Camelia, the Perl 6 bug

IRC log for #parrot, 2012-01-09

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:02 dalek parrot: 651dd87 | Whiteknight++ | / (2 files):
00:02 dalek parrot: re-headerize and fix a typo
00:02 dalek parrot: review: https://github.com/parrot/parrot/commit/651dd87d89
00:04 Coke joined #parrot
00:42 aloha (parrot/parrot) Issues closed : 235 (Cleanup compiler function GCI ) by Alekssasho : https://github.com/parrot/parrot/issues/235
00:56 bacek_at_work ~~
01:19 cotto dukeleto, ping
01:45 cotto http://wiki.call-cc.org/ch​icken-compilation-process
03:18 jsut joined #parrot
03:38 Psyche^ joined #parrot
05:11 cky joined #parrot
05:50 benabik joined #parrot
05:59 alvis joined #parrot
06:18 dukeleto cotto: pong
06:21 dalek parrot: c6a80b8 | dukeleto++ | t/pmc/timer.t:
06:21 dalek parrot: Remove an svn remnant in the Timer tests
06:21 dalek parrot: review: https://github.com/parrot/parrot/commit/c6a80b89f9
06:23 cotto dukeleto, how's your new gig going to mesh with #ps?
06:23 cotto are you still working remotely?
06:23 dukeleto why, oh why are our Timer PMC tests in pasm? And why do they hang on OS X?
06:23 dukeleto cotto: still working remotely, but the last few weeks have just been crazy with holidays and starting the new gig
06:24 dukeleto cotto: are we thinking of changing the time of #ps?
06:24 cotto dukeleto, I figured it was something like that
06:24 cotto nobody's brought it up
06:24 dukeleto cotto: this CHICKEN stuff is interesting
06:25 dukeleto cotto: we kind of fell off the M0 horse. How do we get back on?
06:26 cotto dukeleto, that's what I've been thinking about.
06:26 cotto From where I can see, the next step is getting 6model in core.
06:26 bacek ~~
06:26 zby_home joined #parrot
06:26 cotto That comes with its own set of questions.
06:26 cotto hio bacek
06:26 bacek cotto, aloha
06:26 cotto among which are:
06:27 cotto how do we do the transition so that the nqp folks stay happy and still retain the necessary degree of control over what they need
06:27 cotto that's the big one
06:28 dukeleto bacek: fancy seeing you here
06:29 dukeleto cotto: i am confused by the current status of 6model/nqpq or whatever it is called
06:29 bacek dukeleto, yeah... It's been a while...
06:29 dukeleto bacek: i feel ya. I haven't been in here in quite a while myself
06:29 cotto dukeleto, welcome to the club
06:29 bacek current 6model is nice and shiny.
06:29 cotto I have no idea what qregex or nqpq are
06:29 cotto bacek, and still changing a bit, last I looked (early december-ish)
06:29 benabik NFA based regexen.
06:30 dukeleto and what is the status of Rakudo on the nom branch? Are they still fighting regressions and such?
06:30 bacek http://irclog.perlgeek.de/​perl6/2012-01-08#i_4949636
06:30 bacek cotto, 6model is stable (fsvo)
06:30 bacek in order to properly incorporate it into parrot we have to:
06:30 cotto <3
06:30 bacek 1. Throw away all PMC crap.
06:31 cotto <3<3
06:31 bacek 2. Throw away whole lot of VTABLE crap
06:31 bacek 3. Throw away current PCC.
06:31 benabik More specifically, QRegex is NFA based regex and nqpq is nqp + qregex.
06:31 bacek after first 2 steps number of ops will be reduced to about 100
06:32 dukeleto bacek: QAST?
06:32 bacek ah, yes.
06:32 bacek 4. Throw away IMCC just because you can't teach old dog new tricks.
06:32 dukeleto sounds like 6model in parrot core means rewriting parrot core
06:32 bacek dukeleto, QAST is primitive Captures with bit of accessors
06:32 dukeleto bacek: what is the Q for? Queasy?
06:33 benabik nQp
06:33 bacek dukeleto, yes. And it's only one way to keep parrot alive.
06:33 benabik QAST is NQP's PAST replacement/extension.
06:33 dukeleto bacek: how does that mesh with m0? M0 is, supposedly, the rewrite of parrot core that the least amount of people disagree about
06:34 cotto bacek, do you see a way to make that transition reasonably smooth?
06:34 dukeleto bacek: so it seems we need to know how to bolt 6model to m0
06:35 bacek cotto, nope. And I don't see any value in about 90% of current parrot codebase.
06:35 bacek dukeleto, it's reasonably easy. Just implement "m0 vm". 6model is VM agnostic.
06:35 cotto the part of me that reads Joel is fighting with the part of me that's looked at the code you're talkiing about
06:36 cotto 6model is just code
06:36 cotto m0 is a thingy that runs code
06:36 cotto together, they fight crime
06:36 benabik cotto: I think we can make PMCs a kind of REPR and have the new and old stuff co-exist.
06:36 bacek cotto, Spolsky? "2dn version problem"? Or it was Brooks?
06:36 dukeleto bacek: i have a hard time swallowing the "vm agnostic" koolaid. Not that I don't believe it will be, but I think that will come with huge performance problems
06:37 cotto bacek, yeah.  I remember it from Spolsky's blog, but it could have come from Brooks.
06:37 dukeleto cotto: last i remember we went down a research hole with m0
06:37 cotto benabik, good point
06:37 cotto yeah
06:37 dukeleto cotto: i was trying to research ARM opcodes and got lost in the madness
06:38 dukeleto cotto: suffice to say, ARM is complicated. There is no "ARM". There are a collection of related architectures which are ARMish
06:38 cotto dukeleto, I have a better idea of how it actually needs to be implemented, but it's not the priority until 6model in core is looking more likely.
06:38 cotto yeah.  It explains why I see LWN articles on how the arm kernel code is a mess.
06:39 dukeleto how does 6model relate to a new JIT in the future?
06:39 dukeleto orthogonal? perpendicular?
06:39 cotto I see it as orthogonal.
06:39 benabik dukeleto: Right angles or right angles?
06:40 dukeleto M0 motivates me, because we can't get a JIT without M0. But 6model seems like just another (albeit very nice and shiny) refactor.
06:40 dukeleto No one needs to sell me on the features of 6model. I get it, I buy it. But still: Just another refactor.
06:41 * dukeleto needs to sleep, since he refactored timezones today
06:41 cotto dukeleto, some kind of object system refactor is necessary if we want to support other hlls.
06:41 benabik M0 is movement towards JIT.  6model is movement towards making more languages easier.  (I think.)
06:41 cotto benabik, yes
06:41 dukeleto cotto: yes, totally. But 6model requires an entire rewrite. Not quite an "object system upgrade"
06:41 benabik I _think_ that the problems with Python's object system during GSoC would have been far easier with 6model.
06:42 dukeleto benabik: yes, i agree
06:42 cotto benabik, also my understanding
06:42 cotto dukeleto, different people get excited about different things
06:42 cotto I see 6model as being more immediately necessary, so that's where I plan on focusing my energy.
06:43 cotto dukeleto, are you interested in taking the lead on M0?
06:43 dukeleto cotto: throwing me under the bus, eh?
06:43 cotto (I'm not going anywhere, of course.)
06:43 cotto dukeleto, I'm wary of buses.
06:43 dukeleto cotto: i am not sure what the title of "lead" on M0 would change
06:44 cotto dukeleto, "title"? ;)
06:44 benabik dukeleto: different timezone?  Where'd you wonder off to/from?
06:44 * benabik is a little slow, should probably already be asleep.
06:44 dukeleto I am now on EST visiting family and will be attending Perl Oasis later this week.
06:45 dukeleto cotto: i definitely haven't forgotten about m0, but I sure have too many started projects.
06:45 benabik "Perl, Go-Karts, and Beer"  Sounds good.
06:45 dukeleto benabik: indeed! It is my first time. Should be fun.
06:45 dukeleto I will be talking about Jitterbug
06:46 cotto dukeleto, you do what you have the tuits for.  Since I plan on working on 6model, you're welcome to do with M0 as you see fit and see how far you can take it.
06:46 dukeleto cotto: i will think about how to proceed with m0. If you can do some kind of email to parrot-dev with an m0 brain dump, that would be useful
06:46 dukeleto cotto: sounds good
06:47 * dukeleto .sleep()
06:47 cotto dukeleto, alright.  It might not be until tomorrow though.  I'm up early (and ideally down soon)
06:47 cotto 'night, dukeleto
07:48 jsut_ joined #parrot
07:56 bacek aloha, clock?
07:56 aloha bacek: LAX: Sun, 23:56 PST / CHI: Mon, 01:56 CST / NYC: Mon, 02:56 EST / UTC: Mon, 07:56 UTC / LON: Mon, 07:56 GMT / BER: Mon, 08:56 CET / TOK: Mon, 16:56 JST / SYD: Mon, 18:56 EST
07:59 sorear Why does aloha list both UTC and LON?  I thought everyone knew those were the same.
08:00 sorear ...wait, British Summer Time, nevermind
08:26 mj41 joined #parrot
08:52 snearch joined #parrot
09:33 dalek nqp: bb332d9 | bacek++ | src/pmc/ (3 files):
09:33 dalek nqp: Don't use pmc-specific accessors when not necessary
09:33 dalek nqp: review: https://github.com/perl6/nqp/commit/bb332d93b6
09:37 lucian joined #parrot
09:57 dip joined #parrot
10:16 kshannon Is there anyone here familiar with how big a job it'd be to get the parrot debugger working again?
10:21 alvis kshannon: You might ask/msg Whiteknight. He said, at one point, he wanted to take a look at it.
10:24 moritz wasn't there a gsoc project for reviving the debugger?
10:24 tadzik there was
10:25 tadzik it didn't go exactly according to the plan, ISTR
12:31 mtk joined #parrot
13:23 dukeleto ~~
13:23 dalek parrot.github.com: 0a08012 | alvis++ | / (2 files):
13:23 dalek parrot.github.com: Updated index.html and releases.html to reflect the change over to parrot-docsx repos.
13:23 dalek parrot.github.com: review: https://github.com/parrot/parro​t.github.com/commit/0a08012858
13:24 dukeleto kshannon: soh_cah_toa++ is a good person to ask debugerry stuff
13:24 dukeleto kshannon: he worked on it most recently.
13:25 dukeleto alvis++ # documentation whisperer
13:28 dalek parrot.github.com: 815f0f6 | alvis++ | / (14795 files):
13:28 dalek parrot.github.com: Removed 3.10.0 through 1.0.0 in favor of the change over to parrot-docsx repos.
13:28 dalek parrot.github.com: review: https://github.com/parrot/parro​t.github.com/commit/815f0f674a
13:38 simcop2387 joined #parrot
13:41 alin joined #parrot
13:42 dalek parrot.github.com: 946baa6 | alvis++ | index.html:
13:42 dalek parrot.github.com: Corrected missing '>' on html tag.
13:42 dalek parrot.github.com: review: https://github.com/parrot/parro​t.github.com/commit/946baa6603
13:42 dalek parrot.github.com: 6b9e5f0 | alvis++ | releases.html:
13:42 dalek parrot.github.com: Fixed an oops - to correct the text to the links to the 0.*.* docs
13:42 dalek parrot.github.com: review: https://github.com/parrot/parro​t.github.com/commit/6b9e5f0068
13:55 contingencyplan joined #parrot
14:11 dalek rakudo/nom: 8b620e7 | moritz++ | src/core/Enumeration.pm:
14:11 dalek rakudo/nom: Enumeration.roll
14:11 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/8b620e7896
14:34 PacoAir joined #parrot
14:36 simcop2387 joined #parrot
14:55 dukeleto the timer pmc tests going into an infinite loop on OS X is starting to really bug me
14:55 dukeleto they will feel my wrath, soon
15:37 contingencyplan joined #parrot
16:26 Myhrlin joined #parrot
16:27 dmalcolm joined #parrot
16:38 alester joined #parrot
16:44 alester Did people get spammed from GitHub?  Rick is importing tickets.
16:47 cotto ~~
16:48 cotto alester: !
16:48 alester ?
16:48 cotto (in a good way)
16:48 cotto I'm excited that the import is happening.
16:48 cotto that's all
16:48 alester It's the 2nd attempt
16:48 alester we'll check it again
16:48 alester and then he'll do the final one.
16:48 AzureStone joined #parrot
17:13 alin joined #parrot
17:17 mj41 joined #parrot
17:27 mj41 joined #parrot
17:27 * moritz hasn't got github ticket spam
17:45 zby_home joined #parrot
18:03 alin joined #parrot
18:03 fperrad joined #parrot
18:28 Myhrlin joined #parrot
18:34 he joined #parrot
19:13 mj41 joined #parrot
19:24 alin joined #parrot
19:39 alester OK, get out there and check for GitHub ticket problems!
19:50 alvis joined #parrot
19:51 cotto alester++
20:02 bluescreen joined #parrot
20:18 dalek rakudo/nom: 0838f41 | jnthn++ | src/ (2 files):
20:18 dalek rakudo/nom: Implement .DEFINITE.
20:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0838f41b8d
20:18 dalek rakudo/nom: b1d3c7c | jnthn++ | src/Perl6/Metamodel/ (4 files):
20:18 dalek rakudo/nom: Refactor to use method_table in place of methods inside of MOP.
20:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b1d3c7cd50
20:18 dalek rakudo/nom: 1f302f8 | jnthn++ | src/ (6 files):
20:18 dalek rakudo/nom: Implement :excl and :all for methods, attributes and parents. Make :excl the default.
20:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1f302f8c5e
20:18 dalek rakudo/nom: b21d28b | jnthn++ | src/ops/perl6.ops:
20:18 dalek rakudo/nom: Fix .DEFINITE for container case.
20:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b21d28b540
20:18 dalek rakudo/nom: e65b6b1 | jnthn++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
20:18 dalek rakudo/nom: Fixes to .is_dispatcher and .is_generic.
20:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e65b6b1825
20:18 dalek rakudo/nom: 783b468 | jnthn++ | t/spectest.data:
20:18 dalek rakudo/nom: Run S12-introspection/definite.t.
20:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/783b4680d3
20:32 dalek rakudo/nom: de59b4b | jnthn++ | docs/ChangeLog:
20:32 dalek rakudo/nom: Start a ChangeLog for the month, so we don't have it all to do right before the next release.
20:32 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/de59b4b0c6
20:37 dalek nqp/qbootstrap: f500829 | kshannon++ | .gitignore:
20:37 dalek nqp/qbootstrap: Make git ignore all *.pbc
20:37 dalek nqp/qbootstrap: review: https://github.com/perl6/nqp/commit/f500829b49
20:37 dalek nqp/qbootstrap: e4fe3ad | kshannon++ | src/QRegex/NFA.nqp:
20:37 dalek nqp/qbootstrap: The enumcharlist can be negated.
20:37 dalek nqp/qbootstrap:
20:37 dalek nqp/qbootstrap: Fix the NFA building to pay attention to that.
20:37 dalek nqp/qbootstrap: review: https://github.com/perl6/nqp/commit/e4fe3add50
20:39 dalek rakudo/nom: 6bf68d0 | jnthn++ | t/spectest.data:
20:39 dalek rakudo/nom: Run S14-roles/anonymous.t.
20:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6bf68d0e47
20:50 Kovensky joined #parrot
21:03 dalek rakudo/nom: b281b21 | jnthn++ | src/core/Hash.pm:
21:03 dalek rakudo/nom: Handle various other cases of Hash.delete.
21:03 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b281b21425
21:03 dalek rakudo/nom: f4b327b | jnthn++ | t/spectest.data:
21:03 dalek rakudo/nom: Run S32-hash/delete.t.
21:03 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f4b327bf97
21:03 dalek rakudo/nom: c827edc | jnthn++ | src/core/Array.pm:
21:03 dalek rakudo/nom: Implement Array.delete (based on ng code, but with various tweaks and updatings).
21:03 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/c827edcc3f
21:03 dalek rakudo/nom: 38165a5 | jnthn++ | t/spectest.data:
21:03 dalek rakudo/nom: Run S32-array/delete.t.
21:03 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/38165a575e
21:40 zjmarlow joined #parrot
21:43 PacoAir joined #parrot
21:53 bluescreen_ joined #parrot
22:04 dmalcolm_ joined #parrot
22:30 dalek rakudo/staged-setting-fail: 664f46b | jnthn++ | src/core/control.pm:
22:30 dalek rakudo/staged-setting-fail: Work around pir::... limitation.
22:30 dalek rakudo/staged-setting-fail: review: https://github.com/rakudo/rakudo/commit/664f46b9b2
22:56 dalek rakudo/staged-setting-fail: ab92aff | jnthn++ | src/Perl6/ (2 files):
22:56 dalek rakudo/staged-setting-fail: Fix for the 'Sub doesn't have an SC' issue - CORE and CORE2 both compile now.
22:56 dalek rakudo/staged-setting-fail: review: https://github.com/rakudo/rakudo/commit/ab92aff520
23:01 dmalcolm__ joined #parrot
23:06 snearch joined #parrot
23:23 dngor joined #parrot
23:44 zjmarlow joined #parrot
23:55 bacek_at_work ~~

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

Parrot | source cross referenced