% AndyA has joined #parrot
% tewk has left tewk!~tewk@ekstrom.org
% tewk has joined #parrot
<dalek> r27776 | pmichaud++ | trunk:
<dalek> : [pct]:
<dalek> : * Add keyed_int scope.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27776
<dalek> r27777 | pmichaud++ | trunk:
<dalek> : [nqp]:
<dalek> : * Switch postcircumfix:[] to use keyed_int scope instead of keyed.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27777
<dalek> r27778 | pmichaud++ | trunk:
<dalek> : [rakudo]:
<dalek> : * Refactor lots of stuff dealing with Pairs, argument handling, and lists.
<dalek> :   The resulting code is 127 lines shorter, more correct, gets rid
<dalek> :   of lots of unwanted references to "infix:,", and helps get us closer
<dalek> :   to being able to handle list assignment and list context.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27778
<cognominal> pmichaud++ # when ones remove code and improves thins, that menas he gets it right
<bacek> pmichaud++! infix:, - root of all evil!
<bacek> karma pmichaud 
<purl> pmichaud has karma of 1277
<Tene> Now I need to svn up again before I leave in the morning.
<cognominal> pmichaud++ # never panics 
% teknomunk has left teknomunk!~teknomunk@r74-195-239-111.stl1cmta01.stwrok.ok.dh.suddenlink.net
<cognominal> Unable to find regex 'panic'
<cognominal> Null PMC access in invoke()
<Eevee> panic: panic not found
<Tene> panic: Eevee not Eevee
<Eevee> I'm not?
<Eevee> this raises a lot of questions
<bacek> Only true Eevee can raise a question. You are not Eevee
<Tene> reduce all code to one word: DWIM
<Eevee> that's Perl 7
* Tene sleep
% iblechbot has joined #parrot
<cognominal> when Perl 7 is bootsrapped, it eventually compiles himself whith the unique following line:   DWIM ::= 42
<cognominal> Eevee: this is good if you trapped in a cave with a cyclop
<cognominal> s/you/you are/
<cognominal> pmichaud the tree is broken
<cognominal> I checked with a clean tree (non polluted in my side)
<moritz> which tree?
<moritz> rakudo works fine for me
<moritz> just ran spectest_regression, all pass
% pjcj has left pjcj!~pjcj@84-73-177-217.dclient.hispeed.ch
<Tene> cognominal: more information, please?
<purl> it has been said that more information is available at our Details page.
% ejs has left ejs!~ejs@172-140-124-91.pool.ukrtel.net
<nopaste> "bacek" at 202.7.166.163 pasted "Failed 5/583 test programs. 19/11275 subtests failed." (23 lines) at http://nopaste.snit.ch/13054
<bacek> I'm not sure about latest commits... I didn't run make test for couple of weeks
<nopaste> "cognominal" at 82.67.232.89 pasted "you broke perl6" (13 lines) at http://nopaste.snit.ch/13055
<cognominal> I did make realclean
<cognominal> forget information
<purl> cognominal: I forgot information
<cognominal> forget more information
<purl> cognominal: I forgot more information
<bacek> purl?
<purl> yes, bacek?
<bacek> forgot purl
<bacek> forget purl
<purl> bacek: I forgot purl
<bacek> stupid bot?
<purl> What do you expect for 1 meg of AI?
<bacek> LOL :)
<bacek> is purl implemented in erl6?
<bacek> perl6?
<purl> perl6 is, like, pretty explicit about making them available
<moritz> purl, forget perl6
<purl> moritz: I forgot perl6
<moritz> purl: perl 6?
<purl> somebody said perl 6 was the spec, rakudo and pugs are two of the implementations.
<moritz> purl, perl6 is http://dev.perl.org/perl6/ or done when it's done
<purl> OK, moritz.
<cognominal> autoreference?
<cognominal> autoreference is fun with purl
<moritz> autoreference is also "did you mean autobox?"
<purl> okay, moritz.
<bacek> karma?
<purl> karma is, like, just as useful as the Perlmonks XP :-) or at http://ircxory.jrock.us/
<moritz> karma bacek
<purl> bacek has karma of 7
<bacek> not so much ;)
<cognominal> bacek, no, purl is written in perl 5
<moritz> bacek: make more commits ;)
<bacek> moritz, can you commit patches from #54740 and #54742? :)
<moritz> bacek: I don't have a commit bit (yet)
<bacek> me either
<bacek> karma moritz
<purl> moritz has karma of 21
<bacek> heh. 3 times more :)
<moritz> bacek: and more than half of it due to one patch ;)
<moritz> or rather, one patch and three follow-ups that fixed that patch. And then really fixed it ;)
<bacek> moritz, :)
% ejs has joined #parrot
% ejs has left ejs!~ejs@114-156-124-91.pool.ukrtel.net
% ejs has joined #parrot
% ejs has left ejs!~ejs@2-246-124-91.pool.ukrtel.net
% Ivatar has joined #parrot
<cognominal> we start slowly and accelerates progressively until we attain the Kurzweilian singularity (Perl 7 enlignment)
% ejs has joined #parrot
<bacek> Why dies_op is not implemented in Test.pm?
<moritz> which Test.pm?
<moritz> rakudo: my $x = sub { say "foo" }; $x();
<moritz> bacek: I think because $! isn't implemented yet
<moritz> uhm, 'make testtest' is b0rked
<bacek> shit...
* bacek hates 'infix:,'...
<dalek> allison@perl.org | Bylaws:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?bylaws
<bacek> pmichaud, when you'll be available: check past/pir for "my $x=sub {say 'Hello'}; $x()"
<bacek> infix:, suck all the time...
<dalek> allison@perl.org | Articles of Incorporation:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?articles_of_incorporation
<shorten> dalek's url is at http://xrl.us/bkyfj
<cognominal> pmichaud, my bad. my tree was clean in the sense that  'svn diff' returns nothing but that is not a pristine one
<dalek> allison@perl.org | Articles of Incorporation:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?articles_of_incorporation
<shorten> dalek's url is at http://xrl.us/bkyfj
<dalek> allison@perl.org | Articles of Incorporation:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?articles_of_incorporation
<shorten> dalek's url is at http://xrl.us/bkyfj
<dalek> allison@perl.org | Articles of Incorporation:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?articles_of_incorporation
<shorten> dalek's url is at http://xrl.us/bkyfj
<dalek> allison@perl.org | Articles of Incorporation:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?articles_of_incorporation
<shorten> dalek's url is at http://xrl.us/bkyfj
<dalek> allison@perl.org | Articles of Incorporation:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?articles_of_incorporation
<shorten> dalek's url is at http://xrl.us/bkyfj
% rdice has joined #parrot
<dalek> allison@perl.org | Articles of Incorporation:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?articles_of_incorporation
<shorten> dalek's url is at http://xrl.us/bkyfj
<dalek> allison@perl.org | Bylaws:
<dalek> link: http://www.perlfoundation.org/parrot/index.cgi?bylaws
% Psyche^ has joined #parrot
% ank has left ank!~ank@ppp121-44-210-24.lns1.hba1.internode.on.net
% Patterner has left Patterner!~Psyche@e177234003.adsl.alicedsl.de
% Psyche^ is now known as Patterner
% iblechbot has left iblechbot!~iblechbot@ppp-62-216-197-67.dynamic.mnet-online.de
% Casan has joined #parrot
% kid51 has joined #parrot
<bacek> Hmm...
<bacek> What the difference between 'push @l, (1,2,3)' and 'push @l, [1..5]'?
<bacek> Why first pushes 3 items and second only one?
<moritz> the first flattens, the second doesn't
<moritz> the first is identical to push @l, 1, 2, 3;
<moritz> whereas [...] never flattens
<dalek> r27779 | jonathan++ | trunk:
<dalek> : [rakudo] A few updates to Mutable PMC, and subclass it for Scalar, Array and Hash containers.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27779
<dalek> r27780 | jonathan++ | trunk:
<dalek> : [rakudo] Tests for Mutable PMC.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27780
<bacek> moritz, in which syn it described? And how I can distinguish this calls in PIR?
% Ademan has left Ademan!~dan@h-67-101-146-154.snfccasy.dynamic.covad.net
<bacek> ho. Jonathan is alive! :)
% Ademan has joined #parrot
<moritz> bacek: S03 calls [1,2,3] an "array composer"
<bacek> S03?
<purl> S03 is the operators spec or dev.perl.org/perl6/doc/design/syn/S03.html
<bacek> moritz, thanks.
<moritz> bacek: btw the IRC logs at http://irclog.perlgeek.de/parrot/today turn strings like "S03" and S03:1234 into links automatically
<bacek> Do I understand right that we really need Array class in rakudo?
<bacek> moritz, I don't need logs! I'm here right now! :)
<moritz> bacek: I'm here as well, but I find these links so convenient that I used logs anyway ;)
<moritz> for example r27780 linking to the last changeset
<cognominal> moritz++ # I did not know that
* bacek thinking about writing plug-in for xchat to do same thing.
<moritz> cognominal: half of my commits to the pugs repo where actually tweaks to the IRC logs ;)
<cognominal> I want to move from xchat to limechat once I add some missing features.
<cognominal> were is that in pugs?
<moritz> in misc/irclog/
<moritz> http://svn.pugscode.org/pugs/misc/irclog/lib/IrcLog/WWW.pm adds the markup
<cognominal> next is to rewrite it in Perl 6 :)
% ejs has left ejs!~ejs@148-166-124-91.pool.ukrtel.net
<cognominal> I just got a "natural"  use of regex as a method with a paramter but it is not yet supported
% iblechbot has joined #parrot
% jbf has joined #parrot
% ejs has joined #parrot
<Juerd> Also, if you want to type URLs but not long ones, tnx.nl/S03 redirects.
<Juerd> (Will probably be moved to another domain soon though)
% cognominal has changed the topic of #parrot to: Parrot 0.6.2 "Reverse Sublimation" Released | http://parrotcode.org/ | 18/672 new/open tix | logged in http://irclog.perlgeek.de/parrot/today
% Ademan has left Ademan!~dan@h-67-101-146-154.snfccasy.dynamic.covad.net
% rdice has left rdice!~richarddi@CPE001ff33cb98b-CM00159a01d44c.cpe.net.cable.rogers.com
% Ademan has joined #parrot
% ejs has left ejs!~ejs@38-157-124-91.pool.ukrtel.net
% pnutr has joined #parrot
% pnutr is now known as superNUT
% jbf has left #parrot
% kid51 has left kid51!~jkeen@pool-70-107-13-187.ny325.east.verizon.net
<bacek> moritz, do you have commit permission to pugs?
<moritz> bacek: yes
<moritz> bacek: want one too?
<bacek> moritz, no.
<bacek> moritz, can you add 'lives_ok' into t/spec/fudge 
<moritz> bacek: I can
<bacek> moritz, thanks. Line 76
<moritz> bacek: done
<bacek> moritz, thanks
<moritz> you're welccome, thanks for report
<moritz> ing
<bacek> moritz, ho! Can I send you few... ee.. lot of patches for spectest? With #?rakudo skip 'Not implemented yet' in most of them?
<moritz> bacek: you can, but you can also get a commit bit within 3 minutes
<bacek> moritz, no. I do not want. Really :)
<moritz> ok
<moritz> either nopaste them or send to moritz at faui2k3.org
% jhorwitz has joined #parrot
<nopaste> "bacek" at 202.7.166.163 pasted "S29-list/sort.t patch for moritz" (62 lines) at http://nopaste.snit.ch/13057
<cognominal> bacek++
<nopaste> "bacek" at 202.7.166.163 pasted "S29-list/grep.t patch for moritz" (34 lines) at http://nopaste.snit.ch/13058
% ejs has joined #parrot
% Theory has joined #parrot
% Ademan has left Ademan!~dan@h-67-101-146-154.snfccasy.dynamic.covad.net
<bacek> How I can check arity of passed closure in PIR? E.g. in @l.map { $^a + $^b } ?
<moritz> bacek: patches committed
<moritz> s/mm/m/
<bacek> moritz, thanks
<bacek> moritz++ :)
* bacek found arity method...
% teknomunk has joined #parrot
% Ademan has joined #parrot
% Ademan has left Ademan!~dan@h-67-101-146-154.snfccasy.dynamic.covad.net
% IllvilJa has left IllvilJa!~jilves@emea-netcache1.oracle.co.uk
% petdance has joined #parrot
<Tene> bacek: .arity, or, if you need more information about, say, optional or slurpy params, inspect
% ejs has left ejs!~ejs@233-128-124-91.pool.ukrtel.net
% Psyche^ has joined #parrot
% Patterner has left Patterner!~Psyche@f054000015.adsl.alicedsl.de
% Psyche^ is now known as Patterner
% Zaba has joined #parrot
% Zaba_ has left Zaba_!~zaba@ip102.148.adsl.wplus.ru
<cognominal> I just got a "natural"  use of regex as a method with a paramter but it is not yet supported  <=== I was wrong
% wknight8111 has left wknight8111!~Whiteknig@c-71-230-33-251.hsd1.pa.comcast.net
<pmichaud> (lives_ok) we're wanting to not come up with too many tests that fudge has to recognize, I think
<pmichaud> lives_ok is probably okay for now, but we don't want to go overboard with it
<moritz> I think we should just make everything available that is needed frequently
<pmichaud> with "frequently" being the key word
<moritz> ack -al lives_ok|wc -l
<moritz> 145
<pmichaud> what does lives_ok test that couldn't be easily tested by something else?
<moritz> that something doesn't through an exception
<pmichaud> wouldn't 'ok' do that?
<moritz> no
<moritz> ok's args are evaluated before ok() is called
<moritz> lives_ok expects something Callable
<moritz> and there are many instances of 'eval "string"; ok !$!, "it lived";' that I'd like to convert to eval_lives_ok
<moritz> (or we could use a multi of lives_ok(Str, ...), but that could be confusing)
<dalek> r27781 | chromatic++ | trunk:
<dalek> : [PMC] Tidied PMCProxy code slightly.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27781
<jonathan> hi all
<jonathan> Workshop has no wifi...just a cat-5 for the person who is giving a talk
<moritz> hi jonathan ;)
<jonathan> Thus my general off-lineness, though I did a ci of some hacking I'd done earlier.
<jonathan> moritz: See you got more tests added. :-)
<moritz> jonathan: I don't think it's comitted yet
<dalek> r27782 | chromatic++ | trunk:
<dalek> : [src] Improved C++ building: removed unnecessary casts, improved attribute
<dalek> : ordering (NotFound, RT #54674).
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27782
<jonathan> moritz: Will do now
<jonathan> Just discovered I don't have a patch program on here.
<moritz> d'oh
<jonathan> I didn't hvae lex and yacc either...well, fixing that now.
<moritz> should I send you the plain file?
<cognominal> jonathan : I have a temporary logo for rakudo : http://objectif-brousse.org/rakudo.jpg
% petdance_ has joined #parrot
% petdance has left petdance!~Andy@64.81.227.163
<moritz> btw 'my &x = 1' should fail (because 1 !~~ Callable), but doesn't in rakudo
% petdance_ has left petdance_!~Andy@64.81.227.163
<jonathan> moritz: Once I've got container types/mutable stuff sorted out, plus :instanceof for subs, that will be fixed.
% petdance has joined #parrot
<jonathan> Need to get both of those done first.
<moritz> ok
<jonathan> cognominal: I'm hoping that it's a camel doing martial arts... ;-)
<moritz> .oO( Jew Jitsu http://www.adi.priv.hu/cucc/jew-jitsu.jpg )O
<jonathan> However, they're both on my do really soon list - in fact, the mutable stuff is well underway.
<dalek> r27783 | jonathan++ | trunk:
<dalek> : [rakudo] Add more tests to the known passing list. Patch courtesy of moritz++.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27783
<jonathan> moritz: lol! :-)
<dalek> r27784 | jonathan++ | trunk:
<dalek> : [rakudo] Update makefile to build the scalar, array and hash PMCs that subclass Mutable.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27784
% petdance has left petdance!~Andy@64.81.227.163
<dalek> r27785 | chromatic++ | trunk:
<dalek> : [JIT] Undid improper extern-ization of two x86-JIT-only variables from r27736
<dalek> : (NotFound, also RT #54602).
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27785
<dalek> r27786 | jonathan++ | trunk:
<dalek> : Update MANIFEST to include PMCs I added earlier today.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27786
% braceta has joined #parrot
% teknomunk has left teknomunk!~teknomunk@r74-195-239-111.stl1cmta01.stwrok.ok.dh.suddenlink.net
% teknomunk has joined #parrot
% braceta has left braceta!~Braceta@bl9-219-249.dsl.telepac.pt
<dalek> r27787 | jonathan++ | trunk:
<dalek> : [rakudo] Turns out that calling a PMC Scalar isn't such a good idea. :-)
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27787
* jonathan puts on his armor of C horror protection and hacks on IMCC.
% cognominal has left cognominal!~cognomina@82.67.232.89
% ank has joined #parrot
% cognominal has joined #parrot
<pmichaud> I'm also on limited bandwidth today (cell phone while driving through town)
<pmichaud> jonathan: how's the mutable stuff coming along?
<pmichaud> (if I disappear suddenly, it's because of cell droppage)
% Psyche^ has joined #parrot
<dalek> r27788 | chromatic++ | trunk:
<dalek> : [IMCC] Tidied file, resolving one small TODO.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27788
% Patterner has left Patterner!~Psyche@f054001172.adsl.alicedsl.de
% Psyche^ is now known as Patterner
<jonathan> pmichaud: Well, I actually replaced copy with assign, and :viviself with Mutable rather than Undef, and passed all the sanity tests earlier today.
<jonathan> not committed that patch, actually reverted it since it was simple find/replace
<jonathan> Will do a branch, I guess.
<pmichaud> I'm happy with either branch or trunk, if you are.
<pmichaud> i.e., you can choose.  It's okay if spectest_regression breaks temporarily as long as 'make test' works.
<pmichaud> I looked at the code for .=  last night -- shouldn't we able to treat .=  as just a method call and assign?
<jonathan> I am without wifi at the venue, so if I can get a lot working...
<jonathan> Be careful if you refactor that
<jonathan> Make sure my Dogg
<jonathan> oops
<jonathan> my Dog $fido .= new();
<jonathan> still workis afterwards
<pmichaud> right.  "my Dog $fido"  should cause $fido to be initialized with a Dog protoobject, which then becomes the invocant for .new, yes?
<jonathan> yes
<jonathan> IIRC, there was a little subtlety to getting .= to work when you had a declaration
<pmichaud> hmmmm
<pmichaud> I'm guessing it'll be easier when mutables are in place, because we should be able to get rid of :viviself entirely
<jonathan> ?
<jonathan> We still need the initial thing to be a Scalar or whatever
<jonathan> Not just a null PMC
<pmichaud> right, but I'd bind it immediately
<jonathan> ah, ok
<jonathan> I was viviself-ing it.
<pmichaud> my $a;   should immediately create a Scalar and bind it to $a
<jonathan> OK
<jonathan> Works for me.
<pmichaud> however, that Scalar can be smart enough to respond with false to "exists" if it hasn't been assigned to or vivified yet
<jonathan> well, I put a Failure in it initiallhy.
<jonathan> Need to get it to have a protoobject if it was a class though
<pmichaud> well, if we do  my $a;   then does $a exist yet?
<jonathan> No
<jonathan> So far I'm just forwarding find_method to whatever the value is, though.
<pmichaud> so, what I wnt to have happen is
<pmichaud> that a mutable keeps track of how it should be initialized when it is vivified
<pmichaud> or, alternately, it could go ahead and vivify itself, but report 'false' for exists until it's assigned to
<pmichaud> that probably makes more sense.
<pmichaud> and yes, it could go ahead and get a protoobject at initialization.
<pmichaud> so, "my $a;"   creates a scalar $a and initializes it with a Failure object
<jonathan> That is what I am doing at the moment.
<jonathan> See Mutable PMC.
<pmichaud> "my Dog $a;"  creates a scalar $a and initializes it with a Dog protoobject
<jonathan> Right.
<jonathan> That's what I'm planning.
<pmichaud> so then   "my Dog $a .= new()"   evaluates  "my Dog $a" first, then just calls .new on the protoobject and assigns the result back into $a
<jonathan> The subtlety is that you generate some PAST that initializes it with the proto-object.
<pmichaud> yes, but that's part of "my Dog $a" and not .=
<pmichaud> right?
<jonathan> yeah, but you need to get the PAST::Var out of that to use with .=.
<jonathan> And that's what the special-case code in .= is doing.
<pmichaud> but  "my Dog $a" is a PAST::Var
<jonathan> It's a bit more.
<pmichaud> ??
<jonathan> It's a binding too.
<jonathan> And setting of properties if there's a type constraint.
<pmichaud> the binding returns the thing bound, fwiw
<pmichaud> setting of properties should be part of the viviself of my Dog $a
<dalek> r27789 | chromatic++ | trunk:
<dalek> : [src] Removed an unnecessary const_string() call from MMD registration code.
<pmichaud> for that matter, it can all go into the viviself
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27789
<jonathan> Then viviself needs to be powerful enough to suport that.
<pmichaud> viviself can have any PAST tree
<jonathan> Oh?
<jonathan> I didn't know that.
<jonathan> That makes things easier. :-)
<pmichaud> most things that take strings can also take PAST trees
<pmichaud> string is just a shortcut for a PAST tree to do "new 'Foo'"
<jonathan> But not :name.... ;-)
<pmichaud> right.  But :named can be :-)
<jonathan> (BTW, if :name could take a tree, we can probably tidy some more things up.)
<pmichaud> and if we really want :name to accept a PAST var, we could do that.
<pmichaud> that's a good idea.
<jonathan> Well, or a PAST::Op.
<pmichaud> sorry, I meant PAST tree
<jonathan> Sure
<jonathan> That would let some of the indirect access be cleaner.
<jonathan> OK, I'm going to go and meet up with some people now.
<pmichaud> yes, it would -- we could get rid of some unshifts.
<jonathan> Be sociable. :-)
<jonathan> BTW, my PCT talk at the uni and my Perl 6 talk at NPW today were both well received.
<pmichaud> excellent.
<jonathan> Got people talking about trying to do various languages with PCT.
<pmichaud> I submitted talks for PCT and Rakudo to YAPC::EU
<jonathan> OK
<jonathan> I will submit mine too.
<jonathan> I talked to one of the organizers today.
<jonathan> They know I'm coming and will submit.
<jonathan> Asked if I would co-ordinate with you, what we would talk about. Told him we were way ahead of him on that. :-)
<jonathan> Will probably submit an OO and a types one.
<pmichaud> excellent
<pmichaud> those would be well received also
<jonathan> I think so.
<jonathan> People here seemed to like what they saw.
<jonathan> I talk about Rakudo tomorrow.
<jonathan> Today was about Perl 6 as a language.
<jonathan> Anyway, beer time...laterz.
<pmichaud> later
<cognominal> jonathan, how many people at the workshop?
<cognominal> pmichaud the panic is still broken.
<spinclad> <panic: Help!> ==> <panic: 'Help!'>
<pmichaud> which panic?
<pmichaud> oh, it's not being imported.
<pmichaud> just a sec.
<pmichaud> actually, panic must be working, otherwise we'd never see a syntax error.  So, where is panic broken?
<cognominal> Unable to find regex 'panic'
<cognominal> Null PMC access in invoke()
<cognominal> okm the syntax changed
<cognominal> I am doing a pod5 grammar
<cognominal> I don't get it
<pmichaud> I probably need to be able to see an example of it not working.
<cognominal> I pasted the TOP rule of perl 6
<cognominal> so I should get the same problem with a faulty perl 6
<cognominal> I check
<pmichaud> is the grammar a subclass of PCT::Grammar ?
<cognominal> ho shoot.
<cognominal> you are right
<pmichaud> afk for a while  #  nap
% Zaba_ has joined #parrot
<cognominal> thx
% Zaba has left Zaba!~zaba@ip102.148.adsl.wplus.ru
<dalek> r27790 | rgrjr++ | trunk:
<dalek> :    + Remove gratuitous "__" from names.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27790
% mire has joined #parrot
% pjcj has joined #parrot
% guru has joined #parrot
% tetragon has left tetragon!~seneca@216.126.67.44
% super_ has joined #parrot
% Zaba_ has left Zaba_!~zaba@ip102.148.adsl.wplus.ru
% Zaba has joined #parrot
% tetragon has joined #parrot
% askie has left askie!~askie@81.171.100.207
% askie has joined #parrot
% rdice has joined #parrot
% guru has left guru!~guru@dsl-67-204-22-150.acanac.net
% Casan has left Casan!~IceChat7@users163.kollegienet.dk
% ejs has joined #parrot
% Psyche^ has joined #parrot
% Patterner has left Patterner!~Psyche@e177235034.adsl.alicedsl.de
% Psyche^ is now known as Patterner
<bacek> morning...
% iblechbot has left iblechbot!~iblechbot@ppp-62-216-197-96.dynamic.mnet-online.de
% super_ has left super_!~super@97.102.76.72
<cognominal> when bacek the sentient andipodian wakes up it's time for me to go to bed.
<cognominal> I am not a true programmer that takes the graveyard shift
% jisom has joined #parrot
* jonathan has insomnia. :-(
<Infinoid> at least its a weekend
* Infinoid is trying to stay awake after stuffing his belly full of sushi
<jonathan> Infinoid: Yeah, but I'm giving talks.
<jonathan> Had one on Friday, gave one today, got another tomorrow.
<jonathan> Anyway, went for a walk just now...maybe that helps.
* jonathan likes walking around Stockholm.
<Infinoid> cool
<cognominal> I can't sleep eiher :(
<cognominal> and jonathan must keep strenght to talk Friday at http://conferences.mongueurs.net/fpw2008/schedule
<jonathan> cognominal: I have work meeting here on Monday. I fly home on Tue. I get back at 6pm. On Thursday morning I fly to France for FPW.
<jonathan> So it's all quite exhausting.
<jonathan> if I can get some sleep in between now and then, I'll be fine.
<cognominal> and on the side, you are a core programmer of parrot :)
<jonathan> I'm trying to do that between talks. ;-)
<jonathan> Actually, I have no conferences between FPW and YAPC::EU.
<cognominal> I will meet you at the boarding gate in Orly, btw
<jonathan> So I have time to code, and to get to know my new country.
<jonathan> Boarding gate?
<jonathan> Shouldn't we check in together?
<jonathan> So we can have seats together on the flight?
<jonathan> Bonus hacking! ;-)
<cognominal> yes
<jonathan> OK
<jonathan> Email me (jnthn@jnthn.net) your cell number
<jonathan> I call you once I land from Vienna.
<jonathan> And then we can meet up and go check in.
<cognominal> yes
<jonathan> Cool.
* jonathan is looking forward to Albi conference.
<jonathan> Ok, I will try and sleep now.
<jonathan> Got first talk of the morning (thankfully, at 10am) tomorrow. uh, today.
<jonathan> night all
<cognominal> sent
<jonathan> thanks
<jonathan> will reply with mine
<jonathan> cya
<cognominal> I will bring a mulitplug power cord.
% Ivatar has left Ivatar!~graham@tu055.demon.co.uk
<cognominal> boardind gate != checking gate.   I understand jonathan now.
% ejs has left ejs!~ejs@252-143-124-91.pool.ukrtel.net
% rdice has left rdice!~richarddi@CPE001ff33cb98b-CM00159a01d44c.cpe.net.cable.rogers.com
% mire has left mire!~Frodo@223-172-222-85.adsl.verat.net
% rdice has joined #parrot
<bacek> I still don't understand which part of STD.pm related to @a.map: {...}...
% Limbic_Region has joined #parrot
% jisom has left jisom!~jisom@c-98-212-164-35.hsd1.il.comcast.net
% Zaba_ has joined #parrot
% Zaba has left Zaba!~zaba@ip102.148.adsl.wplus.ru
<pmichaud> @a.map: {...} or  @a.map:{...} ?
<bacek> pmichaud, with space.
<bacek> I try to fix spectest. There is problems with arity...
<bacek> is(("foo","bar").map: { $_.substr(1,1) }, @expected, "map of constant list works");
<bacek> it parsed as map with 3 args and 'is' with 1. Instead of 'is' with 3 and map with 1.
<nopaste> "bacek" at 202.7.166.163 pasted "Implementation of List.map for review" (81 lines) at http://nopaste.snit.ch/13059
% braceta has joined #parrot
<bacek> pmichaud, can you review this one? It supports multiple arity map.
<bacek> say ~(1,2,3,4).map: { $^a + $^b }; returns '3 7' as expected
<pmichaud> the STD.pm rule is methodop
<pmichaud> the line that says
<pmichaud>     | ':' <?before \s> <!{ $+inquote }> <arglist>
<bacek> pmichaud, found it.
<bacek> Looks like src/parser/grammar.pm should have explicit priority for 'dotty'. Similar to --> Methodcall
<bacek> ...in STD.pm
<bacek> correct?
<pmichaud> I think the spectest is wrong.
<bacek> pmichaud, hmm...
<bacek> pmichaud, there is many places with such syntax in spectest.
<bacek> Should we change it and use explicit parentizes around 'map'?
<pmichaud> either that or remove the space before the {
<bacek> if fails to parse without space.
<bacek> Is it 'colonpair' not fully implemented?
* bacek wants someone to check #54740 and #54746 :)
<pmichaud> correct, rakudo doesn't understand :{...} yet.
<bacek> pmichaud, which part of STD.pm not implemented in this case?
<pmichaud> I don't understand the question.
<pmichaud> We have :{...} in the grammar, we just don't have it so that it maps correctly to a closure argument
<bacek> Statement not terminated properly at line 5, near "((\"foo\",\"b"
<bacek> I assume that grammar isn't complete... Am I totally wrong?
<pmichaud> there are parts of the grammar that aren't complete, yes.  But I'm not sure that this is one of them.
<pmichaud> in nopaste #13059, the call to "newclosure" is wrong.
<bacek> pmichaud, why?
<pmichaud> newclosure can only be called from the closure's outer sub
<bacek> pmichaud, it doesn't work without it :(
<pmichaud> (which 'map' is not.)
<pmichaud> correct, there's a problem in that rakudo doesn't always call newclosure appropriately.
<pmichaud> but by the time that 'map' gets called, newclosure will have already been done.
<bacek> my @l = (1..3); say ~@l.map({$_*2});
<bacek> my @l = (1..3); { say ~@l.map({$_*2}); }
<bacek> without newclosure I've got too few arguments passed (0) - 1 params expected
<pmichaud> correct.
<bacek> But it crashes with newclosure... Hmm...
<pmichaud> there's a problem with parrot on its handling of :outer
<bacek> pmichaud, how I can fix it?
<pmichaud> I don't know how to fix it -- it's a deep parrot bug and I'm still trying to come up with a good workaround.
<pmichaud> the bug is rt#47956
<pmichaud> because of that bug,  the {$_*2} closure of the first example can't be given an :outer 
<pmichaud> since it can't be given an :outer, it can't be made into a closure.
<bacek> pmichaud, ouch...
<bacek> we need someone with great "dark magick" skills to fix it.
<pmichaud> I might be able to do a workaround.
<pmichaud> let me make a workaround and let's see where it goes.
<bacek> warp everything into closure and invoke it?
<TimToady> < bacek> it parsed as map with 3 args and 'is' with 1. Instead of 'is' with 3 and map with 1.
<TimToady> that looks like how it's supposed to parse; the : introduces an argument list, and {}, 2, 3 is a list of 3 args
<pmichaud> yes, that's why I think the spectest is wrong there.
<pmichaud> back in a few mins.
<TimToady> yes, it's wrong.
* bacek rejected yesterday commit rights to pugs... May be it was wrong decision :) 
<bacek> ping vany 
<purl> I can't find vany in the DNS.
<bacek> vany, ping!
<pmichaud> bacek: you should get a pugs commitbit if you're going to work on spectests
<bacek> pmichaud, I'm going to work on anything that helps rakudo :)
<pmichaud> that would certainly help :-)
<TimToady> I believe I heard that commitbit was back up
<bacek> pmichaud, It's just to much resonsiblity for me at the moment. I still newbie in parrot/rakudo/Perl6
<pmichaud> it's not too much responsibility.  mistakes are okay -- we'll catch them.
<TimToady> yes, commitbit is up
<TimToady> msg me your email and preferred nick, and I can add you
<TimToady> or most anyone on #perl6 can add you
<bacek> TimToady, bacek@bacek.com, bacek :)
<TimToady> commitbit sent, it's customary to add yourself to AUTHORS as first commit (that is, in pugs repo, not parrot repo)
<TimToady> even if all the rest of your commits are down in t/spec :)
% bacek has left bacek!~bacek@123-243-38-218.tpgi.com.au
<confound> bacek@bacek.bacek
% Zaba has joined #parrot
% Zaba_ has left Zaba_!~zaba@ip102.148.adsl.wplus.ru
% bacek has joined #parrot
<dalek> r27791 | pmichaud++ | trunk:
<dalek> : [p6object]:
<dalek> : * Add 'can' method to P6metaclass.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27791
<dalek> r27792 | pmichaud++ | trunk:
<dalek> : [rakudo]:
<dalek> : * Add Any.can().
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27792
<pmichaud> afk for a while -- will look at newclosure late.r
<bacek> Setup: HsSyck.cabal:14: 'Executable' stanza starting with field 'license'
<bacek> Installation failed for HsSyck at util/build_pugs.pl line 235.
% Limbic_Region has left Limbic_Region!~Limbic_Re@c-68-49-236-220.hsd1.md.comcast.net
% jhorwitz has left jhorwitz!~chatzilla@96.245.16.45
% rdice has left rdice!~richarddi@CPE001ff33cb98b-CM00159a01d44c.cpe.net.cable.rogers.com
% Psyche^ has joined #parrot
<bacek> #perl6?
<purl> #perl6 is at irc.freenode.net.
% Patterner has left Patterner!~Psyche@e177224100.adsl.alicedsl.de
% Psyche^ is now known as Patterner
<nopaste> "bacek" at 202.7.166.163 pasted "Parsing bug in methodop." (58 lines) at http://nopaste.snit.ch/13060
<bacek> $x('world') generates 'infix:,'(x, 'world')...
<pmichaud> bacek: good catch.
<pmichaud> of course, that bug isn't in methodop :-)
<cotto_home> does anyone regularly compile with Intel's compiler?
<dalek> r27793 | pmichaud++ | trunk:
<dalek> : [rakudo]:
<dalek> : * Forget to set the Perl6Object => Object mapping.
<dalek> : * mncharity++ Eevee++ spinclad++
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27793
<bacek> pmichaud, ok-ok :) Not methodop
<pmichaud> I have it fixed now -- just running spectest_regression as a verification before committing
<dalek> r27794 | pmichaud++ | trunk:
<dalek> : [rakudo]:
<dalek> : * Fix bug with $x(1) introduced by r27778
<dalek> : * Reported by bacek++
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27794
<pmichaud> $x(1) fixed in r27794
* tetragon watches Configure.pl collapse after passing in the options for a pure i386 build of parrot on a PPC box
<nopaste> "tetragon" at 69.196.141.26 pasted "Fun with testing an OS X configuration patch" (190 lines) at http://nopaste.snit.ch/13061
<pmichaud> bacek: ping
% petdance has joined #parrot
% tetragon has left tetragon!~seneca@69-196-141-26.dsl.teksavvy.com
<dalek> r27795 | pmichaud++ | trunk:
<dalek> : [rakudo]:
<dalek> : * Set svn:ignore property for src/pmc/mutable.pmc .
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27795
% tetragon has joined #parrot
% teknomunk has left teknomunk!~teknomunk@r74-195-239-111.stl1cmta01.stwrok.ok.dh.suddenlink.net
<dalek> r27796 | pmichaud++ | trunk:
<dalek> : [pct]:
<dalek> : * Make PAST::Block nodes a bit smarter about executing 'newclosure'.
<dalek> :   Ideally we'd execute 'newclosure' almost anytime that we refer
<dalek> :   to a block we just created and avoid the typecheck, but RT#47956
<dalek> :   prevents us from doing that.
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27796
<pmichaud> bacek: r27796 should mean you no longer need newclosure for 'map', 'grep', 'reduce', etc.
% Theory_ has joined #parrot
% Theory has left Theory!~Theory@c-24-21-175-208.hsd1.or.comcast.net
% braceta has left braceta!~Braceta@bl9-219-249.dsl.telepac.pt
<bacek> pmichaud, great!
* bacek works as plumber today... A lot of things to fix at home :)
<bacek> pmichaud, it works!
% shamu has left #parrot
<bacek> pmichaud++
<bacek> Files=39, Tests=666
<bacek> Perl6 _IS_ evil :)
% cotto_home has left cotto_home!~cotto@75-92-237-130.sea.clearwire-dns.net
<dalek> r27797 | pmichaud++ | trunk:
<dalek> : [rakudo]:
<dalek> : * Fix :a<3> Pair notation.  (mncharity++ for noticing this.)
<dalek> : * Add stringification for Pairs (possibly incorrect -- needs
<dalek> :   spec clarification).
<dalek> diff: http://www.parrotvm.org/svn/parrot/revision?rev=27797
<pmichaud> afk # sleep
% shamu has joined #parrot
% shamu has left #parrot
% ank has left ank!~ank@ppp121-44-210-24.lns1.hba1.internode.on.net
% mire has joined #parrot
% cotto_home has joined #parrot
% petdance has left petdance!~Andy@64.81.227.163
% AndyA has left AndyA!~andy@82.152.157.85
% ank has joined #parrot
% AndyA has joined #parrot
% Zaba has left Zaba!~zaba@ip102.148.adsl.wplus.ru