Perl 6 - the future is here, just unevenly distributed

IRC log for #phasers, 2011-04-26

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

All times shown according to UTC.

Time Nick Message
04:33 [particle]1 joined #phasers
04:36 [particle] left #phasers
09:49 birdwindupbird joined #phasers
13:34 birdwindupbird left #phasers
13:41 pmurias joined #phasers
13:47 [particle]1 left #phasers
13:56 [particle] joined #phasers
14:42 birdwindupbird joined #phasers
14:43 birdwindupbird left #phasers
14:57 pmurias left #phasers
15:02 birdwindupbird joined #phasers
15:12 [particle]1 joined #phasers
15:13 [particle] left #phasers
15:20 [particle]1 is now known as [particle]
16:55 birdwindupbird left #phasers
17:55 masak joined #phasers
18:06 _ilbot joined #phasers
18:06 Topic for #phasers is now weekly Rakudo status meetings with phase transitions: Tue 19:00 UTC | IR clogs at http://irclog.perlgeek.de/phasers/today
18:12 colomon joined #phasers
18:54 takadonet joined #phasers
18:59 Util Pre-report: Nothing P6-ish done, but I now have a working laptop. Plan: R* binary for OS X. EOR
19:02 colomon \o
19:02 Util o/
19:02 colomon Pre-reporting one minute early, eh?
19:02 Util :)
19:05 Util I embrace the pre-reporting decision of #parrotsketch, and enourage others to do the same in #phasers. I am just not very timely :)
19:05 masak o/
19:05 jnthn o/
19:07 [Coke] left #phasers
19:08 masak report: Nil. EOR
19:08 * jnthn is writing his still...
19:09 [Coke] joined #phasers
19:11 colomon report: Nothing until today.  Today, blog post, trivial fix to spectest.data, update from TimToady for blog post code, code for next blog post done.  Future: That stuff, plus prepping / giving Perl 6 presentation at Penguicon on Saturday.  EOR
19:12 moritz oh, forgot my "future" part: continue speccing errors
19:13 jnthn ok, here's mine...
19:13 masak oh, same here. I'm trying to get a blog post or two out the door. it's hard work, but I expect to make it eventually.
19:14 jnthn Over the last week, I hacked a little on NQP... ;-)
19:14 jnthn * Roughly 100 commits - see https://github.com/perl6/nqp/commits/master
19:14 jnthn * Totally re-worked packages
19:14 jnthn * The package really does hang off .WHO now - that's the authoritative source for a package
19:14 jnthn * All installations and lookups work through this mechanism
19:14 jnthn * This means that lexical packages Just Work now
19:14 jnthn * Realized that implementing is export and lexical UNIT::EXPORT will also be easy
19:14 jnthn * Separate compilation also works - every module gets a clean GLOBALish and we reconcile them at load time
19:14 jnthn * In compile-and-run mode, the set of packages the compiler maintains is the same one that we use at runtime. In compile-and-pir mode, we rebuild it at "deserialization time", though it's not really serialization yet.
19:14 jnthn * We now locate various objects during the compile time, such as the parent classes and roles and meta-objects, and add them to the compile-time created meta-object instance. Again, same one that we use at runtime in compile-and-run mode.
19:14 jnthn * Wrote a blog post on 6guts
19:15 jnthn Tasks for the week to come...
19:15 jnthn * Create attribute meta-objects at compile time and add them to the meta-object
19:15 jnthn * Detect undeclared attributes during the compile and whine about them
19:15 jnthn * Natively typed attributes
19:15 jnthn * Work out how adding methods to the compile time meta-object will work, given they're not compiled yet
19:15 jnthn * Whatever else I have time for
19:15 jnthn EOR
19:15 birdwindupbird joined #phasers
19:16 masak jnthn++
19:18 jnthn Any other reports? :)
19:19 colomon jnthn++
19:21 jnthn Oh, I forgot to mention, I also accidentally implemented a first cut of supplying custom meta-objects for class/grammar/role etc :)
19:21 moritz jnthn++
19:21 masak o.O
19:22 jnthn masak: https://github.com/perl6/nqp/blob/master/src/how/EXPORTHOW.pm
19:22 jnthn Actually even the default ones for NQP aren't baked in
19:22 jnthn The meta-objects library exports them to the setting, which is how they get found :)
19:22 masak nice.
19:22 jnthn (No pun intended.) ;)
20:02 tadzik oh, I'm late!
20:03 tadzik darn
20:04 jnthn off-by-one? ;)
20:04 tadzik no, the PM meeting, I completely forgot about #phasers :(
20:04 jnthn Aww
20:04 jnthn But PM meeting is a good excuse ;)
20:04 tadzik but I don't think I have anything to report anyway, besides GSoC of course, but that's nothing to be talking about so far
20:11 masak tadzik: tell us your thoughts about it.
20:11 masak what's the first test you'll be writing? :)
20:13 tadzik load.t for the POD classes :>
20:13 tadzik I still think about what is the way to test them so the tests will make sense
20:14 tadzik surely I'll look how other implementors have done it
20:16 masak test behaviours. that way, you don't have to mock a lot, and you'll get a rich API.
20:16 tadzik test behaviours?
20:16 tadzik I may appear to be thinking a bit slow, I've been celebrating a bit :)
20:17 masak yeah, there's a way to do something to an object, and then check with getter methods that the change had some effect.
20:17 masak try to avoid that ;)
20:17 tadzik oh that's a bit stupid :)
20:17 masak much better if the object actually *responds* somehow.
20:17 tadzik that's rather testing how classes work in Perl 6 :)
20:18 masak aye.
20:18 tadzik well, the only thing making sense to test is where the class is something more than just a container
20:18 masak a Pod parser is basically a filter. filters are very well-suited for testing.
20:19 jnthn tadzik: I'll be very happy to assist with the .WHO wiring when you get there, fwiw. :)
20:19 masak not to mention the .HOW wiring ;)
20:19 tadzik masak: well, the Pod parser. But the Pod Nodes, the representation?
20:19 tadzik it's a prerequisite for the parser, and that's the first target of mine
20:19 masak tadzik: hm. I guess they'd need to have getters.
20:20 masak tadzik: but much more important to test whether they serialize correctly, etc.
20:20 tadzik oh, right
20:20 tadzik I need my "gsocmess" repository
20:23 tadzik there we go https://github.com/tadzik/gsocmess :)
21:22 masak \o/
22:09 masak left #phasers
22:58 birdwindupbird left #phasers

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