% particle_ has left particle_!~particle@144.81.84.141
% rafl has left rafl!~rafl@armitage.dreker.org
% rafl has joined #parrotsketch
% particle has joined #parrotsketch
% obra has left obra!~jesse@4.79.3.63
% obra has joined #parrotsketch
<tewk> Report:
<tewk> I'm going to miss parrotsketch today.
<tewk> Haven't done a lot.
<tewk> I have been working on ruby classes and modules.
<tewk> I'm not blocking on anything.
<tewk> I am having problems debugging PGE,  I need to get some help from pmichaud when he has some time.
<tewk> EOR
<obra> ok. what's the right way for him to help you?
<tewk> Actually I gust 
<tewk> just figured it out.
<obra> cool!
% allison has joined #parrotsketch
% Jonathan has joined #parrotsketch
<obra> Jonathan: generally the way things on #parrotsketch work is I go around the room, picking on people. they paste in a report of what they've been up to for the past week, what they'll be working on soon, what's giving them trouble and so on
<obra> are you up for reporting today?
<Jonathan> sure :)
<obra> excellent. Welcome to the Parrot shop ;)
<obra> er. pet shop. whatever 
<particle> i'd like to buy a parrot.
<obra> sorry. we're all out
<PerlJam> surely there's a dead one around somewhere
<particle> you leave me no choice... i'll have to build one myself!
<obra> Ok. Parrot Sketch roll call. Who's around?
% chromatic has joined #parrotsketch
* chip hops up and down, then fixes his chair
* leo_ is here
<obra> ok. today's ordering: tewk (I'll replay), chip, allison, leo, particle, chromatic, pmichaud (if he's here), whomever else pops up that I've missed
* Jonathan waves
<obra> (sorry. I thought I'd typed "Jonathan")
<obra> I'm good at that :/
<obra> Jonathan: why don't you go right after I replay tewk:
<obra> 12:48 <tewk> I'm going to miss parrotsketch today.
<obra> 12:48 <tewk> Haven't done a lot.
<obra> 12:48 <tewk> I have been working on ruby classes and modules.
<obra> 12:48 <tewk> I'm not blocking on anything.
<obra> 12:49 <tewk> I am having problems debugging PGE,  I need to get some help from pmichaud when he has some time.
<obra> 12:50 <tewk> EOR
<obra> 13:23 <obra> ok. what's the right way for him to help you?
<obra> 13:31 <tewk> Actually I gust
<obra> 13:31 <tewk> just figured it out.
<obra> 13:32 <obra> cool!
<obra> So, now I'd like to welcome Jonathan Worthington, who's been hacking on .net -> Parrot
<chip> <applause>
<Jonathan> <bow>
<obra> How's the hacking?
<Jonathan> good
<Jonathan> I'm all enthusiasmed from YAPC :)
<Jonathan> reportage...
<Jonathan> * Tweaks to .Net translator to move away from old (find|store)_global ops
<Jonathan> * Get BCG building on Win32 (thanks to vsoni for fixes on other platforms)
<Jonathan> * Planning to look over BCG more in the coming week
<Jonathan> * Sent proposal to list about reference PMCs, waiting thoughts
<Jonathan> Also taken on writing the first draft of the bytecode PDD
<Jonathan> Will have something together in the next week on that.
<Jonathan> Other than minor odds and ends, that's about all.
<obra>   Very cool.
<obra> Anything you're blocking on?
<Jonathan> Just feedback on the reference PMCs stuff, but got plenty to do while I wait for that.
<obra> cool
<obra> chip: you're up
<chip> Last week:
<chip>   * ENOPARROT for the most part
<chip>   * partially updated pdd07 (coding style), under optimistic assumption I
<chip>     could knock it out and never have to think about it again
<chip> Coming week:
<chip>   * finish new pdd07 (coding style)
<chip>   * plus, the balance of last week's list, in approx. order by priority:
<chip>      - resolve default packfile question (per chromatic & leo)
<chip>      - review leo's PMC reorg proposal
<chip>         ... and see if eliminating STRING* is a Good Thing
<chip>      - do, or add to RT, items from OSCON meeting w/Allison
<chip>      - perhaps start on exceptions & dynamic scoping;
<chip>         these share a lot of concepts, and perhaps infrastructure
<chip>      - in background, make & keep Parrot compilable as C++
<chip>         ... for integration with C++ systems, and for lintish reasons
<chip> 100 RETURN
<obra> blicking oin anything?
<obra> blocking, anything
% Coke has joined #parrotsketch
<obra> allison? how's it going?
<allison> Mainly spent the week on the Parrot compiler tools talk and traveling.
<chip> oopsie, sorry, meatspace interruption
<allison> I had some good conversations at YAPC::EU, including some with Leo and Jonathan (the results of the latter were in Jonathan's report, so I won't repeat).
<chip> no not blocking on anything, pls ga allison
<allison> At the Parrot hackathon, I updated the Punie grammar to precompile the operator precedence rules. (Oh yeah, I need to check that in.)
% smash has joined #parrotsketch
<allison> Also poked at a PGE bug with smash (hi smash!) at hackathon. I need to sum up our results and add them to the ticket.
<allison> EOR
<obra> blocking on anything?
<allison> a bit of quiet time to churn through the IO PDD
<obra> ok.
<obra> leo?
<leo_> * YAPC::EU::2006 is regretfully history
<leo_> * some parrot-related BOF thingies are at: http://use.perl.org/~leo/journal/
<leo_> * more stuff from YAPC:
<leo_> - jonathan is about implementing weak refs to registers and refs to array/hash elemements in parrot core (these are in dotnet already but generally useful)
<leo_> - allison's PGE/TGE talk was very impressive, and we wanted to do some APL/PGE BOF after vaxman's talk about APL, but didn't find a common time slot
<leo_> * GSoC has finished, I've submitted the final evaluation survey
<leo_> * I'm currently improving httpd.pir a bit. I want to use it as the base for my slides presentation software in future - the old one got boring ;-)
<leo_> * ah yes and: YAPC::EU::2007 will take place in Vienna/Austria
<leo_> ^D
<obra> particle: you're up
<particle> another light week...
<particle> committed an example grammar for IO language
<particle> helped coke++ with a JSON parser grammar (which replaces an earlier (and unused) pre-PGE parser i wrote)
<particle> currently dealing with some 'none' changes which affected pge
<particle> waiting on compiler tools refactor (i wish it was a branch, so i could follow and help with work-in-progress)
<particle> thinking about compiler tools test suite (obviously need some design details before this progresses too far)
<particle> goto NEXT
<obra> chromatic: how's things?
<chromatic> Allison and I spent some time discussing segfault prevention, specifically in relation to where and how the default packfile gets initialized.  No consensus.
<chromatic> I've made no real progress on the embedding interface, except reinforcing some of my views.  I hope to get those down in the PDD this weekend.
<chromatic> I'm thinking of how to port Class::MOP to Parrot, but I haven't written even any  experimental code yet.  I think it's worth considering how the Perl 6 layout of packages and modules and namespaces and classes might work for Parrot.
<chromatic> I'm blocking on Pheme while waiting for the new TGE code and the compiler object.  Is there anything I can do to help that along?
<chromatic> I'm also waiting on particle's Win32 test results before checking in Embed::Parrot.
<chromatic> That's it.
<obra> mm. still no patrick.
<obra> Coke: you're up
<Coke>   - add first stab at parrothist.pod
<Coke>   - more C89 cleanup
<Coke>   - more work on JSON
<Coke> that's it.
<obra> what's next for you?
<Coke> get JSON working fully. then back to tcl.
<obra> cool
<obra> And we're to me.
<obra> * Reaction to ponie's demise has been..quite muted. 
<obra> * I'm working with a friend at Rational to try to get a couple Purify licenses for the project
<obra> Question time?
<chip> obra++ purify++   Question about compiler tools refactor: Would a branch be suitable?
<chip> anything in svn can be seen by many more eyes
<particle> i think chromatic and i (at least) would like to see a branch fro that
<chromatic> Definitely.
<chromatic> Less black-hole code is good.
<chip> OK, just name it something with a number because this won't necessariliy be the only refactoring of the compiler tools
<particle> i'll make sure patrick reads today's log
<chip> question on the parrotcode.org svn repo: I can't commit, is that on purpose or by accident?
<chip> (in this case it's because pdd07 =~ s#clip/##)
<particle> talk to robrt. must have combust, and perhaps other prereqs. coke may know more, also
<chip> will ask robrt & coke
<chromatic> Why make Will do it?
<chip> will will!
<obra> because he has experience managing it, presumably
* particle was thinking big picture. it's better to have more than one parrotcode.org committer imo
<obra> Next?
<chromatic> Chip, while you're thinking of the default packfile, Allison and I had philosophical questions about where to prevent segfaults.
<chromatic> It'd be nice to have some general rules about where to check for error conditions.
<chip> I'll include that in pdd07
<chip> My feeling is that C has too many pointer values to clutter up all code with checks
<chip> (emphasis on *all*)
<chromatic> Specifically, can we guarantee that the interpreter will always be in a non-segfaultable state after Parrot_init() or should we add pointer checks through macros, inline functions, or regular functions wherever it's important not to dereference a null pointer?
* leo_ and possibly others want to replace the whole PF code with a bunch of PMCs
<chip> So the embedding interface must be NULL-safe, the extending interface should be NULL-safe (exceptions fine), and no other interfaces have any obligation to check for null pointers.
<chromatic> NULL-safe only on entry points into Parrot through the external API?
<allison> that seems like a reasonable compromise
<chip> Yes.  Additional null checks seem like good for debugging only.
<chromatic> That works for me as a general principle.
<chip> OTOH, "no segfaults when the user has followed the rules" trumps all other rules.
<chip> OK.  I'll add that to pdd07 and you can let me know if I've caused a problem.
<chromatic> Sure, but if someone treats our opaque interpreter pointer as something writable, I'm okay with a little good-natured taunting.
<chip> Yeah.  That's the user-who-doesn't-follow-the-rules "spawn nethack" rule  :-)
<chromatic> Okay, I'll work up a SEGV signal handler with execve.
<chip> um, could you unpack that?
<chip> oh duh
<chip> excellent!
<particle> chromatic: since my 'patch' doesn't like your patch, i'll apply manually today
<chromatic> Which patch?
* Jonathan has to disappear - sorry!
<particle> Parrot::Embed take 2
<obra> take care, Jonathan 
<obra> what else do folks have?
<particle> a desire for a common PAST?
<obra> Also, world peace?
<leo_> just domination
% chromatic has left chromatic!~chromatic@sub17-30.member.dsl-only.net
<particle> mmm... whirled peas
<obra> Ok. I'll catch everybody next week
<obra> *wave*
<leo_> thx Jesse cu
% chromatic has joined #parrotsketch
% chromatic has left chromatic!~chromatic@sub17-30.member.dsl-only.net
% Coke has left #parrotsketch
% Jonathan has left #parrotsketch
% allison has left allison!~chatzilla@sub17-30.member.dsl-only.net
% Nicolashotmail has joined #parrotsketch
<Nicolashotmail> Hello
<Nicolashotmail> What is parrot sketch
<Nicolashotmail> ?
<obra> It's a weekly chat about the development of Parrot, a new virtual machine, designed for dynamic languages like Perl
<leo_> http://www.parrotcode.org/misc/parrotsketch.html
<Nicolashotmail> Thankx, have a good day ! bye.
% Nicolashotmail has left #parrotsketch
% particle_ has joined #parrotsketch
% particle has left particle!~particle@144.81.84.141
% particle_ is now known as particle
% particle has left particle!~particle@144.81.84.141
% particle has joined #parrotsketch
% particle has left particle!~particle@144.81.84.141
% particle has joined #parrotsketch