Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2013-05-26

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:06 woosley joined #perl6
00:06 cognominal__ .u 00a0
00:06 yoleaux U+00A0 NO-BREAK SPACE [Zs] ( )
00:08 sorear [Coke]: my @lines := 'wtf.pl'.IO.lines  ?
00:09 sorear I think that's specced to be lazy.  might not be implemented, also, might interact weirdly with sink context
00:10 sorear to answer your first question, while $data-file.get -> $line { take $line } or while defined my $line = $data-file.get { take $line }
00:19 [Coke] sorear: danke. it seems lazy enough. now to rewrite my stupid script to work lazily.
00:19 [Coke] I wouldn't have to write more elegant code if rakudo was faster. :)
00:26 tgt joined #perl6
00:43 PacoAir joined #perl6
01:41 imIKARi joined #perl6
01:41 sorear my @lines = { $data-file.get } ...^ !*.defined maybe
01:42 FROGGS_ joined #perl6
01:45 woosley joined #perl6
01:50 IKARi__ joined #perl6
01:53 imIKARi joined #perl6
01:54 shinobicl_ joined #perl6
02:17 btyler joined #perl6
02:21 IKARi__ joined #perl6
02:39 imIKARi joined #perl6
03:23 [Coke] I find myself wanting a NOTFIRST phaser. :P
03:27 diakopter [Coke]: I've wanted that for many years
03:27 diakopter srsly
03:28 [Coke] ended up with FIRST my $first = True;   ... $first = False;}
03:28 sorear is that called NEXT?
03:28 [Coke] and then if !$first ...
03:28 [Coke] no, next will fire the first time through also, I think.
03:29 [Coke] r: for 1..3 { NEXT say "eek"}
03:29 camelia rakudo 45ae2d: OUTPUT«eek␤eek␤eek␤»
03:53 preflex_ joined #perl6
03:54 [Coke] r: for 1..10  { my $a = 3; FIRST { say $a} } # masakbug?
03:54 camelia rakudo 45ae2d: OUTPUT«(Any)␤»
04:03 adu joined #perl6
04:05 kshannon joined #perl6
04:08 [Coke] that makes it hard to use FIRST, I think.
04:13 [Coke] opened a ticket.
04:19 diakopter r: for ^5 { if FIRST { return 1} { say $_ } }
04:19 camelia rakudo 45ae2d: OUTPUT«Can not get attribute '$!storage' declared in class 'Parcel' with this object␤  in method reify at src/gen/CORE.setting:5672␤  in method gimme at src/gen/CORE.setting:6101␤  in method eager at src/gen/CORE.setting:6080␤  in block  at /tmp/2MKSphEXIU:1␤␤»
04:19 LordV joined #perl6
04:20 diakopter hard to know whether that's in the compiler
04:20 diakopter oh nm; can't be
04:25 Psyche^ joined #perl6
04:30 imIKARi joined #perl6
04:37 diakopter [Coke]: ping
04:37 shinobicl_ joined #perl6
04:38 [Coke] pong
04:38 dalek perl6-roast-data: e395511 | coke++ | bin/days_since.pl:
04:38 dalek perl6-roast-data: Speed up by 50%
04:38 dalek perl6-roast-data:
04:38 dalek perl6-roast-data: Avoid processing every line in the file - only process those
04:38 dalek perl6-roast-data: lines we need to find a state change.
04:38 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/e395511114
04:39 diakopter [Coke]: emulate NOTFIRST: for ^5 { (state $a)++&& say '_'; .say }
04:40 [Coke] nifty.
04:40 diakopter r: for ^5 { (state $a)++&& say '_'; .say }
04:40 camelia rakudo 45ae2d: OUTPUT«0␤_␤1␤_␤2␤_␤3␤_␤4␤»
04:40 diakopter r: for ^5 { (state $a)++&& print '_'; .print }
04:40 camelia rakudo 45ae2d: OUTPUT«0_1_2_3_4»
04:41 diakopter r: for ^5 { (state $)++&& print '_'; .print }
04:41 camelia rakudo 45ae2d: OUTPUT«0_1_2_3_4»
04:41 diakopter golfed
04:42 [Coke] niecza has been dirty for 312 days. pugs has been clean for 8 days. rakudo has been dirty for 21 days.
04:44 diakopter r: for ^5 { (state$)++&& print '_'; .print }
04:44 camelia rakudo 45ae2d: OUTPUT«0_1_2_3_4»
04:44 diakopter golfed more
04:44 diakopter n: for ^5 { (state$)++&& print '_'; .print }
04:44 camelia niecza v24-51-g009f999: OUTPUT«0_1_2_3_4»
04:45 diakopter r: for ^5 { (our$a)++&& print '_'; .print }
04:45 camelia rakudo 45ae2d: OUTPUT«0_1_2_3_4»
04:45 diakopter golfed more
04:55 kurahaupo joined #perl6
04:59 dalek rakudo/nom: 0e206e0 | coke++ | INSTALL.txt:
04:59 dalek rakudo/nom: win32 spectest smolder prereq notes
04:59 dalek rakudo/nom:
04:59 dalek rakudo/nom: RT #78152
04:59 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0e206e0b03
05:01 kshannon joined #perl6
05:03 [Coke] nr: say lines[0]
05:03 camelia rakudo 45ae2d, niecza v24-51-g009f999: OUTPUT«Land der Berge, Land am Strome,␤»
05:05 * [Coke] wonders if TODO tickets are helpful for rakudo, or if that is better served in a doc, or a wiki page or something.
05:14 dmol joined #perl6
05:20 IKARi__ joined #perl6
05:24 woosley left #perl6
05:26 shinobicl_ left #perl6
05:36 kaare__ joined #perl6
05:38 diakopter [Coke]: what if FIRST and LAST took else blocks
05:47 patspam joined #perl6
05:48 kurahaupo joined #perl6
05:49 ivan`` joined #perl6
05:52 SamuraiJack_ joined #perl6
05:53 Juerd_ joined #perl6
05:53 timotimo_ joined #perl6
05:53 jtpalmer joined #perl6
06:05 shinobicl_ joined #perl6
06:13 imIKARi joined #perl6
06:42 [Coke] diakopter: that would get me the functionality. not sure I like the syntax.
06:50 * [Coke] sighs, as FIRST just isn't specced to do the useful thing from his POV.
06:50 * [Coke] sticks with "my $first = True; loop { ... ; $first = False}
06:52 * [Coke] boggles, as START means first.
06:52 shinobicl_ open question: do you think is still worth it to support relational databases?
06:53 labster yes.
06:53 labster I tell them how awesome they are when they retrieve my data.
06:53 [Coke] .ask timtoady if we should switch the names of FIRST and START, maybe.
06:53 yoleaux [Coke]: I'll pass your message to timtoady.
06:54 labster SELECT awesome FROM TABLE win WHERE cool > 0;
06:54 shinobicl_ i was recently working in a java project, using orm libraries to support my objects on tables, but this looks as soooo much boilerplate. Sadly i only have experience in relational databases but now it just seems awkward
06:54 shinobicl_ converting objects to relational and back
06:56 labster not really familiar with any of this stuff... are you trying to do something where RDBMS is not the right tool?
06:57 shinobicl_ no, just pointing out that you need to use huge frameworks (like Hibernate) to make the database understand your objects relations
06:58 labster is it just me, or is stage parse going much faster on rakudo/parrot?
06:59 shinobicl_ and also, there's clearly a trend for things like NoSql and Cassandra
07:04 labster Of course there is, but that's a matter of choosing the right tool.  If you want to be able to ask analytic questions, you need a RDBMS.  NoSQL and memcached are fine for things where you don't need those features, but only want to store some data.
07:05 [Coke] I have never worked on a project where ORM was a better answer than having someone who knew sql.
07:05 [Coke] (perhaps due to scale of the DB or the objects, Iunno)
07:05 diakopter ok, work on more toy projects! :)
07:07 * [Coke] tries to bisect RT#117957
07:08 * [Coke] also ponders a bisect helper for rakudo, but that is too many yaks deep.
07:08 PerlJam [Coke]: re swapping START and FIRST, the docs for START even say "on first ever execution"
07:10 [Coke] PerlJam: are you agreeing with me about the proposed switch or suggesting that a read of the docs will set everyone straight?
07:10 PerlJam [Coke]: agreeing.
07:11 labster [Coke]: I think FROGGS decided 117957 had something to do with my implementation of IO::Spec::Unix.tmpdir of all things.  Not 100% sure if it was this bug or another one.
07:11 PerlJam Why call something START that's then immediately described as happening "first", but then have something called FIRST that means something else.
07:11 woolfy left #perl6
07:14 sorear Warning: the behavior of "state" is defined in terms of START; if changing START, please explicitly make a decision about what happens to state so that we don't have to guess
07:16 labster anyway, I'm spectesting a workaround for what FROGGS_ found, spectesting now -- though the failure only seems to happen on feather.
07:16 sorear let me just say that "NoSQL" literally means that.  all the popular "NoSQL" databases are functionally equivalent to RDBMSes, minus the SQL parser and query optimizer
07:17 diakopter indexes aren't nearly as powerful too
07:17 diakopter though I guess you can lump that in with the optimizer
07:17 sorear if you look under the hood of say MySQL, you'll find that InnoDB's internal interface is a transactional key-value store
07:17 saxx11 joined #perl6
07:19 sorear "indexes" in an RDBMS are usually implemented as auxiliary tables with the indexed column as primary key.  the SQL layers automatically do transactional updates to keep the index updated
07:19 [Coke] labster: iospec merge is in the list of remaining candidates, I think.
07:19 diakopter sorear: that makes sense
07:20 berekuk joined #perl6
07:20 sorear disclaimer: the RDBMSes I have studied in the most detail are mysql/InnoDB and SQLite.  I don't really know how the big iron commercial SQLs work inside
07:22 labster The idea was that $*TMPDIR called .tmpdir, which references on %*ENV which -- although %PROCESS::ENV was already set in the terms.pm scope -- seems to cause issues on feather.
07:23 masak good morning, #perl6
07:23 diakopter o/
07:24 * masak feels very refreshed
07:24 diakopter \o/
07:24 masak it's amazing what a full night of sleep can do to a person :>
07:24 labster \\o
07:24 sorear good morning masaaaak
07:26 * masak twinkles happily
07:27 masak r: my @a=<a a a>;my @b=<b b b>;for (@a »,« @b) »,« @a -> $a, $b, $c {say "$a $b $c"}
07:27 camelia rakudo 0e206e: OUTPUT«Lists on both side of non-dwimmy hyperop of infix:<,> are not of the same length␤left: 6 elements, right: 3 elements␤  in sub hyper at src/gen/CORE.setting:14054␤  in sub hyper at src/gen/CORE.setting:14042␤  in block  at src/gen/CORE.setting:14033␤  in block  at /…
07:27 * masak adds that to https://rt.perl.org/rt3/Ti​cket/Display.html?id=77746
07:32 SamuraiJack_ joined #perl6
07:41 masak [Coke]: re https://rt.perl.org/rt3/Tic​ket/Display.html?id=118179 , I think that's a non-bug
07:41 masak FIRST {...}*      at loop initialization time, before any ENTER
07:41 masak if it runs befor any ENTER, it also runs before the assignment $a = 3
07:41 masak so it *sees* the lexical... it's just that the lexical hasn't been assigned to yet.
07:42 masak compare 'my $a = 3; BEGIN { say $a }'
07:42 labster 118179 is already rejected
07:43 masak oh! so it is.
07:44 masak sorry about the noise.
07:44 masak on an abstract level, I can agree that the semantics [Coke]++ wanted ought to be easy, too.
07:45 masak &
07:51 FROGGS_ labster: I have some more information for you...
07:51 FROGGS let me get the diff from feather
07:51 sergot hi! o/
07:51 lizmat joined #perl6
07:51 imIKARi joined #perl6
07:52 arlinius joined #perl6
07:53 FROGGS labster: https://gist.github.com/FR​OGGS/0d1a9f4967a2c4d49606
07:54 FROGGS labster: it breaks bool.t if you comment in line 10, 13 or 14
07:55 labster huh?
07:55 FROGGS yeah
07:56 lizmat PerlJam: FWIW, the way I understand START, is that START runs once per closure S04:1464, and has nothing to do with loops
07:57 IKARi__ joined #perl6
07:58 labster Wait, is it the comments that are causing the error, or the regexes?
07:58 FROGGS labster: good question...
07:58 FROGGS I'll comment in and remove the comments
08:00 labster This bug makes no sense to me.
08:02 lizmat r: sub a ( $a=Mu ) { say $a }; a  # not sure if I mentioned this before
08:02 camelia rakudo 0e206e: OUTPUT«Nominal type check failed for parameter '$a'; expected Any but got Mu instead␤  in sub a at /tmp/ifvoQEy7nb:1␤  in block  at /tmp/ifvoQEy7nb:1␤␤»
08:03 lizmat r: sub a ( $a=Mu ) { say $a }; a(Any)  # not sure if I mentioned this before
08:03 camelia rakudo 0e206e: OUTPUT«(Any)␤»
08:03 labster Yeah, isn't that the "default is Any" issue?
08:04 labster sub a ( $a) implies sub a (Any $a)
08:04 lizmat r: sub a ( Mu $a=Mu ) { say $a }; a  # not sure if I mentioned this before
08:04 camelia rakudo 0e206e: OUTPUT«(Mu)␤»
08:04 lizmat labster++
08:04 masak jnthn and I are running a tad late for day 2 of the conf. we'll be sneaking in at the back at around 10:20.
08:05 masak [Coke]: doesn't START do exactly what you wanted from FIRST?
08:06 berekuk joined #perl6
08:26 FROGGS labster: it's not the comment, I commented in line 10 in the diff and removed the comment and it fails again
08:39 denisboyun_ joined #perl6
08:42 domidumont joined #perl6
08:49 woolfy joined #perl6
08:50 moritz good morning everybody
08:50 FROGGS morning moritz
08:51 tadzik hello
08:51 FROGGS o/
08:52 moritz how's plpw going?
08:52 tadzik better than I thought
08:52 tadzik we had a little bit of technical difficulties, but almost nobody noticed :)
08:52 tadzik the power went down in the entire office. Including doors
08:53 sorear can your doors be opened manually or were you trapped?
08:53 sorear how screwed would you have been if the building had proceeded to catch fire? :/
08:53 tadzik worse. They were open. Both ways. No alarm
08:54 tadzik we sorted it out, but it still makes me shiver
08:54 sorear (where I live, I think most doors are required by law to be manually openable?)
08:54 sorear what kind of facility is it?
08:55 tadzik it's a big building with like 10 offices
08:55 tadzik they forgot to turn the power back on after the outage during the night
08:56 tadzik fortunately it's not a problem anymore :0
08:56 sorear you say "office", I imagine a room roughly 5m x 5m x 2.5m with exactly one door, one desk, and three chairs
08:56 tadzik hehe, no. It's an office space for few dozens of people
08:57 tadzik we now have ~40 PLPW atendees just in a ping-pong room
08:57 lizmat it's pretty ok, it sounds worse then it is!
08:57 dmol joined #perl6
08:57 tadzik yeah. Only I know how bad it was :D
08:58 markstacey joined #perl6
08:58 moritz well, a proper table tennis room is at least 14m x 7m
08:59 moritz not too bad for fitting in 40 people :-)
08:59 tadzik well, it's generally a leisure space
08:59 tadzik I'm bad at approximating, but it may be like, 8m*10m, or something
09:00 spider-mario joined #perl6
09:00 tadzik there will be pics :)
09:01 * sorear wonders if there is actually an international standard for ceiling heights
09:01 lizmat not internationally, I'm pretty sure the standards for Dutch offices are different from the ones in e.g. Portugal
09:01 moritz for table tennis courts there is :-)
09:03 sorear lizmat: interesting.  the hotel room I stayed in in Perl had an 8 foot +- 1% ceiling, which was...surprising
09:03 tadzik is it like 2,5 meters?
09:03 sorear it's the standard height in the US, which makes sense because it's a convenient round number here
09:03 lizmat Ah, yes, good old German building standards  :-)
09:03 sorear ...but why do they use it in Perl
09:04 lizmat because it is Germany ?  with generally larger people as well ?
09:04 tadzik *cough* USA *cough*
09:07 sorear 2,4384 m; I can only say the hotel room ceiling was 2,44(1)
09:07 labster FROGGS: try changing the regexes to s/// form or .subst form?  maybe put $path on line 15 as an explicit return value, .=subst on line 14 and see if that fixes part of the problem?
09:08 FROGGS labster: will do, but I already tried .subst(...
09:08 domidumont joined #perl6
09:09 berekuk joined #perl6
09:10 labster Yeah, I'm not on feather, and I'm only just guessing about what it could be.
09:11 * sorear wonders what the joke is about German building standards
09:11 labster If we have to, we can just drop $*TMPDIR -- it's not specced.
09:12 moritz please don't, panda uses it, iirc
09:13 dalek rakudo-debugger: 857a884 | jonathan++ | lib/Debugger/UI/CommandLine.pm:
09:13 dalek rakudo-debugger: Fix debugger on latest Rakudo.
09:13 dalek rakudo-debugger: review: https://github.com/jnthn/raku​do-debugger/commit/857a884250
09:14 lizmat sorear: German houses in general, are larger and better built then most other houses in Europe
09:15 lizmat fwiw, I don't think they do that because it's cheaper, but because they have to
09:18 FROGGS labster: $*TMPDIR is just too handy to be dropped
09:19 FROGGS labster: you can request an account on feather though
09:19 FROGGS brb
09:26 labster moritz: panda doesn't use it, but probably should.  We really need a proper File::Temp.
09:29 * masak is so happy to be at PLPW
09:30 moritz labster: oh, maybe it was LWP::Simple
09:30 FROGGS moritz: I think it was that, yes
09:31 masak "Perl Cannot Be Parsed: A Formal Proof" is on https://news.ycombinator.com/ right now. in other news, bumblebees cannot really fly, and if the Earth really revolved around its own axis like the physicists say, we'd all be thrown off by centripetal forces.
09:31 FROGGS it can easily changed to use IO::Spec.tmpdir, but I want to know why that actually fails on feather
09:35 labster So far as I can see, the only thing those three lines have in common that the others do not is anchors (^$) that aren't in square brackets.  Or not in an alternation.
09:35 masak r: for 1..10  { my $a = 3; START { say $a } }
09:35 camelia rakudo 0e206e: OUTPUT«3␤»
09:35 * masak adds this to #118179
09:36 FROGGS masak: but this is clear, no?   whatever  / 25 ; # / ; die "this dies!";
09:37 FROGGS because you'd know whatever's prototype without any code execution
09:37 masak FROGGS: my point isn't whether the proof is correct or not. (it is.)
09:37 moritz the correct statement would be "Perl cannot be parsed without running parts of it"
09:39 masak FROGGS: my point is that saying "Perl cannot be parsed" is silly, when people are writing and running bazillions of Perl programs every day.
09:39 masak if you're saying that, and the state of the world is what it is, then what you're saying needs to be revised.
09:39 sorear someone should point out that you can't parse lisp without running reader macros
09:40 sorear or parse C without header files
09:40 sorear sleep&
09:40 moritz or parse tex without running tex
09:40 labster whoa, it's way too late for me to read this kind of "proof"
09:41 labster I'm going to get me to a nullary^W bed.
09:41 masak a nullary bed sounds kinda comfy.
09:41 labster sleep &
09:41 sergot n: grammar A { token TOP { <HEAD>\s+Building\s+<name> }; token HEAD { \=\=\> }; token name {.*};}; my $s = "==> Building Foo::Bar\n==> Doing something with Foo::Bar"; say A.new.parse($s);
09:41 camelia niecza v24-51-g009f999: OUTPUT«「==> Building Foo::Bar␤==> Doing something with Foo::Bar」␤ HEAD => 「==>」␤ name => 「Foo::Bar␤==> Doing something with Foo::Bar」␤␤»
09:41 sergot r: grammar A { token TOP { <HEAD>\s+Building\s+<name> }; token HEAD { \=\=\> }; token name {.*};}; my $s = "==> Building Foo::Bar\n==> Doing something with Foo::Bar"; say A.new.parse($s);
09:41 camelia rakudo 0e206e: OUTPUT«「==> Building Foo::Bar␤==> Doing something with Foo::Bar」␤ HEAD => 「==>」␤ name => 「Foo::Bar␤==> Doing something with Foo::Bar」␤␤»
09:41 lizmat at least you won't be able to sink into a nullary bed
09:41 masak good night, labster. dream of parsing/Hamlet fiction.
09:42 sergot How can I gen only Foo::Bar into <name>?
09:43 moritz sergot: make token name more selective
09:43 moritz \S+ would be a decent start
09:43 sergot moritz: I was trying to add \n to TOP token after <name>, it doesnt work
09:44 sergot I want everything after Building to the first appearance of \n :)
09:45 moritz r:  r: grammar A { token TOP { <HEAD>\s+Building\s+<name> }; token HEAD { '==>' }; token name { \S+ }}; my $s = "==> Building Foo::Bar\nblah"; say A.new.parse($s)
09:45 camelia rakudo 0e206e: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Two terms in a rowâ�¤at /tmp/mcFYwyeG2s:1â�¤------> [32mr[33mâ��[31m: grammar A { token TOP { <HEAD>\s+Build[0mâ�¤    expecting any of:â�¤        argument listâ�¤        postfixâ�¤        infix stopperâ�¤        infix or meta-infixâ�¤        statem…
09:45 moritz r: grammar A { token TOP { <HEAD>\s+Building\s+<name> }; token HEAD { '==>' }; token name { \S+ }}; my $s = "==> Building Foo::Bar\nblah"; say A.new.parse($s)
09:45 camelia rakudo 0e206e: OUTPUT«「==> Building Foo::Bar」␤ HEAD => 「==>」␤ name => 「Foo::Bar」␤␤»
09:45 moritz sergot: like this?
09:45 sergot Oh, so what is \S+?
09:45 sergot moritz++ thanks!
09:45 sergot Why does it work so?
09:45 moritz \S is "anything that is not a whitespace
09:46 sergot Ok, it's clear now. Thanks :)
09:47 sergot moritz: one more question. If I want something after \n, Should I write in token TOP ...\n<something>... ?
09:47 sergot Is it "good"? :)
09:48 moritz well, you can do it with \n<something>
09:48 moritz or you can split the input into lines before putting the stuff into the grammar
09:48 bbkr_ joined #perl6
09:48 moritz which is often easier for line-based formats
09:49 sergot I'm trying to parse this: http://feather.perl6.nl/~sergot/modules/err1. To add more features to my modules list. :)
09:49 sergot Ok, thank You moritz. :)
09:49 moritz a grammar might be easier for that
09:50 moritz because the test output is actually a multi-line format
09:50 diakopter I'm looking for someone with moderate experience with CSS on mobile to do some urgent/critical volunteer work to help out yapc austin
09:50 sergot moritz: so, can I call parse with a(n) array/list of lines?
09:50 moritz sergot: no
09:50 moritz sergot: you'd have to call A.parse for each line
09:51 sergot moritz: Ok, I'll try so. thanks again. :)
09:52 diakopter if you have that experience and you're a fast worker, and are willing/able to contribute 15-20 hours total over the next 6 days, contact me please
09:59 diakopter anyone experienced in event/party planning want to sanity check my table layout? [please?]
10:01 PacoAir joined #perl6
10:03 arnsholt I'm don't really know much about table layout, but I can have a look
10:04 lizmat is this about who's sitting where ?   Or something else ?
10:05 diakopter positioning of tables on a floor plan
10:05 diakopter arnsholt: I sent it on gtalk
10:05 moritz IMHO that's best planned with paper cut proportinally to the table size :-)
10:05 lizmat don't put them in front of emergency exits, more than that I don't really have to offer
10:06 diakopter moritz: I did pixel-scaled onto an architect's scale drawing
10:06 diakopter with a grid
10:14 dalek nqp/rak-jvm-support: 113b736 | jonathan++ | src/vm/jvm/runtime/org/perl6/​nqp/sixmodel/reprs/NFA.java:
10:14 dalek nqp/rak-jvm-support: Implement NFA serialize/deserialize.
10:14 dalek nqp/rak-jvm-support: review: https://github.com/perl6/nqp/commit/113b736708
10:15 diakopter lizmat: who needs emergency exits when there are huge bay doors open to the air on a wall? :)
10:16 lizmat diakopter: I can think of some situations, but generally these would involve a large zombie attack
10:17 diakopter I like small zombie attacks better
10:17 moritz http://podcastle.org/2010/09/01/podcastle​-120-some-zombie-contingency-plans-fixed/
10:19 tgt joined #perl6
10:30 sqirrel joined #perl6
10:37 rindolf joined #perl6
10:44 lizmat r: my %h{Any}=Any=>1; say %h.exists(Any)
10:44 camelia rakudo 0e206e: OUTPUT«use of uninitialized value of type Any in string context  in block  at /tmp/NTFY5txVTK:1␤␤False␤»
10:44 lizmat n: my %h{Any}=Any=>1; say %h.exists(Any)
10:44 camelia niecza v24-51-g009f999: OUTPUT«[31m===[0mSORRY![31m​===[0mâ�¤â�¤Postconstraints, and shapes on variable declarators NYI at /tmp/5v0DNvzso8 line 1:â�¤------> [32mmy %h{Any}[33mâ��[31m=Any=>1; say %h.exists(Any)[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.…
10:44 lizmat r: my %h{Any}=Any=>1; say %h.exists(Any)  # expect True
10:44 camelia rakudo 0e206e: OUTPUT«use of uninitialized value of type Any in string context  in block  at /tmp/Diq3EfjBnq:1␤␤False␤»
10:44 * lizmat files rakudobug
10:45 masak lizmat++
10:50 kurahaupo joined #perl6
10:56 dalek rakudo/jvm-support: f4fe4dd | jonathan++ | src/vm/jvm/runtime/org/perl6/rakudo/Binder.java:
10:56 dalek rakudo/jvm-support: Implement type captures in signature binder.
10:56 dalek rakudo/jvm-support: review: https://github.com/rakudo/rakudo/commit/f4fe4dd8b3
11:00 woolfy joined #perl6
11:06 FROGGS .tell labster that the fail goes away if I comment out a single line of the test file >.<
11:06 yoleaux FROGGS: I'll pass your message to labster.
11:07 FROGGS I hate this kind of bug
11:23 espadrine joined #perl6
11:31 colomon Hmmm, MuEvent starting failing tests overnight.
11:32 FROGGS without any changes?
11:32 colomon dunno, my reports don't have any more details yet
11:32 colomon without any changes to MuEvent
11:33 FROGGS :/
11:33 colomon presumably the changes were to Rakudo
11:35 colomon huh, all the tests worked when I just ran them by hand in a different account
11:35 FROGGS it picks the same perl6 binary?
11:36 colomon no
11:36 colomon oh, right!
11:36 colomon duh
11:36 tadzik colomon: maybe they're just flapping
11:37 colomon tadzik: ?
11:37 colomon also passes using the same account and p6 binary.  :\
11:37 colomon tadzik: does MuEvent occasionally fail?
11:38 tadzik colomon: not to my knowledge
11:39 FROGGS r: multi term:<test> { 42 }; multi sub test($a) { $a + 1 }; say test 1;
11:39 camelia rakudo 0e206e: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Two terms in a rowâ�¤at /tmp/nHQ4QNIHo0:1â�¤------> [32mmulti sub test($a) { $a + 1 }; say test [33mâ��[31m1;[0mâ�¤    expecting any of:â�¤        postfixâ�¤        infix stopperâ�¤        infix or meta-infixâ�¤        statement endâ�¤        state…
11:39 colomon trying it with panda now
11:40 tadzik yep, me too
11:40 colomon that worked too
11:40 colomon so I've got no clue.
11:40 tadzik duh
11:41 FROGGS colomon: do you capture that test output somewhere? like backtraces?
11:42 colomon FROGGS: ooo, maybe, but not in a particularly orderly fashion
11:44 colomon well, I've got the stderr dump for the entire smoke run, but it's not sorted (as far as I know) or tagged by module name
11:45 FROGGS maybe you can search for the failing test file name?
11:46 colomon don't think so
11:53 colomon tadzik: I see a line         # don't waste time if it has already been installed once
11:53 colomon in emmentaler.
11:54 colomon does that mean smoker doesn't test modules which have already been installed?
11:54 tadzik well, it's not only about wasting time
11:54 colomon oh, is that already installed in this particular run?
11:54 tadzik yes precisely
11:55 tadzik you get a fresh state when you start emmentaler (or you should've
11:55 tadzik )
11:56 colomon I'm trying to figure out how to add the module being worked on to the STDERR output.  :)
12:01 mattp_ joined #perl6
12:03 a232346 joined #perl6
12:04 colomon rn: $*ERR.say: "Working on blah"
12:04 camelia rakudo 0e206e, niecza v24-51-g009f999: OUTPUT«Working on blah␤»
12:04 colomon okay, I've tried adding that to smoker.
12:05 colomon looks like it works.  :)
12:06 colomon (occurred to me I could always run it by hand and see what it did.)
12:06 a232346 left #perl6
12:15 mtk joined #perl6
12:17 FROGGS r: my $n = 3; say "a,a,a,a,a" ~~  / "a"**$n % [","] / # looks like I need that for perl5's sub calls :/
12:17 camelia rakudo 0e206e: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Quantifier quantifies nothingâ�¤at /tmp/s1P_ecAsTN:1â�¤------> [32mmy $n = 3; say "a,a,a,a,a" ~~  / "a"**[33mâ��[31m$n % [","] / # looks like I need that fo[0mâ�¤    expecting any of:â�¤        postfixâ�¤        infix stopperâ�¤        infix or…
12:17 moritz FROGGS: why?
12:18 moritz you shouldn't do a regex match per sub call; that's way too slow
12:18 FROGGS no, when parsing things like: @a = shift || 1
12:19 * moritz doesn't see the connection yet
12:19 FROGGS you need to know the prototype of shift in order to know that it can take zero or one arg
12:19 moritz oh
12:19 FROGGS and things like: @a = shift @b, 2, 3; where shift only eats the @b
12:20 FROGGS so I need to do: <EXPR>**$count % [','] in token <args> or so
12:20 FROGGS well, basically I need to do that for all possibilities of $count
12:24 masak r: say [+] ('a'..'z').pairs.map: { 1 + .key if .value ~~ any < m a t h > }
12:24 camelia rakudo 0e206e: OUTPUT«42␤»
12:25 masak on perl6-users, eiro asks us to make this shorter.
12:25 masak I think we can do that :>
12:25 FROGGS cool
12:26 FROGGS r: say [+] (0,'a'..'z').pairs.map: { .value ~~ any < m a t h > and .key }
12:26 camelia rakudo 0e206e: OUTPUT«42␤»
12:26 masak r: say [+] ('a'..'z').grep({ any <m a t h> })>>.ord X- ord('a')
12:26 camelia rakudo 0e206e: OUTPUT«325␤»
12:27 masak er :)
12:27 colomon close enough!
12:27 masak :P
12:27 * masak tries to figger out what he did wrong
12:27 masak r: say ('a'..'z').grep({ any <m a t h> })
12:27 camelia rakudo 0e206e: OUTPUT«a b c d e f g h i j k l m n o p q r s t u v w x y z␤»
12:27 moritz r: say [+] 'math'.comb>>.ord X- ord('a')
12:27 camelia rakudo 0e206e: OUTPUT«38␤»
12:27 masak r: say ('a'..'z').grep(any <m a t h>)
12:27 camelia rakudo 0e206e: OUTPUT«a h m t␤»
12:27 masak ah.
12:28 moritz r: say [+] 'math'.comb>>.ord X- ord('a') + 1
12:28 camelia rakudo 0e206e: OUTPUT«34␤»
12:28 colomon r: say [+] (0,'a'..'z').kv.map: { 1 + $^k of $^v ~~ any < m a t h > }
12:28 camelia rakudo 0e206e: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Two terms in a rowâ�¤at /tmp/tuSC0LXXa6:1â�¤------> [32msay [+] (0,'a'..'z').kv.map: { 1 + $^k [33mâ��[31mof $^v ~~ any < m a t h > }[0mâ�¤    expecting any of:â�¤        postfixâ�¤        infix stopperâ�¤        infix or meta-infixâ�¤        sta…
12:28 moritz r: say [+] 'math'.comb>>.ord X- ord('a') X+ 1
12:28 camelia rakudo 0e206e: OUTPUT«818␤»
12:28 colomon r: say [+] (0,'a'..'z').kv.map: { 1 + $^k if $^v ~~ any < m a t h > }
12:28 camelia rakudo 0e206e: OUTPUT«46␤»
12:28 lizmat r: sub a ( $a is copy ) { $a=5 }; a(1)
12:28 moritz r: say [+] 'math'.comb>>.ord X- (1+ord('a') )
12:28 camelia rakudo 0e206e: OUTPUT«Cannot modify an immutable value␤  in block  at /tmp/Dmc8P4X3n2:1␤␤»
12:28 camelia rakudo 0e206e: OUTPUT«34␤»
12:28 * lizmat submits rakudobug
12:28 colomon 46 ?!
12:28 masak r: say [+] 'math'.comb>>.ord X- (ord('a') - 1)
12:28 camelia rakudo 0e206e: OUTPUT«42␤»
12:29 colomon oh
12:29 moritz masak: right, that was what I was trying to do
12:29 colomon r: say [+] ('a'..'z').kv.map: { 1 + $^k if $^v ~~ any < m a t h > }
12:29 camelia rakudo 0e206e: OUTPUT«42␤»
12:29 moritz r: say <m a t h>».org X- ('a'.ord - 1)
12:29 camelia rakudo 0e206e: OUTPUT«No such method 'org' for invocant of type 'Str'␤  in method dispatch:<hyper> at src/gen/CORE.setting:1097␤  in block  at /tmp/B0cksZLtJo:1␤␤»
12:29 moritz r: say <m a t h>».ord X- ('a'.ord - 1)
12:29 camelia rakudo 0e206e: OUTPUT«13 1 20 8␤»
12:30 moritz r: say [+] <m a t h>».ord X- ('a'.ord - 1)
12:30 camelia rakudo 0e206e: OUTPUT«42␤»
12:30 FROGGS cool
12:30 masak r: sub v { ord($^l) - ord('a') + 1 }; say [+] <m a t h>.map(&v)
12:30 camelia rakudo 0e206e: OUTPUT«42␤»
12:31 masak oh, but moritz++' is much shorter :)
12:31 * moritz answers on p6u
12:31 masak moritz++
12:32 prevost joined #perl6
12:32 moritz r: say [+] <m a t h>».ord X- 'a'.pred.ord
12:32 camelia rakudo 0e206e: OUTPUT«Decrement out of range␤  in method <anon> at src/gen/CORE.setting:10070␤  in any  at src/gen/Metamodel.nqp:2504␤  in any find_method_fallback at src/gen/Metamodel.nqp:2492␤  in any find_method at src/gen/Metamodel.nqp:939␤  in block  at /tmp/rKFRJbJpBF:1␤␤»…
12:33 moritz r: say 'a'.ord.pred.chr
12:33 camelia rakudo 0e206e: OUTPUT«`␤»
12:33 moritz r: say [+] <m a t h>».ord X- '`'.ord
12:33 camelia rakudo 0e206e: OUTPUT«42␤»
12:33 * moritz would like to get rid of that -1, but doesn't have any really good ideas
12:34 moritz r: say :36<a>
12:34 camelia rakudo 0e206e: OUTPUT«10␤»
12:34 moritz r: say ('a'..'z').grep(any 'math'.comb)
12:34 camelia rakudo 0e206e: OUTPUT«a h m t␤»
12:35 colomon rn: say (ord('a') - 1).char
12:35 * colomon can never remember what the opposite of ord is
12:35 camelia rakudo 0e206e: OUTPUT«No such method 'char' for invocant of type 'Int'␤  in block  at /tmp/7W5d4b6eHw:1␤␤»
12:35 camelia ..niecza v24-51-g009f999: OUTPUT«Unhandled exception: Unable to resolve method char in type Int␤  at /tmp/dKGSe7uJ7S line 1 (mainline @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4331 (ANON @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4332 (module-CORE @ 582) ␤  at /home/p…
12:35 moritz r: say [+] ('a'..'z').grep(any 'math'.comb).map *.ord - 1
12:35 camelia rakudo 0e206e: OUTPUT«Not enough positional parameters passed; got 1 but expected 2␤  in method map at src/gen/CORE.setting:1396␤  in block  at /tmp/a9ceOOf4kx:1␤␤»
12:35 colomon rn: say 'a'.ord, '`'.ord
12:35 camelia rakudo 0e206e, niecza v24-51-g009f999: OUTPUT«9796␤»
12:35 moritz colomon: chr
12:35 moritz r: say [+] ('a'..'z').grep(any 'math'.comb).map: *.ord - 1
12:35 camelia rakudo 0e206e: OUTPUT«422␤»
12:36 * colomon sees masak already did this...
12:36 moritz r: say [+] ('a'..'z').grep(any 'math'.comb).map: *.ord - ord('a') - 1
12:36 camelia rakudo 0e206e: OUTPUT«34␤»
12:36 moritz r: say [+] ('a'..'z').grep(any 'math'.comb).map: *.ord - ord('a') + 1
12:36 camelia rakudo 0e206e: OUTPUT«42␤»
12:40 colomon r: say [+] ('a'..'z').grep(any 'math'.comb).map: *.ord - ord('`')
12:40 camelia rakudo 0e206e: OUTPUT«42␤»
12:45 saxx11 joined #perl6
12:49 felher r: say [+] <m a t h>».ord»-»('a'.ord -1) #me finds that a bit clearer than X- with one element on the left, but thats probably just taste.
12:49 camelia rakudo 0e206e: OUTPUT«42␤»
12:50 felher And its only shorter, if you use » and not >>. And even then only by one character :)
12:50 moritz but only because you omit the whitespaces
12:50 LoRe joined #perl6
12:51 moritz you can also drop the blank after Z-
12:52 felher moritz: yeah, right. It's really more of a taste thing. :)
12:52 saxx12 joined #perl6
12:52 FROGGS r: say [+] <m a t h>».ord»-»96
12:52 camelia rakudo 0e206e: OUTPUT«42␤»
12:52 FROGGS only "r: 42" would be shorter I guess :P
12:52 felher :D
13:08 FROGGS btw, <m a s a k> is closest to truthness from what I have tested
13:18 Chillance joined #perl6
13:23 colomon masak: pull request sent for HTML::Template.  ;)
13:33 denisboyun_ joined #perl6
13:49 fhelmberger joined #perl6
13:50 dagurval joined #perl6
13:57 fhelmberger joined #perl6
14:06 FROGGS pmichaud / jnthn: if you get the chance, please have a look at https://gist.github.com/FROGGS/5652897
14:21 snearch joined #perl6
14:28 arlinius joined #perl6
14:34 saxx11 joined #perl6
14:39 kaare__ joined #perl6
14:40 sergot joined #perl6
14:43 konundra joined #perl6
14:45 pmichaud FROGGS: what's the purpose of the 5652897 patch?
14:48 pmichaud oh, I see it at the top.
14:48 pmichaud my first response would be to veto the patch, unless there's a compelling use case for it.
14:49 pmichaud NQP is supposed to be a lightweight version of Perl 6.
14:49 pmichaud also, switching  "has int $!min"  to be "has $min" is likely a no-go.
14:50 pmichaud er, "has $!min"
14:51 FROGGS okay, so this should go into rakudo?
14:51 sergot Is there any possibility to get the value from to a variable? https://github.com/rakudo/rakudo/​blob/nom/src/core/control.pm#L187
14:52 pmichaud FROGGS: that's where I'd expect it to be implemented first, yes.
14:52 sergot I wrote IO::Capture::Simple, and it doesnt capture it.
14:52 FROGGS pmichaud: thanks, will try
14:52 sergot github.com/sergot/IO-Capture-Simple
14:53 sergot Any ideas? ;)
14:53 pmichaud sergot: I'm having trouble parsing your question
14:53 pmichaud what are you wanting to get?
14:53 sergot pmichaud: https://github.com/sergot/​IO-Capture-Simple/issues/2
14:54 FROGGS r: my $output_of_ls = shell("ls"); # . .. .git ...
14:54 camelia rakudo 0e206e: OUTPUT«shell is disallowed in restricted setting␤  in sub restricted at src/RESTRICTED.setting:2␤  in sub shell at src/RESTRICTED.setting:7␤  in block  at /tmp/4eUqOovEK1:1␤␤»
14:54 sergot pmichaud: I'm sorry for my weird language. :)
14:54 pmichaud I don't think the shell() function is supposed to return the output.
14:56 kivutar joined #perl6
14:56 sergot pmichaud: yes, but my module gets value from std{in,out,err} to variables. But it cant do this with e.g. shell();
14:57 pmichaud maybe qx or qqx then?
14:57 sergot Am I understandable? :) Im a little bit in hurry.
14:58 sergot The problem is make my module to capture that.
14:58 sergot Ok, have to go. I'll think about it. :)
14:58 pmichaud I don't think shell/run will be the commands you need there, then.
14:58 pmichaud they aren't designed to capture the output.
14:59 sergot pmichaud: My module already captures $*IN $*OUT $*ERR.
15:00 sergot it has problem with capturing shell output.
15:00 sergot o/
15:00 sergot thanks pmichaud++
15:20 xinming joined #perl6
15:26 splitcells joined #perl6
15:37 splitcells How do I 'use' an other script that is in the same dir as the current script?
15:37 moritz you don't 'use' scripts, you 'use' modules
15:37 moritz use lib '.'; use MyModule;
15:48 splitcells on windows use lib '.'; ist not working
15:49 moritz how is it not working?
15:49 splitcells there is not . location
15:49 woolfy left #perl6
15:49 splitcells I am using 'C:/...'
15:49 timotimo_ there is always .
15:49 splitcells not on my machine
15:49 moritz . refers to the current directory, even on windows
15:50 splitcells no
15:50 moritz it may be called MyFolder, but if it's the current working directory, you can still refer to it as '.'
15:50 splitcells I am using windows 8
15:50 moritz if not, your windows is broken, and you should get a refund from Microsoft
15:50 splitcells this does not work
15:50 splitcells use v6; use lib '.'; use Cells;
15:51 moritz and what's the error message?
15:51 splitcells ===SORRY!=== Could not find Cells in any of: ., c:/rakudo/lib/parrot/4.10.0​/languages/perl6/site/lib, c:/rakudo/lib/parrot/4.10.0/​languages/perl6/vendor/lib, c:/rakudo/lib/parrot/4.10.0/languages/perl6/lib, C:\Users\splitcells/.perl6/2013.02.1/lib <<< Process finished.
15:51 moritz and is there a Cells.pm or Cells.pm6 in the current working directory?
15:52 splitcells yes
15:52 splitcells this works
15:52 splitcells use v6; use lib "X:/..."; use Cells;
15:53 timotimo_ do you invoke the script like perl6 Foo.p6 or perl6 X:/.../?
15:53 moritz splitcells: what does  use v6; say dir('.'); say cwd;  produce?
15:53 splitcells on notepad++
15:53 splitcells C:\rakudo\bin\perl6.exe "$(FULL_CURRENT_PATH)"
15:53 splitcells C:\rakudo\bin\perl6.exe "X:\..."
15:54 moritz splitcells: then the current working directory is whatever notepad++'s current working directory is
15:54 moritz splitcells: and not the directory where the script is
15:54 moritz oh
15:54 moritz my fault
15:54 moritz you even said
15:54 splitcells i have to launch it via cmd than?
15:54 moritz in the same dir as current script
15:55 moritz r: say cwd.path.directory
15:55 camelia rakudo 0e206e: OUTPUT«/home␤»
15:55 splitcells perl5 did not have such problem^^
15:55 moritz splitcells: well, with p5 you have to use FindBin
15:55 splitcells no
15:55 splitcells notepad++ : perl "$(FULL_CURRENT_PATH)"
15:55 moritz p5 doesn't automaticially search for modules in the same path as the script
15:56 moritz maybe notepad++ does some magic; no idea
15:56 moritz anyway
15:56 moritz try
15:56 moritz use lib $*PROGRAM_NAME.path.directory
15:56 moritz r: say $*PROGRAM_NAME.path.directory
15:56 camelia rakudo 0e206e: OUTPUT«/tmp␤»
15:57 moritz r: use lib $*PROGRAM_NAME.path.directory; say @*INC[0]
15:57 camelia rakudo 0e206e: OUTPUT«/tmp␤»
15:58 splitcells output .
15:59 moritz splitcells: what does  say $PROGRAM_NAME  report?
16:00 splitcells .
16:00 * moritz doesn't believe it
16:00 splitcells that's no joke -.-
16:00 splitcells use v6; use lib $*PROGRAM_NAME.path.directory; say $*PROGRAM_NAME.path.directory
16:00 moritz and now I was just asking about $*PROGRAM_NAME, not the whole thing
16:01 splitcells u mean only use v6; say $*PROGRAM_NAME.path.directory?
16:01 moritz no
16:01 moritz I mean
16:01 moritz use v6; say $*PROGRAM_NAME;
16:01 splitcells X:\Semac\AlgorithmPrograms\Cells\test.pl6
16:02 moritz ok, next
16:02 moritz use v6; say $*PROGRAM_NAME.path.perl
16:02 splitcells IO::Path.new(basename => "X:\\Semac\\AlgorithmPrograms\\Cells\\test.pl6", directory => ".")
16:03 moritz oh
16:03 moritz splitcells: what version of rakudo are you using?
16:03 splitcells rakudo-star-2013.02.msi6
16:03 moritz that explains why .directory doesn't work yet :(
16:03 ironcame1 joined #perl6
16:04 Gruber joined #perl6
16:04 splitcells is there a newer one?
16:04 moritz there's always a newer rakudo if you build from source
16:04 moritz no idea if there's a newer rakudo star msi
16:05 lee__ joined #perl6
16:05 moritz and I hope there'll be a new one in a few days
16:05 splitcells ty
16:05 * moritz feels he wasn't a great help :(
16:06 splitcells works on cmd
16:07 moritz at least something :-)
16:07 splitcells even without use lib '.';
16:09 Bucciarati_ joined #perl6
16:10 sergot_ joined #perl6
16:10 moritz that's midly surprising for me
16:11 lustlife joined #perl6
16:27 yoleaux joined #perl6
16:29 dpk joined #perl6
16:30 s1n joined #perl6
16:32 yves_ joined #perl6
16:32 crazedpsyc joined #perl6
16:50 colomon joined #perl6
16:51 colomon do we know who gam on github is?  I just fixed his Test-Junkie module, and issued a pull request.
17:00 rindolf joined #perl6
17:08 fhelmberger joined #perl6
17:12 * moritz doesn't know
17:14 lizmat joined #perl6
17:16 woolfy joined #perl6
17:20 moritz in the current star distro-in-git, t/custom-headers-and-content.t and t/get-unsized.t from LWP::Simple fail
17:21 moritz Bailador t/03-response-content.t also fails
17:22 grondilu joined #perl6
17:22 * moritz tries to update all the modules
17:28 domidumont joined #perl6
17:54 Pleiades` joined #perl6
17:56 timotimo_ jnthn++ # fixing the debugger in time for a new star release
18:05 rindolf joined #perl6
18:07 FROGGS moritz: I thought I have fixed that t/get-unsized.t fail
18:11 FROGGS I guess I leave that **$n up to somebody else
18:12 FROGGS making that work in rakudo doesnt help me with my current problem in v5
18:14 FROGGS if I only could translate (@$$$;+) directly to tokens...
18:17 kurahaupo joined #perl6
18:18 berekuk joined #perl6
18:19 rindolf joined #perl6
18:23 FROGGS hmmm, I guess I can
18:27 Juerd [Coke]: No
18:31 SamuraiJack_ joined #perl6
18:52 rindolf joined #perl6
18:56 denisboyun_ joined #perl6
18:58 timotimo_ huh. try require Text::Levenshtein <&distance>; distance("foo", "bar");  --> Method 'Any' not found for invocant of class 'Parcel'  ?!?!?
18:59 FROGGS can you paste with --ll-exception?
18:59 timotimo_ sure
19:00 timotimo_ https://gist.github.com/timo/8b392e697679ec3696e3 - there we go
19:00 timotimo_ now i've got to run
19:10 moritz $ panda install MIME::Base64
19:10 moritz (Mu)
19:10 moritz No such method 'get-project' for invocant of type 'Any'
19:10 * moritz sadpanda
19:11 moritz that's a freshly bootstrapped panda on rakudo 2013.05
19:11 FROGGS timotimo_: damnit, that 'try' is busting it
19:17 PacoAir joined #perl6
19:29 ilbot_nextgen joined #perl6
19:30 ilbot_nextgen joined #perl6
19:31 ilbot_nextgen joined #perl6
19:51 itz__ moritz: I installed panda/rakudo 2013.05 from git yesterday and it works for me
19:57 timotimo_ maybe that's what it does before you've done the first update?
20:00 colomon joined #perl6
20:02 timotimo_ but that shouldn't be.
20:08 moritz $ panda update
20:08 moritz No such method 'update' for invocant of type 'Any'
20:10 timotimo_ can you see in the panda binary what the value of the ecosystem is after it gets created?
20:10 timotimo_ that's really strange
20:11 moritz it's Any
20:11 moritz that's what the error message tells me
20:12 moritz NARF
20:12 timotimo but how? it gets created with Panda::Ecosystem.new and the new method isn't custom
20:12 moritz I accidentally used an old version of panda, not the one I just installed
20:12 moritz cruft--
20:12 timotimo m)
20:12 moritz moritz--
20:13 timotimo that's quite all right
20:13 * moritz is just relieved that he is broken, and not panda
20:20 sorear that's not good.  we can fix panda, but who can fix moritz?
20:22 Khisanth joined #perl6
20:26 moritz sorear: sleep
20:27 labster good postnoon, #perl6
20:27 yoleaux 11:06Z <FROGGS> labster: that the fail goes away if I comment out a single line of the test file >.<
20:27 moritz r: say 42
20:27 camelia rakudo 0e206e: OUTPUT«42␤»
20:28 sorear o/ labster
20:28 labster \o
20:32 labster FROGGS: Yeah, I have no idea about that failure line -- it seems like there's nothing in common between the two parts of code that are failing.
20:33 FROGGS labster: true
20:33 FROGGS there is just something that messes up the GC or so
20:42 timotimo r: say "Foo Bar Baz" ~~ m:P5/(...)(?:\ (...))+/
20:42 camelia rakudo 0e206e: OUTPUT«「Foo Bar Baz」␤ 0 => 「Foo」␤ 1 => 「Baz」␤␤»
20:42 timotimo is this not supposed to match Bar and Baz?
20:43 timotimo perl5: $_ = "Foo Bar Baz"; /(...)(?:\ (...))+/; print $1;print $2; print $3;
20:43 timotimo eval: $_ = "Foo Bar Baz"; /(...)(?:\ (...))+/; print $1;print $2; print $3;
20:44 timotimo p5eval: $_ = "Foo Bar Baz"; /(...)(?:\ (...))+/; print $1;print $2; print $3;
20:44 p5eval timotimo: FooBaz1
20:44 timotimo oh, okay
20:44 moritz timotimo: in p5, a quantified capture only records the last match
20:44 timotimo pcre is so great /s     :P
20:46 tgt joined #perl6
20:48 timotimo what's the way to record each of the results?
20:48 moritz (??{ push @array, $^N }) # or so
20:49 moritz (inside the quantifier, but after the capture
20:49 moritz )
20:53 FROGGS r: my $p = '$'; grammar G { rule TOP { :my $s = $p; :my $n = ''; :my $i = 0; [ <?{ $n = $s.substr($i++, 1) }> <arg($n)> ]+ % ',' }; token arg($n) { \d+ } }; say G.parse( "1,2,3" )
20:53 camelia rakudo 0e206e: OUTPUT«「1」␤ arg => 「1」␤␤»
20:53 FROGGS r: my $p = '$$'; grammar G { rule TOP { :my $s = $p; :my $n = ''; :my $i = 0; [ <?{ $n = $s.substr($i++, 1) }> <arg($n)> ]+ % ',' }; token arg($n) { \d+ } }; say G.parse( "1,2,3" )
20:53 camelia rakudo 0e206e: OUTPUT«「1,2」␤ arg => 「1」␤ arg => 「2」␤␤»
20:53 FROGGS :o)
20:57 labster FROGGS: try binding in terms.pm?  $TMPDIR := IO::Spec.tmpdir
20:58 jnthn FROGGS, pmichaud: I don't think "just shove an AST into min/max" will work. It expects integer literals there, not a QAST tree. The code gen actually needs to know the exact values at present. So does the NFA stuff. I suspect it needs some kind of smarter thing, maybe a different rxtype so we don't put all the complexity into quant itself.
20:58 labster either that, or tell me who I have to bug to get an account on feather :o)
21:00 FROGGS labster: moritz and Juerd can give you an account I think
21:01 FROGGS jnthn: thanks, will keep that in mind if/when I will work at that again
21:01 timotimo moritz: yikes, that's apparently not something you're supposed to be able to do nicely in perl5 :|
21:07 btyler joined #perl6
21:33 snoopy joined #perl6
21:41 tgt joined #perl6
21:48 tgt joined #perl6
21:53 bruges joined #perl6
21:53 kurahaupo joined #perl6
21:55 kurahaupo joined #perl6
22:04 tadzik *whew*
22:04 tadzik PLPW is over. 'twas awesome
22:05 sergot_ o/
22:06 sergot_ tadzik: thank you again for plpw
22:06 sergot_ !
22:07 tadzik you're welcome. Thank you!
22:07 timotimo tadzik: \o/
22:07 tadzik You made it. I just herded cats :P
22:08 tadzik it was fun :)
22:08 sergot_ :)
22:09 sergot It must be done in next year too. :)
22:09 sergot have to * :)
22:09 timotimo have people been recording videos of the talks?
22:10 tadzik I thought it's settled, that Poznań.pm organizes YAPC::Poland next year :)
22:10 sergot timotimo: yes
22:10 tadzik timotimo: yes, everything is recorded
22:10 tadzik timotimo: I'll try to get the video asap, cut it into parts and put it all on YT
22:10 timotimo \o/
22:10 tadzik this'll require some tuits and computing power
22:10 timotimo that's great
22:11 tadzik yeah, and it just happened
22:11 sergot tadzik: do you know when and where will be available photos and videos from the whole PLPW (made by Wendy and others?)
22:11 sergot Wendy, right?
22:11 tadzik one of the attendees asked "can I record this?" "yeah, sure, go ahead"
22:11 tadzik sergot: oh, I have those too
22:11 tadzik but it's like 7 gigabytes
22:11 sergot No way, really?
22:11 tadzik aye
22:11 tadzik full-frame Nikon D800 :D
22:12 sergot So... Could you send me this via postmail?
22:12 tadzik I'll run some perl script to compress those a bit and put it somewhere for people to look at
22:12 tadzik again, when I get some tuits
22:12 sergot Great. :)
22:12 sergot Looking forward. :)
22:13 sorear postmail, eh
22:13 sorear o/ sergot
22:13 tadzik I'm looking through them. They're good :)
22:14 tadzik jest has some pictures too, I believe
22:15 sergot sorear: o/
22:15 sergot sorear: did I wrote something not correctly? :P
22:15 sergot write*
22:18 sorear I just don't think I've heard postmail before.  It's cute
22:19 sergot sorear: :)
22:19 sergot It should be post or mail, shouldn't it? :)
22:19 tadzik snail-mail, maybe :)
22:19 sergot ok, thanks. :)
22:27 chrisdev joined #perl6
22:34 lizmat tadzik++ + pjotr++ + pjotr++ for PLPW!
22:43 adu joined #perl6
22:45 dalek perl6-bench: 2ef95b7 | (Geoffrey Broadwell)++ | TODO:
22:45 dalek perl6-bench: Add a TODO sublist for known-slow tasks; add one item there for the qq:to// unindent
22:45 dalek perl6-bench: review: https://github.com/japhb/pe​rl6-bench/commit/2ef95b72cb
22:45 dalek perl6-bench: a44ac08 | (Geoffrey Broadwell)++ | bench:
22:45 dalek perl6-bench: Chase a Rakudo behavior change
22:45 dalek perl6-bench: review: https://github.com/japhb/pe​rl6-bench/commit/a44ac088b8
22:47 sergot lizmat: o/
22:48 lizmat sergot: /o
22:55 masak o/
22:55 sorear ō
22:56 * lizmat hopes masak is home safe
22:56 masak fsvo "home", yes :)
22:56 masak (teaching tomorrow. I'm in a hotel room, feeling a bit like a character in a William Gibson novel)
22:57 lizmat the life of a globe trotter  :-)
22:57 lizmat see you soon in Austin, TX !
22:57 masak :>
22:57 masak my point exactly.
22:58 masak I bet if the sky weren't night-black, it'd have the color of television, tuned to a dead channel.
23:07 sorear I see what you did there :P
23:08 masak :)
23:12 ajr joined #perl6
23:17 crab2313 joined #perl6
23:22 masak 'night, #perl6
23:25 census joined #perl6
23:28 PacoAir joined #perl6
23:29 timotimo jnthn: i'm a bit bummed that apparently i can't work on $/ in the debugger after a match came through
23:29 timotimo is that right?
23:39 berekuk joined #perl6
23:42 btyler joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo