r29870 | allison++ | pdd25cx: : [pdd25cx] Updating new code from merge to new exception syntax. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29870 % TiMBuS has joined #parrot does t/spec/S12-class/attributes.t fail for anybody else? % iblechbot has joined #parrot moritz: no. I get a segmentation fault moritz: bad luck :) I'm wondering if my checkout is seriously b0rked somehow or perhaps my ~/.ccache moritz: looks like... Or maybe you are on some branch I'm not % Ademan has left Ademan!~dan@h-67-101-43-213.snfccasy.dynamic.covad.net moritz: make realclean? it has been said that make realclean is sometimes need in parrot to work against missing dependencies bacek: did that already ... multiple times ;) moritz: definitely bad luck :) % Debolaz has left Debolaz!~root@nat.andersberle.com % barney has joined #parrot r29871 | fperrad++ | opengl4lua: : experiment another way of OpenGL in Lua diff: http://www.parrotvm.org/svn/parrot/revision?rev=29871 % Ademan has joined #parrot % brettles has left brettles!~brett@access.hillsong.com % daizisheng has left daizisheng!~c066b722@67.159.55.26 % barney has left barney!~bernhard@p549A3707.dip0.t-ipconnect.de % rblackwe_ has joined #parrot % rblackwe has left rblackwe!rblackwe@where.is.allmydata.net % iblechbot has left iblechbot!~iblechbot@228.16-dial.augustakom.net % masak has joined #parrot % bacek has left bacek!~bacek@mcas-151.usr.optusnet.com.au % ruoso has left ruoso!~ruoso@201009046064.user.veloxzone.com.br % donaldh has joined #parrot % cognominal has left cognominal!~cognomina@82.67.232.89 % kj has joined #parrot % cognominal has joined #parrot % brettles has joined #parrot r29872 | kjs++ | trunk: : [pirc/new] update parser a bit based on revision of pdd19. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29872 % iblechbot has joined #parrot % masak has left masak!~user@nl119-201-126.student.uu.se % cognominal has left cognominal!~cognomina@82.67.232.89 % bacek has joined #parrot % Casan has left Casan!~casan@users163.kollegienet.dk % Whiteknight has joined #parrot % cognominal has joined #parrot % jan has left jan!~chatzilla@89-253-66-101.customers.ownit.se % jan has joined #parrot % s1n has left s1n!~s1n@75-16-180-101.lightspeed.allntx.sbcglobal.net % s1n has joined #parrot % Psyche^ has joined #parrot % Patterner has left Patterner!~Psyche@e177237204.adsl.alicedsl.de % Psyche^ is now known as Patterner r29873 | jkeenan++ | trunk: : Skip 39 tests if OS is Sun or Solaris. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29873 r29874 | jkeenan++ | trunk: : SKIP block for tests not yet passing on certain OSes. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29874 % masak has joined #parrot % TiMBuS has left TiMBuS!~Hurf@123-243-167-27.static.tpgi.com.au % ruoso has joined #parrot % Whiteknight has left Whiteknight!~nobody@c-71-230-33-251.hsd1.pa.comcast.net % ruoso has left ruoso!~ruoso@201009046064.user.veloxzone.com.br % bacek has left bacek!~bacek@122.110.2.31 % ruoso has joined #parrot % brettles has left brettles!~brett@60-242-148-254.static.tpgi.com.au % iblechbot has left iblechbot!~iblechbot@ppp-62-216-201-119.dynamic.mnet-online.de % teknomunk has joined #parrot % jq has left jq!~jquelin@merlin.mongueurs.net % gryphon__ has joined #parrot % jq has joined #parrot r29875 | kjs++ | trunk: : [pirc/new] more updates based on recent revision of pdd19 : * removal of pasm-style registers (only symbolic now) diff: http://www.parrotvm.org/svn/parrot/revision?rev=29875 % rdice has joined #parrot % davidfetter has joined #parrot % teknomunk has left teknomunk!~teknomunk@r74-195-239-111.stl1cmta01.stwrok.ok.dh.suddenlink.net % Casan has joined #parrot % iblechbot has joined #parrot % cxreg has left cxreg!~count@62.f9.1243.static.theplanet.com % UltraDM has joined #parrot % jhorwitz has joined #parrot % cxreg has joined #parrot % Andy has left #parrot % uniejo has left uniejo!~uniejo@193.88.64.250 % cxreg has left cxreg!~count@62.f9.1243.static.theplanet.com % Whiteknight has joined #parrot % davidfetter has left davidfetter!~davidfett@start.fetter.org % cxreg has joined #parrot % cxreg has left cxreg!~count@62.f9.1243.static.theplanet.com % UltraDM has left UltraDM!~george@dasapass.avira.com % Ademan has left Ademan!~dan@h-67-101-44-168.snfccasy.dynamic.covad.net % cxreg has joined #parrot % cxreg has left cxreg!~count@62.f9.1243.static.theplanet.com % Ademan has joined #parrot % masak has left masak!~user@nl119-201-126.student.uu.se Hi. I received mail "[svn:parrot] r29867 ... " from svn-commit-parrot@perl.org 10 times. Probably a loop or something. 10 times is pretty much ;) 10 is a little much, but I've also been seeing a lot of duplicates. % cxreg has joined #parrot % wolverian has left wolverian!wolverian@feather.perl6.nl % dalek has left dalek!dalek@feather.perl6.nl % wolverian has joined #parrot % dalek has joined #parrot % Whiteknight has left Whiteknight!~nobody@c-71-230-33-251.hsd1.pa.comcast.net % cxreg has left cxreg!~count@62.f9.1243.static.theplanet.com % Theory has left Theory!~Theory@c-67-160-131-113.hsd1.wa.comcast.net % Theory has joined #parrot ace. t/pmc/os.t is failing on _one_ of my _two_ Cygwin platforms. mmm. different 'expected' output from Perl 5 stat() % nilsl has joined #parrot % nilsl has left #parrot donaldh: bug in some rev of cygwin perl? Infinoid: I'm just tracking it down. r29876 | fperrad++ | trunk: : [Lua] : - give a change to run tests : (lua.pbc is the only mandatory .pbc) diff: http://www.parrotvm.org/svn/parrot/revision?rev=29876 Infinoid: It looks like some rev of cygwin perl is behaving differently for print join( "\n", stat("xpto") ), "\n"; In other words, the default conversion from a number to a string is behaving differently. donaldh: sounds like a testing bug It's possible that one is compiled with different type sizes. % cjfields has joined #parrot I'll need to compare perl -V donaldh: btw if you want to get a commit bit for parrot you have to send all your helpful patches to RT, otherwise most of the developers never see them donaldh: (and send a CLA) moritz: thanks. will do. % cxreg has joined #parrot Okay. Converting the test to $stat = sprintf("%08x" x 13, stat('xpto')) fixes it. donaldh: send patch ;-) I hear you! ;-) % cxreg has left cxreg!~count@62.f9.1243.static.theplanet.com % ruoso has left ruoso!~ruoso@201009046064.user.veloxzone.com.br donaldh: are the numbers different, or is it a CRLF line ending thing? The numbers are different. One comes back negative. awesome. I'd suggest "%08x "x13 though, with the space to make it clear where the numbers begin and end instead of newlines? the sprintf() line you mentioned above has no newlines maybe I misunderstood. Sorry. Bad typing. Was sprintf("%08x\n" x 13, ... oh. that works :) donaldh++ Can you believe I didn't even know about the x operator this morning:-0 and already put to good use ;) 10+ years of perl programming and never found it before. it comes in handy. sort of a poor man's hyperoperator Well I found parrot has a repeat opcode and thought 'perl must do this'. wow. I only have about 4 years of perl experience % cxreg has joined #parrot and I keep finding features now and then r29877 | kjs++ | trunk: : [pirc/new] use flex' built-in line-number-track-keeping option instead of custom code to do that. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29877 Yeah, I read about the differences between Perl 5 and Perl 6 and found myself thinking 'wow, Perl 5 does that?' ;) % cxreg has left cxreg!~count@62.f9.1243.static.theplanet.com r29878 | kjs++ | trunk: : [pirc/new] rename TK_SYM_?REG to TK_?REG : + remove update_location() function, as this is built-in in flex. : + fix alignment a bit : + change alias of TK_PARROT_OP into "parrot instruction" for better error message. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29878 r29879 | moritz++ | trunk: : [rakudo] remove obsolete t/fetchspec, as of RT #57324 diff: http://www.parrotvm.org/svn/parrot/revision?rev=29879 % donaldh has left donaldh!~chatzilla@proxy-sjc-1.cisco.com % Ademan has left Ademan!~dan@h-67-101-44-168.snfccasy.dynamic.covad.net So, I've recieved 11 copies so far of r29867 from the svn-to-mail list. Anyone else getting that? Tene: mj41 reported ~10 copies % Ademan has joined #parrot % masak has joined #parrot is there an easy way to debug a parser produced by mk_language_shell (i.e., where/why a parse is failing)? % cjfields has left cjfields!~cjfields@cjfields.igb.uiuc.edu the parse fails? hrmm you want to add debugging to the parser, then ah, ok. there's none of that built into the produced parser? look at the rule i'm sure it's that i'm rather unfamiliar with the changes to perl6 regex/rules, but was hoping if there was an easier way... will do. thanks. like, token foo { 'foo' || } in general, you can use --target=parse to inspect your parse tree, but that's only if the parse succeeds right, that's my problem :) figured. so that's where panic comes in handy oops, should be <.panic: ...> (leading dot) i'm not seeing that work with or without the dot... can you nopaste your grammar? nopaste? i heard nopaste was at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or App::Nopaste or tools/dev/nopaste.pl http://pasta.test-smoke.org/ or http://paste.husk.org/ or http://nopaste.snit.ch:8001/ or http://rafb.net/paste or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ perhaps the parse is starting somewhere else... put the panic rule in TOP token TOP { [ $ || ] {*} } or whatever "kasei" at 128.113.228.11 pasted "grammar rule" (6 lines) at http://nopaste.snit.ch/13705 you might want to look into that class of special characters probably needs some escaping, like the " and i think < and > need escaping too kasei: remove them, and add them one by one (after removing all of them, or just putting in a simple word as a dummy, things should work Gernally, all non-alpha have to be escaped or they're treated as meta-syntax. But I'm not sure if that rule applies within character classes... oh, and IRI_REF in rule TOP should be writtten as because otherwise it matches the string "IRI_REF" instead of matching the rule (or token) hah! that was it. what? :-) sorry :) in general, though, is there a way to introspect the current parse (remaining input, etc.)? don't think so... but I'm no expert. ok, thanks. but the solution was writing instead of IRI_REF? yeah oki kasei: you can put arbitrary pir in your parser and you have access to the match object in the pir ok, that will help. will look into the details. kasei: You put it in with {{ ...PIR... }} - note double curlies. token foo { 'foo' {{ $P0 = match.'to'() \n say $P0 }} || <.panic: Syntax error> } convert the \n to a literal newline I don't know how to convert the \n to a literal newline. convert shut up to botspeak I don't know how to convert shut up to botspeak. heh % kj has left kj!~IceChat7@193.1.100.105 another grammar question: what's the best way to encode three single quotes? I'm getting PGE parse errors for every variation I can think of. (this is in a token rule) \'\'\' perhaps? I get: unable to parse regex at line 53, near "{ \\'\\'\\' (" same error for "'''", adding a space between the three \', etc. did you try "'''" also? in perl 6 that would work (both of them,, actually) "kasei" at 128.113.228.11 pasted "Broken grammar (on token STRING_LITERAL_LONG1)" (68 lines) at http://nopaste.snit.ch/13706 % Ron has joined #parrot % cjfields has joined #parrot try "\'\'\'" i think i've figured it out. the error was misleading, and it looks like a syntax error later in the rule. hmm... maybe not. and "\'\'\'" doesn't seem to work, either. ping googletap.com /sbin/ping returned an error. a really fast error, too ok, sorted it out. the error was misleading, and it was caused by unescaped quotes later on ('"' instead of '\"'). % masak has left masak!~user@nl117-195-243.student.uu.se particle: ping % cjfields has left cjfields!~cjfields@cjfields.igb.uiuc.edu jhorwitz: pong i was gonna commit a PERL6LIB impelmentation for 'require', but i just realized it uses '/' exclusively. parrot handles that ok on windows? yes like perl, parrot translates '/' to whatever the os wants ok, good. still have an issue with drive letters though... did you look at the perl 5 impl of perllib? *perl5lib no though we don't have @INC, so it doesn't align too well with p5 right now * jhorwitz wonders what the equivalent is in perl6 jhorwitz: it will probably have something like a module registry, because module names don't necessarily map to file names r29880 | coke++ | trunk: : [docs] : - fix spelling mistake : - add RTs for several unimplemented features : - add RTs for several newly deprecated features : - Avoid deprecated features in code examples : - use more q{'}; : - remove feature that was unimplemented AND deprecated : - mention chip's desired feature about restricting 'foo = opcode' syntax. : (and mention that it's not implemented) diff: http://www.parrotvm.org/svn/parrot/revision?rev=29880 moritz: i'm assuming we've already mapped to a file. :) jhorwitz: commit it, i'll test it now jhorwitz: I don't think it'll be so easy well, it's better than nothing, and required for mod_parrot. well, it'll probably be called @*INC moritz: how so? i'm only doing this at the "require" level, and then only if you've "required" a path. it's just one piece of the puzzle. right now rakudo's "us"e translates to a path and calls "require" *use jhorwitz: ok, then I'm fine with it jhorwitz: I just don't think it's a good long term solution jhorwitz: have you coordinated with pmichaud on this? he said he was going to work on this on the airplane today. iirc. Tene: he's working on require/PERL6LIB? i told him i was going to commit this at OSCON.... Ah, perhaps I'm misremembering. 02:27 <@pmichaud> generic planning note: my plan at the moment is to get standalone/precompiled modules working in Rakudo tomorrow while on the plane 02:27 <@pmichaud> this will involve changes to PCT, P6object, and rakudo yes, that's unrelated r29881 | chromatic++ | trunk: : [t] Ensured that the ICU configuration tests clean up all temporary directories : (Andy Dougherty, RT #57386). diff: http://www.parrotvm.org/svn/parrot/revision?rev=29881 Ah. i believe that's so you can, for example, output PIR from the rakudo compiler and run it independently particle, pmichaud: #perl6-soc in 2 actually, it's to speed up the test suite moritz: "use" is pretty useless right now so i see this as a stop-gap solution. it works, and we can tear it out later if needed. among other things jhorwitz: ok, agree cool 'use' is way lower on the roadmap, so anything now is just a prototype and hacky * jhorwitz can attest to that ;-) r29882 | jhorwitz++ | trunk: : [rakudo] 'require' should check paths in PERL6LIB diff: http://www.parrotvm.org/svn/parrot/revision?rev=29882 * particle rebuilds parrot and rakudo particle: Are you on Win32 at the mo? yep particle: Could you try, once your build is done, running S03-operators\value_equivalence.t (IIRC) under the harness? Well, fudge it and run the .rakudo. It was passing all tests that were not fudged for me last night. But the harness reported "dubious". jonathan: it works fine here moritz: Linux? somebody said Linux was linux is linux by another name. i'm currently desktop($work) laptop(building parrot), in online meeting, on phone call, listening to radio, and lunch. jonathan: yes particle: OK, whenever you have time. :-) * jonathan isn't in a hurry. moritz: You can add it to spectest_regression.data if you want, I guess... moritz: Since this suggests it's not a problem with Rakudo itself. jonathan: ok, will do % rdice has left rdice!~richard_d@CPE0014bfafbbd5-CM0011e6ecf48a.cpe.net.cable.rogers.com moritz: It nudges us a little closer to 2,000. ;-) r29883 | particle++ | trunk: : [rakudo] update svn:ignore on src/ops/ diff: http://www.parrotvm.org/svn/parrot/revision?rev=29883 r29884 | particle++ | trunk: : [rakudo] small refactor for clarity diff: http://www.parrotvm.org/svn/parrot/revision?rev=29884 * particle realcleans parrot, due to all dubious results on spectest_regression % wknight-work has joined #parrot particle: You're getting many of them? :-S % wknight-work has left wknight-work!~chatzilla@66.252.102.38 particle: I will probably miss today's call (I'm already on the plane) my report is basically the same as #parrotsketch r29885 | moritz++ | trunk: : [rakudo] added S03-operators/value_equivalence.t to spectest_regression. : Also restored alphabetical order of files in spectest_regression.data diff: http://www.parrotvm.org/svn/parrot/revision?rev=29885 safe travels, pmichaud jonathan: every single test returns status 5 from spectest_regression, that is the plane isn't supposed to depart for another 25 mins, but we're already boarded :-) * particle calls in now pmichaud: Have a good journey, and a great vacation. I'll call in a second, after I've finished sync'ing mail and svn for the flight. (and assuming I can) btw "ocean sushi" in hilo (big island) is fabulous and cheap, if the kids eat sushi they love sushi katie already says "yes, yes yes" .oO( tell the kids one of the weird hackers said "hi" ;) :) oh, the flight is leaving early! wow! everyone's on board already sixperl says enjoy vacation % Ron has left Ron!~rblasch@80-121-78-64.adsl.highway.telekom.at okay, time to disconnect have a safe trip, pmichaud! see you all later bye r29886 | coke++ | trunk: : [docs] add another RT for a newly deprecated item diff: http://www.parrotvm.org/svn/parrot/revision?rev=29886 % jhorwitz has left jhorwitz!~chatzilla@96.245.16.64 % leo has left leo!lt@feather.perl6.nl % leo has joined #parrot % AndyA has left AndyA!~andy@ca93nt.hexten.net r29887 | coke++ | trunk: : [tcl] : - s/'tcl_group'/''/ in preparation for the deprecation of this version of .HLL : - convert some (broken) stub precompiled versions to stub runtime versions : - cleanup some comments. : - use more 'elements' diff: http://www.parrotvm.org/svn/parrot/revision?rev=29887 % AndyA has joined #parrot % gryphon__ has left gryphon__!~gryphon@dsl-209-221-185-54.zipcon.net % Whiteknight has joined #parrot % gryphon__ has joined #parrot % teknomunk has joined #parrot r29888 | coke++ | trunk: : [t] update test name to reflect difference from identically named test diff: http://www.parrotvm.org/svn/parrot/revision?rev=29888 % gryphon__ has left gryphon__!~gryphon@dsl-209-221-185-54.zipcon.net r29889 | coke++ | trunk: : [tcl] Add empty stubs for most of the remaining builtins. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29889 wow, lots of files updating r29890 | Whiteknight++ | gsoc_pdd09: : [gsoc_pdd09] update to trunk r29889 diff: http://www.parrotvm.org/svn/parrot/revision?rev=29890 % iblechbot has left iblechbot!~iblechbot@67.17-dial.augustakom.net % grim_fandango has joined #parrot how do you mark a test as "todo" or mark a test to be skipped? I can never remember how to do it, and I can never find it in documentation anywhere its a tweak to the test's description I think its just a TODO: prefix skipped tests are different, you put them in a SKIP block in perl5 for parrot tests, todo often looks like: pir_output_is( <<'CODE', <<'OUT', 'error', todo => 'not yet implemented' ); you can grep/ack for these, they're everywhere in parrot's t/ :) I'm trying to TODO a whole test file t/op/gc.t exposes problems that I'm trying to ignore oh. would skip_all work? use Parrot::Test skip_all => 'ongoing GC changes'; i can try that it'd get them out of your way, at least. I might want to keep them as TODO, in case they miraculously start passing I don't know of a way to do that for a whole test script at once, but you can tweak the individual tests to do this okay, thanks. I'll play with it okay, putting it all into a TODO: { } block seems to be working % Limbic_Region has joined #parrot great, glad to hear it. by the way, this is all documented in "perldoc Test::More" Yeah, I was looking at that, and was having a hard time distilling the exact answers I was looking for maybe I'm just dense you found the right answer :) Infinoid: How's the packfile stuff going (if you're getting time to work on it, that is)? % kid51 has joined #parrot jonathan: made some progress, not much time. partial introspection is there and working in pir Nice! Are you keeping the branch sync'd with trunk at all? yeah, I do a summary merge once in a while Great. Thanks for working on it! I need more time... :) % brettles has joined #parrot I've worked on it, like, 1 or 2 days in the last month Time. There's never *quite* enough of it... actually, it's proving pretty straight-forward to just wrap the code and data structures that are already in place I was pleasantly surprised by that. http://www.parrotvm.org/svn/parrot/view/branches/pdd13pbc/t/pmc/packfile.t shows the bits which work so far to match bytes \x00..\x20 in a grammar rule, do I need to do something other than put that in a character class? is there syntax for matching bytes? Infinoid: Looking good. :-) Infinoid: The hard part, will be trying to replace the guts that use the packfile, to access it through the PMC interface in the appropriate places. yeah, and that also includes side tools like pdump salutations all I figure I'm going to learn a whole bunch of parrot internals along the way. Sure Hopefully I'll have some tuits to hack on it a bit with you too. I'm taking a week of vacation next month, and should have some decent parrot hacking time then... so hopefully it will move forward a bit Nice. * jonathan is having away-from-computer vacation for a week at end of August. % Dfa-Dom has joined #parrot "away-from-computer vacation"? dear god, why? * Infinoid boggles at the concept % Dfa-Dom has left #parrot Infinoid: I find I need to physically get away from things for a while to be able to relax properly. Plus, it keeps the RSI at bay. I've discovered I'm susceptible to both RSI and burnout, if I don't take such breaks. RSI? RSI is Repetitive strain injury you find information at http://en.wikipedia.org/wiki/Repetitive_strain_injury purl is useful?!?! don't expect it to happen regularly RSI sucks. Repeatedly. r29891 | jkeenan++ | trunk: : Responding to reports of failures of one test in : http://rt.perl.org/rt3/Ticket/Display.html?id=56716 and Smolder and smoke : tests. Relaxing test to look for definedness of result, rather than matching : a particular result string. diff: http://www.parrotvm.org/svn/parrot/revision?rev=29891 % Ademan has left Ademan!~dan@h-67-101-44-168.snfccasy.dynamic.covad.net Tene: Yes. :-| Actually, the thing that really seemed to help mine, was getting one of those split keyboards. I used to use two keyboards, one on each side of me. That was very comfortable. I get RSI from pulling too many ropes when sailing, but not from typing, strangely. as for burnout... changing projects a lot seems to help for that ...and changing languages Yes, agree, though I seem to have got myself involved in a somewhat...long running...one that probably doesn't want to lose me. ;-) Changing languages only bothers me if I've had to deal with a hateful language. after a few years of fixing other people's broken code and bad designs, C has become such a language for me. :( That sounds understandable. The only project I ever write C for these days is Parro.t *Parrot well, at least it has a nice runtime environment (Parrot C code does, I mean) Yes, it's not so bad. There are some hateful areas of the Parrot codebase (IMCC), though. IMCC is a no-go area for me. I've tracked a couple of bugs into the borders of IMCC, and then ran around trying to figure things out and eventually just gave up. sailboat race & is IMCC going to be phased out in favor of pirc, or is that still up in the air depending on which side gets more tuits? One plan is to go straight from POST to PBC, iirc. Yeah, that's a plan too. I think nobody really wants to spend tuits trying to refactor IMCC into something better, but there aren't currently, it seems, enough tuits spare to go into pirc or POST -> PBC. Though it's generally agreed that we want to move away from IMCC eventually. by Christmas? When Perl 6 is out, every day will be like Christmas ! Maybe someone needs to commit something that breaks IMCC to persuade people to reallocate their tuits. That's easy, although not in the way you meant it. Thing is, in my experience, fixing minor things in IMCC, or adding minor things to it, isn't sufficiently long job to make it easier to go and finish PIRC or do POST -> PBC instead. so someone needs to find a major blocker that will necessitate either a massive refactoring or IMCC or a new compiler sounds like a good RT: "IMCC isn't sufficiently broken" :-P s/or IMCC/of IMCC/ % bacek has joined #parrot % Ademan has joined #parrot i can break IMCC, and I can do it convincingly it would be like it never worked in the first place Whiteknight: You think you'll get your GC working eventually? no but I'm going to try! (at least I'm honest) You going to keep working on it after SoC is over? I'm going to have to! But yes, I plan on seeing this project through to the bitter end :) * jonathan is hoping for a sweet end. The deeper I get into my implementation the more problems, vaguarities, and ommissions I find in parrot and the docs so at the very least I'm going to fix some of those r29892 | Whiteknight++ | gsoc_pdd09: : [gsoc_pdd09] reinstate some tests that I've been ignoring, but with a big fat TODO diff: http://www.parrotvm.org/svn/parrot/revision?rev=29892 % davidfetter has joined #parrot r29893 | Whiteknight++ | gsoc_pdd09: : [gsoc_pdd09] major simplification of the Gc_it_hdr data structure, remove unnecessary pointer redirections and dereferences diff: http://www.parrotvm.org/svn/parrot/revision?rev=29893 r29894 | Whiteknight++ | gsoc_pdd09: : [gsoc_pdd09] Some big changes: : * Remove the idea of "cards" entirely. It's expensive and complex, and the net benefits are nil : * Remove all datastructures and macros associated with the old cards : * Simplify arena allocator : * remove the ill-defined "config" member of the Gc_it_data structure, which has never found a use diff: http://www.parrotvm.org/svn/parrot/revision?rev=29894 r29895 | Whiteknight++ | gsoc_pdd09: : [gsoc_pdd09] fix all my function prototypes with make headerizer diff: http://www.parrotvm.org/svn/parrot/revision?rev=29895 % neopallium has joined #parrot % grim_fandango has left grim_fandango!~matt@bas2-kingston08-1128671702.dsl.bell.ca