parrot.org/ - clean up those smolders for the release!
Set by moderator on 20 October 2008.
00:09 AndyA joined
dalek r32290 | kjs++ | trunk: 00:25
: [pirc/new] add missing tokens for PASM
: + support .line and .file directives
: + add nice error messsages for non-supported tokens.
: + heredocs and .include work for pasm mode as well, which is nice.
diff: www.parrotvm.org/svn/parrot/revision?rev=32290
00:29 kj joined 00:39 GeJ joined 00:43 Whiteknight joined
dalek r32291 | jkeenan++ | trunk: 00:50
: Correct spelling errors in 'skip' messages in 6 files: failling => failing.
diff: www.parrotvm.org/svn/parrot/revision?rev=32291
r32292 | allison++ | pdd22io:
: [pdd22io] Moving the common utility functions up a level.
diff: www.parrotvm.org/svn/parrot/revision?rev=32292
r32293 | allison++ | pdd22io: 00:54
: [pdd22io] Update paths and documentation for new location and filename. Fix
: compilation warning about freeing constant C strings.
diff: www.parrotvm.org/svn/parrot/revision?rev=32293
r32294 | allison++ | pdd22io: 00:55
: [pdd22io] Move I/O utility functions to main src/ directory.
diff: www.parrotvm.org/svn/parrot/revision?rev=32294
r32295 | kjs++ | trunk: 01:00
: [pirc/new] refactor PASM grammar bits to re-use PIR grammar rules.
: + this is going to be easy :-)
diff: www.parrotvm.org/svn/parrot/revision?rev=32295
r32296 | allison++ | pdd22io: 01:03
: [pdd22io] Rename or remove duplicate functions from old I/O API so the two can
: peacefully co-exist.
diff: www.parrotvm.org/svn/parrot/revision?rev=32296
r32297 | allison++ | pdd22io: 01:05
: [pdd22io] Change path of src/io/common.c to src/io.c in headerizer tags.
diff: www.parrotvm.org/svn/parrot/revision?rev=32297
r32298 | allison++ | pdd22io: 01:07
: [pdd22io] Change ATTR parsing in pmc2c to handle the attributes in the
: FileHandle data struct. Clean up the formatting to make it more maintainable.
diff: www.parrotvm.org/svn/parrot/revision?rev=32298
dmknopp Hello all, I would like to loop thru all the HLL objects alive in Parrot. It seems to me that the AddrRegistry PMC would be a good place to ask for information on all those objects. 01:11
Is this recommended to do in a HLL using PIR?
chromatic They're not stored in AddrRegistry. 01:12
dmknopp the mem addresses are stored right? 01:13
chromatic No.
dmknopp key : mem addr value: ref count
hmm... ok
chromatic AddrRegistry is only a member of the root set used to anchor otherwise-unanchored PMCs for GC.
PMCs returned to embedding programs, for example.
dmknopp drat. back to square 1 01:14
ok, thanks.
chromatic If an HLL such as Cardinal wanted to track all objects created in that language, it would need something like Ruby's ObjectSpace.
dmknopp :-) thats what I was hoping to implement 01:15
I was hoping AddrReg was a hash of all live objects which I could then tarverse and inspect
chromatic We don't have anything like that now, unless you write C code to scan all PMC pools. 01:16
dmknopp Do you have an suggestions on adding an ObjectSpace like functionality?
ok
chromatic For Parrot in general or a specific HLL?
dmknopp scan all PMC pools, I will have to look into that
im thinking for HLL and Cardinal specificly 01:17
but i guess other languages might need this type of thing as well
chromatic Then you add something to the MOP for that language to cache all created objects with weak references.
dmknopp Are we talking the P6 protoobject 01:18
chromatic Or the Cardinal equivalent.
dmknopp or maybe the find_method :vtable
ok, yeah right now Cardinal uses the P6 protoobject 01:19
chromatic Add an initializer to Cardinal's Object which registers the object in ObjectSpace, and add a finalizer which unregisters it. 01:20
dmknopp Right now, I believe in Cardinal, that all objects that are created go thru the CardinalObject.new. This would be a good place maybe to add the initializing code. 01:21
chromatic That's what I'd do.
dmknopp When you say register with the object in ObjectSpace, Do you mean create a hash of WeakRef PMCs , can I do that? Or does ObjectSpace mean some thing more specific I havent learned about? 01:22
err "register the object in ObjectSpacee"
chromatic It's just a hash with a mark vtable entry which doesn't do anything.
You'd have to write it, but it's easy to write. 01:23
dmknopp ok thanks 01:32
dalek r32299 | allison++ | pdd22io: 01:40
: [pdd22] Keeping the revision history of the ParrotIO PMC tests under the new
: name.
diff: www.parrotvm.org/svn/parrot/revision?rev=32299
r32300 | allison++ | pdd22io: 02:01
: [pdd22io] Updating old ParrotIO tests for FileHandle PMC.
diff: www.parrotvm.org/svn/parrot/revision?rev=32300
r32301 | chromatic++ | trunk:
: [OO] Fixed a potential bug lurking in Class PMC's isa_pmc() vtable entry, where
: the (cached) short name of a class may be an empty string, while the class may
: have a real name. Added tests to t/oo/isa.t which attempt to catch this in the
: scenario where I saw it.
diff: www.parrotvm.org/svn/parrot/revision?rev=32301
r32302 | chromatic++ | trunk: 02:07
: [OO] Revised Class PMC's isa() vtable entry to delegate to isa_pmc() as soon as
: possible, given the idea that it's much faster to compare two PMCs against each
: other than it is to perform repeated STRING allocations and comparisions. If
: that's not always true, it soon will be.
diff: www.parrotvm.org/svn/parrot/revision?rev=32302
r32303 | chromatic++ | trunk: 02:09
: [t] Added some MMD tests I had lying around for debugging.
diff: www.parrotvm.org/svn/parrot/revision?rev=32303
r32304 | chromatic++ | trunk:
: [lib] Tided some Perl 5 code; no functional changes.
diff: www.parrotvm.org/svn/parrot/revision?rev=32304
r32305 | allison++ | pdd22io: 02:13
: [pdd22io] Adding the basic FileHandle PMC.
diff: www.parrotvm.org/svn/parrot/revision?rev=32305
02:55 MariachiElf joined
dalek r32306 | allison++ | pdd22io: 03:17
: [pdd22io] From PDD 22, correct default when no mode string passed is "read".
diff: www.parrotvm.org/svn/parrot/revision?rev=32306
r32307 | allison++ | pdd22io: 03:25
: [pdd22io] FileHandle PMC 'open' and 'close' methods, plus initialize
: platform-specific closed value for 'os_handle' attribute.
diff: www.parrotvm.org/svn/parrot/revision?rev=32307
03:36 Andy joined
dalek r32308 | allison++ | pdd22io: 03:44
: [pdd22io] Fix malloc error on freed C string.
diff: www.parrotvm.org/svn/parrot/revision?rev=32308
r32309 | allison++ | pdd22io: 03:46
: [pdd22io] Explicitly mention the default behavior when no mode string is passed.
diff: www.parrotvm.org/svn/parrot/revision?rev=32309
r32310 | allison++ | pdd22io:
: [pdd22io] Mode string for 'open' can be null.
diff: www.parrotvm.org/svn/parrot/revision?rev=32310
03:49 bacek joined
dalek r32311 | allison++ | pdd22io: 03:50
: [pdd22io] FileHandle PMC: more complete 'destroy' vtable function and fix a bug in 'clone'.
diff: www.parrotvm.org/svn/parrot/revision?rev=32311
03:50 Psyche^ joined
dalek r32312 | allison++ | pdd22io: 03:52
: [pdd22io] Updated main I/O header file with the headerizer.
diff: www.parrotvm.org/svn/parrot/revision?rev=32312
03:56 stockwellb joined
stockwellb I'm trying to get the bigger picture of the 'Role' where would be a good place to look? 03:59
chromatic stockwellb, www.oreillynet.com/onlamp/blog/2006..._obje.html 04:17
stockwellb Thank you chromatic, Role and PIR didn't google well so I thought I'd come to the source. 04:19
cotto Roles are more of a Perl6 thing. Searching for those terms looks like it yields some useful results. 04:21
stockwellb I ran across Roles in composition_1.pir under t/oo so I saw it first as a PIR thing, not something specific to Perl6. 04:22
05:27 tewk_ joined 05:45 MariachiElf joined 05:48 diakopter joined 05:49 Bzek joined 07:07 uniejo joined 07:39 allison joined 07:47 bacek joined 08:27 barney joined
dalek r32313 | chromatic++ | trunk: 08:40
: [src] Added temporary_pmc_new() and temporary_pmc_free(), which
: allocate/initialize and free a PMC with a lifetime within a single C function.
: See the disclaimer in the documentation of temporary_pmc_new(). These are
: highly useful in certain rare circumstances and dangerous outside of them.
diff: www.parrotvm.org/svn/parrot/revision?rev=32313
08:51 bacek joined
bacek hi there 09:15
purl hi, bacek.
bacek purl: glad to see you alive ;) 09:17
purl bacek: huh?
09:28 tomyan joined 09:33 iblechbot joined 10:09 kj joined
dalek r32314 | masak++ | trunk: 10:12
: [rakudo] unmade commit r32205, which broke boolean junction flattening
diff: www.parrotvm.org/svn/parrot/revision?rev=32314
r32315 | kjs++ | trunk: 11:48
: [pirc/new] refactor PASM grammar bits.
diff: www.parrotvm.org/svn/parrot/revision?rev=32315
11:57 Zaba joined 12:22 Lorn joined 12:30 gaz joined 13:41 DietCoke joined
Coke pmichaud: does -t4 work for you now? 13:42
14:02 grim_fandango joined 14:05 gryphon joined 14:12 cognominal joined
dalek r32316 | kjs++ | trunk: 14:14
: [pirc/new] more PASM grammar refactoring, and adding a helper function to match the parse/data flow.
diff: www.parrotvm.org/svn/parrot/revision?rev=32316
Coke I want NYTParrotProf 14:17
PerlJam Coke++ me too! 14:18
Coke ok. I fixed step 1. =-) 14:25
14:36 PacoLinux joined 14:43 gaz joined 15:01 Andy joined
cognominal je suis chez mon pĆØre et j'ai eu le malheur d'essayer de connecter sa machine vista directement Ć  la mienne via un cable Ethernet. 15:21
Maintenant, sa machine est connectƩ Ơ la freebox qui lui donne un IP mais la freebox ne rƩpond pas au paquets envoyƩs par sa machine. 15:22
jonathan Je ne sais pas...
;-)
kj is learning french
cognominal oops
I am not in #perlf
kj mais non!
cognominal everyone is welcome to come in #perlfr to learn french
kj I'm sticking to the Je m'appelle level... 15:23
particle i don't even know how to type 'Ć '
jonathan I'm at the Je suis un merdique on francais level. 15:24
s/on/en/ # see!
kj hon hon hon!
15:27 cognominal_ joined
TimToady ćƒ•ćƒ©ćƒ³ć‚»ćÆå¦»ćŒč©±ć›ć‚‹ć‚ˆļ¼ 15:32
Coke sadly, that doesn't render here. 15:33
cognominal_ reaches for the ikana app on his iphone
I also have an app for kanjis but did not tried it yet. 15:34
pmichaud Coke: (-t4) will test in a bit. 15:40
I probably need to find a crashing program again to test -- I only noticed it when trying to debug other issues
tewk_ NotFound: japhb: did you guys decide which opengl func is broken in NCI jit 15:41
Coke pmichaud: I was able to trigger the bug even with ./tclsh -e 'puts hi'; anything PGE intensive triggered it for me. 15:42
if you find it broken again, open a ticket, I'll poke at it.
pmichaud okay, great.
Coke ... and hopefully jonathan can fix it again. ^_^
jonathan pmichaud: I'm approved to come to the Parrot developer summit. :-) 15:46
So, double dose of jet lag coming up... 15:47
pmichaud indeed
I look forward to seeing you there :-)
jonathan Happily, flying west to the US, I probably will get it less than when flying east back home. 15:48
So should be pretty awake at the summit.
pmichaud probably more awake than me, even :-)
15:49 particle1 joined
jonathan pmichaud: It may prove a good time for us to hash out all of the container stuff. 15:51
kj jonathan: when are you flying? 15:53
jonathan Will fly on Thursday
kj ok, me on Monday 15:54
jonathan oh, you're going to have some time out there first?
kj yep
jonathan nice
jonathan had big vacation recently so doesn't have time/money for that on this trip, sadly 15:55
Just taking the Monday afterwards, then fly home on the Tuesday
kj I'm flying back 19th, I think that's Tues as well
Oh it's Wednes 15:56
jonathan Where you flying from?
kj dublin
i live in limerick, ireland 15:57
jonathan Ah, OK. :-)
jonathan didn't make it to Ireland yet
16:04 jhorwitz joined
Coke there once was a man from limerick. 16:14
moritz whose limericks where just two lines long, and didn't rhyme at all. 16:15
Coke man from limerick / did not stand on tradition / preferred his haiku. 16:18
kj he made up his mind
and added a fourth line
o/~ laa laa la la laaa la o/~ 16:19
jonathan thinks he will hack on Rakudo tomorrow 16:21
Coke what, no tcl!? 16:24
jonathan I haven't thought of any patches to write yet for it, beyond renaming the test target. ;-)
dalek r32317 | kjs++ | trunk: 16:25
: [NEWS] update more PIRC news.
diff: www.parrotvm.org/svn/parrot/revision?rev=32317
Coke I eagerly await the PGE stuff from pmichaud, btw, since most of my sub executions are in PGE (no clue how long they take, but still)
jonathan Yes, I'm looking forward to that too.
particle so is pmichaud, i imagine. 16:26
Coke pmichaud: are the PGE speedups something you can break into chunks and have multiple people hacking on? Or in some way ||ize it so it's not just you?
16:31 hercynium joined
particle coke: not likely. the protoregex and ltm work involve huge refactors 16:34
however, i'm sure they'll be branch-based
jonathan rakudo.org hasn't had any posts other than perl 6 design minutes for ages... :-( 16:35
particle ack, i never wrote an 'is export' post
i suppose mostly because i never got import working, so is export isn't useful
jonathan particle: It'd be really great if you could blog there as you work on the command line implementation stuff too. 16:36
particle yes, i will indeed.
i've promised weekly reports
Coke Hey, uh, Undef is a singleton, right?
jonathan It's the URL I keep giving out to people at conferences to follow Rakudo progress.
particle Method 'panic' not found for invocant of class 'PGE;Match' 16:37
ick.
japhb tewk_: nope, life interfered. Just managed to shrink test case somewhat (in new static-triangle.pir). Still not trivial yet, but at least less cruft.
particle i wish src/gen_actions.pir had some way to link back to the source 16:39
.sub "_block4636" :anon :lexid("391") :outer("382") 16:40
jonathan "Just" needs implementing the HLL debug segment stuff...
particle that isn't helpfully descriptive
even a comment like # src/parser/actions.pm 347 16:41
above the generated code 16:42
16:48 jsut|work joined 16:57 Theory joined
nopaste "Coke" at 72.228.52.192 pasted "Remove some unneeded calls to <new>" (26 lines) at nopaste.snit.ch/14449 16:58
Coke If someone wants to apply that, feel free; I notice no speedups in an small tcl test applying it, but it seems proper to apply. 16:59
17:04 Theory joined 17:42 ruoso joined
particle allison: ping 18:06
allison pong
particle do we have a general inquires email addr at parrot.org? 18:07
allison nope, we can add one
particle info@ would do 18:08
allison alias for one particular person, or multiple people?
nopaste "particle" at 98.232.28.49 pasted "allison: letter of invitation template, for review" (26 lines) at nopaste.snit.ch/14450
allison like, should it just forward to directors? 18:09
particle sure
allison I'll submit a request, but it may not be done in time for the letter. I'd just use your "official" parrot.org email address for this letter. 18:10
particle wow, that's an annoying lack of line wrapping
jonathan allison: Thanks for emails - will book flight today.
particle ah, didn't realize we needed to request the addr. 18:11
ok, yeah, i'll use my addr
allison that is, 'jerry.gay' rather than 'particle' (That was a good suggestion you made to have email address names as well as aliases) 18:12
jonathan: excellent!
purl EGG-see-lent!
particle yep, ok 18:13
allison jerry: I'd put quotes around "Parrot" in 'On behalf of the Parrot Foundation ("Parrot")' (standard legalese) 18:14
nopaste "particle" at 98.232.28.49 pasted "letter of invitation... much better formatting" (46 lines) at nopaste.snit.ch/14451
particle ah, thanks
allison particle: also, "Google's Mountain View Campus" rather than "Google Headquarters" 18:15
or change "Campus" to "campus" non-caps
purl allison: that doesn't look right
allison purl: oh, hush :)
purl allison: excuse me?
moritz purl, change 1 USD to EUR 18:16
purl moritz: 1 USD makes 0.7895 EUR
allison that can't be right
particle change 1 EUR to USD
purl particle: 1 EUR makes 1.2666 USD
particle yep, that's right.
dollar++
allison wow, what happened to the dollar? 18:17
particle the rest of the world is in a worse financial state than the usa, or so it's perceived among currency traders 18:18
shall i stick this template on the wiki, or as a directors-shared google doc? 18:19
allison I'd say google doc, not something we want random people using 18:20
particle that was my thought. too bad we don't have a private doc repo @parrot.org
allison particle: we can probably arrange that 18:21
particle do you have letterhead?
allison I use an Open Office template, see the advisory board invoices 18:22
particle: I can sent you that
or, if you're not using Open Office, I can send you the logo
particle i have ms office, open office, and acrobat pro 18:23
allison particle: excellent
particle: I find that carrying around pre-printed letterhead isn't practical, but I can find a color printer at any local copy shop 18:24
particle yep
i'd like to have software that gen's pdfs on our letterhead
allison particle: open office does great PDFs 18:26
particle: or, do you mean without actually including the logo template in the office doc? 18:27
particle the latter
purl the latter is better
18:32 chromatic joined
allison particle: you'd still need a template that precisely measures how much space to leave around the logo, at which point, might as well just have the logo in the doc 18:35
particle: letterhead sent
particle allison: received 18:45
18:51 rindolf joined
rindolf Hi all. 18:53
19:09 chromatic joined
Coke allison: are you on the board on TPF? or just TPaF? 19:27
particle www.perlfoundation.org/board_of_directors 19:29
davidfetter anybody going to OSDC? 19:30
kj OSDC? 19:33
purl OSDC is probably the Open Source Developpers Conference or at www.osdcon.org/ or at www.osdc.com.au/
19:34 cognominal joined
davidfetter anybody? 19:38
purl i heard anybody was there?
davidfetter larts purl
kj it's very far :-( 19:42
chromatic I don't have a criminal record.
19:43 peepsalot joined 19:45 sjansen joined
PacoLinux Real programmers don't need abstract concepts to get their jobs done: they are perfectly happy with a keypunch, a FORTRAN IV compiler, and a beer. :) : www.ee.ryerson.ca:8080/~elf/hack/realmen.html 19:49
dalek r32318 | kjs++ | trunk: 19:50
: [pirc/new] pretty much finish the PASM grammar stuff and actions and do memory cleanup in regalloc.c
diff: www.parrotvm.org/svn/parrot/revision?rev=32318
davidfetter kj, it's pretty far for me, too 19:51
kj if I had some big company to back me up... :-) But trips that far are costly for a student. 19:52
Coke chromatic: that just means you were never caught.
or had it expunged! 19:53
chromatic or don't have a turntable.
Infinoid guess they're a package deal nowadays 20:00
chromatic Class <-> Role :: Module :: Package <-> Turntable ? 20:01
Coke 2 * $turntable ...
chromatic Throw in some \\, [], T, and a, and you have valid Haskell code there. Hm.
dalek r32319 | kjs++ | trunk: 20:05
: [pirc/new] fix some stuff.
: + disallow PIR registers in PASM mode
: + disallow all unknown characters in PASM lexer
: + clean up some output.
: + PIRC now does (almost properly) {PIR,PASM}->{PIR,PASM}
diff: www.parrotvm.org/svn/parrot/revision?rev=32319
particle kjs: i doubt pasm allows macros, because pasm is human-readable bytecode. 20:06
bacek morning everybody
kj particle: ok, that's good news. that saves me a lot of messing with code.
who uses PASM anyway... 20:07
chromatic Some of the tests, which is awfuul.
bacek davidfetter: I'm probably going to OSDC. It's not so far from me ;) 20:08
davidfetter bacek, cool. interested in a joint table with postgres?
bacek, or do you know anybody who'd be up for staffing one? 20:09
bacek davidfetter: I don't have much experience with postgres...
davidfetter that's fixable ;) 20:10
where are you?
bacek right now? At home ;) 30 km from Sydney, CBD 20:12
20:20 mberends_ joined 20:31 masak joined 20:37 rdice joined
davidfetter wonders whether he should invest in some anti-drop-bear gear 20:46
Infinoid definitely 20:47
protective headgear with spikes
allison Coke: on the board of both TPF and PaFo
allison idly wonders if PaF would catch on better than PaFo 20:49
particle or ParF :) 20:50
chromatic I parf with excitement!
allison Narf! Poit!
diakopter fran. 20:51
particle ParF and the technicolor VirM
allison Are you thinking what I'm thinking, Pinky? I think so Brain, but why does Broccoli look so green? 20:52
particle__
particle++
dalek r32320 | allison++ | pdd22io: 21:02
: [pdd22io] Remove the unused and useless mode integer value (it's all passed in 'flags').
diff: www.parrotvm.org/svn/parrot/revision?rev=32320
Coke wishes chrome had better help 21:03
21:04 MariachiElf joined
Coke finds #chromium on freenode and is helped. 21:27
21:30 mberends_ joined
Coke the new osx deprecation warning ticket is a duplicate. 22:00
(if anyone wants to go RT spelunking)
22:03 iblechbot joined
dalek r32321 | kjs++ | trunk: 22:18
: [pirc/new] disable parrot's search path for .include files. This needs more thought/testing. + prevent double free().
: + fix a forgotten argument.
diff: www.parrotvm.org/svn/parrot/revision?rev=32321
22:18 rob joined 22:19 TiMBuS joined
dalek r32322 | kjs++ | trunk: 22:21
: [pirc/new] woops, typo.fixed previous commit.
diff: www.parrotvm.org/svn/parrot/revision?rev=32322
r32323 | kjs++ | trunk: 22:28
: [pirc/new] disable construction of an interpreter in hdocprep, it segfaults for some reason on linux.
diff: www.parrotvm.org/svn/parrot/revision?rev=32323
22:36 johbar joined 22:38 dmknopp joined 22:46 gryphon joined 22:53 Psyche^ joined 23:06 Limbic_Region joined 23:12 bacek_ joined
jonathan hopes United Airlines are good 23:19
23:22 Hinrik joined
bacek_ purl: #perl6? 23:35
purl #perl6 is at irc.freenode.net.
jonathan OK folks, flight booked, will be at Parrot Developer Summit. 23:37
funding++
23:44 Whiteknight joined

Powered by Cro and the RakuĀ® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!