Camelia, the Perl 6 bug

IRC log for #parrot, 2011-06-30

Parrot | source cross referenced

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:04 whiteknight although, it's going to take a while to get it working
00:19 benabik_ benabik: Just sign off already. :-P
00:36 atrodo bacek_at_work: is there a way to get my box to work without that patch besides adding more memory?
00:36 atrodo oh, and bacek++
00:37 theory left #parrot
00:37 bacek_at_work atrodo, we need this patch anyway. Doesn't matter how much memory do you have.
00:44 atrodo bacek_at_work: then how am i the first to find it?  is it just random on who triggers it?
00:45 bacek_at_work atrodo, totally random
00:45 atrodo interesting
00:45 atrodo looks like ipfy is going to have a large time gap then
00:50 whiteknight yay! Its building!
00:52 dalek parrot/whiteknight/packfilewrapper: 7dc0e22 | Whiteknight++ | / (7 files):
00:52 dalek parrot/whiteknight/packfilewrapper: Move segment-related functions from src/packfile/api.c to new file src/packfile/segments.c. Add a header file pf_private.h to that subsystem for intra-subsystem declarations. src/packfile/api.c is for external-facing interfaces. Things that are private to the subsystem should not be there
00:52 dalek parrot/whiteknight/packfilewrapper: review: https://github.com/parrot/parrot/commit/7dc0e2276d
00:52 whiteknight needs a hell of a lot of cleanup, but the hard work is done
00:55 kurahaupo joined #parrot
00:55 benabik__ joined #parrot
00:58 cotto left #parrot
01:00 benabik_ left #parrot
01:06 whiteknight I can't wait to get this branch merged now
01:06 whiteknight I'm excited about cleaning up this subsystem
01:15 colomon joined #parrot
01:16 cotto joined #parrot
01:17 benabik_ joined #parrot
01:21 cotto ~~
01:22 woosley joined #parrot
01:22 benabik left #parrot
01:22 benabik__ is now known as benabik
01:24 cottoo joined #parrot
01:25 kid51 joined #parrot
01:28 cotto left #parrot
01:28 soh_cah_toa joined #parrot
01:33 colomon left #parrot
01:36 whiteknight I'll keep up with the cleanup tomorrow. Now, bed
01:36 dukeleto joined #parrot
01:38 whiteknight left #parrot
01:38 colomon joined #parrot
01:40 benabik left #parrot
01:40 benabik_ is now known as benabik
01:40 bubaflub joined #parrot
01:44 cottoo left #parrot
01:44 cottoo joined #parrot
01:45 cottoo dukeleto, a bunch of us are in the lobby if you want to hang out
01:45 cottoo is now known as cotto
01:56 dukeleto left #parrot
01:58 colomon left #parrot
01:59 kid51 left #parrot
02:00 colomon joined #parrot
02:01 davidfetter left #parrot
02:09 colomon_ joined #parrot
02:14 cottoo joined #parrot
02:14 colomon left #parrot
02:14 colomon_ is now known as colomon
02:16 cotto left #parrot
02:20 benabik ~~
02:21 contingencyplan left #parrot
02:32 cottoo hio benabik
02:33 benabik hio cottoo
02:33 cottoo I'm going to have some interesting M0 stuff to talk about tomorrow
02:33 benabik Got my Morse keyer to work, so hopefully won't be as distracted by it tomorrow.  :-D
02:33 benabik M0++
02:33 benabik cottoo: Really really interested.  :-D
02:33 cottoo benabik, great.  I'm excited to get something crappy enough to elicit feedback.
02:34 benabik Don't undersell yourself.
02:40 cottoo This is like hitting myself in the hand with a hammer.  Using a slightly higher-level language than M0 feels amazing after writing M0 directly.
02:42 benabik :-D
02:45 lichtkind left #parrot
02:47 soh_cah_toa left #parrot
03:02 c0tt0 joined #parrot
03:02 c0tt0 msg dukeleto chromatic++'s commit bit on parrot.git on github got disabled.  Can you flip it back?
03:02 aloha OK. I'll deliver the message.
03:05 cottoo left #parrot
03:10 Coke is there a desugared way to say $P1['foo'] ?
03:10 sorear try using pbc_disassemble
03:11 sorear ...hah, pbc_disassembe can't disassemble it
03:12 sorear set_p_p_kc P0,P1[?]
03:12 sorear stupid Key pmcs
03:12 sorear can't we kill them yet?
03:13 sorear um
03:13 c0tt0 that'd be keyed get
03:13 sorear WHY do pbc files contain the GC flags word
03:13 sorear "custom_mark"
03:15 c0tt0 is now known as cotto
03:21 leto joined #parrot
03:21 benabik left #parrot
03:23 leto ~~
03:24 dalek parrot/m0-prototype: 087b5c0 | chromatic++ | src/m0/c/ (2 files):
03:24 dalek parrot/m0-prototype: Added skeleton of C-based M0 interpreter.
03:24 dalek parrot/m0-prototype: review: https://github.com/parrot/parrot/commit/087b5c0986
03:24 sorear yay chromatic!
03:25 colomon \o/
03:26 leto woot, M0-in-C lives!
03:27 cotto some day it'll even parse m0b headers
03:27 cotto It'll be great.
03:29 sorear leto?
03:29 sorear you sold your duchy?
03:31 leto sorear: sigh. Evidently my irc problems was a zombie screen process with a duplicate dircproxy process
03:31 * leto kills his zombie
03:31 leto is now known as dukeleto
03:31 dukeleto :: How Leto Got His Duchy Back ::
03:32 sorear opbots names
03:32 sorear opbots trust colomon
03:32 slavorg Ok
03:32 slavorgn Ok
03:32 sorear opbots trust benabik
03:32 slavorg But I already trust benabik
03:32 slavorgn Ok
03:33 dukeleto i find it very odd that chromatic's commit bit on parrot.git got flipped
03:33 * dukeleto attempts to use dircproxy again
03:33 dukeleto left #parrot
03:35 dukeleto joined #parrot
03:35 * dukeleto got his groove back
03:40 benabik joined #parrot
03:49 Coke do they stale out?
03:51 cotto they shouldn't.
03:54 ingy is now known as \o
03:55 \o is now known as ingy
03:55 Coke ingy: ~~
03:56 benabik ingy: o/
03:57 theory joined #parrot
03:57 dukeleto ingy: just pushed my first perl 6 module, and it even has a few passing tests
04:04 dalek parrot/m0-prototype: 80907fe | chromatic++ | src/m0/c/m0_interp.c:
04:04 dalek parrot/m0-prototype: Added m0b file magic number verification.
04:04 dalek parrot/m0-prototype: review: https://github.com/parrot/parrot/commit/80907fea19
04:16 benabik left #parrot
04:18 colomon left #parrot
04:21 dalek parrot/m0-prototype: 23aed14 | chromatic++ | src/m0/c/m0_interp.c:
04:21 dalek parrot/m0-prototype: [m0c] Added m0b version check to file loading.
04:21 dalek parrot/m0-prototype: review: https://github.com/parrot/parrot/commit/23aed141b8
04:21 benabik joined #parrot
04:22 colomon joined #parrot
04:26 jimmy joined #parrot
04:26 jimmy is now known as JimmyZ
04:32 cottoo joined #parrot
04:33 ingy dukeleto++
04:33 cottoo If anyone needs an M0 newbie task, there are a couple that'd be useful.  The new C interp needs tests for all possible exit conditions and it'd be nice to have an optional test that runs the tests against the C interp under valgrind and makes sure there aren't any bad reads or writes.
04:33 cottoo Now that we're starting to get more tests, it'll also be useful to think about the best way to run the M0 tests against multiple interps.
04:34 cotto left #parrot
04:36 cottoo is now known as cotto
04:48 JimmyZ left #parrot
04:50 particle left #parrot
04:51 fperrad joined #parrot
04:52 particle joined #parrot
04:53 cotto left #parrot
04:54 dalek rakudo/nom: 28b06a2 | pmichaud++ | src/core/Exception.pm:
04:54 dalek rakudo/nom: Add Exception.Str, noted by Coke++.
04:54 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/28b06a25df
05:04 plobsing left #parrot
05:19 cotto joined #parrot
05:39 cotto left #parrot
05:44 colomon left #parrot
05:51 cotto joined #parrot
06:10 benabik left #parrot
06:33 contingencyplan joined #parrot
06:33 kurahaupo left #parrot
06:38 mj41 joined #parrot
06:43 theory left #parrot
07:23 fperrad left #parrot
08:16 daniel-s joined #parrot
08:38 Kulag left #parrot
08:39 Kulag joined #parrot
08:50 Kulag left #parrot
08:56 Kulag joined #parrot
09:03 Drossel joined #parrot
09:04 Kulag left #parrot
09:15 Drossel left #parrot
09:20 SHODAN joined #parrot
09:34 fperrad joined #parrot
09:37 dalek rakudo/nom: a7f7e86 | moritz++ | t/spectest.data:
09:37 dalek rakudo/nom: another passing test file
09:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a7f7e86309
09:37 dalek rakudo/nom: 29d90cf | moritz++ | src/core/Exception.pm:
09:37 dalek rakudo/nom: simplify Exception.pm
09:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/29d90cf6d6
10:15 woosley left #parrot
10:22 dalek rakudo/nom: 53ecd28 | jonathan++ | src/pmc/perl6lexpad.pmc:
10:22 dalek rakudo/nom: Handling of $_, $/ and $! in nested blocks. May well need tweaking, but should essentially work, especially for $_.
10:22 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/53ecd287b6
10:22 dalek rakudo/nom: 79d17be | jonathan++ | t/spectest.data:
10:22 dalek rakudo/nom: Two more passing test files.
10:22 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/79d17be26d
10:22 dalek rakudo/nom: 1c85c66 | jonathan++ | src/Perl6/Actions.pm:
10:22 dalek rakudo/nom: Mark mainline as needing fresh lexicals.
10:22 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1c85c662d7
10:22 dalek rakudo/nom: 1b841ba | jonathan++ | NOMMAP.markdown:
10:22 dalek rakudo/nom: Update nommap.
10:22 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1b841ba4b9
10:35 JimmyZ joined #parrot
10:55 dalek rakudo/nom: 028df0a | jonathan++ | src/Perl6/Grammar.pm:
10:55 dalek rakudo/nom: Fix declarations of packages inside evals. Also unbusts the REPL a bit more.
10:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/028df0a8b7
10:55 dalek rakudo/nom: e65e8ff | jonathan++ | t/spectest.data:
10:55 dalek rakudo/nom: Another passing test file.
10:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e65e8ff934
11:03 contingencyplan left #parrot
11:03 JimmyZ left #parrot
11:21 dalek rakudo/nom: b90e016 | jonathan++ | src/core/Mu.pm:
11:21 dalek rakudo/nom: Make Mu.new have slurpy named parameter, and pass that on to bless, in preparation for BUILD support.
11:21 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b90e016eb8
11:21 dalek rakudo/nom: af6b27e | jonathan++ | t/spectest.data:
11:21 dalek rakudo/nom: Another passing test file.
11:21 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/af6b27e5bd
11:23 dalek rakudo/nom: 8f3b029 | pmichaud++ | tools/lib/NQP/Configure.pm:
11:23 dalek rakudo/nom: Update NQP::Configure to perform 'git pull' on branches when needed.
11:23 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8f3b029466
11:23 dalek nqp: 879fe3c | pmichaud++ | src/ (3 files):
11:23 dalek nqp: Add nqpevent opcodes for logging events (e.g., subroutine entry).
11:23 dalek nqp: review: https://github.com/perl6/nqp/commit/879fe3c59d
11:23 dalek nqp: a4083b4 | pmichaud++ | tools/lib/NQP/Configure.pm:
11:23 dalek nqp: Update NQP::Configure to perform 'git pull' on branches when needed.
11:23 dalek nqp: review: https://github.com/perl6/nqp/commit/a4083b42e0
11:35 dalek rakudo/nom: 630c220 | jonathan++ | src/Perl6/SymbolTable.pm:
11:35 dalek rakudo/nom: Re-using constants built from literals: good. Re-using the PAST node every time we see one of them: bad. Can get .named(...) and other such things. So now we make a fresh PAST node each time, and cache the constant itself.
11:35 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/630c220c65
11:39 colomon joined #parrot
11:50 dalek rakudo/nom: e4d01bb | moritz++ | t/spectest.data:
11:50 dalek rakudo/nom: two more passing test files
11:50 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e4d01bbece
11:50 dalek rakudo/nom: 60dcfb9 | pmichaud++ | / (3 files):
11:50 dalek rakudo/nom: Add initial (possibly cheating) version of assignment metaop.
11:50 dalek rakudo/nom:
11:50 dalek rakudo/nom: This version works fine, but doesn't generate a new sub and
11:50 dalek rakudo/nom: might not work out okay with nested metaops.  If not, then we'll
11:50 dalek rakudo/nom: fix this when we fix all the generation of all of the metaops
11:50 dalek rakudo/nom: (as soon as we figure out how/where to do that :-).
11:50 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/60dcfb9c5a
11:59 Coke left #parrot
11:59 Coke joined #parrot
12:00 lucian joined #parrot
12:06 cotto left #parrot
12:08 dalek rakudo/nom: 866d02f | moritz++ | t/spectest.data:
12:08 dalek rakudo/nom: 5 more passing spectest files, pmichaud++
12:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/866d02f3e3
12:09 bluescreen left #parrot
12:10 mtk joined #parrot
12:13 JimmyZ joined #parrot
12:14 benabik joined #parrot
12:20 bluescreen joined #parrot
12:28 kid51 joined #parrot
12:30 SHODAN left #parrot
12:31 cotto joined #parrot
12:33 whiteknight joined #parrot
12:34 ambs joined #parrot
12:38 fperrad left #parrot
12:39 whiteknight good morning, #parrot
12:39 kid51 good morning, whiteknight
12:39 kid51 Did you get a chance to work on vtable_substr
12:40 fperrad joined #parrot
12:40 whiteknight I completely forgot about it last night.
12:41 kid51 http://trac.parrot.org/parrot/ticket/1682
12:42 whiteknight yeah, I know where it is, I just didn't look at it last night
12:42 whiteknight I want to test it once with Rakudo and then I'm going to merge it
12:42 whiteknight so if somebody else beats me to it, that's fine too
12:42 kid51 Thx
12:49 cotto ~~
12:49 whiteknight good morning cotto
12:53 dalek rakudo/nom: a65c17f | moritz++ | src/core/ (3 files):
12:53 dalek rakudo/nom: allow hashes to store Mu values
12:53 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a65c17fe21
12:53 dalek rakudo/nom: 279c947 | moritz++ | t/spectest.data:
12:53 dalek rakudo/nom: one more passing test file
12:53 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/279c94724f
12:54 dalek rakudo/nom: 9d2ad0d | pmichaud++ | src/Perl6/Actions.pm:
12:54 dalek rakudo/nom: Revert 5bce9d, commit 630c22 fixes the overall problem with re-use of
12:55 dalek rakudo/nom: PAST nodes for constants and so this workaround is no longer needed. jnthn++
12:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9d2ad0dc96
13:02 benabik left #parrot
13:03 tadzik hello #parrot!
13:04 whiteknight hello tadzik
13:05 kid51 left #parrot
13:06 bubaflub morning #parrot
13:06 tadzik hello bubaflub
13:06 tadzik I finally have my holidays
13:08 bubaflub tadzik++
13:08 bubaflub how'd the exams go?
13:09 tadzik as expected :) 2/3 passed, physics for september
13:09 tadzik I was permitted to take only one attempt, so I didn't even tried that now, I need some time to learn it properly
13:10 dalek parrot/whiteknight/packfilewrapper: 3440910 | Whiteknight++ | src/pmc/packfileview.pmc:
13:10 dalek parrot/whiteknight/packfilewrapper: rename variable, on suggestion from cotto++
13:10 dalek parrot/whiteknight/packfilewrapper: review: https://github.com/parrot/parrot/commit/3440910c06
13:11 tadzik I can now get back to hacking and everything
13:11 tadzik and eating properly
13:12 tadzik I must've lost a few kgs during the last weeks
13:13 whiteknight Ha, I had that same experience
13:14 tadzik now that I think of it, my food for the last 2 days have been a kebab, few beers and liters of mountain dew
13:14 whiteknight At the end of my first year I was completely out of money, so I bought the only things I could afford: a small bottle of jelly, a small container of peanut butter, and a loaf of white bread
13:14 whiteknight I ate PB&J sandwiches every meal of every day for about a week and a half
13:15 whiteknight on my way home I was shaky, nauseous, and had a terrible headache
13:17 dalek nqp: de3c519 | pmichaud++ | src/ (2 files):
13:17 dalek nqp: Add --nqpevent=<logfile> option to HLL::Compiler.
13:17 dalek nqp: review: https://github.com/perl6/nqp/commit/de3c51919a
13:18 tadzik during the last exam session I almost became nuts, or at least completely broken psychologically, that's why I decided to leave the Physics failed. It was just too much on my head
13:21 cottoo joined #parrot
13:21 cotto left #parrot
13:28 plobsing joined #parrot
13:31 cotto joined #parrot
13:31 cottoo left #parrot
13:38 colomon bubaflub: what do I need to get your GMP interface up and running, just to see if it works for me?  (Before trying to integrate it anywhere, I mean.)(
13:38 bubaflub colomon: you'll need GMP > 4.3, Parrot > 3.2 (i believe) and a copy of Winxed (which comes with recent parrots)
13:38 bubaflub colomon: sorry, i didn't get setup.pir stuff fixed yesterday so you'll have to compile three files on your own
13:39 bubaflub colomon: to run the tests you'll need Rosella but that isn't strictly necessary
13:39 colomon I'd love to run the tests, though, just to make sure everything's working.  :)
13:40 colomon any hints on determining which version of GMP I have installed?
13:41 bubaflub colomon: here's the commands needed to build stuff: https://gist.github.com/1056256
13:41 colomon will it work with GMP 5.0.2?
13:41 bubaflub colomon: yes, that's what i'm using
13:42 bubaflub the GMP API is stable and backwards compatible back to 3 I believe.  i asked other people who maintain bindings and they said they maintain back to 4.2 or 4.3
13:44 colomon bubaflub: where do I get your source from?
13:44 kid51 joined #parrot
13:44 moritz wow, that's confusing... debian ships with libgmp3-dev, and iti says the version is 2:4.3.2+dfsg-1
13:45 moritz from that I could have understoof if the package was named libgmp2-dev or libgmp4-dev, but the 3 is totally opaque to me
13:45 dukeleto ~~
13:46 dukeleto moritz: i think they append numbers to their own packages when they add incompatible versions of the packages, or some junk. That is definitely anti-intuitive
13:46 dukeleto colomon: https://github.com/bubaflub/parrot-gmp
13:47 colomon moritz: yeah, I was getting the same sort of nonsense trying just looking at the library files.  but it appears I got 5.0.2 installed which
13:47 colomon ack, thought I deleted that last not-quite-a-thought!
13:47 dukeleto colomon: i am not sure his setup.pir works yet, search #parrot archives from the last few days for "GMP" to see how to manually compile it
13:48 colomon dukeleto: bubaflub++ already told me how to manually compile it.  :)
13:48 dukeleto colomon: cool
13:48 colomon dukeleto: is the hackathon starting yet?
13:48 dukeleto colomon: i am eating breakfast, but the room is ready. I assume people will filter in soon
13:49 dukeleto colomon: kid51 was trying to get in at 8am, before they opened it
13:49 kid51 Magnolia room is now open
13:53 colomon bubaflub: I get No NCI thunk available for signature `void (ptr, ptr, long)'
13:53 bubaflub colomon: ahhhhhhhhhhhhhhh bummer.
13:53 * kid51 and particle present; where are the rest of you lazy bums?
13:54 colomon kid51: busy hacking in my hotel room.
13:54 bubaflub colomon: the short answer is you need libffi until generate some NCI thunks
13:55 colomon bubaflub: is it just a matter of installing libffi somewhere on my system?
13:55 bubaflub colomon: yep.
13:56 colomon hmmm... that doesn't seem to have helped?
13:56 colomon I installed libffi 3.0.9.
13:56 bubaflub colomon: re-build parrot, the configure step should detect libffi
13:57 colomon ah
13:57 bubaflub colomon: so here's the long version:
13:58 bubaflub colomon: parrot doesn't include (nor could it) every possible NCI thunk which means not every possible C signature is covered; lamentably GMP has functions that are not covered in core parrot.  the real solution is to generate my own NCI thunks, include them with my project, have them compile when you run setup.pir, and dynamically load them into parrot.  the lazy way to do it is to install libffi.  i'm doing the lazy way for now but will
13:58 bubaflub remove that dependency before the end of GSoC.
13:59 colomon fair enough
13:59 bubaflub sorry this is so rough around the edges
13:59 colomon eh, it's what I get for being an early adopter
14:00 colomon kid51: the great thing about me working from my room for the moment is I'm not driving the rest of you nuts with my quest to listen to every recording of fiddler player MacDarra O Raghallaigh off of youtube.  ;)
14:00 colomon but I'll be down soon, there aren't very many videos up, alas.
14:00 dalek parrot-gmp: 582c5f0 | bubaflub++ | README.md:
14:00 dalek parrot-gmp: ugh, libffi is also required. colomon++
14:00 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/582c5f0486
14:01 dalek TT #1761 closed by jkeenan++: Inconsistent release tags
14:01 dalek TT #1761: http://trac.parrot.org/parrot/ticket/1761
14:02 dukeleto bubaflub: you have people wanting to use parrot-gmp. A shiny setup.pir would be very nice...
14:02 bubaflub dukeleto: yeah.  rl interrupted my hacking yeseterday.
14:02 colomon built-in thunks too.  I'd love to have every nom developer using parrot-gmp by next week....  ;)
14:02 dukeleto bubaflub: alas, that happens
14:03 dukeleto bubaflub: colomon++ and I will be in the same room soon, bugging you in parallel
14:03 dukeleto bubaflub: you have been warned :)
14:03 bubaflub dukeleto: i feel like there is a debugging joke here somewhere
14:03 colomon sure: you do the debugging, we'll stop bugging you.  :)
14:05 colomon bubaflub: okay, I ran your examples and my computer says 32 34 342.  is that right?
14:05 bubaflub colomon: yep
14:05 colomon \o/
14:05 bubaflub colomon: if it wasn't working you'd get segfaults or other random explosions
14:05 colomon okay, I'm going to pause youtube and join the hackathon
14:06 bubaflub colomon: open up examples/basic.winxed to see how some of that works
14:06 colomon left #parrot
14:08 fperrad left #parrot
14:09 NotFound bubaflub: the setup file is a lot easier to write in winxed than in pure pir.
14:09 colomon joined #parrot
14:09 bubaflub NotFound: i was heading that direction - especially with the hash notation and whiteknight's Rosella example
14:09 dukeleto bubaflub: NotFound makes a good point
14:09 dukeleto bubaflub: you may be better off with a setup.winxed
14:09 dukeleto bubaflub: BUT, if you want Rakudo to easily be able to subsume parrot-gmp, setup.pir might be better
14:10 bubaflub dukeleto: well, i can include both setup.winxed and setup.pir in the repo
14:10 bubaflub like i do now with some of the other stuff
14:10 dukeleto bubaflub: sure
14:10 NotFound dukeleto: yes, but you don't need to write it, winxed can do.
14:10 whiteknight winxed -o setup.pir -c setup.winxed
14:10 whiteknight check both in
14:11 dukeleto whiteknight++
14:12 bubaflub bingo
14:12 fperrad joined #parrot
14:12 NotFound You can even write and test it in json and the paste it into winxed.
14:13 benabik joined #parrot
14:14 benabik ~~
14:17 dalek TT #1761 reopened by jkeenan++: Inconsistent release tags
14:17 dalek TT #1761: http://trac.parrot.org/parrot/ticket/1761
14:17 dalek TT #1761 closed by jkeenan++: Inconsistent release tags
14:17 dalek TT #1761: http://trac.parrot.org/parrot/ticket/1761
14:19 soh_cah_toa joined #parrot
14:22 dalek parrot/whiteknight/packfilewrapper: 99afbb7 | Whiteknight++ | src/embed.c:
14:22 dalek parrot/whiteknight/packfilewrapper: Fix small issue in Parrot_pbc_read caused when I changed the return type of Parrot_pf_read_pbc_file to PackFile*. Fixes t/src/extend.t failures
14:22 dalek parrot/whiteknight/packfilewrapper: review: https://github.com/parrot/parrot/commit/99afbb7520
14:23 JimmyZ_ joined #parrot
14:23 darbelo joined #parrot
14:25 benabik_ joined #parrot
14:25 colomon soh_cah_toa: https://github.com/colomon/Math-Vector
14:28 bubaflub colomon: that looks neat
14:28 JimmyZ left #parrot
14:28 JimmyZ_ is now known as JimmyZ
14:30 FalseNames joined #parrot
14:31 benabik left #parrot
14:31 benabik_ is now known as benabik
14:31 JimmyZ left #parrot
14:36 JimmyZ joined #parrot
14:41 colomon bubaflub: dukeleto keeps saying you have nqp bindings, too?
14:42 bubaflub colomon: well, nqp examples are planned
14:42 colomon :)
14:43 dalek parrot/soh-cah-toa/hbdb: 74902a2 | soh_cah_toa++ | t/tools/hbdb/cmds.t:
14:43 dalek parrot/soh-cah-toa/hbdb: Removed (hbdb) prompt from cmds.t test
14:43 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/74902a236d
14:48 bubaflub colomon: there was talk of having an NCI system in NQP that doesn't relied upon Parrot's
14:48 bubaflub colomon: but no work in that direction that i know of
14:48 benabik (Reinventing the wheel)--
14:49 colomon bubaflub: I'm talking with pmichaud and jnthn on #perl6 about how to get things into nqp-nom.
14:49 bubaflub colomon: i'm loitering there now
14:50 colomon ah, good.
14:50 pmichaud Reinventing the wheel is what we do.
14:50 pmichaud The whole point is that some of our existing wheels aren't good enough. :)
14:50 dalek rakudo/nom: 6fff467 | Coke++ | src/core/operators.pm:
14:50 dalek rakudo/nom: add enough <eqv> to pass t/01-sanity/06-eqv.t
14:50 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6fff4677aa
14:52 pmichaud And we know that the NCI wheel in Parrot has some.... issues.
14:52 bubaflub pmichaud: indeed.  i'm working with plobsing in getting a more round wheel.
14:52 dalek parrot/hbdb: e7a9429 | cotto++ | config/gen/makefiles/root.in:
14:52 dalek parrot/hbdb: add make target for hbdb tests
14:52 dalek parrot/hbdb: review: https://github.com/parrot/parrot/commit/e7a94296f0
14:52 whiteknight ideally reinvented wheels make their way back upstream to parrot core
14:52 bubaflub colomon: either way, almost all of the code is generated (plus some templates) so it wouldn't be hard to target something else
14:52 pmichaud the other point is that NQP is going to want to work with other backends as well, and will therefore likely want some NCI-based interface for those too (which probably won't be Parrot's NCI)
14:54 bubaflub pmichaud: perhaps an NCI abstraction class?  that way each backend (like Parrot or .NET or JVM or whatever) can implement the messy details
14:54 pmichaud bubaflub: perhaps... although 6model classes are pretty close to the VM already
14:55 pmichaud so we probably don't want too big of an abstraction there.
14:55 * bubaflub knows very little about 6model details
14:56 bubaflub pmichaud: some abstraction i imagine is necessary; i'm thinking more of a unified interface for all NCI shuffling to take place through.  then again no such system is perfect.
15:06 dalek rakudo/nom: 37ff459 | Coke++ | src/core/operators.pm:
15:06 dalek rakudo/nom: add undefine, pass t/01-sanity/06-eqv.t
15:06 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/37ff459759
15:17 darbelo_ joined #parrot
15:17 darbelo left #parrot
15:20 kid51 dukeleto: Are you still experiencing the problem in http://trac.parrot.org/parrot/ticket/2031 (configuration on Gentoo)?
15:25 dalek website: rohit_nsit08++ | JavaScript : Object System
15:25 dalek website: http://www.parrot.org/conte​nt/javascript-object-system
15:39 benabik Important hackathon result: GMP is now pronounce "Gump"
15:41 colomon To avoid confusion with the GIMP, you see.
15:42 whiteknight I always pronounce it "gimp" in my head
15:42 whiteknight the image manipulation program is "the gimp"
15:43 * PerlJam always says "gee em pee" in his head
15:45 whiteknight That's just crazy
15:45 * jnthn__ is crazy too :)
15:52 * sjn agrees :)
15:54 mj41 left #parrot
15:57 dukeleto also, the mascot for parrot-gmp (parrot gump) should be a shrimp boat (as suggested by TimToady++)
15:59 kid51 left #parrot
16:13 SHODAN joined #parrot
16:19 dalek parrot/soh-cah-toa/hbdb: 1fc4ba1 | soh_cah_toa++ | t/tools/hbdb/cmds.t:
16:19 dalek parrot/soh-cah-toa/hbdb: Added test for 'help' command but open3() is causing problems so it needs to be modified in the future
16:19 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/1fc4ba1b96
16:20 theory joined #parrot
16:25 dalek rakudo/nom: 29d4eb1 | jonathan++ | / (4 files):
16:25 dalek rakudo/nom: Implement role punning. Also fixes Foo ~~ SomeRole.
16:25 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/29d4eb1133
16:25 dalek rakudo/nom: 45e6e3b | jonathan++ | NOMMAP.markdown:
16:25 dalek rakudo/nom: Remove a completed nommap item.
16:25 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/45e6e3bcd6
16:27 JimmyZ left #parrot
16:31 dodathome joined #parrot
17:02 estrabd joined #parrot
17:03 colomon for those at the hackathon: http://www.valetgourmet.com/
17:07 cotto I'm thinking about the syntax and name for the M0 overlay language.  For a name I'm thinking "mole".  For syntax, https://gist.github.com/1048210#file_syntax
17:10 cotto dukeleto, ^
17:12 benabik Molé?
17:12 colomon dukeleto: (828) 236 - 3839
17:12 colomon dukeleto: http://www.orientalpavilion​.com/contents.aspx?menuId=4
17:14 cotto holé molé
17:20 benabik cotto: Is there a difference between S and P in M0?
17:21 cotto benabik, only notationally
17:23 benabik Is it worth introducing it at M0?  It seems magical.  ;-)
17:26 hercynium joined #parrot
17:30 benabik consts might be slightly easier to read with an = in there: `const type name = value`
17:31 cotto The reason I didn't do that is because that constants are declarative, so there's no runtime component.
17:35 benabik Well, yes, but it flows together without the punctuation.   Eh, style preference.
17:39 plobsing left #parrot
17:41 benabik s/chunnks/chunks
17:43 benabik Do returns in composed ops affect the chunk using them?  Can composed ops declare constants?
17:44 benabik Okay, I think I'm done.
17:44 cotto benabik, composed ops don't have returns
17:44 benabik Looks pretty nice.
17:44 cotto thanks
17:44 cotto I'm just doing this for soh_cah_toa's notifications.
17:57 dmalcolm joined #parrot
18:02 FalseNames left #parrot
18:23 soh_cah_toa whiteknight: are you gonna be at ppw in october?
18:25 dalek rakudo/nom: 91491d8 | moritz++ | t/spectest.data:
18:25 dalek rakudo/nom: two more passing spec tests
18:25 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/91491d8636
18:30 contingencyplan joined #parrot
18:30 whiteknight ppw?
18:30 soh_cah_toa pittsburgh perl workshop
18:31 whiteknight oh, probably not.
18:31 soh_cah_toa lame
18:31 whiteknight 2011 is not a good year for whiteknight to travel
18:32 cotto left #parrot
18:33 soh_cah_toa sure it is. i think it's b/c you don't want us to discover that you're really just a 50-year-old creep pretending to be a good hacker ;)
18:34 whiteknight I will *definitely* be creepy when I am 50
18:34 whiteknight no, I'm pouring every last dollar I have to my name into buying a new house
18:35 whiteknight and I just found out yesterday that my lovely apartment complex is going to charge us an extra month's rent for breaking lease
18:35 whiteknight we had a "home buyer option" in our lease, which is supposed to make it easier for us to buy a home and get out of the lease if we need to
18:36 whiteknight but the penalty for using the home buyer option is apparently the same as just breaking the lease and walking away from it, except we need to give 60 days notice
18:37 FalseNames joined #parrot
18:38 soh_cah_toa na, i believe you
18:47 mj41 joined #parrot
18:52 davidfetter joined #parrot
19:08 * Coke wonders why it takes so long to clone parrot.git
19:10 tadzik it's huge
19:13 pmichaud these days I often just (1) make an empty directory, (2) copy (recursively) .git from an existing parrot repo into the next directory, (3) chdir to the new directory, (4) git checkout .
19:15 pmichaud bwahahahahah
19:15 pmichaud wrong chan
19:18 dalek nqp: 9a65d79 | pmichaud++ | src/Regex/Cursor.pir:
19:18 dalek nqp: Add an optional match argument to !cursor_pass, to let us explicitly specify the match to be used.
19:18 dalek nqp: review: https://github.com/perl6/nqp/commit/9a65d79573
19:18 soh_cah_toa left #parrot
19:19 soh_cah_toa joined #parrot
19:20 FalseNames left #parrot
19:20 FalseNames joined #parrot
19:22 lichtkind joined #parrot
19:27 benabik aloha?
19:27 benabik aloha: aloha?
19:27 aloha benabik: aloha is simple Bot::BasicBot::Pluggable hosted on https://github.com/bacek/aloha
19:32 daniel-s left #parrot
19:33 dalek rakudo/nom: 11ec306 | pmichaud++ | / (4 files):
19:33 dalek rakudo/nom: Refactor metaop handling, add R metaop.  More to come in a bit.
19:33 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/11ec306523
19:36 bluescreen left #parrot
19:41 cotto joined #parrot
19:44 dalek rakudo/nom: 8bd86ac | moritz++ | / (2 files):
19:44 dalek rakudo/nom: start with Rat literals
19:44 dalek rakudo/nom:
19:44 dalek rakudo/nom: Sadly it does not work yet, but maybe it inspires jnthn++ to tell me how to work around the serializer limitations
19:44 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8bd86ac97e
19:49 dodathome left #parrot
19:52 bluescreen joined #parrot
20:04 dalek parrot: f33bce0 | NotFound++ | runtime/parrot/library/distutils.pir:
20:04 dalek parrot: add rule 'pir_winxed' to build and clean steps of distutils
20:04 dalek parrot: review: https://github.com/parrot/parrot/commit/f33bce0ec1
20:04 hercynium_ joined #parrot
20:05 hercynium left #parrot
20:05 hercynium joined #parrot
20:08 whiteknight heh, that's one thing I can delete out of Rosella now
20:09 whiteknight NotFound: I had a bunch of winxed routines for distutils in Rosella. winxed_winxed, pasm_winxed (for includes), etc
20:09 whiteknight we can merge all that stuff into core distutils
20:10 hercynium_ left #parrot
20:10 NotFound whiteknight: Maybe will be better to first rewrite distutils in winxed or nqp.
20:11 darbelo_ left #parrot
20:11 NotFound It's boring and unuseful to write such stuff in pir
20:17 lucian_ joined #parrot
20:19 whiteknight I agree with that. I've considered doing such a rewrite before
20:19 whiteknight What I really really really want to do is rewrite the osutils module to use namespaces so we don't clobber common names
20:20 whiteknight rewriting that in winxed would be easy.
20:21 cotto +1, but who's going to do it?
20:21 whiteknight I might
20:21 NotFound Yes. distutils also use global names, but it's a one shoot program, no problem with it.
20:21 whiteknight Like I said, I've thought about doing it already
20:21 whiteknight There's no reason not to namespace it though. It's just good practice
20:22 lucian left #parrot
20:22 NotFound Yeah, but there is a problem with deprecation and transition period.
20:22 whiteknight actually, I take that back. Namespaces in Parrot are not "good" at all
20:24 darbelo joined #parrot
20:24 NotFound I'm wondering what will be the best place to put winxed includes. Current include search paths, or something different?
20:24 NotFound I mean the new winxed includes, not pasm generated from winxed.
20:24 whiteknight how are they going to be installed? As .winxed source files or as .pbc files?
20:24 whiteknight if .pbc, there's nothing different about them from normal includes
20:25 whiteknight oh, okay
20:25 whiteknight runtime/winxed
20:25 NotFound No, the includes are going to be winxed files, supposed to be used just for const, and class and extern function declarations.
20:27 NotFound Supposed for a now, maybe enforced in the future.
20:28 ambs left #parrot
20:28 ambs joined #parrot
20:29 bubaflub whiteknight: +1 to re-writing distutils in winxed and adding some nice winxed type functions
20:30 whiteknight left #parrot
20:36 darbelo left #parrot
20:39 theory_ joined #parrot
20:41 theory__ joined #parrot
20:43 theory left #parrot
20:43 theory__ is now known as theory
20:47 theory_ left #parrot
20:57 SHODAN left #parrot
21:02 kthakore joined #parrot
21:02 kthakore Hallo ...
21:02 kthakore it seems parrot broke perl6 NativeCall.pm
21:03 benabik Broke how?
21:04 cotto What works before that's broken now?
21:04 cotto *worked
21:07 mj41 left #parrot
21:09 kthakore http://paste.scsys.co.uk/117887
21:09 kthakore cotto: results
21:09 kthakore cotto: SDL.pl work with NativeCall.pm circa Dec 2010
21:09 kurahaupo joined #parrot
21:10 benabik LTA
21:11 colomon kthakore: try running from NativeCall.pm6 instead of NativeCall.pir?
21:11 kthakore colomon: I am ... not running anything
21:11 kthakore use NativeCall
21:11 kthakore in perl6
21:11 colomon errr... try deleting /home/kthakore/Documents/Development/p​erl/rakudo/parrot_install/lib/3.5.0-de​vel/languages/perl6/lib/NativeCall.pir
21:12 kthakore colomon: ok
21:13 kthakore colomon: no dice
21:13 colomon what does it say now?
21:13 kthakore same error
21:14 kthakore nope ... never mind
21:14 kthakore it only crash at exit
21:14 dalek parrot/soh-cah-toa/hbdb: dd98cd2 | soh_cah_toa++ | / (2 files):
21:14 dalek parrot/soh-cah-toa/hbdb: The "help" command works (sort of). Added "quit" command (sort of). Does not build
21:14 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/dd98cd247a
21:14 dalek parrot/soh-cah-toa/hbdb: e52afe3 | soh_cah_toa++ | src/hbdb.c:
21:14 dalek parrot/soh-cah-toa/hbdb: This is just for cotto to see
21:14 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/e52afe3e37
21:14 kthakore so I don 't need the NativeCall form that http://github.com/jnthn/zavolaj.git
21:17 bluescreen left #parrot
21:19 bluescreen joined #parrot
21:20 kthakore colomon++ thank you
21:23 lucian_ left #parrot
21:28 bluescreen_ joined #parrot
21:28 dukeleto cotto: where is your M0 gist?
21:29 kthakore colomon: can you exaplain ... why that worked tough?
21:30 colomon all I know is that something goes wrong when you compile NativeCall.pm6.  Long-standing issue.  I'm not sure if jnthn++ knows the reason or not.
21:32 cotto dukeleto, https://gist.github.com/1048210
21:36 ambs left #parrot
21:42 Psyche^ joined #parrot
21:46 fperrad left #parrot
21:47 Patterner left #parrot
21:47 Psyche^ is now known as Patterner
21:48 dalek nqp: e00bafe | pmichaud++ | src/HLL/Grammar.pm:
21:48 dalek nqp: Avoid PMCNULL error when converting non-existent hash element to string.
21:48 dalek nqp: review: https://github.com/perl6/nqp/commit/e00bafee83
21:48 dalek nqp: 6d9b526 | pmichaud++ | src/ (2 files):
21:48 dalek nqp: Add --rxtrace option, temporarily disable .lineof calculation for now.
21:48 dalek nqp: review: https://github.com/perl6/nqp/commit/6d9b526e45
22:02 benabik aloha: botsnack
22:02 benabik karma imcc
22:02 aloha imcc has karma of -25.
22:17 whiteknight joined #parrot
22:19 zby_home joined #parrot
22:20 whiteknight good afternoon, #parrot
22:21 estrabd left #parrot
22:21 hercynium left #parrot
22:23 benabik o/ whiteknight
22:23 whiteknight hello benabik
22:23 cotto hi whiteknight
22:25 dalek parrot/whiteknight/packfilewrapper: f62841b | Whiteknight++ | src/packfile/ (2 files):
22:25 dalek parrot/whiteknight/packfilewrapper: Move function prototypes from src/packfile/output.c to src/packfile/pf_private.h
22:25 dalek parrot/whiteknight/packfilewrapper: review: https://github.com/parrot/parrot/commit/f62841b56d
22:25 dalek parrot/whiteknight/packfilewrapper: 6a025d9 | Whiteknight++ | src/p (2 files):
22:25 dalek parrot/whiteknight/packfilewrapper: Make sure to get the PackfileView referenced from Sub constants, so when we mark a Sub we also mark the PackfileView that owns it
22:25 dalek parrot/whiteknight/packfilewrapper: review: https://github.com/parrot/parrot/commit/6a025d9252
22:25 whiteknight hello cotto
22:26 whiteknight kid51: ping
22:28 dmalcolm left #parrot
22:29 bubaflub left #parrot
22:31 dalek parrot: d1cbeb0 | Whiteknight++ | / (11 files):
22:31 dalek parrot: Merge branch 'vtable_substr'
22:31 dalek parrot: review: https://github.com/parrot/parrot/commit/d1cbeb0652
22:33 dalek parrot: c2ea702 | Whiteknight++ | NEWS:
22:33 dalek parrot: +NEWS entry about VTABLE_substr
22:33 dalek parrot: review: https://github.com/parrot/parrot/commit/c2ea702dbd
22:39 whiteknight some of the functions marked with PARROT_EXPORT in src/packfile/*.c absolutely should NOT be exported
22:40 whiteknight it's a little distressing, actually
22:45 dalek TT #2140 created by whiteknight++: Deprecate all PackFile_* functions
22:45 dalek TT #2140: http://trac.parrot.org/parrot/ticket/2140
22:50 zby_home left #parrot
22:53 bubaflub joined #parrot
22:53 mtk left #parrot
23:05 janus left #parrot
23:10 alester left #parrot
23:31 benabik Are Parrot strings (not String PMCs) immutable?
23:31 NotFound benabik: mostly.
23:31 benabik Mostly?
23:32 NotFound benabik: you can only mutate them locally, with string just created and before passing to any external part.
23:32 NotFound And deep inside parrot internals, of course.
23:33 whiteknight inside the string subsystem they can be played with. Outside that they are typically considered immutable
23:33 preflex left #parrot
23:35 benabik If I use Parrot_str_new(INTERP, "text"), will the result _ever_ change out from under me?
23:35 NotFound benabik: no
23:35 benabik Awesome.
23:35 whiteknight quite
23:35 benabik String PMCs are mutable though.  Alright.
23:36 NotFound If it does, is a bug, report it ASAP
23:36 benabik Someone pointed me at ropes and I thought a Rope PMC would be awesome.
23:36 whiteknight I looked at one of those a while back
23:36 whiteknight I played with a prototype but never got too far
23:37 benabik The tricky part will probably be having continuations to generate more data.
23:37 NotFound benabik: you also have StringBuilder and ByteBuffer for mutable things that can play with strings.
23:38 benabik NotFound: The point is to have highly efficient non-mutable strings.
23:41 NotFound benabik: note that we are probably going to change object models soon, so spending time in highly optimized PMCs may a lose of time.
23:41 dalek rakudo/nom: a413eb1 | pmichaud++ | src/core/List.pm:
23:41 dalek rakudo/nom: Fix List.reverse.
23:41 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a413eb145b
23:41 dalek rakudo/nom: df9d49b | pmichaud++ | / (3 files):
23:41 dalek rakudo/nom: Add reduction metaops for triangle and right associative operators.
23:41 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/df9d49b952
23:41 dalek rakudo/nom: a854856 | pmichaud++ | / (2 files):
23:41 dalek rakudo/nom: Merge branch 'nom' of github.com:rakudo/rakudo into nom
23:41 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a854856e40
23:41 preflex joined #parrot
23:41 benabik NotFound: Soon?  I like soon.
23:41 NotFound FSV of 'soon' ;)
23:42 benabik Bah.  Getting my hopes up.
23:46 colomon left #parrot
23:47 benabik left #parrot
23:47 soh_cah_toa left #parrot
23:50 dukeleto the hackathon is migrating to the bar...
23:51 whiteknight I sense the end of hacking is nigh
23:52 whiteknight what have y'all been hacking on anyway?
23:52 cotto left #parrot
23:55 dalek rakudo/nom: 0a8368a | pmichaud++ | src/core/metaops.pm:
23:55 dalek rakudo/nom: Let the triangle reduction op handle infinite lists, somewhat.
23:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0a8368a8b9
23:55 FalseNames left #parrot

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

Parrot | source cross referenced