weekly Rakudo status meetings with phase transitions: Tue 19:00 UTC | IR clogs at irclog.perlgeek.de/phasers/today
Set by moderator on 5 March 2011.
04:29 cognominal joined
sorear PRE REPORT: 07:29
DID:
* New model for sandboxing: no more SAFE.setting. Dangerous subs are marked "is unsafe" and are patched out by the loader. 07:30
* Tene++ recommends hardening the OS jail and doing away with language sandboxing in p6eval entirely
* gave moritz++ a commit bit 07:31
* Parcels support ?~+ prefix ops by stringifying
er, by listifying
* made tiny progress towards a 6modelish separation of concerns in the object system 07:32
* fixed up Threads module
* added general Monitor and ObjectPipe classes to Threads - I'd like to see some cool stuff done (moritz++ talked about wanting to play with object pipes a while ago, iirc) 07:33
* niecza only recompiles if the source files are newer by at least 2 milliseconds, to work around stupid rounding errors in the CLR double<->string converters; speeds up builds massively on Windows 07:34
* added "class A::B" syntax, and support for "use A::B" as well; NieczaPassBegin.pm6 et al should go away soon 07:35
* my, has, state, and anon variables can now have type constraints
* .= operator added
WILL DO:
* try to get yapsi running. (goal for March)
* try to finish the STable/HOW split 07:36
BLOCKERS:
* none
EOR
07:47 colomon joined 08:12 mariolone joined, mariolone left 09:33 mariolone joined, mariolone left 12:18 [Coke] left 12:35 _ilbot left 12:36 _ilbot joined
moderator weekly Rakudo status meetings with phase transitions: Tue 19:00 UTC | IR clogs at irclog.perlgeek.de/phasers/today
diakopter PRE REPORT: EOR (again) 18:26
er, actually I did fix a slight omission in S32/Str and worked on printf for a while in niecza; it's pending/WIP 18:27
Util Pre-report: Added code to build a binary .dmg package of R* for OS X. Code review welcome. Actual binary will be available for testing later today. EOR 18:51
PerlJam Util: do you need to be on OSX to build it? 18:53
Util PerlJam: Yes. 18:54
I expect that other users will write procedures/code to package for Ubuntu, Redhat, Win32, etc. While it may be possible to merge all the Linux processes into one build stream, the Win32 and Mac processess will have to be run on their respective platforms. 18:55
18:56 ashleydev joined
colomon \\o 19:00
Util Hello
colomon well 19:05
I guess if Util++ has already reported, and no one else has shown up, I'm up next. :)
Util Please, whenever you are ready :) 19:06
colomon Basically, I got smash++'s benchmark harness running on my Linux box, generated up-to-date benchmark numbers, and then cobbled together some graphs with moritz_++'s help.
I've now got some suggestions from sorear++ for improving the graph format. 19:07
I've also started working on additional benchmarking of masak's contest's p5 problem, trying to see how the codes do when they are given less extreme input. 19:08
EOR
Util Are the current graphs publicly visible from the web?
colomon yes 19:09
justrakudoit.wordpress.com/2011/03/...k-results/
oh, I should add that the graphs suggest we've had a couple of performance regressions in the last two months. I haven't had time to look into it further yet. 19:10
Util colomon++ 19:12
diakopter colomon++ Util++ 19:17
colomon diakopter: good point! Util++ for OS X work!!! 19:18
diakopter Util: I have no clue who has been doing the windows packages of parrot/rakudo that go to sourceforge... does anyone here know? 19:20
Util jnthn, I think, did the first (and only) Win32 .msi for R*; it is available on GitHub. 19:22
I had lost track of the SourceForge uploads, and am thrilled to see that they are current. 19:23
Looks like fperrad is doing the work there. 19:24
diakopter oh
colomon fperrad++ 19:25
Util fperrad++ indeed!
colomon where's there a link to those? 19:36
(the windows .msi, I mean) 19:37
Util colomon: sourceforge.net/projects/parrotwin32/files/ 19:38
colomon thanks!
PerlJam If the sourceforge .msi is current, maybe we should put it on github.com/rakudo/star/downloads ? 19:42
Util I suspect that the sf .msi is of Rakudo, not Rakudo Star.
PerlJam oh, I thought it was R* 19:43
Util I have not verified; I just suspect
If it *is* Star, then I agree (to put it on GitHub.../downloads) 19:44
PerlJam In any case ... what can we do to help fperrad++ build MSIs for R* ? :) 19:45
diakopter email him?
colomon Is there an actual Rakudo in there? All the filenames I just saw looked like parrot stuff. 19:51
PerlJam it looks to me like he's naming the files parrot-rakudo 19:53
where parrot is the parrot version and rakudo is the rakudo version
for example: Parrot-3.1.0-Rakudo-20110215
He's just explicit about the combinations 19:54
Util I think email will be best; he has not spoken in #parrotsketch since Nov 2009, but his last commit to parrot was 3 days ago, so he *is* active. 19:56
A preview R* Darwin binary is ready for inspection - s3.datasaw.com/Rakudo_Star_2011-01_...ment_3.dmg 20:16
tadzik hrm, did we forgot about #ps or it's just that no one wants to start or what? :) 21:34
besides those few who have reported 21:35
colomon tadzik: errr.... you're like two and a half hours late. 21:46
but please feel free to report. :)
tadzik well, I was not reporting for I didn't see any "official start now" :) 21:51
colomon 19:00 UTC, I give an opening wave, and Util says Hello. 21:55
We seemed to be the only ones around. :)
tadzik my bad then :)
bah, schade :) 21:56
colomon usually there are about five waves, of course, but we were short-handed. ;) 21:57
tadzik that reminds me of a PM meeting when only 2 of us shown up, and the other guy was half an hour late :) 22:03
moritz_ is back 22:09
sorry, forgot to submit a report... but it wasn't very exciting anyway 22:10
22:24 [Coke] joined