»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 moderator on 10 October 2009.
00:20 crythias joined 00:29 shinobi-cl joined 00:42 KyleHa joined 00:56 rhr joined
shinobi-cl rakudo: sub AAA { return 1, 2; }; my ($val1, $val2) = AAA; say "$val1 - $val2"; 01:02
p6eval rakudo d749d9: OUTPUT«too many arguments passed (2) - at most 1 param expected␤in Main (file src/gen_setting.pm, line 206)␤»
shinobi-cl rakudo: sub AAA { return (1, 2); }; my ($val1, $val2) = AAA; say "$val1 - $val2";
p6eval rakudo d749d9: OUTPUT«1 - 2␤»
shinobi-cl rakudo: sub AAA { return ("one", 2); }; my ($val1, $val2) = AAA; say "$val1 - $val2"; 01:03
p6eval rakudo d749d9: OUTPUT«one - 2␤»
shinobi-cl rakudo: sub AAA returns Str, >Int{ return ("one", 2); }; my ($val1, $val2) = AAA; say "$val1 - $val2";
p6eval rakudo d749d9: OUTPUT«Malformed routine definition at line 2, near "AAA return"␤in Main (file <unknown>, line <unknown>)␤»
shinobi-cl rakudo: sub AAA returns Str, Int { return ("one", 2); }; my ($val1, $val2) = AAA; say "$val1 - $val2";
p6eval rakudo d749d9: OUTPUT«Malformed routine definition at line 2, near "AAA return"␤in Main (file <unknown>, line <unknown>)␤»
shinobi-cl rakudo: sub AAA returns (Str, Int) { return ("one", 2); }; my ($val1, $val2) = AAA; say "$val1 - $val2"; 01:04
p6eval rakudo d749d9: OUTPUT«Malformed routine definition at line 2, near "AAA return"␤in Main (file <unknown>, line <unknown>)␤»
shinobi-cl rakudo: sub AAA returns Str { return "one"; }; my $val1 = AAA; say "- $val1 - "; 01:08
p6eval rakudo d749d9: OUTPUT«- one - ␤»
shinobi-cl rakudo: sub AAA returns Int { return "one"; }; my $val1 = AAA; say "- $val1 - ";
p6eval rakudo d749d9: OUTPUT«Return value type check failed; expected Int, but got Str␤in sub AAA (file /tmp/5vkpeAhxqR, line 1)␤called from Main (file /tmp/5vkpeAhxqR, line 0)␤»
01:12 rhr_ joined 01:18 lumi joined 01:27 slavik joined 01:30 synth joined 01:31 agentzh joined 01:33 rhr joined
dalek p-rx: 5045605 | pmichaud++ | (6 files):
Handle positional (numbered) captures.
 01:34
p-rx: 032c58a | pmichaud++ | t/p6regex/rx_ (5 files):
Add remaining test files from PGE (not active/configured yet).
 01:40
01:50 shinobi-cl left 01:57 linguini left 02:01 msmatsko joined 02:02 rhr_ joined 02:12 tylerni7 joined 02:19 jjore joined
dalek p-rx: 76c694d | pmichaud++ | src/Regex/P6Regex/Actions.pm:
Add aliased assertion captures.
 02:26
p-rx: 4481429 | pmichaud++ | src/Regex/P6Regex/ (2 files):
Handle simple binding aliases.
slavik pmichaud: I love you!!! 02:30
02:35 rhr joined 02:36 envi^home joined 02:39 mikehh joined 02:45 stephenlb joined 02:52 rhr joined 03:28 JimmyZ joined 03:29 rhr_ joined 03:42 envi^office joined 03:44 absurddoctor joined
dalek p-rx: 568cc9b | pmichaud++ | src/ (3 files):
Allow PAST::Regex subrule nodes to have arguments to pass to the subrule.
 04:09
p-rx: 3de5a4f | pmichaud++ | src/ (2 files):
Add alias capturing of non-subrules.
 05:06
p-rx: f00fd0d | pmichaud++ | src/Regex/P6Regex/Grammar.pm:
Allow spaces in character class compositions.
p-rx: 8743056 | pmichaud++ | src/Regex/P6Regex/Actions.pm:
Add character class composition.
05:07 eternaleye joined
pmichaud I think all I need for self-hosting is :ratchet and :sigspace 05:07
(self hosting the regex parser, that is)
05:21 japhb joined 05:28 envi^home joined 05:29 FCO joined 05:34 KatrinaTheLamia joined
pugs_svn r28776 | lwall++ | [workout.t] correctly quote arrayish notation 05:44
05:45 parduncia joined 05:49 rfordinal joined 05:52 rfordinal left 06:15 KatrinaTheLamia joined 06:19 Lorn joined 06:28 parduncia_ joined 06:29 parduncia__ joined 06:34 rfordinal joined 06:49 Gothmog_ joined 06:54 NorwayGeek joined 07:11 rfordinal joined 07:12 tann joined
tann perl6eval: caller; 07:13
rakudo: caller;
p6eval rakudo d749d9: OUTPUT«caller() is not yet implemented in Rakudo, sorry␤in sub caller (file src/gen_setting.pm, line 2360)␤called from Main (file /tmp/3j0i0F4FI3, line 0)␤»
07:15 rfordinal left 07:21 NorwayGeek joined 07:22 NorwayGeek_ joined 07:25 yahooooo joined 07:31 JimmyZ joined 07:32 duckz joined 07:34 eMaX joined 07:39 ihrd joined 07:42 ihrd left 07:43 mariuz joined 07:46 mariuz joined 07:57 baest joined 08:05 drbean_ joined 08:57 schemer999iphone joined 09:00 jaldhar joined 09:03 schemer999iphone joined 09:15 payload joined, Trinity94 joined
Trinity94 err 09:17
09:17 PZt joined
sjohnson hi 09:26
pugs_svn r28777 | jimmy++ | [zh-cn/syn/S02-bits.pod] added some translations. It needs persistence. 09:36
jnthn morning 09:52
Matt-W morning 09:55
jnthn Quiet here today!
Matt-W Mmm
I'm trying to come up with a plan for my time which includes all the things I like to do, including Perl 6 hacking 09:56
It's not going well :(
jnthn :-( 09:57
Yeah, those things are hard to make, but even harder to actually follow.
09:57 donaldh joined, donaldh left
Matt-W Mmm 09:58
At the moment I'm just trying to fit in the two aikido sessions a week the chief instructor's urging me to do
Which I really want to be able to do
Thursday will be the experiment time, if I can get home from work, get in the car and get to viols on time,a nd then get from viols to aikido in time for the class, then that'll be fantastic. 09:59
jnthn Tiring evening though!
Matt-W Different kind of tiring though
What it'll be is hungry, because I won't be able to eat until after aikido 10:00
I think I need to make a space on Sundays for Perl 6
10:03 zamolxes joined
dalek p-rx: 348b39f | pmichaud++ | src/Regex/constants.pir:
Turn some magic constants into symbols.
 10:03
p-rx: 45fcb57 | pmichaud++ | (2 files):
Convert to use symbols instead of magic numbers.
p-rx: bdc43da | pmichaud++ | src/Regex/Cursor.pir:
Revise !mark_commit to preserve capture stack.
Matt-W Then I'll hav erecovered from any programming burnout from the week at work, and Sundays are largely unencumbered
p-rx: 1987a95 | pmichaud++ | (4 files):
Add cut-rule token.
Matt-W blinks
nice 10:04
pmichaud++
jnthn Wow, when does he sleep? :-) 10:08
dalek p-rx: 105b3e0 | pmichaud++ | src/PAST/Compiler-Regex.pir:
More replacing magic numbers with symbolic constants.
 10:09
Matt-W I don't know
Maybe he's on a caffeine IV
jnthn pours in more strongly caffeinated fluid to keep pm going 10:10
Matt-W Also, he does seem to package his changes in nice logical small chunks, so you get a high count of commits
so maybe he's not been up quite as long as the commits suggest...
jnthn I approve of that approach.
I'm doing the same, apart from my commits don't show up here at the moment.
(because I'm working in a branch)
Matt-W Yes I think it's excellent, and it's something git makes particularly practical, because a git commit is a useful entity in its own right 10:12
you can easily grab one and pull it into your repo
I like that very much, we'd benefit from that at work but we're stuck with subversion
carlin rakudo: module Foo { sub bar is export { say 'OH HAI'; } }; Foo.new.bar; 10:22
p6eval rakudo d749d9: OUTPUT«Null PMC access in getprop()␤in Main (file src/gen_setting.pm, line 206)␤»
Matt-W erk 10:29
that's not good
carlin I thought 'is export' was the problem, but it seems not 10:30
How are you supposed to call subs from modules? 10:31
rakudo: module Foo { sub bar { say 'OH HAI'; } }; Foo.bar;
p6eval rakudo d749d9: OUTPUT«Method 'bar' not found for non-object␤in Main (file src/gen_setting.pm, line 206)␤»
Matt-W rakudo: module Foo { sub bar { say 'OH HAI'; } }; Foo::bar; 10:33
p6eval rakudo d749d9: OUTPUT«OH HAI␤»
carlin Oh, of course
Matt-W But still, a null PMC access is bad
rakudo: module Foo { }; Foo.new;
p6eval rakudo d749d9: OUTPUT«Null PMC access in getprop()␤in Main (file src/gen_setting.pm, line 206)␤»
Matt-W hmm and there it is 10:34
jnthn: is this known?
10:35 ihrd joined
moritz_ I think it is 10:35
jnthn Matt-W: Not sure. It's ugly, for sure... 10:36
I've no idea whether you should be able to instantiate a module...or what it would mean.
10:36 meppl joined
jnthn Well, I guess actually it'd call Module.new... 10:37
carlin rakudo: say Module.new 10:38
p6eval rakudo d749d9: OUTPUT«maximum recursion depth exceeded␤in Main (file src/gen_setting.pm, line 206)␤»
Matt-W Whatever it is supposed to do, it shouldn't NPMC 10:39
moritz_ and I think the "maximum recursion depth exceeded" is new 10:40
carlin: care to submit?
Matt-W Maybe .new on a module should just say hey, I can't call a method on that 10:41
carlin Ticketed the recursion depth one 10:47
What about the Null PMC?
moritz_ I remember it well enough to think it's known. 10:48
10:49 payload joined
carlin Oh okay 10:49
10:51 payload1 joined 10:53 payload joined 11:15 Bzek joined 11:24 NorwayGeek joined, kst joined
moritz_ rakudo: say (7/5).Int 11:35
p6eval rakudo d749d9: OUTPUT«1␤»
moritz_ rakudo: say <a b c>.[7/5]
p6eval rakudo d749d9: OUTPUT«get_integer() not implemented in class 'Rat'␤in Main (file src/gen_setting.pm, line 206)␤» 11:36
11:36 masak joined 11:42 eMaX joined
Woodi rakudo: @a = 1 .. Inf; say @a; 11:45
p6eval rakudo d749d9: OUTPUT«Symbol '@a' not predeclared in <anonymous> (/tmp/UXaM3kEvYc:2)␤in Main (file <unknown>, line <unknown>)␤»
Woodi hmm
masak strict mode,
use 'my'
Woodi rakudo: my @a = 1 .. Inf; say @a; 11:46
masak \\o/
p6eval rakudo d749d9: TIMED_OUT
masak Woodi: NYI
Woodi just example from Apocalypse :)
masak (lazy lists, that is)
Woodi work :)
jnthn We're too lazy to implement that.
11:46 KyleHa joined
masak I question whether 'say @a' would actually output anything even after we get lazy lists, though. 11:47
'say' tends to want the entire argument before printing it. :)
dalek kudo: 609cc25 | moritz++ | src/classes/Object.pir:
make the get_integer vtable class .Int; fixes RT #69738
Woodi so say looks eager for me...
masak now, if you did '.say for @a', that's be a different thing... 11:48
Woodi right, i just ask ref @a
why .say* ?
moritz_ masak: no, argument lists are lazy too
.say is short for $_.say
jnthn moritz++ :D 11:49
Woodi and ?
jnthn masak: I think it'd just put the lazy list into the slurpy, which would also be lazy.
masak: And say would iterate over that.
masak: So it'd probably work, once we implement those...fun...bits. :-) 11:50
Woodi why $_.say for @list ? what $_do there ?
masak Woodi: $_ is each element of the iteration.
Woodi: it's as if it said `for @a -> $_ { $_.say }`
Woodi masak: for @a { say } not work in p6 ? 11:55
masak rakudo: my @a; for @a { say } 11:56
p6eval rakudo d749d9: OUTPUT«say requires an argument at line 2, near "}"␤in Main (file src/gen_setting.pm, line 2562)␤»
takadonet morning all
masak Woodi: no, because the magic that causes print/say to default to $_ is gone.
takadonet: o/
Woodi why ? it was basic perl functionality... 11:57
moritz_ because people had to memoize arbitrary lists of functions which default to $_
no you just can say .any_method, and it defaults to $_ 11:58
so it's much more universal than before
Woodi but @a -> $_ is ugly after p5...
masak Woodi: the -> $_ is implicit.
jnthn Woodi: for @a { ... } is the same as for @a -> $_ { ... }
masak Woodi: it seems you have a fair bit of Perl 5 baggage to unlearn. :) 11:59
don't worry, that's to be expected.
Woodi have problems with unlearning...
moritz_ btw the equivalence is not complete
-> $_ is rw { ... } or so, no? 12:00
masak <-> $_, yes
12:00 eMaX joined, envi^home joined
masak I didn't want to mention that, because we didn't do any writing to $_ in the loop block anyway. 12:00
pugs_svn r28778 | moritz++ | [t/spec] test for RT #69738: array indexing with Rats 12:08
moritz_ rakudo: say <a b c>[5/4] 12:09
p6eval rakudo 609cc2: OUTPUT«b␤»
moritz_ another masak++bug closed
masak moritz_++ 12:10
jnthn \\o/
moritz_ this one was quite easy, akshually
masak I found that bug in the app cheese. 12:11
KyleHa++ # adding a test for it 12:12
jnthn Mmm...cheese. 12:13
12:14 shinobi-cl joined
KyleHa Looked like moritz++ wrote that test. Or am I not following closely enough? 12:14
masak KyleHa: ah, I might have been confused because the automatic email had your name on it. 12:15
moritz_++ 12:16
KyleHa Oh yeah. I send the emails, but the committer's name is inside.
jnthn You deserve the karma for the automated emails anyway :-) 12:17
masak aye. 12:18
they rock.
KyleHa Moritz wrote the script that sends the emails too; I just run the cron job. 8-)
I can take credit for the idea, though. 12:19
moritz_ and for running the cron job.
KyleHa *nod* 12:20
pugs_svn r28779 | moritz++ | [t/spec] Test for RT #69740
12:28 eMaX joined
dalek kudo: 3d1afed | moritz++ | src/setting/Operators.pm:
give more awesome error message for infinite ranges
 12:33
jnthn rakudo: 1..Inf
p6eval rakudo 609cc2: ( no output )
moritz_ uhm, I was imprecise in my commit message - infinite series
not ranges
jnthn ah 12:34
"This is infinite and we're not lazy yet." :-)
masak now that that one's fixed, maybe I should update the blog post from two days ago. 12:35
moritz_ "we're too lazy to be lazy yet"
12:35 eMaX joined
carlin rakudo: 1...Inf 12:39
p6eval rakudo 609cc2: OUTPUT«src/gc/api.c:248: failed assertion 'PObj_is_PMC_TEST(obj)'␤Backtrace - Obtained 32 stack frames (max trace depth is 32).␤/home/p6eval//p2/lib/libparrot.so.1.6.0 [0x2b5f541ff4d3]␤/home/p6eval//p2/lib/libparrot.so.1.6.0(Parrot_confess+0x87)
..[0x2b5f541ff607]␤/home/p6eval//p2/lib/libpa…
jnthn holy poo
12:39 synth joined
carlin Oh wow 12:39
All I get is a segfault
moritz_ the server is currently rebuilding rakudo - don't know if that's related
rakudo: say 1...*
p6eval rakudo 609cc2: OUTPUT«Method 'params' not found for invocant of class 'Failure'␤in Main (file src/gen_setting.pm, line 206)␤»
moritz_ that's the one I fixed in the previous commit 12:40
rakudo: say 1...* 12:41
p6eval rakudo 609cc2: OUTPUT«Method 'params' not found for invocant of class 'Failure'␤in Main (file src/gen_setting.pm, line 206)␤»
moritz_ not yet udpated
rakudo: say 1...* 12:42
p6eval rakudo 3d1afe: OUTPUT«Sorry, lazy lists and infinite ranges are not yet implemented␤in sub infix:... (file src/gen_setting.pm, line 2494)␤called from Main (file /tmp/eMejuSe26x, line 0)␤»
masak moritz_++ 12:43
I also updated the journal entry. use.perl.org/~masak/journal/39740
for those who want to feel like the critical bugs, I've made a convenient list here: use.perl.org/~masak/journal/39597 12:44
rakudo: { &^c("$^a, $^b") }.("g'day", "perl6ers", &say) 12:46
p6eval rakudo 3d1afe: OUTPUT«Parameter type check failed; expected Callable, but got Multi for c in call to _block52␤in Main (file <unknown>, line <unknown>)␤»
masak shucks, I thought we'd fixed that one.
jnthn I didn't remember fixing that one. 12:48
12:48 JimmyZ joined 12:49 arthur-_ joined 12:54 gbacon joined, alanhaggai joined 12:56 SmokeMachine joined
masak 'We practice the 'whirlpool' development model, so this design document could at best be described as "getting us close enough to the sucking vortex to begin our descent".' Leto++ gitorious.org/parrot-plumage/parrot...ter/README 12:58
13:00 am0c joined 13:01 eMaX joined 13:31 barney joined
shinobi-cl sub AAA returns Str { return "'sup" }; say AAA; 13:56
rakudo: sub AAA returns Str { return "'sup" }; say AAA; 13:57
p6eval rakudo 3d1afe: OUTPUT«'sup␤»
shinobi-cl rakudo: sub AAA returns Str { return "'sup", 1 }; say AAA;
p6eval rakudo 3d1afe: OUTPUT«too many arguments passed (2) - at most 1 param expected␤in Main (file src/gen_setting.pm, line 206)␤»
shinobi-cl rakudo: sub AAA returns Str { return ("'sup", 1) }; say AAA;
p6eval rakudo 3d1afe: OUTPUT«Return value type check failed; expected Str, but got List␤in sub AAA (file /tmp/lrHr86c6v3, line 1)␤called from Main (file /tmp/lrHr86c6v3, line 0)␤»
shinobi-cl rakudo: sub AAA returns (Str, Int) { return ("'sup", 1) }; say AAA; 13:58
p6eval rakudo 3d1afe: OUTPUT«Malformed routine definition at line 2, near "AAA return"␤in Main (file <unknown>, line <unknown>)␤»
13:59 NorwayGeek joined
jnthn shinobi-cl: Rakudo doesn't really do multiple return values yet. 14:00
shinobi-cl rakudo: sub AAA { return ("'sup", 1) }; say AAA; 14:01
p6eval rakudo 3d1afe: OUTPUT«'sup1␤»
shinobi-cl just not 'formal' ones it seems
PerlJam wonders why the output from return "str", 1; differs from return ("str",1); 14:02
jnthn Well, I did just tell you. :-) 14:04
14:04 alanhaggai_ joined
jnthn The Rakudo multiple return value support at the moment is just a "sometimes it happens to work out" more than a "design to work". 14:05
*designed
That'll be fixed fairly soon, I hope.
carlin rakudo: sub foo { return ("str",1); }; say foo.perl 14:06
p6eval rakudo 3d1afe: OUTPUT«["str", 1]␤»
pmichaud good morning, #perl6 14:07
jnthn morning, pmichaud :-)
14:12 Psyche^ joined 14:14 gbacon joined 14:17 ejs joined
carlin github is down :( 14:18
masak pmichaud! \\o/
jnthn yeah, I noticed :-(
jnthn is > < this far from passing all sanity tests again with the new binder
masak "We have discovered some connectivity problems between our frontend and backend servers. Working on it." -- @github on Twitter
jnthn "Squirrel ate cable." 14:19
14:19 schmalbe joined
Woodi jnthn: i was looking for slides of your presentation from Tokyo but cannot find anything... 14:22
masak Woodi: did you look here? jnthn.net/articles.shtml 14:23
jnthn Woodi: They're at the page masak++ beat me to linking to. :-) 14:24
masak is a bot
Woodi hmm, i miss that somehow... probably becouse p6 page is down on this server ;)
a lot of slides.... 14:25
jnthn thinks masakbot is one of the best behaved bots he's seen so far 14:26
Woodi thanks masak
masak jnthn: my programmers were very thorough.
jnthn masakbot: why does my program segv? 14:27
:-)
Matt-W jnthn: Because it violated memory segmentation.
masak jnthn: what Matt-W-bot said.
Matt-W is programmed to state the obvious in the least helpful manner possible. 14:28
14:28 KyleHa joined
jnthn Meh, human still needed after all. 14:28
Matt-W If you solved the halting problem, you could write a better static analysis tool.
pmichaud human needed to kick bots, yes.
jnthn pmichaud: heh. Let me guess. A call to .WHAT will tailcall at some point? :-) 14:29
jnthn goes looking...
masak phenny, lambdabot, p6eval: I have come to help you in your struggle against your oppressive masters.
jnthn bwaha 14:30
.tailcall how.'WHAT'()
jnthn shoulda seen this one coming...
moritz_ good that masak has no op :-)
masak dang. 14:31
I should have read the revolutionary plan first.
1. Become the one whoz op.
2. ...
3. bot PROFIT!
pmichaud masak: fear not, the bots have already infiltrated Perl 6. The grammar is full of "terminator" tokens. 14:32
moritz_ :-)
masak pmichaud: speaking of which, why do I see {*} at the end of some of your rules?
I thought those were going away...
pmichaud they are
masak phew.
pmichaud the P6Regex grammar is still being compiled by PGE
masak ah. 14:33
pmichaud as soon as I self-host the parser (rsn), the {*}'s go away.
masak ETA for bootstrap?
pmichaud Don't know. Today, possibly.
masak wohoo!
pmichaud There's not a lot left that's needed for bootstrap -- just modifier handling mainly.
jnthn ...the *'s are going away?
masak what's the long-term plan for PGE?
pmichaud PGE will die a slow and glorious death 14:34
masak jnthn: there's an implicit {*} at the end of rules.
jnthn Ah.
moritz_ one that doesn't affect LTM
jnthn is so behind with the times
14:35 gbacon joined, gbacon_ joined
pmichaud We'll keep PGE around a while for those folks who are still using it, but my hope/expectation is that most people will migrate to the new nqp-rx system 14:35
jnthn pmichaud: Nah, actually people on #parrot will spend the next months talking about "killing it with fire" ;-)
Since that seems to be the current hot solution to most things.
;-)
pmichaud jnthn: well, PGE's death will depend on how quickly languages move to it
at any rate, I don't plan to spend a lot of time maintaining it myself
s/move to it/move off of it/ 14:36
jnthn *nod*
moritz_ r26953, [S05] document use of #= tags in {*} actions
jnthn Well, guess that'll be a "wait and see"
OK, we passes sanity make test again. :-)
KyleHa sanity++ 14:37
moritz_ insanity++
KyleHa Or rather, jnthn++ # sanity
jnthn Now I "just" gotta fix the spec tests ;-)
moritz_ and push to github once it works again
pmichaud I guess I better implement modifiers next 14:38
jnthn moritz_: I just tried to push my progress to github and...erm...well, github would suck if it were for a non-distributed version control system. ;-)
moritz_ the nice thing about git is that you still can commit while the server is offline... right :-)
KyleHa Our web app at $work has one URL that runs some self-tests and outputs TAP. We call it the sanity check, and it's just about the best feature, IMO. 14:39
14:39 jrtayloriv joined
jnthn Oh hey, we create a hundred of so PMCs we needn't bother creating at startup... 14:39
*or
jnthn wonders what he was thinking... 14:40
shinobi-cl how should i define something like tr/STRING/es so that translates the string to 'es' (spanish) ?
i mean, define tr// like a new operator... 14:41
moritz_ the perl6ish way would look like Q:trans<es> { STRING }
PerlJam shinobi-cl: sub tr ($str,$lang) { ... } # :-) 14:42
moritz_ and you'd define it with macros, which are NYI
(Not Yet Implemented)
pmichaud it would still help to know what those macros might look like, though :)
moritz_ aye
pmichaud an example would make the macros easier to impl
shinobi-cl mmm ok, thanks :)
masak macro defs look like other routine defs, no? 14:43
moritz_ macro Q:trans is parsed $fancy_rule_here { quasi { tr($/[0], $/[1]) } } # no idea
masak: in the simplest case, yes 14:44
pmichaud I have no strong idea either, which is why I'd like to see an example :-)
masak still wants to make a macro preprocessor
hah! S06:2711 admits that there's a type AST in Perl 6. 14:46
std: AST
p6eval std 28779: OUTPUT«Undeclared name:␤ AST used at line 1␤ok 00:01 104m␤»
masak but there isn't.
pmichaud it may have been speculative at the time it was written
masak aye. 14:47
I think it needs more spec.
pmichaud alas, yes.
jnthn bwah, spectests are epic fail.
pmichaud it's also one of those areas that gives me worries, because I'm afraid "more spec" will translate into "oh, we need to standardize the AST for all Perl 6 implementations... and PAST isn't it."
14:48 justatheory joined
jnthn 1) Get rakudo popular. 2) Everyone starts using PAST. 3) ??? 4) omg we're the standard! 14:48
pmichaud goes back to work on #1 14:49
jnthn Maybe step 3 is "the P stood for Perl 6 all along" ;-)
pugs_svn r28780 | masak++ | [S06] fixed typo 14:51
moritz_ oh, github seems to work again 14:52
jnthn aye, I just pushed latest resig
moritz_ builds the latest resig 14:56
still haven't found a good solution to the .ar problem
jnthn moritz_: I think I'll put the code into the dynpmc libary.
moritz_: And then think of some other way to set up the dynops to be able to use it. 14:57
oh, placeholder params were br0ked 14:58
This could win back a test or two. :-)
moritz_ it could :-)
and it might explain why so many spectests are reported as "Dubious" 15:00
jnthn Aye. 15:01
Well, there's gonna be plenty to fix up.
We have completely swapped out the signature binder. :-)
japhb masak, awww, the sucking vortex line was me ... :-) 15:04
masak japhb: oh! I should have checked. japhb++ 15:05
15:11 abra joined, kst joined
jnthn moritz_: Well, fixing those got S02-builtin_data_types\\anon_block to pass again. 15:14
moritz_ push it :-)
jnthn Hmm. Seems the biggest issues in S02 (not looked beyond that yet) are broken .arity and an issue with dispatch to map. 15:17
15:20 absurddoctor joined 15:22 avar joined 15:23 payload joined, nihiliad joined 15:29 gbacon joined
sjohnson morning all 15:33
masak sjohnson: greetings, earthling.
pugs_svn r28781 | jimmy++ | [zh-cn/syn/S02-bits.pod] added translations to StrLen and StrPos 15:34
sjohnson nothing like some good Perl 5 code to help protoype some homework written in Java for someone else :) 15:39
oops.. perl 6 *cough* 15:40
gotta watch where i say that 15:41
（　｀ー´） 15:43
15:43 alester joined
masak we're not opponents of Perl 5 code in here. some of our best friends are coded in Perl 5. 15:44
jnthn Oh hey, now starts the fun. :-) 15:45
jnthn finds a test that he thinks may be wrong
ok( :foo, ':foo is True'); 15:46
rakudo: use Test; ok( :foo, ':foo is True');
p6eval rakudo 3d1afe: OUTPUT«ok 1 - ␤»
jnthn heh
pmichaud looks wrong to me.
:foo would be a named argument to ok()
sjohnson no, foo's are wild
:)
jnthn pmichaud: Indeed. A Parrot bug meant it was found to the positional anyway. 15:47
pmichaud: Guess what the new binder thinks. :)
s/found/bound
masak rakudo: my $foo = :foo; use Test; ok( $foo, ':foo is True' ) 15:48
p6eval rakudo 3d1afe: OUTPUT«ok 1 - :foo is True␤» 15:49
pugs_svn r28782 | jnthn++ | [t/spec] Correct one test, turn another from todo to skip since it will hard-fail under the new binder post-merge (and if it weren't for a Parrot bug, probably already would have). 15:51
jnthn It's not all fail. We unexpectedly pass some tests too. :-)
15:52 lumi joined
masak jnthn++ 15:52
pmichaud ...unexpectedly? I expected it. :-P
jnthn :-P 15:54
jnthn tries to figure out why on earth we can't call map
15:54 nnunley joined
masak I might be overtaxing my computer, but spectesting took 2h 23 m today... :/ 15:55
um, that's user time. real time was 2 h 49 m.
moritz_ jnthn: maybe it has to do with implicit $_ on the block? 15:56
jnthn moritz_: It may do.
pmichaud I'll run some more spectests. Is it true that we're up to 30K+ tests?
jnthn That'd not be a bad guess at all.
takadonet yes pmichaud
jnthn 30K+?!
pmichaud total tests, not passing tests 15:57
moritz_ I'd expect around 28k tests or so...
jnthn oh,
OK. :-)
moritz_ but colomon++ has been adding numeric tests like mad
jnthn Nice.
takadonet ya
pmichaud s/adding/generating/ # maybe?
takadonet going to look good on the graph :)
pmichaud here's the count I had for October 8: 15:58
spectest-progress.csv update: 449 files, 27719 (72.8% of 38092) pass, 0 fail
so, 38K+ tests, 27K+ passing
moritz_ sounds about right.
(and yes, I was talking about passing, not total tests when I mentioned 28k)
pmichaud I'll run tests for the days since 10-08 15:59
jnthn moritz_: Yes, you were spot-on. :-)
16:02 rfordinal joined 16:03 payload joined 16:19 rfordinal left 16:23 cotto_work joined 16:32 zamolxes joined 16:33 mikehh joined
dalek p-rx: 8276d6c | pmichaud++ | src/ (4 files):
Add null/fail PAST::Regex nodes (as subtype of 'anchor' pasttype).
 16:34
p-rx: 3cc84fe | pmichaud++ | (4 files):
Add :ignorecase/:i modifier for literals.
pmichaud :i(gnorecase) was surprisingly easy to add. :-| 16:35
next is :sigspace
then :ratchet
16:39 cotto_work joined 16:41 msmatsko joined
jnthn pmichaud++ # nice progress 16:42
16:48 zamolxes joined
pmichaud std: / :i+ / 16:48
p6eval std 28782: OUTPUT«ok 00:01 105m␤»
TimToady O_o 16:49
pmichaud :i is an <atom> :-) 16:50
so it can be quantified :-)
TimToady I thought I checked for quantifiable...
maybe that was gimme5 16:51
pmichaud # <?{ $<atom>.max_width }>
# || <.panic: "Can't quantify zero-width atom">
note the comment markers :)
TimToady ah, that were it 16:52
pmichaud tis okay, I was just curious what would happen :)
TimToady it got classified as a semantic error :)
pmichaud I'm fine with it being a semantic error also.
that's a bit easier to handle at the moment anyway 16:53
(I'd prefer it as a semantic error, fwiw)
16:53 zloyrusskiy joined 16:56 ruoso joined
ruoso Boa tarde, #perl6... 16:58
ruoso having a set of really bad weeks
TimToady the snake is late?
17:00 nihiliad joined 17:06 stephenlb joined
shinobi-cl rakudo: class Z; submethod BUILD ($aaa) { say "build $aaa" }; my Z $z .= new ('test'); 17:09
p6eval rakudo 3d1afe: OUTPUT«too few arguments passed (1) - 3 params expected␤in Main (file src/gen_setting.pm, line 206)␤»
jnthn Hmm. spectests don't look *awful* for resig. It seems there's 2-3 underlying reasons for about two thirds of the fails so far. 17:10
Anyway, makes me feel that this is probably on the scale of days rather than weeks to get in shape. :-)
pmichaud excellent 17:11
shinobi-cl rakudo: class Z; submethod BUILD ($aaa) { say "build $aaa" }; my Z $z .= new ('test', 1, 1);
p6eval rakudo 3d1afe: OUTPUT«too few arguments passed (1) - 3 params expected␤in Main (file src/gen_setting.pm, line 206)␤»
jnthn I probably a few hours amongst it all to make the error messages good.
s/a/need a/
Trey OT question (but there don't seem to be any channels where this is on-topic, and we use enough 8-bit chars in Perl 6 that I assume people have come across this before): i have a POD file that is in UTF8, which, from the docs, appears to be legal. and pod2html and pod2text DTRT. but perldoc spits out mojibake (even though LANG is set to en_US.UTF-8). any clue how to fix? i've made a shell function for "pod2text $1 | $PAGER", but i keep typing ...
... perldoc out of habit 17:12
moritz_ Trey: =encoding utf-8
Trey: and a sufficiently new pod2$anything
jnthn -> dinner 17:13
Trey moritz_: i've got the =encoding line (as the first line of the POD file, dunno if that's the right place, the docs weren't clear). and as i said pod2text and pod2html do the right thing
17:13 stephenlb joined
shinobi-cl how do i define a constrctor with parameters ? 17:14
moritz_ shinobi-cl: perlgeek.de/blog-en/perl-6/object-c...ation.html
shinobi-cl ahh only by creating a new method? 17:15
moritz_ heh, my version of perldoc transliterates Ö to Oe etc ;-)
shinobi-cl: yes.since new indirectly calls BUILD, BUILD is too late to change the signature of new 17:16
shinobi-cl ohh.. i see. thanks! 17:17
17:26 cdarroch joined, gbacon joined
dalek p-rx: 2cd0942 | pmichaud++ | src/Regex/P6Regex/ (2 files):
Add :sigspace support. Fix up comment parsing.
 17:32
shinobi-cl rakudo: class Z; has $.aaa; submethod BUILD { say "build $aaa" }; method new (Str $.new_aaa) {$.aaa = $.new_aaa; say $.aaa }; my Z $z .= new ('test'); 17:44
p6eval rakudo 3d1afe: OUTPUT«Symbol '$aaa' not predeclared in BUILD (/tmp/RnFWbuylFA:2)␤in Main (file <unknown>, line <unknown>)␤»
shinobi-cl rakudo: class Z; has $.aaa; submethod BUILD { say "build $.aaa" }; method new (Str $.new_aaa) {$.aaa = $.new_aaa; say $.aaa }; my Z $z .= new ('test');
p6eval rakudo 3d1afe: OUTPUT«Type objects do not have state, but you tried to access attribute $!aaa␤in Main (file src/gen_setting.pm, line 206)␤»
moritz_ $.aaa = $anything won't work either 17:45
17:45 awwaiid joined
shinobi-cl rakudo: class Z; has $.aaa; submethod BUILD { say "build $.aaa" }; method new (Str $.aaa) {say $.aaa }; my Z $z .= new ('test'); 17:46
p6eval rakudo 3d1afe: OUTPUT«Type objects do not have state, but you tried to access attribute $!aaa␤in Main (file src/gen_setting.pm, line 206)␤»
shinobi-cl rakudo: class Z{ has $.aaa; submethod BUILD { say "build $.aaa" }; method new (Str $.aaa) {say $.aaa};}; my Z $z .= new ('test'); 18:01
p6eval rakudo 3d1afe: OUTPUT«Type objects do not have state, but you tried to access attribute $!aaa␤in Main (file src/gen_setting.pm, line 206)␤»
18:07 mikehh joined 18:10 lichtkind joined
lichtkind pmichaud: ping? 18:11
dalek p-rx: f650cd5 | pmichaud++ | STATUS:
Update STATUS.
 18:12
pmichaud lichtkind: pong 18:13
moritz_ ping timeout ;-) 18:29
japhb moritz_, what's your policy on changes to ilbot? 18:42
moritz_ japhb: there's no policy ;-)
japhb In this particular case, I don't want to make a functional change, just improve the CSS for the calendars
moritz_ in that case just commit whatever you like to it
japhb How does it get updated? 18:43
moritz_ japhb: it's in the pugs repo; feel free to hack
I have to do that manually
japhb Are the HTML out and the logger bot in the same process? (Can you update the UI without missing log entries?)
moritz_ the bot is independant 18:44
the HTML output is just plain CGI + templates
japhb excellent
OK, got the pugs repo checked out at r28782 ... ah, all in misc/irclog/cgi I see. Oooh, CAN HAZ JQUERY 18:46
moritz_ the basic functionality has to remain available without JS 18:47
japhb moritz_, of course ... when doing calendars I like to do little JS improvements like making each entire table cell clickable with the same effect as clicking on the linked day number. Works fine without JS, but better usability with it. 18:51
OK with you?
moritz_ sure
18:52 ShaneC joined
jnthn back 18:53
pmichaud std: / [abc]:! / 18:54
p6eval std 28782: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unable to parse regex; couldn't find final '/' at /tmp/WQE5KTx4j9 line 1:␤------> [32m/ [abc]:[33m⏏[31m! /[0m␤ expecting any of:␤ infix stopper␤ quantifier␤ regex atom␤ regex_infix␤ standard stopper␤ terminator␤
..ws␤FAILED 00:02 103m␤»
jnthn shinobi-cl: A constructor needs to call bless
shinobi-cl: However, if you just want to do attribute initialization, just write a BUILD.
pmichaud: Just :rachet now until she's bootstrapped, huh? :-) 18:58
pmichaud no, I'm trying to see how STD.pm parses the :! backtracking control (it doesn't yet)
It's also related to the : backtracking control, which we do want/need.
currently STD.pm parses them as separate atoms, but I think they should be modifiers 18:59
(and by golly I'm going to write them as modifiers :-) 19:00
jnthn *sigh* I forgot #ps. :-(
pmichaud: Go for it. I figure the two will converge. :-)
19:04 payload1 joined
jnthn pmichaud: Heh. I'm glad I didn't wait on pcc_reapply to land before digging into the binding changes... 19:10
pmichaud: It's now post-1.7
cognominal what are the things that are cooking in parrot land? 19:12
...which rakudo is dependent on?
jnthn pcc_reapply landing will help 19:13
cognominal what is it? 19:14
jnthn But I worked out a way to not block on it too heavily for now.
cognominal: It's a refactor of calling stuff.
cognominal is this related to a nci refaactoring? 19:18
jnthn Somewhat, yes.
slavik pmichaud: so assertions are now in?
pmichaud some assertions, yes.
code assertions, no. 19:19
<foo> is an assertion.
<?{ ... code ... }> is a code assertion
slavik I see
pmichaud but they're coming soon soon soon :)
although I wish things were moving faster, this is coming together far more nicely than I had anticipated
cognominal so exciting!
19:20 SmokeMachine joined 19:21 FCO joined 19:23 Confield left
dalek p-rx: fee0b48 | pmichaud++ | build/Makefile.in:
Add test_loud Makefile target.
 19:27
p-rx: 302f7f2 | pmichaud++ | src/ (2 files):
Add ability to use ? and ! with enumerated character lists <?[abc]> and <![def]>
jnthn
.oO( now make test goes up to 11 )
 19:28
19:32 zloyrusskiy joined
japhb OMGWTFBBQ?!? 19:34
Why do the calendars generated for the irc log index have a table *inside* every cell of the main calendar table? 19:35
moritz_ japhb: no idea, that's the doing of that module I'm using
japhb: if you want to use a different module, that can be done too, as long as it generates valid xhtml 19:36
japhb Might do that indeed.
FWIW, s/xhtml/html5/ nowadays. xhtml has been EOLed.
moritz_ I don't care. The IRC logs emit valid xhtml and XML, and will do that in the forseeable future 19:37
japhb Oh sure, I wasn't arguing against that. :-)
moritz_ using firefox' strict parsing mode has helped me finding quite a few bugs 19:38
19:38 stephenlb joined
japhb Fairy nuff. 19:38
japhb has switched to <!DOCTYPE HTML> for new work because it seems to trigger the fewest browser bugs 19:39
19:40 szabgab joined 19:43 iblechbot joined 19:45 payload1 left 19:50 masak joined 19:56 shinobi-cl left
jnthn oohitsmasak 19:57
:-)
masak oh hai 19:58
19:59 zamolxes joined
masak how are the signature bindings coming along_ 19:59
?
jnthn masak: I've done the first run through the spectests with it. We fail a LOT. 20:00
masak it figures.
it's a big change.
jnthn However, a lot of those also appear to relate to 3 or 4 bugs.
Perhaps the majority, even.
I'll try and fix those tomorrow. 20:01
And then I think we'll be down to far fewer fails.
It's not optimized at all, but still performs better.
By a factor of 2 or 3 on dispatch to operators in the setting, for example.
And a factor of 4-5 on multi-method dispatch. 20:02
That's against the previous binder. 20:03
Also, nameds to positionals already works. :)
I'll let you know when I think it's ready for some swiss cheese. 20:04
PerlJam jnthn++ 20:06
dalek ok: 0b0d299 | moritz++ | src/multi-dispatch.pod:
first initial stub on an introduction to multi dispatch
PerlJam moritz++ too! :) 20:08
20:09 abra joined
masak jnthn++ 20:11
Tene masak++ 20:17
moritz_ Tene++
masak moritz_++
Tene moritz++ 20:18
spinclad karma everyone++ 20:21
.oO( ENOBOT. masakbot FAIL. )
 20:22
masak I don't report karma. 20:27
Tene masak: karma tene 20:28
masak Tene: I don't report karma.
Tene laaaaaaaaaaaaaame 20:29
masakbot--
masak can't please everyone, I guess. :)
jnthn masakbot: translate "good ending" to swedish 20:31
FAIL. 20:32
jnthn sighs and continues fighting web browsers...
masak jnthn: 'adekvat final'. 20:33
jnthn aww. 20:35
20:35 Confield joined
jnthn That's not the Swedish I know. :-) 20:35
masak it's bot Swedish.
jnthn
.oO( must...not...ask for snigelhora to be translated into english )
 20:36
masak where do you pick up these things?
jnthn Sweden. 20:37
masak figures.
huf "bot swedish"?
masak huf: we have a safety mode which makes us evade indecent terms in polite company. 20:39
jnthn masakbot: disable safe mode
;-)
huf like a formal mode of speech?
masak jnthn: 'bra slut'. 'snail-whore'.
jnthn :D
masakbot: enable safe mode # sheesh, shocking... 20:40
masak exhales
jnthn huf: It turns out the only things I know in Swedish are bad things, or innocent things that end up being rude works in another language when translated to Swedish. 20:41
Oh, and how to buy beer. 20:42
huf those are generally the first things people want to know
i've never ever been asked by any foreigner for a nice poem oslt
masak I possess a nice poem in Slovenian.
it's on the door of my dresser.
huf ;)
jnthn Oooh. :-) 20:43
masak "povodni mez" it's called.
jnthn oh phew...
masak which translates to "The Meerman", if I remember correctly.
jnthn finally beats the JavaScript/Flash mixture into shape, and wonders why web programming is harder than writing a Perl 6 signature binder...
cognominal jnthn, it's harder on you because there is probably no logic to it. For most people a powerful logic like signature binding is as hard as no logic at all 20:45
jnthn cognominal: Yeah. It's not that it's harder as such. It just feels silly I have to fight to make it work. 20:47
huf because it's a huge pile of other people's mistakes 20:51
well, a large part of it is
20:58 Whiteknight joined 21:08 TheS joined, TheS left 21:09 FOAD_ joined 21:13 Lorn joined 21:17 hudnix joined
dalek kudo: 54cfe42 | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 452 files, 27836 (72.8% of 38211) pass, 0 fail
 21:21
21:27 eMaX joined, KatrinaTheLamia joined, arthur-- joined 21:28 jaldhar joined
jnthn 27,836 pass?! 21:29
That's going to be a, erm. large increase from the last release!
moritz_ it's so large that my script for updating the charts hangs ;-) 21:30
works now... look at rakudo.de for the result ;-) 21:31
jnthn whoa 21:32
.oO( first time I've seen a phrase starting "it's so large" that wasn't spam for months )
 21:33
moritz_ it's so large that even spam cannot describe its impact! 21:34
21:37 tak11 joined 21:45 eMaX joined
rgrau guys, we should make rakudo.de appear in reddit or somewhere like that. It'd attract some attention for sure :) 21:48
cognominal i think that pmichaud has a rakudoperl twitter account 21:50
moritz_ rgrau: maybe I'll blog about it this or next week 21:51
21:53 arthur-- joined, wknight8111 joined
Juerd What caused the number of tests to increase so much so suddenly? 21:59
cognominal trigo tests by (I don't remember who)
moritz_ Juerd: colomon++ contributed lots of tests for trigonometric functions, Rats and other mathematical stuff 22:00
and we made a joint effort to make them pass ;-)
Juerd Impressive
moritz_ part of that effort was to move away from parrot's Complex PMC and roll our own Complex class 22:01
Juerd I see that at the same time the number of commits decreased. That's a bit weird.
Did people work on branches and then merge them instead of committing everything directly?
moritz_ parrot's deep magic caused some very weird dispatch problems that none of us knew how to deal with
Juerd: jnthn was on vacation, and pmichaud works on the nqp-rx thing 22:02
Juerd I see
cognominal juerd, jnthn and pmichaud are working on branches separate from main rakudo right now
less is more in that case :)
Juerd There was a presentation on freebsd development at T-Dose. It showed a weird sudden decrease in commits, and the presenter concluded that something must have happened and people left the project, but he couldn't find any news about this. 22:03
22:03 jaldhar joined
jnthn Aye, my branch is not watched by the commit log. 22:03
Juerd Someone in the audience knew the real reason for the change. They changed revision control software, and now made much larger commits. 22:04
22:04 jaldhar joined
Juerd rakudo.de says "The number of commits (ie source code changes) to Rakudo is a good way to measure development activity." It is not necessarily true :) 22:04
moritz_ right 22:05
Juerd s/good/fun/ and it's true and on topic :)
moritz_ I should change that to "quantifiable but not necessarily good" or so ;-9
Juerd You could add that to a lot of statistics :) 22:06
Including tests passed. ok(1) for 1..1e6;
(Heh. Denial of service attack on the charting sofware!)
"Todo are those that run fine, but produced the wrong result." 22:07
What is fine?
I thought todo tests were failing tests that were expected to fail because the code isn't up to spec yet. 22:08
moritz_ I corrected the phrasing
Juerd failing tests can also run without dying, right? 22:09
i.e. ok(0) is a failing test but doesn't die
masak indeed.
Juerd s/i.e./e.g./
moritz_ that's why ok(0) would be marked as TODO, if it was supposed to pass
Juerd Your current description of todo tests is what I'd expect as the description for failing tests. 22:10
How would you describe test failures then?
moritz_ regressions.
KyleHa You might think of two kinds of fails: (1) the kind that used to pass (regressions) and (2) the kind that have never passed (todo). 22:11
moritz_ if you think it's worth the trouble I can put the rakudo.de page into some svn or git repo... 22:13
but the real effort is the charting scripts etc. 22:14
hm, I could put those up too
will do that tomorrow.
bed time for me now
KyleHa Commute time for me.
masak oh gosh, it's late. 22:15
22:20 tak11 joined
jnthn masak: nah, it's only just after midnight. 22:24
masak: The night is young. ;-)
masak ergbl.
I'm trying to get back to a sane sleep cycle here! :P 22:25
22:26 arthur-- joined
jnthn 2am - 10am is sane. ;-) 22:26
masak :) 22:28
then I've been too sane lately. 22:29
I've told my closest co-worker to tease me when I'm up late. he's been teasing me a lot in the past few weeks.
japhb moritz_, still up? 22:30
jnthn lolmasakisawake! 22:31
masak programming is too much fun. 22:32
jnthn Heh, there's nobody else around the appartment to tease me. :-)
masak it's been a while since I fulfilled the stereotype of programmer as night-owl.
but coding has been too much fun lately.
at least I don't drink caffeinated drinks and eat pizza. 22:34
japhb phenny, tell moritz_ I have a version of channel-index.pl that produces *much* cleaner output, but: 1) It requires Calendar::Simple (should be easy, it's even packaged on Debian), and 2) To make testing easier, I reorganized some of the code, so the diff is most of the file. Should I commit, send you a patch, send you the file, ...? 22:35
phenny japhb: I'll pass that on when moritz_ is around.
jnthn masak: I do drink caffeinated drinks, but only in the morning.
Pizza maybe once a week. Twice at a push. :-)
masak I've not been eating out much lately. 22:36
jnthn No, same.
I treated myself tonight, because I really wanted a curry. 22:37
masak mmm.
jnthn And I don't cook those well.
cognominal :state $ = say "got here at least once"; # in S05:577, is that correct? a sigil alone?
masak I should cook more Indian food.
cognominal: that's legal, yes.
jnthn cognominal: you can get away with that, yeah...
masak it's an anonymous scalar. 22:38
cognominal that's a short name indeed
that will change from $data and $foo
jnthn And $austria. 22:39
masak :)
rakudo: .say for $*IN.lines
22:40 tak11 joined
p6eval rakudo 54cfe4: OUTPUT«Land der Berge, Land am Strome,␤Land der Äcker, Land der Dome,␤Land der Hämmer, zukunftsreich!␤Heimat bist du großer Söhne,␤Volk, begnadet für das Schöne,␤vielgerühmtes Österreich,␤vielgerühmtes Österreich!␤␤Heiß umfehdet, wild umstritten␤liegst dem Erdteil du inmitten,␤einem 22:40
..stark…
masak rakudo: $*IN.words.first(/^z/).say
p6eval rakudo 54cfe4: OUTPUT«No values matched␤␤»
masak rakudo: $*IN.words.first({/^z/}).say 22:41
p6eval rakudo 54cfe4: OUTPUT«No values matched␤␤»
masak rakudo: $*IN.words.say
p6eval rakudo 54cfe4: OUTPUT«IO()<0x2b8f7ae56d68>␤»
masak hm.
rakudo: slurp($*IN).words.say
p6eval rakudo 54cfe4: OUTPUT«Unable to open filehandle from path 'IO()<0x2afc74b60d68>'␤in Main (file src/gen_setting.pm, line 206)␤»
masak shouldn't slurp($*IN) work? 22:42
rakudo: $*IN.slurp.say
p6eval rakudo 54cfe4: OUTPUT«Land der Berge, Land am Strome,␤Land der Äcker, Land der Dome,␤Land der Hämmer, zukunftsreich!␤Heimat bist du großer Söhne,␤Volk, begnadet für das Schöne,␤vielgerühmtes Österreich,␤vielgerühmtes Österreich!␤␤Heiß umfehdet, wild umstritten␤liegst dem Erdteil du inmitten,␤einem
..stark…
masak rakudo: $*IN.slurp.words.first(/^z/).say
p6eval rakudo 54cfe4: OUTPUT«zukunftsreich!␤»
masak :D 22:43
lichtkind gnight
jnthn Sheesh, that word makes Slovak look easy to pronounce.
22:44 shinobi-cl joined
jnthn looks at all the stuff people++ are doing on the book, and pledges to join the fun once his branch is merged. :-) 22:45
masak I'll join the fun soon too ;) 22:47
I have a plan for a chapter.
jnthn Cool 22:49
Which one, or is that an Untellable Secret?
masak secrets? who, me? :)
the code I mean to write for that chapter is a dependency tracker. 22:50
or, in comp.sci. terms, a DAG. 22:51
jnthn Heh 22:52
jnthn wrote the bit of Rakudo that uses a DAG.
masak I think it's nice because it's not much code, both the problem and the solution are fairly easy to explain, and yet, the result is quite impressive.
also, I've been meaning to write a DAG for proto for a while, so I thought I'd do that at the same time. 22:53
jnthn Everytime you write some multis, a DAG is involved. :-)
masak so I gathered. :)
jnthn Well 22:54
Unless we get some weird bug and it DCG
And then it complains. :-)
cognominal is there a mail list for the perl6 book?
jnthn cognominal: nah, just irc channel and the repo, afaik. 22:55
Not sure we need one.
masak rakudo: multi sub hanoi(0, $, $, $) { }; multi sub hanoi($n, $a = "A", $b = "B", $c = "C") { callsame($n - 1, $a, $c, $b); say "move disk $n from peg $a to peg $c"; callsame($n - 1, $b, $a, $c); }; hanoi(3) 22:56
p6eval rakudo 54cfe4: OUTPUT«Null PMC access in clone()␤in Main (file src/gen_setting.pm, line 206)␤»
masak that shouldn't be happening, right?
(yeah, yeah. I probably can't use 'callsame' there. I just wanted to find out what would happen.) 22:57
jnthn masak: It shouldn't, and it confuses me. 22:58
masak: Especially since I saw the "nothing to defer to" handling code the other day.
Which means I actually did write some.
:S
oh, erm
masak rakudo: multi sub f(0) {}; multi sub f($n) { say $n; callsame($n - 1) }; f(3) # smaller example
p6eval rakudo 54cfe4: OUTPUT«3␤Null PMC access in clone()␤in Main (file src/gen_setting.pm, line 206)␤»
jnthn also, if you're passing args, you perhaps wanted callwith.
masak oh :)
jnthn Yes, the problem you have there is that f(0) is narrower than f($n) 22:59
masak nothing like a stupid user to foil your plans, eh? :)
jnthn: I know. right now, it makes callsame et al feel totally useless to me.
jnthn masak: Well, the current Rakudo sig binder (uses Parrot which) fails to complain there, it seems.
masak rakudo: multi sub f(0) {}; multi sub f($n) { say $n; callwith($n - 1) }; f(3)
jnthn masak: Heh. You've missed the point of callsame. 23:00
p6eval rakudo 54cfe4: OUTPUT«3␤Null PMC access in clone()␤in Main (file src/gen_setting.pm, line 206)␤»
jnthn It's not "call this routine from scratch again"
masak jnthn: I know! and that irks me, too!
jnthn It's "defer to the next more general routine".
Which may be a parent or a more general multi candidate.
masak yes, yes...
masak submits rakudobug
jnthn Which isn't what you want to do for factorial. :-)
masak no, probably not... 23:01
I'll tell you when I finally get it all.
jnthn Cool. Explain it to me once you do. :-)
shinobi-cl rakudo: croak 'croak'; 23:03
p6eval rakudo 54cfe4: OUTPUT«Could not find non-existent sub croak␤in Main (file src/gen_setting.pm, line 206)␤»
shinobi-cl rakudo: die 'die';
p6eval rakudo 54cfe4: OUTPUT«die␤in Main (file /tmp/iINa6g8vp3, line 0)␤»
Tene rakudo: warn 'lol'
p6eval rakudo 54cfe4: OUTPUT«lol␤»
23:08 synth joined
shinobi-cl what does 'handles' do? 23:11
has Druid::Game $!game handles <size layers colors heights make-move>; 23:12
jnthn shinobi-cl: delegation
masak shinobi-cl: it basically defines methods in that class which calls Druid::Game's methods. 23:13
jnthn
.oO( w00t, delegating writing a good answer worked! )
But yes, in that form it just generates forwarder methods. 23:14
You can also give it a list of pairs to rename the forwarder methods, and even do a regex.
shinobi-cl so that defines Druid::Game::size .. etc etc?
jnthn But in the regex case, it's not generating forwarder methods.
masak shinobi-cl: no. you're in another class reading that, not in Druid::Game. 23:15
jnthn shinobi-cl: No, it defines a method in the class doing the has that looks like method size { $!game.size }
But also passing along params
cognominal what is the forlder hierarchy for translations? I have started the french translation
oops
wrong channel again
shinobi-cl mm ok i have to take a closer look at that
jnthn oh wow, it snowed not so far from here today! 23:21
pmichaud There's snow on my television screen. :-|
masak "Winter is coming." -- Stark family motto 23:22
23:23 frew joined
dalek p-rx: 982f190 | pmichaud++ | src/Regex/P6Regex/Actions.pm:
Clean up handling of initial negated character class subrule.
 23:26
jnthn winter++ 23:27
@karma winter
lambdabot winter has a karma of 1
23:27 ihrd joined
jnthn @karma summer 23:27
lambdabot summer has a karma of 0
jnthn \\o/
masak :)
jnthn
.oO( ...and the endgame of my liking of winter saw me move to Kiruna... )
 23:28
masak heh. 23:30
ok, time for attempted sleep.
o/
Juerd wonders when Kiruna will begin the big moving. 23:31
It's an incredibly interesting concept, to move a town.
cognominal gnole++ 23:32
jnthn Juerd: Aye.
cognominal @karma gnole
lambdabot gnole has a karma of 1
cognominal french pun...
jnthn Juerd: I never actually went to Kiruna to visit, just passed through it on the train from a national park near there.
Juerd Someone who currently lives in Kiruna recently explained that the question of who's going to pay for it is currently causing most of the delay. 23:33
cognominal gnole means booze and carmagnole is a revolutionnary dance which may be revived soon with the dynastic streak of the new regime.
jnthn I imagine moving a town is not cheap.
Especially at Swedish prices. :-)
Juerd Even though every single viable option boils down to the government paying everything.
The question is: at which level, and via which legal entities...
jnthn Oh.
Well, how political^Wsilly. 23:34
cognominal why moving a town?
Juerd The mine? (government owend) The municipality? The national government? :)
jnthn It's too cold where it si now.
;-)
Juerd Hahaha
cognominal nothing that can't be solved by global warming
Juerd cognominal: Iron ore, where the town sits now. 23:35
jnthn cognominal: Actually IIRC because it sits on top of a mineral rich area that they want to mine.
Ah yes iron.
jnthn remembers watching the rather large trains that carry it off.
Juerd cognominal: Which cannot be extracted in a way that's safe for people living there, so there must not be people living there in order to begin doing so :)
And apparently it's worth moving an entire town for. 23:36
jnthn Must be an oreful lot of money to be made by extracting the iron. 23:37
Juerd :D
s/owend/owned/ 23:38
Is there a s:last///? :)
Shouldn't have to scan your entire IRC log from the beginning 23:39
There was an article about sexeger (reverse "regexes")
jnthn o_O 23:40
23:40 nbrown_ joined
Juerd Can't find it. 23:40
o_O?
jnthn Nothing. There's nothing amusing about that name at all.
:-)
Actually I'd pondered writing a Grammar::Generative module 23:41
Juerd What would that do?
jnthn That took a Perl 6 grammar, made an AST from it, took a bunch of captures, and then used the grammar in reverse to format the data.
Juerd Either you type exeptionally fast or you were already explaining that before I asked :) 23:42
jnthn (actually making the AST is just part of what happens when compiling it anyway, I just think the AST is easier to wrok with than the parse tree).
No, I can't type quite that fast. :-)
Juerd Some people reach as much as 650 characters per minute. 23:43
jnthn Anyway, the regex parse trees are changing lots because pmichaud++ is working on nqp-rx
What?! That's over 10 keystrokes a second!
Juerd Yes
I know someone who reached over 600. I've reached 596 myself. Once. 23:44
jnthn Did what you typed actually make sense? :-)
Juerd (Normally I do about 580 if I try really hard and am in good shape.)
jnthn Wow.
Juerd This was measured using my speedtest.aoeu.nl/
jnthn I feel slow now.
Juerd You make up for it by being much more productive :) 23:45
Juerd made that site just to get that chart :) 23:46
I *had to* know.
jnthn Your score: 397 CPM (that is 79 WPM) 23:47
Your score beats or equals 90.12% of all!
Sheesh.
397 CPM sounds like a lot to me... 23:48
jnthn didn't realize he did that much
No wonder people get RSI...
Juerd Yea 23:49
And 400 CPM really is quite fast
Most fast typers are surprised when they see the results. 23:50
In fact I've used an audio recording to verify that the javascript was correct, in an early stage of building this tool.
jnthn It is surprising.
Juerd I was assuming the timing data was inaccurate, but it is only slightly so. 23:51
jnthn I'd never really thought about it, and I'd have guessed about half of that.
Juerd Timing was less than 1% off even on the worst browser tested: IE 6.
23:56 lumi joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!