% iblechbot has joined #parrot
<spinclad> japhb: if Piper is like Tweety (and from its announcement it sounds just like) then it will announce to the channel each time it restarts, and /msg everyone who enters on its watch.
% alvar has joined #parrot
<spinclad> ^ (re one-time announcement)
% spinclad has left #parrot
% spinclad has joined #parrot
% alvar has left alvar!~alvar@sub87-230-120-57.he-dsl2plus.de
<spinclad> and sure enough, Piper just /msg'd me.
<spinclad> thanks, Piper!
% alvar has joined #parrot
% contingencyplan has left contingencyplan!~contingen@76.186.27.146
* teknomunk is away: Gone away for now.
% teknomunk is now known as teknomunk_afk
% AndyA has left AndyA!~andy@82.152.157.85
% cosimo_ has joined #parrot
<spinclad> particle: trivial bikeshedding: move logs to http://www.parrotcode.org/misc/parrot-logs/ ?
% contingencyplan has joined #parrot
% cosimo_ has left #parrot
% Alias has joined #parrot
% AndyA has joined #parrot
% HG`` has joined #parrot
% HG` has left HG`!~wells@user-544636b6.lns2-c12.dsl.pol.co.uk
% cognominal_ has left cognominal_!~cognomina@82.67.232.89
% ruoso has joined #parrot
% cognominal_ has joined #parrot
% barney has joined #parrot
% zaphod has joined #parrot
<svnbotlt> r25424 | kjs++ | trunk:
<svnbotlt> : [docs] add a document about PAST nodes. Not sure about location, but these docs need to be re-organized anyway.
<svnbotlt> : manifest is updated accordingly, and file props are set. yay!
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25424
<svnbotlt> r25425 | kjs++ | trunk:
<svnbotlt> : [docs] fix some spelling errors and fix one  {{XXX todo.}} Don't want to copy too much from pdd26, so refer to it where possible.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25425
% kjs_ has joined #parrot
<svnbotlt> r25426 | kjs++ | trunk:
<svnbotlt> : [docs] add a bit more explanation and fix a couple of things in the past_building_blocks.pod doc.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25426
* barney got beaten by chromatic in commenting http://eli.thegreenplace.net/2008/02/01/a-subjective-comparison-between-perl-and-ruby/
<lathos> chromatic pays too much attention to what Ruby people think. :)
<barney> Actually Eli is mostly a Perl guy and his comparison looks quite fair to me
<lathos> My point is still true. :)
<svnbotlt> r25427 | bernhard++ | trunk:
<svnbotlt> : [Rakudo]
<svnbotlt> : Clean up src/utils/Makefile
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25427
% AndyA has left AndyA!~andy@82.152.157.85
% AndyA has joined #parrot
<Alias> lathos: chromatic pays too much attention to what EVERYBODY thinks
% contingencyplan has left contingencyplan!~contingen@cpe-76-186-27-146.tx.res.rr.com
<svnbotlt> r25428 | kjs++ | trunk:
<svnbotlt> : [docs] add an example for 'keyed' access to past_building_blocks.pod
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25428
% contingencyplan has joined #parrot
% cognominal_ has left cognominal_!~cognomina@82.67.232.89
<svnbotlt> r25429 | bernhard++ | trunk:
<svnbotlt> : Add a tip to past_building_blocks.pod
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25429
% cognominal_ has joined #parrot
<kjs_> barney++ # past tip + copyright fix, thanks!
% barney has left barney!~bernhard@p549A16F4.dip0.t-ipconnect.de
<svnbotlt> r25430 | kjs++ | trunk:
<svnbotlt> : [ecmascript] fix assignment and += operators for now. solution is a bit crappy, but it works for now. (why does infix:= + "assign" not work?)
<svnbotlt> r25431 | kjs++ | trunk:
<svnbotlt> : [ecmascript] add one arith. test.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25431
% szbalint has left szbalint!comet@dev.perl.hu
% Alias has left Alias!adam@CPE-124-188-112-79.nsw.bigpond.net.au
% Alias has joined #parrot
% Alias_ has joined #parrot
% Alias has left Alias!adam@CPE-124-188-112-79.nsw.bigpond.net.au
% zaphod has left #parrot
% Andy has joined #parrot
<svnbotlt> r25432 | jonathan++ | trunk:
<svnbotlt> : [rakudo] Very preliminary work on inheritance. Doesn't work unless you declare classes in reverse order, due to an ordering change in the PAST -> POST translation phase.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25432
<svnbotlt> r25433 | kjs++ | trunk:
<svnbotlt> : [cardinal] start using PCT. "x=1" works :-)
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25433
% alvar has left alvar!~alvar@sub87-230-120-57.he-dsl2plus.de
% alvar has joined #parrot
% alvar has left alvar!~alvar@sub87-230-120-57.he-dsl2plus.de
% alvar has joined #parrot
<kjs_> does anybody know whether <after > assertion is implemented in pge?
% alvar has left alvar!~alvar@sub87-230-120-57.he-dsl2plus.de
<kjs_> apparently yes :-) (it worked!)
% Andy has left Andy!~Andy@adsl-75-57-198-137.dsl.emhril.sbcglobal.net
<svnbotlt> r25434 | kjs++ | trunk:
<svnbotlt> : [ecmascript] think to have solved the newline issue in cardinal, using 'after' assertion. 
<svnbotlt> : + added keywords as reserved.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25434
<kjs_> woops; wrong tag
<svnbotlt> r25435 | kjs++ | trunk:
<svnbotlt> : [cardinal] add operator table. just a matter of finishing now :-)
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25435
<TimToady> yow, while thinking about the problems of intuiting the left side of assignment, I kept wishing for a list assignment operator so I could leave = as item assignment
<TimToady> and then it occurred to me that binding to the right uses ->, so binding to the left really ought to use <-
<TimToady> which frees up := for list assignment
<jonathan> What do related-ish things like ::= or =:= become if you do that?
% cognominal_ has left cognominal_!~cognomina@82.67.232.89
<jonathan> I find the reasoning kinda appealing though...
<TimToady> ::= would presumably be slice assignment, if anything
<TimToady> and ::= turns into <<- maybe
<TimToady> about the only downside I can see is the ambiguity with $x<-1
<TimToady> meaning $x < -1
<TimToady> but maybe strictness says you don't bind to bare variables without a declarator anyway
<TimToady> and no one would ever say my $x < -1
<TimToady> also lets us say things like $foo := 1,2,3 and mean it
<TimToady> or more to the point $foo.meth := 1,2,3 and be clear
<TimToady> but it's definitely gonna mindbend some folks
<jonathan> $foo.meth = 1,2,3; # 1,2,3 as a list is passed to meth
<TimToady> and it's a big enough change that we'll actually have to have a migration strategy
<TimToady> no, that's 2,3 in a void context now
<TimToady> and it's not passed to meth in any case, it's assignment to the proxy returned by meth
<jonathan> Sure.
<jonathan> But with := you'd be assigning the 1,2,3 as a list to that proxy?
<TimToady> it also rationalizes the fact that :(Dog $fido = 1, Cat $fifi = 2) parses without eating the comma
<TimToady> item assignment and list assignment would become two very distinct operators
<TimToady> parsed differently, and with different context on the right
<TimToady> but in any case, either would attempt to assign to the lvalue on the left appropriately
<jonathan> Is : sufficiently associated in people's minds with lists elsewhere?
<TimToady> I think so, it's also used in "print $fh: 1,2,3"
<jonathan> I'd suggest @= for list assignment, but it's probably used for something and looks ugly.
<jonathan> But the @ reminds you right away of lists.
<TimToady> I went through that thinking several times, but it has problems
<TimToady> visual ambiguity
<TimToady> but worse it conflicts with detecting two terms in a row
<TimToady> and the whole syntactic house of cards collapses if we can't do that
<TimToady> but yes, I definitely considered @= and $= and =$ and =@ and hated them all
<TimToady> I also considered =:, but we can't really get away with that if we keep := as binding
<TimToady> on the other hand -> is already binding
<TimToady> a <- form wouldn't be entirely symmetrical, missing the block, and maybe requiring "my" or whatever, but it's more for mental space than pure symmetry
<jonathan> I'm using binding in NQP quite a bit, due to the lack of assignment, and thus having to bind to it where I'm not declaring it.
<jonathan> If NQP is to remain a subset of Perl 6 and remain free of assignment, not sure how that'll play out.
<jonathan> Or I might be missing something. :-)
<TimToady> well, to the first approximation, the meaning of binding doesn't change at all, just the s/':='/<-/
<jonathan> Oh, of course. The ambiguity wiht with :=, not <-.
<TimToady> and we'll just teach the mathematicians to put whitespace around < like they already should be
<jonathan> Sorry.
<jonathan> Oh, no.
<TimToady> if you don't have assignment at all, you don't have any ambiguity yet. :)
<jonathan> It is with <-
<jonathan> True.
<TimToady> and we already require whitespace before <
<TimToady> so anyone who writes $x< is already hosed
<TimToady> and nobody's gonna write $x <-1...well, almost nobody
<TimToady> I'm sure someone will manage...
<jonathan> Yeah. :-S
<jonathan> Maybe the person correcting the compiler error they got from $x<-1 by putting a space after the x.
% Theory has joined #parrot
<svnbotlt> r25436 | kjs++ | trunk:
<svnbotlt> : [cardinal] add basic function definition (but have to finish and check). + small refactoring for identifiers.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25436
<TimToady> I think <- as binding also enhances teachability
% Theory has left Theory!~Theory@dsl093-038-250.pdx1.dsl.speakeasy.net
% alvar has joined #parrot
% alvar has left alvar!~alvar@sub87-230-120-57.he-dsl2plus.de
% alvar has joined #parrot
% alvar has left alvar!~alvar@sub87-230-120-57.he-dsl2plus.de
% cognominal_ has joined #parrot
% particle has left particle!~particle@c-24-19-3-148.hsd1.wa.comcast.net
% alvar has joined #parrot
% davidfetter has joined #parrot
% particle has joined #parrot
<particle> TimToady: what about ,= ?
<particle> i've thought of implementing push as ,= but it does feel kinda listy
<particle> s/but/as/
<jonathan> Given , constructs a list, ,= would feel to me to behave more like any other op-then-=
<jonathan> @foo ,= 4,5,6; # @foo = @foo,4,5,6
<svnbotlt> r25437 | kjs++ | trunk:
<svnbotlt> : [cardinal] Implement if statement and fix a bit with functiondefs.
<svnbotlt> : This is too easy :-)
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25437
<jonathan> Argh. Just changed a getclass I was emitting to a get_class since the former is deprecated and...breakage.
% Limbic_Region has joined #parrot
<svnbotlt> r25438 | kjs++ | trunk:
<svnbotlt> : [cardinal] get basic function call working, but not multiple statements.
<svnbotlt> : + added statement modifiers (but not working yet).
<svnbotlt> : + add "print" builtin.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25438
<svnbotlt> r25439 | kjs++ | trunk:
<svnbotlt> : [cardinal] forgot to test before commit :-| fixed syntax error and add var. decl.
<svnbotlt> r25440 | kjs++ | trunk:
<svnbotlt> : [cardinal] forgot something in comp_stmt, which is now fixed and now the tests are passing. Yay!
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25440
<Limbic_Region> interesting - when did the decision to log the channel happen?  I thought it came up just yesterday
<jonathan> Limbic_Region: I think it did, and then there was a rather fast decision...
* Limbic_Region doesn't mind
<Limbic_Region> it just seems odd that after years of having it come up with the same disposition it would change
<kjs_> where can these logs be found?
<svnbotlt> r25441 | kjs++ | trunk:
<svnbotlt> : [cardinal] 
<svnbotlt> : * small update to test, which now uses a stmt modifier
<svnbotlt> : * added "alias" function (guessed about impl. can't be very wrong)
<svnbotlt> : * allow newlines at top of input.
% DarkWolf84 has joined #parrot
<jonathan> One core Parrot fix later, I've got a working first cut of compile-time role composition in Perl 6.
<svnbotlt> r25442 | kjs++ | trunk:
<svnbotlt> : [cardinal] check for reserved words in identifier.
<svnbotlt> : + add "not expr".
<svnbotlt> r25443 | jonathan++ | trunk:
<svnbotlt> : [core] Generalize Parrot_oo_extract_methods_from_namespace to take the namespace as a parameter, so we can use it in the Role PMC too.
<particle> kjs: /topic
<particle> lunch &
% cognominal_ has left cognominal_!~cognomina@82.67.232.89
% cognominal_ has joined #parrot
% kjs_ has left kjs_!~IceChat7@ip565fd420.direct-adsl.nl
<svnbotlt> r25444 | chromatic++ | trunk:
<svnbotlt> : [src] Add header change missing from r25443 (namespace argument added to
<svnbotlt> : Parrot_oo_extract_methods_from_namespace).
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25444
<jonathan> Oops. Did the header change and then only checked in src/
<svnbotlt> r25445 | jonathan++ | trunk:
<svnbotlt> : [nqp] Add =:= operator to NQP.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25445
<svnbotlt> r25446 | jonathan++ | trunk:
<svnbotlt> : [rakudo] Initial implementation of roles and compile time role composition. Has an ordering issue, as with the inheritance patch earlier, which may be a PAST -> POST translation bug. Simply compiles to use Parrot's role implementation directly.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25446
% lichtkind has joined #parrot
<lichtkind> @seen fglock
% lichtkind has left #parrot
% mj41 has left mj41!chatzilla@pc-jurosz.ro.vutbr.cz
% iblechbot has left iblechbot!~iblechbot@ppp-62-216-200-29.dynamic.mnet-online.de
% tcoppi has left tcoppi!~thisnukes@dhcp-west-76.resnet.nmt.edu
% tcoppi has joined #parrot
% jjore has left jjore!~jjore@c-24-16-241-176.hsd1.wa.comcast.net
% Theory has joined #parrot
% Andy has joined #parrot
% slightlyoff has joined #parrot
% IllvilJa has joined #parrot
% avar has left avar!avar@hlagh.mtfnpy
% avar has joined #parrot
% IllvilJa has left IllvilJa!~jilves@emea-netcache1.oracle.co.uk
% szbalint has joined #parrot
% HG`` has left HG``!~wells@user-544636b6.lns2-c12.dsl.pol.co.uk
<svnbotlt> r25447 | petdance++ | trunk:
<svnbotlt> : use STREQ() macros
<svnbotlt> r25448 | petdance++ | trunk:
<svnbotlt> : check a size in the preprocessor rather than C code
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25448
<svnbotlt> r25449 | petdance++ | trunk:
<svnbotlt> : check for capabilitieis in the preprocessor instead of C code
<svnbotlt> r25450 | petdance++ | trunk:
<svnbotlt> : localizing and consting
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25450
% alvar has left alvar!~alvar@sub87-230-120-57.he-dsl2plus.de
<svnbotlt> r25451 | petdance++ | trunk:
<svnbotlt> : use the STREQ macro
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25451
<svnbotlt> r25452 | petdance++ | trunk:
<svnbotlt> : Mark some functions as PARROT_PURE_FUNCTION, and make Parrot_forall_header_pools an ignorable result.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25452
% jjore has joined #parrot
<Andy> WHO LOVES TO TWEAK SOURCE CODE?
<Andy> WHO?
<Tene> You do?
<Andy> I DO, YES!
<Andy> I'm baffled by some of the warning: will never be executed
<Tene> I'd like a warning like that applied to me.
<Andy> haha
* Tene is looking for something parrot-related to hack on tonight.
<svnbotlt> r25453 | petdance++ | trunk:
<svnbotlt> : shimmed some unused interps, and fixed a few ARGINs that should have been ARGMODs
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25453
<svnbotlt> r25454 | petdance++ | trunk:
<svnbotlt> : consted & localized.  Removed my first HISTORY block.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25454
% ruoso has left ruoso!~ruoso@a81-84-26-6.cpe.netcabo.pt
% japhb has left japhb!~geoff@76.191.190.8
% Andy has left Andy!~Andy@64.81.227.163
% Theory has left Theory!~Theory@dsl093-038-250.pdx1.dsl.speakeasy.net
<Tene> Anyone up to help me figure out some PCT/pir stuff?
* Tene trawls `find . -name '*.pir'` for examples to copy.
% japhb has joined #parrot
% Limbic_Region has left Limbic_Region!~Limbic_Re@c-68-49-236-220.hsd1.md.comcast.net
% Andy has joined #parrot
* teknomunk_afk is back.
% teknomunk_afk is now known as teknomunk
% slightlyoff has left slightlyoff!~slightlyo@204.14.154.209
% slightlyoff has joined #parrot
<Tene> Sure, everybody is active on weekday mornings and gone on Saturday nights.
<Tene> lolcode's AST for function declarations is weird.
<Tene> Working on making it more like rakudo's.
<Tene> I really need to learn either svk or git-svn
<arcady> Tene there's a git mirror of parrot on repo.or.cz
<arcady> and it seems like it's kept up to date
<Tene> Hmm.  That's an option.  Thanks.
<svnbotlt> r25455 | petdance++ | trunk:
<svnbotlt> : Show the number of PARROT_API that we found for --apilist
<svnbotlt> r25456 | petdance++ | trunk:
<svnbotlt> : ld() certainly doesn't need to be part of the API
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25456
<svnbotlt> r25457 | petdance++ | trunk:
<svnbotlt> : removed an extra return from the function
<svnbotlt> r25458 | petdance++ | trunk:
<svnbotlt> : localizing vars
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25458
<svnbotlt> r25459 | chromatic++ | pdd17pmc:
<svnbotlt> : [lib] Minor cleanup in Parrot::Pmc2c::PMCEmitter.  It's tidier and slightly
<svnbotlt> : easier to read now.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25459
% Theory has joined #parrot
<peepsalot> does parrot support perl 5 in some way?  i read there was a ponie project that is now dead?
<Tene> peepsalot: yeah, ponie is dead.
<peepsalot> and there is nothing else at the moment that is set to take it's place in supporting perl 5?
<Andy> I think I understood about 4% of what chromatic proposed.
<svnbotlt> r25460 | petdance++ | trunk:
<svnbotlt> : Consted.  Removed some duplicated CONST_STRING calls.
<svnbotlt> diff: http://perlsix.org/svn/parrot/revision/?rev=25460
<arcady> ponie was more along the lines of supporting parrot in perl 5
<arcady> really, gradually replacing the guts of the perl interpreter with parrot stuff
<Tene> wow, I've been fighting with myself for at least two hours.
<Tene> Oh, right, I still can't get the arity of anything.
<Tene> The PAST indicates that the arity is being set properly, but in the expression handler, .arity on the sub returns 0.
<Tene> ... oh, all the params are being set to :optional
<Tene> Okay, got that all worked out.