Perl 6 - the future is here, just unevenly distributed

IRC log for #phasers, 2011-11-16

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

All times shown according to UTC.

Time Nick Message
14:34 jnthn I'll miss #phasers due to an evening $dayjob task
14:34 jnthn Quick preport...
14:34 jnthn * Much of the week was sucked up by $dayjob, travel, attending London Perl Workshop and visiting
14:35 jnthn * LPW went well - gave two talks
14:35 jnthn * Since I got back, worked with moritz++ got get the last things blocking bigint dealt with, and it got merged
14:35 jnthn * Started to dig into the NFA branch of NQP
14:35 jnthn In the next week...
14:35 jnthn * Tuit supply way better than the last couple of weeks
14:35 jnthn * (hopefully) :)
14:36 jnthn * Plan to focus on NFA branch and getting us some protoregex support in nom
14:36 jnthn * Will also fix that bigint boolification bug before the release
14:36 jnthn * Probably other bits
14:36 jnthn Planning
14:37 jnthn * We're missing a Rakudo release person this month - sign up! ;)
14:37 jnthn * I'm aiming that we are in shape for a Rakudo * relesae in December. This means I plan to focus on protoregexes, NCI and other bits that unblock the ecosystem.
14:37 jnthn EOR
16:37 Util Pre-report:
16:37 Util (I may be in-and-out of #phasers due to $WORK.)
16:37 Util # Done:
16:37 Util * Fixed typos in Perl 6 specs.
16:37 Util * Found conflict in [S03] w.r.t. left-hand-side of xx operator.
16:37 Util * Unravelled and committed about half of my 5-year "Perl 5 to Perl 6 translator" code slush:
16:37 Util = https://github.com/Util/Blue_Tiger#readme
16:37 Util Feedback/review welcome from all perspectives
16:37 Util # Plan to do:
16:37 Util * Finish committing the Blue Tiger code slush:
16:37 Util = is_structure_like()
16:37 Util = for my $i (@a) {...}    to   for @a -> $i {...}
16:37 Util = other control structure changes.
16:37 Util * Brush off Mac binary creator for upcoming Rakudo Star
16:38 Util # Blockers:
16:38 Util * $WORK
16:38 Util EOR
16:40 tadzik pre-report (classes during #phasers): closed a bunch of tickets, we're under 450. Abandoned the idea of a META.info dedicated parser due to regex mess. Plans: more ticket closing, maybe. EOR
16:57 colomon pre-report:
16:57 colomon * Did nothing p6ish except worry that the state of the module tools is sadly lacking in both nom and niecza.
16:57 colomon plan:
16:57 colomon * Hope to get the mandelbrot script working in Gtk (note: potential Advent calendar topic)
16:57 colomon * Continue advocating for the modules
16:57 colomon EOR
17:05 diakopter pre-report ($work during #phasers): rolled back most of the optimization attempts in nqplua. Stalled on porting QRegex due to the lack of labels/goto in Lua 5.1. They exist in lua 5.2, and the author of luajit mentioned to me they're on his TODO but with no known time horizon. Started digging into QRegex NFA branch to increase its bus number. Tried to debug protoregexes.  EOR
17:32 benabik joined #phasers
17:32 Util Hello
17:32 mikemol joined #phasers
17:33 TimToady o/
17:33 mikemol o/
17:33 mikemol Just a fly on the wall. :)
17:33 * TimToady too :)
17:33 colomon \o
17:34 Util \o
17:34 TimToady actually, in my case, the Walls have ears
17:34 mikemol I *really* want to get around to implenting NQP in C++, but that's not going to happen any time soon. But it's something I actively ponder a lot. :)
17:39 TimToady porting Perl 6 to C++ could make it a super gaming engine, so you maybe you need to find some young gamer to do it in his basement when he should be studying :)
17:40 mikemol I said NQP, not Perl 6. You've still got NOM to worry about, there. :)
17:40 TimToady 6model should be able to handle C++ style objects pretty well, eventually
17:42 * TimToady suspects that most of the poeple who are serious about #phasers can't attend today
17:42 TimToady is there anything serious that anyone wants to discuss?
17:42 mikemol Perl6 to C++ has at least a couple snags. First is eval(). Second is that C++ classes aren't mutable at runtime. You can modify instance data, but you lost the ability to modify the classes the moment you compiled.
17:42 * PerlJam is belatedly here
17:42 * mikemol lets the real discussion continue. :)
17:43 Util As pre-reported, looking for feedback on Perl 5 to Perl 6 translator: https://github.com/Util/Blue_Tiger
17:43 PerlJam all of the pre-reports look like "I won't be at #phasers because of $work or $school"
17:45 colomon Util: I'll try to take a proper look at it sometime soon.  Util++
17:45 sorear aaagh
17:45 sorear hi, I'm here, I thought #phasers was in 45, accursed DST
17:46 TimToady for a moment, thought someone had stabbed you
17:47 sorear did: discussed unicode stuff a bit with TimToady (who wants to see case, space, and hyphen sensitive <:General_Category<Other_Letter>> (or is that <:Gc<Lo>> ?)) and khw
17:47 sorear started work on a C# Unicode Character Database library
17:48 TimToady \o/
17:48 Util colomon: thanks!
17:48 * TimToady wonders if it shouldn't be in Perl 6 someday
17:48 sorear will do: more unicode stuff, planning to talk to the TR18 folks about their interpretation of grapheme mode
17:48 mberends joined #phasers
17:48 sorear very few tuits this month.
17:48 sorear EOR
17:49 sorear PerlJam: I didn't make it because I overslept
17:49 TimToady or in the C# subset of Perl 6...
17:51 TimToady sorear: btw, I'm okay with simple short/long aliases, or at least okayer than with casefolding
17:52 TimToady the long ones could require declaration, if there's overhead
17:57 diakopter oh; I didn't realize #phasers was now
17:57 diakopter it's supposed to be 9:30 PST 10:30 PDT?
17:58 PerlJam 1730 UTC
18:00 mberends o/
18:00 mberends sorry for late arrival
18:01 mberends report: gave a great talk about Niecza, Gtk and Tetris at London Perl Workshop. .EOR :)
18:03 colomon mberends++
18:06 Util mberends++
18:06 mberends I'll commit the slides into niecza/docs :)
18:09 Util I will certainly read them!
18:10 sorear me too!
18:11 colomon So, I promised to push this: anyone have a feel for what needs to be done to get module tools working in nom and niecza?
18:11 colomon Right now it feels like there's a lot of cool stuff going on, but the core state of Perl 6 has regressed eighteen months or so....
18:12 PerlJam colomon: could you define "module tools" for me?
18:12 colomon pies / panda / ufo / whatever
18:13 PerlJam thanks
18:13 TimToady colomon: the core state of rakudo has regressed, but I don't see much regression in other implementations...
18:13 colomon PS we also need to figure out how to allow people to install both niecza modules and rakudo modules at the same time.
18:13 colomon TimToady: niecza never had those tools working at all.  That's not a regression for niecza, but it's not helpful, either.
18:18 sorear \o/ mberends
18:19 Util A roadmap of all the current pieces of the puzzle would be helpful.    pies / panda / ufo, what is deprecated, which does what part of the job, and which fails on what implementations.
18:19 sorear +1
18:19 sorear and _what the job is_
18:19 colomon +1
18:20 PerlJam something like http://perl6.org/compilers/features.html but for modules and applications would be nice
18:22 colomon sorear: at the big picture level, I'd say the job is to be able to A) easily install modules and B) easily run their test suite.  Also would be good to be able to C) easily create new modules for the ecosystem.
18:23 mberends _what the job is_ almost deserves to be specced, as an addition to https://github.com/rakudo/rakudo/blob/nom/docs/S11-Modules-proposal.pod and maybe moved to another location.
18:24 [Coke] can we  add the "works with" <impl> <latest version> info to  http://modules.perl6.org/ ? That seems like a useful place for it.
18:24 PerlJam [Coke]++
18:25 colomon [Coke]: that's a great idea, but right now we don't have any way to generate that information other than running each module by hand.
18:25 Util Tornado warning here. Signing off; will backscroll later.
18:25 [Coke] colomon: Have to start somewhere, yes.
18:26 TimToady Util++ gives new meaning to the term "duck typing"
18:28 diakopter Alabama/Georgia border?
18:28 colomon diakopter: yes
18:28 colomon TimToady++ # had to read that one three times before I got it, but then I laughed very, very loudly.  :)
18:33 colomon I wish tadzik++ were here, as he has a clear idea what panda (pies?) needs done to run on nom.  That would be a good start, I guess.
18:51 colomon oh, hey, one other issue for the p6 community: are we going to try to get some p6 tasks into the TPF part of the Google Code-In?
18:54 PerlJam that would be cool
18:55 PerlJam I can't think of many "high school level" problems though.
19:01 colomon here's a list of the p5 ones: http://wiki.enlightenedperl.org/gci2011/gci2011/ideas
19:01 colomon I'd say there are probably lots of open p6 issues easier than the medium difficulty / high difficulty tasks there
19:06 diakopter what is perl5i
19:07 PerlJam It's something that tries to be perl6-like but keep everything in a perl 5 frame (if that makes enough sense)
19:07 sorear http://search.cpan.org/~mschwern/perl5i-v2.9.1/lib/perl5i.pm
19:07 colomon https://github.com/schwern/perl5i
19:15 sorear TimToady: how doy ou feel about <:gEnErAlCaTeGoRy<Lo>>?  not strictly a perversion of smartmatching to fuzzily match the _keys_
19:17 sorear (slight reminder: property names and values (except Uppercase_Mapping and suchlike, which you shouldn't be using directly in regexes anyway) are restricted to printable ASCII characters, whereupon case-folding is simple and linguistically well defined
19:22 TimToady I'm okay with canonicalizing those
19:24 TimToady but I still don't think that casefolding has much place in programming, being steeped in Unix culture
19:28 * sorear out
19:29 * jnthn back
20:23 * tadzik is here
20:24 jnthn o/ tadzik
20:25 tadzik ohai
20:25 jnthn tadzik: I'm working on protoregexes atm.
20:25 tadzik \o/
20:25 tadzik supercool
20:25 jnthn tadzik: Hopefully once that's back, we can get close to having JSON::Tiny back.
20:25 colomon tadzik: lots for you to comment on in the backlog here.  ;)
20:25 tadzik are there any unanswered questions about the ecosystem zoo?
20:26 jnthn tadzik: Mostly, I want to know what - other than protoregexes - we need.
20:26 tadzik to be perfectly honest, I never tried pies/panda on niecza
20:26 tadzik jnthn: nothing from the things I could catch at compile time
20:26 tadzik jnthn: I'll grab something to eat, fake some ecosystem data and see how the whole thing works
20:26 jnthn tadzik: OK, great
20:27 jnthn tadzik: Getting that, then NCI back up are my main goals for the week or two ahead.
20:28 tadzik I'm thinking sometimes about my 6model-ish parrot object wrapper
20:28 tadzik s/my/my idea of/
20:28 tadzik but that's another story
20:28 jnthn :)
20:28 colomon tadzik++
20:30 tadzik for now it's nom time
20:30 jnthn enjoy

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