»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo: / pugs: / std: , or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by diakopter on 25 January 2010.
00:06 ggoebel left 00:07 meppl left 00:12 QtPlatypus joined 00:16 jferrero left 00:20 meppl joined 00:25 cotto_work left 00:29 IllvilJa left, wknight8111 left, hercynium left, ash_ left, andy1 left, rgrau left, [particle] left, simcop2387 left, renormalist|work left, gfldex left, f00li5h[HireMe] left, f00li5h[HireMe] joined 00:30 sjn joined, rgrau joined 00:32 jferrero joined 00:38 cotto_work joined 00:46 snarkyboojum left 00:47 Chillance left 00:49 rgrau left 01:02 snarkyboojum joined 01:05 simcop2387 joined, yinyin joined 01:07 meppl left 01:13 lue joined
lue hello! 01:14
m-locks hi 01:17
why does rakudo complain about use v6;
01:17 wolf2k_ubuntu joined
m-locks or am i supposed to use a v6 module on perl5 01:18
crap
i bet i can't find that one ported
TimToady alpha: use v6;
p6eval alpha 30e0ed: ( no output )
TimToady rakudo: use v6
p6eval rakudo b9b10e: ( no output ) 01:19
m-locks so somethings up
TimToady it doesn't complain on my machine
m-locks i was going to try out some of those modules there 01:20
that svg-plot for instance
lue bkeeler: ping
bkeeler hi there 01:21
01:22 hercynium joined
lue I think I may want to ssh to that server :). Runlevel 3 made no tangible difference (and I just spent a few minutes FIXING the hd after shutting down) 01:23
(of course, whenever I shut down normally, I have to fix the hd, but it was more than usual) 01:24
01:25 jferrero left
m-locks hmm 01:28
"load_bytecode" couldn't find file 'SVG::Plot::Pie.pir'
it's supposed to look for .pm i guess 01:29
01:30 lestrrat is now known as lest_away
bkeeler You may need to compile the SVG::Plot::Pie code to pir with perl6 --target=pir 01:30
Though obviously use *should* be fine with loading .pm files directly 01:31
m-locks well, gonna try out that --target first, dunno whats wrong there
bkeeler I guess it just hasn't been implented in master yet 01:32
m-locks well its not working that great either 01:33
bkeeler It looks like use doesn't actually do anything at compile time, it just emits code to run a !use sub in cheats/use.pir
m-locks ok but how do i supply the module name 01:34
bkeeler Looks like alpha might be your best bet for playing with that SVG code
lue Perl6: the only language to admit it cheats :) 01:35
bkeeler Though like I said, you should be able to './perl6 --target=pir file.pm > file.pir' or similar to compile it manually
Well, the idea is that the cheats are temporary hacks which should go away eventually 01:36
lue I know. But still...
m-locks i guess trying out modules are just temporary hacks then too
maybe i'm missing some directory setting or something 01:38
lue (this is great. I'm doing stuff and it's not affecting my computer at all. I'm ecstatic :D) 01:39
01:44 mssm left
lue bkeeler: I am amazed at your server. Must want. 01:53
bkeeler It was a semi-decent server a few years ago, but it does the job 01:54
lue what you call semi-decent I call really fast! 01:55
bkeeler Well, in comparison to G3, yeah heh 01:56
lue When you use a G3 forever, anything's bound to amaze you :D 01:57
lichtkind good night 02:02
lue good night o/
02:05 lest_away is now known as lestrrat 02:11 spray joined
spray pointme: mubot 02:12
no?
lue what are you trying to do
spray comments on bottom of: howcaniexplainthis.blogspot.com/200...ts-do.html 02:13
namely, get bot code
lue ...we don't have a pointme bot :/ 02:15
02:15 sbilik joined, TimToady sets mode: +vv buubot dalek, TimToady sets mode: +v ilogger2, TimToady sets mode: +vvv lisppaste3 p6eval phenny
spray do you know where I could obtain the code for the bots? 02:16
(assuming its available)
lichtkind lue: thanks
02:16 lichtkind left
lue How would I run changes to the test suite in the pugs repo as opposed to running make spectest in the rakudo repository? 02:19
02:21 sbilik left 02:24 ShaneC left
bkeeler LIke if you just want to run one test? 02:28
colomon make t/spec/S32-trig/sin.t 02:29
for example
bkeeler Yep, what he said
That takes care of fudging too if need be
OK, food time. Back later....
lue alright. I really don't want to call make spectest every time I change a test :) 02:30
colomon you can also run it directly -- ./perl6 t/spec/S32-trig/sin.rakudo to get the fudged version, for instance. 02:32
but I prefer using the make form except for special cases (like running in the debugger).
02:33 eternaleye left
lue It says `nothing to be done for [filename here]' :/ 02:34
colomon did you say make blah/.rakudo ? 02:35
that doesn't work. make always has to be blah/.t
TimToady I think that's abuse of make
colomon is an abuser. 02:36
lue then what should I do !?
02:37 PhilKenSebben joined
colomon ignore TimToady and use make. 02:37
lue (I did use .t btw)
colomon can you post what you typed and what it said back? 02:38
(paste if it is very long?)
lue make t/spec/S32-trig/sin.t (this gave me:)
make: Nothing to be done for `t/spec/S32-trig/sin.t'.
colomon what directory did you run it from? 02:39
TimToady THAT'S A CORRECT RESULT IF IT ALREADY EXISTS!!!!
lue pugs, the top level
colomon Oh!
if you want to test rakudo, you need to do it from the rakudo directory.
lue :/ I just want to run the tests as I change them... and I'd rather change them in pugs directly 02:40
colomon TimToady: That's not how Rakudo's makefile is set up.
TimToady and using make that way prevents you from using make to *write* any tests
it's just completely bogus 02:41
it's not design smell, it's design rotten to the core
lue still wants to run tests w/o make spectest ... 02:42
TimToady someday I'll tell you what I really think... :)
lue would ./perl6 be alright ? 02:44
PhilKenSebben laughs "Ha ha haaaaaaaa!"
02:44 PhilKenSebben left
colomon and maybe writing a new script that does exactly what make blah/test.t does only without using the make structure would be a good test for a wandering newbie. :) 02:44
lue: yes, but you'll have to fudge fudged tests by hand. 02:45
if you do it that way.
TimToady fudgeall should work just fine for that
lue do I HAVE to fudge tests run w/ perl6?
colomon (And I don't have a clue how to do that.)
well, you can avoid it if there is no fudging in the test file. 02:46
or if you don't mind it not working. :)
TimToady often the test file won't even compile without fudging
lue Isn't fudging for, when you want the test to work, but you don't mind lieing a little (aka The Politician Method)
TimToady unlike a politician, it helps to cut down noise 02:48
basically, it allows you test against some subset language rather than real Perl 6 02:50
lue so, fudging is temporary (also like politicians' lies nowadays)
TimToady yes, the very fact of fudging is considered a test failure, when it comes to real Perl 6
the acid test will be passing all the *.t tests without fudging 02:51
but fudging is useful for weeding out results you aren't (yet) interested in
lue alright, I'll have a look at all that when appropos. I'll add myself to the AUTHORS file when I have something to upload. 02:52
02:53 Trashlord joined
dalek kudo/master: 7c90f14 | pmichaud++ | src/core/ (4 files):
Refactor some methods in/into Iterator/Iterable.
 03:02
colomon pmichaud: BTW, IMO Range.Num should have been implemented more smartly than just creating the iterator and counting it. That was only ever intended to be a quick way to get .Num working. 03:06
Though of course, it is easily overriden. 03:08
lue util/prove6 seems like a candidate for running tests *coughcough* 03:13
colomon prove6? 03:15
lue idk, it's in the README 03:16
colomon it's in pugs/util 03:17
looks like it is designed for pugs. 03:18
lue Anything in particular that needs to be testable? Or just look in the TODO? 03:19
colomon I'm not sure I follow you? 03:20
lue Anything of critical importance in terms of testing?
colomon Ah, good places for you to be adding tests, you mean? 03:22
lue yes. 03:23
colomon ponders...
lue I don't want to waste time on an operator in S02 when the whole of S22 is untested :)
TimToady there's an operator in S02? :)
lue just an example, not meant to be real :) 03:24
colomon we could use a lot more up-to-date series tests, for example.
lue (We smile too much) :)
colomon: will look at that. :) 03:25
colomon but that's going to be hard for you, because it's tricky and doesn't work properly yet.
so that's not an actual example. :)
TimToady better too much than too little
colomon (Hopefully it will be a good target for tests next week...)
I'm trying to think of areas that shocked me when I looked at them. 03:26
lue I'll just look around, see what isn't there, look in TODO, etc... :)
colomon rakudo: say 5 min 4 03:29
p6eval rakudo 7c90f1: OUTPUT«4␤»
colomon rakudo: say 4 min 5 min 2
p6eval rakudo 7c90f1: OUTPUT«2␤»
colomon That's one of the ones I ran into. I'm not sure there are *any* real tests for that form of min / max.
either that or I completely missed them somewhere in there. 03:30
There are tests for the meta-reduce form of those operators, but not the regular forms. 03:31
lue rakudo: say 4 min 3 min 5 03:32
p6eval rakudo 7c90f1: OUTPUT«3␤»
lue will look at minmax
colomon though that test itself may not work? I'll take a look, too....
03:36 quietfanatic left
lue rakudo: say min(3,4,2) 03:38
p6eval rakudo 7c90f1: OUTPUT«invoke() not implemented in class 'Integer'␤current instr.: 'perl6;Any;_block730' pc 256913 (src/gen/core.pir:14638)␤»
colomon rakudo: say (3,4,2).min 03:39
p6eval rakudo 7c90f1: OUTPUT«2␤»
colomon both forms should work, I think...
oh, wait. 03:40
03:41 mikehh_ joined, mikehh left
colomon rakudo: say min({$^a <=> $^b}, 2, 5, 3); 03:41
p6eval rakudo 7c90f1: OUTPUT«2␤»
TimToady alpha: say [min](3,4,2) 03:42
p6eval alpha 30e0ed: OUTPUT«2␤»
03:43 quietfanatic joined
lue should min without the [] works as well? 03:43
colomon lue: the sub form of min must take the Ordering argument first.
so it works like I used it there, but not like you used it.
03:44 mikehh_ left, mikehh_ joined
TimToady have to use [min] for the other way 03:45
colomon and [min] doesn't work yet in master. :(
TimToady which is why I used alpha... 03:46
lue where would min/max tests be? I don't see them... 03:48
colomon S32-list
though I guess you might be able to make an argument that the operator form could go in S03 or something. 03:49
lue yeah, .min , .max , and .minmax are defined in S03
colomon max= is tested in S03-operators/assign.t 03:51
lue waitaminute. In S32-list/minmax.t it tests if @array.min is -3, when the lowest element in that list is -9 ! 03:52
colomon what line?
(and I'm trying to get that test file fudged enough to run right now. :) 03:53
lue line 20, according to my count 03:54
03:54 thirdfourth joined
colomon huh, that test passes, too. 03:56
lue rakudo: @rray=<5 -3 7 0 1 -9>; say @rray.min;
p6eval rakudo 7c90f1: OUTPUT«Symbol '@rray' not predeclared in <anonymous>␤current instr.: 'perl6;PCT;HLLCompiler;panic' pc 137 (compilers/pct/src/PCT/HLLCompiler.pir:101)␤»
lue rakudo: my @rray=<5 -3 7 0 1 -9>; say @rray.min;
p6eval rakudo 7c90f1: OUTPUT«-3␤»
lue ō.–
colomon oh, it's doing a string comparison, not a numeric comparison.
lue ...of course (?)
TimToady because <> isn't up to current spec 03:57
those should come out as Ints
lue so, should that test be force-failed for now?
colomon alpha: @rray=<5 -3 7 0 1 -9>; @rray.sort.perl.say
p6eval alpha 30e0ed: OUTPUT«Symbol '@rray' not predeclared in <anonymous> (/tmp/Xr5ugCURto:10)␤in Main (file <unknown>, line <unknown>)␤»
colomon alpha: my @rray=<5 -3 7 0 1 -9>; @rray.sort.perl.say
p6eval alpha 30e0ed: OUTPUT«["-3", "-9", "0", "1", "5", "7"]␤»
lue ō.–
colomon It shouldn't be force-failed, it should be fixed -- by the spec, the test is wrong. 03:58
lue the negative numbers don't play nice... (maybe avoid <> for now?)
TimToady the dwimminess of <> was added not very long ago
thirdfourth Hi All. Quick question: would you say that Perl 6 has less or more magic than Perl 5? By "magic", I mean language features that just work because they're handy and practical, but don't necessarily fit terribly well with the rest of the language features.
lue rakudo: my @rray=[5 -3 7 0 1 -9]; say @rray.min; 03:59
TimToady less, if you define it that way
thirdfourth Sorry -- that came out wrong: I meant, "hi all. I have a quick question I was hoping to get an answer to tonight"
p6eval rakudo 7c90f1: OUTPUT«Confused at line 11, near "my @rray=["␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
lue rakudo: my @rray=5 -3 7 0 1 -9; say @rray.min;
p6eval rakudo 7c90f1: OUTPUT«Confused at line 11, near "my @rray=5"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
lue rakudo: my @rray=«5 -3 7 0 1 -9»; say @rray.min; #please work
p6eval rakudo 7c90f1: OUTPUT«-3␤»
colomon I would say it has more magic stuff that works, but less magic stuff that doesn't fit.
TimToady in general, instead of any specific magic, we've designed in general principles that happen to come out the same way :) 04:00
lue rakudo: my @rray=[5, -3, 7, 0, 1, -9]; say @rray.min; #please work
p6eval rakudo 7c90f1: OUTPUT«5 -3 7 0 1 -9␤»
colomon lue: give me a minute to finish my current reworking of minmax.t.
lue rakudo: my @rray=5, -3, 7, 0, 1, -9; say @rray.min; #please work
p6eval rakudo 7c90f1: OUTPUT«-9␤»
lue colomon: alright
\o/
thirdfourth @TimToady and @colomon: Thanks for the answers. Yes, TT, that's the kind of magic I mean. 04:01
lue .u 3
phenny U+0033 DIGIT THREE (3)
lue .u 9
phenny U+0039 DIGIT NINE (9)
TimToady so, as an example
instead of having $a and $b as special inside sort, we have placeholder arguments that can be used anywhere 04:02
and sort {$^a <=> $^b} falls out naturally from that
lue thirdfourth: the fact that we can get rakudo to do anything is magic enough :D
TimToady and instead of certain functions magically dealing with $_ by default, we always use .foo to explicitly call any method on $_ 04:03
04:04 wolf2k_ubuntu left
lue --location=!COMPUTER & 04:05
thirdfourth @TimToady: thanks for the examples. :) 04:06
colomon alpha: my @rray=<5 -3 7 0 1 -9>; say min(@rray)
p6eval alpha 30e0ed: OUTPUT«No applicable candidates found to dispatch to for 'min'␤in Main (file <unknown>, line <unknown>)␤»
TimToady there are many, many more
04:07 justatheory left
TimToady the parsing of blocks is not special between built-ins and user-defined blocks 04:07
so you don't have to track which blocks require a final ; as you do in C or Perl 5 04:08
thirdfourth I'm looking to put a weekend into learning the basics (would like to get a jump on everyone before Rakudo Star :) ). Will probably start with the "Perl 5 to 6" chapters at perlgeek.de/en/article/5-to-6 , and also the Perl 6 Advent calendar.
04:08 lestrrat is now known as lest_away
TimToady that's a good place to start 04:08
04:10 thirdfourth left 04:15 mikehh_ is now known as mikehh 04:16 mikehh left, mikehh joined 04:20 justatheory joined
pugssvn r29802 | colomon++ | [t/spec] Refudge to work with Rakudo. Still known errors where the tests do not conform to current spec. 04:25
04:34 petdance joined
colomon lue: Okay, I've pushed my changes to rakudo and t/spec. If you'd like to play around with minmax.t, I'd suggest updating both, then looking at change that first array from < > to ( ) and then figure out what the new answers should be. 04:39
off to bed now...
dalek kudo/master: a1ed7ff | (Solomon Foster)++ | src/core/IO.pm:
Implement simple version of sub lines to get minmax.t working.
 04:40
kudo/master: d6129bc | (Solomon Foster)++ | t/spectest.data:
Turn on minmax.t.
lue will do colomon, will do. ō/ (salute) 04:42
04:43 stephenlb left 04:49 eternaleye joined
lue rakudo: my @rray=(5 -3 7 0 1 -9); say @rray.min; 04:58
p6eval rakudo 7c90f1: OUTPUT«Confused at line 11, near "my @rray=("␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
04:58 cotto_work left
lue rakudo: my @rray=(5,-3,7,0,1,-9); say @rray.min; 04:58
p6eval rakudo 7c90f1: OUTPUT«-9␤»
lue I can't get the tests to run on their own (grumble grumble). Plan B then. 05:12
05:14 Jeffrey_ joined, payload left 05:15 Jeffrey_ left
lue rakudo: my @array=(5,-3,7,0,1,-9); say min({$^a <=> $^b }, :values(@array)) 05:15
p6eval rakudo d6129b: OUTPUT«Unexpected named parameter 'values' passed␤current instr.: 'min' pc 233168 (src/gen/core.pir:5314)␤» 05:16
lue alpha: my @array=(5,-3,7,0,1,-9); say min({$^a <=> $^b }, :values(@array))
p6eval alpha 30e0ed: OUTPUT«sh: ../rakudo-alpha/perl6: No such file or directory␤»
lue :( stupid rebuilding
alpha: my @array=(5,-3,7,0,1,-9); say min({$^a <=> $^b }, :values(@array)) 05:18
p6eval alpha 30e0ed: OUTPUT«Unexpected named parameter 'values' passed␤in Main (file src/gen_setting.pm, line 324)␤»
lue pugs: my @array=(5,-3,7,0,1,-9); say min({$^a <=> $^b }, :values(@array))
p6eval pugs: OUTPUT«*** Named argument found where no matched parameter expected: (values,Ann (Pos (MkPos "/tmp/ouKNDem2vH" 1 60 1 66)) (Var "@array"))␤ at /tmp/ouKNDem2vH line 1, column 32 - line 2, column 1␤»
lue std: my @array=(5,-3,7,0,1,-9); say min({$^a <=> $^b }, :values(@array)) 05:19
p6eval std 29802: OUTPUT«ok 00:01 109m␤»
spinclad rakudo: my @array=(5,-3,7,0,1,-9); say min({$^a <=> $^b }, |@array)
p6eval rakudo d6129b: OUTPUT«-9␤»
spinclad lue: std would accept that (it parses) regardless of whether min takes a named :values argument 05:20
lue Ok, thanks! (rewriting minmax.t test...) 05:22
alpha: my @array=(5,-3,7,0,1,-9); say (@array.min: { abs $^a <=> abs $^b })
p6eval alpha 30e0ed: OUTPUT«0␤»
lue alpha: my @array=(5,-3,7,0,1,-9); say (min({ abs $^a <=> abs $^b }, @array)) 05:24
p6eval alpha 30e0ed: OUTPUT«0␤»
lue alpha: say ((-10..10).min: { abs $^a <=> abs $^b })
p6eval alpha 30e0ed: OUTPUT«0␤»
lue rakudo: say ((-10..10).min: { abs $^a <=> abs $^b }) 05:26
p6eval rakudo d6129b: OUTPUT«too many positional arguments: 2 passed, 1 expected␤current instr.: 'perl6;Attribute;accessor_helper_ro' pc 3312 (src/gen/Attribute.pir:400)␤»
lue alpha: say (1..10).min: { ($_-3) * ($_-5) } 05:27
p6eval alpha 30e0ed: OUTPUT«4␤»
lue rakudo: say (1..10).min: { ($_-3) * ($_-5) }
p6eval rakudo d6129b: OUTPUT«too many positional arguments: 2 passed, 1 expected␤current instr.: 'perl6;Attribute;accessor_helper_ro' pc 3312 (src/gen/Attribute.pir:400)␤»
lue alpha: my @array=(5,-3,7,0,1,-9); say @array.max 05:28
p6eval alpha 30e0ed: OUTPUT«7␤»
lue alpha: my @array=(5,-3,7,0,1,-9); say (@array.max: { $^a <=> $^b }) 05:30
p6eval alpha 30e0ed: OUTPUT«7␤»
lue alpha: say ((-10..9).max: { abs $^a <=> abs $^b }) 05:31
p6eval alpha 30e0ed: OUTPUT«-10␤»
lue alpha: my @array=(5,-3,7,0,1,-9); say (max({ $^a <=> $^b }, @array)) 05:32
p6eval alpha 30e0ed: OUTPUT«7␤»
lue rakudo: my @array=(5,-3,7,0,1,-9); say (max({ $^a <=> $^b }, @array))
p6eval rakudo d6129b: OUTPUT«7␤»
lue alpha: my @array=(5,-3,7,0,1,-9); say (@array.max: { abs $^a <=> abs $^b }) 05:33
p6eval alpha 30e0ed: OUTPUT«-9␤»
lue alpha: say 42.min
p6eval alpha 30e0ed: OUTPUT«42␤»
lue alpha: say 42.max
p6eval alpha 30e0ed: OUTPUT«42␤»
lue alpha: say (1, Inf).max 05:34
p6eval alpha 30e0ed: OUTPUT«Inf␤»
lue alpha: say (0, NaN).min; say (Inf, NaN).min; say (0, NaN).max; say (Inf, NaN).max; 05:38
p6eval alpha 30e0ed: OUTPUT«0␤Inf␤0␤Inf␤» 05:39
lue Oh hoh! minmax.t got it horribly wrong!
rakudo: say (0, NaN).min; say (Inf, NaN).min; say (0, NaN).max; say (Inf, NaN).max;
p6eval rakudo d6129b: OUTPUT«0␤Inf␤NaN␤NaN␤»
lue pugs: say (0, NaN).min; say (Inf, NaN).min; say (0, NaN).max; say (Inf, NaN).max;
p6eval pugs: OUTPUT«NaN␤NaN␤0␤Inf␤»
lue the hell... ō.–. Methinks noone's reading the spec. Three P6 implementations, three results, all different from what minmax.t says. 05:40
pmichaud colomon: (Range.Num) -- sure, if you have a smarter mechanism for computing the range, then the class can override it. My intent is that if you have an Iterable (and don't know much else about it), then the calculation for .Num is to count the iterations. :-) 05:42
TimToady minmax should probably be redefined in terms of Range objects
lue will blotch out those tests until further notice...
05:43 gfx joined
lue TimToady: you see the problem? Three implementations giving three results, none matching what was set in minmax.t! *pull hair* 05:44
pmichaud lue: that's why we have a set of standard tests :) 05:45
lue yes. Where's the info in the spec when you need it? (Is NaN greater than Inf? How's about 0 vs NaN? argh!) 05:46
How can I make ALL IMPLEMENTATIONS skip a test (besides removing it)? 05:47
pmichaud comment it out?
explicitly add a skip(...) call?
lue I'm looking for the fancy #? skip command, but that'll do.
just surround the offending code in skip( [code] ) ? 05:48
pmichaud no
skip(5, 'reason for skipping 5 tests');
(I might have the args backwards there.)
or, perhaps more likely, todo(...) 05:49
lue for example, there's a #?rakudo 4 skip '[op] NYI' line in their. I just want to make it skip all implementations, not just rakudo
s/their/there/
pmichaud in this case, using #? is likely incorrect, because the reason for skipping the tests has to do with the fact that the tests are wrong, not because any implementations have difficulty with them.
lue so, just comment it out, and change the # of planned tests? 05:50
pmichaud that's what I'd do, yes.
lue will do.
pmichaud and in the comment it out, explain why they're commented out.
lue of course.
The guy before me had rakudo skip the first two of four NaN related ones, but that's not going to help. 05:51
05:56 crazed left, kaare joined 05:57 kaare is now known as Guest87962 05:58 petdance left
pugssvn r29803 | lue++ | added my er, name, to AUTHORS 06:04
06:05 crazed joined
pugssvn r29804 | lue++ | [t/spec] switched @array from <> to (), changed tests accordingly, and blotched out NaN-related tests. 06:06
lue I'm done for the night. Cheers! 06:07
06:07 lue left 06:08 JimmyZ joined 06:41 dual left 06:43 lest_away is now known as lestrrat 06:46 justatheory left 07:18 riffraff joined 07:20 Su-Shee joined 07:29 samlh left 07:30 snarkyboojum left 07:42 PZt left 07:43 avar left 07:44 avar joined 07:46 JimmyZ left 07:50 PZt joined 07:51 JimmyZ joined 07:54 ihrd joined, ihrd left 07:56 Guest87962 left 07:59 iblechbot joined 08:00 JimmyZ left
Su-Shee good morning. 08:29
mathw morning 08:42
moritz_ good morning 08:45
08:46 barney joined 08:47 pmurias joined 08:48 dual joined 08:56 gfx left 09:04 dakkar joined
dalek kudo/master: 94ada1f | moritz++ | (4 files):
track change of macro names of PObj_active_destroy_ macros in parrot

Also bumps PARROT_REVISION to r44371		 09:16
09:18 meppl joined
moritz_ rakudo: my $c = { say $_.perl; say @_.perl }; $c(2, 3, 4, 5) 09:32
p6eval rakudo d6129b: OUTPUT«Mu␤[2, 3, 4, 5]␤»
moritz_ jnthn: please take a look at the evalbot output above, and #63974 09:33
there are tests in t/spec/S06-signature/sub-ref.t (which we don't run yet)
and I think the last two tests in t/spec/S06-signature/sub-ref.t are a topic for pmichaud++ (closures) 09:36
09:36 uniejo joined
dalek kudo/master: e9b6253 | moritz++ | t/spectest.data:
moritz-- does not understand the simple format of t/spectest.data; fixing
 09:39
09:42 cotto left 09:43 cotto joined 09:45 yinyin left 09:46 Mips128 joined 09:47 xabbu42 joined
pugssvn r29805 | moritz++ | [t/spec] unfudge some tests for rakudo 09:47
r29806 | moritz++ | [t/spec] fudge sub-ref.t for rakudo (I have opened tickets for the new failures)
09:49 uniejo left, uniejo joined
lisppaste3 colomon pasted "latest rakudo doesn't build for me" at paste.lisp.org/display/95476 09:50
09:51 mssm joined
dalek kudo/master: 2c4f69b | moritz++ | t/spectest.data:
re-enable sub-ref.t
 09:51
09:51 uniejo left 09:52 uniejo joined 09:55 cotto left, kaare joined 09:56 kaare is now known as Guest44360
moritz_ phenny: tell masak marcus.nordaaker.com/2010/02/cpanmi...superstar/ it seems like people like module installers with less output noise than CPAN.pm :-) 09:57
phenny moritz_: I'll pass that on when masak is around.
09:58 lestrrat is now known as lest_away 10:10 cotto joined 10:16 quester_ joined 10:21 jferrero joined 10:24 masak joined
masak oh hai #perl6. :) 10:25
phenny masak: 09:57Z <moritz_> tell masak marcus.nordaaker.com/2010/02/cpanmi...superstar/ it seems like people like module installers with less output noise than CPAN.pm :-)
10:25 orafu left
masak oh gasp... :) 10:25
10:25 orafu joined
moritz_ :-) 10:25
masak "And when I say check out, I mean it literally." that's a meme of mine (I think). it's from this blog post: use.perl.org/~masak/journal/37303 10:27
hm, David Green proposes 'whenever' on p6l. I'm not immediately charmed by the idea. 10:31
10:31 lest_away is now known as lestrrat
masak I do see where he's coming from, though. 10:32
moritz_ neither am I
masak maybe a good idea for a module.
lisppaste3 quester pasted "Latest rakudo doesn't build for me either" at paste.lisp.org/display/95477 10:35
masak quester_: did you try removing your installed Parrot and doing a clean checkout? 10:36
moritz_ wait
quester_: git pull, and use --gen-parrot
I just fixed that error for latest parrot, and bumped PARROT_REVISION
quester_ masak: Oh... no, I just let Configure.pl do its thing with Parrot. 10:38
moritz_ then your rakudo is too old 10:39
quester_ moritz: Okay, will do. Thank you very much.
moritz_: I did the git pull and perl Configure.pl --gen-parrot again, and got exactly the same error about ...bin/parrot: symbol lookup error: dynext/perl6_group.so: undefined symbol: PObj_active_destroy_SET 10:49
moritz_ quester_: which branch are you on? 10:50
git branch | grep '\*' 10:51
10:52 mikehh left
quester_ moritz: * master 10:52
10:52 mikehh joined
quester_ has mv'ed his problematic rakudo out of the way and is doing a fresh install from scratch... will be done in 20-25 minutes 10:59
uniejo also gets same error for a clean checkout of master branch.
11:00 sathiya joined
sathiya how could i contribute to perl6 development ? 11:00
i know perl5 pretty well ..
oh nobody here to help me ? 11:02
then why is this statement at perl web: Go to #perl6 (irc.freenode.net) and someone will be glad to help you get started. 11:03
pmurias sathiya: hi ;)
sathiya hi ..
pmurias sathiya: you only waited 3 minutes
;)
sathiya yeah yeah .. i thought there would be so much traffic . 11:04
but it is not ..
ok sorry if something is wrong ?
let us get started ?
;)
quester_ I think it's kind of a slow time for #perl6. There is a web page on helping with Perl 6 at www.perlfoundation.org/perl6/index....evelopment
frettled masak: Hee-hee, the Temporal/TZ discussion is really flowing. Who'd have thunk it. 11:05
frettled is glad to have contributed in a small way to some extra insanity, muahahaha.
masak frettled: I'm simultaneously encouraged and scared by the discussion on p6l. 11:06
moritz_ sathiya: what kind of thing would you want to do for Perl 6?
sathiya some contribution like ... 11:07
in development or in documentation
in packaging or release ..
moritz_ sathiya: we need people to write tests, compilers, modules, documentation, a module installer and/or infrastructure...
sathiya oh ok, then i can write some modules ..
masak sathiya: so, you're basically offering up tuits? that's great! let me tell you a little about my projects. :)
sathiya: heard of proto?
sathiya no .. 11:08
moritz_ sathiya: Rakudo has monthly releases, you could volunteer as a release manager
masak finds links
sathiya: github.com/masak/proto/blob/master/README
sathiya: this is the closest to CPAN we have right now.
sathiya ok ..
masak it needs to be replaced. 11:09
badly.
well, actually not as badly anymore, since we recently did a rewrite of it. mberends++
11:09 payload joined, payload left
masak but we would like to usher in an era where S11 is used more in practice. it will require some deep thinking and planning and stuff like that. 11:10
frettled masak: Yes, me, too. I think it's important to have something sane to handle time comparison, but I also think it's important to have something that allows for SUS compatibility.
11:10 payload joined
masak probably also quite some collaboration with Plumage. 11:10
11:10 frettled sets mode: +o masak
sathiya rakudo is for parrot ?! -- right ... but what is that ? 11:10
masak can haz @
11:10 payload1 joined
masak sathiya: ok, so Parrot is a VM. think JVM. 11:10
moritz_ sathiya: Rakudo is a compiler. 11:11
11:11 payload left
masak sathiya: actually, I shouldn't be governing what you want to do. I should let you stay in here for a while and soak in the environment. eventually you will find something you like. 11:11
lunch &
11:12 payload1 left, payload1 joined, payload2 joined
sathiya @masak: ok, let me do that ... thanks for your help anyway. 11:13
JVM for java, Parrot for Perl ..
masak sathiya: welcome! trust me, there are things to do... :)
11:17 payload1 left
quester_ My make died with ...rakudo/parrot_install/bin/parrot: symbol lookup error: dynext/perl6_group.so: undefined symbol: PObj_active_destroy_SET ... make: *** [perl6.pbc] Error 127 ... again, on a clean install; it's the same error as my nopaste about 40 minutes ago. 11:18
moritz_ quester_: what's the revision number of your parrot? 11:21
11:21 ggoebel joined
quester_ moritz_: It's r44371. 11:22
moritz_ hrm 11:23
quester_: did you re-configure after 'git pull' in Rakudo?
quester_ moritz_: Um... do you mean perl Configure.pl --gen-parrot? If so, yes, I did. 11:25
moritz_ that's very odd 11:26
quester_ moritz_: In the Amsterdam release, Parrot defined it: /parrot/include/parrot/pobj.h:#define PObj_active_destroy_SET(o) PObj_flag_SET(custom_destroy, o) 11:30
moritz_: That doesn't appear to be in the current Parrot. Was it supposed to be replaced with something else?
11:33 ggoebel left
lisppaste3 quester pasted "Parrot pobj.h diff between Amsterdam and tonight's Rakudo" at paste.lisp.org/display/95482 11:34
jnthn moritz_: Eww, that test has tripped me up before. :-/ 11:35
moritz_ quester_: github.com/rakudo/rakudo/commit/94a...46e4cf06b0 11:41
that's what you should get when you do a 'git pull'
and that should fix the build error for you 11:42
quester_ Okay, thank you again!
moritz_ quester_: when you do a 'git show', what's the first line of the output? 11:43
quester_ moritz_: commit 2c4f69b24a8d5d17d33570f5ef13c6003f597fa0 11:44
moritz_ that's the latest 11:45
quester_ moritz_: It's doing the make now...
moritz: So it was just a change in the way the bit was named in Parrot? 11:47
moritz_ yes
and if you always used the parrot revision recommended by build/PARROT_REVISION, I have no idea why you should have ever seen a build error 11:48
quester_ Hmm... yes. I've just been letting Configure.pm --gen-parrot take care of that. 11:49
moritz_ and that should build the correct parrot rev, yes
11:50 payload2 left, payload1 joined, payload joined
quester_ moritz:_: I just did a grep -ri PObj_active_destroy_SET . (in the rakudo directory) and found ./src/pmc/objectref_pmc.template: PObj_active_destroy_SET(SELF); 11:51
moritz_: and got another dynext/perl6_group.so: undefined symbol: PObj_active_destroy_SET when the build ended. I will try vi'ing that on my local copy and see what happens. 11:52
moritz_ quester_: I missed that occurrence 11:53
(because I used 'ack PObj_active src', and ack ignores file types it doesn't know about_
s/_/)/
testing now, and I'll push if it works 11:54
quester_ moritz_: Oh, I see. I think we've arrive in Wonderland now. [Re: Haddocks' Eyes / The Aged Aged Man / Ways and Means / A-sitting on a gate]
s/arrive/arrived/ 11:55
moritz_ stupid 11:56
I've also changed that in generated .c files
so the .template file wasn't used for building the .c files, they had a newer time stamp than the .pmc and .template files
=> my build worked, everybody elses failed
dalek kudo/master: 4071ac3 | moritz++ | src/pmc/objectref_pmc.template:
catch another instance of PObj_active_destroy_SET noticed by quester++
 11:57
quester_ Is going to back to git pull. The make I did with the change vi'ed locally worked fine. 12:01
uniejo A new checkout of master branch now compiles fine. 12:08
Thanks. moritz++ quester_++ 12:09
moritz_ uniejo++ for trying it out
quester_ moritz_: I'm sorry, do you mean that the generated .c files are distributed with a git pull and are also supposed to be regenerated by make...? 12:10
moritz_ quester_: no
quester_ uniejo: You're very welcome.
moritz_ quester_: I changed them locally with a perl -pi -e 's/../../' src/*.pmc src/*.c
quester_: and that gave them newer time stamps, so 'make' did not regenerate them 12:11
quester_: so the (still wrong) .template file did not matter, and my build went to completion
quester_ moritz_: D'oh. Of course. So... a make clean; make would have caught that, but not just a make? (I'm thinking I'm going to make the same mistake one of these days Real Soon Now.) 12:13
moritz_ quester_: right 12:14
or even a new Configure.pl would have helped
that also cleans
quester_ moritz_: I see. Thanks again! 12:15
moritz_: BTW, after the git pull both make and make test worked. 12:16
moritz_ great
colomon phenny: tell lue Looks like you did a nice job on minmax.t. lue++ 12:17
phenny colomon: I'll pass that on when lue is around.
quester_ moritz: It's been fun, and educational. Good localtime; I'm off to bed now. 12:18
moritz_ sleep well
12:19 quester_ left 12:21 k23z__ joined, sathiya left 12:22 lambdabot joined 12:24 bluescreen joined 12:32 riffraff left 12:34 ggoebel joined 12:37 pmurias left
colomon did the spectests just get faster again? I just got 424 wallclock secs, which is the best I can remember it being since trig was added back in. 12:37
12:42 PZt left 12:44 ggoebel left
moritz_ 437s here (on two cores) 12:45
I think bacek++ worked like mad on the GC, maybe that gave us some improvements :-) 12:46
12:47 ruoso joined 12:51 masak left
ruoso Bom dia, #perl6 12:52
colomon o/ 12:53
12:54 avar left 12:55 avar joined
ruoso pmichaud, re your iterable commit -- note that evaluating the number of elements in an iterable might require early evaluation of its sub-iterators... 12:56
so maybe the iterable should be allowed to return a NotYetEvaluated failure when asked for the number of elements... 12:58
and you have a special way to force the early evaluation if necessary
13:03 clintongormley joined 13:18 ruoso left 13:22 ignacio_ joined 13:29 k23z__ left
takadonet morning all 13:29
frettled good TAI ;) 13:30
13:31 iblechbot left 13:34 nbrown left
jnthn
.oO( At a TAI restaurant, you always know exactly when the food will come )
 13:34
13:34 nbrown joined 13:35 ruoso joined
m-locks that proto needs to be updated for some small changes 13:39
in Install.pm s/has Ecosystem/has ::Ecosystem/, and also s/undef/*.notdef/ of course 13:40
13:44 riffraff joined
ruoso TimToady, After yesterday's argument on datetime, I think I came to the conclusion that providing a coercion from Instant to Numeric is just as misleading as it would be providing a coercion from Str to an utf8 buf automatically... So I'd like to propose dropping that coercion entirely... 13:54
13:54 masak joined 13:56 jferrero left
masak re the names on www.perlfoundation.org/perl6/index....akudo_star -- I think 'Rakudo star' should have a capital 'S'; and I've never actually heard the name 'Rakudo Whatever'. 14:07
I'm a big fan of the description 'pivotal release'. it captures perfectly what it is.
14:08 stepnem joined
masak it's "Lisbon" in English. 14:08
also, the final quote doesn't really provide an excellent example of what the earlier part of the paragraph talks about. 14:09
m-locks aye
masak the proper term for 'almost quote' is 'paraphrase'. 14:10
m-locks masak: proto didn't build, small changes needed to be made in Install.pm
masak m-locks: trying to build it on alpha or Rakudo master? 14:11
m-locks master i guess
or from that git
s/has Ecosystem/has ::Ecosystem/ and s/undef/*.notdef
then it works :)
masak m-locks: I think you're the first to attempt that. :) 14:12
but if those are the only changes required. I'd guess you're on alpha after all...
m-locks but it started to build perl6 even if i had existing perl6 installed in the dir already
masak yes. 14:13
m-locks so i kinda never made to the end
masak I don't know exactly what you mean by that.
m-locks i hit ctrl-c cos i didn't want to build perl6 again, cos i already ahd built it in the dir
masak 'in the dir'? 14:14
m-locks the dir in config.proto
that says here it is supposed to be
masak oh, I see.
yeah, it shouldn't do that.
consider filing a bug report to the github issue tracker about that.
m-locks i shall look into it
masak we really shouldn't be caring about installing perl6 at all anymore if we find it installed globally. 14:15
proto is a bit behind the times in that way.
m-locks i see
masak m-locks: I'm backlogging right now. I see you're checking out a number of Perl 6 projects. 14:19
14:19 k23z__ joined 14:20 clintongormley left
masak m-locks: just be aware that all of those have been written for Rakudo alpha, and (as far as I know) none of them has been rebased to work on the new Rakudo master. 14:20
m-locks oh ok
it explains the changes required
masak in other words, if you're testing things with the aim of getting things to work, you'll be a great lot happier with Rakudo alpha than with Rakudo master right now. :)
which brings me to today's little project.
I'm planning to set up a nightly build of November, Form.pm and Druid. 14:21
build them on alpha, run the tests, and report successes and failures in a cute HTML table. 14:22
ditto Rakudo master in a table next to it.
the goal being that the 'master' table should be 100% green by the March Rakudo release.
14:23 hanekomu joined
m-locks is that git clone of rakudo an alpha release? 14:23
masak which git clone?
the one that proto makes?
by the way, I use the term 'alpha' here to refer to 'the old development branch of Rakudo, the one that was actively developed between 2006 and January 2010'. 14:24
m-locks oh ok 14:25
masak alpha: say "Here I am! I'm old!"
p6eval alpha 30e0ed: OUTPUT«Here I am! I'm old!␤»
m-locks so i guess it is the master then
masak yes.
m-locks that i dug up
yes, all fine and good
masak it's master that's eventually going to become Rakudo Star.
m-locks i hear ya :)
k23z__ has anyone here tried solving the rubik's cube with A* ? is it feasible ?
masak k23z__: hah, I actually tried that many many years ago. 14:26
k23z__: I ran out of hard drive space. :P
at depth 5 or 6 or so.
k23z__ masak, vimeo.com/9640169
masak, in Perl5
masak, I'm not sure if every configuration is 6-solvable or 5-solvable
masak I did it in Turbo BASIC, IIRC>
s/>/./ 14:27
14:28 pmurias joined
pmurias ruoso: hi 14:28
ruoso: under what circumstances can we inline a block 14:29
masak k23z__: I'm pretty sure it isn't. especially not if you don't count 180 degree rotations as one move. 14:31
14:31 colomon left
masak for me, it was mostly a fun thing to try. I was young and naive. :) 14:31
14:34 ggoebel joined 14:37 rgrau joined
PerlJam good $localtime #perl6 14:37
m-locks to you too :)
masak PerlJam: \o
<TimToday> minmax should probably be redefined in terms of Range objects 14:45
as in, 'minmax should get a new definition that makes use of Range objects'? 14:46
14:47 ignacio_ left, iblechbot joined 14:51 bkeeler joined 14:54 mssm left 14:58 Mips128 left 15:03 iblechbot left 15:06 Guest44360 left 15:09 ignacio_ joined 15:11 justatheory joined, masak left 15:12 masak joined, rv2733 joined 15:13 mssm joined
m-locks FFFFUUUUUUUU 15:16
Error: Target org.macports.fetch returned: ghc is not yet supported on Mac OS X 10.6.x (SnowLeopard)
so no pugs for me
PerlJam for what do you need pugs if you have rakudo? :) 15:17
m-locks just to try it out, but seems that i don't need it 15:18
15:18 crazed left, crazed joined
m-locks pugs svn had all the synopsis and other documentation which was nice to get 15:19
rakudo does not provide such documentation
at least i was unable to find any specs 15:20
the tests are good pieces of code, though
PerlJam Hmm. Interesting observation.
Should Rakudo (or Rakudo *) bundle the Synopses as part of the distribution? 15:21
m-locks i would say yes
PerlJam Certainly yes for Rakudo *
(IMHO)
m-locks it would need a doc browser though
pmurias m-locks: you can checkout the synopsis from the pugs repo without installing haskell
m-locks well yes, but i needed to download them, my internet access is limited from time to time 15:22
15:22 charsbar left, charsbar joined
pmurias you need to install Rakudo too 15:23
m-locks that i have done yes :)
but was looking for language synopses
pmurias meant download
you are aware of perlcabal.org/syn 15:24
?
m-locks yes, but as i said my internet access is limited
pmurias limited? 15:25
m-locks it's on and off
let's just say i don't have one
15:25 ash_ joined 15:31 vorner joined
masak m-locks: I'd recommend doing a checkout of svn.pugscode.org/pugs/docs/Perl6/Spec 15:31
m-locks: I use git-svn for that, but SVN is fine too if you don't plan to write a lot to the spec. :) 15:32
Su-Shee plans for rakudo * are still april, yes? 15:33
m-locks yes, but that's non-wgettable
because of robots.txt 15:34
masak m-locks: it requires an SVN client.
m-locks oh ok, i can svn just the perl6 spec?
masak Su-Shee: yes, but not necessarily the April release, IIU pmichaud C.
PerlJam m-locks: yep
masak m-locks: yes, that was my point. (and PerlJam's)
m-locks check, i was new to svn 15:35
rgrau is there any doc browser on the way? u4x?
Su-Shee and is there a place where I can follow something like DBI6?
15:35 hanekomu left
PerlJam Su-Shee: no one is writing DBI for perl 6 that I know of. 15:36
Su-Shee: there has been some work with SQLite and such though
masak rgrau: I have half-baked plans for u4x. as a part of keeping the effort honest/visible, one of the first things I'll do is probably HTML serialization, so that one can browse the u4x documentation online.
15:36 mberends joined
jnthn is back 15:36
PerlJam Su-Shee: and someone (simon?) looked at getting DBDI working with Rakudo
masak rgrau: but right now, each time I think of u4x, I realize that that time should go into the Perl 6 Book, which has a tighter schedule.
jnthn omg book1 15:37
jnthn needs to give tuits to that.
Su-Shee PerlJam: didn't someone say a couple of months ago that dbi6 will be modeled after jdbc and stuff?
m-locks what happened to that doing java2perl6 on some java DB*
mberends oh hai masak, jnthn, Su-Shee
jnthn Su-Shee: afaik, that's still The Plan.
Su-Shee PerlJam: the sqlite stuff I know of, yes.
jnthn Or at least, it seemed to be last time I tallked to tbunce
Su-Shee jnthn: hm, i see. 15:38
rgrau masak: yep... I may have some tuits in near future. maybe I could give a try hacking a bit on u4x. I'll let you know if that happens
jnthn yayitsmberends!
PerlJam Su-Shee: DBDI is the low-level interface that gets us database drivers, DBI is built entirely on top of that (or would be)
Su-Shee hi mberends :)
jnthn Su-Shee: well actually...
15:38 ggoebel left
jnthn Su-Shee: More that the jdbc *driver* interface would be taken. 15:38
PerlJam Su-Shee: and then, of course, things like DBIC woudl be build on top of that :)
jnthn Su-Shee: The high-level bits that everyday folks use would be rather more Perly. :-)
Su-Shee PerlJam: I've just summized "perl 6 database bindings under "DBI6", sorry.
PerlJam Su-Shee: there's at least 3 levels of play here and I'm not sure anyone is actively working on any of them. 15:39
15:39 [particle] joined
Su-Shee PerlJam: hm. 15:39
jnthn should try and get Blizkost to work again, so at least DBI for Perl 5 is usable. 15:42
PerlJam too many interesting projects, not enough hours in the day
jnthn PerlJam: Spot on.
mberends jnthn: trying to bring back @*ARGS out of builtins-old/guts.pir into glue/run.pir fails, because the new args pmc does not seem to have a 'list' or an 'Array' method. Any idea where I should look for a solution? 15:43
masak mberends: good TAI to you :) 15:44
jnthn: [re tuits to book] you and me both :)
mberends masak: good CET to you :)
masak rgrau: having synergetic collaboration with someone over u4x might be just what that project needs.
PerlJam masak: have you been working on the book "off-line"?
masak PerlJam: does thinking about it count? :) 15:45
m-locks lol
PerlJam masak: as logn as there some writing in the future before the tentative publication :)
m-locks getting there...
PerlJam has been worrying over the book lately.
I've been meaning to write stuff, but Other Things keep coming up
masak PerlJam: indeed. it's been dormat of late. 15:46
PerlJam: same here.
PerlJam: my current plan is to write stuff up as Rakudo features come back online.
for example, I have the nice poker hand example with enums and subtypes...
masak goes to find the gist...
mberends is focusing on getting proto working with new master 15:47
m-locks does it not work
pmichaud good morning, #perl6 15:48
PerlJam good morning pmichaud
m-locks good morning, sir
15:48 jferrero joined
masak pmichaud: \o 15:48
this one. gist.github.com/244255
15:48 Psyche^ joined
masak kinda proud of that. 15:48
mberends m-locks: unfortunately proto does not work with new master, for quite a few reasons
masak but it's written with old-skool enums. I'd like to get new-skool enums working in Rakudo master, and then write up a chapter on that. :)
PerlJam masak: yeah, that's nice (except you found the one with an extra face card ;) 15:49
masak mberends++ # proto tuits
PerlJam: 'an extra face card'?
oh right!
m-locks mberends: i managed to compile it but didn't try it yet to actually setup any libs 15:50
ash_ my nq-nqp parser can now parse most of the test files for nqp without error, now to move onto node building so i can walk the node's to build the code
masak should fix that
so, 'jack' is the correct name for 11, right?
mberends plans to hijack jnthn++ next week to make proto work again
PerlJam masak: yes. 15:51
masak: (though "knight" works for me in lieu of "jack" too)
masak PerlJam: fixed.
15:51 Patterner left, Psyche^ is now known as Patterner
mberends m-locks: it won't work, because for example it cannot read the PERL6LIB environment variable yet 15:51
m-locks ok, what about tweaking it to use config.proto 15:52
jnthn mberends: Yes, we can hack on proto things. :-) 15:53
mberends: That would be cool. :-)
masak \o/
jnthn morning, pmichaud 15:54
mberends m-locks: that will be a fallback strategy for working around anything that alpha has, that we cannot put into master. But initially proto can be used as a medium size test case of everythingg that master *should* be able to do.
m-locks env variables are so last millenium
jnthn masak: On book, for me a lot of it is that Rakudo itself has been in need of Epci Attention of late.
mberends: (@*ARGS) maybe make a Parcel and .Seq it. 15:55
oh, wait, is @*ARGS meant to be mutable, I wonder
masak jnthn: yes. understandable, and predictable. (or we should have predicted it, at least.)
m-locks mberends: aye, it seems nice if it works
jnthn masak: Well, I sort of did and din't.
mberends jnthn: thanks, that's just the sort of thing I can investigate during de-commute()
masak jnthn: it's nice if @*ARGS is mutable. then one can shift things off it. 15:56
jnthn mberends: OK, what masak++ said. I reckon on maybe just taking the PMC array and trying like
$P0 = new ['Array']
$P0.'!STORE'(the_rpa)
Or some wuch
mberends :) thanks, that mey be enuf to go on 15:57
speling--
jnthn masak: ng took longer than I expected, and I hadn't anticipated needing some weeks anti-burnout break over Christmas/New Years/early January. :-)
masak: Anyway, Rakudo is getting into better shape again now. :-) 15:58
masak jnthn: Viem. :)
jnthn masak: btw, what is "skrattande"?
I parsed your twit up until that word :-)
masak jnthn: 'laughing'.
jnthn ah. :-) 15:59
masak it's a gerund in this case, I'd guess.
mberends "schateren" in .nl :-)
PerlJam "ande" eq "ing"?
masak PerlJam: yes. both for gerunds and participles. :)
jnthn ooh, handy to know :-) 16:00
masak PerlJam: actually, it's / <[ae]>nde /, depending on the verb type.
m-locks en skrattande h􏿽xE4st
PerlJam so in the early days of IRC I picked up Finnish words from all the Finns that were around. Perhaps though masak's tweets I'll resurrect that tradition :)
masak m-locks: Klad Hest! :) www.snelhest.se/ 16:01
m-locks voidaanhan me suomeakin puhua :)
jnthn
.oO( En skrattande rulltrappa. )
16:01 pyrimidine joined
masak jnthn: :D 16:01
jnthn Mental image win. :-)
m-locks lol @ hestmatematik
masak m-locks: you from .fi? 16:04
16:08 ignacio_ left
m-locks masak: aye 16:09
hejsan
masak hej, hej :) 16:10
Uppsala, Sweden here.
Tene Huh. I never knew where you were from, masak.
masak Tene: you never asked :P
Tene I guess I could have asked, but it never occured to me.
Yeah. :) 16:11
masak I like that about IRC.
apart from the waketimes (and sometimes not even from that), you can't tell where people are from. and it doesn't matter much either.
jnthn never made it to Uppsala yet. Maybe I have to visit and distract masak++ from bug filing for a bit :-)
m-locks haha 16:12
masak the only other phenomenon I know which does that flattening-out of nationality, besides the Internet, is Esperanto.
16:12 ash_ left, ash_ joined
masak jnthn: I'd be happy to have you come here and try to distract me from filing bugs :P 16:12
16:12 ash_ left
pmichaud notes the word "try" 16:12
masak pmichaud: yes. I've really streamlined the process by now. 16:13
pmichaud: you said during a talk that you suspected that I have a script for filing bugs. that's not exactly true. I *am* a script for filing bugs. :P
Tene masak: any process that doesn't have an IRC bot can still be further streamlined. 16:14
pmichaud Tene: masak is also an IRC bot :)
masak makes happy circling motions in the air
TimToady is a Usenet bot that got out of hand 16:15
Tene I was an IRC bot for a while, but then I met people IRL.
pmichaud: around what time of year might the hypothetical west-coast hackathon happen? 16:16
masak that sounds like a false dichotomy to me :)
pmichaud Tene: I'm open for suggestions
I had been hoping to do it pre-Rakudo *, but my schedule is rapidly filling up :-(
anyway, my guess is that it'll likely be May timeframe; otherwise we're looking at July/OSCON sort of thing I guess 16:17
there's a small possibility that I'll be in California for an extended period of time in the next couple of months, however. 16:18
Tene If things go well, I'm probably moving to CA during March.
TimToady any of you are always welcome at our house
jnthn masak: Man, you make a visit sound most challenging.
pmichaud TimToady: thanks -- very possible I'll take you up on that :)
m-locks west-coast hackathon sounds epic
Tene shows up at TT's at 4AM 16:19
jnthn pmichaud: fwiw, if it's more July-ish and there's funding, there's a chance I can make it.
pmichaud jnthn: that would be OSCON, then. And any funding would likely be used to get me to OSCON in that case :)
jnthn: but I'll see if we can get you there also
that would be awesome
jnthn pmichaud: OK, well, don't plan around me too much.
masak jnthn: if you come here, you can admire my Japanese table (the only piece of furniture in my room). 16:20
jnthn pmichaud: But it's more likely I can make something in June/July than in May.
masak: Minimalism win. :-)
pmichaud jnthn: okay. between German Perl Workshop and YAPC::NA, June is somewhat booked already
jnthn pmichaud: ah, ok
pmichaud (and YAPC::NA won't be "west coast" :-)
jnthn nod 16:21
masak jnthn: it's funny. I looked at all my furniture in 2003, and then I threw it out. haven't looked back since.
jnthn: now, a floor on the other hand, that's a versatile thing. you can use it for paperwork, hosting guests, or ballroom dancing :) 16:23
jnthn You make a good point :-)
pmichaud masak: yes, I think every living area should have one.
masak pmichaud: :)
16:23 ignacio_ joined, PZt joined
TimToady floors make /me feel very old very fast 16:24
pmichaud masak: (poker example in p6) that is awesome
masak thank you. 16:25
will be even awesomer with the new enum semantics.
pmichaud masak: is it in an article somewhere?
masak no.
I actually don't remember why I wrote it :)
PerlJam masak: because it was awesome
masak I can go back and check the logs. 16:26
pmichaud perhaps as a bug locating device?
masak no, don't think so.
mberends de-commute() &
16:26 mberends left
masak I wanted to try subtypes a bit, in a... non-bugfinding way. 16:26
Tene masak: do you have a bed, or do you use the floor for sleeping too? 16:29
masak Tene: mattress.
the bed was actually the first piece of furniture I threw out. it was awful.
nowadays, it feels slightly odd to sleep in beds, when I'm a guest at other people's. 16:30
Tene That's what I did at my last place. My bed now is a set of drawers with a mattress on top.
(I'm really low on space in my current apt.)
m-locks my bed is a stack of 3 mattresses on top of each other 16:31
PerlJam Hammocks make nice "beds" :)
mdxi i went to an estate sale and scored a really nice old bed with a basically unused boxspring and mattress for $170. and an amazingly seasoned cast iron frying pan.
dead people have the best stuff 16:32
Tene www.ikea.com/us/en/catalog/products/80121338
masak pmichaud: seems I wrote it Just For Fun. irclog.perlgeek.de/perl6/2009-11-27#i_1777770
m-locks lol @ cuteness driven development 16:34
perl6 monks do it zen style
TimToady
.oO(make me one with Whatever)
 16:35
PerlJam TimToady: here's a Whatever sandwich for you |*| 16:36
jnthn std: |*| 16:37
p6eval std 29806: OUTPUT«===SORRY!===␤Missing term at /tmp/dKpCeHwa4I line 1 (EOF):␤------> |*|⏏<EOL>␤ expecting any of:␤ POST␤ bracketed infix␤ infix or meta-infix␤ postfix␤ postfix_prefix_meta_operator␤ prefix or term␤ term␤FAILED
..00:01 108m␤»
masak std: sub circumfix:<| |>() {}; |*|
p6eval std 29806: OUTPUT«===SORRY!===␤Missing term at /tmp/x5gi1pg57f line 1 (EOF):␤------> sub circumfix:<| |>() {}; |*|⏏<EOL>␤ expecting any of:␤ POST␤ bracketed infix␤ infix or meta-infix␤ postfix␤ postfix_prefix_meta_operator␤ prefix
..or term␤ term␤FAILED 00:01 11…
TimToady STD doesn't do circumfixes
jnthn alpha: sub circumfix:<| |>() {}; |*| 16:38
masak oh :/
p6eval alpha 30e0ed: OUTPUT«Confused at line 10, near "|"␤in Main (file <unknown>, line <unknown>)␤»
jnthn aww
:-)
masak neither does Rakudo...
masak submits rakudobug
TimToady: why doesn't STD.pm? is it particularly difficult?
PerlJam Now that I look at |*|, it occurs to me that it might have made a good reduction op
|+| 1..$n
masak still not too late...
(but put it in a module) 16:39
TimToady [*] is easier to type on my keyboard :)
and it fitss with the use of [op] on any op
fits, even
and you can't have a circumfix that starts the same as a prefix, anyway 16:40
PerlJam yeah, |||| would have been a bit of pain as compared with [||] :)
sbp ikea — my %tools{"Allen"};?
TimToady language design is hard, let's go study math 16:42
PerlJam TimToady: so, using |...| for absolute value is out ? :)
16:42 nihiliad joined
masak PerlJam: we could always make a module that serializes Perl 6 code to LaTeX, and changes (...).abs to |(...)| 16:43
16:43 nihiliad left, nihiliad joined
m-locks heh 16:44
Tene perl6 --target=LaTeX
TimToady you might be able to make |...| work if you defined a postfix:<|> that redefined a prefix:<|> on the same term
16:44 lisppaste3 joined
TimToady but such a reverse redefinition would be considered a language design smell 16:45
masak no kidding.
PerlJam TimToady: be sure to codify those design anti-patterns for future users of Perl 6
:-)
masak TimToady's Smell Of The Day. 16:46
16:46 nsh joined
PerlJam kinda like Bertie Bott's Every Flavor Beans except with odors 16:47
masak the puns of late really stink.
16:47 Achilles333 joined 16:48 frettled sets mode: +o masak
PerlJam I guess that would be largely similar to MJD's red flags stuff 16:48
16:48 frettled sets mode: +o PerlJam
frettled is a rampant kanelbullebot. 16:48
Achilles333 hi folks
masak frettled++
Achilles333: hi!
Achilles333 How can I contribute to perl6?
frettled You just did! :D
masak Achilles333: first off, you can stay here and enjoy the atmosphere. :)
PerlJam Achilles333: what do you want to do? what are you interested in?
m-locks could you code a working DBI wrapper
16:49 barney left
masak m-locks: I did that :) no-one seems to care... 16:49
m-locks masak lol
ok so umm
we have DBI module?
jnthn masak: Oh?
masak Squerl!
come on, people, keep up :)
jnthn Oh.
frettled masak: I think some of us went «oooh, yay!» when we saw it, though.
jnthn remembers that name
PerlJam doesn't remember squerl at all 16:50
masak it's in Web.pm. it's seriously undertested.
m-locks we need that 6PAN already...
masak I'd be very happy to see people pick it up.
Tene masak: but it's not *called* Dee Bee Eye, so it doesn't count.
TimToady perhaps you should find a good tester
m-locks masak: is it documented?
Achilles333 perlJam: I am interested in coding, documentation and organizing/cleaning things
m-locks documentation would rule...
masak m-locks: only in blog posts :/
m-locks: shall I dig them up for you?
m-locks so here we go!
masak: please :) 16:51
masak does so
m-locks Achilles333 could rip all blog documentation and compile them into module documentation files
specifying which sub and method did what
frettled And test Squerl and look at MySQL and PostgreSQL implementations? :D
PerlJam or work on u4x
m-locks u4x mos def 16:52
Achilles333 m-locks: Ok. give me more pointers to acheive to do that
Tene Achilles333: for whatever you do, you'll probably want to start by downloading and compiling Rakudo.
PerlJam Achilles333: what Tene said
Tene rakudo.org/how-to-get-rakudo 16:53
PerlJam just realized that he has a latent small negative bias towards the name Achilles that comes from reading too much Orson Scott Card
masak m-locks: use.perl.org/~masak/journal/39686 and then use.perl.org/~masak/journal/39894
Tene Many people start contributing by adding and organizing tests.
m-locks thx m8
TimToady sentences PerlJam to reading GEB 16:54
masak seconded.
TimToady not that Achilles comes off terribly well there either...
Tene I need to finish GEB someday. I got distracted after the first half or so and haven't gotten around to finishing it.
TimToady but better than the Illiad...
masak frettled: Squerl is at least made to support many SQL implementations. that's part of the beauty of Suquel, its port-parent.
m-locks btw does that november wiki support SQL 16:55
masak m-locks: no.
m-locks ok
masak but if Squerl becomes mature enough, then sure...
Tene Achilles333: the spec tests are available over svn at svn.pugscode.org/pugs/t/spec/
frettled masak: port-parent, me like 16:56
PerlJam If I hadn't just read "Ender in Exile" a couple of weeks ago, I probably wouldn't have the bias. It's activation energy would have been far to low to register
Achilles333 Tene: Ok. Downloading Rakudoooo....
m-locks how to get squerl? or is it embedded in some module 16:57
frettled Mumble, looks like I won't make it to YAPC::NA this summer, too much goings-on. Maybe I'll try for OSCON instead.
m-locks: Web.pm
PerlJam pmichaud: is it snowing in Plano? Apparently it's snowing in Austin
pmichaud PerlJam: no, the snow all went south of here, I think. 16:58
Tene YAPC::NA would be fun to attend.
16:58 payload1 left, payload left
pmichaud yes, looks like all of the precipitation went south of us. According to radar, Dallas is getting ice/snow/rain mix right now, but it's completely dry here. 17:00
masak m-locks: it currently comes with Web.pm. 'proto install web' will install it.
TimToady radar shows a big storm hitting us just now, should start raining any minute now 17:01
frettled Yay.
Well, today has been a positively _balmy_ -7 °C on the riverside, so I'm quite happy.
…except for the soaring power prices.
PerlJam needs to move out of Corpus Christi again.
frettled heh 17:02
TimToady it has definitely been an El Niño year for us, which we really needed after four years of drought
frettled The system price is down from € 134.80/MWh (Monday) to 92.70 (Wednesday), though. 17:03
17:03 kensanata joined
frettled TimToady: four years? Yikes. 17:03
I've sort of noticed the unrelenting brushfires, but, uhm. 17:04
17:04 colomon joined, bluescreen left
frettled Uhm, nevermind, the prices I cited were for the entire Nordpool area. Oslo's prices are roughly 50% higher. 17:05
17:06 bluescreen joined 17:09 cotto_work joined
[particle] el niño here means sun, crocuses in january, and daffodils in february 17:13
and spring won't be over until july
m-locks wheres that 17:15
pmichaud el niño here means "the boy" :-) 17:16
TimToady yes, well, it does in Peru too, where they named it
but I didn't realize that Spanish was the national language of Texas :P 17:17
masak maybe we should call them "Spanish quotes", then.
pmichaud it's a component of Texas' national language, I think :)
17:18 cdarroch joined, cdarroch left, cdarroch joined
PerlJam we've ceceded and no one told me?!? 17:18
pmichaud > my @a = <the 5 quick brown foxes>; say @a.sort.join(' ');
5 brown foxes quick the
lambdabot <no location info>: parse error on input `='
colomon \o/
17:18 pmichaud sets mode: +b *!*lambdabot@*.members.linode.com, lambdabot was kicked by pmichaud (i hate you))
masak :) 17:18
pmichaud > my @a = <the 5 quick brown foxes>; say @a.sort({ +$_ }).join(' ');
the quick brown foxes 5
jnthn pmichaud: Yay, sort! :-D 17:19
pmichaud I actually had sort working (1-liner) last night.
17:19 ggoebel joined
pmichaud The part that makes it longer is handling the arity-1 case 17:19
colomon indeed.
pmichaud (I kept running into bizarre null pmc and type mismatch errors)
17:20 colomon left
jnthn Odd 17:20
TimToady so, if we can detect $_ usage in a closure, why can't we detect it in ~~ .truthy?
pmichaud at present we're not doing that
> my %h = <a 1 b 2 c 3>; say %h.perl; 17:21
Method 'perl' not found for invocant of class ''
:-(
17:21 payload joined
jnthn NYI 17:21
17:21 payload1 joined
pmichaud decides to eliminate the N after lunch 17:22
(unless someone beats me to it :)
17:23 ggoebel left
TimToady actually, I guess we don't detect $_ usage in a closure; it's just the way the arg defaults, so never mind :) 17:23
pmichaud right.
m-locks oh yeah you could hack that proto working in master 17:24
now saying Null PMC access in get_iter()
Tene I just sent a commitbit for pugs to Achilles333.
masak m-locks: I'd recommend using proto with Rakudo alpha for now :)
m-locks yes, i'm just too lazy for makeinstalling alpha ;) 17:25
masak laziness can be a very useful force multiplier. so, um, keep up the good work, I guess. 17:27
m-locks haha, just living up to the principle 17:28
17:29 athenot joined 17:30 rgrau left
norg 2 17:31
17:31 stephenlb joined, riffraff left
[particle] m-locks: seattle 17:31
dalek kudo/master: 069df7a | pmichaud++ | src/core/Seq.pm:
Initial version of Seq.sort . Seems to have issues with
 17:32
kudo/master: 9b33a8d | pmichaud++ | src/core/Seq.pm:
Refactor Seq.sort to resolve some arity < 2 bugs (and separate
17:33 ignacio_ left 17:34 Achilles333 left, meteorjay joined
masak rakudo: subset ReallyLargeInt of Int where { $_ > 1e12 }; say 2e12 ~~ ReallyLargeInt 17:36
p6eval rakudo 4071ac: OUTPUT«Confused at line 11, near "subset Rea"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
masak ok. oh well.
jnthn Didn't put back subset yet.
masak so after I implement named enums, should I do subtypes, perhaps :)
s/ps/ps?/
jnthn If you don't get beaten. :-) 17:37
pmichaud
.oO(we're down to the point of beating our developers? shudder)
17:38 riffraff joined
jnthn er, beaten *to it*. :-) 17:38
masak jnthn: :P 17:39
jnthn: English is your native tongue, that what you said? :)
oh right, it was Lolcat. now I remember.
jnthn Plz I can haz extra English classes? 17:40
PerlJam pmichaud: why did you use .? in the call to &by.?arity ?
17:41 rgrau joined
pmichaud some of our subs (multis) don't have .arity yet 17:41
jnthn
.oO( what's the arity of a multi... )
PerlJam jnthn: what does the spec say? :)
jnthn PerlJam: I can't recall that it akshually does. :-) 17:42
PerlJam .count is all of the parameters and .arity are the required parameters, right?
PerlJam should read the spec
pmichaud time for lunch 17:43
m-locks S06-signature/arity.t
jnthn PerlJam: what does the spec say? :)
:-)
17:43 riffraff left
jnthn PerlJam: But yes :-) 17:43
m-locks we should have a .pod-reader here that searches for operators and such 17:44
from specs
PerlJam how do you syntactically distinguish between multis for purposes of calling .arity on them? 17:45
m-locks The number of required parameters a subroutine has can be determined by calling its .arity method
TimToady use of .arity is a Bad Smell, except perhaps for implementing infix:<...>
PerlJam I think it fits well for use in .sort 17:47
TimToady yes, there too, but not for anything map-like
jnthn PerlJam: &multisub.candidates
PerlJam Hmm. 17:49
17:49 azawawi joined
azawawi hi 17:49
PerlJam how would I say (for sorting), "use this one particular multi"? like now we can say things like @a.sort(&foo), would I have to say something like @a.sort(&foo($,$)) ? 17:50
assuming foo is a multi
jnthn There is some NYI syntax for that, I forget exactly how it looks.
17:50 cdarroch left
jnthn Maybe &foo<Any,Any> or some such. 17:50
PerlJam candidates are only distinguished by type? i.e. multi sub foo ($a is ro, $b) { }; multi sub foo ($a is rw, $b) { } are indistinguishable and triggers an exception ? 17:52
jnthn btw, s/ro/readonly/ and it's default anyway 17:53
rw should be a bit tighter
Actually, given I've refactored the binder, that may be implementable. 17:54
Quite easily.
17:54 alester left
jnthn (If anyone wants to play, it's probably just a couple of lines in Perl6MultiSub to tweak the sub that compares the narrowness of two candidates.) 17:55
pugssvn r29807 | azawawi++ | [Cursor] The syml folder is now created properly inside the temp directory 17:56
PerlJam "is rw" is considerent a constraint?
jnthn Yeah 17:57
PerlJam Hmm.
masak kinda makes sense.
jnthn I was planning to just consider it one.
The other thing that needs doing in that bit of code (also easy) is that for multi foo(|$c (Int $a, Int $b)) we need to promote the nested signature into the one that the multi-dispatcher cares about. 17:58
PerlJam then, according to the spec, tie-breaking is handled by declaration order
jnthn PerlJam: Right.
17:58 dakkar left
jnthn PerlJam: But "has a constraint" is always tighter than "no constraint" 17:58
And is readonly = not a constraint. 17:59
It's just the default
So :($x is rw) will always sort into a tigheter narrowness level than :($x)
The only time you'll hit the declaration order fun is if you have a 18:00
:($x is rw) and a :($x where 42)
18:01 iblechbot joined
pugssvn r29808 | azawawi++ | [Cursor] Fixed an uninitialized error that was caused by my earlier commit. azawawi-- 18:08
PerlJam Do any of you guys have any good techniques for focusing? I find I'm continually distracted and I always feel like I have to find "other time" to do things. 18:09
m-locks meditation is fun and enhances brain abilities 18:10
masak TimToady: what do you think of this S12 patch? gist.github.com/312499
vorner PerlJam: I use headphones and music to filter out external interference, but I'd use something for the internal one too
18:10 Trashlord left
masak PerlJam: what m-locks said. also, off-line time. 18:11
m-locks music works too, but sometimes just all quiet and still is needed
masak actually, non-comuter time. it's amazing what a focus drain the computer is for me. :/
azawawi masak: hi there :)
m-locks :/
masak azawawi: oh hai! how's life? 18:12
PerlJam going off-line is the easy thing, but still I find myself distracted.
vorner some people say a girlfriend helps too, but I have varying results there
PerlJam It's like I have ADD or something. (I don't know that that's clinically the case though)
vorner: I have a wife and 4 kids. I don't think I need a girlfriend too ;)
azawawi masak: in Padre-land... :)
18:13 ShaneC joined
vorner PerlJam: ok, cross that one out :-)) 18:13
18:13 bluescreen left
m-locks well wife and 4 kids are the definition of distraction 18:14
masak by the way, does anyone remember why the commas-between-adverbs-optional rule was abolished?
or was it a commas-before-adverbs-optional rule?
PerlJam masak: I'm sure TimToady knows :) 18:15
masak PerlJam: yes, but I was hoping someone who's here would remember :)
jnthn PerlJam: I found that accepting that there just was going to be times when I couldn't focus, and actually stepping away from things to relax/mess around rather than trying to do stuff and failing, kinda helped when me focus more when I did come to do stuff.
PerlJam I think it's another one of those two-terms kind of things, but I don't remember at all
masak PerlJam: yes, it certainly is. nowadays.
PerlJam: but it wasn't always. 18:16
Tene PerlJam: what really helps for me is formalizing my "working time" and "non-working time". For me personally, it was "Put on headphones, play specific music, sit in a specific place, now I'm working", and I would only do those things when I was trying to focus, and stop as soon as I was doing something else.
PerlJam jnthn: That's the one thing I've come to realize and sometime do (though there's a tendency for me to keep trying despite any discernable progress)
Tene: you know what? I think that may be part of my problem. I've trained myself to be distracted in certain places rather than the other way around. 18:17
Tene PerlJam: one other thing that helps for me is to start with a list of one item, "Enumerate tasks in detail", then break it out into "enumerate the steps in Foo, in Bar, etc." and work my way down from there. 18:18
Yes, I've definitely done that before. :)
PerlJam specifically, at work, I tend to get interrupted alot, so rather than "fix" that, I've got it in my head that "at worked you will always be distracted" 18:19
s/worked/work/
Tene PerlJam: I have a wife and 0 kids, so I have plenty of excess distraction to spend on my girlfriend. ;) 18:21
PerlJam Tene: I hope your wife doesn't get on the net much or she's okay with that arrangement :) 18:22
Tene For a while, tea was my "Okay now I'm really focusing" signal. 18:24
I know I've used a few others...
Yes, the latter.
masak re-reads perlbuzz.com/2007/12/where-is-perl-...t-die.html with interest 18:25
PerlJam anyway, thanks for the input all; I'm going to spend the rest of the day trying to focus and Get Stuff Done :-) 18:26
I may even get in the zone for an hour or so :)
PerlJam &
Tene PerlJam: The trigger that worked the BEST EVER for me was sending $20 to [email@hidden.address] via paypal.
fwiw
you might try it.
18:27 cdarroch joined, cdarroch left, cdarroch joined, masak left
Su-Shee I like tiny tasks. then I can mentally trick myself into "oh, well this tiny thing I'll get done in 10 minutes, I'll just do it now.." 18:28
18:28 bluescreen joined
Su-Shee (4 hours later...:) 18:28
Tene lesswrong.com/lw/1sm/akrasia_tactics_review/ has an excellent overview, with plenty of links, on a variety of anti-akrasia tactics 18:29
Su-Shee "akrasia"? the greek version of "procrastination"? 18:30
Tene "Failure of will"
procrastination is the most common case, yes.
Su-Shee ah. 18:31
in a recent paper about procrastination I've read the author derives it from a lack of impulse control. 18:32
18:32 ggoebel joined
Tene There are other posts on lesswrong that talk about studies of willpower, how various activities deplete willpower and lead to poorer impulse control, etc. 18:33
pugssvn r29809 | lwall++ | [S02] make Instant completely opaque as suggested by ruoso++.
r29809 | Instants are now considered a more type basic than epochs, which
r29809 | are just particular named instants. All culturally aware time
r29809 | can be based on calculations involving instants and durations.
r29809 | (Durations are still considered interconvertible with Reals.)
jnthn s/more type basic/more basic type/ ? 18:34
Su-Shee Tene: I worked as a projectmanager - I had to manage people procrastinating.. :/
18:38 synth left, azawawi left
TimToady std: anon class is Mammal {...} 18:39
p6eval std 29808: OUTPUT«===SORRY!===␤Unable to parse class definition at /tmp/UELjqzDu2y line 1:␤------> anon class is ⏏Mammal {...}␤ expecting trait␤FAILED 00:01 107m␤»
jnthn std: anon class :: is Mammal {...} # no, you gotta do this 18:40
p6eval std 29808: OUTPUT«ok 00:01 105m␤»
jnthn Otherwise is swallows "is" as the class name.
*it
TimToady phenny: tell masak re anon class is, see irclog.perlgeek.de/perl6/2010-02-23#i_2024709 18:41
phenny TimToady: I'll pass that on when masak is around.
TimToady STD is your friend, except when it isn't...
18:42 quietfanatic left
diakopter std: ->(){}<> <-{} 18:43
p6eval std 29808: OUTPUT«ok 00:01 108m␤»
18:48 eternaleye left
pmichaud back 18:53
18:57 hercynium left
TimToady jnthn: &foo:($,$) with a colon so that it's parsed as a sig 18:58
or &foo:(Any,Any) 18:59
arnsholt diakopter: Nice. Any idea what it should mean? =) 19:00
TimToady it means diakopter is insane, obviously :)
pmichaud ...and that Perl 6 thinks insanity is okay. 19:01
TimToady there is that...
[particle] see! perl 6 is still perl 5! line noise! 19:05
19:08 zard1989 joined
Su-Shee well it'll be like in vim - every time you make a typo, you'll discover a new feature/possibility. 19:09
TimToady it will not, of course, actually run
a closure doesn't respond to .<>
jnthn TimToady: Thanks. :-) 19:12
19:12 ggoebel left 19:13 lichtkind joined 19:26 k23z__ left 19:28 riffraff joined 19:29 payload1 left, pjcj_ joined, payload left, payload1 joined, jferrero left 19:30 payload joined 19:31 xabbu42 left 19:32 rv2733 left 19:34 hercynium joined 19:36 ruoso left 19:38 payload left 19:39 payload1 left 19:45 riffraff left 19:46 zard1989 left 19:47 riffraff joined 19:50 ruoso joined
ruoso TimToady, just another note... Durations can also be culturally aware... "1 day" is a valid gregorian duration... 19:54
and is not well defined in terms of seconds 19:55
it is certainly more reasonable for them to be able to be coerced to the number of seconds
but when you add "1 day" it behaves differently then "86400 seconds" 19:56
19:59 alester joined
TimToady that is not a Duration 19:59
and 86400 POSIX seconds is not a Duration either :) 20:01
and you cannot coerce 1 day to a number of seconds without knowing which day you're talking about 20:04
we will only use the basic Duration type to discuss atomic seconds in Perl 6 20:05
it is more accurate to say that cultural units can have varying durations, but durations are not aware of culture 20:06
(unless you have a relativistic culture... :) 20:07
ruoso so what "1 day" is?
in perl5 DateTime familiy... it's a DateTime::Duration 20:09
20:09 athenot left 20:10 frettled sets mode: +o ruoso
frettled «1 day» is a problem, yes. 20:11
DST implies that it can be one hour shorter or longer than the norm, leap seconds can make it one or more seconds longer. 20:12
20:12 eternaleye joined
frettled Then we have those fscked-up decisions when governments change time zones. 20:13
When I proposed looking at TAI for the time specification, I believe we were perusing a document about this nice soup. Hmm. 20:14
Of course, that was probably on IRC…
I don't even recall if I specifically suggested TAI or something else, heh. 20:15
TimToady "1 day" is a cultural/spacetime-dependendent algorithm for deriving one instant that is approximately 24 hours from another instant, that's all
without knowing the culture, the space, or the time, you don't know the duration 20:16
20:17 rv2733 joined
frettled This was one of the links: hpiers.obspm.fr/eop-pc/earthor/utc/...econd.html 20:17
ruoso TimToady, I meant: what type is "1 day", since it's not an Instant and not a Duration? 20:18
TimToady maybe we should call them Durances, as in "durance vile"
ruoso heheh... 20:19
TimToady and since durances are usually measured in cultural units
ruoso but if the Duration is always TAI, why isn't it just a Number?
TimToady N days/years in jail
frettled ruoso: I think perhaps considering durations merely as "atomic seconds" would be better.
that is, when you want that kind of precision 20:20
TimToady because numbers are unitless
and we want to keep our dimensional analysis straight
ruoso so Duration is just a tagged Number
TimToady yes, like Kilograms 20:21
20:21 riffraff left
ruoso er... Mass would be the correct analogy 20:21
or else we need to call it AtomicSeconds instead of Duration 20:22
TimToady nodnod
ruoso and that probably makes sense
(calling it AtomicSeconds instead of Duration)
TimToady ugly
ruoso but when will you need to specify that type anyway?
TimToady it's fine to treat one particular unit as privileged
especially when it's the gcd like seconds are 20:23
ruoso gcd? 20:24
TimToady well, no units above that are necessarily evenly divisible
ruoso (gcd as in Greatest common divisor?) 20:25
TimToady yes
ruoso but what is the use case for declaring the type? 20:26
TimToady mental abstraction
ruoso but when would you need it?
TimToady when you're thinking :) 20:27
ruoso heh...
alright... when would you use it in a code example? 20:28
TimToady when the error message tells you "Illogical attempt to multiple two Durations"
*multiply
ruoso ahm... ok 20:29
Tene TimToady: "... available precision subset precision ..."?
TimToady and so that you can know that when you add one to an Instant, you get another Instant
but when you add it to a Duration, you just get more Duration
Tene TimToady: "... created either via any of <only one branch of either>." 20:31
TimToady and more generally, to prevent you from confusing foot-pounds with newtons, if you're NASA
frettled :)
ruoso TimToady, okay... point taken... so Duration and Durance?
frettled And even more generally, if you're doing time keeping. ;)
TimToady Tene: feel free to clean up sloppy prose where it's obvious
and we all know that Durances are kinda vile :)
frettled hehe
TimToady I think it's nice that they're related words though 20:32
Tene TimToady: "... perhaps the year year 2000 epoch ..."
frettled I can see use cases for these thingies. And I _know_ there are use cases for stuff like Unix time_t according to SUS (sometimes referred to as POSIX)
Tene TimToady: I'm about to be AFK for a while. Leaving notes here to maybe someone will get it if Idon't when I get back.
frettled TimToady: VileTime!
20:33 alester left
TimToady VT:: 20:33
and everyone will think it's VirtualTime :)
ruoso TimToady, it could be in the namespace of the given calendar/cultural-mindset 20:37
Gregorian::Durance
20:37 alester joined
ruoso Gregorian::DateTime 20:37
TimToady that's insufficient
you need the triple, culture, space, time to know what's when 20:38
that triple is what I was proposing we call TZ for old time's sake
by which I don't mean POSIX TZ
ruoso TimToady, but that are attributes.... 20:39
TimToady maybe just Z for Zone, and Zulu
ruoso the calendar set modifies even which attributes are that
TimToady calendar set is part of culture
ruoso right right...
TimToady I make no claims that the triple is a completely orthogonal vector space :) 20:40
but they seem to be the minimal useful vector
ruoso but I insist... the pivot is the culture
space and time are just attributes around that pivot
hence... GregorianCalendar 20:41
TimToady yes, that's what I was calling that TZ
where it's really for Twilight Zone
I really hate long names for common types... 20:42
ruoso but they won't be commonly used as the long names
frettled I like just "Time", because it's sort of all-encompassing, but "Tm" is a nice abbreviation.
ruoso we're going to need a quote qualifier for each of those anyway
TimToady sez who 20:43
if there are coercions, quote qualifiers will merely be duplicate
ruoso as in... 20:44
TimToady needless complexity
frettled Would "Tm" clash significantly with something, or have some hoplessly inhibiting ambiguity?
ruoso Gregorian::DateTime $x = "2009-01-03T20:00:00+0000"; # ?
20:45 uniejo left
TimToady .= new is one way 20:45
20:45 uniejo joined
ruoso TimToady, but there is another axis 20:46
which is the date format
representation format, I mean
TimToady you mean preferred output format, since presumably we recognize more than one input format for some systems 20:47
seems like a subset of culture, mostly
ruoso I meant input format, which is considerably harder than output
frettled Culture::Duration
TimToady Durance
frettled Or perhaps just Cult::Durance
ruoso q:ISO8601{2009-01-03T20:00:00+0000}
TimToady let's just not reinvent locales, since that is by definition "badly" 20:48
allbery_b that'd be my answer
TimToady locales make a lousy type system 20:49
allbery_b ...but localization makes lousy types, so that's fair
TimToady Cal:: maybe 20:51
20:52 ggoebel joined 20:53 uniejo left, ruoso left
TimToady T:: as the top of a large hierarchy of cultural/space/time/customization concepts 20:54
frettled locales still need to be taken care of, though :(
TimToady as in T-30 seconds :)
locales are cultural info
and customization
and in general, Perl 6 ignores locales unless specifically instructed to pay attention 20:55
and said instruction can be either lexical or dynamic in Perl 6, I suspect
frettled yup
though some sort of localization interface against (and I mean against) Unix and Wintendo would be nice. :) 20:56
TimToady "The question is, who's to be master, that's all." --Humpty Dumpty
20:57 hercynium left
frettled I think our most significant problem with this time-and-culture discussion is that it's very hard to be clear, concise and precise about something that's so woolen, voluminous and blurry. 20:58
TimToady yes, when the top-level is the cultural dimension, word choice becomes problematic 20:59
20:59 mssm left
TimToady but it does seem rather healthy to divorce all of this conceptual mumbo-jumbo from the underlying linear types 21:00
which will presumably be the medium of cross-cultural communication
21:00 riffraff joined
TimToady and highlight places where the conceptual mumbo-jumbo is ill-defined 21:01
see "1 day"
TimToady wanders off for a durance or so & 21:03
frettled :D
pugssvn r29810 | tene++ | [S02] Small prose fixes. No functional changes. 21:06
21:06 athenot joined
frettled \o/ 21:07
allbery_b "1 second" is iffy if the "second" is a "daka" 21:08
("dakah")
sorry, that'd be minute, I don't recall the proper name for a second in that system
if any
21:14 bluescreen left 21:18 lmc_ joined, lmc_ left 21:19 pyrimidine left
sjohnson hi guys 21:24
Tene hi
frettled hiho 21:25
21:27 jferrero joined 21:28 hercynium joined 21:31 hercynium left, jferrero left, athenot left, cdarroch left, rgrau left, charsbar left, rgrau joined 21:32 charsbar joined, cdarroch joined, cdarroch left, cdarroch joined 21:36 daemon joined, awilcox joined 21:39 ggoebel left 21:41 awilcox left 21:48 charsbar left, PacoLinux left, nadim left, gabiruh joined, hatseflats joined, nadim joined
mathw Evening 21:54
jnthn oh hai, mathw 21:56
mathw how're things?
21:58 PacoLinux joined
jnthn About as normal as I could hope for, given I move country next week. :-) 21:58
frettled jnthn: ooh, congrats!
AIU, I haven't been paying attention. Which country?
jnthn frettled: I'm moving to Sweden
frettled w00t 21:59
jnthn frettled: Aren't you somewhere in Scandinavia?
frettled jnthn: surströmming all year for you!
jnthn: Yes, Oslo, Norway.
Are you moving to somewhere near masak++?
jnthn When I make my millions, I'll come and visit Oslo for a beer.
;-)
frettled he
and that's all you'll be able to afford! ;) 22:00
jnthn Not too near to masak++...but not too far away either. :-)
frettled 100 SEK: 81.670, 82.470 NOK
Darnit, still stable.
mathw oooh I liked Sweden
where in Sweden are you going, jnthn?
jnthn mathw: Lund
frettled nice 22:01
mathw oh right near Denmark
jnthn Aye
mathw not very near where I went
jnthn Well, not as near as, say, Malmo, or Helsingborg. But still very close. 22:02
mathw a lot nearer than Stockholm or Uppsala
I forget that Sweden extends that far south
and gets so close to Denmark
so that there are bridges
jnthn Yeah
I can probably get to Copenhagen in < 1 hour by train.
mathw which is awesome 22:03
jnthn Yeah
mathw where can I get to in < 1 hour by train
...
jnthn Well, I've had Vienna about an hour away for the last couple of years too. :-)
mathw Leicester
Derby
chesterfield
how exciting
jnthn THRILLING!
No.t
*Not
mathw ooh ooh ooh
Mansfield!
oh wait no Mansfield's a shithole
jnthn omg how do you resist going there like, all the time?
;-) 22:04
mathw because I like to keep my aikido within the dojo...
jnthn lol
mathw there was a self-defence special class today
outside in the rain
in a car park full of puddles
and, apparently, cat poo
all in the name of realism
jnthn Serves you right for wishing wetness on me the other day. :-P
mathw I went to the normal class and did my testing instead
jnthn Aww. 22:05
:-)
mathw I'm not quite mad enough to do the outdoor classes yet
I'm sure it'll come
jnthn Work on Rakudo.
It'll come.
:-)
mathw lol
you're probably right 22:06
I have a sudden urge to call you sensei
lichtkind so wiki has 6 new articles and over 27 edit made today :) 22:08
mathw \o/
It's great to see the progress on the wiki updating
lichtkind :)
mathw lichtkind++
jnthn hopes none of them were spam
22:08 Chillance joined
jnthn lichtkind++ :-) 22:08
lichtkind thanks 22:09
no the new stuff is about audrey, allison , chromatic, damian, jonathan leto
and historic implementations
so most people have now an article
mathw: do want your own? :) 22:10
mathw it would be ridiculously short
lichtkind or just lurker? :)
mathw I don't get much done
lichtkind most are already :)
doesnt matter its community im not your boss
mathw: so what you were involved? 22:11
mathw mine would say something like 'Wants to do things, but spends too much time playing music and going to aikido classes'
22:11 kensanata left
lichtkind aikido is important 22:11
mathw my senseis would be pleased to hear that :) 22:12
lichtkind even if some say the only one ever mastered aikido was ujeshiba
mathw it depends where you draw the line
is there even anywher eyou can draw it?
nobody stops learning
lichtkind mathw: ujeshiba defeated samurai with swords with its empty hands :) 22:14
mathw I'm not denying he was good
lichtkind he was aura sensitive 22:15
mathw: so you hack on rakudo or parrot?
mathw I think my name's on one commit somewhere in rakudo :)
22:16 athenot joined
sjohnson is the anticipated rakudo release still coming around the mountain in a month or two? 22:17
jnthn
.oO( she'll be coming round the mountain when she comes... )
sjohnson: April is still the target. :-)
mathw the bit that always got me was the bit about riding three white horses
or was it six? 22:18
which is just even more ridiculous
must be mini horses
anyway, time for me to have a bath before all my muscles seize up
jnthn ride(all(@horses))
mathw goodnight
jnthn night, mathw
sjohnson sweetness 22:20
pmurias lichtkind: aura sensitive = you mean he had supernatural powers? 22:34
lichtkind pmurias: from standpoint of an cow rational thinking is supernatural :) 22:36
22:37 wknight8111 joined 22:41 Su-Shee left
pmurias lichtkind: i assume that means yes 22:42
22:43 rv2733 left
lichtkind pmurias: that ment i dont like it call supernatural :) 22:48
pmurias lichtkind: you train aikido? 22:51
lichtkind pmurias: no i only read about ujeshiba, but what i heard sounds familiar
22:54 iblechbot left, ggoebel joined
pmurias lichtkind: do you train a martial art/combat sport? 22:55
lichtkind pmurias: i did
22:56 ggoebel left
pmurias lichtkind: what? 22:57
lichtkind pmurias: karate/ taiji/ wushu
but only shotokan
22:58 ggoebel joined, colomon joined 23:12 IllvilJa joined
pugssvn r29811 | masak++ | [S12] found/fixed a case of a missing comma between adverbs 23:15
23:17 ggoebel left 23:22 ihrd joined 23:23 ihrd left 23:24 Exodist left 23:27 quester joined, pmurias left 23:37 charsbar joined 23:48 riffraff left 23:51 colomon left

Powered by Cro and the Raku® Programming Language.

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

Thank you!