»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by sorear on 25 June 2013.
00:01 martinzaremba joined
lizmat feels all alone and also calls it a day 00:13
sleep&
japhb lizmat: Not all alone,
woolfy lizmat is definitely not alone!
japhb just all the people still around are multitasking or eating dinner or somesuch
woolfy is 1,5 meter away
woolfy hits lizmat 00:14
japhb guesses a table between.
With something 1,6 meter long?
lizmat ouch!
woolfy I juest leaned over and smacked her on her shoulder
alone... puh!
japhb It's a true geek that responds to being smacked in meatspace with "ouch!" in IRC space 00:15
woolfy yeah, here it was complete silence (except from he sniggering softly) 00:16
s/he/she/ darn...
japhb I mentally s/he/her/ 00:17
woolfy shite
her indeed
English is fiddicult
sleep well all
00:19 martinzaremba left 00:20 martinzaremba joined
japhb same to you 00:22
00:24 rurban joined 00:25 martinzaremba left 00:28 rurban left 00:29 eternaleye left 00:30 eternaleye joined 00:36 hoverboard joined 00:50 fhelmberger joined 00:55 fhelmberger left 01:00 anaeem1 joined 01:05 anaeem1 left 01:19 SevenWolf left 01:29 Sqirrel left 01:30 FROGGS left 01:36 wooden joined, wooden left, wooden joined, Su-Shee_ joined 01:39 Su-Shee left 01:41 xenoterracide joined 01:42 FROGGS joined 01:43 jnap left 01:50 klapperl_ joined
TimToady rosettacode.org/wiki/Stable_marriag...lem#Perl_6 <-- wish the highlighting knew about ' inside identifiers... 01:53
01:53 klapperl left
TimToady with the recent additions by Grondilu++ and FROGGS++ we have now passed both C and J, and are now in 4th place :) 01:59
colomon \o/ 02:01
02:02 hoverboard left 02:09 Pleiades` left 02:17 Pleiades` joined 02:23 Pleiades` left 02:24 eternaleye left 02:28 eternaleye joined 02:33 eternaleye left 02:34 kaare_ joined 02:37 xragnar_ joined, xragnar is now known as Guest43817, Guest43817 left, xragnar_ is now known as xragnar 02:39 sorear left 02:42 eternaleye joined 02:43 sorear joined 02:47 sorear left 02:48 molaf joined 02:50 fhelmberger joined 02:53 sorear joined 02:55 fhelmberger left 02:58 sorear left 03:03 sorear joined 03:08 raiph joined 03:09 SevenWolf joined, molaf left
japhb agrees with colomon 03:10
03:12 sorear left 03:15 sorear joined 03:22 martinzaremba joined 03:24 johnjohn101 joined
johnjohn101 hi perl 6 03:25
03:25 Alula left 03:28 raiph left 03:32 martinzaremba left 03:33 martinzaremba joined
TimToady howdy johnjohn101 03:36
03:36 Alula joined
johnjohn101 what's the word on perl 6? 03:36
TimToady exciting 03:37
johnjohn101 moarvm is now the latest?
03:37 hoverboard joined
TimToady that would be more than one word :P 03:37
but yes, it's getting to be the leader
but in another sense, all the backends are the "latest", since we're developing in parallel 03:39
r: say $*VM.name
camelia rakudo-moar d78b63: OUTPUT«moar␤»
johnjohn101 what ever happened to perl 6 on java vm?
camelia ..rakudo-jvm d78b63: OUTPUT«jvm␤»
..rakudo-parrot d78b63: OUTPUT«parrot␤»
TimToady there it is ^^
johnjohn101 is there going to be an official release soon? 03:40
TimToady what is this word, "official"?
TimToady looks around for officers... 03:41
johnjohn101 meaning it's considered production ready, i guess
TimToady we'll just keep making it better, and eventually you won't be able to stand not using it... :)
when it's production ready depends on what you're producing... 03:43
I probably wouldn't use it for plane avionics just yet
johnjohn101 is odbc available for it yet?
probably the biggest part of my perl 5 coding 03:44
TimToady not sure whether DBIish supports that yet, but yeah, that is important
could probably get at JDBC via jvm 03:46
I think Tim Bunce has some plans for what he wants to do with Perl 6, but not sure where he is on that... 03:47
but I'm sure we could always use more help to push the db stuff forward 03:48
johnjohn101 he's done a good job with perl 5 for sure
TimToady we don't really have many dbfolk here on channel, since we're mostly in eat-your-own-dogfood mode still, and compilers don't need much in the way of databases 03:50
so DBIish is what we have so far
03:51 kaare_ left
TimToady looks like it supports mysql, pg, and sqlite currently (and TestMock) 03:52
there's a lot of TODO comments in it, though
so I'd consider it more of a proof of concept 03:53
johnjohn101 thanks, i may get some time in july on vacation to mess with it
TimToady cool
03:59 hoverboard is now known as totescalientebab 04:00 totescalientebab is now known as pinklipsbaby 04:08 pinklipsbaby is now known as hoverboard 04:16 BenGoldberg left 04:18 SamuraiJack joined 04:20 xenoterracide left 04:23 anaeem1 joined, kaare_ joined, johnjohn101 left 04:27 perigrin left 04:48 perigrin joined 04:49 perigrin left 04:50 perigrin joined 04:51 fhelmberger joined, anaeem1 left 04:52 thou left 04:54 anaeem1_ joined 04:55 fhelmberger left 04:57 REPLeffect joined 05:04 rurban_ left 05:07 rurban_ joined 05:17 araujo left 05:36 xinming_ left 05:37 hoverboard left, FROGGS left 05:40 jack_rabbit left 05:41 jack_rabbit joined 05:42 xinming joined 05:55 kurahaupo joined 05:56 woosley joined, hoverboard joined 05:59 bjz joined 06:04 jack_rabbit left 06:17 kaleem joined 06:22 jsn_ joined, bjz left 06:23 bjz joined 06:25 FROGGS joined
FROGGS o/ 06:26
TimToady \o 06:27
06:49 bjz left
jnthn o/ 06:50
06:50 rindolf joined 06:51 fhelmberger joined 06:54 anaeem1_ left, martinzaremba left 06:55 anaeem1_ joined 06:56 fhelmberger left 07:01 dayangkun left 07:02 dayangkun joined
sergot morning o/ 07:05
07:06 zakharyas joined
yakudza r: say $*VM.name 07:10
camelia rakudo-moar d78b63: OUTPUT«moar␤»
..rakudo-jvm d78b63: OUTPUT«jvm␤»
..rakudo-parrot d78b63: OUTPUT«parrot␤»
07:11 darutoko joined 07:12 ilbot3 joined
sergot m: if 1 { LAST { say "run me!" } } 07:15
camelia ( no output )
jnthn That only makes sense in a loop.
sergot right :) 07:16
jnthn (You can attach phasers to whatever block you like, I guess, but for LAST something has to be there to actually fire it.)
07:20 dmol joined
sergot jnthn++ thanks 07:28
07:28 pdcawley left 07:30 igorsutton joined, pdcawley joined 07:33 jsn_ left, spider-mario joined 07:36 pecastro left 07:39 woolfy left, woolfy joined
nwc10 Why did Apple honour Python as the dynamic language to (effectively) dump on? Are people more likely to be writing iPhone apps in Python than (say) Ruby? 07:40
jnthn nwc10: They dumped on it? 07:42
igorsutton nwc10: Support for Python was available way before Ruby. I remember having Python/PyObjC as an option to write Mac apps in the past.
jnthn musta missed something...
jnthn hasn't read the new language's manual yet. :) 07:43
igorsutton nwc10: Made by Sofa from Amsterdam had a Mac application written in Python.
nwc10 jnthn: Apple announced a new language, named Swift, which is (sort of) "Objective-C without the C."
igorsutton nwc10: And AFAICS Swift is much more like Go than Python.
jnthn nwc10: Yes, that bit I know...it's the Python connection that isn't clear to me.
nwc10 and the slides from WWDC show speed comparisons with Python and IIRC Objective C 07:44
jnthn Oh.
nwc10 yes
jnthn OK.
Now I understand :)
igorsutton nwc10: Although some of its syntax reminds PyObjC bindings.
nwc10 now you understand what I don't understand
er, add () to make that make sense
igorsutton nwc10: As I said, people have used PyObjC to make Mac apps in the past. 07:45
nwc10 www.cnet.com/news/apples-swift-expl...-it-means/ -- The other concern, which isn't much of a worry in my mind, is that by lowering the barrier of entry with an easier to learn, easier to use language, you'll result in more rookie developers launching apps that simply aren't very good. 07:46
so, SWIFT? :-)
07:46 virtualsue joined
igorsutton nwc10: This happens all the time, i.e. PHP 07:46
nwc10 I was thinking more metacpan.org/release/JMCNAMARA/Acm...-PERL-0.01 07:48
07:49 hoverboard left 07:51 denis_boyun_ joined 07:55 Su-Shee_ is now known as Su-Shee
jnthn notices that Swift dropped the parens around conditions in conditionals/loops, like Perl 6. 08:04
08:11 bjz joined 08:12 rindolf left 08:13 rindolf joined 08:14 Sqirrel joined, denis_boyun_ left 08:23 zakharyas left, zakharyas joined
igorsutton jnthn: yes! :) 08:27
jnthn: And dropped also the semicolons as well :D 08:28
jnthn igorsutton: I don't really care for that choice, tbh. 08:29
igorsutton jnthn: Less is more IMO
jnthn I'd say this is one place where it isn't. :) 08:30
C# really needs to steal "let x = ...;" outside of Linq queries. I've wanted that for ages.
igorsutton Why?
(about the semicolons)
timotimo nwc10: make the language as hard as possible, otherwise all the newbies will be publishing their code! 08:31
igorsutton timotimo++
08:31 donaldh joined
jnthn Makes it obvious what's happening in a multi-line expression, and provides a clear visual indication of "end of utterance", just like natural languages typically do with a "." 08:31
igorsutton jnthn: That's true, but usually when writing code you use one line per statement anyways. 08:33
jnthn That becomes less and less true for me, as my code generally tends to become less and less imperative. YMMV. 08:34
igorsutton jnthn: Right, I do see that. But usually like that you're more chaining operations than inlining them in one line.
08:35 dakkar joined
igorsutton jnthn: Do you have an example so I understand? 08:35
jnthn Anyway, don't see anything earth-shattering in Swift. It's got some nice ideas; happy to see destructuring, a not-busted-like-C#/Java string model, type inference, lambdas. Not so fond of the lambda syntax, mind. 08:36
nwc10 why are C# strings busted, and is it the same reason as Java? 08:37
igorsutton jnthn: I don't consider it earth-shattering, but it is good to get refreshed :)
I particularly liked the let/var model 08:38
jnthn igorsutton: github.com/jnthn/perl6-reactive-sa...-helper.p6 is full of multi-line statements.
nwc10: Yes. UTF-16. 08:39
nwc10 aha
So, like JavaScript too?
arnsholt No, JS is UCS-2, which is terrible
nwc10 er, so it can't even think about Red Apples and Green Apples? 08:40
arnsholt Why is UTF-16 a problem, beyond being no-actually-fixed-width and such?
nwc10 what a 0xD83D 0xDCA9
arnsholt Yeah, UCS-2 is fixed width 16-bit, while UTF-16 encodes astral characters with surrogate pairs
ren1us I'm actually just skimming the language specs now, but one thing jumped out at me as just being a weird thing to bother with 08:41
"To use a reserved word as an identifier, put a backtick (`) before and after it. For example, class is not a valid identifier, but `class` is valid. The backticks are not considered part of the identifier; `x` and x have the same meaning."
igorsutton jnthn: They're multi-line statements, true. But usually one statement per line neverthless.
arnsholt ren1us: I dunno, might be nice. Sometimes you really do want a variable with the same name as a reserved word 08:42
jnthn arnsholt: It's a problem if you ask for the length of a string and get the number of 16-bit things. Or you do a substring operation and can chop a surrogate pair in half....
ren1us: Such things are very useful in any code-gen scenario.
arnsholt When doing metaprogrammy stuff in Java, you frequently want a variable "class", for example (usually called clazz in Java, IIRC)
jnthn: Yeah, that's bad. But more of a problem with the implementation than UTF-16 though, no? 08:43
jnthn C# provides @ for this...
masak antenoon, #perl6
jnthn @class, for example
arnsholt "Oh, UTF-16 isn't fixed width after all? Let's patch it up as best we can..."
jnthn arnsholt: It's not UTF-16 that's bad, it's not distinguishing between opaque strings that are at least at codepoint level vs. an array of 16-bit numbers 08:44
arnsholt *That* I can agree is a problem! =
=)
With the potential for causing some pretty nasty bugs too, I suspect
igorsutton Usually in ObjC you get a reference for the data as UTF8String or something. Being strings in Swift NSStrings, I think you get the same sort of access. 08:45
nwc10 jnthn: we are live with hackathon: act.useperl.at/apw2014/talk/5565 act.useperl.at/apw2014/talk/5566 08:46
jnthn yay 08:47
ooh, and up to September to decide what to talk about. yay.
08:48 SevenWolf left
nwc10 might be useful for scheduling if you submit a talk that is a placeholder/teaser, and we can put it live 08:48
details to be confirmed later
timotimo salzburg is 6 hours by train, 4.5 hours by car :| 08:49
nwc10 where are you?
Who can you share a ride with?
timotimo i'm in karlsruhe; potentially jean AKA corecatcher might go there 08:50
08:51 pippo joined
jnthn 6 hours isn't an especially long train trip... :) 08:51
lizmat good *, #perl6!
08:52 fhelmberger joined
lizmat timotimo: we're going by car, and usually pass by Mannheim anyway, so Karlsruhe may just be short detour for us 08:52
timotimo jnthn: if you're used to it! :)
lizmat: i shall ponder my attendence
lizmat cool!
jnthn timotimo: do the trans-sib some day, that'll get you used to it :P
timotimo how many days is that again?
08:53 pecastro joined
jnthn 6-7 I think, though it's more fun to get off every so often and look around a bit :) 08:53
timotimo mhm
jnthn dragged it out over 3 weeks, and the longest chunk of time actually spent on the train was 36 hours or so
timotimo it goes mostly cold places, doesn't it? 08:54
jnthn Wouldn't mind doing it again, though will probably do the BAM or Trans-Mongolian next time... :)
timotimo: Depends on time of year. :)
timotimo mhh, look at the time, i ought to get going. 08:55
nwc10 5 was a drag from Vienna to Erlangen
moritz nwc10: when you're next in or around Erlangen, please ping me :-) 08:56
nwc10 I think that the real drag part was that the ICE wasn't very E
08:56 denis_boyun_ joined, fhelmberger left
moritz didn't read backlog, just responds semic-mechanically to location names 08:56
nwc10 trains should not be slower than cars
timotimo you got dragged? wow, that sounds really uncomfortable!
08:58 denis_boyun_ left 09:08 anaeem1_ left 09:09 anaeem1_ joined 09:11 anaeem1_ left 09:12 anaeem1_ joined
dalek rlito: 647112c | (Flavio S. Glock)++ | / (3 files):
Perlito5 - parser - localize compiler state before BEGIN
 09:13
09:13 anaeem1 joined, anaeem1_ left
dalek kudo/nom: 17f0230 | (Elizabeth Mattijsen)++ | src/core/Process.pm:
Take #2 on implementing $*USER and $*GROUP

Please note that this adds one qx() for every start. Somehow, I can't get this to be on-demand using Proxy. Pointers in that direction welcome!		 09:18
09:19 woolfy left
masak m: month 09:21
camelia rakudo-moar d78b63: OUTPUT«===SORRY!=== Error while compiling /tmp/LaTRxwVvWt␤Undeclared routine:␤ month used at line 1␤␤»
masak m: constant m = month
camelia rakudo-moar d78b63: OUTPUT«===SORRY!===␤Cannot invoke null object␤»
masak submits LTA rakudobug
r: constant m = month
camelia rakudo-moar d78b63: OUTPUT«===SORRY!===␤Cannot invoke null object␤»
..rakudo-parrot d78b63: OUTPUT«===SORRY!===␤Could not find sub &month␤»
..rakudo-jvm d78b63: OUTPUT«===SORRY!===␤java.lang.NullPointerException␤»
jnthn lizmat: uh, that's not going to be good for our startup time :( 09:24
Not to mention it's unportable.
lizmat: Why not just make the class obtain the information if requested?
masak could it be made on-demand in some way?
right. 09:25
timotimo 0.00user 0.00system 0:00.00elapsed 100%CPU (0avgtext+0avgdata 1016maxresident)k ← a run of "id" :P
so it'll all be our qx overhead
masak oh, lizmat++ even uses the word 'on-demand' in her commit comment :)
lizmat did you read the commit message ?
I *tried* to get it to use Proxy.new
masak read the commit message on-demand :P
jnthn lizmat: Yes, did you read what I said?
lizmat: I said the *class* should obtain the info on demand.
Why does it need Proxy? 09:26
timotimo to react to an access to the scalar
though if we have an object
lizmat hmmm... I though of IdName as a generic class, not specific for $*USER
jnthn So give it a constructor arg that populates things on demand? 09:27
lizmat ah, ok, I see what you mean (I think)
will ponder that during the day 09:28
jnthn This is the kind of commit where I'd rather we didn't have the feature at all than have it done this way...sorry.
lizmat well, I agree... hence my request for pointers :-)
jnthn Anyway, passing a generator to the class that it triggers on .id or .name being used for the first time is probably enough. 09:29
lizmat jnthn: do you want me to revert it now or can you wait until I fix this later today ?
jnthn lizmat: If you'll fix it later today, that's fine.
(as in, I can wait)
lizmat okidoki, might be past midnight local, but still before midnight UTC :-)
jnthn :) 09:30
lizmat sightseeing&
jnthn have fun o/
09:30 lizmat left 09:45 denis_boyun joined 09:47 pecastro left
masak m: my @foo = <abc def ghi>; .say for @foo.map(*.subst(/^./, '').uc) 09:50
camelia rakudo-moar d78b63: OUTPUT«BC␤EF␤HI␤»
moritz m: say for <abc def ghi>.map(*.substr(1).uc) 09:52
camelia rakudo-moar d78b63: OUTPUT«␤␤␤»
moritz m: .say for <abc def ghi>.map(*.substr(1).uc) 09:53
camelia rakudo-moar d78b63: OUTPUT«BC␤EF␤HI␤»
09:53 pecastro joined 09:57 kurahaupo left
timotimo m: .substr(1).uc.say for <abc def ghi> 09:59
camelia rakudo-moar d78b63: OUTPUT«BC␤EF␤HI␤»
10:00 virtualsue left 10:01 rindolf left
nwc10 jnthn: gah. I can create a placeholder talk for you, but I can't hide it from the public, and if I delete it, you stop being a confirmed attendee 10:05
er, "committed"
ACT can commit you without your consent :-)
but it seems that ACT (at least the web admin interface) can't sign someone up for the conference without their consent 10:06
jnthn nwc10: Hm, I joined the conf... 10:07
nwc10: Does that mean you need me to submit the placeholder?
10:07 dwarring left
nwc10 yes, you joined. But I can't mark you as "confirmed" and make your bold 10:07
{your name,you} bold
10:11 anaeem___ joined 10:13 anaeem1 left 10:14 anaeem___ left 10:17 anaeem1 joined
jnthn nwc10: Well, filed you a placeholder one in case it's useful. 10:23
10:28 donaldh left
nwc10 jnthn: I think "yes" (we may change our mind, but for now it is accepted) 10:30
it was more "I can't make you a placeholder without putting words into your mouth"
10:31 donaldh joined
masak TIL the term 'upvar': a variable rooted not in the currently executing routine's lexpad. 10:33
a quick search seems to indicate that Tcl uses this term a lot.
moritz would just call it an "outer" 10:34
jnthn iirc lua uses that terminology too 10:36
10:37 virtualsue joined 10:41 fhelmberger joined
Woodi hi today :) 10:49
masak: how you think, what functionalities ufo4c should have ?
11:09 synopsebot left 11:10 masak left, d4l3k_ joined, pmichaud left, Util joined, PerlJam joined, dalek left, pmichaud joined, d4l3k_ is now known as dalek 11:11 tadzik joined, sergot joined 11:12 masak joined, masak is now known as Guest21530 11:13 FROGGS left 11:27 synopsebot joined 11:45 anaeem1 left, anaeem1__ joined 11:50 vendethiel left 11:51 anaeem1__ left 11:52 anaeem1_ joined 11:57 ppabian joined, ppabian is now known as bbkr_ 11:59 Sqirrel left 12:00 vendethiel joined 12:01 FROGGS joined 12:05 pecastro left 12:06 kivutar joined 12:10 anaeem___ joined 12:12 Guest21530 is now known as masak 12:13 anaeem1_ left
masak <Woodi> masak: how you think, what functionalities ufo4c should have ? 12:21
Woodi: what is ufo4c?
12:21 sqirrel joined
jnthn Presumably, a ufo but for C :) 12:21
tadzik so, a headerizer? 12:23
I'd love myself a C++ headerizer
it may just make working with C++ a bit more bearable 12:24
masak jnthn: that was my guess, too. 12:26
I just wanted to make sure. I haven't heard about this project before.
not really sure why Woodi is asking me this ;) 12:27
I mean, if he's scratching an itch, shouldn't the features be pretty clear already?
if not, what's the goal of the project? who is the prospective end user? maybe ask them, not me? :) 12:28
12:28 pecastro joined
bbkr_ rakudo: my $x = 3; print $x.WHAT 12:33
camelia rakudo-{jvm,moar} 17f023: OUTPUT«use of uninitialized value of type Int in string context␤»
..rakudo-parrot 17f023: OUTPUT«use of uninitialized value of type Int in string context in block at /tmp/tmpfile:1␤␤»
bbkr_ hm, why print behaves differently than say? 12:34
jnthn Because print does .Str and say does .gist
(No, I don't know why. Yes, I did argue for both to do .gist. No, I didn't win that debate. :)) 12:35
bbkr_ ah, some old context that say=print+"\n" was stuck in my head. thanks for explanation 12:36
jnthn Well, that's how I always used to have it too, and part of me wishes it were so simple :)
Alas, it's not...
masak ISTR it had something to do with printing to filehandles.
which needs to be low-level enough. 12:37
I *think* that was the reason.
I don't like it.
we have too many ways to stringify things.
jnthn The Str/gist distinction makes sense *in general*.
This is just a place I was a bit surprised to see it crop up. 12:38
12:38 guru joined
masak aye. 12:38
12:38 guru is now known as Guest74357
masak but we have not just Str and gist, we also have pretty and DUMP. 12:38
12:39 Guest74357 is now known as ajr_
masak and probably some fifth one I can't recall right now. 12:39
jnthn DUMP is a Rakudo-specific debugging aid. And .pretty...uh...is that actually a thing?
r: say "foo".pretty
camelia rakudo-{parrot,jvm,moar} 17f023: OUTPUT«No such method 'pretty' for invocant of type 'Str'␤ in block at /tmp/tmpfile:1␤␤» 12:40
[Coke] sees tcl & upvar in the backlog. squee!
FROGGS *g*
jnthn I know .pretty has been discussed. I don't recall it making spec.
masak oh, phew. 12:41
[Coke] -1 on yet another .Str variant. :)
masak this is one of the places where I think Python's got it very right.
[Coke] (if we need a pretty print, I think .perl(:pretty) would be better.)
jnthn No mention of .pretty in S02, anyway.
masak m: say [1, 2, {foo => 3}] 12:42
camelia rakudo-moar 17f023: OUTPUT«1 2 {"foo" => 3}␤»
masak o.O
do the same thing in Python, and it automatically does (the analogue) of .gist on the array.
>>> print [1, 2, { 'foo': 3 }] 12:43
[1, 2, {'foo': 3}]
that is, stringification *is* .gist *is* .perl
12:43 kaare_ left
masak it's hard to argue with the simplicity of that. 12:43
dalek rlito: 00aef6f | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js - optimize throw() calls
masak I also find it hard to defend the default stringification of Perl 6.
in some sense, I feel that is a relic of Perl 5 that we didn't get rid of yet. 12:44
we did manage to get rid of it with True and False \o/
also -- and I *guess* this is an argument in my favor -- the out put is OUTPUT«1 2 {"foo" => 3}␤» these days. it used to be OUTPUT«1 2 "foo"\t3␤» 12:45
don't know when it changed or why. maybe it just happened.
but it is converging on the sanity I'd prefer.
12:45 jnap joined
masak jnthn: re Swift and semicolons, it seems that the consequence for Swift programs is long lines of code: github.com/fullstackio/FlappySwift...cene.swift 12:52
so you could call Swift a "horizontall-scroll-bar-oriented language" 12:53
sergot searchengineland.com/big-win-duckdu...ion-193081 12:55
12:58 pdcawley left 12:59 anaeem___ left 13:00 pdcawley joined
jnthn masak: Maybe the plan is to drive sales of a forthcoming ultra-widescreen product :P 13:04
[Coke] I also am not a fan of having multiple methods of stringification. (4 if you count pretty .perl, which I would rather have than .gist or .perl)
13:09 thou joined
[Coke] m: my $b=[]; say \$b # unrelated 13:10
camelia rakudo-moar 17f023: OUTPUT«:()␤»
13:13 thou left
masak :() <-- monkey smiley 13:15
13:16 [particle] joined
dalek kudo-star-daily: d8197f9 | coke++ | log/ (13 files):
today (automated commit)
 13:21
[Coke] no change in modules status. (moar & jvm still have module failurs) 13:22
13:22 vendethiel left 13:24 vendethiel joined 13:28 [particle] left 13:42 vendethiel left 13:43 zengargoyle joined 13:45 vendethiel joined 13:47 btyler joined, ajr_ left 13:48 guru joined, guru is now known as Guest39451 13:49 Guest39451 is now known as ajr_ 13:58 zengargoyle left 14:00 thou joined 14:02 ajr_ left, skids left 14:03 sqirrel left, ajr joined 14:04 ajr is now known as Guest75201 14:05 Guest75201 is now known as ajr_, uniejo joined 14:10 pecastro left 14:12 uniejo_ joined 14:15 Alula left, rurban joined 14:16 Alula joined, pecastro joined, Gothmog_ left 14:18 rurban left
bbkr_ sergot++ # HTTP stuff 14:19
Woodi is back 14:20
dalek rlito: e36dc1a | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js - code cleanup
 14:21
Woodi headizier ? what's headizier ?
but OK, question was way to simple... I'm sory for that... 14:22
moritz only knows "headerizer"
Woodi ok, what is das ? :)
masak Woodi: not too simple, no. out-of-context, yes.
Woodi ah, right :) 14:23
moritz Woodi: in the parrot sources, it's a tool that writes header files for you
masak Woodi: for comparison, I would have been similarly confused if you had asked me what features the projects "Elephants on Stilts" should have.
Woodi scan .c and write .h ? nice :)
14:23 [particle] joined 14:24 pecastro left
moritz iirc it scans op definitions 14:24
not .c files
jnthn masak: I know you're confused, but please tell us! :D
masak jnthn: well, if it's anything like Ruby on Rails, I guess it's a web MVC framework. 14:25
moritz so, what features should it have?
:-)
masak jnthn: but it somehow involves... elephants? I dunno. here is where the confusion begins in earnest.
jnthn: maybe they are there to remind you not to write bloated controllers?
Woodi Elephants on Stilts project is... hmm.. self describing
moritz elephants are blobs that might contain huge object graphs, obviously 14:26
masak Woodi: now, see, this is where our disagreements begin.
jnthn Maybe the stilts reflect layered architecture.
masak Woodi: to me, the name only creates more questions.
Woodi: as did the project name "ufo4c".
Woodi ok, I was just lazy and imagined it to not write makefile manually... my bad for laziness ;) 14:28
colomon laziness is good. 14:30
14:30 uniejo left 14:31 kaare_ joined 14:32 uniejo_ left, Gothmog_ joined, anaeem1 joined
moritz Woodi: there are tools that aim to allow that 14:36
cmake is one of them
FROGGS I don't like cmake 14:40
moritz also, automake
colomon has a Perl 5 script to do that sort of thing for $work 14:42
14:42 dmol left 14:43 oetiker_ left, oetiker_ joined, oetiker_ left
tadzik a friend of mine wrote a makefile generator which works pretty well 14:44
Woodi: github.com/Vifon/vfnmake it may come to your liking
it even supports qt weirdness
I like it because it's zero-conf 14:45
14:48 oetiker__ joined, oetiker__ left
moritz you like that up to the point where you need conf :-) 14:49
moritz is just being sarcastic
14:51 crazed- joined, crazed- left 14:52 crazed- joined, oetiker left
sergot bbkr_++ thanks :) 14:53
14:53 oetiker joined, oetiker left, telex left, oetiker joined 14:54 telex joined, oetiker left 14:55 oetiker joined, crazedpsyc left, crazed- is now known as crazedpsyc 14:57 oetiker left 14:58 oetiker_ joined 14:59 oetiker joined 15:00 rurban joined 15:03 kaleem left
masak m: class D {}; class C { has Array[D] %.ds; method foo { %.ds<bar>.push: D.new } }; C.new.foo 15:04
camelia rakudo-moar 17f023: OUTPUT«Type check failed in assignment to '%!ds'; expected 'Array[D]' but got 'Array'␤ in method push at src/gen/m-CORE.setting:1623␤ in method foo at /tmp/o9AcgHoKoI:1␤ in block at /tmp/o9AcgHoKoI:1␤␤»
masak status report on typed containers: not there yet. :/ 15:05
as far as I can understand, the above fails because the auto-vivification helpfully creates an Array for me.
and that Array is not an Array[D], so the typecheck fails. 15:06
masak submits rakudobug
15:06 denis_boyun left
masak m: class D {}; my Array[D] %ds; %ds<bar>.push: D.new # golf'd 15:07
camelia rakudo-moar 17f023: OUTPUT«Type check failed in assignment to '%ds'; expected 'Array[D]' but got 'Array'␤ in method push at src/gen/m-CORE.setting:1623␤ in block at /tmp/xN_lIMlwRy:1␤␤»
jnthn m: class D {}; my Array[D] %ds; say %ds<bar> 15:10
camelia rakudo-moar 17f023: OUTPUT«(Array[D])␤»
jnthn Ah, good, so we do know the type...
15:14 skids joined 15:22 anaeem1 left
Woodi tadzik: wow, vfnmake looks quite nice :) just name need to be changed :) 15:23
15:24 anaeem1 joined, cognominal left
tadzik I think the author is actually actively looking for a new one :) 15:24
15:25 cognominal joined
Woodi :) 15:26
15:29 sqirrel joined 15:30 igorsutton left
Woodi but to be on topic: I think UDP is state of art in games industry, at least in last ~20 years :) before telnet was used for reliable gaming :) 15:31
15:35 treehug88 joined, pecastro joined
Woodi and about asm/jit and some such: djb maked some asm code generator, he use(d) it to implement fast cryptography algos and wining some contests :) cr.yp.to/qhasm.html maybe some speedups could be borrowed ? 15:36
15:38 thou left 15:41 thou joined, bjz left 15:43 zakharyas left
TimToady re: "upvar", I believe it is dynamic, not lexical 15:46
so I wouldn't call it "outer"
[Coke] each proc in tcl has its own var scope. That's not lexical? 15:49
set a 3 ; proc barf {} { puts $a } ; barf ; #barfs 15:51
TimToady it talks about the "caller"; if you go "up" do you get to the caller or the outer scope? doesn't help distinguish when those are the same scope 15:52
15:55 fhelmberger left 15:56 SamuraiJack left
[Coke] ah, yes, it's all based on caller. 15:57
16:02 araujo joined 16:03 bbkr_ left
masak oh, ok. 16:05
a bit disappointed at that. I like the term better when I misunderstood it.
TimToady If only the world would conform to our misunderstandings, we'd all be much happier, except for when we talk to each other. 16:08
Except, of course, when we all misunderstand the same way... 16:09
16:10 denis_boyun_ joined
masak :) 16:12
I think I understand what you mean. 16:13
m: class Tail { method wag {} }; class Dog { has Tail $.tail handles 'wag' .= new(|%_) }; say "alive" 16:17
camelia rakudo-moar 17f023: OUTPUT«===SORRY!=== Error while compiling /tmp/sSxBHybie8␤Cannot use .= to initialize an attribute␤at /tmp/sSxBHybie8:1␤------> as Tail $.tail handles 'wag' .= new(|%_)⏏ }; say "alive"␤»
masak this is directly from S12:1456. 16:18
synopsebot Link: perlcabal.org/syn/S12.html#line_1456
masak which one is right, S12 or Rakudo?
jnthn Well, Rakudo is right in so far as you cannot actually do it right now :P 16:19
I saw an RT about it last night when browsing, though, so you're not the first to want it. :)
donaldh I'm getting a race condition / deadlock in async IO on both JVM and Moar: gist.github.com/donaldh/090b39bb1f4732ad4b82
masak jnthn: S12 is the first to want it ;) 16:20
jnthn: ok, then I know it's a NYI.
jnthn donaldh: I'm aware of like at least two of those that I didn't get around to dealing with yet. :) So probably. :)
masak: Yeah. I guess it may be a mildly tricky transform. 16:21
donaldh jnthn: okay. any pointers and I can help investigate.
jnthn donaldh: Well, it's mostly a missing bit of Supply infrastructure that needs writing.
So that it doesn't start trying to give you the packets until you .tap'd. 16:22
shopping, bbiab
donaldh yes, sleep 0.1 after connecting allows it to run through. Will look at pending supply jnthn++ 16:25
masak detrain & 16:28
16:28 denis_boyun_ left, fhelmberger joined 16:30 donaldh left 16:32 lizmat joined 16:36 lizmat left 16:39 fhelmberger left 16:41 guru joined, sqirrel left 16:42 guru is now known as Guest30712 16:43 ajr_ left 16:48 Rotwang joined 16:49 dakkar left
dalek rlito: 4022f76 | (Flavio S. Glock)++ | src5/lib/Perlito5/ (2 files):
Perlito5 - js - move context code to subroutine
 16:49
rlito: 9b990c5 | (Flavio S. Glock)++ | src5/lib/Perlito5/constant.pm:
Perlito5 - fix last commit
16:51 Guest30712 left 16:53 virtualsue left
cognominal I have a question about protoregex (in nqp, but it may apply to Perl 6 because relying mostly on the same code). Is there a way to know which regex matched in the proto group. And, given a protomatch list the submatches. In nq/nqp/32-protoregex.t, if I add say(+%( $<symbols> )), it barks : Cannot find method 'hash' 16:55
github.com/perl6/nqp/blob/master/t...otoregex.t
I want to fix nqp --target=parse and that seems one of the problems 16:56
I suppose, in some sense, there is no such thing as a protomatch, and not a way to access the real match 16:58
17:01 treehug8_ joined 17:03 treehug88 left 17:04 treehug88 joined 17:06 treehug8_ left
TimToady $<sym> should countain the symbol that matched 17:06
*contain
cognominal the following test succeeds: ok( $<symbols><sym> eq 'abc', 'correct proto candidate match' ); 17:08
TimToady okay $<symbols><sym> then :) 17:09
cognominal but for dumping I need to enumerate the $<symbols> submatches, that means enumerate some form of hash and array
here only the hash is relevant because there is no positional capture 17:10
I will update the test so we can discuss from there. 17:11
TimToaday : my first question is about the spec. Is there a way to identify what rule matched in a proto regex group? 17:14
17:16 sqirrel joined 17:24 go|dfish left 17:26 pecastro left, xfix joined 17:27 go|dfish joined
timotimo has solid ground under his feet again 17:33
17:37 hoverboard joined
cognominal github.com/cognominal/nqp/compare/...ick_pull=1 17:40
This is quite different from what I said to TimToady, put problems indeed, I was probably lead astray by other problems when trying to dump the parse. 17:42
dalek kudo/nom: 68060fe | (Tobias Leich)++ | src/core/JSON/Pretty.pm:
strip a call to eval (leftover from assimilating JSON::Pretty)
 17:45
17:46 spider-mario left
TimToady in general, which rule matched is only implicit in the returned match data (as "sym"), and we don't provide any kind of pointer by default to which rule matched, since that would be useless most of the time, much like an automatic index on for loops would impose overhead on everyone whether they use it or not 17:47
jnthn +1 17:48
TimToady if you need to know in a particular instance, you can always install an action to record it
cognominal ok 17:49
17:49 kaare_ left
cognominal now there is the problem of unrecorded positional matches 17:49
TimToady doesn't understand what you expect to be recorded if you didn't tell it to record it... 17:50
17:51 spider-mario joined
cognominal oops 17:52
17:52 kaare_ joined, kaare_ left
cognominal good, I added four successful test thens :) 17:52
cognominal updates the patch 17:53
TimToady otoh, there's also .caps and .chunks, which may provide you some more ways to see what matched 17:54
dalek rlito: 8264620 | (Flavio S. Glock)++ | / (59 files):
Perlito5 - js - use more regular syntax
 17:55
rlito: 57d21d2 | (Flavio S. Glock)++ | t5/ (57 files):
Perlito5 - fix last commit
TimToady those will return matches in textual order rather than as a tree 17:56
17:56 dmol joined
TimToady I believe they don't impose any extra overhead on normal matches, since they can be constructed by traversing the existing tree and then sorting; dunno if the current implementation is doing that though 17:57
cognominal github.com/cognominal/nqp/compare/...ick_pull=1 17:58
masak yes, they do.
s/they do/it does/
cognominal Don't know how to cancel my previous pull request.
17:59 Alina-malina left
cognominal sorry for wasting your precious time 18:00
masak not at all. 18:01
flood-filling algos need to have a certain degree of redundancy. 18:02
TimToady +1
the ants only have to be pulling the food toward the nest on average, not always individually :) 18:03
I believe that's even specced... 18:04
S01 sez: "Mostly, we're just a bunch of ants all cooperating (sort of) to haul food toward the nest (on average)." 18:06
18:08 ren1us left
masak that's a sentence with three hedges in it :) 18:08
18:09 cognominal__ joined
masak I'd only paid attention to the last two until today. 18:09
TimToady ants don't find hedges to be much of a barrier
masak s/hedges/caveats/
TimToady doesn't know how to tie a caveat...
jnthn ties a caveat on a hedge 18:10
TimToady the ants hope there's gravy on the caveat 18:11
18:11 cognominal left
masak .oO( it's very exclusive caveat, it's Russian ) 18:12
TimToady that's *caviat*, silly
jnthn
.oO( should putin more effort with these puns... )
 18:13
masak looked at "caviat", and his brain immediately suggested "demacsog"
TimToady should take somethign for my ukraine 18:14
masak maybe u craned your neck too much?
TimToady it ain't no crimea 18:15
.oO(black sea humor...)
masak haha
masak .oO( Май work here is дан ) 18:17
PerlJam keeps looking for the pun in "haha"
TimToady it was a "your mom" joke in Japanese
vendethiel well hello, cheerful people. 18:18
TimToady what about the rest of us?
masak vendethiel!
vendethiel o/, masak !
masak tries not to think about anything in Latin
vendethiel oh c'mon, just do whatever :). It was childish of me to make a snarky comment 18:19
jnthn masak: What happens in Vegas :P
masak jnthn: :D
jnthn: that must be your favorite Latin proverb...
jnthn Yes :)
TimToady that would be at Caesar's, right?
masak .oO( what happens at Caesar's is total craps ) 18:20
18:20 Alina-malina joined
TimToady masak is hoovering up all the nevada puns... 18:21
jnthn Dam it.
18:21 denis_boyun_ joined
masak no puns left for desert. 18:21
TimToady but at least we've had a blast 18:22
masak hey, it's easy, with a bit of Investigation, to make Hoover puns.
dylanwh this is just all night buffett of puns.
jnthn masak: You could really clean the floor with them... 18:23
vendethiel though a bit stodgy for bad english speakers trying to understand them :P
masak jnthn: except, that would suck.
TimToady a vacuum abhors nature too 18:24
18:24 eiro left
TimToady sorry I mentioned Hoover, it set a bad president... 18:25
18:25 eiro joined
[Coke] wonders if "comprehension" is another word for thunk. 18:26
PerlJam [Coke]: like in the phrase "list comprehension"? I've occacionally wondered about that usage myself. 18:27
TimToady
.oO(He comprehended after I thunk him upside the head.)
jnthn Well, I comprehend things better after I've thunk about them... :)
[Coke] sees someone talking about comprehensions in re: swift, and in context, sounded like a thunk.
jnthn [Coke]: I've not seen it as another word for it, though if for is map then I guess the body of the comprehension is thunked...
TimToady I'd just take it as a word meaning "consider all these bits together as a unit" 18:29
18:30 dmol left
TimToady that's all I apprehend in the term 18:30
18:32 sqirrel left
TimToady 3. Logic The sum of meanings and corresponding implications inherent in a term. 18:33
so I think it means something more like "list generator" 18:34
or "list as defined by this term"
But who can comprehend what mathematicians mean when they use terms differently from mere mortals? 18:35
TimToady knows what roots are, and most of 'em aren't square.
PerlJam TimToady--the Punniator. 18:36
TimToady and "graph" means "writing"
that Trogdor Punninator to you 18:37
18:38 guru joined, guru is now known as ajr_
masak m: class Widget { has $.x; method set_x { has $!x_set; die "Already set" if $!x_set++; say "Setting x" } }; given Widget.new { .set_x; .set_x } 18:40
camelia rakudo-moar 17f023: OUTPUT«Setting x␤Already set␤ in method set_x at /tmp/eNF8phGCaJ:1␤ in block at /tmp/eNF8phGCaJ:1␤␤»
masak this pattern just occurred to me.
declaring the attribute inside of the method, if it's only used in that method anyway.
TimToady heh, I didn't even notice till you mentioned it :) 18:41
masak though this is probably better expressed as a 'is once_only' trait or something.
jnthn Yeah, some people were like "argh why not forbid this", and here's a nice use case :P
masak yeah. 18:42
TimToady There is no pattern so good that someone, somewhere, won't turn it into an antipattern.
which pretty much explains the entire course of western theology :) 18:44
18:47 dwarring joined, cognominal joined 18:49 cognominal__ left
dalek rlito: a81c661 | (Flavio S. Glock)++ | / (4 files):
Perlito5 - js - use more regular syntax; more 'eval' tests
 18:50
18:53 virtualsue joined
[Coke] wait, there are non-anti-patterns in theology? 18:54
vendethiel (aren't non-anti-patterns just patterns ?) 18:57
masak only if you believe in the law of the excluded middle. 19:05
19:08 darutoko left, kivutar left, kaare_ joined
dalek rl6-roast-data: 1e2fe9c | coke++ | / (6 files):
today (automated commit)
 19:14
[Coke] still no clean rakudo runs. :|
S02-magicals/pid.rakudo.jvm........................ 1 0 1 0 2 1 todo PASSED: assigning to $*PID does not die 19:24
that todo passes on jvm & parrot if someone wants to re-fudge for moar.
FROGGS r: $*PID = 2
camelia rakudo-moar 68060f: OUTPUT«Cannot modify an immutable Int␤ in block at /tmp/tmpfile:1␤␤» 19:25
..rakudo-{parrot,jvm} 68060f: OUTPUT«Cannot modify an immutable value␤ in block at /tmp/tmpfile:1␤␤»
TimToady r: my $*PID = 2;
camelia ( no output )
FROGGS s/assigning/declaring/ or so
[Coke] was just going off presence of TODO PASSED in logoutput 19:26
TimToady most things in PROCESS should be considered immutable
r: GLOBAL::<$PID> = 2; say $*PID
camelia rakudo-{parrot,jvm,moar} 68060f: OUTPUT«2␤» 19:27
TimToady r: PROCESS::<$PID> = 2;
camelia rakudo-moar 68060f: OUTPUT«Cannot modify an immutable Int␤ in method assign_key at src/gen/m-CORE.setting:9412␤ in sub postcircumfix:<{ }> at src/gen/m-CORE.setting:2690␤ in block at /tmp/tmpfile:1␤␤»
..rakudo-jvm 68060f: OUTPUT«Cannot modify an immutable value␤ in method assign_key at gen/jvm/CORE.setting:9353␤ in sub postcircumfix:<{ }> at gen/jvm/CORE.setting:2690␤ in block at /tmp/tmpfile:1␤␤»
..rakudo-parrot 68060f: OUTPUT«Cannot modify an immutable value␤ in method assign_key at gen/parrot/CORE.setting:9343␤ in method assign_key at gen/parrot/CORE.setting:1828␤ in sub postcircumfix:<{ }> at gen/parrot/CORE.setting:2694␤ in sub postcircumfix:<{ }> at gen/parrot…»
TimToady cool
19:27 virtualsue left 19:29 skids left
masak how do I ask "are $o1 and $o2 of exactly the same type?" ? 19:33
$o1.WHAT === $o2.WHAT ? 19:34
jnthn masak: Should work, I think. 19:38
moritz m: say Mu === Mu 19:39
camelia rakudo-moar 68060f: OUTPUT«Type check failed in binding <anon>; expected 'Any' but got 'Mu'␤ in block at /tmp/9XOP341fMS:1␤␤»
moritz no, doesn't work
because === must autothread 19:40
jnthn Oh...
moritz $o1.WHAT.WHICH === $o2.WHAT.WHICH should work though
m: say Mu.WHICH === Mu.WHICH
camelia rakudo-moar 68060f: OUTPUT«True␤»
moritz m: say (Mu but role { }).WHICH === (Mu but role { }).WHICH 19:41
camelia rakudo-moar 68060f: OUTPUT«True␤»
moritz m: say (Any but role { }) === (Any but role { })
camelia rakudo-moar 68060f: OUTPUT«True␤»
jnthn Apart from .WHICH never got fixed to properly factor in "what metaclass is this
moritz huh.
m: say (Any but role { }) ~~ (Any but role { }) 19:42
camelia rakudo-moar 68060f: OUTPUT«False␤»
moritz what? they don't smart-match, but they are identical? WTF?
I believe that (Any but role { }) === (Any but role { }) returning True is a bug
jnthn moritz: They're not identical.
Right, me too
As I said, it's .WHICH to blame. 19:43
19:46 [particle] left
mr-foobar very interesting talk on c++ channel9.msdn.com/Events/Lang-NEXT/...14/Keynote #tangent 19:47
FROGGS will watch after $dayjob 19:50
moritz FROGGS: still at $dayjob? :(
FROGGS moritz: doing server maintenance from home 19:51
Su-Shee hey, why is there no noah's arc emoji perl6 example to counter swift? ;)
19:51 [particle] joined 19:56 kaare_ left, martinzaremba joined
vendethiel which reminds me, masak, are you coming to les journées perl :) ? 20:04
(or, really, anybody here is free to answer, I'd love to physically chat a bit with y'all) 20:05
(except for you, camelia.)
mr-foobar m: my \♥; 20:06
camelia rakudo-moar 68060f: OUTPUT«===SORRY!=== Error while compiling /tmp/j9fRiEjeBi␤Malformed my␤at /tmp/j9fRiEjeBi:1␤------> my ⏏\♥;␤ expecting any of:␤ scoped declarator␤»
masak vendethiel: would love to come, but can't this year. 20:07
vendethiel ah, too bad, must a meta-monad in your way :)
20:08 skids joined
mr-foobar \ was a unicode escaping thingy right ? 20:08
vendethiel it's just a sigil-less variable, innit ? 20:09
masak the latter. 20:10
mr-foobar: you have to define 'sub term:<♥> {}', I think. 20:11
which is reasonable, since you're using a non-alphanumeric character.
mr-foobar masak: any example program ? I'd like `♥("foo")` to print `I ♥ foo` 20:14
vendethiel m: sub prefix:<♥>(Str $love) { say "I ♥ $love"; }; ♥("foo"); 20:15
camelia rakudo-moar 68060f: OUTPUT«I ♥ foo␤»
vendethiel :D
vendethiel can peurlsiks now
mr-foobar vendethiel++
20:17 lizmat joined
masak m: sub prefix:<♥>(Str $love) { say "I ♥ $love"; }; ♥"foo" 20:20
camelia rakudo-moar 68060f: OUTPUT«I ♥ foo␤»
masak parentheses not required.
mr-foobar Su-Shee ^^ 20:25
Su-Shee oh I know, it works :) 20:28
mr-foobar m: sub prefix:<😂>(Str $hate) { say "😂 $hate"; }; 😂"bash" 20:29
camelia rakudo-moar 68060f: OUTPUT«😂 bash␤»
Su-Shee it even works (only half-legit) in perl 5.. so I don't really see what the fuss about swift is..
looks like I have to write it myself.. though I need an editor which actually can display the stupid unicode emoji-ranges.. 20:30
mr-foobar err wrong unicode .. well swift has optional typing which is great
it's written by the llvm people 20:31
Su-Shee "wrong unicode"? there is a right one now? ;)
masak Perl 6 has optional typing too. 20:32
we like to call it "gradual typing", though.
20:32 hoverboard left
dalek rlito: b98377f | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js - code cleanup
 20:33
jnthn masak: I think what swift is calling optional typing is what you'd call monad me maybe... 20:34
Su-Shee masak: well I'd say we don't need to make a fuss about swift feature-wise..
20:34 rurban left
jnthn masak: The language is statically typed, with type inference to ease the burden. 20:34
masak oh, nullable types. :)
yes, those are very nice. 20:35
mr-foobar Su-Shee: I should have used 😭
20:37 martinzaremba left
mr-foobar my half-reading of the swift book tells me they are using the wrong unicode (utf-16) 20:38
FROGGS now I'd like to .split(/\r\n\r\n/) a Buf :/
20:38 martinzaremba joined
Su-Shee mr-foobar: javascript is utf-16 too if I remember correctly. 20:39
masak mr-foobar: if you wish to contribute to Unicode enlightenment, please say "the wrong encoding", and not "the wrong unicode"...
Su-Shee mr-foobar: also, that's the encoding, unicode is the standard/consortium... 20:40
mr-foobar masak: wokay. I kinda think of unicoding as a verb ( as code part is common with encode ) 20:42
20:42 kivutar joined
Su-Shee mr-foobar: don't. unicode comes in many encodings. 20:42
masak what Su-Shee++ said. 20:44
dalek rlito: 65f462d | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js - code cleanup
FROGGS how do I find the position of a sequence of numbers in a list?
mr-foobar europeans :) 20:45
20:45 kurahaupo joined 20:46 kivutar left 20:47 japhb_ joined
Su-Shee mr-foobar: actually, the unicode consortium came from the US.. there's also ISO 10646. 20:49
masak FROGGS: not a single number, but a sequence of them? 20:51
jnthn m: my @a = 1, 3, 5, 9, 42, 69, 2; my @b = 9, 42; say @a.first-index({ @a[state $i .. $i++ + @b] ~~ @b }) + 1 # or something... 20:52
camelia rakudo-moar 68060f: OUTPUT«3␤»
FROGGS masak: yes
20:52 Rotwang left
FROGGS jnthn++ 20:53
jnthn oh, that isn't quite right
FROGGS ohh?
jnthn my @a = 1, 3, 5, 9, 42, 69, 2; my @b = 3, 5; say @a.first-index({ @a[(state $i = -1) .. $i++ + @b] ~~ @b }) 20:54
m: my @a = 1, 3, 5, 9, 42, 69, 2; my @b = 3, 5; say @a.first-index({ @a[(state $i = -1) .. $i++ + @b] ~~ @b }) 20:55
camelia rakudo-moar 68060f: OUTPUT«1␤»
jnthn m: my @a = 1, 3, 5, 9, 42, 69, 2; my @b = 1, 3, 5; say @a.first-index({ @a[(state $i = -1) .. $i++ + @b] ~~ @b })
camelia rakudo-moar 68060f: OUTPUT«0␤»
FROGGS I like it :o)
jnthn Previous one got the wrong answer if it was at the start of the array 'cus I compensated for the off-by-one wrongly.
FROGGS yeah, I see it
now I can fake up my Buf splitting :o) 20:56
thanks
21:01 dayangkun left
dalek rlito: a5b388d | (Flavio S. Glock)++ | / (3 files):
Perlito5 - js - code cleanup
 21:03
21:05 rurban joined, lizmat left 21:06 skids left 21:07 kurahaupo left 21:11 firnsy joined, firnsy left, firnsy joined 21:17 dayangkun joined 21:18 martinzaremba left 21:19 martinzaremba joined
masak 'night, #perl6 21:28
jnthn 'night, masak 21:31
21:36 lizmat joined 21:38 martinzaremba left 21:40 treehug88 left 21:41 martinzaremba joined 21:42 martinzaremba left 21:43 rurban left 21:51 woolfy joined 21:56 LLamaRider joined 21:58 rurban joined, martinzaremba joined 21:59 SevenWolf joined 22:04 hoverboard joined 22:05 donaldh joined 22:11 ajr_ left 22:12 pippo left 22:13 virtualsue joined 22:15 spider-mario left 22:18 LLamaRider left 22:22 anaeem1 left
donaldh jnthn: I fixed the Supply more/tap race github.com/rakudo/rakudo/pull/274 22:25
jnthn: you already had code for a paused Supply
22:25 virtualsue left
jnthn donaldh: oh, but I don't like/want that solution :( 22:26
donaldh oh.
Well it corrects the current solution :)
jnthn Yeah, but it adds a dependency on something I don't think is the right way to solve the problem...
donaldh okay. np 22:27
lizmat could we at least pull this? so at least this solution doesn't race ?
jnthn Knowing that fixes it for you is worthwhile, though...since it means the thing I have in mind (Supply.on_demand(-> $s { ... }) will do it.
donaldh I didn't realise you had bigger change planned.
lizmat so, shall I pull it ? For now ? 22:28
jnthn lizmat: So I can revert it and replace it in a day or two's time?
donaldh It allows me to get on with writing spectests for Async io
jnthn donaldh: True, though you could do that with it applied locally :)
donaldh Yep. Happy to.
22:29 BenGoldberg joined
lizmat if it's only a day or two, then by all means, don't let me stop you :-) 22:29
jnthn Yeah. I have a full day for Perl 6 things on Thursday so can certainly get to it then.
lizmat cool 22:30
donaldh jnthn++
22:33 denis_boyun_ left 22:34 denis_boyun_ joined
dalek kudo/nom: a8798f7 | (Elizabeth Mattijsen)++ | src/core/Process.pm:
Don't bother trying to set $*USER/GROUP on Windows
 22:36
22:37 martinzaremba left 22:42 hoverboard is now known as water 22:43 PZt joined 22:49 dmol joined 22:54 skids joined, rurban left
jnthn 'night, #perl6 22:56
lizmat good night jnthn 22:57
donaldh 'night
donaldh also
lizmat night donaldh !
22:57 donaldh left 23:03 dayangkun left 23:06 woolfy left 23:07 denis_boyun_ left 23:10 water left, btyler left 23:16 ugexe_ joined, Gothmog_1 joined, Tene_ joined, Tene_ left, Tene_ joined, erdic_ joined 23:20 dylanwh_ joined, clkao left, Gothmog_ left, Gothmog_1 is now known as Gothmog_, erdic left, integral left, djanatyn left, dylanwh left, dylanwh_ is now known as dylanwh, erdic_ is now known as erdic, djanatyn joined, integral joined, integral left, integral joined 23:22 clkao joined 23:23 water joined 23:25 water is now known as hoverboard 23:31 BenGoldberg left 23:33 BenGoldberg joined
timotimo well, in the next few hours i'll sleep, but maybe i'll write the p6weekly after that 23:35
23:38 rurban joined
lizmat gnight timotimo 23:39
lizmat will go to bed soon as well
colomon \o 23:40
lizmat sleep& 23:46
23:46 lizmat left
timotimo sergot: i'm not saying you should be the one to do it, but HTTP::Message could implement the hash interface so that we can have {'Vary'}:delete for example 23:50
or say $my_message<Vary Date Content-Type>:kv
it may be as easy as putting "has %!headers handles "postcircumfix:<{ }>" or something 23:52
er, with proper quotes of course
23:56 hoverboard left 23:57 BenGoldberg left 23:59 lizmat joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!