»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!
Set by sorear on 4 February 2011.
tadzik leontopod: I'm quite sure you can 00:01
istr a pal showing such thing on a PM meeting 00:02
leontopod cool 00:05
00:21 PacoAir left 00:25 nimiezko left 00:30 lisa_schwarzhaar left
timotimo is there any protocol to uploading a new pdf snapshot of the perl 6 book? 00:39
00:41 lestrrat left
timotimo i get lots of kpathsea: Invalid fontname '*' contains ' ' 00:42
leontopod timotimo, torrents?
00:42 lestrrat joined
timotimo hm, i meant something like "what to do before and after doing that" 00:43
also: this pdf has only 5 pages :\
probably due to missing fonts 00:44
i have to install acroread-data and 9base? 00:45
didn't help. i give up. someone with a functioning environment can do it some time 00:46
make html works, but isn't very pretty to look at 00:47
jnthn sleep & 00:59
01:05 scott__ joined 01:10 benabik joined 01:20 lisa_schwarzhaar joined 01:22 noam_ joined 01:23 skeebitt22 joined
skeebitt22 hello all 01:23
01:26 skeebitt22 left
whiteknight every time jnthn posts about his progress on Rakudo, I find myself thinking: "I wish all these features were in Parrot core" 01:27
01:36 whiteknight left 01:38 thou left 01:47 whiteknight joined 01:49 retupmoca joined 01:51 zpmorgan joined
sorear good * #perl6 01:52
zpmorgan .u �
phenny U+FFFD REPLACEMENT CHARACTER (�)
01:53 retupmoca left 01:56 retupmoca joined 02:25 preflex_ joined, skids joined, preflex left, preflex_ is now known as preflex
sorear colomon: yeah, user-defined operators are hosed, and I'm struggling to figure out why 02:37
skids nom: role M { multi method Str (Int:D $self:) { "OHAI" } }; my $a = 1 but M; $a.say; 02:40
p6eval nom 828e4d: OUTPUT«Ambiguous dispatch to multi 'Str'. Ambiguous candidates had signatures:␤:(Int:D , Mu %_!)␤:(Int:D $self, Mu %_!)␤␤␤»
skids so being defined in a role doesn't give that "inner or derived scope"? 02:42
Or I guess Int's .Str is similarly defined... hmm. 02:44
sorear I don't think "inner or derived scope" affects MMD in Rakudo yet
skids Ah.
02:47 xinming left 02:50 cooper left 03:03 unobe left 03:05 unobe joined, lestrrat left 03:07 lestrrat joined 03:09 whiteknight left 03:24 yertalert joined 03:32 xinming joined 03:35 wolfman2000 joined
colomon sorear: o/ 03:36
03:40 orafu left, orafu joined
colomon niecza: say 'a' ∈ <a b c d e> 03:47
p6eval niecza v14-32-g760a302: OUTPUT«True␤» 03:48
colomon sorear: I feel like an idiot for not figuring out what the problem was with that this morning... looked repeatedly at non-class code trying to access a private attribute.
TimToady: btw, it does seem to work now. now we need a honking large test file for it... ;) 03:49
sorear++
bedtime
sorear to be fair, the error message is quite LTA
sleep well
03:53 zpmorgan left
retupmoca aww, NativeCall doesn't support callbacks :( 04:04
TimToady so fix it! :) 04:07
retupmoca I will, if I can figure out how
skids Is there a way for a role to reject being runtime mixed into something. Like e.g. a role that does not like to be mixed into odd numbers? 04:16
sorear are you sure you don't want to write a subtype instead? 04:20
dalek ecza: 20f241a | sorear++ | src/NieczaFrontendSTD.pm6:
Dehose and slightly optimize user-defined operators
 04:22
skids contemplates that. Meanwhile JOOC:
std: role A[Int $self: ] { } 04:23
p6eval std 48335fc: OUTPUT«ok 00:01 110m␤»
04:23 Su-Shee_ joined
TimToady STD does not check anything on that level 04:24
sorear skids: $self will always be an undefined value 04:25
roles are mixed into types, not values
I think whatever you're doing, you're doing very very wrong
skids probably. 04:26
04:26 Su-Shee left
skids So basically a role never has any idea as to the type of thing it is being applied to, other than it can supply method longnames with different invocants... 04:31
(above I was more aiming at rola A [::T $self:] and hoping it might mean "undef of type I'm being applied to") 04:32
sorear You can already access that via ::?CLASS 04:33
skids Yes, what I was wondering really was was that was semantically meaningful, just out of curiosity (JOOC). 04:37
subtype doesn't seem to graft to the possible usage mode I was exploring very well, no. 04:46
nom: role A { "hi".say; }; 1 but A; 2 but A; # is that specced to happen on each mixin? 04:55
p6eval nom 828e4d: OUTPUT«hi␤hi␤»
dalek ecza: 1b3d1e8 | sorear++ | src/NieczaFrontendSTD.pm6:
Reenable overriding predefined operators with different precedence, oops
 04:56
skids I thought stuff like that was supposed to only run once.
doy well, it has to run each time if the role has parameters, at least 04:57
not sure whether or not it makes sense to have a special case for roles without parameters 04:58
sorear one of the odder features of new-model roles is that they take the class as a parameter
not the base class - the nascent class
doy that is odd 04:59
sorear so the role has to be reinvoked for each new class created, even without explicit parameters
arguably, 'but' should be doing some caching to not create lots and lots of "Int but A" classes
skids Is that just to be considered an implementation side-effect, or reliable behavior going forward? 05:01
TimToady we are deliberately prejudicing things in favor of immutable semantics, so modifying things in place is a second-class citizen 05:02
we want implementations to freely cache identical mixins and not duplicate anon classes 05:03
05:04 Chillance joined
skids but once per class/class-parameter-combo, we could do an "if $*CLASS ~~ /.../ warn(...)" or something in there? 05:05
TimToady Perl 6 is fundamentally FP more than fundamentally OO, so we prefer things to be stateless whenever possible
we also prefer the more generic code not to play guessing games with less generic code 05:06
because the generic code will never be generic enough to anticipate all usages 05:07
wanting to do so generally indicates a design smell
05:08 birdwindupbird joined
TimToady certainly if the generic code must have some invariants, that's possible to enforce with things like PRE/POST though 05:08
sorear I suspect that skids is trying to use roles for something fundamentally different from genericity 05:09
TimToady we may eventually end up with a COMPOSE phaser, but I suspect it will be frequently abused for bad purposes
skids really I'm just exploring options at this point. 05:11
sorear what sort of options? perhaps we can help 05:13
TimToady ain't we all, but it helps to understand the underlying philosophy, and go with it rather than against it
skids Sure. I wouldn't want to produce frankencode, nobody would want to use it. 05:14
sorear: "How is this going to look to the user" and "what about the user who prefers to do things this way?" types of options.
TimToady that tends to shade into the user that says "I only learned to think this one way, so why can't I use this hammer on everything?" :) 05:16
05:16 Khisanth left 05:17 retupmoca left
skids Sure, but also there's merit to preserving tmwtdi :-) 05:17
TimToady yes, but "more than one" is not the same as "all" 05:18
sorear ffff
I have a non-bisectable failure in S06-operator-overloading/sub.t 05:19
TimToady fortissississimo
sorear tries updating mono
05:20 Patterner left 05:27 lisa_schwarzhaar left 05:33 Khisanth joined
leontopod fortissimo is a great jug wine 05:34
05:35 Psyche^ joined
tadzik it's italian for 'very loud', iirc 05:35
05:35 Psyche^ is now known as Patterner
moritz colomon: yes, the prisoner's harness is *the* test harness 05:42
05:44 snearch joined
moritz perl6: my Num $s;for 1e0..1e3 -> Num $i {$s+=1e0/$i**2};say $s 05:46
p6eval niecza v14-34-g1b3d1e8: OUTPUT«1.6439345666815615␤» 05:47
..rakudo 828e4d: OUTPUT«Nominal type check failed for parameter '$i'; expected Num but got Int instead␤␤»
..pugs b927740: OUTPUT«1.6439345666815598031390580238222155896521␤»
moritz submits rakudobug
nom: for 1e0 .. 1e0 { say .WHAT }
sorear perl6: say pi**2/6
p6eval nom 828e4d: OUTPUT«Int()␤»
pugs b927740: OUTPUT«1.6449340668482261867549189638081666666667␤»
..niecza v14-34-g1b3d1e8: OUTPUT«1.6449340668482264␤»
..rakudo 828e4d: OUTPUT«1.644934063089␤»
sorear #optimized that for you
sorear builds a new mono to better investigate the segfaults in sub.t 05:49
moritz nom: for 1.1e0 .. 2.1e0 { say .WHAT } 05:50
p6eval nom 828e4d: OUTPUT«Num()␤Num()␤»
moritz nom: for 1e0 .. 2e0 { say .WHAT }
p6eval nom 828e4d: OUTPUT«Int()␤Int()␤» 05:51
05:57 kaleem joined 06:06 Patterner left 06:07 jj__ joined 06:09 jj__ left 06:10 xinming left, xinming joined 06:12 ponbiki joined 06:15 Psyche^ joined 06:16 Psyche^ is now known as Patterner 06:22 thou joined
moritz nom: say (1e0 .. 2e0).perl 06:23
p6eval nom 828e4d: OUTPUT«1e0..2e0␤»
leontopod is Haskell actually used to DO anything? 06:24
moritz supposedly, yes 06:26
bonsaikitten leontopod: every now and then, yes
moritz xmonad is used quite a bit 06:27
en.wikipedia.org/wiki/Haskell_%28pr...plications 06:28
06:35 woosley joined 06:38 snearch left 06:46 Timbus|Away joined, zby_home_ left 06:48 jnthn_ joined 06:49 clkao_ joined, timotimo_ joined, huf_ joined, broquain1 joined, lumi____ joined 06:50 Timbus left, Timbus|Away is now known as Timbus, lumi___ left, aloha left, jnthn left, broquaint left, timotimo left, clkao left, yertalert left 06:51 aloha joined 06:52 araujo left, hillu left, hillu joined, araujo joined 06:54 takesako left, huf left 06:55 takesako joined, broquain1 left, wtw joined 06:56 thou left, birdwindupbird left, tty234 left, DarthGandalf left, PerlJam left, birdwindupbird joined, orafu left, avar left, _jaldhar joined 06:57 Su-Shee_ left, betterworld left, [Coke] left, [Coke] joined, broquaint joined, betterworld joined, jaldhar left, Trashlord left, Tedd1^ joined, PerlJam joined 06:58 avar joined, avar left, avar joined, tty234 joined, chra left, Tedd1 left, orafu joined, Su-Shee joined, REPLeffect left 06:59 REPLeffect joined, chra joined 07:02 araujo left, sudokode left, Gothmog_ left, noam_ left, _ilbot left, slavik left, c1sung left, djanatyn left, eviltwin_b left, scott___ joined 07:03 scott__ left, noam_ joined 07:04 preflex left 07:05 REPLeffect left, REPLeffect joined 07:06 preflex joined 07:07 araujo joined, Gothmog_ joined, _ilbot joined, slavik joined, c1sung joined, djanatyn joined, eviltwin_b joined, _ilbot left, c1sung left 07:08 Gothmog_ left, Gothmog_ joined, c1sung joined, thou joined 07:09 sudokode joined 07:11 DarthGandalf joined
sorear fff. new master mono doesn't work *at all* with niecza 07:13
07:14 drbean_ joined, _ilbot joined, y3llow_ joined, kaleem left, japhb_ joined 07:15 jrockway_ joined, sunnavy_ joined, zamolxes_ joined, jevin_ joined, Maddingu1 joined, gfldex_ joined, Vlavv` left, x3nU left, zamolxes left, jrockway left, rafl left, cj left, y3llow left, gfldex left, sunnavy left, Maddingue left, x3nU__ joined, rafl joined 07:16 y3llow_ is now known as y3llow, Vlavv` joined, drbean left, japhb left, jevin left, cj__ joined, prammer left, drbean_ is now known as drbean, flussence_ joined, prammer joined 07:17 ranguard left, mikemol left, flussence left 07:18 ranguard joined 07:19 simcop2387 left, kst`` joined 07:21 rsimoes2 joined 07:22 simcop2387 joined, pochi joined 07:23 mikemol_ joined, xinming left, tty234_ joined 07:24 mikemol joined, xinming joined, x3nU__ left, pochi_ left, jlaire left, mikemol left, cognominal_ joined, gfldex__ joined, ponbiki left
benabik There's a local shop that does all their work in Haskell. I think they do financial forecasting and the like. (Or maybe that was two people who were on campus the same day. Not sure.) 07:25
07:25 leontopod left, leontopod joined 07:26 hillu_ joined, tty234 left, rsimoes left, gfldex_ left, ab5tract left, hillu left, ab5tract joined 07:27 thou left, cognominal left, wolfman2_ joined 07:28 arnsholt_ joined, [particle]1 joined, y3llow_ joined, kst` left 07:29 jlaire joined 07:30 athomason_ joined 07:31 nsh_ joined, sorear_ joined, awwaiid_ joined, szabgab_ joined
sorear_ yes, Haskell is quite popular with high finance. 07:31
they are not the only user though.
07:32 scottp_ joined, sorear left, sorear_ is now known as sorear, Juerd_ joined, pmichaud_ joined, rhr_ joined, uno_be joined 07:33 betterwo1ld joined, x3nU joined, wolfman2000 left, orafu left, awwaiid left, nsh left, ocharles left, athomason left, y3llow left, Khisanth left, Juerd left, bacek left, arnsholt left, rhr left, pmichaud left, scottp left, szabgab left, renormalist left, betterworld left, [particle] left, unobe left, jfried left, bacek joined, Juerd_ is now known as Juerd 07:34 ocharles- joined, jfried joined, x3nU is now known as Guest6426, OuLouFu joined, OuLouFu left, OuLouFu joined, OuLouFu is now known as orafu, ocharles- is now known as ocharles, y3llow_ is now known as y3llow 07:37 xinming left 07:40 xinming joined 07:44 Khisanth joined 07:46 pothos_ joined 07:47 d4l3k_ joined, sunnavy joined 07:48 hillu joined, skids_ joined, bluescreen100 joined, jaldhar_ joined 07:49 hillu_ left, sunnavy_ left 07:50 dalek left, lumi____ left, pothos left, lumi___ joined, skids left, bluescreen10 left, Grrrr left, _jaldhar left, _ilbot left, pnu_ left, pnu joined, xinming_ joined, _ilbot joined, pothos_ is now known as pothos, p6eval left 07:51 xinming left 07:52 d4l3k_ left 07:53 dalek joined, ChanServ sets mode: +v dalek, simcop2387 left 07:54 simcop2387 joined, dalek left 07:55 p6eval joined, dalek joined, sorear left, sorear joined 07:56 drbean left, ChanServ sets mode: +v p6eval, diakopter left, diakopter joined, ChanServ sets mode: +v dalek, pmichaud_ left, pmichaud joined, colomon left 07:57 Woodi left 07:58 Woodi joined 08:00 Grrrr joined 08:02 drbean joined 08:10 thou joined 08:23 sunnavy left 08:24 sunnavy joined 08:26 ab5tract left, mj41 joined
Su-Shee good morning everyone. 08:29
08:29 Woodi left, rhr_ left, szabgab_ left, awwaiid_ left, nsh_ left, tty234_ left, zamolxes_ left, c1sung left, Gothmog_ left, DarthGandalf left, timotimo_ left, jnthn_ left, woosley left, Patterner left, sftp left, yeltzooo left, nebuchadnezzar left, eiro left 08:32 rhr joined
moritz \o 08:32
08:32 sftp joined 08:35 woosley joined, timotimo joined, nebuchadnezzar joined, eiro joined, Woodi joined, szabgab_ joined, awwaiid_ joined, DarthGandalf joined, c1sung joined, Gothmog_ joined, jnthn_ joined, AAAA9AI joined, Patterner joined, yeltzooo joined, tty234 joined, szabgab_ left, DarthGandalf left, tty234 left, tty234 joined, szabgab joined, tty234 left 08:37 woosley left, DarthGandalf joined 08:38 nsh joined 08:39 zamolxes joined, tty234 joined 08:40 ab5tract joined, woosley joined, Patterner left, Psyche^ joined, Psyche^ is now known as Patterner 08:41 am0c joined 08:42 nebuchad` joined, eiro left, Woodi left, awwaiid_ left, c1sung left, Gothmog_ left, jnthn_ left, yeltzooo left 08:44 d4l3k_ joined, nebuchadnezzar left 08:45 yeltzooo joined, Gothmog_ joined, awwaiid joined, Woodi joined, c1sung joined, dalek left, d4l3k_ is now known as dalek, ChanServ sets mode: +v dalek
sorear o/ moritz, Su-Shee 08:46
08:51 simcop2387 left, pothos left, Khisanth left, jfried left, ocharles left, cj__ left, apejens left, BooK_ left, sirhc left, frettled left, buubot_backup left, krunen left, bbkr1 left 08:52 clkao joined, woosley left, scottp joined 08:53 buubot_backup joined, Trashlord joined 08:54 dukeleto left, scottp_ left, clkao_ left, mikemol_ left 08:55 dukeleto- joined, pothos joined, Khisanth joined, jfried joined, ocharles joined, apejens joined, BooK_ joined, frettled joined, sirhc joined, krunen joined, simcop2387 joined, pothos left, dukeleto- is now known as dukeleto 08:56 pothos joined 08:58 mikemol joined 08:59 sorear_ joined, PerlPilot joined, nebuchad` is now known as nebuchadnezzar 09:00 cj joined 09:01 domidumont1 joined 09:02 sorear left, PerlJam left, am0c left, Chillance left, Juerd left, domidumont left 09:03 Chillance joined 09:04 am0c joined, preflex_ joined, preflex left, preflex_ is now known as preflex 09:08 slavik1 joined 09:09 _jaldhar_ joined, tarch joined, geekosaur joined 09:10 sunnavy_ joined, sftp_ joined, gfldex joined 09:11 timotimo_ joined, pothos_ joined 09:12 GlitchMr joined, slavik left, eviltwin_b left, jnthn joined, sunnavy left, jrockway_ left, sftp left, araujo left, timotimo left, jlaire left, jaldhar_ left, y3llow left, pothos left, DarthGandalf left, gfldex__ left, sorear_ left, y3llow_ joined, jrockway joined 09:13 jlaire joined, pothos_ is now known as pothos, araujo joined 09:14 y3llow_ is now known as y3llow 09:16 sorear joined, DarthGandalf joined 09:18 sunnavy joined, mj41_nb joined 09:20 Maddingu1 is now known as Maddingue 09:21 arnsholt_ is now known as arnsholt, sunnavy_ left, mj41 left, bacek left 09:25 pernatiy left 09:26 bacek joined, AmazingPudding joined 09:30 revdiablo left, bonsaikitten left, rsimoes1 left 09:31 revdiablo joined, revdiablo left, revdiablo joined, dakkar joined 09:35 AmazingPudding is now known as bonsaikitten 09:36 zamolxes_ joined, diakopte1 joined 09:37 diakopter left, c1sung left, zamolxes left 09:39 rsimoes1 joined 09:40 c1sung joined 09:43 nsh left 09:44 PerlPilot left, Trashlord left, japhb_ left, PerlJam joined, nsh joined 09:45 diakopte1 left, japhb_ joined 09:46 REPLeffect left 09:48 szabgab left, szabgab joined 09:49 REPLeffect joined
dalek kudo/nom: 3bfaa06 | moritz++ | src/core/Range.pm:
Range iteration fixes

  * allows big integer ranges
  * numifies the upper limit if the lower one is numeric		 09:52
kudo/nom: e035eed | moritz++ | src/core/Range.pm:
Num ranges now produce Num lists
09:53 diakopter joined 09:57 Trashlord joined 09:58 daxim joined
dalek ast: 53c3cd2 | moritz++ | S03-operators/range.t:
test ranges of bigints and nums
 09:59
10:00 huf_ is now known as huf 10:05 szabgab_ joined 10:06 szabgab left, PZt left
masak good antenoon, #perl6 10:31
10:33 PZt joined
masak [backlog] heh, I'd hate to be remembered as "the guy who used Perl 6 in production". :P 10:33
put Perl 6 to the test people! all of you!
s/test/test,/
also, my the first project that I ever planned to write in Perl 6 was actually a supercool RPG game. 10:35
kind of a framework for scriptable adventure game worlds. 10:36
10:37 kst`` is now known as kst
masak also, in the department of "sentences no-one could predict": "I can code C/C++ already and my hair is sacred." :) 10:37
10:39 slavik1 left 10:40 slavik joined 10:41 xinming_ left, xinming joined 10:46 xinming left
snarkyboojum some uses of Perl 6 in production might imply production Haskell use! :O 10:52
10:52 bacek left, xinming joined 10:53 bacek joined
snarkyboojum but seriously, I've used both Haskell and Perl 6 in production, though that could just a sign that I shouldn't be given prod access ;) 10:53
masak right tool for the job. that's all there is to it. 10:56
snarkyboojum someone else say something, quickly!
ah - masak to the rescue
though "right tool" isn't necessarily simple either ;) 11:00
11:00 Juerd joined 11:02 xinming left 11:04 xinming joined 11:08 MayDaniel joined
masak indeed, it isn't. 11:09
but it tells us something about the needs to keep up-to-date on tools :)
I gave a Perl 5 course in January. in preparing the material, I learned a lot of little tidbits about tools. 11:10
11:11 snearch joined 11:13 lestrrat left 11:14 lestrrat joined
dalek kudo/nom: a61dbd5 | moritz++ | tools/build/NQP_REVISION:
bump NQP revision
 11:26
11:26 scott___ left
dalek ast: d93480e | moritz++ | S0 (2 files):
two rakudo unfudges
 11:33
11:38 lestrrat left 11:39 Home joined, lestrrat joined
Home rakudo: say "test" 11:40
p6eval rakudo e035ee: OUTPUT«test␤»
Home rakudo: say <STDIN>
p6eval rakudo e035ee: OUTPUT«===SORRY!===␤Unsupported use of <STDIN>; in Perl 6 please use $*IN.lines (or add whitespace to suppress warning)␤at /tmp/kbkZxc4TNg:1␤»
11:41 Home left, snearch left, snearch joined
moritz welcome Home! 11:43
masak Home just wanted to test p6eval and our excellent error messages; he already left, fully satisfied. 11:45
moritz I know he left, I just couldn't resist the pun :-) 11:46
11:46 lestrrat left
masak I know :) 11:46
11:48 snearch left 11:49 snearch joined, lestrrat joined 11:59 mtk left 12:02 lestrrat left 12:03 lestrrat joined 12:04 Bucciarati left 12:13 bluescreen100 left 12:15 pernatiy joined 12:16 sftp_ left 12:25 not_gerd joined
not_gerd hello, #perl6 12:25
phenny not_gerd: 08 Feb 21:51Z <jnthn> tell not_gerd that's up to the REPR to do. See P6opaque - it keeps it around in its REPR_data.
12:25 bluescreen100 joined
[Coke] yawns 12:33
12:42 _jaldhar_ left 12:45 xinming left 12:47 xinming joined
moritz jnthn: fwiw the 'bs' branch in nom fails to build here 12:51
In file included from serializationcontext.c:29:
../6model/sixmodelobject.h:143: error: redefinition of typedef ‘STable’
../6model/serialization.h:49: note: previous declaration of ‘STable’ was here
12:52 bbkr left 12:54 flussence_ is now known as flussence
jnthn moritz: Curious - mine doesn't so much as warn me, let alone error. 13:00
not_gerd jnthn: typedefs are semantically definitions, and re-definitions are not allowed (if they are not tentative) 13:03
jnthn not_gerd: Where by tentative you mean? 13:06
oh, I see 13:07
duh
not_gerd int x; int x = 5; is legal as long as you only assign once
it's a technical term from the C standard
jnthn ok
dalek p/bs: 686b47b | jnthn++ | src/6model/sixmodelobject.h:
Fix typedef redefinition that busted the build on some compilers (moritz++ for reporting).
 13:08
not_gerd jnthn: gist.github.com/1789578 is what I'm using...
moritz nom: class A { $.x handles <Numeric> = 5; method x() { return $!x * 2 } }; say A.new.Numeric 13:10
p6eval nom a61dbd: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2␤»
moritz nom: class A { $.x handles <Numeric>; method x() { return $!x * 2 } }; say A.new(x => 5).Numeric 13:11
p6eval nom a61dbd: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2␤»
jnthn Missing "has"
moritz nom: class A { has $.x handles <Numeric>; method x() { return $!x * 2 } }; say A.new(x => 5).Numeric
p6eval nom a61dbd: OUTPUT«5␤»
moritz jnthn++
so the attribute takes care of the 'handles', not the accessor
jnthn Correct. 13:12
moritz makes sense if you think about it, because it means you can delegate to private attributes too :-)
jnthn Well, the generated delegation method looks up the attribute
Pretty sure traits.pm has the implementation of handles
moritz nom: multi a($x) { }; multi a(*@a) { }; say a(3) 13:14
p6eval nom a61dbd: OUTPUT«Nil␤»
13:22 MayDaniel left
dalek ast: 3a7fb4f | moritz++ | S12-class/anonymous.t:
RT #80024, anon class is anon
 13:26
13:26 tokuhirom joined
moritz ok, I have a feature request 13:28
I want both 'CATCH' and 'default' to take statements too, not only blocks
you can write try foo(); without a block 13:29
but sometimes I also want to write CATCH default $died = 1;
or similar stuff
jnthn Thing is that you expect CATCH to get $_ to the exception. And if there's no block there it could be a little misleading. 13:33
s/get/set/
Maybe that's not a fatal issue
Are there use cases for CATCH without a block other than using default? 13:34
moritz sure 13:36
CATCH when X::JustOneException { ... }
but I don't think topicalizatioin is a problem; after all smart-matching is also a mini-topicalizer 13:37
jnthn True
moritz as is statement_control:sym<when>
erm, s/when/given/
though 'when
is too
or not, dunno. I forgot :-)
13:39 xinming left, skids_ left
masak CATCH, being a phaser, looks perfectly normal in expression form. 13:39
but default doesn't, to my eyes. 13:40
moritz well, because your eyes aren't yet used to it
masak of course.
but I believe it's also indicative of a real pattern.
13:40 colomon joined
masak default belongs to a group of things that always take a block. 13:40
CATCH doesn't.
13:40 xinming joined
masak I just wanted to call attention to that through my bias :) 13:40
jnthn Yeah, default does bother me a tad more than CATCH 13:41
moritz well, I request both and get one. That's not too bad by #perl6 feature request standards :-) 13:42
masak :D
we're here to help. :)
CATCH say "oops!" when X::JustOneException; 13:43
should be OK, right?
it reads extremely well.
moritz but won't work 13:44
masak wrong nesting?
moritz because statement-modifying 'when' doesn't succeed the block
masak oh!
moritz thus the 'oops!' will be printed, and then the exception is rethrown
masak ...unless that's what you want... 13:45
13:48 thou left, xinming left, xinming joined 14:01 ggoebel left 14:06 ggoebel joined 14:08 JimmyZ joined 14:14 not_gerd left
colomon moritz: okay, I have a strategy which is slightly better than tit-for-tat. :) 14:16
14:16 ggoebel left 14:17 ggoebel joined, diakopter left, diakopter joined 14:19 REPLeffect left
moritz colomon: submit! submit! 14:20
colomon I'm not sure I should before the last possible minute. ;)
moritz you should
you can also replace your submission with later submissions, if you want
colomon I'm more concerned with keeping my techniques secret. ;) 14:21
14:23 szabgab_ left, szabgab joined 14:24 Chillance left 14:25 Chillance joined 14:31 xinming left 14:32 sftp joined
moritz well, I won't publish any submissions before the deadline, and my own strategies are already written and their SHA1 published 14:32
14:32 REPLeffect joined 14:33 xinming joined 14:35 [particle]1 is now known as [particle] 14:38 timotimo_ is now known as timotimo 14:39 packetknife joined
tadzik friday, friday! 14:39
14:41 PacoAir joined
moritz TCIF indeed! 14:42
tadzik TCIF? 14:44
moritz Thank Chrunchy It's Friday
tadzik heh
it also means the end of the exam session \o/
moritz \o/ time for tadzik++ to write a Prisoner's Dilemma strategy! 14:45
tadzik yay!
[Coke] , running a niecza auto-un-fudge, realizes he has another fudge patch he never pushed. whoops. 14:47
[Coke] will see if it gets any bigger first.
uno_be win 6 14:52
14:59 xinming left, fasad joined 15:00 xinming joined 15:01 MayDaniel joined
etneg_ checks if niezca has a site 15:04
etneg_ sits back and relaxes till one's up
tadzik niecza 15:05
moritz etneg_: anything missing from github.com/sorear/niecza ? 15:06
etneg_ doesnt look like it 15:07
is that a trick question 15:09
moritz no, it's a real question 15:11
the question is if we have a good reason to build and maintain a website for niecza
or if the github page is enough
etneg_ i think so
didnt you guys say it was like totally awesome or something
heh
sorry i dont code:D
moritz it *is* totatlly awesome 15:12
you should really start :-)
etneg_ well considering it's on par with rakudo and rakudo has a website 15:13
i'd think niecza should go in the same direction
my 2$ opinion 15:14
moritz thanks for your opinion 15:16
etneg_ and reading more, it lookslike it's being compared to raudo
rakudo
so certainly a website 15:17
moritz I also think it needs a website to be taken seriously, but I'm not (yet?) ready to volunteer to make one
etneg_ possibly a wiki page would be a good start 15:18
moritz somehow I have bad experience with wiki pages 15:21
15:21 tokuhirom left
etneg_ well it's a start 15:21
:D
moritz it's what you do if you create a page to then abandon, hoping that the rest of the world will maintain it
[Coke] accidentally makes feather REALLY slow for a litle while. 15:22
*little
flussence just wrote a completely nonsensical PD strategy which does only slightly worse than tit-for-tat 15:25
dalek ast: 702aa34 | coke++ | S (14 files):
niecza auto-unfudge
15:38 sftp_ joined 15:39 sftp left 15:43 Psyche^ joined 15:47 Patterner left, Psyche^ is now known as Patterner, MayDaniel left 16:00 birdwindupbird left
leontopod I think pushing perl 6 in a functional programming direction is a disaster. 16:01
You are going to push it right into academia.
16:05 am0c left
TimToady that's only one of the places we're going to push it :P 16:05
flussence
.oO( nobody uses functional languages! just look at how SQL flopped! )
 16:08
16:08 wtw left
TimToady but the P still stands for Practical and Pathological 16:08
16:09 Trashlord left
TimToady you fundamental mistake is to assume that if something is better for academia, it must necessarily be worse for everything else, and this is...not yet proven... :) 16:10
*your 16:11
daxim sql is counted chiefly among the declarative languages, not the functional ones
leontopod Is that really you, Larry? 16:12
wow, I am talking to Larry Wall! =)
TimToady I'll never tell...
daxim: declarative and functional are not exclusive concepts in my head 16:13
daxim certainly, but I was surprised to see sql named as an example of a non-floppy functional language 16:14
TimToady sure, and dduncan++ shows up here regularly to point out how floppy SQL is compared to whatever he's calling his new language these days :) 16:16
16:18 JimmyZ left
TimToady anyway, both of those are really mushy concepts because it's really hard to make something that is theoretically pure and also practical, so people compromise on their designs to varying degrees, but still call them "declarative" and/or "functional" 16:18
16:20 domidumont1 left 16:22 Trashlord joined
TimToady Perl 6 is multiparadigmatic, which means it can't be monodogmatic about any of them. 16:22
au still thinks P stands for Polymorphic... 16:23
TimToady :D
[Coke] au: hey!
au /o
TimToady for sure, Perl 6 is becoming more existential, despite having been cursed and recursed for having too many lambdas... :) 16:27
16:28 daxim left
leontopod THUS SPAKE LARRYTHUSTRA 16:28
au well, we did start with a rather cursive syntax in the first place... :) 16:29
TimToady "It's full of asterisks!"
anyway, been there, have the T-shirt, which I still wear fondly from time to time
au *delighted* 16:30
leontopod actually an "It's full of asterisks!" T-shirt would rule
TimToady though I often have to point out that it's an *imaginary* timeline, and the "hack, hack, hack" bit looks more like a long-term graph of the stock market 16:31
16:31 kaare_ joined 16:32 yertalert joined
[Coke] my god. it's full of whatever. 16:32
16:33 thou joined, thou left
leontopod the stock market is a big ponzi scheme 16:33
TimToady still, recent recessions notwithstanding, stock market curves to tend upward over the long term, so I guess you might say I'm bullish on Perl 6 16:34
leontopod maybe Perl 6 is a big ponzi scheme! =D
TimToady shh!
leontopod to suck in more and more programmers!
yertalert lol 16:35
TimToady so far we suck at that :)
16:38 wala joined 16:39 p6eval left, wala left
flussence Perl 6 seems more like building pyramids than a pyramid scheme sometimes... but that just means it'll last 5000 years 16:40
16:41 diakopter left, dalek left 16:42 diakopter joined
TimToady hmm, maybe we need to install some traps for grave robbers... 16:42
16:43 dalek joined, ChanServ sets mode: +v dalek 16:44 MayDaniel joined
TimToady actually, they're already there: the attempt to backport Perl 6 features to Perl 5 has already demonstrated some of those traps :) 16:48
16:49 MayDaniel left
flussence (are they still trying to backport "." and "~"?) 16:50
[Coke] there was a patch for it. 16:52
TimToady dunno; all I know is it's much easier to break everything at once than one piece at a time
and the cumulative pain of evolving Perl 5 into a Perl 6 one bit a time would be much greater than the pain we've had developing Perl 6, and would take much longer 16:54
jnthn evening, #perl6
leontopod Perl 5 is undergoing ferocious development
what are they up to Perl 5.14 now? 16:55
geekosaur most recent release, yeh 16:56
TimToady sure, it's ferocious; nevertheless, it's ferocious like a lion chained to a gatepost, so it doesn't get so far down the road as it otherwise might... 16:57
leontopod Is it true that Perl 5 has preprocessor definitions that are thousands of characters long? 17:01
I love the C preprocessor.
geekosaur has not looked but it would not be uniquein having such
leontopod A simple text preprocessor, why can't we have that for Perl 6? 17:02
jnthn TimToady: I haz a silly question. In "my $x = [+] 1,2,3", what is it that makes the %item_assignment prec limit not cause <initializer> to stop parsing at the first comma, but swallow the rest of the list as the operand to the meta-op?
leontopod Why gussy it all up?
timotimo leontopod: who stops you from using m4 for perl5?
leontopod because m4 doesn't come with Windows?
timotimo neither does perl5
or c++
or c
17:03 p6eval joined, ChanServ sets mode: +v p6eval
yertalert Or anything useful 17:03
leontopod perl -P, baby
jnthn yertalert: Hey, we got Notepad!
leontopod Notepad got some kind of software award. 17:04
yertalert lol
leontopod But they couldn't figure out who actually wrote it.
yertalert It amazes me that so many windows apps have remained unchanged since 3.1
leontopod hey, windows 3.1, trumpet winsock, and netscape navigator was a pretty peppy combination on a 28k modem back in the day 17:05
yertalert But, I do prefer an extremely light weight editor like notepad for just scratching out ideas
That's why I have leafpad on Linux 17:06
leontopod I think Microsoft should give away Windows 98 for free.
yertalert and it has line numbering!
lol, why?
tadzik so we can play games, obviously 17:07
17:07 kaleem joined
benabik I wonder how long you can have a copy of Win98 connected to the internet before it's riddled with viruses. 17:08
moritz does it even come with a network stack? :-)
timotimo i think it was less than 30 minutes the last time i read about that
and that's already quite a long time ago
benabik I think the network stack started being standard with 95...
yertalert They still have 98 viruses? 17:09
timotimo windows 95 is the best windows
because it has hover 950
yertalert I think XP is the best 17:10
moritz likes fvwm best
moritz trolls a bit
yertalert It was a huge scam by MS to see just how long millions of people could go without a modern operating system
moritz nom: say 1.join.perl 17:11
p6eval nom a61dbd: OUTPUT«"1"␤»
leontopod I still have win 2k pro sp4, but its ip address is 10.0.0.4 17:12
it's behind a pretty good linux firewall/nat box
17:13 Tedd1^ left
yertalert lol, linux has to protect windows 17:13
dalek kudo/nom: c3fb7d6 | moritz++ | src/core/ (3 files):
small X::AdHoc refactor

Now it's finally the way I want it to be.
fasad Is perl6's philosophy "Perl 5 done right!" ? 17:16
[Coke] b: say 2012-1998
p6eval b 1b7dd1: OUTPUT«14␤»
[Coke] b: say 2012-1995 ; # whoops
p6eval b 1b7dd1: OUTPUT«17␤»
moritz fasad: it's more like "Perl done right!" 17:17
TimToady righter, or a least righterish 17:19
jnthn: probably the same thing that makes -2**64 treat ** as tighter than the - 17:20
cognominal_ maxtaco.github.com/coffee-script/ # pasted by rgs in #perlfr
TimToady jnthn: it doesn't look like the reduce rule is doing anything special other than setting the precedence to %list_prefix 17:21
17:22 MayDaniel joined
TimToady perl6: say -2**3 17:22
p6eval pugs b927740, rakudo a61dbd, niecza v14-34-g1b3d1e8: OUTPUT«-8␤»
TimToady whew!
cognominal_ TimToady do we have a construct similar to the icecoffeescript "await/defer" in Perl 6? 17:23
…possibly using threads instead. 17:24
fasad ubuntu 17:25
^^ignore that 17:26
TimToady
.oO(ignore ubuntu?)
jnthn TimToady: Yeah, thus my confusion. EXPR is called from initializer with preclim of %item_assignment. It calls termish, which will parse the [+] as a prefix, then <term>, which parses the 1. Then back in EXPR we parse an infixish, which gets the , and this then has $inprec le $preclim as true, so it's considered the last TERM 17:29
Or so I'd expect it to work, and that's how it's happening in Rakudo. Somehow I'm missing a trick in STD. 17:30
17:32 MayDaniel left, MayDaniel joined 17:33 mj41_nb left
jnthn I had wondered if it was an effect of the term-forcing that reduce does, but it'd see that this particular case leaves it as a normal prefix. 17:33
fasad TimToday: :D
jnthn *seem 17:34
17:34 fasad left
jnthn TimToady: Oh. My assumption that STD gets it right because Niecza gets it right may well be wrong. 17:39
TimToady: Niecza has a term:reduce and then parses <args> after it.
17:41 wolfman2_ left
jnthn This is perhaps telling 17:42
std: my $x = [+] 1, $y;
p6eval std 48335fc: OUTPUT«===SORRY!===␤Variable $y is not predeclared (declarators are tighter than comma, so maybe your 'my' signature needs parens?) at /tmp/P7RKd7vEnl line 1:␤------> my $x = [+] 1, ⏏$y;␤Check failed␤FAILED 00:01 112m␤»
jnthn std: my $x = [+] (1, $y);
p6eval std 48335fc: OUTPUT«===SORRY!===␤Variable $y is not predeclared at /tmp/qhkDMu3KQE line 1:␤------> my $x = [+] (1, ⏏$y);␤Check failed␤FAILED 00:01 113m␤» 17:43
leontopod so I installed ubuntu for the very first time about two weeks ago and wouldn't you know it, the new window manager looks like something that runs on a cell phone or pad computer.
jnthn This implies (iiuc) that it's marked declend in MEMOS before it sees the $y
So I'm suspecting STD parses this the way Rakudo's stdinit branch does, which would seem to be not what at least the spectests expect, nor what folks will in general. 17:44
benabik leontopod: Yeah, Unity looks a little odd. Haven't played with it much myself. (I mostly use 10.04 because it's what we have at school (LTS)) 17:45
benabik should talk in less parentheticals.
leontopod That's the way everything is going: cell phones and pad computers
TimToady jnthn: yes, we should probably switch reduce to do the <args> trick
17:46 uno_be left
jnthn TimToady: OK, I'll do what Niecza does, or something close to that. 17:46
TimToady and I have confirmed that viv does in fact misparse it
jnthn TimToady: Phew. I was going nuts trying to figure out what I'd missed on method EXPR :P
s/on/in
[Coke] jnthn: This is the "punch list" of items that need addressing before 17:47
releasing a nom-based distribution.
Does that need to be replaced with a new goal, or just removed?
(from NOMMAP.*)
jnthn [Coke]: Oh, I planned to fold nommap into ROADMAP (if any of it isn't already)
heh, yeah, it also claims that the Order enumeration is missing, which it ain't. 17:49
benabik [Coke]: your nick makes me keep wondering what infix:<Coke> does. :-D
jnthn [Coke]: Will take care of it this evening. 17:50
leontopod Ritalin/Cocaine, Amphetamine/Adderall 17:51
masak au! \o/
[Coke] jnthn++
TimToady jnthn: though it does point out a subtle flaw in EXPR that could perhaps be corrected by noticing that we've passed a precedence loosener and fixing the limiter 17:52
17:53 birdwindupbird joined
masak au: we're a bunch of people who want to kick-start the development of Pugs again, and bring it up to spec. 17:54
last I counted: [Coke], ingy, and me. 17:55
leontopod hi masak =)
masak hi Teratogen =) 17:56
leontopod oh, right
17:56 leontopod is now known as Teratogen, Teratogen left, Teratogen joined
masak both nicks are excellent, by the way. 17:56
Teratogen well, the way I got stuck with Teratogen was
this was back on efnet in 1994
I got invited by Plutonium (an irc operator at the time) to be an op on #sex 17:57
great
masak "leontopod" puts me in mind of dandelions.
Teratogen but I was still trying to figure out a nickname because I was so new to irc
masak "Teratogen" sounds like something that generates Terats.
araujo masak, I could help with testing too ...
:P
Teratogen I saw the word "teratogen" in a novel, I think Jurassic Park
so I started using it
and every time I changed my nick to something else pixiedust would holler at me. 17:58
masak araujo: testing? please elaborate. :)
Teratogen so I ended up using Teratogen more and more.
And now I am kind of stuck with it.
araujo masak, well, with pugs .. at least initially
masak araujo: you're testing Pugs?
Teratogen I am still banned on efnet #perl
it's been a decade, and the ban is still on.
*giggle*
araujo masak, not now .... but i have been checking a bit its code 17:59
17:59 pernatiy left
[Coke] masak: should we install haskell globally on feather? 18:00
18:00 dakkar left
TimToady
.oO( #perl6, the Halfway House of banned personages attempting to re-enter society... :)
 18:01
Teratogen I remember the apostrophic jihad. 18:02
jnthn TimToady: (fixing the limiter) that's kinda what I expected to find STD's EXPR doing :)
Teratogen That is where Tom Christiansen attacked anybody who said dont instead of don't 18:03
im instead of i'm
etc
[Coke] etc. ;)
Teratogen well, yeah, etc.
baw
TimToady Well, Tom is a prescriptivist at heart, but he's mellowed somewhat of late, at least some of the time... 18:04
18:04 thou joined 18:08 not_gerd joined
not_gerd hello again, #perl6 18:08
masak prescriptivism should be outlawed. 18:18
[Coke]: yes, I think we should.
TimToady that would be proscriptivism, not prescriptivism :)
cognominal_ masak, try harder for a self-contradiction: "perscriptivism must be outlawed" 18:19
TimToady though I freely admit that the practitioners of each are deeply intertwingled :)
18:20 birdwindupbird left
[Coke] moritz: hey, can you install the tools we need to build pugs.hs on feather? ;) 18:25
18:26 Tedd1 joined 18:30 birdwindupbird joined 18:32 kaleem left, kaleem joined
Su-Shee today, I had a horrible experience. please hurry with perl 6. I never ever want to experience working with java and eclipse again. 18:32
[Coke] su-shee; I do that. and also cold fusion! 18:34
Su-Shee [Coke]: I'm honestly lost for words after that experience.
[Coke] as with many things, it gets easier with practice. 18:35
I like the eclipse bit, actually, as I can afford not to remember all the java. 18:36
Teratogen Will Java truly ever be completely FOSS?
I have a funny feeling that the guy who runs Oracle is going to keep that from happening.
Su-Shee [Coke]: I simply avoided Java until today and then decided to give Android app development a try.. 18:37
Teratogen Su-Shee, yes, I want to develop android apps in Perl please!
Su-Shee I think I try again when JS/HTML5 is the main stack.. 18:38
not_gerd Teratogen: possibly, you can do that right now - there's Mono for Android... 18:39
Rakudo on Android is probably not going to happen any time soon, though - the Parrot build system can't really deal with cross-compilation 18:41
18:44 skids joined 18:55 kaleem left 19:01 unobe joined
masak Su-Shee: the Android stack on Eclipse isn't so bad. it's based on auto-generation of a bunch of stuff, but considering that, it works quite well. 19:05
colomon I keep on fantasizing about setting sorear++ up with a Android machine and Mono for Android... 19:07
wonder if that could fit into Google Summer of Code.... 19:08
moritz [Coke]: I'm updating GHC on feather, in the hope that it'll make compiling pugs.hs easier 19:09
masak moritz++ 19:10
19:18 thou left, buubot_backup left
[Coke] moritz++ 19:18
19:20 colomon left
sorear good * #perl6 19:33
masak sorear! \o/ 19:34
19:39 buubot_backup joined 19:42 bluescreen100 left 19:43 buubot_backup left
moritz $ ghc -v 19:45
Glasgow Haskell Compiler, Version 7.4.1, stage 2 booted by GHC version 7.0.4
that's on feather
[Coke] moritz: looks like Pugs.hs also needs cabal for hacking. 19:48
also, thank you. ;) 19:49
moritz [Coke]: yes, I'm now following the dependencies
cabal-install installed. 19:50
19:52 icwiener joined
[Coke] moritz++ 19:52
dalek kudo/stdinit: 1d2af87 | jnthn++ | src/Perl6/ (2 files):
Switch reduction meta-op parsing to be a term, which resolves the last couple of regressions caused by switching to using <initializer>.
jnthn moritz: Think I've got stdinit fixed up now. :)
moritz \o/
jnthn Wanna do a test run yourself, or shall I just go ahead and merge it?
moritz I'm spectesting origin/stdinit merged into nom right nwo 19:53
jnthn moritz: With the above patch?
moritz jnthn: yes
jnthn moritz++
oh noes, spam in the RT queue! 19:54
jnthn suspects meeting thousands of singles would take a LOT of time 19:55
moritz unless you find a way to massively parallelize
19:56 bluescreen100 joined
jnthn nom: say 645 - 80 19:57
p6eval nom c3fb7d: OUTPUT«565␤»
dalek ast: da13ad1 | moritz++ | S04-statements/try.t:
tests for &die, try and X::AdHoc
 19:58
20:03 bluescreen100 left, bluescreen10 joined 20:04 ksi joined
moritz which pugs repo is the most recent one? 20:06
20:06 buubot_backup joined
dalek kudo/nom: c9d2827 | jnthn++ | / (5 files):
Merge branch 'nom' of github.com:rakudo/rakudo into nom
 20:07
kudo/nom: 525ec75 | jnthn++ | / (2 files):
Remove NOMMAP; update ROADMAP.
20:07 y3llow left, pothos left
jnthn oops, I accidentally the merge commit 20:09
moritz jnthn: all tests pass on stdinit + nom, plus a bunch of passing TODOs
20:09 birdwindupbird left, y3llow joined
moritz well, a few exception tests fail, but I'm to blame for that 20:09
(and i want to fix them over the weekend)
20:09 pothos joined
jnthn moritz: yay 20:10
moritz: OK, please do go ahead and push the merge :)
20:11 y3llow left 20:12 y3llow joined
moritz done. 20:12
dalek kudo/nom: a1180bc | moritz++ | src/ (3 files):
Merge remote branch 'origin/stdinit' into nom
20:12 pothos left 20:13 pothos joined 20:14 y3llow left, y3llow joined, not_gerd left 20:15 pothos left 20:16 pothos joined, y3llow left, y3llow joined
cognominal_ moritz, the goal is to make rakudo grammar closer to std resulting in better error messages? 20:17
20:18 y3llow left, Trashlord left 20:19 y3llow joined
masak jnthn: yeech, a merge commit! don't you check the log before you push? :P 20:23
moritz cognominal_: not only that, but it also makes some action methods easier, and other stuff possible that wasn't possible before 20:25
20:25 Trashlord joined 20:26 cooper joined
cognominal_ moritz++ 20:26
PerlJam moritz++ following the Perl tradition of making the impossible possible :)
[Coke] blog.chromium.org/2012/02/future-of...today.html updates to JS that may sound familiar. 20:32
20:33 fasad joined
moritz PerlJam: jnthn++ did most of the work, I just merged the branch 20:34
jnthn Just been through RT a bit and shuffled some tests over to testneeded thanks for moritz++'s $*HAS_SELF work :) 20:38
er
some tickets
sorear is stdinit just the initializer change?
jnthn sorear: yeah 20:39
sorear: Well, and $*HAS_SELF
sorear: I didn't do defterm yet.
sorear moritz: website and blog are instances of the same problem - I don't understand the audience well enough to write for it
[Coke] I would love to read posts for niecza like those that jonathan does on 6guts. 20:40
dalek ast: dfc580a | moritz++ | S0 (2 files):
rakudo unfudges
TimToady even Moses needed Aaron...
masak sorear: I read the Niecza release announcements with great interest every time. 20:42
PerlJam too
moritz sorear: one solution is to just write stuff for an imaginary audience, and then listen to feedback
masak sorear: but don't let people here bully you into taking valuable time off from coding to blog for someone else's sake. :)
PerlJam though I find I don't play with niecza regularly like I do with Rakudo
masak I'm starting to.
20:43 alester joined
masak Rakudo is still my main man, but Niecza is definitely getting there (in terms of how much I use it). 20:43
moritz sorear: also the announcements talk about "what", but I'd love to read more about the "why"s (why is $feature hard to implement right now, ...)
PerlJam It is always a a little bit of happiness when I write something that both Rakduo and Niecza can run :)
jnthn OK so... 20:44
nom: my $foo = 42; ($foo, 'stuff')[0] = 24; say $foo;
p6eval nom a1180b: OUTPUT«Cannot assign to a readonly variable or a value␤␤»
moritz ... and that happens quite often these days, as long as I don't use "handles" :-)
jnthn RT#105368 complains about this.
I just fixed it and discovered we have the following test
moritz jnthn: (unrelated) stdinit closed 3 different bug reports about [+] parsing :-)
jnthn my $x = 44;
$rt66304 = ( 11, $x, 22 );
dies_ok { $rt66304[1] = 'rw' }, 'variable List element is immutable'; 20:45
moritz: oh, nice! :D
Both of these tests can't be right :) Which way is it? :)
masak jnthn: mutable. 20:49
PerlJam wishes ack had a "paragraph mode" sometimes
masak otherwise, how would cool things like @array[0, 1, 2].reverse work? :)
jnthn masak: Good good.
alester alas
jnthn std: [xx] 1,2 20:50
p6eval std 48335fc: OUTPUT«ok 00:01 109m␤»
benabik Maybe regex based context?
-C /^$/,/^$/ (Backwards until empty line, forwards until empty line) 20:51
PerlJam benabik: that's ... interesting. 20:52
benabik PerlJam: I wasn't terribly serious. But, yes, interesting.
Or, perhaps, "interesting' 20:53
sorear we need a good extraction tool
moritz sometimes I want a tool that searches only code examples in POD 20:54
so either within C<...> or code blocks
because grepping for things like 'has' or 'does' is simply unproductive otherwise
PerlJam yep
that would be the next "killer feature" for ack even :) 20:55
sorear PerlJam: ack -f | xargs perl -00ne 'print if /.../' might DWYW 20:56
dalek d: 52f3895 | larry++ | STD.pm6:
turn reduceop parsing into a term
 21:00
TimToady niecza: constant @foo = 1,@foo; say @foo[1] 21:02
p6eval niecza v14-34-g1b3d1e8: OUTPUT«␤»
TimToady sorear: any clue how far that is from working?
sorear if by 'working' you mean 'carping with a decent error message', that could probably be done today 21:04
sorear wonders if there is a way to configure the regex(?) used by git grep -W 21:05
TimToady no, I mean working working, and in doing what I mean :) 21:06
*as in
sorear that would require considerable thinking. :) 21:07
making , thunk the RHS is not reasonable, but maybe something can be done at the level of constant @foo = ...
TimToady it's just a lazy list; we have to persuade @foo to return an appropriately lazy iterator 21:08
[Coke] TimToady: what do you expect that to print? 21:09
TimToady 1
it's the boiled down version of the hamming recursion
[Coke] what is @foo at that point? 1,1 ?
or 1,1,1... ? or?
TimToady constant @hamming = 1, dedup (@hamming X* 2) M (@hamming X* 3) M (@hamming X* 5); 21:10
it's a fancy way of saying 1 xx * via recursive definition 21:11
Haskellers do this sort of thing all the time
21:13 GlitchMr left
TimToady and Python includes hamming in their tests of whether iterators are correctly defined, so this bears thinking about 21:13
PerlJam ... and if *those* languages can do it, surely Perl 6 should be able to ;) 21:14
TimToady we should at least understand why we can't, if not, and how best to work around it 21:15
if it actually requires strict laziness, for example, we might just define 'constant' that way 21:16
Teratogen Haskell is pretty much an academic language, isn't it?
PerlJam Teratogen: no
TimToady among other things
Teratogen I don't hear about avionics or web sites being written in Haskell, so...
*shrug* 21:17
PerlJam Teratogen: GHC has made Haskell more than just "academic"
TimToady and I think we already had the avionics discussion
Teratogen The still use a multithreaded version of Algol called "Jovial" in avionics software. 21:18
That is an OLD language.
From, like, 1963.
PerlJam before the dawn of time!
TimToady as I said, we already had this precise discussion; let's not have it again
Teratogen Ok =(
21:18 snearch left
Teratogen Sorry, I am bipolar and forgetful . =( 21:18
TimToady more like OCD, but we'll let it pass :) 21:19
PerlJam makes a note to try that trick on Teratogen in the future ;)
Teratogen No!
I am already on four medications!
PerlJam Teratogen: don't you remember? We talked about this just the other day!
:-) 21:20
Teratogen Stop it.
TimToady PerlJam: don't torment the Teratogen
it's the best Teratogen we've got, so we'll have to be nice to it
PerlJam point taken.
TimToady anyway, rakudo can't seem to do it either, quite 21:22
nom: my @foo; @foo.plan: 1, @foo; say @foo[1] 21:23
p6eval nom a1180b: OUTPUT«Method 'plan' not found for invocant of class 'Array'␤␤»
TimToady hmm, thought we had that, oh well
flussence pugs: my @foo = 1,@foo; say @foo[1]; 21:24
p6eval pugs b927740: OUTPUT«␤»
masak Teratogen: if you forget things from day to day, I understand much better why you've been asking us to "just get Perl 6 out the door" so often.
masak hugs Teratogen
flussence huh, usually pugs surprises with hard code like that
Teratogen I have some short term memory problems, probably from the medications I am taking
TimToady that's an eager assignment
Teratogen My long term memory is ok
PerlJam flussence: it did surprise you, just not the way you thought it would :)
21:24 MayDaniel left
masak Teratogen: where does short term memory end and long term memory start for you? 21:25
TimToady pugs: my @foo := (1,@foo); say @foo[1]
flussence pugs: constant @foo = 1,@foo; say @foo[1]; # I don't think it supports constants...
p6eval pugs b927740: OUTPUT«␤»
pugs b927740: OUTPUT«pugs: *** Undeclared variable: ("@foo",MkPad (padToList []),[PCompiling {pc_pad = MkMPad {mp_id = 140456300146673, mp_pad = <ref:0x7fbe87a55461>}}])␤ at /tmp/izJFdXmd50 line 1, column 19-23␤»
Teratogen masak, hard to tell
PerlJam Teratogen: I tend to have problems with medium-term memory myself
Teratogen I've made modifications to my jeep, and now thinking about it I don't know how I made them.
TimToady pugs: constant @foo = (1,@foo); say @foo[1] 21:26
p6eval pugs b927740: OUTPUT«pugs: *** Undeclared variable: ("@foo",MkPad (padToList []),[PCompiling {pc_pad = MkMPad {mp_id = 140704615526385, mp_pad = <ref:0x7ff8586e7d99>}}])␤ at /tmp/pjf9djXthZ line 1, column 17-25␤»
Teratogen Like "What the hell was I thinking when I did that?"
Same with computer programs.
PerlJam (where "medium term" is in the 1 hour - 1 day range)
Teratogen I look at some old stuff that I wrote and I can't figure out what the fuck I was doing.
flussence that's normal
PerlJam I can remember things for about an hour or 2, but then I forget them until a day or so later
21:27 jeffreykegler joined
TimToady niecza: constant @foo = gather { take 1; take @foo }; say @foo[1] 21:30
p6eval niecza v14-34-g1b3d1e8: OUTPUT«(timeout)»
TimToady kaboom!
21:30 pernatiy joined 21:31 bluescreen10 left
TimToady niecza: my @foo := gather { take 1; take @foo }; say @foo[1] 21:32
p6eval niecza v14-34-g1b3d1e8: OUTPUT«(timeout)»
TimToady take seems to be eagerizing the return of @foo somehow
21:33 colomon joined
TimToady niecza: my @foo := gather { take 1; take $_ for @foo }; say @foo[1] 21:33
p6eval niecza v14-34-g1b3d1e8: OUTPUT«1␤»
TimToady \o/
niecza: my @foo := gather { take 1; take $_ for @foo }; say @foo[582] 21:34
21:34 ab5tract left
p6eval niecza v14-34-g1b3d1e8: OUTPUT«Unhandled exception: Circular data dependency in list iteration, or last fetch threw exception␤ at /tmp/j494A5xugU line 1 (ANON @ 4) ␤ at <unknown> line 0 (KERNEL dogather @ 1) ␤ at /home/p6eval/niecza/lib/CORE.setting line 2308 (GatherIterator.reify @ … 21:34
TimToady aww
masak PerlJam: I used to meditate quite a bit, and I remember medium-term memory actually getting a boost from that. some days I got a little snapshot replay of things I'd forgotten during the day.
21:35 jeffreykegler left
PerlJam masak: define "meditate". Did you have something to help you meditate? 21:35
Su-Shee PerlJam: "a cushion"? ;) 21:36
masak PerlJam: no.
TimToady meditation is what happens when you stop doing everything else :)
PerlJam TimToady: I call that "sleep" :)
TimToady you have to stop that too
masak &
PerlJam This guy www.ted.com/talks/shawn_achor_the_h..._work.html talks about meditation as one of the things that leads to happiness. 21:37
TimToady if I'm not careful, I spend all day meditating by default :)
I think it just comes down to "Don't be so busy thinking that you can't think."
Su-Shee PerlJam: if you prefer neuro-science - medidation is rather well researched by now. 21:38
PerlJam I can't get my brain to stop talking to itself some times.
Su-Shee PerlJam: it's basically a certain physical training of your brain to make it pulse in certain waves.
PerlJam: that would be the "training" part.. 21:39
TimToady nom: my @foo := gather { take 1; take $_ for @foo }; say @foo[582]
p6eval nom a1180b: OUTPUT«splice() not implemented in class 'Mu'␤␤»
TimToady whee!
Su-Shee PerlJam: if you compare all kinds of meditation schools, you'll notice that they all teach several aspects of focus.. "focus on something" "blend out everything" etc. 21:40
TimToady that's for the ADHD folk; those of us on the autistic spectrum can already shut everything out because it's our only alternative to paying attention to everything 21:42
however, I think I'll meditate on the insides of my eyelids for now 21:43
zzz &
21:47 bluescreen10 joined
dalek kudo/nom: 9070eeb | jnthn++ | src/core/Parcel.pm:
Don't strip elements of their containers when indexing into a Parcel. Fixes RT#105368.
 21:48
kudo/nom: 2cfeacd | jnthn++ | src/ (2 files):
Thunk the LHS of the xx operator, as current spec requires.
ast: 4f47254 | jnthn++ | S02-types/lists.t:
Remove a couple of wrong tests, some corrections and some unfudges for Rakudo.
 21:49
21:50 rir joined 21:56 yertalert left 21:57 pmurias joined
pmurias hi 21:57
anyone else having trouble with git pull'ing niecza from github?
21:58 rir left 22:10 tarch left
PerlJam pmurias: nope. just did it fine 22:13
pmurias works again 22:16
dalek ecza: 3b576df | (Paweł Murias)++ | p5test/p5test.pl:
Fix typo in p5test.
22:19 alvis left
benabik github is suffering from a DDoS. There are sporadic interruptions in service. 22:19
jnthn std: sub foo() { state $z++; $z } 22:25
p6eval std 52f3895: OUTPUT«ok 00:01 110m␤»
pmurias why would anyone DDoS github? 22:26
benabik No idea, but that's what they say is happening: github.com/blog/1036-about-this-we...ailability 22:28
flussence one of the comments on their blog explained it nicely
"cvs fans"
.oO( but a true cvs fan would use a non-distributed DoS, and they'd have to log in first... )
 22:30
jnthn CVS has...fans? :)
timotimo www.quickmeme.com/meme/3619or/ 22:31
22:34 bkolera joined 22:41 benabik left 22:42 benabik joined
dalek ast: 5df18e5 | jnthn++ | S12-attributes/instance.t:
Fudge updates for S12-attributes/instance.t.
 22:52
kudo/nom: f05a588 | jnthn++ | src/Perl6/ (2 files):
Fix handling of self scoping in attribute initializers.
kudo/nom: 6f3d34b | jnthn++ | t/spectest.data:
Turn on S12-attributes/defaults.t.
kudo/nom: 9f9b446 | jnthn++ | src/Perl6/Grammar.pm:
Enable use of :state/:constant/:our declarations in regexes.
22:53 kaare_ left 23:17 bkolera left
sorear benabik: I wonder if this is related to the freenode DoS 23:23
23:34 bkolera joined 23:35 tarch joined 23:36 Trashlord left
jnthn rest & 23:36
23:40 whiteknight joined 23:43 dorlamm joined 23:44 pmurias left
dalek kudo/nom: a39d86a | jnthn++ | docs/ChangeLog:
More ChangeLog updates.
 23:47
23:50 wolfman2000 joined
masak 'night, #perl6 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!