% PerlJam has joined #parrotsketch
% allison has joined #parrotsketch
% chromatic has joined #parrotsketch
% pmichaud has joined #parrotsketch
% woggle has joined #parrotsketch
<chromatic> Hello all.  I am wearing my temporary jesse pants today.
<chromatic> Allison, you're first up alphabetically.  What's new?
* chip tries to imagine jesse pants
<allison> I finished off the TGE refactors I was working on last week.
<allison> TGE can now generate complete grammars in PIR from the .tg source files.
<pmichaud> allison++
<allison> (no need for intermediate PIR files)
<allison> I also update Punie to use the feature.
<allison> Now I've put a hold on TGE refactors for a while to work on Punie some more.
<allison> (Though, I'm already finding that further planned TGE refactors would be handy in Punie...)
<allison> In Punie I'm working on variables and assignment.
<allison> Nearly have that finished.
<allison> Subroutines are next and I hope to have them finished this week.
<chromatic> Are you blocking on anything?  Are you having fun?
<allison> I'd like to have as much finished as possible before I give a talk on it in 2 weeks. :)
<allison> I am having fun.
<allison> Not blocking on anything except time.
<allison> EOR
<chromatic> Thanks, Allison.  Chip?
<chip> back
<chip> So.  My diabolical plan continues with extra evil, as I plan to still review those pdds by two weeks from today.
<chip> Real life hates me, and I hate it back.  :-P
<chip> I'm not blocking on anybody else's work, only on the flow of archetypal ideas and images from my own cranium.
<chip> ^D
<chip> [oh, and the fun ratio is high when RL isn't interfering]
<chromatic> Excellent.  chromatic?
<chromatic> Well thanks, not-jesse.
<chip> [I also have a quick question for the language implementors in the house]
<chromatic> Leo's suggestion of parrot -G makes all of the Pheme tests run when run directly.   Of course, there are still multidispatch errors and segfaults on various tests when running through make test.
<chromatic> But that does make it likely it's a GC issue.
<chromatic> That's sort of blocking me now, but I'm also short of time with a couple of other projects, so it's not awful.
<chromatic> That's it.
<chromatic> Thank you, c.
<chromatic> Okay, Leo?
<leo_> * patched TGE to not use properties at all for .id handling
<leo_>   (cleaner code, less resource usage and faster)
<leo_> * was at APW 2006 http://use.perl.org/~domm/journal/29873
<leo_> * tried to fix chromatic's GC issues - no success til yet
<leo_> EOR
<chromatic> Are you blocking on anything?  Are you having fun?
<leo_> not blocking and having fun (the usual RL terms apply, though)
<chromatic> Excellent.  Matt?
<allison> leo, thanks, that will all be replaced in a coming refactor as we're going to proxy objects over tacking an id onto existing nodes, but it's a nice intermediate step
<chromatic> Hm, we'll come back to Matt.  Nicholas?
% Coke has joined #parrotsketch
<chromatic> No Nicholas either.  Next up, Jerry.
<particle> it's been a quiet parrot week for me. at this point i'm thinking real life will interfere with any serious progress until the chicago hackathon.
<particle> i've had PDD review on my task list for some time
<particle> if i see chip make progress on text, i'll look it over, and also pay attention to format
<particle> waiting for a response from patrick on a perl6 parser issue, but not blocking
<particle> .end
<chromatic> Having fun?  Blocking?
<particle> fun, yes. blocking, no.
<chromatic> Great!  Jonathan?
% mdiep has left mdiep!~matt@c-24-18-232-4.hsd1.wa.comcast.net
<particle> there's no jonathan here
<particle> oh, there is!
<chromatic> He seems somewhat silent though, so... Patrick?
<pmichaud> hi, all
<pmichaud> first, apologies for my extended silence -- rl issues intruded in a nasty way (house purchase fell through)
<pmichaud> things are cleaned up now, I have limited net access until Thursday
<pmichaud> as a result, no new items to report for this past week
<pmichaud> tomorrow I have an entire day in front of a computer (sans net) to do p6/parrot stuff, which will be nice
<pmichaud> my expected plan is to write p6 versions of the TGE transformations -- i.e., to specify the TGE rules using p6 code.  This will primarily be for documentation initially, but also to help others understand what is going on
<pmichaud> when we have a mini-p6 available, it may become the official transformation code
<pmichaud> tomorrow I will also be writing information for people who are interested in helping at the hackathon, and expect to publish it thu or fri
<pmichaud> thursday I will be available all day irc and code to work on parrot
<pmichaud> I just received word that my second oscon presentation was accepted (on parsing in parrot) so I'm looking forward to preparing that
<pmichaud> thursday I plan to get arrays working in p6
<allison> pmichaud: as soon as you have a routine I can call to pass in mini-p6 and get back PIR, I can build that in as a syntax option for TGE
<pmichaud> we'll see how things go after that.  real life issues will keep me a little busy on saturday and again on monday, but other than that I'm likely to be heavily p6-focused through the end of the hackathon
<pmichaud> allison:  it'll be based on compreg, undoubtedly
<allison> indeed :)
<pmichaud> end of report
<chromatic> Thanks, Patrick.  Fun?  Blocked?
<pmichaud> blocked only on real life time; looking forward to having fun again once I'm past that :-)
<chromatic> Alrighty.  Anyone else an Official Implementor with an Official Implementor Badge?
<leo_> is woggle here?
* chip brandishes his Chairman Larry Decoder Ring
<woggle> Yes.
<chromatic> Alright, it's woggle then Will.
<woggle> Okay, so I implemented enough STM backend that single-threaded stuff works mostly as expected.
<woggle> I also revived the old threading code which had broken when calling conventions changed.
<woggle> To do that, I added slurpy support to NCI and flatten support to the va_list calls. (On the stm branch only at this point.)
<woggle> The old threading code will probably need interface work as the thread type distinction are not implemented and it's not clear that they will be -- and if they are, it's not clear exactly what the semantics would be.
<woggle> I'm also going to need to look at sanely makign shared read-only things, but I guess I'm going to have to talk on the list for that.
<woggle> EOR
% chromatic has left chromatic!~chromatic@sub17-30.member.dsl-only.net
% chromatic has joined #parrotsketch
<chromatic> Thanks, woggle.  Will?
<allison> (we're having network trouble, which is why chromatic and I keep dropping off and on)
<chromatic> Hm, don't hear anything from Will.
<chromatic> Chip, you had a question for Implementors?
<chip> swapped out.  hm.
<chip> ah, yes.  keys vs. arrays.
<chip> the question is: Are 'key' pmcs convenient enough to work with?  I'm thinking that since some of the namespace ops use them, we should just use them consistently for all namespace operations where I'd spec'd using arrays instead.
<chip> If they're not convenient enough, they probably should be made more convenient anyway.
<chromatic> It's a lot easier to create keys from literals than arrays....
<particle> i think jonathan worthington might have ideas about that, chip
<leo_> yes - because keys were inside parrot earlier (the only PMC constant with special cased handling overall)
<leo_> I'm in favor of creating constant arrays ('lists') instead
<leo_> in the long run at least (when r/o interfaces are available)
<chip> er
<chromatic> I'd have to play with the syntax and semantics of dealing with namespaces for which I don't know the complete name at compile time to give a better answer.
<chip> OK, I"m getting a definite "maybe".  :-)
<chip> works for me tho
<leo_> a key is just a specialized PMC constant, with extra packfile/pbc code
<chromatic> Alrighty.  Did I miss anyone?  Are there any other questions?
% Jonathan has joined #parrotsketch
<particle> oh, hey, nobody mentioned that jonathan committed the .net => pbc converter this week
<chromatic> You're hardly a nobody, particle!
* particle wonders how chromatic switches pants so quickly
<chromatic> Velcro.
<chromatic> That seems like a wrap for this week.  Thank you all.
* particle sings YMCA
% chromatic has left #parrotsketch
<pmichaud> back, all -- was disconnected by starbucks :-(
* pmichaud reviews log
% Jonathan has left #parrotsketch
% pmichaud has left #parrotsketch
<chip> --starbucks++
% mdiep has joined #parrotsketch
% Coke has left Coke!~wcoleda@host-69-95-19-3.roc.choiceone.net
% mdiep has left mdiep!~matt@c-24-18-232-4.hsd1.wa.comcast.net
% Coke has joined #parrotsketch
% particle has left #parrotsketch
% particle has joined #parrotsketch
% Coke has left Coke!~wcoleda@64.241.37.140
% Coke has joined #parrotsketch
% Coke has left Coke!~wcoleda@cpe-72-228-52-192.nycap.res.rr.com
% allison has left allison!~chatzilla@sub17-30.member.dsl-only.net
% Coke has joined #parrotsketch
% mdiep has joined #parrotsketch
% woggle has left #parrotsketch
% Coke has left Coke!~wcoleda@cpe-72-228-52-192.nycap.res.rr.com
% Coke has joined #parrotsketch
% Coke has left Coke!~wcoleda@cpe-72-228-52-192.nycap.res.rr.com
% Coke has joined #parrotsketch
<spinclad> ie, { temp starbucks -= 1 }
<chip> until it's time for another fix
<spinclad> so that's decrement starbucks one coffee, credit to thee?
* spinclad just had a hot choco there on the way home