above
20:20 MayDaniel joined
andrei_alexandre "perl6" "representation traits" ain't a winner either :D 20:20
sorear Right, I just made up the term 20:21
Google is frighteningly fast; I googled a term once and got the irclog as the first hit, less than an hour after the term was used
PerlJam sorear: you should write an article about your made-up terms :)
S02:568 makes me think that sub foo ($a is rw) { ... } my int $x = 7; foo($x); # is perfectly valid. 20:22
andrei_alexandre PerlJam: you mean this? github.com/perl6/roast/blob/master/...#L543-L568 20:23
sorear I'm an implementor. I reserve the right to declare parts of the synopses "completely insane" and refuse to implement them. 20:24
andrei_alexandre I found perlcabal.org/syn/S02.html, what's 568 referring to?
sorear :)
PerlJam andrei_alexandre: line number.
andrei_alexandre: have a look at irclog.perlgeek.de/perl6/today
andrei_alexandre: my text will be a link to the appropriate section 20:25
andrei_alexandre thanks!
sorear PerlJam: how I think this will work is that autoboxes are read-only 20:26
andrei_alexandre so thats' perlcabal.org/syn/S02.html#line_568
PerlJam andrei_alexandre: read the third sentence of that paragraph.
sorear: why read only?
sorear my int $foo might be stored in a register window or dedicated stack memory; it doesn't make any sense to allow read-write pointers to it
PerlJam Hmm 20:28
20:29 molaf joined
PerlJam I was kind of thinking that $a would be an Any that knows how to translate reads/writes into a form that's useful for the int $x (or carp if that can't be done) 20:29
so ... sub foo ($a is rw) { $a = "hi" }; my int $x = 7; foo($x); # dies horribly 20:30
20:30 Entonian joined
andrei_alexandre so let's seee 20:30
PerlJam but ... sub foo ($a is rw) { $a = 32 }; my int $x = 7; foo($x); # works fine
andrei_alexandre third sentence of paragraph (and the following paren: "However, Perl will try to make those look like their corresponding uppercase types if you treat them that way. (In other words, it does autoboxing. Note, however, that sometimes repeated autoboxing can slow your program more than the native type can speed it up.)"
sorear Constructing those objects is going to get very expensive very quickly. I'd rather not bother.
andrei_alexandre So it does autoboxing
but autoboxing doesn't quite work for pass by reference 20:31
hence my question
sorear andrei_alexandre: autoboxing doesn't quite work for pass by reference.
andrei_alexandre that's what I think
boxing an int works, you box it and you pass it around
sorear andrei_alexandre: when you say "my int $foo", you're leaving the land of ponies and butterflies and entering the world of native types, which is a much less pleasant place 20:32
andrei_alexandre boxing a reference to int... is only possible if there's a "reference to int" type around
sorear if you don't want to deal with the ramifications of native types, don't use them
andrei_alexandre I don't want to use/avoid explicit types
just to figure the design decision
20:33 molaf left
PerlJam maybe TimToady will read this and decide what's Right and True and patch the spec for us ;) 20:33
sorear I've asked TimToady about this more than once. 20:34
20:39 atomicstack left 20:44 risou left 20:47 andrei_alexandre left 20:52 deepaknagarajan joined, pythonian4000afk is now known as pythonian4000 20:55 deepaknagarajan left 20:59 hudnix left 21:00 hudnix joined 21:01 MayDaniel left 21:02 Italian_Plumber joined 21:10 Trashlord left 21:14 pyrimidine left 21:20 slavik left 21:25 bluescreen left 21:28 justatheory joined
masak 'night, #perl6 21:34
21:34 masak left 21:37 knewt_ joined, knewt_ is now known as knewt2, Entonian left 21:40 knewt2 left 21:41 knewt_ joined, knewt_ is now known as knewt2, jlaire left 21:51 jlaire joined, icwiener left 21:55 Schwern left 21:58 Mowah left 22:09 justatheory left, justatheory joined
alester I'm working w/Damian on getting input from folks about threads in Perl 6 22:11
Gonna run a blurb on Perlbuzz asking people for input about what works and doesn't work with concurrent processes.
per his mail from a little while ago
Anyone willing to be a point of contact for collecting those? I'm guessing that people won't want to sign up for perl6-whatever if they don't have to, and the lower the barrier to entry, the better. 22:12
22:12 mavrc left 22:19 edon left 22:26 estrabd is now known as estrabd_amf 22:32 Alias left 22:45 Entonian joined 23:15 jferrero left 23:17 Trashlord joined 23:19 sunnavy left, sunnavy joined 23:22 bluescreen joined 23:25 timbunce left 23:30 alester left 23:33 rgrau_ left, rgrau_ joined 23:44 orafu left 23:46 orafu joined 23:58 rgrau_ left

Powered by Cro and the Raku® Programming Language.

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

Thank you!