|
weekly Perl 6 status meetings with phase transitions: Wed 17:30 UTC | IR clogs at irclog.perlgeek.de/phasers/today Set by moderator on 30 September 2011. |
|||
|
03:10
ashleydev joined
|
|||
| sorear | DID: much more hacking on /serialize. Freeze/thaw is now mostly working; niecza can compile and run modules without dependencies (that means -L NULL). The setting can be compiled but cannot yet be used | 16:41 | |
| EOR | 16:42 | ||
| erm, WILL DO: implement dependencies (loading and linking), generally chase regressions | |||
|
17:08
masak joined
|
|||
| sorear | o/ masak | 17:10 | |
| masak | preport: blogged macros intro. ate, thought, and slept macros. not much actual coding, though. plan to ameliorate in the coming week. EOR | ||
| sorear | masak: did your grant get accepted? | 17:12 | |
| colomon | Pre-report: | 17:15 | |
| * Added close, Cool.rand, pick, and unlink to niecza. | |||
| * Lots of related spectest fudging (plus atan2.t fudging) for niecza | |||
| * Have a fully working version of my Tune Reminder project: github.com/colomon/TuneReminder | |||
| * Simple p6 classes to implement a running log of tunes I have practiced and suggest which ones to practice next | |||
| * Simple Gtk user interface (screenshot: github.com/colomon/TuneReminder/bl...minder.png ) | |||
| * I've actually started using it as a practice tool. :) | |||
| * Blogged on proposed ways to create FatRat literals | |||
| * Tried to convince TimToady that having rand xx 10 call rand ten times was a bad idea | |||
| * (I mean, I think it's a lovely facility to have, but it makes the equally reasonable case where you actually only wanted one call to rand uglier, and does so in a way which is not consistent with other related operators) | |||
| Plans: | |||
| * More polishing on Tune Reminder | |||
| * Possibly a blog post with a fuller (and probably still futile) explanation of why I don't like that rand xx 10 thing | |||
| * Look around for more LHF in niecza if I get the chance | |||
| EOR | 17:16 | ||
| moritz | What I did: | ||
| * a few small Rakudo patches | |||
| * applied patches | |||
| * wrote a file with BEGIN tests | |||
| * tried to reason about why BEGIN doesn't closure as other blocks (ie tried to understand the last section of S04). Only partial success | 17:17 | ||
| * helped a bit with IO::Select | |||
| * tested macros a bit for masak++, found a Null PMC access | 17:18 | ||
| * used native ints a bit in the rakudo setting | |||
| What I plan to do: | 17:19 | ||
| * exception stuff. Really :-) | |||
| EOR | |||
| wil miss phasers, but backlog | |||
| tadzik | pre-report: noticed Parrot's Socket.poll, implemented unspecced IO::Socket.poll, example in gist.github.com/1296356 Also successfully invited moritz++ to hack on Select | 17:21 | |
| [Coke] | PR: DID: more rakudo queue wrangling and niecza/rakudo fudging, t/spectest.data updating. Someday I'll actually write code. :P | 17:22 | |
| masak | sorear: no, but they're gonna discuss it on the 23rd. | ||
| sorear: I have a draft about the specifics of D4 in the works, since both commenters and TPF asked about it. | |||
| jnthn | Preport: | 17:23 | |
| Week was mostly swallowed by $dayjob. Fixed a few bugs, landed the optimizer branch and blug. | |||
| Good news is that the $dayjob workfest is largely over, so I'll have tuits again. Should be able to largely give Monday and Tuesday to Rakudo work, and should have a bit of time tomorrow or Friday too. | 17:24 | ||
| No specific plans; expect to pick off a few more RTs, maybe hack a little more on the optimizer, maybe add some missing feature(s). :) | 17:25 | ||
| EOR | 17:26 | ||
|
17:28
mls joined
|
|||
| colomon | o/ | 17:30 | |
| jnthn | o/ | ||
| PerlJam | greetings. | 17:31 | |
| pmichaud | greets | ||
| jnthn | o/ pmichaud | 17:32 | |
| TimToady lurques | |||
| PerlJam is at home with a sick kid today | |||
| mls | Report: mostly did parrot hacking (this led to a 12% startup speedup). Am looking at implementing stacked lexpads. EOR | ||
|
17:33
diakopter joined
|
|||
| sorear | moritz: the last section of S04 has a very oldspec feel and I wouldn't trust it too much | 17:33 | |
| o/ diakopter | |||
| diakopter | hi | ||
| jnthn | mls: "stacked"? | 17:34 | |
| mls | I don't want to do a call for every block with lexicals | ||
| PerlJam | my report: did -- nothing. will do -- Rakudo compiler release tomorrow. EOR | 17:36 | |
| diakopter | my report: I was bitten by the p6 bug again; sprixel was resurrected (again) as an interpreter in C. starting on 6model porting today. | ||
| sorear | niecza release on the 31st | ||
| I've also been helping diakopter a bit | 17:37 | ||
| masak | "p6 bug" -- which one? :) | 17:38 | |
| SCNR | |||
| [Coke] read "the p6 bug" and thought: which one, there's hundre... oh. | |||
| diakopter | heh | ||
| jnthn | :P | ||
| [Coke] | (*&#$; caught up with masak! | ||
| colomon | sorear: will you merge the serialization into master before niecza's release, do you think? | ||
| sorear | colomon: likely | ||
| colomon | \\o/ | ||
| PerlJam | Apparently we all thought the same thing about sorear's report ... not sure that's a good thing. | 17:39 | |
| er, s/sorear/diakopter/ | |||
| diakopter | oh, I also quieted blackberry chatmosphere users from #perl6 | ||
| jnthn | yay! | ||
| colomon | diakopter++ | ||
| masak | \\o/ | 17:40 | |
| tadzik | "it's like a thousand of voices suddenly cried in terror" | 17:43 | |
| TimToady | "and then were silent, apparently..." | 17:47 | |
| mls | jnthn: if you have some spare time there's also my enter/leave branch... | 17:49 | |
| jnthn | mls: ah, yes. | ||
| mls: I will have time to look at that in the next couple of days. | |||
| Well, I should :) | |||
| mls | awesome. | ||
| jnthn | mls: Were there Parrot changes to make this possible? | 17:50 | |
| mls | No, only rakudo changes | ||
| jnthn | ah, ok. | ||
| if 1 { LEAVE { say 'x' }; die }; CATCH { say 'y' } | 17:51 | ||
| What's the output of this on the branch? | 17:52 | ||
| mls | should be y, x, then the exception | ||
| (but I'm currently at home, will tell you tomorrow) | 17:53 | ||
| ( the CATCH block rethrows the exception in case you were wondering...) | 17:54 | ||
| jnthn | yeah, I was mostly concerned that we ran the handler before unwinding. | ||
| mls | hmm, could also be y, exception, x... | ||
| sorear | no, it couldn't | 17:55 | |
| mls | no, handlers are only run when unwinding | ||
| sorear | oh I guess it could | ||
| jnthn | :) | ||
| sorear | we haven't specced exactly when unhandled exceptions are printed | 17:56 | |
| TimToady | it better run handlers before unwinding | ||
| sorear | y must proceed x, that is for sure | ||
| jnthn | TimToady: What does the "no handler" case count as? :) | ||
| TimToady | it should effectively be an outermost CATCH | ||
| jnthn | OK. | ||
| mls: So there's your answer :) | |||
| TimToady | that will unwind everything and then exit | ||
| sorear | TimToady: but does it print in the CATCH body (before unwind) or outside the block (after unwind)? | 17:57 | |
| jnthn | TimToady: Will it print the message first, then unwind? | ||
| TimToady | but it would be the choice of that CATCH whether the die or the x came out first | ||
| in general, CATCH comes before unwind, even if this one is different | 17:58 | ||
| so I think probably the x should come after | |||
| mls | I think my current code implements it that way | 17:59 | |
| sorear | TimToady: is the extreme similarity between Perl6 exceptions and Windows SEH a matter of convergent evolution? | ||
| TimToady | well, possibly we're both converging on Lisp :) | 18:00 | |
| for some definition of Lisp | |||
| masak | according to smug Lisp weenies, everyone is... :) | ||
|
18:07
benabik joined
|
|||
| Util missed #phasers meeting, but had nothing to report anyway. Will backscroll. | 18:52 | ||
|
19:02
benabik left
19:09
diakopter left
19:43
[Coke] joined
19:45
diakopter joined
19:46
diakopter left
23:04
[Coke] joined
23:10
[Coke] joined
23:19
[Coke] joined
23:25
[Coke] joined
|
|||
