Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2013-03-28

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:14 BenGoldberg left #perl6
00:14 cosimo joined #perl6
00:20 dalek v5: b2e16c8 | (Tobias Leich)++ | / (3 files):
00:20 dalek v5: regexes are working now
00:20 dalek v5: review: https://github.com/rakudo-p5/v5/commit/b2e16c8756
00:21 timotimo_ just like that o_O
00:23 kst joined #perl6
00:23 FROGGS *g*
00:23 FROGGS gnight
00:31 masak 'night, #perl6
00:32 xinming joined #perl6
00:41 denisboyun_ joined #perl6
00:46 kurahaupo FROGGS: was it intentional to switch leading '0' from decimal to octal?
00:53 sorear kurahaupo: leading 0 is octal in perl 5
00:55 raiph joined #perl6
01:04 labster joined #perl6
01:10 kurahaupo sorear: I'm well aware of that; ditto for C & C++. I'd also question removing the warning/recommendation to use '0o' instead.
01:11 anuby joined #perl6
01:15 Targen joined #perl6
01:19 sorear kurahaupo: FROGGS is modifying Rakudo's parser into a perl 5 parser, so using p5 syntax rules is kind of a given, no?
01:20 hypolin joined #perl6
01:42 stevan_ joined #perl6
01:56 kurahaupo joined #perl6
02:10 thou joined #perl6
02:22 daniel-s_ joined #perl6
02:31 thou joined #perl6
02:36 sqirrel_ joined #perl6
02:36 FROGGS_ joined #perl6
02:38 thou_ joined #perl6
02:38 cognominal joined #perl6
02:38 toddr_ joined #perl6
02:41 kurahaupo sorear++ thanks for the explanation
02:45 am0c joined #perl6
02:46 lustlife joined #perl6
02:55 fgomez joined #perl6
02:58 thou joined #perl6
03:01 c1sung joined #perl6
03:05 thou joined #perl6
03:08 orafu joined #perl6
03:16 preflex_ joined #perl6
03:25 thou joined #perl6
03:27 colomon joined #perl6
03:46 thou_ joined #perl6
03:49 bruges joined #perl6
04:06 thou joined #perl6
04:28 thou joined #perl6
04:34 thou joined #perl6
04:47 thou_ joined #perl6
04:49 rindolf joined #perl6
04:54 kurahaupo joined #perl6
05:03 dwoldrich joined #perl6
05:07 thou joined #perl6
05:21 kurahaupo left #perl6
05:25 xinming joined #perl6
05:25 denisboyun_ joined #perl6
05:28 thou_ joined #perl6
05:28 SamuraiJack joined #perl6
05:31 kaleem joined #perl6
05:35 kaare_ joined #perl6
05:40 SamuraiJack joined #perl6
05:43 pyjery joined #perl6
05:46 dmol joined #perl6
05:48 thou joined #perl6
05:49 odoacre joined #perl6
05:59 dalek rakudo/extend-dump: 0cfabe7 | (Geoffrey Broadwell)++ | src/core/ (2 files):
05:59 dalek rakudo/extend-dump: Remove/clean up debugging output a bit
05:59 dalek rakudo/extend-dump: review: https://github.com/rakudo/rakudo/commit/0cfabe7dfb
05:59 dalek rakudo/extend-dump: 334a90b | (Geoffrey Broadwell)++ | src/core/ (4 files):
05:59 dalek rakudo/extend-dump: Improve DUMP output a bit for Bool, Parcel, Range, and a few misc cases
05:59 dalek rakudo/extend-dump: review: https://github.com/rakudo/rakudo/commit/334a90b5ff
05:59 dalek rakudo/extend-dump: b91487f | (Geoffrey Broadwell)++ | src/core/ (2 files):
05:59 dalek rakudo/extend-dump: Minor whitespace cleanups
05:59 dalek rakudo/extend-dump: review: https://github.com/rakudo/rakudo/commit/b91487ff37
05:59 dalek rakudo/extend-dump: eddcaa5 | (Geoffrey Broadwell)++ | src/core/Mu.pm:
05:59 dalek rakudo/extend-dump: Add comment with desugared version of $attr.get_value(self), which jnthn believes might fix an edge case later
05:59 dalek rakudo/extend-dump: review: https://github.com/rakudo/rakudo/commit/eddcaa5ea9
05:59 dalek rakudo/extend-dump: ac1c9b4 | (Geoffrey Broadwell)++ | t/02-rakudo/dump.t:
06:00 dalek rakudo/extend-dump: First attempt at tests for DUMP, starting with simple literals
06:00 dalek rakudo/extend-dump: review: https://github.com/rakudo/rakudo/commit/ac1c9b4cda
06:00 dalek rakudo/extend-dump: b15d183 | (Geoffrey Broadwell)++ | src/core/Mu.pm:
06:00 dalek rakudo/extend-dump: Comment out debug output in DUMP routines in Mu.pm
06:00 dalek rakudo/extend-dump: review: https://github.com/rakudo/rakudo/commit/b15d18312c
06:00 dalek rakudo/extend-dump: d3e1010 | (Geoffrey Broadwell)++ | src/core/Mu.pm:
06:00 dalek rakudo/extend-dump: Begin implementing indented DUMP output; Mu.pm changes only for first pass
06:00 dalek rakudo/extend-dump: review: https://github.com/rakudo/rakudo/commit/d3e1010c4e
06:08 thou_ joined #perl6
06:09 arlinius joined #perl6
06:11 kaleem joined #perl6
06:29 thou joined #perl6
06:31 kaare_ joined #perl6
06:39 pupoque joined #perl6
06:49 thou_ joined #perl6
06:53 kaare_ joined #perl6
06:54 thou joined #perl6
06:59 thou_ joined #perl6
07:07 daxim joined #perl6
07:12 thou joined #perl6
07:16 diakopter joined #perl6
07:32 thou_ joined #perl6
07:36 daniel-s_ joined #perl6
07:41 thou joined #perl6
07:53 domidumont joined #perl6
07:57 daniel-s_ joined #perl6
07:58 kaare__ joined #perl6
08:00 labster joined #perl6
08:01 thou_ joined #perl6
08:06 domidumont joined #perl6
08:08 kaare__ joined #perl6
08:10 thou joined #perl6
08:12 rindolf joined #perl6
08:14 thou joined #perl6
08:19 rindolf joined #perl6
08:20 drbean joined #perl6
08:25 rindolf joined #perl6
08:32 woolfy joined #perl6
08:33 thou joined #perl6
08:34 daniel-s_ joined #perl6
08:45 Heather joined #perl6
08:45 Heather Hello #Perl6 !
08:45 moritz \o Heather
08:45 pyjery left #perl6
08:46 * Heather is writing own Code Style Guide to fit own Code Style ...
08:46 labster Hi Heather
08:46 Heather labster hi
08:46 labster I read your blog, Heather++
08:46 Heather labster huh, cool :)
08:46 thou joined #perl6
08:46 labster It makes me want to write my own blog, just to show that Perl 6 has users.
08:46 Heather labster usually I lazy to write stuff to blog :)
08:47 labster me too
08:47 moritz labster: please do
08:47 Heather take a look https://github.com/Heather/Beauty-and-Freedom !!! It's brand new CODE STYLE ! It need users
08:47 Heather *needs
08:48 moritz convincing people to adapt a new code style is even harder than convincing them to adapt a new programming language :-)
08:49 Heather moritz people must understand that Beauty and Freedom can be better then standard :P that's my point
08:49 Heather *than
08:49 * Heather always use 'then' wrong
08:49 labster I'm not even sure what style I'm using :/
08:50 Heather labster then use Beauty and Freedom :D it will fit anything >_<
08:50 Heather - If you're not agreed with some terms or don't like some Code Style you're Free to use Custom style you like
08:51 Heather - If you desired to use Beauty-and-Freedom Style Guide you can also mix it with any other Style Guides you like
08:51 Heather ...
08:51 Heather \o/
08:51 labster moritz: I will write something, once I get this patch to File::Spec out
08:57 kivutar joined #perl6
08:57 thou joined #perl6
09:00 labster I just realized I have the first working version of IO::Path.filesystem (though I call it IO::Path::More.volume).
09:01 moritz labster: that might be an extension that should go into core
09:01 Heather labster btw if you liked my blogpost feel free to G+ and Twit it :D
09:01 labster It needs File::Spec to work, which means that at least a portion of File::Spec needs to go into core.  And... it needs work.
09:02 moritz labster: when I wrote IO::Path I was aware that at some point, we need volumes, but since I don't have any windows system to test, I left it out for the first version
09:02 FROGGS_ labster: do you have a solution for case-insensitivity (insanity?) testing?
09:05 labster FROGGS_: What I submitted should work, though it's not the most elegant thing.  I keep thinking I should move it to Unix.pm, but there are issues with that, mainly with everything subclassing from Unix.
09:06 thou joined #perl6
09:07 labster moritz: Did you write the spec for IO::Path, or just the implementation?  Because if it's the spec, I have some issues to discuss.
09:07 moritz labster: both
09:08 lizmat morning #perl6
09:08 Heather lizmat hey
09:08 moritz \o lizmat
09:08 labster hi lizmat
09:09 lizmat FROGGS_: I think there should be a presentation about your Perl 5 in Rakudo work at the YAPC::NA
09:09 lizmat would you be able to give such a presentation?
09:09 lizmat hi all!
09:10 tadzik hello lizmat
09:12 labster moritz: ok.  first of all, I think "filesystem" should be renamed to "volume".  Filesystem implies any(<zfs ntfs fat32 ...>) to me, which isn't what we'd be returning.  Volume is the P5 File::Spec terminology, and it seems to make more sense to me.
09:12 moritz labster: agreed
09:14 labster Next is stringification.  What was the logic behind having File::Path.Str return just the basename?  I have to say, I was surprised to see it return that rather than the whole path.
09:14 FROGGS_ lizmat: in north 'murica?
09:15 moritz labster: the reason was making dir() backwards compatible
09:15 moritz labster: in retrospect that's probably a bad idea
09:15 moritz so we need a deprecation notice, and wait 'til after the release
09:16 user____ joined #perl6
09:16 lizmat FROGGS: yes, in Texas even!
09:16 FROGGS_ lizmat: I'm afraid I don't have the money to get there :/
09:17 labster I was just looking around at File::Find, and File::Find::Result objects returns the whole path when stringified too.
09:18 moritz ah yes, but File::Find isn't the only user of dir()
09:18 tadzik yeah, I was imitating Perl 5's File::Find and find(1)
09:19 am0c joined #perl6
09:20 labster that was more of a similar behavior issue but I guess you're already convinced :)
09:22 lizmat FROGGS_: lack of money is fixable, lack of free time to travel and do a presentation is not
09:22 moritz labster: yes, I'm quite boring to talk to. You don't even have to ramp up all your arguments to convince me :-)
09:23 moritz labster: will you patch the specs?
09:23 labster huh, do I even have access to them?
09:24 moritz does https://github.com/perl6/specs/ say "Read-Only access" next to the URL?
09:24 FROGGS_ lizmat: out of curiosity, how do I fix that? :o)
09:24 moritz should be read-write
09:25 lizmat FROGGS_: you could be sponsored
09:25 FROGGS_ that is ... interesting ...
09:27 labster moritz: I do have access
09:27 labster .oO (  The path from hacker to language implementor is surprisingly short around here )
09:30 moritz labster: iirc you submitted a pull request to the ecosystem repo, so I put you into the 'perl6' team, which we use for nearly all repos in the 'perl6' organization
09:30 * lizmat afk for ~ 2.5 hours&
09:30 moritz because we're far too lazy to do any fine-grained access control
09:31 moritz so, everybody who submits a new module gets write access to the specs repo too :-)
09:31 labster see, laziness is a virtue
09:31 moritz and roast, and evalbot, and perl6-examples, and most of the other repos
09:31 masak greetz, #perl6
09:32 moritz \o masak
09:32 labster moritz: also, please take a look at IO::Path::More, and see if there's anything there you feel should be part of the spec.  Probably not stuff like .find
09:32 moritz labster: URL to the module/repo?
09:32 labster https://github.com/labster/p6-IO-Path-More
09:33 moritz the is_absolute and absolute and relative methods might be worth importing
09:33 moritz (though as is-absolute)
09:34 moritz maybe method parent
09:36 labster I get lazy and use underscores because I'm trying to port perl 5
09:37 moritz that's fine for a module, but not for core :-)
09:43 jaldhar joined #perl6
09:44 labster Oh, oh!  forgot the last issue
09:44 labster r: my $p = '.';  say $p.=path.WHAT for ^6;
09:44 p6eval rakudo a23458: OUTPUT«(IO::Path)␤(Str)␤(IO::Path)␤(Str)␤(IO::Path)␤(Str)␤»
09:45 fhelmberger joined #perl6
09:46 labster .path alternates object types, which is strange.  I don't know what to do about that.
09:47 moritz finding a better name for the string representation of a path would help
09:48 daniel-s_ joined #perl6
09:48 labster Str?   fullpath?
09:48 moritz +1
09:49 moritz just Str will be fine, I think
09:50 labster And IO::Path.path returns itself, I suppose.
09:51 moritz aye
09:58 FROGGS_ labster: you have commit bit now for File::Spec
09:59 FROGGS_ feel free to refactor it if you think it makes sense
09:59 DarthGandalf joined #perl6
10:00 labster FROGGS_: Thanks.  I was about to ask for it, but I got sidetracked on the specs.
10:00 FROGGS_ labster: I'm really sorry that I can't help you with it right now
10:01 labster no, please keep working on perl5, that's awesome
10:01 jaldhar joined #perl6
10:01 FROGGS_ labster++ # I love you
10:03 masak :)
10:03 labster hugme: hug FROGGS_
10:03 * hugme hugs FROGGS_
10:04 FROGGS_ *g*
10:06 Heather UPDATE #PERL6 SET USER = USER + 1
10:07 masak there's a nice warm atmosphere in here today :)
10:07 Heather masak that's Stockholm syndrome
10:07 tadzik :D
10:07 tadzik It's funny because masak is not that far from Stockholm
10:08 Heather >_<
10:08 masak heh.
10:09 masak Heather: so you're saying Perl 6 somehow holds us hostage, and we're starting to have sympathy for it?
10:10 Heather masak yep :P
10:11 Heather Osama Ben Camelia
10:12 labster .oO (Camelia with a long beard?)
10:12 Heather labster ... Richard Ben Camlia ? >_<
10:13 * Heather always imagine only one person when hear something about bread
10:14 tadzik the gingerbread man! \o/
10:14 labster Oh wow, it is getting late here in California (Ï€ AM).  I'll think about the specs tomorrow.
10:14 labster night everyone
10:14 Heather labster good night
10:15 * Heather wants more MMO & RP elements on #perl6 o_O
10:16 masak Heather: um. http://strangelyconsistent.org/blog/perl-6-is-my-mmorpg
10:16 Heather masak yes I saw it :)
10:17 masak also, http://github.com/masak/crypt
10:17 Heather well "SAW" is not what I wanted to say
10:18 Heather masak how to play it?
10:18 tadzik Heather: you can help me write Quest :)
10:18 tadzik although it's not quite RPG, and definitely not MMO :)
10:18 masak Heather: should be just running bin/crypt in that repo.
10:19 SamuraiJack joined #perl6
10:19 Heather tadzik quest?
10:19 masak Heather: oh, and it has a dependency (Adventure::Engine), but it should all install through panda.
10:19 masak errand &
10:19 Heather masak I use CRYSTAL MAIDEN !
10:19 Heather -_-'
10:21 Heather masak so first I will generate an gentoo ebuild for it )
10:23 tadzik Heather: https://github.com/tadzik/Quest
10:23 tadzik it's supposed to be a fun and addictive game
10:23 tadzik it's not, yet ;)
10:23 eiro joined #perl6
10:24 Heather for now MMO part is must have part of any game, even roguelike
10:24 tadzik I still prefer single-player games
10:25 tadzik this way I can stop playing whenever I want and no one gets angry
10:25 Heather well, I don't play single-player games even if they are very cool, I feel it boring
10:25 tadzik I see
10:28 Heather tadzik https://github.com/tadzik/Quest/pull/1
10:30 Heather it's strange...
10:30 Heather how github highlights it?
10:31 tadzik Heather: merged, thanks
10:32 Heather tadzik well, but it doesn't helped
10:32 Heather I was thinking about making github to highlight the file
10:32 Heather tadzik wrong file mode maybe?
10:33 Heather I guess ...
10:33 daniel-s_ joined #perl6
10:36 Heather tadzik https://github.com/tadzik/Quest/pull/2
10:36 Heather this way
10:37 tadzik Heather: hilight it how?
10:38 Heather tadzik https://github.com/Heather/Quest/blob/master/bin/quest
10:39 Heather strange?
10:39 Heather tadzik github is failing saying that " Heather 17 minutes ago Shebang " is last commit
10:39 tadzik hmm, it just thinks it's Perl 6
10:39 tadzik erm, Perl 5
10:40 tadzik why, it is the last commit
10:40 Heather tadzik no, https://github.com/Heather/Quest/commits/master
10:40 tadzik I'm confused. What's the problem?
10:40 tadzik I've only just merged pull-2
10:41 Heather good :)
10:42 Heather now it highlights
10:42 Heather but as perl5 seems like, yes
10:42 Heather tadzik : what means # I'm a bad person ? )
10:44 tadzik I probably wrote something ugly in the code
10:45 Heather tadzik why? :)
10:46 tadzik where did you find that?
10:46 * Heather can't do anything awhile rust recompiling. 95% RAM 35% Swap
10:46 Heather tadzik in bin file beginning, clear sub
10:47 tadzik ayes
10:47 tadzik because it's not Pure-perl in any way
10:47 tadzik it feels hacky, I'd prefer it gone and replaced with a sane solution
10:47 tadzik ANSI code for example
10:47 tadzik but when I tried an ANSI code it didn't work quite well
10:49 tgt joined #perl6
11:04 denisboyun_ joined #perl6
11:23 toddr_ joined #perl6
11:23 immortal joined #perl6
11:23 perigrin joined #perl6
11:23 immortal joined #perl6
11:23 bonsaikitten joined #perl6
11:23 bonsaikitten joined #perl6
11:23 shachaf_ joined #perl6
11:23 Pleiades` joined #perl6
11:23 `patch` joined #perl6
11:23 Pleiades` joined #perl6
11:23 huf joined #perl6
11:23 masak joined #perl6
11:23 mst joined #perl6
11:23 nemesys joined #perl6
11:23 anocelot joined #perl6
11:23 Util joined #perl6
11:23 pjcj joined #perl6
11:24 patspam joined #perl6
11:24 GlitchMr- joined #perl6
11:24 Hor|zon joined #perl6
11:24 pnu joined #perl6
11:24 bloonix joined #perl6
11:24 BooK joined #perl6
11:24 robins joined #perl6
11:24 sitaktif1 joined #perl6
11:24 takesako__ joined #perl6
11:25 masak well, that was fun.
11:25 LordV joined #perl6
11:28 yx joined #perl6
11:31 mangala joined #perl6
11:31 Gothmog_ joined #perl6
11:31 integral_ joined #perl6
11:31 clkao joined #perl6
11:31 PerlJam joined #perl6
11:31 krunen joined #perl6
11:31 Ayiko joined #perl6
11:31 Tene joined #perl6
11:31 xalbo joined #perl6
11:31 mikemol joined #perl6
11:31 ponbiki joined #perl6
11:31 sergot joined #perl6
11:31 snarkyboojum joined #perl6
11:31 mikec__ joined #perl6
11:31 jrockway joined #perl6
11:32 cibs_ joined #perl6
11:32 Heather masak ?
11:33 masak Heather: a netsplit.
11:33 census joined #perl6
11:33 Heather hm...
11:44 FROGGS_ beer with us? np
11:51 dalek v5: aecb96f | (Tobias Leich)++ | lib/Perl6/P5 (2 files):
11:51 dalek v5: add support for unless+else constructs
11:51 dalek v5: review: https://github.com/rakudo-p5/v5/commit/aecb96f996
11:52 cibs joined #perl6
11:52 Heather bear with us
11:52 Heather and balalaika
11:57 dalek v5: c1d7375 | (Tobias Leich)++ | t/spectest.data:
11:57 dalek v5: skip base/lex.t for now
11:57 dalek v5: review: https://github.com/rakudo-p5/v5/commit/c1d7375d7b
12:04 PacoAir joined #perl6
12:13 dalek v5: d13e413 | (Tobias Leich)++ | test.pl:
12:13 dalek v5: we have better tests than this
12:13 dalek v5: review: https://github.com/rakudo-p5/v5/commit/d13e4137d0
12:13 dalek v5: b0d3beb | (Tobias Leich)++ | Makefile:
12:13 dalek v5: add fulltest target
12:13 dalek v5: review: https://github.com/rakudo-p5/v5/commit/b0d3beb293
12:13 dalek v5: 2c5ad30 | (Tobias Leich)++ | .gitignore:
12:13 dalek v5: ignore *.v5 files
12:13 dalek v5: review: https://github.com/rakudo-p5/v5/commit/2c5ad30c1e
12:14 Heather could be fun but because of Perl and ML I want to visit some arabic country alike Egypt ...
12:19 masak Heather: I fail to see the connection. "because of Perl and ML" you want to visit some Arabic country?
12:19 masak is this about readability again? I thought we talked about this. :P
12:20 Heather masak It's about camel in logo
12:20 Heather ...
12:26 yx joined #perl6
12:33 lizmat is anybody here going to the QA Hackathon in Lancaster in ~ 2 weeks?
12:34 lizmat (as in non-lurkers)
12:37 kaleem joined #perl6
12:44 PacoAir joined #perl6
12:48 * lizmat takes that as a no :-)
12:51 [Coke] lizmat: ooh. if I was on one of my increasingly less rare trips to Leeds, I'd try to swing by.
12:52 lizmat the reason I'm asking, is that there is apparently going to be some serious CPAN hacking going on there
12:53 lizmat and I was wondering whether it would make sense to make sure that Perl 6 modules / distributions could be handled as well
12:54 lizmat either as a multi-version distribution (like with my Devel::MaintBlead support module) or in some other wat
12:55 dalek v5: 19b9cc8 | (Tobias Leich)++ | lib/Perl6/P5Actions.pm:
12:55 dalek v5: <blorst> -> <sblock>, this allows to use BEGIN blocks
12:55 dalek v5: review: https://github.com/rakudo-p5/v5/commit/19b9cc80a1
12:56 xinming joined #perl6
12:57 [Coke] argh. "where did panda install itself, and why do I have to keep changing my path to find it."
12:57 FROGGS_ std: use v5; print -d "."
12:57 p6eval std 86b102f: OUTPUT«ok 00:01 52m␤»
12:57 crab2313 joined #perl6
13:00 * [Coke] modularizes his .bashrc slightly so he can change a variable pointing at a parrot revision to track where panda got to
13:01 FROGGS_ [Coke]: panda should be bundled with the compiler release, and it should be always there and always working and always something and ...
13:01 [Coke] I'm not dealing with releases. I'm building from source.
13:01 [Coke] perhaps that is my problem.
13:03 FROGGS_ then panda must be made when rakudo gets recompiled
13:03 [Coke] ecosystem question: do we care if the ecosystem name doesn't match the module name? I see some that match except for a - instead of a ::, for example.
13:04 FROGGS_ hmmm, at least I dont care
13:04 [Coke] FROGGS_: Yes. every time I reinstall rakudo, I have to eithe rebootstrap panda OR completely reinstall it, depending on whether the underlying parrot changed.
13:07 tadzik [Coke]: what do you mean by the ecosystem name?
13:07 [Coke] as it appears on modules.perl6.org (the name in the META.info)
13:07 tadzik also, I've been thinking about reworking the installation process a bit, so these sort of things are not a problem anymore
13:07 FROGGS_ ++tadzik
13:08 tadzik ie imagine 'panda install Foo' saying "seems like you recompiled rakudo recently, do you want me to rebootstrap all the things?"
13:08 tadzik also getting rid of the PATH insanity we have now
13:08 FROGGS_ that would be sweet
13:08 [Coke] That would be nice.
13:09 [Coke] I'm trying to figure out how to get panda to let me install "YaMl"
13:09 [Coke] (or rather, correct me helpfully)
13:09 tadzik marry it to Text::Levenshtein
13:09 tadzik or even not
13:09 tadzik we could add a fallback, so if there's no such module it tries comparing lc($arg) to map { *.lc } @modules
13:10 tadzik then a "did you maybe mean $other-thing?"
13:10 [Coke] yes, that was going to be the first step, then maybe TL.
13:10 [Coke] I'm trying to suss out where that would go, though.
13:13 tadzik there can be some controversial issues with that, as moritz commented on the GH ticket
13:13 moritz suggesting stuff is fine
13:13 moritz autocoreccting not
13:14 moritz because then you might put slightly wrong names into some automated script, and once a module exists with that slightly different name, you suddenly install a different module
13:14 moritz but I think I wrote that already :-)
13:14 [Coke] moritz: yes, that's what we said.
13:14 [Coke] "calm down."
13:15 moritz ok, great :-)
13:15 masak a program should never *ever* automatically correct faulty input.
13:15 masak if you said it already, it bears repeating. :)
13:15 [Coke] ok, I'm not working on this, then. :P
13:16 masak may I suggest extremely nice error messages instead?
13:16 masak the right way to go ahead from the idea of automatic correction -- every single time through history -- is to pimp the error messages to say "did you mean YAML?"
13:17 [Coke] That's what I was *going* to do, yes.
13:17 [Coke] Holy crap, people.
13:17 masak sorry :/
13:18 [Coke] "I wish to make perl 6 nicer" "You're going to do it wrong!"
13:18 masak [Coke]: recently, after our nicer levenstein variable error messages, there have been suggestions that Rakudo automatically correct the source code. I guess I'm a bit touchy about the thought of that.
13:18 masak [Coke]: didn't mean to get it all over you.
13:20 dayangkun joined #perl6
13:21 [Coke] Is there a way to get the 6POD out of an installed module ?
13:21 [Coke] I wish to look at documentation from the command line.
13:21 FROGGS_ masak: automatically correct variables... and that wasn't a joke?
13:24 tadzik [Coke]: perl6 --doc /path/to/module.pm
13:24 tadzik maybe p6doc is capable of doing that
13:24 [Coke] tadzik: "ew"
13:24 [Coke] (see also complaints about "where did panda get installed"
13:24 tadzik ah
13:25 [Coke] something like perl6 --doc Module::Name would be spiffier.
13:25 masak FROGGS_: it was a suggestion by someone who hadn't seen the consequences of such a suggestion. I forget who.
13:25 tadzik it's even specced
13:25 tadzik I think it was me :)
13:25 masak FROGGS_: I think it's even been suggested two or three times recently, by different novices.
13:25 arnsholt timotimo: I can't reproduce your Zavolaj problem (it's also been reported as Zavolaj #23) =(
13:25 FROGGS_ masak: ohh, I just remember me suggesting that as a joke :o)
13:25 masak heh.
13:25 tadzik ah, in the source code
13:26 arnsholt So I'm guessing it's probably something GC-related
13:26 FROGGS_ masak: I'm a guy who would tike to force 'strict' for Perl 5, even for one-liners... I would never like to have source-code-autocorcection
13:27 [Coke] ... Did anyone every implement Perl6::Format for perl6 ?
13:27 [Coke] *ever
13:28 fgomez joined #perl6
13:28 [Coke] https://github.com/mathw/form/ ?
13:29 masak yes, that's the one effort I know about.
13:30 masak it's a couple of years sleeping now.
13:30 masak mathw_: ping :)
13:31 [Coke] modules.perl6.org is written in perl5 ?
13:32 tadzik yes
13:32 tadzik http://feather.perl6.nl/~sergot/modules/ is all Perl 6
13:33 [Coke] ... so it's already been ported to perl6, but the main site is using the p5 version?
13:33 tadzik yeah
13:33 tadzik also, the latter has a per-module pages and tests results
13:34 [Coke] it also hasn't been updated in a while.
13:34 tadzik yeah
13:35 tadzik sergot: ping
13:35 [Coke] Is sergot's version going to become the main version at some point? blocking on anything?
13:36 tadzik not sure
13:37 * [Coke] sees that p6prove is also under started.
13:37 [Coke] er, *underway
13:37 [Coke] which then makes the last word useless. ;)
13:38 tadzik yep
13:39 skids joined #perl6
13:41 * lizmat wonders where panda is specced?
13:42 bluescreen10 joined #perl6
13:45 hoelzro morning #perl6
13:46 cibs_ joined #perl6
13:47 tadzik lizmat: it's not, not really
13:47 tadzik actually, Pies is supposed to be a spec and Panda the implementation
13:51 [Coke] what do folks think about porting individual pure p5 modules vs. the v5 work going on rcently?
13:52 nwc10 a) long(er) term the ported version useful
13:52 nwc10 b) short term the ported version means more code written in Perl 6, which is also useful
13:52 nwc10 in the middle, having a way to automatically convert stuff is useful
13:52 lizmat I see the v5 work as important for darkpan
13:53 nwc10 oh yes, that too
13:53 lizmat porting CPAN modules (be they pure perl or not) is a task that will need to be done as needed
13:53 lizmat and while doing that, will make a good selection of useful CPAN modules in the process
13:54 lizmat anyway, that's my .02
13:57 masak I think something using PPI could get pretty far converting Perl 5 to Perl 6.
13:57 masak in fact, ooh! *that's* what my first CPAN module is going to be. obviously.
13:59 masak the idea being: if Perl::Critic can do a good job finding stuff to improve, a converter module can do a decent job sidegrading code to Perl 6, using known automatic conversions.
14:00 masak and it's also completely fine for it give up in some cases, or suggest improvements that can only be realistically carried out by a human.
14:02 colomon Hmmm... porting tests well should be a priority, no?
14:02 lizmat so, as a sort of "ufo" that would build an annotated Perl 6 source code version?
14:02 kaare__ joined #perl6
14:03 masak colomon: that's a good point.
14:03 masak colomon: and maybe smaller scripts.
14:03 masak lizmat: :)
14:07 cibs joined #perl6
14:08 raiph <hoelzro> where are the various dispatchers ... <FROGGS> in nqp's C land
14:09 timotimo well, the multiple dispatch is now implemented in nqp, too
14:09 timotimo or ... "only"
14:09 raiph hoelzro: what timotimo said
14:10 arnsholt timotimo: I'm not having luck reproing your bug, but it's been reported on GitHub as well
14:10 timotimo read that, thanks
14:12 timotimo Rakudo-2003.03 - hehe, wat? :)
14:12 masak blast from the past!
14:13 * lizmat is considering porting Sys::RunAlone to Perl 6
14:13 lizmat nice small module
14:13 bluescreen10 joined #perl6
14:13 timotimo is that when you only want to have one instance of your program?
14:13 mtk joined #perl6
14:14 lizmat yup
14:14 timotimo sounds nice
14:14 PerlJam that's an odd name though.  "run alone"?  without any friends?
14:15 tadzik like libunique
14:15 tadzik which is now GtkApplication, apparently
14:16 PerlJam especially given a widely known term for "only one instance" -- singleton.
14:16 lizmat "WARNING Unique is now in maintenance mode, and its usage is strongly discouraged"
14:16 PerlJam (or, perhaps I'm just curmudgeonly todya :)
14:16 colomon masak: I'm now reading up on PPI....
14:17 masak PPI is great.
14:17 masak Alias++ at his best, solving unsolvable problems.
14:17 timotimo arnsholt: in #22 ("how to store binary data in a buffer"), you refer to lots of plumbing having to be done; is that refering to vmarray?
14:17 lizmat Sys::RunAlone is much, much simpler
14:17 lizmat it just attempts to lock the source file at INIT time, and exits if it can't
14:19 tadzik clever
14:20 lizmat I can't take all the credit: "Inspired by Randal Schwartz's mention of using the DATA handle as a semaphore
14:20 lizmat on the London PM mailing list."
14:20 lizmat that was back in 2005  :-)
14:23 Psyche^ joined #perl6
14:32 am0c joined #perl6
14:33 xilo joined #perl6
14:33 cibs_ joined #perl6
14:43 crab2313 joined #perl6
14:57 colomon masak: hmmm, corollary to my previous comment: porting p5 testing modules early on should probably be a priority.
15:01 arnsholt timotimo: Yeah, that's the vmarray stuff
15:02 masak colomon: would be interesting to know which ones are the most used p5 testing modules.
15:02 colomon masak: yes, it would.
15:03 masak colomon: as long as it's just Test::Simple and Test::More, we're already set.
15:05 bonsaikitten joined #perl6
15:06 bluescreen10 joined #perl6
15:06 tadzik and Test::Fatal
15:08 timotimo Test::Fatal is the one where it deletes your hard drive if the tests fail?
15:09 masak seems fair to me.
15:13 colomon Is that like the old product that would eject your floppy disk at high speed if it detected any signs of a system crash?
15:14 * colomon suspects that ad was in Creative Computing circa April 1982.....
15:15 timotimo ha, that's cute
15:15 timotimo also i imagine painful
15:15 arnsholt timotimo++ # Backtrace from -g compiled Parrot/NQP
15:15 arnsholt I <3 you a little bit now =)
15:21 timotimo :)
15:21 timotimo didn't even realise there was nothing wrong with the backtrace until now
15:28 arnsholt And there is an NQP frame in there: #21 is the invoke handler
15:29 arnsholt Oh, could you add to the bug thread or the Gist if the backtrace is from the tweaked test file with loop, or the plain one?
15:29 arnsholt (I'm guessing with loop?)
15:31 lizmat wrt to porting Test modules, Test::Effects ?
15:31 masak lizmat: nice. didn't know about that one before.
15:32 * lizmat was exposed to it again at the SPW by the closing keynote of TheDamian
15:33 lizmat this would also mean Test::Trap
15:35 timotimo it's from the one with the loop, yes
15:35 timotimo do you also want a stacktrace from without the loop?
15:38 arnsholt Yeah, do that as well. Might come in handy
15:46 thou joined #perl6
15:46 arnsholt But from the bottom of the stack trace it looks like something weird's happening during GC
15:47 arnsholt *sigh* I really need to get my testing thing running
15:47 am0c joined #perl6
15:47 timotimo oh
15:47 timotimo another thing
15:48 timotimo using -Ilib instead of -Iblib/lib i get no segfault for the non-loop version
15:49 * [Coke] finds http://retester-daoswald.dotcloud.com/ in his feedly. perl5 regex online debugger.
15:49 timotimo and for the loop version as well
15:52 timotimo arnsholt: can you make any sense of that?
15:53 timotimo serialisation and gc interoperating in a strange way?
15:55 thou joined #perl6
15:55 moritz the two most "fun" type of problems to debug interacting
15:57 timotimo how does one properly differentiate fun as in "perl6 is -Ofun" and "when playing dwarf fortress, you have lots of opportunities to have Fun"? is it just the capitalisation?
16:02 SamuraiJack_ joined #perl6
16:04 bbkr joined #perl6
16:07 [Coke] We've had a p6 project manager in the past. Is anyone currently filling that roel?
16:07 [Coke] *role?
16:12 timotimo isn't that pm's role or something?
16:12 timotimo given pmichaud { say $_ ~~ Perl6::Project::Manager }
16:13 Targen joined #perl6
16:17 [Coke] I wouldn't have guessed that, no.
16:20 [Coke] tadzik: what's left todo on https://github.com/tadzik/Template-Mojo ?
16:21 [Coke] (also, why is it Template::Mojo instead of 5's Mojo::Template ?)
16:22 masak "A leader is best when people barely know that he exists" -- Lao Zi
16:29 dmol joined #perl6
16:30 Chillance joined #perl6
16:34 arnsholt timotimo: Odd. If I -Iblib/lib or -Ilib it doesn't matter on my end
16:35 domidumont joined #perl6
16:37 PerlJam Coke: I don't understnad "p6 project manager".  We don't have a "Perl 6 project" as such.  We have a "Rakudo project" and a "Niecza project" and so forth though.  And I'd say we have a "Perl 6 product", but that's something different.
16:38 PerlJam [Coke]: who was the past perl6 project manager?
16:39 masak how many have we had?
16:41 _sri [Coke]: i like the name, keeps Mojo::Template free for the official port
16:42 PerlJam google tells me it was obra
16:43 PerlJam (I never would have guessed that :)
16:44 masak that sounds vaguely familiar.
16:45 PerlJam From Aug 2007 -- http://www.theperlreview.com/Interviews/jesse-vincent-200708.html
16:45 PerlJam It's so weird reading that now :)
16:54 [Coke] jesse is the one I remember. Allison just mentioned she was doing the job for a while.
16:54 [Coke] PerlJam: some (myself included) would argue that that's part of the problem. :)
16:54 PerlJam [Coke]: I might just agree with you.
16:55 [Coke] though I think rakudo is "winning" through sheer force of jnthn++.
16:55 PerlJam .oO( jnthn is Charlie Sheen? )
16:56 arnsholt More like jnthn is the Chuck Norris of compiler hacking =D
16:59 masak I work with him. I can confirm he is Chuck Norris.
17:00 kaleem joined #perl6
17:01 * [Coke] ponders something project schedule like with no dates so we can document (and potentially agree) on what direction we're heading in.
17:04 masak that's partly what the planned production-readiness site is meant to do.
17:04 masak kind of like a compiler comparison chart, but focusing on "how far along are we with these desirables that would enable us to say 'yep, it's ready for serious use'"
17:07 arnsholt The Zavolaj fails in the recent past are really motivating me to whipping up this automated testing stuff I talked about recently as well
17:08 arnsholt We really need to get something like that running
17:08 masak +1
17:09 [Coke] wasn't "emmentaler" the answer to that question?
17:09 masak yes.
17:09 [Coke] what's the current state of that?
17:09 masak sergot++ got something up and running.
17:09 masak tadzik linked to it earlier today in the backlog.
17:09 masak but it hasn't been updated in a while.
17:10 arnsholt Does emmentaler test the compilers as well?
17:16 flussence joined #perl6
17:19 masak arnsholt: don't think so. I see why that would be nice-to-have too, especially on multiple platforms. but it feels like a separate thing from testing the module ecosystem.
17:19 domidumont1 joined #perl6
17:20 FROGGS_ it should be kept separate, cpantesters only cares about modules too
17:20 FROGGS_ you could treat perl as a module though
17:20 FROGGS_ (on the websites)
17:20 arnsholt True, true
17:21 [Coke] arnsholt: I'm already testing the compilers:
17:21 [Coke] https://github.com/coke/perl6-roast-data
17:22 [Coke] (3 of them, anyway)
17:22 [Coke] ah. so emmentaler is not just "test the modules", but is also the answer to  "dogfood modules.perl.org?"
17:23 masak seems so.
17:24 flussence g'morning, *
17:25 masak o/
17:25 thou joined #perl6
17:26 * [Coke] trips over something in java 1.7 and wonders if he's expecting something he shouldn't
17:27 raiph Heather: I glanced at cm. the MAIN sub has me wondering. do you know about automatic usage/help features? and using multi subs to process various command line options? http://perlcabal.org/syn/S06.html#Declaring_a_MAIN_subroutine
17:27 flussence hoelzro: ping? perl6-posix/META.info has a stray whitespace at «[ "NativeCall " ]» and panda doesn't like that.
17:28 arnsholt What I'm thinking of is something to automatically build Rakudo and NQP (at least for the time being) with --gen-parrot and against Parrot HEAD, and test a selection of modules
17:28 raiph (for others reading along, i'm talking of https://github.com/Heather/CrystalMaiden/blob/master/bin/cm )
17:28 arnsholt At least test Zavolaj, since it's pretty intertwined with the compiler guts, and I tired of randomly finding out it's broken because of some internals change I missed =)
17:29 [Coke] arnsholt: I think the "selection of modules" testing is handled by sergot's emmentaler project.
17:29 arnsholt Right. I should probably look at emmentaler first
17:29 * [Coke] checks to see which version of parrot he's testing rakudo with daily.
17:29 * [Coke] is just using --gen-parrot
17:30 * [Coke] suspects that's appropriate for roast tests.
17:30 tadzik [Coke]: I think Mojo::Template is a bad name :)
17:31 tadzik esp. in our ecosystem, where it has nothing to do with Mojo, but a lot to do with Templates
17:32 [Coke] and once Mojo exists?
17:32 tadzik then we'll see
17:32 tadzik _sri | [Coke]: i like the name, keeps Mojo::Template free for the official port
17:32 tadzik sounds good to me
17:32 [Coke] ... who is going to do the "official" port if not you? ;)
17:32 tadzik _sri, of course :)
17:32 [Coke] and why is this version not it? ;)
17:32 tadzik I'm only doing unofficial ports
17:33 [Coke] (_sri) ... ah. well that's fine, then!
17:33 tadzik of Dancer, of Mojo::Template... ;)
17:34 * tadzik decomutees
17:35 dalek Perlito: 4f7a85e | (Flavio S. Glock)++ | src5/lib/Perlito5/X64/Assembler.pm:
17:35 dalek Perlito: Perlito5::X64::Assembler - Operand set_disp8, set_disp32
17:35 dalek Perlito: review: https://github.com/fglock/Perlito/commit/4f7a85e983
17:46 tgt joined #perl6
18:20 hoelzro good * #perl6
18:20 colomon \o
18:20 spider-mario joined #perl6
18:23 hoelzro can anyone help me understand what '<O('%methodcall')>' does?
18:23 hoelzro (in Grammar.pm)
18:24 arnsholt Sets the precedence of the thingy
18:24 arnsholt Which the OPP (in EXPR) uses to sort out grouping
18:25 PerlJam precedence and associativity, n'est pas?
18:25 arnsholt And associativity, that's true
18:26 hoelzro where is the O token/rule defined?
18:27 hoelzro and more importantly, where is the code that actually handles calls?
18:28 masak hoelzro: in HLL::Grammar
18:28 hoelzro ah ha
18:28 masak (in nqp)
18:28 hoelzro masak: was that for O, or call handling?
18:28 * hoelzro is looking at that callsame bug
18:29 FROGGS_ and in Perl6::Grammar, search for Perl6::Grammar::O(
18:29 FROGGS_ Perl6::Grammar line 2885
18:29 masak hoelzro: that was for O
18:29 hoelzro ah, ok
18:30 hoelzro so it seems that callsame + routine traits doesn't work because Routine.can('postcircumfix:<( )>') is undefined
18:31 thundergnat joined #perl6
18:31 hoelzro so I think a "dummy" implementation should be present so callsame can do its job
18:32 * PerlJam wonders if there are any other "unimplemented" operators like that.
18:33 masak hoelzro: I think you're simply the first to try and overload invocation like that. it makes a lot of sense to do it -- it's just not been done before.
18:33 hoelzro I'm guessing calling semantics are implemented in Parrot land
18:34 hoelzro masak: yay, I'm first =)
18:34 PerlJam hoelzro: that means you need to blog about it ;)
18:34 masak not sure the calling semantics are that tied to Parrot these days.
18:34 hoelzro ;)
18:34 hoelzro I'm guessing it's done in a dispatcher somehow
18:34 thundergnat Good morning (UGT) perl6
18:35 hoelzro my current technique relies on inserting nqp::say and recompiling =/
18:35 hoelzro thundergnat: ahoy!
18:35 PerlJam hoelzro: does monky patching not work so that you don't have to endure the long compile?
18:36 hoelzro PerlJam: good question; haven't tried
18:36 PerlJam heh, "monky patching" ... that's what happens on perlmonks  :)
18:36 masak thundergnat! \o/
18:37 thundergnat masak \o\
18:37 thundergnat errr...  \o/
18:37 masak /o/
18:38 PerlJam hoelzro: I bet you can Routine.HOW.add_method() at the top of your script before you add the trait.
18:38 masak hoelzro: it just struck me... did you explore doing '$r()' rather than 'callsame' as a workaround?
18:38 thundergnat I had mentioned a few days ago that some infinite lists have extremely slow .first methods. Well, I put up a gist exploring the boundary of the problem and a possible fix.   https://gist.github.com/thundergnat/5265633
18:38 PerlJam masak++ oh, good idea
18:38 masak hoelzro: need to pass in the right args too. but that's a detail.
18:39 thundergnat It may not be the best way to go about it, but it exposes the problem better at least.
18:39 hoelzro masak: no, I did not...
18:39 hoelzro good call!
18:41 hoelzro masak: infinite recursion =/
18:41 [Coke] thundergnat++
18:41 masak hoelzro: oh! right, 'does'...
18:42 dalek perl6-roast-data: 64a9dd8 | coke++ | / (3 files):
18:42 dalek perl6-roast-data: today (automated commit)
18:42 dalek perl6-roast-data: review: https://github.com/coke/perl6-roast-data/commit/64a9dd8093
18:42 raiph hoelzro: fwiw, afaik the dispatcher is now in nqp (not C)
18:42 kaare_ joined #perl6
18:43 thundergnat [Coke]: I even went so far as to recompile Rakudo with those methods added and it still passes roast... but it might be attacking the problem from the wrong direction.
18:43 [Coke] niecza still has 18 failures. (which as I recall colomon couldn't duplicate - anyone else want to do a spec test run on niecza and see if they can find some tests to fudge?)
18:43 hoelzro I'm not entirely clear which dispatcher does what
18:43 hoelzro ex. jnthn's work and src/Perl6/Metamodel/Dispatchers.pm
18:43 [Coke] thundergnat: pmichaud is probably the person to weigh in on that.
18:44 thundergnat [Coke]: tru
18:44 thundergnat *e
18:46 masak can anyone explain to me why https://twitter.com/AlgebraFact/status/317343468282331137 is true?
18:47 mmcleric joined #perl6
18:47 masak I mean, those are the only two groups of order four I know of, up to isomorphism.
18:47 masak but that doesn't mean it's true/proven. :)
18:48 hoelzro it's the *binder* that jnthn ported to NQP, right?
18:48 mtk joined #perl6
18:49 * masak .oO( full of women )
18:49 hoelzro haha
18:50 colomon joined #perl6
18:55 raiph hoelzro: ah. as you were then. :)
18:56 raiph hoelzro: a commit comment by jnthn: "Rip out most of the C-based multi-dispatcher. Only leftover bits are the compile-time dispatch analysis used by the optimizer, which still needs porting."
18:57 hoelzro I see
18:57 hoelzro where is the dispatcher source located?
18:57 hoelzro I could only find src/Perl6/Metamodel/Dispatchers.pm
19:00 raiph https://github.com/rakudo/rakudo/commit/6191f6d65ce46a451c3f4e64d3d1d1ace011b457 ?
19:02 hoelzro I just want to find the code that implements postcircumfix:<( )> =(
19:03 arnsholt That should be somewhere in Routine or something
19:03 arnsholt Lemme grep for a 'sec
19:05 arnsholt hoelzro: Try src/core/Routine.pm:85
19:05 arnsholt Oh, wait. No
19:08 moritz r: say (sub () { }).can('postcircumfix:<( )>')
19:08 p6eval rakudo a23458: OUTPUT«␤»
19:09 moritz iirc there's some 6model stuff which handles invocations without having to go through postcircumfix:<( )>
19:09 arnsholt Oh, right. That's probably it
19:10 raiph r: say nqp::can(sub () { }, ('postcircumfix:<( )>')
19:10 p6eval rakudo a23458: OUTPUT«[31m===[0mSORRY![31m===[0m�Unable to parse expression in argument list; couldn't find final ')'�at /tmp/D7uzLWyAWK:1�------> [32m:can(sub () { }, ('postcircumfix:<( )>')[33m�[31m<EOL>[0m�    expecting any of:�        postfix�»
19:11 raiph r: say nqp::can(sub () { }, 'postcircumfix:<( )>')
19:11 p6eval rakudo a23458: OUTPUT«0␤»
19:12 raiph gotta run, but https://github.com/rakudo/rakudo/blob/nom/src/Perl6/Metamodel/BOOTSTRAP.pm ?
19:13 moritz afaict BOOTSTRAP just calls that method, not installs it
19:21 arnsholt hoelzro: I'm completely stumped. Sorry =/
19:24 hoelzro yeah, me too =/
19:25 masak this may be the *weirdest* Perl 6 code I've ever written: https://gist.github.com/masak/5266081
19:25 masak (but, man do I love Perl 6)
19:25 masak also, it works. :)
19:26 shachaf_ joined #perl6
19:31 masak but I think that answers my own question about Z_4 and Z_2 X Z_2 being the only two groups of that size.
19:33 perigrin safari doesn't like your infix operator.
19:34 perigrin unless ... you're intentionally using an empty box?
19:34 masak no, it's a cute little 'x' here.
19:34 * perigrin blames mobile safari then.
19:34 masak .u ⨯
19:34 yoleaux U+2A2F VECTOR OR CROSS PRODUCT [Sm] (⨯)
19:34 masak yoleaux: yo! :)
19:34 perigrin yeah works fine in prompt.app ... fails miserably in mobile safari.
19:35 perigrin chrome doesn't like it either so I blame webkit or whatever the iOS HTTP widget is.
19:38 flussence .u ×
19:38 yoleaux U+00D7 MULTIPLICATION SIGN [Sm] (×)
19:39 flussence I guess 2A2F is the more technically correct one
19:40 lue hello world o/
19:43 masak .u 2A2F
19:43 yoleaux U+2A2F VECTOR OR CROSS PRODUCT [Sm] (⨯)
19:43 masak flussence: oh! point. will change.
19:44 masak that's the one I was looking for. missed it somehow.
19:48 labster joined #perl6
20:25 thirdm joined #perl6
20:26 sorear .u TENSOR
20:26 yoleaux No characters found
20:27 awwaiid masak, what is that FIRST business?
20:30 diakopter .u 2297
20:30 yoleaux U+2297 CIRCLED TIMES [Sm] (⊗)
20:30 diakopter eh
20:30 sorear .u 2299
20:30 yoleaux U+2299 CIRCLED DOT OPERATOR [Sm] (⊙)
20:31 sorear masak: 00D7 has a unicode alias of "cartesian product"
20:33 xinming joined #perl6
20:36 cognominal joined #perl6
20:37 sorear 2297 "tensor product" may also be appropriate..  2A2F is rather not
20:38 diakopter o_O
20:42 masak awwaiid: it's a nice way to put a "header" into a print loop.
20:44 awwaiid so those happen onthe first iteration only?
20:44 awwaiid fancy
20:44 masak yep.
20:44 PerlJam hence the name :)
20:45 awwaiid the all-caps threw me I guess. used to perl5 all-caps labels being strewn about
20:45 PerlJam masak: why didn't you use a FIRST block?  Accident of implementation?  or you just like the repeated FIRST?
20:45 * colomon seems to recall desperately wanting FIRST a few years back, but has never actually used it since it actually got implemented.
20:46 sorear awwaiid: it's all caps because it's viewed as a generalization of BEGIN/CHECK/INIT/END
20:46 awwaiid interesting
20:46 * lue still likes  now - BEGIN now  for execution time
20:46 PerlJam awwaiid: we call those things "phasers" because they are all about when things are executed
20:46 awwaiid I tend to discourage/avoid BEGIN/CHECK/INIT/END whenever possible
20:46 sorear "phasers" is the term we use now, because they correspond to a generalized notion of "execution phase changes"
20:46 PerlJam (in what phase of execution)
20:47 * awwaiid absorbs twice as fast
20:48 sorear after a phaser you can have a block or statement, a syntactic category that we named "blast" because puns
20:48 masak PerlJam: just like the repeated FIRST.
20:48 awwaiid excellent priorities, sorear
20:53 sorear awwaiid: don't credit me, they were already blasts when I joined
21:01 fgomez joined #perl6
21:06 bruges joined #perl6
21:08 arnsholt sorear: I think Rakudo calls them blorst, these days =)
21:09 sorear arnsholt: when i joined STD called them blasts and Rakudo called them blorsts.  Still not fixed? :(
21:09 arnsholt Not AFAIK
21:09 fgomez joined #perl6
21:10 PerlJam blorst still fits ... block or statement
21:10 PerlJam :)
21:12 masak I think 'blast' won. despite someone's touching defense on p6l of 'blorst'.
21:14 kaare_ joined #perl6
21:19 rindolf joined #perl6
21:19 hoelzro ok, I think I may have figured out how this works
21:19 hoelzro sort of.
21:20 hoelzro I think it's routine_def (or whatever it's called)
21:21 hoelzro that's at least the part that associates the QAST block with the Perl 6 Block object, I think
21:22 colomon joined #perl6
21:24 hoelzro ah ha!
21:24 hoelzro found it!
21:24 hoelzro BOOTSTRAP.pm line 555
21:27 thou joined #perl6
21:33 thirdm joined #perl6
21:33 thirdm left #perl6
21:34 PerlJam hoelzro: how does that line relate to your problem?
21:34 * PerlJam doesn't immediately see it.
21:34 hoelzro PerlJam: well, it and 556 show me that Parrot has an "invoke" hook that gets bound to whatever's in the $!do attribute
21:34 hoelzro this is from digging around in other places, too
21:42 am0c joined #perl6
21:47 gdey joined #perl6
21:49 timotimo moritz: any new ideas WRT EXPORT_hash?
21:50 hoelzro (EXPORT improvements)++
21:56 * masak .oO( IMPORT exprovements )
21:58 hoelzro so, it seems the QAST block gets associated with the code object at compile time...
21:58 hoelzro ...but I'm not sure where it actually gets compiled to bytecode
22:01 hoelzro wow, this is hard =/
22:01 hoelzro I should really be doing my Coursera stuff.
22:01 hoelzro or visiting with family.
22:03 hoelzro moar typo fixes! https://github.com/rakudo/rakudo/pull/119
22:04 thou joined #perl6
22:08 labster Okay, I've never done a commit to the specs yet, so how does this IO::Path look?  https://gist.github.com/labster/5267241
22:09 woolfy left #perl6
22:15 hoelzro ok, I haven't figured out how to fix the callsame bug (yet)
22:15 hoelzro but I *do* have a workaround: my $clone = $r.clone; $r does role { method postcircumfix:<( )>(|) { say 'before'; my $result = $clone(); say 'after'; return $result; } }
22:19 xilo joined #perl6
22:29 thou joined #perl6
22:34 skids joined #perl6
22:35 masak hoelzro: nice.
22:36 masak awkward, but better than no workaround. :)
22:47 masak 'night, #perl6
22:47 dalek rakudo/nom: b176ca8 | (Rob Hoelz)++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
22:47 dalek rakudo/nom: More typo fixing
22:47 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b176ca8d09
22:47 dalek rakudo/nom: 25a9d6d | (Tobias Leich)++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
22:47 dalek rakudo/nom: Merge pull request #119 from hoelzro/nom
22:47 dalek rakudo/nom:
22:47 dalek rakudo/nom: More typo fixing
22:47 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/25a9d6d4c5
22:53 dalek specs: 6960683 | (Brent Laabs)++ | S32-setting-library/IO.pod:
22:53 dalek specs: Change IO::Path stringification, add new methods.
22:53 dalek specs: .Str returns full path instead of basename
22:53 dalek specs: .path is now a no-op (so Str.path.path still returns IO::Path)
22:53 dalek specs: .contents is now documented - lists the directory
22:53 dalek specs: .is-absolute, .is-relative, .absolute, .relative added
22:53 dalek specs: review: https://github.com/perl6/specs/commit/696068389a
22:55 revdiablo joined #perl6
22:56 thou joined #perl6
23:21 [Coke] rn: ('A' ... 'a').say
23:22 p6eval niecza v24-35-g5c06e28: 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 [ \ ] ^ _ ` a␤»
23:22 p6eval ..rakudo 25a9d6: OUTPUT«(timeout)»
23:24 FROGGS_ rn: ('A' ... 'Z').say
23:24 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28: 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␤»
23:24 FROGGS_ rn: ('_' ... 'a').say
23:24 p6eval rakudo 25a9d6: OUTPUT«(timeout)»
23:24 p6eval ..niecza v24-35-g5c06e28: OUTPUT«_ ` a␤»
23:25 FROGGS_ rn: say '_'.ord
23:25 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28: OUTPUT«95␤»
23:25 FROGGS_ rn: say 'a'.ord
23:25 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28: OUTPUT«97␤»
23:29 [Coke] rakudo: my %h=('a'=>1,'b'=>2); my $h_ref=\%h; say $h_ref.{'a'}
23:29 p6eval rakudo 25a9d6: OUTPUT«(Any)␤»
23:29 tgt joined #perl6
23:29 [Coke] masak: please check https://rt.perl.org/rt3/Ticket/Display.html?id=76462 - I think this is probably correct autoviv behavior now.
23:30 [Coke] r: eval "{\n\n\n\n\'}"
23:30 p6eval rakudo 25a9d6: OUTPUT«[31m===[0mSORRY![31m===[0m�Confused�at /tmp/p_KRuLhO3q:1�------> [32meval "{\n\[33m�[31mn\n\n\'}"[0m�    expecting any of:�        postfix�»
23:33 dalek v5: a71daf8 | (Tobias Leich)++ | lib/Perl6/P5 (2 files):
23:33 dalek v5: let -d be parsed as term:filetest rather that an unknown sub with prefix
23:33 dalek v5:
23:33 dalek v5: I added *a lot* debug stuff for this, which can be turned on by setting $DEBUG.
23:33 dalek v5: review: https://github.com/rakudo-p5/v5/commit/a71daf88d2
23:38 FROGGS_ r: say '.'.IO.d
23:38 p6eval rakudo 25a9d6: OUTPUT«True␤»
23:48 dalek v5: e0e5402 | (Tobias Leich)++ | lib/Perl6/P5 (2 files):
23:48 dalek v5: dispatch -e "foo" to "foo".IO.e
23:48 dalek v5: review: https://github.com/rakudo-p5/v5/commit/e0e540254d
23:50 [Coke] rn: +(23..2300000000)
23:50 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28: OUTPUT«(timeout)»
23:51 [Coke] rn: 0..^3.3
23:51 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28:  ( no output )
23:51 [Coke] rn: say 0..^3.3
23:51 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28: OUTPUT«0..^3.3␤»
23:51 pmichaud good evening, #perl6
23:51 [Coke] rn: say 0..^3.3.list
23:51 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28: OUTPUT«0..^1␤»
23:51 [Coke] rn: say (0..^3.3).list
23:51 p6eval rakudo 25a9d6, niecza v24-35-g5c06e28: OUTPUT«0 1 2 3␤»
23:52 [Coke] pmichaud: Allo.
23:52 FROGGS_ hi pmichaud
23:52 * [Coke] wonders if +(23..2300000000) should emit a number quickly.
23:53 FROGGS_ well, it could cheat
23:54 [Coke] I am looking at a patch which makes it cheat, aye.
23:54 pmichaud [Coke]: I suspect it wouldn't be too hard to manage, no.
23:54 pmichaud er
23:54 [Coke] RT# 82312. seeing if it breaks anything...
23:54 pmichaud if there's an easy + reliable way to emit a number quickly, I'd go for that :)
23:55 [Coke] ok. I'll get this applied.
23:56 pmichaud not sure if Numeric needs a :D invocant there
23:56 [Coke] test for that? +Range ?
23:59 [Coke] > +Range
23:59 [Coke] Cannot look up attributes in a type object

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

Perl 6 | Reference Documentation | Rakudo