|
weekly Rakudo status meetings with phase transitions: Tue 19:00 UTC | IR clogs at irclog.perlgeek.de/phasers/today Set by moderator on 8 February 2011. |
|||
|
01:07
cognominal joined
|
|||
| sorear | DID | 07:40 | |
| Was asked by pmurias what I thought niecza's niche should be. I have no answer ... | |||
| Added Z and X (meta)ops | 07:41 | ||
| Redesigned TODO lists for niecza and hopefully made them friendlier | |||
| Added here document de-indenting | |||
| Took a bunch of minor items off the TODO - callsame, %(), constant @/%, @.foo, return $x, $y | 07:43 | ||
| Implemented the control exception -> goto optimization | |||
| Made qw[] and related forms work in more cases | |||
| Made ? quantifier not imply a list | |||
| Started fudging spectests | 07:44 | ||
| WILL DO | |||
| Focus a lot more on spectests | |||
| Try to knock down a couple of the "Medium"s from TODO | |||
| BLOCKERS | |||
| Tuits, TODO ideas | 07:45 | ||
| EOR | |||
| (now sleep for real.) | 07:46 | ||
|
08:07
diakopter joined
09:38
jnthn left,
jnthn joined
13:16
[Coke] left
13:23
[Coke] joined
17:56
moritz_ left,
eternaleye left,
diakopter left
18:06
moritz_ joined,
eternaleye joined
18:10
diakopter joined
18:16
lue joined
18:28
takadonet joined
18:32
lue left,
lue joined
18:35
pyrimidine joined
|
|||
| Util | Pre-report: No forward progress; all tuits were eaten by server crash (now fixed). EOR | 18:35 | |
| colomon | Pre-report: Blog post on simplifying fractions, started implementing gcd on local copy of Rakudo. Running behind on BigInt / FatRat project because $work suddenly got very busy. EOR. | 18:38 | |
| moritz_ | colomon: can't the public gcd sub simply call pir::gcd? | 18:46 | |
| anyway, my report: hacked in the new nqp repo. Ported some code from PIR to nqp. Incorporated my new command line parser. | |||
| I would port some more code, but I'm a bit afraid I might slow down parsing signficantly | 18:47 | ||
| since nqp doesn't do natively typed variables yet | 18:48 | ||
| EOR | |||
| I might miss the actual #phasers meeting, depending on $daughter | 18:49 | ||
|
18:49
mberends joined
|
|||
| Util | Pre-queued question; regarding: Perl 6 .batch method being replaced with .munch method, or sometimes by .[^$n]. | 18:51 | |
| Around August 2010, there was IRC traffic and repo commits about all this. | |||
| I think that the work must be incomplete, because the S07 still contains .batch (plus a reference to .batch in S03), and no spec contains .munch. | |||
| Correct? Thoughts? Is this on anyone's worklist? | 18:52 | ||
| Also, I cannot find the discussion of the *why* of batch/munch change. Any pointers? | |||
| EOQ | |||
|
18:54
pyrimidine left
|
|||
| moritz_ | Util: I think there's a difference... one of those methods is destructive, the other not | 18:55 | |
| but I can't remember the specifics | |||
| pmichaud++ should know | |||
|
18:55
pyrimidine joined
|
|||
| jnthn | pmichaud++ at some point should get the docs in sync too :) I think that's the plan, anyways. | 18:56 | |
| Util | moritz_: Thanks. I was holding off of that issue ( .batch !== .munch due to destruction ), hoping that it would be answered by a pointer to the discussion. Good to know, though. | 18:57 | |
|
18:57
[particle]1 is now known as [particle]
|
|||
| PerlJam | I thought the name was changed just to emphasize the destructive nature. | 18:57 | |
| .munch consumes, .batch is not clear on that | 18:58 | ||
| moritz_ | ah right, that must have been it | ||
| lue | REPORT: Didn't do anything remotely P6 since about Jan. 31. :/ | 19:01 | |
| PLANS: Continue learning C++ for my P6-on-GCC project, get my Pod6 parser out the door by the end of the month. | |||
| EOR | |||
| jnthn figures it must be the real #phasers time now :) | 19:04 | ||
| mberends | \\o | ||
| jnthn | o/ | ||
| jnthn haz a report ready | |||
| Here comes :) | |||
| I missed last week's #phasers - sorry. Main news I shoulda posted then: | |||
| * NQP gets roles | |||
| * $?CLASS is generic, as should be (Rakudo gets that wrong) | |||
| * Did lots of thinking about roles and how to factor them in 6model generally, which will help doing the full-blown Rakudo ones | 19:05 | ||
| This week wasn't especially productive... | |||
| * Fixed DispatcherSub cloning in NQP, which will fix some SEGVs that were reported | |||
| * Added "is parrot_vtable" which works through the meta-model to allow Parrot v-table overrides in 6model | |||
| * It builds a flattened array indexed by v-table slot, so lookups are fast | |||
| * Used it to map the NQPMu.Str method to get_string, so say(SomeClass) works again | |||
| * Sketched out how I think a bunch of things in the Rakudo-on-6model will look. Ranges from "pretty confident" to "wild guess" :-) | |||
| * Wrote NOMMAP - the ROADMAP for the nom branch | |||
| Should have more time this week. Want to... | |||
| * Get the new NQP referenced from Rakudo's nom branch | |||
| * Get the class and native meta-objects to build | |||
| * Get grammar and actions buildable with the new NQP | |||
| * Gut stage 1 compiler to be just meta-objects + grammar + actions | |||
| * Get some very basic setting to build (e.g. with Mu in it) | |||
| * If time, get proto and multi bits in place | |||
| * If time, work on static lexpad bits | |||
| .EOR | |||
| mberends | wow! jnthn++ | 19:06 | |
| lue | .oO(You're a fast typist!) |
||
| PerlJam | Util: .munch probably comes from discussion starting at irclog.perlgeek.de/perl6/2010-06-04#i_2403514 | ||
| Util | PerlJam: Thanks! Just what I was looking for | 19:07 | |
| moritz_ | anybody else got a report? | ||
| mberends | a tittle bittle | 19:08 | |
| * no work done, if that's what you're thinking ;) | 19:09 | ||
| * pulled out of FOSDEM because of family vacation opportunity | |||
| * will talk at perlworkshop.nl | 19:10 | ||
| * will host a p6 hackathon at home 6-8 march | |||
| colomon | moritz_: yes, gcd is a one-line function. It's just a matter of carefully getting it into the code, running all the tests, adding tests (which I've already done), etc. Just an example of how short my time has been. :) | ||
|
19:11
lue left
|
|||
| mberends | * contact me if you think you can come to the Netherlands around those dates. | 19:11 | |
| * will talk at yapc::ru 14-15 may | |||
| .eor | |||
| jnthn | Stroopwafels and Perl 6 hacking...what could be better? | 19:12 | |
|
19:13
lue joined
|
|||
| mberends | and a fridge full of pivo :) | 19:13 | |
| jnthn | Oh, it just got better! | ||
| PerlJam | jnthn: beer? | ||
| jnthn | \\o/ | ||
| jnthn will be there :) | |||
| mberends does the preparatory beer sampling for quality control | 19:14 | ||
| jnthn | mberends++ # testing | ||
| mberends | test driven drinking! | 19:15 | |
| PerlJam | I guess that's better than drinking driven tests | 19:16 | |
|
19:16
colomon left
|
|||
| jnthn | One note on the Rakudo nom branch. In the next few days, I'm going to start tearing things up a bit in there, so if you find it unbuildable or useless in the next week or two, don't be so surprised. :) I do plan to merge changes from master in reasonably often, so there's no reason to cease doing improvements there - unless they're specifically related to the object model. | 19:17 | |
| If that changes, I'll be sure to mention it. I think we'll have to wait and see whether we actually merge nom or if we say "master is now beta and nom is now master", as we did with the ng branch. | 19:19 | ||
| moritz_ | sounds sane | 19:21 | |
| jnthn | Either way, it should be a much, much less painful transition. | ||
| At the moment, I expect the main issues we'll see hit people are changed multi semantics, my vs our scoping bits, and anyone who was relying on the MOP in a way that won't carry over. | 19:23 | ||
| moritz_ | jnthn: are natively typed attributes available in nqp syntax yet? | 19:26 | |
| jnthn | moritz_: Not yet | ||
| moritz_: I'll have to do that soonish. | |||
| moritz_ | jnthn: do you know any LHFs that I might port from PIR to NQP? or generally features that I can implement without deep wizardry? | 19:27 | |
| jnthn | I guess giving you the "switch Match to 6model" task would be too cruel, though you can have a go if you want. :) | 19:28 | |
| Let me look for other bits :) | |||
| I didn't catch up on what you've ported so far... ;) | |||
| Oh! | 19:29 | ||
| One thing you maybe could look at is in NQPMu.pm | |||
| Setting up some kinda CREATE/BUILD/bless style constructor thingy. | |||
| It's not so much a porting task as a "write it" task, but it doesn't involve writing PIR at least :) | 19:30 | ||
| moritz_: The other thing that would be mighty helpful is to look at the install target and make sure it installs stuff to names that don't conflict with nqp-rx. | |||
| Dunno if any of those tasks sound interesting? :) | 19:31 | ||
|
19:35
masak joined
|
|||
| masak seems to have missed tonight's phasers | 19:36 | ||
| I don't have much of a report anyway. I'm blogging a little and writing far too little Perl 6 code at present. :) | |||
| moritz_ | jnthn: I'll take a look | ||
| jnthn: the installing thing doesn't sound too interesting, but the rest does | |||
|
19:37
takadonet left
|
|||
| jnthn | moritz_: Heh. The installing thing isn't interesting for anyone. :) | 19:37 | |
| I probably block on it Real Soon Now though, so it may well fall to me if nobody else touches it. The *best* person to do it would be Pm. :) | |||
| github.com/perl6/nqp/issues | 19:38 | ||
| I've put some issues there and will add more LHF with time. | |||
| As well as some MHF and HHF and AF :) | 19:39 | ||
| masak | issues. jnthn haz 'em :P | ||
| jnthn | :'( | ||
| :P | |||
| masak hugs jnthn :) | |||
| moritz_ | jnthn: if it's as simple as prefixing some names with a new prefix, that works | ||
| jnthn | moritz_: That's what I was thinking of for a first take, yes :) | ||
| PerlJam | or suffix | 19:40 | |
| nqp-nom for instance : ) | |||
| er, :) | |||
| jnthn gets more issues :) | 19:44 | ||
|
20:22
pmichaud_ joined
20:27
cognominal left,
pmichaud left
20:28
cognominal joined
20:43
mberends left
20:56
mberends joined
21:08
mberends left
21:39
takadonet joined
21:58
pyrimidine left
22:03
masak left
|
|||