#parrot Parrot 2.1.1 Released! | parrot.org/ | Channel log: irclog.perlgeek.de/parrot/today | Tasks: Fix HLL bugs! Fix and test corevm target!
Set by moderator on 9 March 2010.
03:25 hercynium joined
dalek kudo: fefb298 | (Solomon Foster)++ | t/spectest.data:
Turn on S03-operators/misc.t.
 03:32
03:39 Austin joined
dalek rrot: r44917 | petdance++ | trunk/src/pmc/eventhandler.pmc:
localizing vars and consting
 03:49
dukeleto 'ello 04:08
sorear hello 04:12
dukeleto sorear: how goes it?
sorear how goes what? 04:13
dukeleto sorear: the heat death of the universe ;) It is just a saying. 04:14
Infinoid life, the universe, and everything?
04:17 Andy joined
dalek rrot: r44918 | petdance++ | trunk/src/pmc/file.pmc:
consting
 04:23
Austin sings, "Henry Kissinger! How I'm missing ya. And wishing you were here!" 04:27
04:27 silug joined
sorear dukeleto: are you dalek' 04:31
04:31 janus joined
sorear dukeleto: are you dalek's operator? 04:31
plobsing dalek? 04:36
purl i think dalek is #parrot's spammy little rss bot or (see: dalek plugins)
plobsing dalek plugins?
purl i heard dalek plugins was github.com/Infinoid/dalek-plugins/tree/master
04:37 eternaleye joined
plobsing sorear: Infinoid is a good bet if you're looking for dalek stuff 04:37
sorear yeah, I was trying to find out how to get it to correctly credit me on commits 04:39
looks like I'm going to have to somehow get into the main Parrot CREDITS 04:40
dalek izkost: 65ca056 | dukeleto++ | src/pmc/p5 (5 files):
Get rid of __attribute__ warnings
 04:47
sorear ooh, there're more than the two of us on blizkost? 04:48
dalek rrot: r44919 | plobsing++ | trunk/src (2 files):
eliminate useless use of private1 flag on NCI PMC
 04:55
dukeleto sorear: i can add you to the CREDITS, if you want 05:14
sorear: i read your recent commits to blizkost, they are very interesting. It makes blizkost make a lot more sense 05:15
sorear: I am hacking on PL/Parrot, Parrot embedded in Postgres, so the explanation of how data marshalling is done in blizkost is very enlightening 05:17
sorear Actually, nothing in that document even exists yet. I was writing out my ideas so that I could get jnthn to check them 05:41
the implementation phase is *next*
dukeleto sorear: well, i like your ideas :) 05:45
sorear: if you send me a patch to the parrot CREDITS, i can apply it 06:11
sorear but... I haven't even done anything 06:14
dukeleto sorear: if you have committed to blizkost, you have done something 06:15
sorear oh
I thought I had to have worked *on Parrot*
dukeleto sorear: not really. the CREDITS file is used to track karma for any Parrot-based project
sorear++ 06:17
karma sorear
purl sorear has karma of 11
sorear @karma sorear
dukeleto sorear: well, i am going to bed. talk to you later!
sorear oh, I guess that wiki page was out of date 06:18
sleep well
dalek izkost: 9fb0458 | (Stefan O'Rear)++ | docs/SEMANTICS:
Third, much simplified, ready to use draft
 06:25
sorear ... why does Parrot CREDITS have a U+FEFF ZERO WIDTH NO-BREAK SPACE as the seventh character? 06:28
some processing tool must have screwed up a BOM
07:20 fperrad joined
dalek kudo: 60f9b3b | (Solomon Foster)++ | t/spectest.data:
Move some S03-operators tests back into their proper location, delete duplicate, turn on S03-operators/short-circuit.t.
 07:33
07:37 chromatic joined
dalek kudo: 345f63d | (Solomon Foster)++ | t/spectest.data:
Turn on S04-statement-modifiers/if.t.
 07:50
08:18 snarkyboojum joined, snarkyboojum left 09:07 iblechbot joined
sorear Is there such a thing as a "PMC class with no public constructor" ? 09:12
09:22 eternaleye joined 09:27 bacek joined
bacek o hai 09:31
mikehh hi bacek 10:01
bacek aloha mikehh
mikehh bacek: the problems still with codetest in ops_pct are all related to nqp files being treated as perl or pir 10:04
bacek hmm... We have to teach codetest how to deal with nqp.
I suspect we will have more and more nqp stuff inside
mikehh as fart as I can see all perlcritic failures are on nqp files 10:05
excuse me far 10:06
the other problem is ext/nqp-rx/src/gen/settings.pm which really should not be tested 10:09
bacek yes, it's another generated file 10:11
mikehh src/ops/core_ops.c is still being removed by make realclean - needs sorting out there 10:12
not sure if it should be removed or out of svn control 10:13
bacek it shouldn't be removed. At the of the day we will have all src/ops/*.c files committed into vcs 10:14
10:36 AndyA joined
dalek rrot: r44920 | bacek++ | branches/pcc_hackathon_6Mar10/t/op/calling.t:
Update test to use get_results after invokecc.
 10:53
rrot: r44921 | bacek++ | branches/pcc_hackathon_6Mar10/src/sub.c:
Propogate current signature into callee context in continuation unwind environment.
rrot: r44922 | bacek++ | branches/pcc_hackathon_6Mar10/src/ops/core.ops:
Remember freshly built signature in current context in set_returns.
rrot: r44923 | bacek++ | branches/pcc_hackathon_6Mar10/src/call/args.c:
Fix handlig "Pi..." signatures: actually append invokant if it's first argument.
 11:10
rrot: r44924 | bacek++ | branches/pcc_hackathon_6Mar10 (3 files):
Pass error flag to check into fill_params. We are usually don't check RESULTS_COUNT
bacek seen chromatic 11:12
purl chromatic was last seen on #parrot 13 hours, 14 minutes and 3 seconds ago, saying: msg whiteknight See r44911; I think we're very close. The interaction of :vtable and :method might not be quite right yet though.
bacek seen allison 11:13
purl allison was last seen on #parrot 1 days, 16 hours, 53 minutes and 16 seconds ago, saying: :) [Mar 12 18:19:44 2010]
bacek msg chromatic t/op/calling.t down to 3 failures. 2 related to "op result_info" and can be ignored for now. calling_97 is a beast. It fails to find vtable override. 11:15
purl Message for chromatic stored.
bacek msg allison t/op/calling.t down to 3 failures. 2 related to "op result_info" and can be ignored for now. calling_97 is a beast. It fails to find vtable override.
purl Message for allison stored.
11:22 payload joined 12:02 lucian joined
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#32671), fulltest) at r44924 - Ubuntu 9.10 amd64 (gcc with --optimize) 12:05
12:44 fperrad joined 12:52 joeri joined
dalek kudo: 2751645 | (Solomon Foster)++ | src/core/ (2 files):
Fix Num.ACCEPTS(Complex), and implement Rat.ACCEPTS.
 13:45
kudo: 8f0ca97 | (Solomon Foster)++ | t/spectest.data:
Turn on S03-smartmatch/any-num.t.
14:20 iblechbot joined 14:23 _shockwave joined 15:08 Psyche^ joined 15:48 Myhrlin joined 15:59 tetragon joined 16:07 kid51 joined 16:21 payload joined 16:26 theory joined 17:01 eric_j joined 17:06 chromatic joined, davidfetter joined 17:40 Austin_away left, Austin_away joined
Austin Good morning, Parrot. 17:41
17:58 Whiteknight joined 18:02 _2x2l joined
dukeleto Austin: hola 18:10
Austin What's new, duke? 18:11
mikehh rakudo - spectest_smolder #32674 - All tests successful - some TODO passes - on parrot r44924 - Ubuntu 9.10 amd64 (gcc with --optimize) 18:16
Whiteknight hello everybody 18:20
purl Hello Dr. Nick!
18:23 hercynium joined
Austin hello, whiteknight 18:25
Whiteknight hey Austin 18:26
Austin What's up today?
Whiteknight not much. Just got back from a night in Lancaster 18:30
I've decided to give up my life of technology, move to a farm and raise...I don't know...goats? 18:31
Austin sings, "As I walk through the valley where I harvest my grain, I take a look at my wife and realize she's very plain!" 18:32
Whiteknight but I've been milking and plowing so long that even izejiel thinks my mind is gone
Austin *ezekiel
www.youtube.com/watch?v=xo74Dn7W_pA
Whiteknight yeah, stupid fingers today 18:33
Austin tonight we're gonna party like it's 1699
So what's up in Lancaster, other than smoked turkey legs? 18:36
Whiteknight nothing really. my cousin got married, so big shindig for that 18:37
Austin Well, congratulations to your cousin.
Was this wedding Friday, or Saturday? 18:38
Whiteknight sa
sat
Austin YOu RON?
You
19:11 allison joined 20:06 kurahaupo joined
Whiteknight Austin: I would *really* like to have an option to make unit tests happen in a specific order 20:18
or, at least to "pin" certain tests to the top of the order 20:19
20:29 kurahaupo joined
dalek kudo: d631c8f | (David Romano)++ | (9 files):
Merge remote branch 'upstream'
 20:30
kudo: e013cab | (David Romano)++ | t/spectest.data:
Merge remote branch 'upstream'
kudo: 5d15313 | (David Romano)++ | (3 files):
Merge remote branch 'upstream'
kudo: 77ee50d | (David Romano)++ | t/spectest.data:
Turn on S06-other/main-eval.t remove unneeded comment lines in t/spectest.data
20:35 bacek joined
cotto hi bacek 20:48
do you know why t/compilers/pct/complete_workflow.t fails in ops_pct?
bacek cotto, no idea... 20:50
dalek rrot-linear-algebra: b62053b | Whiteknight++ | src/pmc/nummatrix2d.pmc:
change nummatrix2d.intialize_from_* methods to fill row-first.
 20:52
purl dalek: that doesn't look right
rrot-linear-algebra: ccfc0bf | Whiteknight++ | t/pmc/nummatrix2d.t:
finish the conversion of nummatrix2d tests to kakapo. Also, fix all test failures
cotto have you had any tuits to look into it? 20:57
bacek cotto, can you try to sync with trunk, regenerate core_ops.c with ops2c and rerun tests? 20:59
cotto sure 21:00
21:01 payload joined
dalek rrot: r44925 | bacek++ | branches/pcc_hackathon_6Mar10/src/call/args.c:
Don't try to clone_key_arg for PMCNULL argument.
 21:01
rrot: r44926 | bacek++ | branches/pcc_hackathon_6Mar10/src/ops/core.ops:
Don't reuse old signature in set_returns
cotto prepares for stupid merge conflicts
dalek rrot: r44927 | bacek++ | branches/pcc_hackathon_6Mar10 (5 files):
More fixes to C side of PCC:

  - Accept NULL singnatures in fill_params_from_varargs.
  - Propogate current_signature into parent contenxt in pop_contenxt.
  - Update call_obj before filling results in C PCC. CallSignature can be
   during invocation.
kudo: 9ffe679 | jonathan++ | src/pmc/perl6multisub.pmc:
Tweak to make sure Perl6MultiSub matches a (somewhat dubious) Parrot heuristic.
 21:10
kudo: 558c638 | jonathan++ | src/ (2 files):
Apply patch from colomon++ to move Int.ACCEPTS into the setting.
21:11 hercynium joined 21:12 shockwave joined
Austin Whiteknight: ping 21:21
Whiteknight Austin: pong 21:22
Austin What order?
purl order is deliberately random. (Although not random enough.)
Whiteknight Austin: depends, but there are cases where I want certain tests to happen first
Austin Would it be enough to sort the method names asciibetically, or do you want something more? 21:23
Oh
Just "Run these 4 first, then all the rest?"
Whiteknight I don't know what would be best, honestly. There obviously is benefit to the randomness
yeah, that would be perfect. And obviously we could expand that list to cover all tests if we want perfect ordering 21:24
Austin Sure.
What's the reason for the order?
Whiteknight well, in my tests for PLA, I'm wanting to test the VTABLEs before the METHODs
so when a failure happens, it's easier to see where the problem originated 21:25
Austin That sounds like a lot of methods to be typing in one at a time. How many vtable test-methods are there?
Whiteknight I'm still formulating what I want 21:26
I'm thinking about breaking into separate files, but then I need to tweak my harness to run the files in the right orders
Austin Heh
How about all the vtables in random order, then the methods?
dalek kudo: 8915a61 | jonathan++ | t/spectest.data:
Turn S02-builtin_data_types/declare.t back on.
 21:27
Whiteknight I don't know what I need. Let me think about it and I'll get back to you
Austin Okay. 21:28
What I'm thinking is a loader that makes the suite in the right order.
Whiteknight would it be reasonably easy to, on request, run the tests in the order they are defined in the file? 21:29
Austin So if you just want to test the vtables first, we could make the test prefix be "test_vtable_..." and load, and then make the test prefix be "test_method_..." and load, and the sub-groups are random (because hash keys are randomized, bacek++) but the groups are relatively ordered.
Whiteknight or do we not have that ordering information at the time the test is run?
Austin I think we could do that. (The information isn't explicitly available, but you can query a sub for its offset) 21:30
bacek
.oO( OrderedHash ftw? )
I did fix it in 2.1
Austin bacek: I don't think NameSpace or Class returns an ordered hash of methods.
bacek Ah... 21:31
dalek kudo: a2ceb6f | jonathan++ | t/spectest.data:
Turn unspace.t back on.
 21:33
bacek But you can change Class PMC to use OrderedHash instead of Hash :)
class.pmc:510
Austin Okay. And if I define methods A, B, A, A, B, A, what order are they in? 21:34
bacek A,B
Austin Heh 21:35
bacek Redefinition doesn't update position
dalek rrot: r44928 | cotto++ | failed to fetch changeset:
[opsc] sync branch with trunk, update manifest and core_ops.c
bacek cotto, any luck with complete_workflow? 21:38
cotto nope. still fails 21:41
bacek sigh... 21:42
How it's can be?
Austin whiteknight: I'll look at ordering. Open a ticket when you figure out what you want, please? 21:43
cotto needs a nap
Whiteknight Austin: sure thing
bacek have to go to $dayjob
Austin code.google.com/p/kakapo-parrot/
21:44 lucian joined
Austin How do I know if a particular ticket / changeset has made it into a release (and if so, which release)? 21:48
nm. testcases are cheap 21:53
Whiteknight: N4N, but suite already has a .sort that does asciibetical. 21:57
Whiteknight oh, okay
Austin Just put .sort.run instead of .run
That's a guaranteed order, but not very convenient. 21:58
dalek rrot-linear-algebra: f60dadf | Whiteknight++ | t/ (2 files):
fix a weird non-failure failure. Also, fix an issue where our simpleton test harness craps out on complex lines of TAP output
 21:59
rrot-linear-algebra: 0081b86 | Whiteknight++ | t/pmc/nummatrix2d.t:
add tests for VTABLE nummatrix2d.get_string
rrot-linear-algebra: a40d08f | Whiteknight++ | t/pmc/nummatrix2d.t:
fix/add nummatrix2d tests
22:03 lucian_ joined 22:26 patspam joined
dalek kapo: dd07c2b | austin++ | (3 files):
Added ordering to UnitTest/Loader.nqp
 22:30
Austin Whiteknight: I just pushed a loader that has an extra ordering step.
dalek kudo: 17766db | jonathan++ | src/Perl6/Actions.pm:
Add a missing twigil in signature check from STD.
 22:54
kudo: 1d5c09f | jonathan++ | t/spectest.data:
Turn S06-signature/errors.t back on.
kudo: 7a56f35 | jonathan++ | src/ (3 files):
Get $x.$y() idirect method calls working again.
kudo: e704ef9 | jonathan++ | src/Perl6/Actions.pm:
The illegal twigil detection was a little...over-eager...to complain. Fixed in this patch.
Whiteknight Austin: A very general idea to resolve the issue would be a hook in the suite where I pass a callback to sort the methods 23:21
Austin ^^ 23:22
I just pushed a Loader with an ordering method. :)
By default, it's @tests.unsort.
Whiteknight oh, nice
Austin I'm thinking if you want the vtable/method partition, we can do that by putting a little extra naming into play 23:23
test_vtable_foo..., test_method_bar...
dalek parrot: 708d0a2 | dukeleto++ | Makefile:
Give psql the -AX flags when running tests. selenamarie++
 23:27
23:31 Andy joined 23:32 payload joined
Whiteknight Austin: so how do I do the ordering, override order_tests? 23:33
Austin Yeah. Check the test case.
method order_tests(@tests) { @tests.sort }
or 23:34
{ my @vt := @tests.grep: -> $name { $name.substr(0, 6) eq 'test_v' }; my @meth := @tests.grep: -> $name { $name.substr(0, 6) eq 'test_m' } ; @vt.append: @meth; } 23:35
or 23:36
{ @tests.sort.reverse }
(same thing - all test_v before test_m)
Then override default_loader in your testcase class to use the loader. 23:38
method default_loader() { My::Custom::Loader.new; }
purl, pla? 23:40
purl i guess pla is really focusing on fast 2D data buffers
Austin purl, parrot-linear-algebra?
purl austin: wish i knew
Austin how is this possible? 23:41
Whiteknight purl, pla is Parrot-Linear-Algebra 23:42
purl ...but pla is really focusing on fast 2D data buffers...
Whiteknight purl, pla is also Parrot-Linear-Algebra
purl okay, Whiteknight.
23:43 snarkyboojum joined
Whiteknight Parrot-Linear-Algebra is a package of matrix datatypes for Parrot 23:44
Austin Who is Austin Bain, and why does he have my name, anyway?
Whiteknight Parrot-Linear-Algebra?
purl Parrot-Linear-Algebra is a package of matrix datatypes for Parrot
Austin github: austin hasn’t done anything yet 23:45
Whiteknight send him an email, see if you can get the name
Austin And "Austin Hastings" is invalid, just like "Austin_Hastings" is invalid.
What kind of fucking idiot monkeys work at github? 23:46
Whiteknight only the best fucking idiot monkeys
Austin "Oh, nobody would every want a name with separate words in it - that's ridiculous. What do you think this is, the FUTURE?"
Whiteknight change your nick to "Austmeister" 23:47
purl Whiteknight: that doesn't look right
sorear register austin on gitorious just to be confusing
Austin I'll go with "aghast"
Which I am, at github's weak-ass login filter...
sorear++
Too late, though. 23:48
(I'm already 'austin' on gitorious) 23:49

Powered by Cro and the Raku® Programming Language.

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

Thank you!