weekly Perl 6 status meetings with phase transitions: Wed 17:30 UTC | IR clogs at irclog.perlgeek.de/phasers/today
Set by moderator on 4 December 2011.
jnthn Probably will make at least the first bit of #phasers...here's a preport 15:49
* Some 6model design work, especially on representing arrays and generalizing the notion of representations beyond just objects
* A refactor to bring things in line with the design
* Worked on CArray representation
* Made TypeObj but SomeRole now work
* Fixed a couple of bugs in parametric roel handling
* Implemented typed hashes
* Partially implemented types arrays
* Made it so only a Callable can be assgined to a my &foo
* Refactored type handling; we now enforce sigil-implied roles on binding
* Ported .WALK, getc, IO.x from ng and enabled various tests
* Implemtned ff, ^ff, ff^ and ^ff^
* Merged nom version of Zavolaj into master, fixed win32 example, updated todo list
Plans
* Got a lot of $dayjob things over the next week, which may limit how much Perl 6 time I get 15:50
* Get CArray usable in NativeCall
* Work on NativeCall tests
* Resolve more regressions and bugs in nom
Worries
* Not sure I'll have net access from 21st to just after Christmas. Certainly won't have a lot of time (family stuff). So really need to work out the nom/nomstar release in most senses before then.
* Also could do with volunteers for actually doing the releases
EOR
btw, I can probably give 17th, 18th and 20th over to Perl 6 things and not have distractions
Maybe next Tuesday also.
17:20 colomon joined
colomon pre-report: 17:28
* Did Day One Advent post
* Added crude but usable user interface to Gtk-mandelbrot
* Cheered wildly when sorear++ added threading to Gtk-mandelbrot
* Pointed out "is cached" as a useful traits demo
* Implemented "is symmetric" in nom with much help from jnthn++
* Started to add chord support to ABC
* Wrote emergency Advent post, to use when we don't otherwise have a post
* In the process, got rounders.t fully working in Niecza, and added more tests to it
plan:
* Continue polishing Gtk-mandelbrot (improving threading performance would be a big plus)
* Write Advent post on Gtk-mandelbrot
* Try to figure out how to output chords into Lilypond (it's incredibly awkward)
EOR
o/ 17:31
jnthn o/ 17:36
Wow, lonely here tonight. 17:37
17:38 benabik joined
jnthn Well, if there's nothing to discuss, it means I can go for dinner sooner ;) 17:39
colomon so, I didn't see your pre-report, but it's safe to assume you rocked. jnthn++
are we really ready for nomstar? 17:40
(backlogged)
colomon doesn't have any idea how the current spectest passes compare to b's 17:42
we certainly do seem to be doing a lot of cool stuff we weren't before, on the pro side
jnthn colomon: Panda works again, aside from some test fails get in the way of its bootstrap. NativeCall almost works enough for everything it did before, which'll mean we get mysql back. We're within 1000 spectests of before. And there's 2 weeks of development time left.
colomon any idea what's needed to fix the bootstrap? 17:43
jnthn Or close to being within 1000 or so
colomon: stop Test::Mock failing its tests :)
colomon: Panda itself actually works.
colomon tadzik++ 17:45
17:46 felher joined
tadzik hello 17:50
colomon \\o 17:51
jnthn o/ tadzik 17:52
tadzik report: won the battle of Austerlitz; plans: advent blogging, random hacks in random time, nothing in particular
17:54 diakopter joined
jnthn
.oO( O(random) )
17:56
tadzik possibly some Dancer-related stuff 17:59
jnthn OK, seems its really quiet here today, and I'm hungry, so I'm gonna go nom :)
tadzik I feel a subtle need for that :)
from the outside world
PerlJam hungers too
jnthn ...are you talking about Dancer or food? :)
tadzik Dancer, Dancer
speaking of food, I'll be home no sooner than 1.5 hours :/ 18:00
(lectures on 6 PM)--
jnthn will be home in about 2 days...
;)
tadzik :)
jnthn I let my students go before 6 PM. :)
ok, food time. bbl
18:39 benabik left
sorear helo 18:50
[Coke] o/ 18:51
moritz peeks in for a short moment 20:50
I've looked at random nom regressions in roast 20:51
and foud a few common causes for most of the failures
s/most/some/
regex: combinations of % and :s
regex: :i
custom operators
chained series operators
take [1, 2] flattens
missing i Complex constant
autovivification
presence of protos does not imply 'multi' on un-qualified subs
moritz goes to bed 20:52
jnthn moritz: thanks
22:20 Tene joined