% rdice has joined #parrotsketch
% pmichaud has joined #parrotsketch
% kid51 has joined #parrotsketch
% jonathan has joined #parrotsketch
% particl1 has joined #parrotsketch
% particle has left particle!~particle@c-24-19-3-148.hsd1.mn.comcast.net
% smash has joined #parrotsketch
* smash hello everyone
<kid51> Howdy, smash, you fine fellow!
* kid51 had to say it 'cause purl's not here.
% barney has joined #parrotsketch
% allison has joined #parrotsketch
<pmichaud> okay, no Coke or chromatic here today
<allison> let's see... barney, jonathan?
<barney> yep
<allison> anything to report?
<barney> hacked on Eclectus
<barney> .eor
<allison> jonathan mentioned he was on his way out the door
<allison> particle, pmichaud?
<pmichaud> I haven't seen particle this morning, so I'll go ahead and report
<pmichaud> parrot:
<pmichaud> * updated pbc_to_exe to eliminate compiler warnings, speed improvements
<pmichaud> * allow pbc_to_exe and the files it creates to run from any directory
<pmichaud> * pbc_to_exe is still tied to the build directory, though -- fixing this will require cleaning up parrot's "make install"
<pmichaud> pct:
<pmichaud> * fixed the --target=pir option for interactive mode
<pmichaud> nqp:
<pmichaud> * applied fixes from Tene for hashes and arrays
<pmichaud> perl6:
<pmichaud> * watched others (particle, jonathan, coke, etc.) add new features to perl6
<pmichaud> * refactored documentation, updated README, submission guidelines, STATUS
<pmichaud> * added docs/compiler_overview.pod
<pmichaud> ** I'm hoping compiler_overview.pod can be a start for writing generic compiler tools documentation
<pmichaud> * added 'module' keyword and namespaces
<pmichaud> * refactored infix:<~~> (smart matching)
<pmichaud> * added infix:<//> (defined-or)
<pmichaud> * wrote some Perl 6 articles for perlbuzz, journal posts
<pmichaud> * added a 'make perl6' target to generate a standalone executable with pbc_to_ex
<pmichaud> e
<pmichaud> * added @*ARGS support, so hanoi.pl now runs
<pmichaud> * refactored some RT tickets
<pmichaud> * created a metaticket for high-priority perl6 items
<pmichaud> * eager to see Coke's spiffy new online RT summarizer
<pmichaud> plan:
<pmichaud> * initiate official test suite refactoring and review
<pmichaud> * write two articles about adding features to perl6
<pmichaud> * update PCT and perl6 to use 'copy' opcode
<pmichaud> * update perl6 ROADMAP
<pmichaud> blockers:
<pmichaud> * fglock is blocking on dynamic :outer  (RT#49182)
<pmichaud> * many new hackers and compiler users will probably need RT#43269, or a workarou
<pmichaud> nd
<pmichaud> EOR
<allison> busy couple of weeks! :)
<allison> jonathan?
<allison> (he mentioned he was still here)
<jonathan> * Gave a lightning talk about implementing Perl 6 given statement, then checked it in at the end of the talk, at the Israeli Perl Workshop.
<jonathan> * Started digging into the junctions implementation. Much left to do, but also quite a bit done.
<jonathan> * Other little tiny things in languages/perl6/
<jonathan> * Auto-threading of junctions on sub calls is next up.
<jonathan> * Not much Parrot core work, but I will try and get back into that soonish.
<jonathan> EOR
<allison> smash?
<allison> tewk?
<smash> no report soz, bu i have a question
<allison> kid51?
<kid51> Yes several points
<kid51> A lot of refactoring of the configuration system which, if done correctly, should mostly be invisible to other Parrot developers.
<kid51> Most prominently, revised the subs used by config/auto/*.pm to build and run little C programs which probe the OS for config values -- such as cc_build, cc_run -- now are methods in Parrot::Configure::Compiler and must be invoked as, e.g., $conf->cc_build().  These methods are inherited by Parrot::Configure object.  So now, when these methods are invoked they have access to all data stored in Parrot::Configure object at
<kid51>  time of invocation.
<kid51> In many config step classes, I've refactored the guts of runstep() out into internal methods/subroutines and have written tests directly for them in t/configure/1??-*.t.  This boosts the test coverage considerably (see http://thenceforward.net/parrot/coverage/configure-build/coverage.html).
<kid51> Coke showed me how to search for all tickets which I've taken (as distinct from those I opened).  This has enable me to go through several old tickets and resolve them.
<kid51> There's a guy who works in NYC (and who is not named Dan!) who is showing interest in the project.  I'm hoping to be able to achieve F2F hacking with him.
<kid51> Next week will be attending a BSD users meeting.  Would like to recruit smoke testers for BSDs besides FreeBSD and Dragonfly.  Cf.:  http://rt.perl.org/rt3/Ticket/Display.html?id=49276.
<kid51> EOR
<allison> allison?
<allison> - Spent the past 2 weeks removing the tendrils of the old event system so I could turn it off. Timers, sleep, NCI callbacks, and thread suspension for GC now on new concurrency scheduler.
<allison> - Ended up solving some mod_parrot thread-related crashes in the process, which was a nice side-effect.
<allison> - Checked in changes with a few intermittently failing tests to get broader platform testing and review. Thanks to all for your reports.
<allison> - Currently working on implementing one more significant piece of PDD 25, which should resolve the remaining failing tests.
<allison> - Updated the debian packaging metadata in the repository.
<allison> - Wrote a debian packaging guide.
<allison> - Wrote a guide to setting up a chroot environment for testing and generating packages in various versions of Debian and Ubuntu.
<allison> EOR
<allison> any other reports before questions?
<leo> minireport
<leo> * fixed the scheduler.c breakage and made it compile on Win32 with Jonathan
<leo> * will try to find some time slots and round tuits for parrot dev in 2008
<pmichaud> leo++ (welcome back)
<leo> thx
<allison> leo++
<allison> questions, smash you had one?
<smash> yes, anything o i can do to help that debian package thing ?
<allison> read through the chroot instructions and packaging guide and try them out
<allison> that will help catch any steps I didn't document
<smash> ok
<allison> I'd like it to be as repeatable as possible
<allison> (similar to our standard release guide)
* smash nods
<allison> other questions?
<kid51> re:  Our smoke test reporting system
<kid51> It's < optimal, IMHO.
<kid51> I'd like to be able to search it by tester, by individual test, etc.
<allison> the submission process, or the interface to review them?
<kid51> Who, if anyone, is "in charge" of it?
<kid51> Can we recruit someone to overhaul it?
<kid51> allison:  Main concern is interface and what's behind it -- but, then again, my own smoke was submitted and not displayed.
<kid51> I'd like to get more info out of it a la cpan.testers
<allison> certainly we can recruit someone for improvements
<allison> someone was working on it at YAPC::EU, but haven't heard from them since then
<kid51> I assume it's hosted at perl.org -- but where is the code housed?
<pmichaud> (hosted at perl.org) yes
<barney> wasn't that tools/util in the Parrot repo?
<barney> smokeserv-server.pl
<pmichaud> barney++
<pmichaud> there's even a README :-)
* barney suggests a tools/smoke dir
<allison> barney: agreed
<pmichaud> are there enough items to warrant a separate smoke/ dir?  (I'm not opposed, just asking)
<leo> is the smokeserv-server.pl on a machine, where folks can change that file?
<pmichaud> I suspect it's wherever the other parrotcode.org files are
<allison> kid51: are you offering to find a volunteer?
<allison> pmichaud: those are in the perl.org combust repository, all flat templated html
<kid51> allison:  I think that, after studying what's in tools/util/, I'd like to start a wiki page that would be an evolving spec for a better smoke server
<allison> kid51: sounds good
<allison> I think that wraps up that one. other questions?
<allison> Okay, back to coding. Happy New Year everyone!
<pmichaud> happy 2008, all
<leo> -""-
% pmichaud has left #parrotsketch
* kid51 nods at #parrotsketch and gives him a small kipper
<barney> a guats neis
% kid51 has left #parrotsketch
% chromatic has joined #parrotsketch
% barney has left barney!~bernhard@p549A0B8E.dip0.t-ipconnect.de
% chromatic has left #parrotsketch
% allison has left #parrotsketch
% jonathan has left #parrotsketch
% rdice has left rdice!~richarddi@CPE001217e365c7-CM00159a01d44c.cpe.net.cable.rogers.com
% rdice has joined #parrotsketch
% rdice has left rdice!~richarddi@CPE001217e365c7-CM00159a01d44c.cpe.net.cable.rogers.com
% particl1 is now known as particle