weekly Rakudo status meetings with phase transitions: Tue 19:00 UTC | IR clogs at irclog.perlgeek.de/phasers/today
Set by moderator on 8 June 2010.
02:24 colomon joined 12:26 bphillips joined, bphillips left 13:16 spinclad joined 15:29 bphillips joined 15:34 ash_ joined 16:04 ash_ joined 17:34 eternaleye joined 17:55 eternaleye joined 17:59 ash_ joined 18:14 eternaleye joined 18:57 masak joined
masak o/ 19:00
sorear Hi.
jnthn oh hai
sorear FOR ONCE, I make a #phasers 19:01
spinclad \\o
jnthn \\o/
jnthn waits a moment to see it pmichaud or colomon or moritz materialize 19:02
masak writes some Perl 6 in the meantime :) 19:03
jnthn Guess I can paste my report. :-) 19:04
* Hacked with mberends++ at the weekend
* Dug into S12 and S14 triage - trying to get the last remaining bits alpha did back in place
* Handles is done, which is nice - much happier with the implementation this time
* WALK passes all tests from master and more; closed 3 tickes from that
* Got *.foo curried at compile time
* Helping mmcleric++ work on improving whatever handling further
* Won't be a very productive next few days - doing family stuff
* Speaking in Belgium on Saturday
* No travel in July, so should be good for hacking. 19:05
</rapport>
masak :)
jnthn masak: Anything to report? How's Buf going? 19:06
masak I had a bit of a dip in activity last week.
but now I seem to be back in the saddle. 19:07
implemented .[] today; will be looking at different encodings next.
also want to look at pls in the next few days, since that's probably going to be part of R*.
ditto Web.pm. 19:08
jnthn How are we doing against the GSoC schedule?
masak looks
schedule's here: gist.github.com/360097
one milestone was yesterday:
2010-06-21 Completed Str.encode/Buf.decode. Started on reading and writing binary files. 19:09
so, I'm "mostly done" with Str.encode/Buf.decode. they do work for UTF-8.
need to expand them into other Parrot-supported encodings.
now that .[] is implemented, all the previous milestones are done, too. 19:10
so I'd say mostly on track.
.eor
[Coke] .
moritz_ surfaces 19:11
jnthn masak++ # thanks
m\\o/ritz_: Anything to report?
moritz_ well, not all that much 19:12
I'm trying to organize the book and other contributors
patched a few LHFs in rakudo
blogged about Math::Model with great success (~5k readers)
I don't get much actual hacking on rakudo done :( 19:13
.eor
masak oh, and this week's rakudobug count: 12. slightly above average. :)
jnthn masak: It didn't go unnoticed. 19:14
masak :>
jnthn moritz_: Working on nice modules is a Good Thing. :-)
ON pls 19:15
Is there general agreement that i'll be what we use for module insallation etc in R*?
er, it'll :-) 19:16
sorear wait, we have the technology to duplicate jnthn and include copies in r* tarballs, now?
jnthn *installation
sorear: That'd be a quite horrifying download. :-)
$ tar -xfz rakudo-star.tar.gz
jnthn extracted...your beer is all gone! 19:17
moritz_ jnthn: I don't think we need proto/pls for installing modules in rkakudo start
*star
masak +1
moritz_ we can just walk the module folders in a given order
and execute the make files (which we can either ship, or generate with ufo)
jnthn aye, true 19:18
It is a convenience I guess.
masak but surely there are modules which we don't ship with Rakudo Star?
jnthn Alas, one people may well expected.
moritz_ right
masak I think there should be some easily accessible (as in "you'll stumble over it") documentation showing (1) how to install a module manually, and (2) how to do it with pls. 19:19
jnthn +1 19:21
masak: Do you think pls stands a good chance of being in a usable state by R*? What can @other do to help with it? 19:22
masak I hope to have it in a usable state this week. that's almost a promise.
I'll put together my proof-of-concept tonight.
jnthn Ah, that's well ahead of schedule then. :-)
19:23 eternaleye joined
masak then people++ can help me hack on it, when they see how it fits together. 19:23
jnthn OK.
[Coke], sorear: Anything to report/add? :-) 19:24
Or ask...
jnthn figures we're probably missing pm and colomon today due to YAPC::NA fun :-)
sorear no 19:25
[Coke] Still planning on doing the release - also, we'll be picking your brainz after R* for more "what you need from parrot" items.
masak I have an impertinent question. are there any items on the list at use.perl.org/~masak/journal/39597 that we believe/know will not be fixed by R*? 19:26
jnthn [Coke]: OK. Conveniently, I expect various Rakudo devs - myself and Pm included - will meet at YAPC::EU, which is shortly after R*, to discuss the way forward.
[Coke]: So I suspect the results of that will very much guide what we'd like to see most in Parrot.
moritz_ masak: there's a patch for --version, it just needs review from pm 19:27
masak moritz_: good to hear.
jnthn masak: How up to date are the (Update:)s?
masak jnthn: mostly. 19:28
not more than five are un-updated.
moritz_ but for example i don't expect adverbial blocks for R*
jnthn masak: rt.perl.org/rt3/Ticket/Display.html?id=65962 is for example done for a while :-) 19:29
masak ok.
jnthn: thanks, updating. 19:30
jnthn rt.perl.org/rt3/Ticket/Display.html?id=67372 also
Oh, it's not marked resolved 'cus it needs tests.
But the issue is fixed.
masak anyway; I'm mostly very hopeful wrt R*, after last week's meeting, and after pmichaud++'s full return to blazing activity.
jnthn masak: I think the others still stand. 19:31
I think we've a good chance of resolving the majority of outstanding ones. 19:32
The command line ones should all be relatively straightforward, I imagine.
masak \\o/
this list is as of September 2009, mind. there are probably a few new unfixed egregorious ones. 19:33
jnthn It's a good list to have, anyways.
masak s/egregorious/egregious/ :)
moritz_ I always found the mixture of PCT, NQP and dont-know-what-else that does the command line parsing quite opaque
jnthn moritz_: ugh.
moritz_ but maybe I'm just missing the magical hint to the right documentation
jnthn moritz_: I'm tring to find the blog post you wrote a while ago with Rakudo * things too, but missing it. 19:35
moritz_ jnthn: I'll dig it up for you
jnthn oh, found it
moritz_ perlgeek.de/blog-en/perl-6/rakudo-s...-list.html 19:36
jnthn Thanks
moritz_ /<foo>/ is still open
closure cloning
jnthn Yeah, various of them are.
A couple of those in the nice but not essential list are likely on my plate.
Both stand a good chance of getting my attention in July. 19:37
esp the user defined operators hiding candidates one.
moritz_ one that's not on my list, but would be nice to have, is topicalization of $_ on the RHS of ~~ 19:38
masak +1 19:39
jnthn moritz_: *nod*
moritz_: I'm confident we'll have that one for R*.
It'd win back a bunch of S03 tests, I imagine.
Main issue is that I'm I'm not sure how to do it elegantly. :-)
19:41 colomon_ joined
colomon_ o/ 19:41
sorry I'm late
jnthn \\o/ 19:42
colomon_: No worries
colomon_: Feel free to report what you got up to this week. :-)
colomon_ well, YAPC
I'm tearing apart the trig tests right now, too. 19:43
been working on it in the background during the conference.
I currently feel I have a solid plan to have fewer tests and test more things, but it is proving a bit painful to write the code to generate the tests.
earlier I got ACCEPTS working for Numeric and Real, and make == work properly again. 19:44
and before that, I did some series work. we're pretty close to the spec there now, I think. 19:45
masak colomon_: will you commit the code to generate the tests somewhere?
colomon_ masak: as always, it is in S32-trig. 19:46
masak \\o/
colomon_ :)
masak didn't know that until now
colomon_ I updated it yesterday (without actually updating the tests), but I'm working on bigger changes now.
I'm hoping to get the trig tests happy, then check in the new ones all at once, so there's one big flux in the total test count instead of several. 19:47
jnthn Sounds sensible. 19:49
colomon_ making good progress, but I've been hitting frustrating Rakudo limitations all afternoon.
masak colomon_++
jnthn colomon_: Any in particular?
masak limitations?
colomon_ too many digits in my numeric constants, trying to do two lazy xx * operations at the same time, etc.
jnthn colomon_: Can always play masak^W^Wfile tickets. :-)
colomon_ I'm sure these are known.
masak those two are files.
*filed
jnthn ah, OK 19:50
colomon_ Two xx * is just the closure bug yet again. It just snuck up on me.
19:50 eternaleye joined
colomon_ and too many digits is old and troubling.... 19:51
gotta run to the keynote. o/
masak it'll be *so* nice when the closure bug is... closed.
jnthn colomon_: Enjoy o/
masak: Indeed.
OK, any more, or is our work here today done? :-)
moritz_ raises hand 19:52
I'm currently thinking about how we do the shipping of Rakudo Star
I currently envision a perl script that git-clones rakudo and some modules 19:53
and puts them into a dist/ dir
and then I want to write a do-it-all build script that's copied into dist/
which then in turn first builds rakudo, then all the modules 19:54
does that sound sane?
jnthn I was thinking more than everything would be in the tarball
spinclad build and install?
jnthn *that 19:55
moritz_ jnthn: I plan to put all of dist/ into the tarball
jnthn moritz_: ah, I see
moritz_: Then yes, sane.
To me anyway
But would be good to know what pmichaud++ thinks too.
masak sounds sane, yes. 19:56
moritz_ I'll commit something soonish 19:57
jnthn moritz_++
having something to play with would be great.
moritz_ you now have something to play with. 20:03
the script is noisy, but it's not user-facing 20:04
jnthn Great. :-)
I'll try and have a play with it soon. 20:05
In the meantime, let's adjourn to #perl6 :-) 20:06
Thanks all
masak \\o/
20:07 masak left 20:12 colomon joined 20:39 ash_ joined 20:41 [Coke] joined